Findings Of Fact The Respondent owns property in Lake County, Florida which adjoins North Lake Holly. North Lake Holly is a fresh water lake. On an undetermined date between December, 1975 and September, 1976, the Respondent caused a horseshoe-shaped basin to be dredged along the shoreline of North Lake Holly adjoining his property. The fill material taken from the dredged area was deposited along the shore of the lake to farm a beach. The basin is approximately 90' long, 50' wide, and 6' deep. The Respondent has erected a dwelling house on his property, and it appears that the dredging was done in order to transform the shoreline of the lake from a vegetated littoral zone to a beach and boat basin. The Department confirmed the violations in December, 1976, and sought to negotiate a restoration plan with the Respondent. The formal Notice of Violation was issued an November 17, 1977. The dredged area was previously a shallow littoral zone dominated by wetlands vegetation. The most prevalent vegetation was sawgrass, but there were also abundant quantities of cattails, maidencane, arrowhead, and willows. The dredging activity relates to only a small portion of the shoreline of North Lake Holly. The activity nonetheless has resulted in the alteration of the characteristics of the lake. The marsh area which fringes the lake serves as habitat for fish and other wildlife, and also serves to filter runoff which enters the lake from the uplands. The Respondent's activities have obliterated a portion of the wildlife habitat, and provide an avenue for some uplands runoff to be discharged directly into North Lake Holly without the benefit of being filtered through wetlands vegetation. The quality of waters in central Florida lakes is related directly to the amount of development along the shoreline. The greater degree of alteration of the shoreline, the greater degree of deterioration of water quality, and the greater the deterioration of wildlife habitat. A project of the magnitude of that accomplished by the Respondent may have no clearly measurable impact upon water quality and wildlife habitat since the rest of North Lake Holly is surrounded by a broad littoral zone. The only impact that the project can have is, nonetheless, adverse. If a project such as the Respondent's is approved, the Department could not, consonant with due process and equal protection concepts prohibit further such alterations of the shoreline. It is likely that some aquatic vegetation will reestablish itself along the shoreline of the dredged area. Such a natural restoration will not, however, alleviate the negative impacts of the Respondent's dredging. The steep inclines of the dredged area will allow only a very narrow rim of vegetation, which cannot be expected to provide habitat and protect water quality to remotely the extent of the, previous undisturbed broad littoral zone. Furthermore, in the time since the project was completed, no significant vegetative zone has reestablished itself. It is possible for the Respondent to gain access to the lake for boating and other recreational purposes without totally obliterating the littoral zone that was in the area. The Department has offered a restoration plan which would accomplish this result. The Respondent undertook the dredge and fill activity without seeking a permit from the Department, and he continues to operate what amounts to a stationary installation which will serve as a source of pollutants to North Lake Holly without any valid permit issued by the Department. The Department has spent $229.41 in assessable costs in investigating and attempting to rectify the illegal dredge and fill activity undertaken by the Respondent.
The Issue The issue in this case is whether the Respondents, Kelly Endres and Ifrain Lima (Endres/Lima), are entitled to an Environmental Resource Permit (ERP) that would allow use of 0.535 acres of previously impacted wetlands for the construction of a single-family residence and associated structures, a 30' x 30' private dock with a 4' access walkway, and a 12' wide boat ramp (Project) at 160 Long Acres Lane, Oviedo, Florida (Property).
Findings Of Fact The following Findings of Fact are based on the stipulations of the parties and the evidence adduced at the final hearing. The Parties The Department is the administrative agency of the state statutorily charged with, among other things, protecting Florida's air and water resources. The Department administers and enforces certain provisions of chapter 373, part IV, Florida Statutes, and the rules promulgated, thereunder, in the Florida Administrative Code. Under that authority, the Department determines whether to issue or deny applications for ERPs. Respondents Endres/Lima own the Property and are the applicants for the ERP at issue in this consolidated proceeding. Petitioner Meier is a neighboring property owner to the south of the Property. Petitioner Meier's property includes a single-family residence with accessory structures and is located on Long Lake. Petitioner Meier is concerned that the NOI provides inadequate environmental protections and that there will be flooding on adjacent properties from the Project. Petitioner Hacker is the neighboring property owner adjacent to the south of the Property. Petitioner Hacker's property includes a single-family residence with accessory structures and is located on Long Lake. He is concerned with the completeness of the application for the Project, the calculation of wetland impacts, that reasonable assurances were provided, and that the Department's NOI ignores willful negligence and allows disparate treatment of Respondents Endres/Lima. Petitioner Kochmann is a property owner with a single-family residence and accessory structures located on Long Lake. She is concerned that the NOI is based on a misleading application and provides no evidence that the Respondents Endres/Lima made reasonable efforts to eliminate and reduce impacts detrimental to the environment. History of the Project and Application On April 12, 2018, Respondents Endres/Lima applied for an ERP for proposed wetland impacts associated with a planned single-family home on the Property. This was the first ERP application for the Property. The Department sent a Request for Additional Information (RAI) on April 24, 2018, and a second RAI on November 2, 2018. Respondents Endres/Lima provided a Mitigation Service Area Rule Analysis for "As If In-Basin" for the Lake X Mitigation Bank for the St. Johns River Water Management District Basins to the Department via email on May 10, 2018. Respondents Endres/Lima submitted revised plans to the Department on September 19, and October 30, 2018. On January 7, 2019, the Department denied the ERP application. The Department and Respondents Endres/Lima, on July 18, 2019, entered into a Consent Order (CO). The Department found, and Respondents Endres/Lima admitted, that approximately 0.80 acres of jurisdictional wetlands were dredged and filled without a valid ERP from the Department; and was done with improperly installed erosion and sedimentation controls. On August 22, 2019, Respondents Endres/Lima submitted a second ERP application. The Department sent an RAI on September 20, 2019, to which Respondents Endres/Lima responded on December 19, 2019. In addition, Respondents Endres/Lima reserved 0.60 of forested Uniform Mitigation Assessment Method (UMAM) wetland credits from the Lake X Mitigation Bank and provided the Department with an updated site plan and Lake X Mitigation Bank credit reservation letter. The Department issued an NOI on February 7, 2020, which was timely published in the Sanford Herald on February 9, 2020. Respondents Endres/Lima provided timely proof of publication to the Department on February 13, 2020. Consent Order and Compliance A warning letter was issued to Respondents Endres/Lima on January 30, 2019, for the dredging and filling of approximately 0.80 acres of forested wetlands and improper installation of erosion and sedimentation control. The CO, executed on July 18, 2019, required Respondents Endres/Lima to cease any dredging, filling, or construction activities on the Property, submit an application for an Individual ERP within 30 days, and pay $5,599.00 in penalties and the Department's costs and expenses. After the issuance of an ERP, Respondents Endres/Lima were also required to implement the restoration actions outlined in the CO. Respondents’ Endres/Lima’s application, dated August 19, 2020, was submitted to the Department on August 22, 2020. Respondents Endres/Lima paid the CO's penalties and costs, and had multiple meetings with the Department to complete the requirements of the CO. Respondents Endres/Lima’s expert, Mr. Exner, testified that he began working on a restoration plan for the Property, which will be provided to the Department once an ERP is issued. Permitting Criteria The Department reviewed the complete application and determined that it satisfied the conditions for issuance under Florida Administrative Code Rule 62-330.301, and the applicable sections of the ERP Applicant's Handbook Volume I (AH Vol. I). The Department also considered the seven criteria in rule 62-330.302 and section 373.414(1)(a), and determined that implementing the Project would not be contrary to the public interest. Water Quantity, Flooding, Surface Water Storage and Conveyance Respondents’ Endres/Lima's civil engineering expert, Mr. Herbert, testified that according to the drainage design, the Property would have swales on either side of the proposed residence to slope water away from the residence. There would also be a conveyance swale on the north property boundary to convey water from the street area and front yard toward the restoration and wetland areas with ultimate discharge to Long Lake. He stated that the elevation of the road at the front of the Property would be at 47.4 feet, and the elevation at the terminus of the swale would be at 45 feet. This would allow a 2.4-foot vertical fall for the swales to convey water to the lake. The design would preserve pre-development surface water flow over the Property to Long Lake, which is the lowest elevation in the area, and will ensure that storm water does not flood adjacent properties. Mr. Herbert also testified that the Project design would maintain pre-development water storage capacity. The imported fill that is currently on the Property in the flood plain would be removed and reshaped so that the lake elevation would be maintained and water can flow correctly. Elimination or Reduction of Impacts and Mitigation Respondents Endres/Lima provided the Department with design modifications to reduce impacts associated with the Project. These included a 15-foot restoration buffer along the lake front's northern shoreline, an elevated access walkway five feet above the wetland restoration area to the proposed dock, limiting the width of the access walk to four feet, and limiting the boat ramp width to a single-lane. In June 2015, an informal wetlands determination was conducted for the Property. The informal determination concluded that the entirety of the Property were wetlands. However, this was an informal determination and was not binding. In October 2016, before the first permit application was submitted, Mr. Exner did a wetlands delineation flagging prior to the Property being cleared or disturbed. Mr. Exner testified that, in his opinion, the Property was not all wetlands because large pines near the road had no high water marks, adventitious growth around the bases, or evidence of pine borer beetles along with other indicators of upland habitat. This wetland delineation was part of the permit submittal, was shown on the plans, was accepted by the Department, and was used for the preparation of the UMAM scoring. Mr. Exner's wetland delineation line was used by the Department to help determine and map the wetland impacts identified in the CO. The direct impact area was assessed at 0.54 acres with a secondary impact area of 0.02 acres for a total impact of 0.56 acres, and a functional loss score of 0.364. Respondents Endres/Lima reserved 0.6 forested UMAM mitigation credits, almost double the amount of functional loss under the UMAM assessment, agreed to purchase 0.46 credits. The excess mitigation bank credits implement part of a plan that provides regional ecological value and greater long-term ecological value than the area of wetland adversely affected. Secondary and Cumulative Impacts The Project's UMAM analysis assessed 0.02 acres, or 870 square feet, of secondary impacts. These impacts would be fully offset by the mitigation proposed for the Project. Petitioners' expert, Mr. Mahnken, noted three areas where he thought the application was incomplete. The first was that the site plan did not call out the location of the secondary impacts. However, Part III: Plans of Section B of the application, does not require that the site plan show the location of the secondary impacts. The application requirements for "plans" requires only the boundaries and size of the wetlands on the Property and provide the acreages of the upland areas, wetland impact areas, and the remaining untouched area. Second, Mr. Mahnken questioned the calculation performed to determine the secondary impact acreage. However, Mr. Mahnken read the information incorrectly and stated that the secondary impact area was 0.002 acres, or 87 square feet, when the UMAM score sheet clearly showed that the secondary impact area is 0.02 acres, or 870 square feet. In addition, the Department's witness, Ms. Warr, testified that even if the Department were to use Mr. Mahnken's analysis, the result would have been the same, i.e., the requirement to purchase 0.46 mitigation credits. Thus, Petitioners failed to support their claim that the Project would have adverse secondary impacts. Third, Mr. Mahnken asserted that cumulative impacts were not adequately addressed. He testified that the assessment for the Property using spill over benefits, in his opinion, was not enough to fully offset the impacts of the Project. Mr. Mahnken acknowledged, however, that his opinion was open to debate, and that he had not conducted any rigorous hydrologic evaluation in reaching his opinion. Respondents Endres/Lima had submitted a report prepared by Breedlove, Dennis & Associates (BDA Report) with their application in order to demonstrate compliance with section 10.2.8, ERP AH Vol. I, regarding cumulative impacts. The BDA Report utilized peer-reviewed hydrologic data that was reviewed and approved by the South Florida Water Management District, and was accepted by the Department pursuant to section 373.4136(6)(c). This was consistent with the Property's location within the mitigation service area for the Lake X Mitigation Bank. The Project is located within the Econlockhatchee River drainage basin, which is a nested basin within the larger St. Johns River [Canaveral Marshes to Wekiva] drainage basin. The Lake X Mitigation Bank is located outside of the Econlockhatchee River drainage basin, but the Project is located within the Lake X Mitigation Bank service area. The BDA report determined that: In summary, the Lake X Mitigation Bank is a regionally significant mitigation bank site that has direct hydrological and ecological connections to the SJRWMD basins, to include the cumulative impacts basin in which the subject property is located (i.e., SJRWMD Basin 19). The size, biodiversity, and proximity of the mitigation bank site to the SJRWMD basins, and the regionally significant hydrological connection between the mitigation bank site and the contiguous SJRWMD mitigation basins, supports the use of this mitigation bank site “as if in basin” mitigation for the Lima/Endres Wetland Fill Project. Additionally, the evaluation of factors, to include connectivity of waters, hydrology, habitat range of affected species, and water quality, demonstrates the spillover benefits that the Lake X Mitigation Bank has on the St. Johns River (Canaveral Marshes to Wekiva) mitigation basin, which includes the Econlockhatchee River Nested basin, and demonstrated that the proposed mitigation will fully offset the impacts proposed as part of the Lima/Endres Wetland Fill Project “as if in-basin” mitigation. The Lake X Mitigation Bank will protect and maintain the headwaters of two regionally significant drainage basins [i.e., the Northern Everglades Kissimmee River Watershed and the Upper St. Johns River Watershed (to include the nested Econlockhatchee River basin)], and will provide resource protection to both river systems (SFWMD Technical Staff Report, November 29, 2016). Furthermore, the permanent protection and management of the Lake X Mitigation Bank will provide spillover benefits to the SJRWMD basins located within the permitted MSA. Mr. Mahnken stated that his review of the Project did not include a hydrologic study and only looked at basic flow patterns for Long Lake. By contrast, the BDA Report included an extensive hydrologic study, looked at all required factors in section 10.2.8(b), ERP AH, Vol. I, and determined that the Project would be fully offset with the proposed mitigation. Thus, Respondents Endres/Lima provided reasonable assurance that the Project will not cause unacceptable cumulative impacts. Water Quality Rule 62-330.302(1)(e) requires that Respondents Endres/Lima provide reasonable assurance that the Project will not adversely affect the quality of receiving waters such that the state water quality standards will be violated. The conditions of the ERP would require the use of best management practices including a floating turbidity curtain/barrier, soil stabilization with grass seed or sod, and a silt fence. Respondent Endres/Lima's experts, Mr. Herbert and Mr. Exner, testified that there is an existing turbidity barrier in the lake around the property and a silt fence around the east half of the Property. While these items are not required by the Department until construction of the Project, part of the silt fence and the turbidity barrier are already installed on the Property and will be required to be repaired and properly maintained in accordance with the conditions of the ERP and Site Plan SP-2. Mr. Herbert testified that the Property will be graded in a manner that will result in a gentle sloping of the lake bank in the littoral zone, which would allow revegetation of the lake bank. Outside of the restoration area and the undisturbed wetlands, the backyard would be covered with grass to prevent migration of sand and soil discharging into the lake. Mr. Exner testified that the grass swales proposed for the Project would provide a considerable amount of nutrient uptake and filtration of surface water on the Property. Also, in the restoration area next to the lake, the restoration plan includes a dense planting plan with native species that have good nutrient uptake capability. Impacts to Fish and Wildlife Rule 62-330.301(1)(d) requires that Respondents Endres/Lima provide reasonable assurance that the Project will not adversely impact the value of functions provided to fish and wildlife and listed species by wetlands and other surface waters. Mr. Exner testified that, in his review of the Property, he did not identify any critical wildlife habitat. He visited the Property multiple times and he did not see any osprey nests, deer tracks, animal scat, gopher tortoises, or sand hill cranes. The Department's Ms. Warr testified that the Florida Fish and Wildlife Conservation Commission database was reviewed, and did not show any listed species in the area. Publication of Notice Petitioners argued that the notice published in the Sanford Herald on February 9, 2020, did not meet the requirements of section 373.413(4). Despite the notice having no effect on their ability to timely challenge the proposed ERP, Petitioners argued that the published notice was insufficient because the notice itself did not provide the name of the applicants or the address of the Project, only a link to the Department's permit file. Unlike the notice required in section 373.413(3), where a person has filed a written request for notification of any pending application affecting a particular designated area, section 373.413(4) does not specify the contents of the published notice. Section 373.413(4) does not require the published notice to include the name and address of the applicant; a brief description of the proposed activity, including any mitigation; the location of the proposed activity, including whether it is located within an Outstanding Florida Water or aquatic preserve; a map identifying the location of the proposed activity subject to the application; a depiction of the proposed activity subject to the application; or a name or number identifying the application and the office where the application can be inspected. In response to the published notice, the Department received approximately ten petitions challenging the NOI, including the petitions timely filed by Petitioners. Therefore, Petitioners were not harmed by any information alleged to have been left out of the published notice. Ultimate Findings Respondents Endres/Lima provided reasonable assurance that the Project will not cause adverse water quantity impacts to receiving waters and adjacent lands; will not cause adverse flooding to on-site or off-site property; and will not cause adverse impacts to existing surface water storage and conveyance capabilities. Respondents Endres/Lima provided reasonable assurance that the Project complied with elimination and reduction of impacts, and proposed more than adequate mitigation. Respondents Endres/Lima provided reasonable assurance that the Project will not cause adverse secondary impacts to water resources; and unacceptable cumulative impacts to wetlands and other surface waters within the same drainage basin. Respondents Endres/Lima provided reasonable assurance that the Project will not cause adverse water quality impacts to receiving water bodies. Respondents Endres/Lima provided reasonable assurance that the Project will not adversely impact the value of functions provided to fish and wildlife, and listed species by wetlands, or other surface waters. Petitioners failed to prove lack of reasonable assurance by a preponderance of the competent substantial evidence.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Department enter a Final Order granting Respondents’ Endres/Lima's ERP application. DONE AND ENTERED this 1st day of December, 2020, in Tallahassee, Leon County, Florida. S FRANCINE M. FFOLKES Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us COPIES FURNISHED: Filed with the Clerk of the Division of Administrative Hearings this 1st day of December, 2020. Jay Patrick Reynolds, Esquire Department of Environmental Protection 3900 Commonwealth Boulevard, Mail Station 35 Tallahassee, Florida 32399 (eServed) Neysa Borkert, Esquire Garganese, Weiss, D'Agresta and Salzman 111 North Orange Avenue Post Office Box 398 Orlando, Florida 32802 (eServed) Tracy L. Kochmann 249 Carolyn Drive Oviedo, Florida 32765 (eServed) Shelley M. Meier 208 Long Acres Lane Oviedo, Florida 32765 (eServed) Brian Hacker 170 Long Acres Lane Oviedo, Florida 32765 (eServed) Lea Crandall, Agency Clerk Department of Environmental Protection Douglas Building, Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399 (eServed) Justin G. Wolfe, General Counsel Department of Environmental Protection Legal Department, Suite 1051-J Douglas Building, Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399 (eServed) Noah Valenstein, Secretary Department of Environmental Protection Douglas Building 3900 Commonwealth Boulevard Tallahassee, Florida 32399 (eServed)
The Issue Whether the Respondents have polluted by dredging and filling within the landward extent of waters of the state, to wit: Choctawatchee Bay, without a permit for said dredging and filling. Whether the Orders for Corrective Action requiring removal of the fill material and restoration of the disturbed wetlands are reasonable and appropriate.
Findings Of Fact Ralph Ritteman has owned some interest in property which has been developed as a subdivision known as Sunset Point, including Sunset Point Addition, since approximately 1970. This property adjoins Choctawhatchee Bay and the Intercoastal Waterway in Walton County, Florida. In early 1984, a subdivision plat was recorded far lots 1-13. That plat showed two areas specifically not to be a part of it. Those two areas were the site of dredging and filling activities by Ralph Ritteman, wherein he had eleven ponds excavated and the spoil placed on the property. The Department of Environmental Regulation asserts that the excavation of the ponds and the placement of the spoil occurred in jurisdictional wetlands of the state. Ritteman asserts that the property is not jurisdictional and that no permit was needed. The primary dredge and fill activities occurred between June and October, 1984. Ritteman represented that these were the dates of the activity in an after-the-fact permit application which he filed with DER but later withdrew. In the course of his testimony in this proceeding, Ritteman took the position that he did the dredging and filling after a November 14, 1985, seminar presented by DER regarding wetland regulations wherein he was misled by a document distributed by DER entitled "State of Florida Joint Application for Permit," which covered dredge and fill guidelines. Specifically Ritteman testified that he did the dredging and filling after that seminar. It can only be concluded that Ritteman's testimony in that regard is false and that Ritteman did the unpermitted dredging and filling in 1984 and knew at the time that his activities were at the every least questionable. Specifically, (1) The 1983 plat shows these exact wetland areas as excluded; (2) A February 14, 1985, buyback agreement between Ritteman and Jerry Johnson, a purchaser of a lot on which the dredging and filling activity had occurred, showed that there was an existing concern about future action by a public agency to require restoration of the property to its prior condition; (3) John Brett, a Respondent herein because he purchased a lot from Ritteman in the affected area, bought the lot in 1985 with the existing ponds in place except for a land bridge which Ritteman had excavated (in 1985) and the fill placed for Brett to use as a homesite; and (4) Richard Sczcepanski, a Respondent herein, bought his lot in February, 1985, and the ponds and spoil were already in existence. Further, observation of Ritteman during his testimony and appearance at this proceeding leads this fact finder to the conclusion that Ritteman was less than candid in all of his testimony and dealings. After engaging in this unpermitted dredging and filling activity, Ritteman divided the affected area into lots and sold these lots to John and Dorothy Brett, Richard Szczepanski, Joe Williams, Jerry Johnson, Mohamed Yazdi, and Reza Toossi. A plat of the newly created lots was recorded as the Sunset Point Addition. All of these purchasers were named by DER in its Notice of Violation. Only the Respondents herein requested a hearing. The lots were sold by Ralph Ritteman and the Florida-Minnesota Land Company. However, that corporation's authority to do business in Florida was revoked on November 10, 1983, by the Secretary of State's Office. The DER discovered the unpermitted activity in 1986 and conducted an investigation to determine if the property impacted by the dredging and filling had been jurisdictional wetlands. The sites described in the Notice of Violation are vegetated with plant species consisting of black needlerush (Juncus roemerianus), sawgrass (Cladium iamaicense), salt meadow cordgrass (Spartina patens), salt grass (Disticalus spicata), and giant reed (Phraomites australis). The Department's investigation, using core samples, located the former surface of the undisturbed wetland beneath approximately 1 1/2 feet of dredged spoil material; the plant species Juncus roemerianus was also identifiable beneath the layer of spoil material placed on top of it by Ralph Ritteman. A beach berm is present at the shoreline interface of the marsh areas with Choctawhatchee Bay. Beach berms such as this one are typically built up in most marshes by the wave action. On the site are piles of unconsolidated fill material that was excavated from the pond areas. Included in this excavated material is muck and black silt-type material associated with salt marsh and gray clay material which was the underpan or confining layer. The Soil Survey Report of 1985 for Walton County shows that on the south site of the dredging activities, the soil type (prior to the excavation and filling) was Duckston muck, which is found in frequently flooded areas in very poorly drained sandy soils in marshes bordering salt water bays. Duckston muck consists of a 4-inch surface layer of black muck over loamy sand. The northern site soil type is Dirego muck, also found in frequently flooded areas with very poorly drained organic soils that occur in tidal marshes. Dirego muck consists of about 28 inches of muck overlying fine sand and loamy fine sand. A 1982 aerial photo clearly shows the delineation between the marshgrass area and the uplands. The current conditions at the site are entirely consistent with the delineation shown in this photo. There are remaining wetlands at the site in an area denoted as the homeowners park on the plat map of the Sunset Point Addition. There is an interchange of water between the remaining wetlands and the bay. Based on all the data, including, but not limited to, aerial photographs, remnant and existing vegetation, site observations, topography, hydrological data and soil types, it is clearly established that, prior to this unpermitted activity, there was a regular, periodic interchange and exchange of water between these wetlands and waters of the state. The two areas of unpermitted activity clearly fall within the jurisdictional wetlands of the state. Choctawhatchee Bay is brackish water and is tidally influenced. In a misapplication of the statute and rule, Ritteman offered into evidence a survey which purported to show the 1 in 10 year flood event elevation. This elevation line was set by surveyor Basil Boles memory of rainfall and his observation of the rack-line or detritus on the beach as it existed in October, 1988. This elevation was not developed by the appropriate engineering techniques required by Section 403.8171, Florida Statutes, and is therefore given no weight. The unpermitted dredging and filling has resulted in pollution as defined by statutes and it eliminated and destroyed plant life in jurisdictional wetlands. It also eliminated the interchange of waters and the contribution of that interchange to the ecology and viability of the marsh system in the area. The Department expended in excess of $494.23 in investigating this violation, but it sought only $494.23 in the Notice of Violation.
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Department of Environmental Regulation enter a Final Order and therein: Find Ralph Ritteman guilty of the violations charged for the unpermitted dredge and fill activity within the landward extent of waters of the state. Order Ritteman to bear the cost of and to perform restoration as specified in the Orders for Corrective Action. Order Ritteman to pay $494.23 to the Department of Environmental Regulation for the investigation of this violation. Order John and Dorothy Brett, Joe Williams and Richard Szczepanski to provide cooperation and site access during the restoration activities. DONE and ENTERED this 18th day of January, 1989, in Tallahassee, Florida. DIANE K. KIESLING Hearing Officer Division of Administrative Hearings The Oakland Building 2900 Apalachee Parkway Tallahassee, FL 32399-1550 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 18th day of January, 1989. APPENDIX TO RECOMMENDED ORDER IN CASE NOS. 88-2560/88-3532/88-3533 The following constitutes my specific rulings pursuant to Section 120.59(2), Florida Statutes, on the proposed findings of fact submitted by the parties in this case. Specific' Rulings on Proposed Findings of Fact Submitted by Petitioner Department of Environmental Regulation Each of the following proposed findings of fact are adopted in substance as modified in the Recommended Order. The number in parentheses is the Finding of Fact which so adopts the proposed finding of fact: 1 (1, 6 & 7); 2 (3 & 5); 3(6 & 8); 4(9); 5(10); 6(15); 7(16); 8 first paragraph (2); and 9(18). The last two paragraph of proposed finding of fact 8 are subordinate to the facts actually found in this Recommended Order. Specific Rulings on Proposed Findings of Fact Submitted by Respondent Ralph Ritteman 1. The proposed findings of fact and conclusions of law and argument are so intermixed in Rittemans proposed order that specific rulings are difficult. However, to the extent that proposed facts are not actually contained in this Recommended Order, they are rejected as being unsupported by the credible, competent, substantial evidence. Specific Rulings on Proposed Findings of Fact Submitted by Respondent Szczpanski 1. The proposed findings of fact and conclusions of law and arguments are so intermixed in Respondents proposed order that specific rulings are difficult. Further, Respondents attempt to introduce new evidence regarding current condition of the property and the adjoining waterbodies and engineering standards. To the extent that proposed findings of fact are not actually contained in this Recommended Order, they are rejected as being unsupported by the credible, competent, substantial evidence introduced at hearing. COPIES FURNISHED: Dale H. Twachtmann, Secretary Department of Environmental Regulation Twin Towers Office Building 2600 Blair Stone Road Tallahassee, FL 32399-2400 Richard L. Windsor Assistant General Counsel Department of Environmental Regulation Twin Towers Office Building 2600 Blair Stone Road Tallahassee, FL 32399-2400 Ralph Ritteman Post Office Box 1747 Santa Rosa Beach, FL 32459 John Brett 532 Clifford Street Ft. Walton Beach, FL 32548 Joe Williams 10 Marlborough Road Shalimar, FL 32579 Richard Szczepanski Post Office Box 855 Shalimar, FL 32579
Findings Of Fact Charlotte Highlands is an approximately 97-acre mobile home subdivision in Charlotte County, Florida. The roads in the subdivision are unpaved. The stormwater sheet flow in the area is from west to east. To the east of Charlotte Highlands is a 21-acre hardwood swamp, the wetlands in question in this proceeding. Stormwater from the 97-acre subdivision west of the wetlands and from the 250 acres west of the subdivision flows to the east into the wetlands. Water flows out of the wetlands to the east, from the 21-acre wetlands through a stream into Myrtle Slough. Myrtle Slough is part of the waters of the State. The County wishes to create a stormwater drainage system for Charlotte Highlands. Under the County's plan, stormwater from the 97-acre subdivision would be discharged into the wetlands owned by Desrosiers Brothers. Although the County and the Department view this project as involving only the discharge of stormwater from the 97-acre subdivision into the wetlands, the stormwater discharged would include the stormwater flowing into the 97-acre subdivision from the 250 acres located directly west of the subdivision. The County met with individuals from the Southwest Florida Water Management District, and that agency questioned the method of calculations used by the County in determining the amount of runoff into the proposed drainage system. Although new calculations of stormwater runoff volume were performed by the County, those new calculations were not provided to the Department in the County's permit application. The wetlands in question contain cypress, maples, laurel oak, bay trees, percia, dahoon holly, buttonbush, ferns, palmetto, and wet pine. Some of these species, especially the maples, cannot withstand much flooding. The outflow from the wetland into Myrtle Slough is via a natural stream. Although there are some indications that some excavation may have taken place in the stream, such as the spoil located near the cattle watering pond near the mouth of the wetlands, water flows from the wetlands to Myrtle Slough through a natural watercourse with no man-made connections. The hydroperiod is the length of time water stays in a wetlands before it drains out of the wetlands. This determines the water level, the critical factor affecting a wetland's ability to perform its vital functions. If the rate or volume of either the inflow or outflow of a wetlands is altered enough, the water level changes, usually with adverse environmental consequences. Certain species of flora will die off if the water level rises too much. Others require high water levels for their survival. In order to assess the effects of a proposed alteration to such a system, one must determine the existing high pool and low pool. Donald H. Ross established the high and low pools for the County. He went to the wetlands and observed the stain, rack, and lichen lines on tree trunks. He also observed the cypress buttress. Ross also determined the invert of the stream, the elevation at which water first starts to run in it. Based solely on this site visit, the County determined the high pool in the wetlands to be at 14.8 NGVD and the low pool to be at 14.1 NGVD. No rainfall data was collected and analyzed; no hydrological studies were performed; no observations were made over a period of time. There are two aspects of this project which can alter the hydroperiod of the wetlands. The first involves the amount of water entering the wetlands, and the second involves the amount of water leaving the wetlands. Currently, runoff from the 97-acre subdivision as well as the 250-acre area west of the subdivision drains toward the wetlands. The County intends to pave the roads in the subdivision and construct a system of swales. Although the paving will increase the impervious surface by an insignificant amount, the runoff will be delivered to the wetlands faster. Accordingly, peaks in water level will occur more suddenly with increased water arriving more quickly. Stormwatr is discharged into wetlands to take advantage of the pollutant-filtering functions of wetlands vegetation. To realize this function, the water must be held in the wetlands for a certain amount of time. The County intends to accomplish this by the installation of a control structure, known as a weir, which will regulate the amount of water leaving the wetlands. The County proposes to construct a weir on the stream between the wetlands and Myrtle Slough approximately 100 feet from the mouth of the wetlands. The top of the weir for this system will be set at 14.8 NGVD, the high pool established by Ross for the County. The weir will also have an orifice set at 14.1 NGVD, the low pool established by Ross and the County, which will allow a constant flow of water out of the wetlands at that elevation. The control structure will cause water to remain in the wetlands for a longer period of time, which will raise the water level in the wetlands by some amount. In order to accurately predict this amount, it is necessary to determine the storage capacity of the wetlands. The County calculated that a storage capacity of 177,761 cubic feet would be required for the wetlands to contain the first one-half inch of rainfall from the 97-acre subdivision. No calculations have been made as to the storage capacity required for the wetlands to contain the first one inch of rainfall from the 97-acre subdivision as well as the 250-acre area that drains into the subdivision which then drains toward the wetlands. The County has failed to establish the hydroperiod of the wetlands. Having failed to establish the hydroperiod of the wetlands, the impact of its project on the wetlands cannot be determined. As an alternative to this project the County considered rerouting the stormwater away from the wetlands. Diverting necessary water from the wetlands would result in the desiccation of the wetlands. However, an increased water flow if not properly discharged would likely result in an over impoundment of the wetlands. Either approach would have an adverse impact on a productive wetland system, such as the wetlands involved here, and a change in the vegetation would adversely impact the wetland's ability to treat the discharge. The treatment of stormwater in wetlands is a relatively new technique. Although some projects have been approved in other parts of the State, projects such as that proposed by the County have not been used yet in southwest Florida.
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is, RECOMMENDED that a Final Order be entered denying Charlotte County's application for a wetlands stormwater discharge facility permit. DONE and RECOMMENDED this 8th day of October, 1987, at Tallahassee, Florida. LINDA M. RIGOT, Hearing Officer Division of Administrative Hearings The Oakland Building 2009 Apalachee Parkway Tallahassee, Florida 32399-1550 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 8th day of October, 1987. APPENDIX TO RECOMMENDED ORDER, CASE NO. 87-0243 Although Charlotte County filed a document called Proposed Findings of Fact and Conclusions on the Evidence, rather than setting forth any findings of fact the County simply makes what it calls a Comparison of Evidence on Issue 1 and a Comparison of Evidence on Issue 2, listing under each heading excerpts from the testimony of each of the witnesses in this proceeding. Accordingly, no rulings are made herein on Charlotte County's proposed findings of fact since it is determined that there are none. Desrosiers Brothers' proposed findings of fact numbered 1-9, 15, 17, 24, 26, 27, and 38 have been adopted either verbatim or in substance in this Recommended Order. Desrosiers Brothers' proposed findings of fact numbered 10-12, 19-21, 23, 25, 29-37, 40, and 41 have been rejected as not constituting findings of fact but rather as constituting argument of counsel or recitations of the testimony. Desrosiers Brothers' proposed findings of fact numbered 13, 14, 16, 18, 22, 28, and 39 have been rejected as being unnecessary or subordinate to the issues under consideration herein. The Department's proposed findings of fact numbered 1, 2, 14 in part, 15, 16 in part, 17 in part, 18-22, 27, and 28 in part have been adopted either verbatim or in substance in this Recommended Order. The Department's proposed findings of fact numbered 5 and 6 have been rejected as not constituting findings of fact but rather as constituting argument of counsel or recitations of the testimony. The Department's proposed findings of fact numbered 16 in part, and 17 in part have been rejected as being unnecessary or subordinate to the issues under consideration herein. The Department's proposed findings of fact numbered 3, 4, and 7-13 have been rejected as being contrary to the weight of the evidence in this cause. The Department's proposed findings of fact numbered 14 in part, 23-26, and 28 in part have been rejected as not being supported by the evidence in this cause. COPIES FURNISHED: Dale Twachtmann, Secretary Department of Environmental Regulation 2600 Blairstone Road Tallahassee, Florida 32399-2400 Philip J. Jones, Esquire 201 West Marion Avenue Suite 301 Punta Gorda, Florida 33950 Matthew G. Minter, Esquire 18500 Murdock Circle Port Charlotte, Florida 33948-1094 Richard Grosso, Esquire Department of Environmental Regulation 2600 Blairstone Road Tallahassee, Florida 32399-2400 =================================================================
The Issue The issues to be resolved in this proceeding concern whether an environmental resource permit (number 4-109-0216-ERP) (the ERP) should be modified to allow construction and operation of a surface water management system (the project) for a residential development known as EV-1, in a manner consistent with the standards for issuance of ERPs in accordance with Florida Administrative Code Rules 40C-4.301 and 40C-4.302.
Findings Of Fact The applicant MCCDD is a unit of special purpose government established in accordance with the provisions of Chapter 190, Florida Statutes for purposes enunciated by that statute. MCCDD has applied for the permit modification at issue in this proceeding. The District is a special taxing district created by Chapter 373, Florida Statutes. It is charged with preventing harm to the water resources of the district and to administer and enforce Chapter 373, Florida Statutes, and related rules promulgated thereunder. Petitioner Larsen was born in Daytona Beach, Florida. Sometime early in 2002 she apparently moved to the Crescent Beach area and lived for 5-6 months. Crescent Beach is approximately 30 minutes from the EV-1 site. Since October 2002, Petitioner Larsen has been a resident of Live Oak, Florida. She resided for most of her life in Daytona Beach, approximately one hour and 20 minutes from the site. She has been involved with the approval process of the entire Palencia Development (DRI) since 1998, of which the subject parcel and project is a part. The Petitioner likes to observe wildlife in natural areas and to fish, swim, and camp. Ms. Larsen has visited the Guana River State Park (Park) which borders the Tolomato River. Her first visit to the Park was approximately one to two years before the DRI approval of the Palencia project. Ms. Larsen has used the Park to observe birds and other wildlife and to fish. She has fished the Tolomato River shoreline in the Park, and also at the Park dam located across the river and south about two and one-half miles from the EV-1 site. Ms. Larsen has seen the Tolomato River some 30 to 40 times and intends to continue using the Tolomato River and the Guana River State Park in the future. On several occasions she and Petitioner Billie have visited "out-parcel" residents of the Palencia development and viewed wildlife and birds and walked the Marshall Creek area and the marsh edge viewing various bird species. In June 2003, after this litigation ensued, she, her niece and out-parcel resident Glenda Thomas walked a great deal of the subject site taking photographs of wildlife. In July 2003, Larsen and Billie participated in a fishing boat trip in the Marshall Creek area. In September 2003, she and Petitioner Billie kayaked on two consecutive days in the Tolomato River and in Marshall Creek, observing various wildlife such as endangered Wood Storks. Petitioner Larsen has been actively involved for the past 12 years as an advocate for the protection of indigenous or native American burial, village and midden sites on private and government property. Petitioner Billie is a spiritual leader or elder of the Independent Seminole Nation of Florida. In that capacity he sees it as his responsibility to protect animals, rivers, trees, water, air, rains, fish, and "all those things." The Independent Traditional Seminole Nation consists of approximately 200 persons, most of whom reside in Southern Florida. Mr. Billie lives in Okeechobee, Florida, several hours distant by automobile from the project site. About 10 to 30 years ago Billie visited the Eastside of Tolomato River, to visit the beach, the river and other areas in what is now Guana State Park. He visited the dike or dam area and walked along the river front in what is now the Park. He checked on burial sites along the Tolomato River in what is now Guana State Park. Billie first visited the Palencia property about five years ago and has been back a number of times. He has observed various forms of wildlife there and has visited out-parcel owners in the development area to ensure that they do not destroy any burial sites. Billie considers himself an environmental and indigenous rights advocate charged with maintaining the earth and resources for the next generation and preserving sacred and burial sites of indigenous people. He has in the past assisted governmental entities in preserving sacred indigenous sites and burial sites and has participated in the reburials of human remains and their belongings. Sometime ago Billie went on a boat ride on the Tolomato River. Since the filing of the Petition in this proceeding he has been in a kayak on the Tolomato River twice and once in a boat in the vicinity of Marshall Creek. He has also observed Marshall Creek from Shannon Road. He has been on the EV-1 site three times, all in conjunction with this litigation. His concerns with the EV-1 project in part stem from alleged impacts to an indigenous burial ground which he feels he identified, due to the presence of "a lot of shell." However, all of the shell was located in a previously constructed road bed off of the EV-1 project site. He testified that he has had no training with regard to identification of archeological sites, but that he can "feel" if a burial site is present. He believes that the EV-1 project will adversely affect everyone just like it adversely affects him. The Project The project is a 23.83-acre, single-family residential development and an associated stormwater system known as EV-1. It lies within the much larger Marshall Creek DRI in St. Johns County, Florida. The project is in and along wetlands associated with the Tolomato River to the east and wetlands associated with Marshall Creek, a tributary of the Tolomato River, to the north. The project consists of thirteen residential lots, two curb and gutter roadway segments with cul- de-sacs (Hickory Hill Court and North River Drive), paved driveways to individual lots, concrete and pvc stormwater pipes, two stormwater lift stations, perimeter berms, four stormwater run-off storage ponds, and an existing wet detention stormwater pond, which was previously permitted and located south and west of the EV-1 site. The project will also have on-site and off- site wetland mitigation areas. All portions of the EV-1 site are landward of the mean high waterline of the adjacent water bodies. The project plan calls for permanent impacts to 0.82 acres of wetlands. A total of 0.75 acres of that 0.82 acre wetlands is comprised of fill for four access crossings for roads and driveways and a total of 0.07 acres is for clearing in three areas for boardwalk construction. MCCDD proposes to preserve 6.47 acres of forested wetlands and 5.6 acres of saltmarsh wetlands, as well as to preserve 10.49 acres of upland buffers; to restore 0.05 acres of salt marsh and to create 0.09 acres of salt marsh wetlands as mitigation for any wetland impacts. The EV-1 mitigation plan is contiguous to and part of the overall Marshall Creek DRI mitigation plan. The Marshall Creek DRI is also known as "Palencia." The upland buffers are included to prevent human disturbance of the habitat value of off-site wetlands. The upland buffers on the EV-1 site range from 25 feet in areas that do not adjoin tidal marshes to 50 feet in areas which front the Tolomato River or Marshall Creek. Within the 25-foot buffers restrictions include (1) no trimming of vegetation and (2) no structures may be constructed. Within the 50-foot buffers the same restrictions apply, except that for 50 percent of the width of each lot, selected hand trimming may be done on branches 3 inches or less in diameter between 3 and 25 feet above the ground surface. The buffers and other preserved areas will be placed in conservation easements, ensuring that they will remain undisturbed. The Stormwater Management System The 23.83 acre drainage area of the EV-1 project is divided into two types: (1) "Developed Treated Area" consisting of the houses, a portion of each residential lot, all driveways, sidewalks and both cul-de-sac roadway sections, comprising 11.27 acres and (2) "Undeveloped Buffer Area" consisting of the undeveloped portion of the residential lots or 12.56 acres. The buffer areas are located between the developed treated area and the surrounding receiving water. The developed and undeveloped areas of each lot will be separated by earthen berms. The berms will be constructed within each lot and will be a minimum of one foot high above existing ground level at the landward ledge of the natural buffer area. When water falls on the house and the surrounding yard it will be directed through grading to the berm of the lot. Once it reaches the berm it will be collected in a series of inlets and pipes; and once collected within the pipe system it will be stored within the collection system and in several storage ponds. The developed areas storage systems consisting of the inlets, pipes and storage ponds are then connected to two stormwater lift stations that transfer the stored runoff to an existing wet detention pond, known as the EV-2 pond, which is located immediately adjacent to the EV-1 project area. There are two pumps and a wet well in each pump station. The combination of storage ponds, piping systems, the wet wells and the pump stations provide storage of the entire required treatment volume which is 61,000 cubic feet. Actually, the system has been designed to treat 65,000 cubic feet, somewhat in excess of the required treatment volume. Even when the pumps are not running these components of the system are able to completely contain the required treatment volume. The system has been designed to capture and treat in excess of 1.5 inches of runoff. This is the runoff that would be generated from a 5.3 inch rainfall event which is expected to occur less than once per year. This l.5 inches of runoff would generate the required 61,000 cubic feet of treatment volume. In order to ensure that the design volume is not exceeded, the applicant has limited the amount of impervious service on each lot to a maximum of 10,000 square feet. In order to ensure that the on-lot ponds in the collection system are hydrologically isolated, they have been designed to be either completely lined or constructed with "cut- off walls" placed in soils with either a hard pan layer or a layer of low permeability. This would prevent the ponds from de-watering nearby wetlands by removing any hydrologic communication between those wetlands and the ponds. Further, the liners and cut-off walls will isolate the pond from the effects of groundwater. This will ensure that the ponds can be maintained at the designed water level and that, therefore, the collection system will have the required storage volume. The EV-2 pond provides for wet detention treatment and was previously permitted and constructed as part of the EV-2 project. In order to accommodate the additional flow from the EV-1 site, the existing orifice will be plugged and an additional orifice will be installed. No changes will be made to the shape, depth, width, or normal water elevation of the EV- 2 pond. The EV-2 pond discharges into wetland systems that are directly connected to the intracoastal waterway. The EV-2 pond discharges into a wetland system and has a direct hydrologic connection to the intracoastal waterway north of the Matanzas inlet. The District rules do not contain a legal definition of the intracoastal waterway; however, for the purpose of determining whether a project discharge constitutes a direct discharge to the intracoastal waterway, the waterway includes more than the navigable channel of the intracoastal waterway. (Projects that have a direct discharge to the intracoastal waterway north of the Matanzas inlet are not required to demonstrate that the post-development peak rate of discharge does not exceed the pre-development peak rate of discharge, because this criterion was designed to evaluate the flooding impacts from rainfall events.) Flooding in water- bodies such as the intracoastal waterway is not governed by rainfall, but rather by tides and storm surges. The system design includes a clearing and erosion control plan and specific requirements to control erosion and sediment. The system design incorporates best management practices and other design features to prevent erosion and sedimentation, including (1) capturing turbidity; (2) sodding and grassing side slopes; (3) filtering water; (4) use of siltation fences during construction; (5) removing sediment; (6) early establishment of vegetative cover; and (7) keeping water velocities low, at less than 2 feet per second. The EV-2 pond is hydrologically isolated from groundwater influence because it was constructed with cut-off walls placed into a hard pan, impermeable layer. The EV-2 pond appears to be working properly, with no indication of adverse groundwater influence. The system has been designed to prevent adverse impacts to the hydro-period of remaining wetlands. The wetlands are hydrated through groundwater flow. The groundwater will still migrate to the wetlands as it did in the pre-development condition. The cut-off walls and liners in the ponds will prevent draw-down of groundwater from the wetlands. No septic tanks are planned for the project. The system is designed based on generally accepted engineering practices and should be able to function as designed. The pumps are three inch pumps that can handle solids up to two and one-half inches in diameter. Yard grates have one-inch slots that will prevent anything larger than one inch diameter from entering the system. Additionally, solids would accumulate in the sump areas. Finally, even if there were a power outage, the system can store the full treatment volume, without discharging, until power is restored. Flood Plain Consideration The 100-year flood elevation for the EV-1 site is 7.0 feet NGVD. The finish flood elevation of the houses will be 8.0 feet. The streets and roadways have been designed to be flood free in accordance with the St. Johns County criteria relating to flooding. The 10-year flood elevation for the EV-1 site is 4.1 feet NGVD. The project will result in filling 2,691 cubic feet of fill in areas below the 4.1-foot NGVD elevation which will include 2,456 cubic feet for "Hickory Hill" and 235 cubic feet for "North River." Thus, 2,691 feet of water will displaced in the 10-year floodplain of the Tolomato River as a result of the EV-1 project. This fill will result in a rise in water elevation in the Tolomato River of 0.0002 feet, which is less than the thickness of the single sheet of paper and is statistically insignificant. If other applicants were to impact the 10-year floodplain to the same extent, there would be no adverse cumulative impact in the flood storage capability of the floodplain. The Tolomato River/intracoastal waterway does not function as a floodway because it is more influenced by wind and tide than by stormwater runoff. Therefore, the project will not cause a net reduction in the flood conveyance capabilities of a floodway. Surface Water Each roadway and master driveway is provided with culverts to ensure redundant, multiple paths for water flow. For this reason, the wetland fill will not significantly impact the flow of water. These redundant connections also ensure that the water velocities are low, reducing the likelihood of erosion. In order to ensure that erosion will not occur, surface water velocities will be less than two feet per second and steep slopes (greater than two percent) will be sodded. The project does not impound water other than for temporary detention purposes. The project does not divert water to another hydrologic water basin or water course. Water Quality The Tolomato River and Marshall Creek, its tributary, are classified as Class II water bodies pursuant to Florida Administrative Code Rule 62-302.400. The designated use for Class II water is for shellfish harvesting. The Tolomato River is the receiving water for the EV-1 project. The Marshall Creek and Tolomato River Class II waters do not meet the applicable Class II water quality standards for total fecal coliform bacteria and for dissolved oxygen (DO). Water sampling indicates that sometimes the regulatory parameters for fecal coliform and for DO are exceeded in the natural occurring waters of Marshall Creek and the Tolomato River. The EV-2 pond has a large surface area and the top of the water column will be the most well-oxygenated due to contact with the atmosphere. Any water discharging from the pond will come from the surface of the pond which is the water containing the highest oxygen content in the entire water column of the pond. Thus, discharges from the EV-2 pond will not violate water quality standards for DO and the construction and operation of the project will actually improve the water quality in the receiving waters with respect to the dissolved oxygen parameter. Bacteria such as fecal coliform, generally have a life span of a few hours to a few days. The EV-2 pond will have a detention time, for water deposited therein, of approximately 190 days. This lengthy residence time will provide an ample opportunity for die-off of any coliform bacteria in the water column before the water is discharged from the pond. Additionally, there will be substantial dilution in the pond caused by the large volume of the pond. No new sources of coliform bacteria such as septic tanks are proposed as part of the EV-1 project. The fecal coliform discharge from the pond will thus be very low in value and will lead to a net improvement in the water quality of the receiving water-body. In fact, since the commencement of construction on the Marshall Creek DRI phases, a substantial and statistically significant decrease in fecal coliform levels has been observed in the main channel of Marshall Creek. The applicant has provided a detailed erosion control plan for the construction phase of the EV-1 project. The plan requires the use of best erosion and sediment control practices. In any location that will have slopes exceeding a two percent gradient, sodding will be provided adjacent to roadways or embankments, thereby preventing erosion. The EV-1 project design is based on generally accepted engineering practices and it will be able to function and operate as designed. The liner and cut-off wall components of the pond portions of the project are proven technology and are typical on such project sites which are characterized by high groundwater table and proximity to wetlands. The pump stations component of the project design is proven technology and is not unusual in such a design situation. The pump stations have been designed according to the stringent specifications provided for wastewater lift station pumps in sewer systems which operate with more frequency and duration of running times and therefore, more stressful service, than will be required for this system. Once constructed, the surface water management system will be operated and maintained by the applicant, which is a community development district. An easement for access in, on, over and upon the property, necessary for the purpose of access and maintenance of the EV-1 surface water management system, has been reserved to the community development district and will be a permanent covenant running with the title to the lots in the project area. The portions of the river and Marshall Creek adjacent to the project have been classified by the Department of Environmental Protection as conditionally restrictive for shellfish harvesting because of fecal coliform bacterial levels, which often exceed state water quality standards for that parameter. The boundary of the conditional shellfish harvesting area is the mean high water elevation. The EV-1 project site is located above the mean high water elevation. None of the wetland areas within the project site are able to support shellfish due to the characteristics of the wetlands and the lack of daily inundation of the high marsh portion of the wetlands. No shellfish have been observed on the EV-1 site. The EV-1 project will not result in a change in the classification of the conditionally restricted shellfish harvesting area. The project will not negatively affect Class II waters and the design of the system and the proposed erosion controls will prevent significant water quality harm to the immediate project area and adjacent areas. The discharge from the project will not change the salinity regime or temperatures prevailing in the project area and adjacent areas. Wetland Impact The 23.83-acre site contains five vegetative communities that include pine, flatwood, uplands, temperate hardwood uplands, wetland coniferous forest, wetland mixed forest and salt marsh. Several trail roads that were used for site access and forestry activities traverse the site. The project contains 0.82 acres of wetlands. The wetland communities are typical and are not considered unique. Most of the uplands on the main portion of the site exhibit the typical characteristics of a pine flatwood community. Some of the road-crossing areas within the EV-1 boundary are wetland pine flatwoods; these areas are dominated by pines and a canopy, but are still considered wetlands. There is also a very small area of high marsh vegetative community within the EV-1 boundary. Most of the site, both wetlands and uplands, has been logged in the past. The wetlands are functional; however, the prior logging operations have reduced the overall wildlife value of the site, including that of the wetlands, due to the absence of mature trees. All of the wetlands on the EV-1 site are hydrologically connected to and drain to the Marshall Creek and Tolomato River systems. The wetlands on the site are adjacent to an ecologically, important watershed. To the east of the EV-1 site, the Tolomato River and Marshall Creek are part of the Guana Marsh Aquatic Preserve. The Guana River State Park and Wildlife Management Area is also to the east of the EV-1 site. All the wetlands and uplands on the EV-1 site are located above the elevation of the mean high water line and therefore are outside the limit of the referenced Aquatic Preserve and Outstanding Florida Water (OFW). Direct Wetland Impact Within the site boundary there will be a total of 0.82 acres of wetland impacts in seven areas. MCCDD proposes to fill 0.75 acres of the wetlands to construct roads to provide access to the developed uplands and selectively clear 0.07 acres of the mixed forested wetlands to construct three pile-supported pedestrian boardwalks. The fill impacts include 0.29 acres within the mixed forested wetlands, 0.32 acres within the coniferous wetlands, and 0.14 acres within the high salt marsh area. The direct impacts to wetlands and other surface waters from the proposed project are located above the mean high water line of Marshall Creek and the Tolomato River. The first impact area is a 0.25-acre impact for a road crossing from the EV-2 parcel on to the EV-1 site. 0.14 acres of the 0.25 acres of impact will be to an upper salt marsh community and 0.11 acres of impact is to a mixed forested wetland. This impact is positioned to the south of an existing trail road. The trail road has culverts beneath it so there has been no alteration to the hydrology of the wetland as a result of the trail road. This area contains black needle rush and spartina (smooth cord grass). The black needle rush portion of this area may provide some foraging for Marsh Wrens, Clapper Rails and mammals such as raccoons and marsh rabbits. The fresh-water forested portion of this area, which contains red maple and sweet gum, may provide foraging and roosting and may also be used by amphibians and song birds. Wading birds would not likely use this area because the needle rush is very sharp- pointed and high and will not provide an opportunity for these types of birds to forge and move down into the substrate to feed. The wading birds also would be able to flush very quickly in this area and their predators would likely hide in this area. The second impact area is a 0.25-acre impact to a pine flatwoods wetland community and will be used for a road crossing. It is in a saturated condition most of the time. The species that utilize this area are typically marsh rabbits, possums, and raccoons. The third impact area is a 0.18-acre impact to a mixed forested wetlands for a roadway crossing on the south end of the project. The impact is positioned within the area of an existing trail road. The trail road has culverts beneath it, so there will be no alteration to the hydrology of the wetland as a result of the road. This area is characterized by red maple, sweet gum and some cabbage palm. There will be marsh rabbits, raccoons, possums, some frogs, probably southern leopard frogs and green frogs in this area. Wading birds would not likely use this area due to the same reasons mentioned above. The fourth impact area is a 0.07-acre impact for a driveway for access to Lot two. This area is a mixed forested wetland area, having similar wildlife species as impact areas three and seven. The fifth impact area is a 0.02-acre clearing impact for a small residential boardwalk for the owner of Lot six to access the uplands in the back of the lot. The proposed boardwalk will be completely pile-supported and will be constructed five feet above the existing grade. This area is a mixed forested wetland area, having similar species as impact areas three and seven. Wading birds would also not likely use this area for the same reasons delineated above as to the other areas. The sixth impact area is also a 0.02-acre clearing impact similar to impact area five. The proposed board walk would be located on Lot five and be completely pile-supported five feet above the existing grade. This area is a mixed forested wetland area similar to impact area five. Deer will also use this area as well as the rest of the EV-1 site. Wading birds will probably not use this area due to the same reasons mentioned above. The seventh impact area is a 0.03-acre impact for two sections of a public boardwalk (previously permitted) for the Palencia Development. The proposed boardwalk will be completely pile-supported, five feet above the existing grade. This is a pine-dominated area with similar wildlife species to impact area two. All these wetlands are moderate quality wetlands. The peripheral edges of the wetlands will be saturated during most of the year. Some of the interior areas that extend outside the EV-1 site will be seasonally inundated. Secondary Impacts The applicant is addressing secondary impacts by proposing 8.13 acres of 25-foot wide (or greater) upland buffers and by replacing culverts at the roadway crossings to allow for wildlife crossing and to maintain a hydrologic connection. Mitigation by wetland preservation is proposed for those areas that cannot accommodate upland buffers (i.e., the proposed impact areas). Under the first part of the secondary impact test MCCDD must provide reasonably assurance that the secondary impact from construction, alteration and intended or reasonably expected uses of the project will not adversely affect the functions of adjacent wetlands or other surface waters. With the exception of wetland areas adjacent to the road crossings, MCCDD proposes to place upland buffers around the wetlands where those potential secondary impacts could occur. The buffers are primarily pine flatwoods (pine dominated with some hardwood). These buffers encompass more area than the lots on the EV-1 site. The upland buffers would extend around the perimeter of the project and would be a minimum of 25 feet and a maximum of 50 feet wide, with some areas actually exceeding 50 feet in width. The buffers along the Marshall Creek interface and the Tolomato River interface will be 50 feet and the buffers that do not front the tidal marshes (in effect along the interior) will be 25 feet. These upland buffers will be protected with a conservation easement. No activities, including trimming or placement of structures are allowed to occur within the 25-foot upland buffers. These restrictions ensure that an adequate buffer will remain between the wetlands and the developed portion of the property to address secondary impacts. The restriction placed on the 25-foot buffers is adequate to prevent adverse secondary impacts to the habitat value of the off-site wetlands. No types of structures are permitted within the 50- foot buffers. However, hand-trimming will be allowed within half of that length along the lot interface of the wetland. Within that 50 percent area, trimming below three-feet or above 25-feet is prohibited. Trimming of branches that are three inches or less in diameter is also prohibited. Lot owners will be permitted to remove dead material from the trimming area. The 50-foot buffers will prevent secondary impacts because there will still be a three-foot high scrub area and the 50 foot distance provides a good separation between the marsh which will prevent the wading birds, the species of primary concern here, from flushing (being frightened away). None of the wetland area adjacent to uplands are used by listed species for nesting, denning, or critically important feeding habitat. Species observed in the vicinity of Marshall Creek or the adjacent Tolomato River wetland aquatic system include eagle, least tern, brown pelican, and wading birds such as the woodstork, tri-color blue heron, and snowy egrets. Wading Birds will typically nest over open water or on a island surrounded by water. Given the buffers proposed by MCCDD, the ability of listed species to forage in the adjacent wetlands will not be affected by upland activities on the EV-1 site. The adjacent wetlands are not used for denning by listed species. Under the second part of the secondary impact test, MCCDD must provide reasonable assurance that the construction, alteration, and intended or reasonably expected uses of the system will not adversely affect the ecological value of the uplands to aquatic or wetland dependent species for enabling nesting or denning by these species. There are no areas on the EV-1 site that are suitable for nesting or denning by threatened or endangered species and no areas on the EV-1 site that are suitable for nesting or denning by aquatic and wetland dependent species. After conducting on-site reviews of the area, contacting the U.S. Fish and Wildlife Service and the Florida Wildlife Commission and reviewing literature and maps, Mr. Esser established that the aquatic and wetland listed species are not nesting or denning in the project area. There is a nest located on uplands on the first island east of the project site, which was observed on October 29, 2002. The nest has been monitored informally some ten times by the applicants, consultants and several times by personnel of the District. The nest was last inspected on October 14, 2003. No feathers were observed in the nest at that time. It is not currently being used and no activity in it has been observed. Based on the absence of fish bones and based upon the size of the sticks used in the nest (one-half inch) and the configuration of the tree (crotch of the tree steeply angled) it is very unlikely that the nest is that of an American Bald Eagle. It is more likely the nest of a red-tailed hawk. Historical and Archeological Resources Under the third part of the secondary impact test and as part of the public interest test, any other relevant activities that are very closely linked and causally related to any proposed dredging or filling which will cause impacts to significant historical or archeological resources must be considered. When making a determination with regard to this part of the secondary impact test the District is required by rule to consult the Division of Historical and Archeological Resources (the Division) within the Department of State. The District received information from the Division and from the applicant regarding the classification of significant historical and archeological resources. In response to the District's consultation with the Division, the Division indicated that there would be no adverse impacts from this project to significant historical or archeological resources. As part of the Marshall Creek DRI application, a Phase I archeological survey was conducted for the entire area of the DRI, including the EV-1 project area. The Phase I survey of the Marshall Creek DRI area revealed nine archeological sites. At the end of the Phase I survey, five of the nine sites were recommended to be potentially eligible for the National Register of Historical places and additional work was recommended to be done on those five sites, according to Dr. Ann Stokes, the archeologist who performed the Phase I survey and other archeological investigation relevant to this proceeding. One of the sites considered eligible for listing on the National Register of Historic Places was site 8SJ3146. Site 8SJ3146 was the only site found in the area near the EV-1 project site. The majority of the EV-1 project site lies to the east of this archeological site. The entry road leading into EV-1 crosses the very southeastern edge or corner of the 8SJ3146 archeological site. Shovel tests for archeological remains or artifacts were conducted across the remainder of the EV-1 property and were negative. Ceramic shards were found in one of the shovel tests (shovel test number 380), but it was determined by Dr. Stokes that that ceramic material (pottery) had been within some type of fill that was brought into the site and the ceramics were not artifacts native to that site. Therefore, it was not considered a site or an occurrence. There was no evidence of any human remains in any of the shovel test units and there was nothing to lead Dr. Stokes to believe that there were any individuals buried in that area. (EV-1) Because a determination was made that 8SJ3146 was a potentially significant site, a "Phase II assessment" was conducted for the site. During the Phase II assessment five tests units were established on the site to recover additional information about the site and assess its significance. The test unit locations (excavations) were chosen either to be next to an area where there were a lot of artifacts recovered or where an interesting type of artifact had been recovered. Test units one through four contained very few or no artifacts. Test unit five however, yielded faunal bones (animal remains), pottery and a post mold (post molds are evidence of support posts for ancient structures). After the Phase II assessment was conducted, site 8SJ3146 was considered to be significant, but the only part of the site that had any of the data classes (artifact related) that made it a significant site was in the area of the very southwest portion of 8SJ3146, surrounding test unit five. Dr. Stokes recommended that the area surrounding test unit five in the very southwestern portion of 8SJ3146 be preserved and that the remainder of the site would not require any preservation because the preservation of the southwestern portion of the site was the only preservation area which would be significant archeologically and its preservation would be adequate mitigation. That southwestern portion of the site, surrounding unit five, is not on the EV-1 site. Dr. Stokes recommended to the applicant and to the Division that a cultural resource management plan be adopted for the site and such a plan was implemented. A Phase I cultural resource survey was also conducted on the reminder of the EV-1 site, not lying within the boundaries of 8SJ3146. That survey involved shovel tests across the area of the EV-1 project area and in the course of which no evidence of archeological sites was found. Those investigations were also reported to the Division in accordance with law. The preservation plan for site 8SJ3146, as to preservation of the southwest corner, is now called an archeological park. That designation was shown to be adequate mitigation for this site. The preservation area is twice as large as the area originally recommended by Dr. Stokes to be preserved; test unit five is within that preservation area. Dr. Stokes's testimony and evidence are not refuted by any persuasive countervailing evidence and are accepted. They demonstrate that the construction and operation of the EV-1 project will not adversely affect any significant archeological or historical resources. This is because any effects to site 8SJ3146 are mitigated by the adoption of the preservation plan preserving the southwest portion of that archeological site. Under the fourth part of the secondary impact test, the applicant must demonstrate that certain additional activities and future phases of a project will not result in adverse impacts to the functions of wetlands or result in water quality violations. MCCDD has demonstrated that any future phase or expansion of the project can be designed in accordance with the District's rule criteria. Mitigation of Adverse Impacts The permit applicant has proposed mitigation to offset adverse impacts to wetland functions as part of its ERP application. The proposed mitigation consists of 0.05 acres of wetlands restoration, 12.07 acres of wetland preservation (including 6.47 acres of mixed forested wetlands and 5.60 acres of salt marsh), 10.49 acres of upland preservation (which includes buffers and additional upland areas) and 0.09 acres of salt marsh creation. The mitigation for the EV-1 project will occur on-site and off-site; 10.49 acres of upland buffer are being committed to the project. The upland buffers are on-site; the rest of the mitigation is off-site and is adjacent to EV-1. There will be 5.6 acres of salt marsh preservation and 6.47 acres of forested wetland preservation. All of the mitigation is on land lying above the mean high water elevation and is outside the aquatic preserve and the OFW. The salt marsh restoration will occur by taking out an existing trail road that is in the northeast section of the site and the salt marsh creation site is proposed at the tip of lot number one. The preservation of wetlands provides mitigation value because it provides perpetual protection, ensuring that development will not occur in those areas, as well as preventing agricultural activities, logging and other relatively unregulated activities from occurring there. This will allow the conserved lands to mature and to provide more forage and habitat for wildlife that would use those areas. The functions that are currently being provided by the wetlands to be impacted will be replaced and exceeded in function by the proposed mitigation. Additionally, MCCDD did not propose any impacts on site that could not be offset by mitigation. The EV-1 project will not adversely affect the abundance and diversity and habitat of fish and wildlife. The mitigation for the proposed project is also located within the same drainage basin as the area of wetlands to be adversely impacted. MCCDD has proposed mitigation that implements all or part of a plan of regional ecological value and the proposed mitigation will provide greater long-term ecological value than the wetlands to be impacted. The plan of regional ecological value consists of the land identified in the DRI as well as the lands that have been permitted as mitigation up to date and the proposed EV-1 mitigation lands. The plan includes lands that have been added to the plan since the approval of the Marshall Creek DRI. The mitigation proposed for the impact to wetlands and other surface waters associated with the project is contiguous with the Guana River Marsh Aquatic Preserve, with previously preserved wetlands and upland islands and with Marshall Creek. When implemented the mitigation plan will create wetlands and preserve wetlands and uplands with functions similar to the impacted wetlands and those wetlands will be connected through wetland and upland preservation to the Guana River Marsh Aquatic Preserve. Corridors and preservation areas important for wildlife movement throughout the whole Palencia site have been set aside. As development progresses towards the eastern portion of the Marshall Creek site, it is important to add preservation areas to the whole larger plan. The lands proposed to be added as mitigation for the EV-1 project will add to the value of the previously preserved lands from other phases of the DRI and development by helping to maintain travel corridors and forage areas for wildlife, to maintain water quality in the adjacent marsh and to maintain fish and wildlife benefits of the aquatic preserve. MCCDD has provided more mitigation than is typically required by the District for such types of impact. The upland preservation ratios for example range from about three-to-one to twenty-to-one. MCCDD is providing upland preservation at a near twenty-to-one ratio. Salt marsh preservation ratios are typically required to be sixty to one and MCCDD is providing mitigation at twice that ratio. Concerning fresh-water forested preservation, the District usually requires mitigation at a twenty to twenty-five-to-one ratio and the applicant is proposing a thirty to one preservation ratio. Additional mitigation will be provided beyond what is required to mitigate the adverse impacts for each type of impact anticipated. Although proposing more mitigation may in some instances not provide greater long-term ecological value than the wetlands to be adversely affected, the mitigation proposed by MCCDD will provide greater long-term ecological value. The Petitioners contend that a chance in circumstances has occurred which would adversely affect the mitigation plan as a plan of regional ecological value. They claim its efficacy will be reduced because of a proposed development to a tract of land known as the Ball Tract which would, in the Petitioners' view, sever connection between the Marshall Creek site and the 22,000-acre Cummer Trust Tract also known as "Twelve mile swamp." Although a permit application has been submitted to the Florida Wildlife Commission for the Ball Tract property, located northwest of Marshall Creek and across U.S. Highway 1 from Marshall Creek and the EV-1 site, no permit has been issued by the District for that project. Even if there were impacts proposed to wetlands and other surface waters as part of any development on the Ball Tract, mitigation would still be required for those impacts, so any opinion about whether the connection would be severed between the project site, the Marshall Creek site and the Cummer Trust Tract is speculative. The Petitioners also sought to establish changed circumstances in terms of reduced effectiveness of the plan as a plan of regional ecological value because, in their opinion, Map H, the master plan, in the Marshall Creek development order plan, shows the EV-1 project area as being located in a preservation area. However, Map H of the Marshall Creek DRI actually shows the designation VP for "Village Parcel" on the EV-1 site and shows adjacent wetland preservation areas. Although Map H shows a preservation area adjacent to the EV-1 parcel, the Petitioners infer that EV-1 was not proposed for development. That is not the case. Map H contains a note that the preservation areas (as opposed to acreages) are shown as generalized areas and are subject to final design, road crossings and final wetland surveys before they were exactly delineated. Therefore, in the DRI plan, the EV-1 area was not actually designated a preservation area. Surface Water Diversion and Wetland Draw-Down Water will not be diverted to another basin or water course as a result of the EV-1 project. Water captured by the treatment system and discharged from the EV-2 pond, will flow back through wetlands that meander through the project site. The EV-1 project will not result in significant diversion of surface waters. The project will also not result in a draw-down of groundwater that will extend into adjacent wetlands. Each of the storage ponds on lots 1, 3, and 7 and between lots 9 and 10 has been designed to include cut-off walls around the perimeter of the ponds and the storage pond on lot 7 will be completely lined. The cut-off walls will be installed in a soil strata that has very low permeability. The cut-off walls and liner will restrict the movement of groundwater from the wetlands into the storage ponds. As a result, the zone of influence of each storage pond will not extend far enough to intercept with the adjacent wetlands. The Public Interest Test The public interest test has seven criteria, with each criteria having equal weight. The public interest test applies to the parts of the project that are in, on or over wetlands, and those parts must not be contrary to the public interest unless they are located in, on or over an OFW or may significantly degrade an OFW; then the project must be clearly in the public interest. It is a balancing test. The EV-1 project, however, is not located in an OFW. The Public Health Safety and Welfare Criteria The parts of the project located in, on and over wetlands will not adversely affect the public health, safety or welfare. These parts of the project will not cause any adverse impact on flood stages or flood plains and discharges from the system will not harm shell fishing waters. This factor is thus considered neutral. Conservation of Fish, Wildlife or Their Habitat The mitigation from this project will offset any adverse impacts to fish wildlife or their habitat. Therefore this factor is considered neutral as well. Fishing, Recreational Value and Marine Productivity There is no recreational activity or fish nursery areas within the project limits and the project will not change the temperature of the aquatic regime. None of the impacts associated with the EV-1 site are within the mean high water line of the marine aquatic regime. The activities are not going to interact with the tidal regime and they cause negligible impacts. Concerning marine productivity, the wetland impacts are landward of the marine system; therefore, impact on marine productivity is not applicable. Thus this factor is considered neutral. Temporary or Permanent Nature The project will be of a permanent nature. Even though the project is permanent, this factor is considered neutral because the mitigation proposed will offset any permanent adverse impact. Navigation and the Flow of Water The parts of the project located in, on and over wetlands will not adversely affect navigation. These parts will also not impound or divert water and therefore will not adversely affect the flow of water. The project has been designed to minimize and reduce erosion. Best management practices will be implemented, and therefore, the project will not cause harmful erosion. Thus this factor is also considered neutral. Current Condition and Relative Value of Functions Being Performed The current condition and relative value of the functions being performed by the areas affected by the proposed activity, wetlands areas, will not be harmed. This is because any adverse impacts to the wetlands involved will be more than offset by the mitigation proposed to be effected. Therefore, there may well be a net gain in the relative value and functions being performed by the natural areas and the mitigation areas combined. Thus this factor is neutral. Works of the District The proposed project will not cause any adverse impact to a work of the District established in accordance with Section 373.086, Florida Statutes. Shoaling The construction and operation of the proposed project to the extent it is located in, on or over wetlands or other surface waters will not cause any harmful shoaling.
Recommendation Having considered the foregoing Findings of Fact, Conclusions of Law, the evidence of record, the candor and demeanor of the witnesses, and the pleadings and arguments of the parties, it is, therefore, RECOMMENDED that a Final Order be entered by the St. Johns River Water Management District granting MCCDD's application for an individual environmental resource permit with the conditions set forth in the technical staff report dated September 24, 2003, in evidence as St. John's River Water Management District's Exhibit 3. DONE AND ENTERED this 9th day of February, 2004, in Tallahassee, Leon County, Florida. S P. MICHAEL RUFF Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with Clerk of the Division of Administrative Hearings this 9th day of February, 2004. COPIES FURNISHED: Deborah J. Andrews, Esquire 11 North Roscoe Boulevard Ponte Vedra Beach, Florida 32082 Veronika Thiebach, Esquire St. Johns River Water Management District Post Office Box 1429 Palatka, Florida 32178-1429 Marcia Parker Tjoflat, Esquire Pappas, Metcalf, Jenks & Miller, P.A. 245 Riverside Avenue, Suite 400 Jacksonville, Florida 32202-4327 Stephen D. Busey, Esquire Allan E. Wulbern, Esquire Smith, Hulsey & Busey 225 Water Street, Suite 1800 Jacksonville, Florida 32202 Kirby Green, Executive Director St. Johns River Water Management District Post Office Box 1429 Palatka, Florida 32178-1429
The Issue The issues are whether IMC Phosphates Company is entitled to an environmental resource permit for phosphate mining and reclamation on the Ona-Ft. Green extension tract, approval of its conceptual reclamation plan for the Ona-Ft. Green extension tract, and modification of its existing wetland resource permit for the Ft. Green Mine to reconfigure clay settling areas, relocate mitigation wetlands, and extend the reclamation schedule.
Findings Of Fact Parties, Phosphate Mining, and Physiography Respondent IMC Phosphates Company, a Delaware general partnership authorized to do business in Florida (IMC), has applied to Respondent Department of Environmental Protection (DEP, which shall include predecessor agencies) for an environmental resource permit (ERP) to mine phosphate rock at the Ona-Ft. Green extension tract (OFG), approval of a conceptual reclamation plan (CRP) to reclaim the mined land at OFG, and modification of a previously issued wetland resource permit (WRP) to relocate and shrink clay-settling areas (CSAs), relocate mitigation wetlands, and extend the reclamation schedule at the Ft. Green Mine, which is an existing mine that is immediately west and north of OFG. Except for the submerged bottom of Horse Creek, which is sovereign submerged land, IMC owns all of the land on which OFG will be located, except for a 1.8-acre parcel owned by Valerie Roberts in Section 16, which is described below with the other sections forming OFG. IMC is negotiating with Ms. Roberts to purchase her land, and she has authorized IMC to pursue mining permits for the entire parcel, including her land. IMC Global, Inc., owns 80 percent of IMC. IMC Phosphates MP Inc., a Delaware corporation, is the managing general partner of IMC. As a successor to International Mining and Chemical Corporation, IMC has been in business for over 100 years. IMC is the largest producer of phosphate in the world. References in this Recommended Order to phosphate mining companies include all forms of business organizations. At present, IMC is operating four phosphate mines in Florida. The largest is the Four Corners Mine, which extends into Hillsborough, Polk, Manatee, and Hardee counties and three river basins. IMC also operates the Hopewell Mine in Hillsborough County, the Kingsford Mine in Hillsborough and Polk counties, and the Ft. Green Mine. Petitioner Charlotte County is located south of Sarasota and DeSoto counties and west of Glades County. The majority of Charlotte Harbor lies within Charlotte County. Charlotte Harbor is a tidal estuary at the mouths of the Peace and Myakka rivers. An Outstanding Florida Water and an Aquatic Preserve, Charlotte Harbor provides critical habitat for a variety of species. Charlotte Harbor is now an estuary of national significance under the U.S. National Estuary Program. Directly or indirectly, Charlotte Harbor supports 124,000 jobs and generates $6.8 billion in sales annually. To protect this unique natural resource, Charlotte County has adopted a local government comprehensive plan directing residential densities away from Charlotte Harbor. Charlotte County has also expended over $100 million in sanitary sewer capital expenditures for, among other things, the protection of Charlotte Harbor, such as by replacing private residential septic tanks with central sewer. Charlotte County's opposition to phosphate mining and reclamation in the Peace River basin is based on concerns about reduced river flows, reduced abundance and diversity of fish species, the loss of wetlands and first-order streams, and degraded water quality. Petitioner Peace River/Manasota Regional Water Supply Authority (Authority) is an agency authorized by Section 373.196(2), Florida Statutes, and created by interlocal agreement among Charlotte, Sarasota, DeSoto, and Manatee counties. The purpose of the Authority is to supply potable water to several suppliers in southwest Florida. Relying exclusively on the Peace River as its source of raw water, the Authority withdraws water from the Peace River two miles downstream of the point that Horse Creek empties into the Peace River. This point is about midway between Arcadia and Charlotte Harbor. As discussed below, the Authority's permit to withdraw water from the Peace River is dependent upon flows at a point upstream of the confluence of Horse Creek and the Peace River. The Authority's current water use permit expires in 2016. From its water treatment plant, which is located near the withdrawal point, the Authority pumps finished water to Charlotte, Sarasota, and DeSoto counties and the City of North Port. Approximately 250,000 persons rely on these suppliers, and, thus, the Authority, for their potable water. At present, the Authority is obligated to supply 18 million gallons per day (mgd), but anticipates demand to increase to 32 mgd by 2015. Petitioner Sarasota County (Sarasota County) owns and operates a water utility system, which currently supplies 24 mgd of potable water to 125,000 persons. Sarasota County obtains potable water from its wellfields, Manatee County, and the Authority, from which it may take up to 3.6 mgd. By 2017, Sarasota County plans to take 13.7 mgd of potable water from the Authority, partly to offset anticipated reductions in the amount of potable water presently being supplied by Manatee County. By 2017, the Authority will supply over half of Sarasota County's potable water. Sarasota County also shares Charlotte County's concerns about the overall environmental integrity of Charlotte Harbor, a small part of which is in Sarasota County. Intervenor Lee County (Lee County) is immediately south of Charlotte County. Nearly half of Charlotte Harbor lies within Lee County. Tourism produced an estimated $1.8 billion to Lee County's economy in 2002. Tourists are attracted to Lee County in part due to the high quality of Charlotte Harbor and its unique chain of barrier islands, passes, sounds, and bays that are integral to local fishing and boating. Lee County shares Charlotte County's concerns about the overall environmental integrity of Charlotte Harbor. Lee County is concerned about, among other things, degraded water quality from the discharge of turbid water, increased pollutant loads to the Peace River and Charlotte Harbor, adversely affected freshwater flows in the Peace River, and the consequences of the phosphate mining industry's inability to restore secondary tributaries, which provide base flow and environmental benefits to Charlotte Harbor. Petitioner Alan R. Behrens (Behrens) resides in Wimauma, Florida, which is in Hillsborough County. He has owned two five-acre tracts along Horse Creek since 1985 and owns a 2.5-acre lot in DeSoto County that fronts Horse Creek for 100-200 feet. The Horse Creek property is 10-15 miles downstream from OFG. Behrens has canoed the entire main stem of Horse Creek from the Peace River to OFG. On May 9, 2004, Behrens canoed up Stream 4w, which is a tributary of Horse Creek on OFG and is described in detail below. Behrens is a founder of Petitioner DeSoto Citizens Against Pollution, Inc. (DCAP), which was incorporated in 1990 as a Florida not-for-profit corporation and has operated in that status continuously since that time. DCAP's purpose is to protect fish, wildlife, and air and water resources; promote public health and safety; increase public awareness of potential environmental hazards; and discourage activities that may be adverse to public health or the environment. DCAP has 52 members, of whom 27 reside in Hardee County, 23 reside in DeSoto County, and two reside in Sarasota County. A substantial number of DCAP's members use Horse Creek for swimming, boating, canoeing, and fossil hunting. At least nine DCAP members own property abutting Horse Creek. Behrens and many DCAP members use wells on their property for potable water. Behrens and DCAP members are concerned that the clay- settling areas described below will increase flooding, the project will adverse affect the timing and volume of the flow and degrade the water quality of Horse Creek, the project will destroy wildlife habitat that--even if reclaimed--will be lost for many years, and the project will cause spills that will destroy fish and wildlife and adversely affect the ability of Behrens and DCAP to enjoy Horse Creek. OFG is in northwest Hardee County, about one-half mile east of the Manatee County line. OFG is about six miles south- southeast of the Four Corners, where Hardee, Manatee, Polk, and Hillsborough counties meet. OFG is about 35 miles east of Bradenton, 12 miles west of Wauchula, several miles south of State Road 62, and 2000 feet north of State Road 64. OFG represents the southernmost extent of phosphate mining in the Peace River basin to date. A nonrenewable resource for which no synthetic substitutes exist, phosphate is an essential nutrient and a major component of manufactured fertilizer. Less important uses of phosphate are for animal feed, soft drinks, and cosmetics. Mining phosphate rock and processing it into phosphoric acid or phosphorus make possible high-yield agriculture, which, by producing more food crop on less land, may reduce worldwide pressure to convert native habitat to improved agricultural land uses. Phosphate is available in limited quantities. Three- quarters of the recoverable phosphate rock in the United States is found in Florida, mostly in discrete deposits ranging from north-central Florida to Charlotte Harbor. Ten to fifteen million years ago, when peninsular Florida was submerged marine bottom, dead marine organisms accumulated as bone and shell on the ocean floor. These accumulations formed the Bone Valley Formation, which, as the seas withdrew and the peninsula emerged, occupies the lower part of the surficial aquifer at the site of OFG. Briefly, the main elements of the proposed activities in these cases, roughly in the order in which they will take place, are relocating wildlife; constructing a ditch and berm system around the area to be mined; removing topsoil from certain donor areas; removing the overburden and depositing it in rows of spoil within the mine cut; removing the underlying phosphate matrix and slurrying it to a nearby beneficiation plant at the Ft. Green Mine for processing to separate the phosphate rock from the sand and clay tailings; slurrying the clay tailings from the beneficiation plant to two CSAs at the southern end of the Ft. Green Mine; slurrying the sand tailings from the beneficiation plant back to the mine cut to backfill the excavation; applying topsoil to certain areas or green manuring areas for which topsoil is unavailable; applying muck to certain areas; contouring the reclaimed land to replicate pre-mining topography; analyzing the post-reclamation hydrology; reclaiming wetlands, streams, and uplands on the reclaimed landscape of OFG; maintaining and monitoring the reclaimed wetlands, streams, and uplands until DEP releases IMC from its ongoing reclamation obligations; correcting any problems in reclaimed areas; and removing the ditch and berm system and reconnecting the reclaimed mined area to the areas adjoining it. In the Findings of Fact, this Recommended Order uses "reclaim" to describe the process by which, post-mining, IMC and its reclamation scientists will construct wetlands, other surface waters, and wetlands at OFG. Likewise, in the Findings of Fact, this Recommended Order uses reclamation and mitigation interchangeably. In the Conclusions of Law, this Recommended Order discusses distinctions in these terms. IMC plans to use multiple draglines to dig a series of long, linear trenches in the mined areas of OFG. Each dragline will first remove overburden and place it in piles parallel to the trench being excavated. After removing the overburden, each dragline will remove the phosphate matrix, which consists of phosphate rock, sand, and clay, and deposit it in shallow depressions. Adding water from the mine recirculation system to the phosphate matrix, IMC will slurry the phosphate matrix to the Ft. Green beneficiation plant, which is about 12 miles from OFG. At the beneficiation plant, the phosphate rock will be separated from the sand and clay tailings, again using water from the mine recirculation system. After recovering the phosphate rock, IMC will slurry the sand tailings, which do not retain water, from the Ft. Green beneficiation plant to OFG for backfilling into the mined trenches with the overburden. Not used in the reclamation at OFG, the clay tailings, which retain water for an extensive period of time, will be slurried to the CSAs O-1 and O-2 on the Ft. Green Mine. CSAs O- 1 and O-2 are the subject of the WRP, which is discussed below. The volume of the clay leaving the beneficiation plant is greater than the clay in situ, pre-mining, because the slurrying process has saturated the clay. The CSAs provide a place to store the saturated clay while it drains and decreases in volume. The clay-settling process takes a long time, extended by IMC's intention to fill the CSAs by stages to make the most efficient use of the areas designated for the settling of clay. By stage-filling the CSAs, IMC will initially install the clay to a considerable height, using an embankment of approximately 50-60 feet. The water that separates from the clay will then drain across the sloped CSA until it enters the mine recirculation system for reuse. The remaining clay will dry and consolidate. After refilling each CSA approximately three times over about ten years, IMC will allow the clay to settle and consolidate a final time. When the clay has consolidated sufficiently to support agricultural equipment, IMC will regrade the area, reduce the side slopes, and remove the embankments, leaving the CSAs at a finished elevation 20-25 feet above the surrounding grade. Given the ongoing nature of IMC's phosphate mining operations, it is likely that some sand and clay tailings from OFG will go elsewhere, rather than return to the OFG mine cuts and CSAs O-1 and O-2, and that some sand and clay tailings from non-OFG mining operations will go to the OFG mine cuts and CSAs O-1 and O-2. However, these facts are irrelevant to the issues raised in these cases, except for consideration of IMC's sand- tailings budget, which is discussed below. Phosphate mining and reclamation practices have changed dramatically in the past 40 years. Although mining operations and reclamation practices are discussed below in detail, one development in mining and one development in reclamation bear emphasis due to the resulting reductions in water losses to the drainage basin. As explained below, mining operations are dependent upon large volumes of water, which flow through the mine recirculation system. Before 1963, phosphate mining pumped roughly 3000 gallons of water for each ton of mined phosphate rock. By the mid-1970s through 1990, the industry had reduced its groundwater consumption to 1500 gallons per ton of mined rock. From 1991 to 1999, the industry again reduced its groundwater consumption from 1200 gallons per ton to 650 gallons per ton, partly by achieving a 97 percent rate of water- recycling in the mine recirculation system. During roughly the same period, phosphate reclamation activities have expanded considerably. Prior to July 1, 1975, reclamation of mined land was voluntary, encouraged only by the availability of state funds to offset reclamation costs. Today, post-mining reclamation is required by law. As a consequence, post-mining reclamation 30 years ago was relatively modest in scope and intensity. One important development in reclamation practices is the phosphate mining industry's transition from early reclamation techniques that relied on relatively inexpensive contouring of the overburden that remained in the mine cuts following the extraction of the phosphate ore. These reclamation practices--aptly called Land-and-Lakes reclamation-- yielded post-reclamation excavations, such as reclaimed lakes or deep marshes, that, compared to pre-mining conditions, retained considerable volumes of surface water. The resulting increase in surface water area, compared to pre-mining surface water area, meant substantial loss of water from the drainage basin due to increased evapotranspiration. More recent reclamation practices, such as those proposed for OFG, feature more extensive backfilling of the mine cuts with tailings to restore pre-mining topography. The result is that less water is lost to evapotranspiration by retention in newly created lakes and deep marshes and more is timely held and passed by the natural drainage conveyances through detention, attenuation, runoff, and base flow--eventually entering the main basin river in volumes, rates, and times (relative to storm events) comparable to pre-mining conditions. Located near the western divide of the Peace River basin, OFG is near a topographical high point marking the divides among five drainage basins. From north to south, the four other basins are drained by the Alafia River, Little Manatee River, Manatee River, and Myakka River. OFG is located toward the bottom of an escarpment where the Polk Uplands descends into the DeSoto Plain. OFG is located almost entirely within a portion of the Horse Creek basin or sub-basin within the Peace River basin. This Recommended Order shall refer to the drainage basins that form the larger Peace River basin as sub-basins. A small portion of the western edge of OFG is within the West Fork Horse Creek (West Fork) sub-basin, and a small portion of the eastern edge of OFG is within the Brushy Creek sub-basin. OFG is toward the upper end of the Horse Creek sub-basin. The West Fork and Brushy Creek sub-basins within OFG contain no streams or stream segments and only, between them, about a half dozen wetlands of one-half acre in size or greater. Obviously, as separate sub-basins, these two areas on OFG are relatively far from Horse Creek. West Fork joins Horse Creek a couple of hundred feet south of OFG and just north of State Road 64. Brushy Creek joins Horse Creek six miles southeast of OFG. Horse Creek joins the Peace River at Ft. Ogden, about 40 miles south of OFG and 15 miles northeast of the mouth of the Peace River at Charlotte Harbor. The Peace River basin comprises about 2350 square miles and extends from its headwater lakes in north Polk County to Charlotte Harbor. By comparison, the Horse Creek sub-basin comprises about 241 square miles, or roughly ten percent of the Peace River basin. At Charlotte Harbor, the average flow of the Peace River is about 1700 cubic feet per second (cfs). By comparison, Horse Creek, at its confluence with the Peace River, flows at an average rate of about 170 cfs--again ten percent of the average rate of flow of the Peace River. West Fork, at its confluence with Horse Creek, flows at an average rate of about 10 cfs. The largest tributary on OFG flows at an average rate of about 0.75 cfs. Forming a little south of Four Corners, Horse Creek is one of five major tributaries of the Peace River. An ecological backbone of this region of Florida, Horse Creek is the only long-term, reliable flowing water system between the Manatee River on the west and Peace River on the east. OFG occupies the upper reaches of Horse Creek. Horse Creek is in good condition, notwithstanding 100 years of nearby cattle ranching. Most of Horse Creek is Class III waters, although a segment near the Peace River is Class I waters. Horse Creek is a moderately incised stream at OFG, especially over its southern two-thirds running through the mine site. Over the little more than three miles that Horse Creek flows through OFG, the streambed drops from nearly 120 feet National Geodetic Vertical Datum (NGVD) at the north end to about 75 feet NGVD at the south end. Within OFG, the valley that Horse Creek occupies is also relatively well-defined. The northern half of the streambed of Horse Creek within OFG is mostly around 100 feet NGVD. The highest adjacent elevations on OFG are about 120 feet NGVD. At least partly for this reason, most of the tributary streams, except in the flat northern portion of OFG, are also well-incised. OFG extends about 4 1/2 miles north to south, and ranges from 2/3 to 2 1/2 miles from east to west, for a total area of about 6 1/2 square miles. Lying entirely within Township 34 South, Range 23 East, OFG, from its northernmost border, occupies three sections, which are, from north to south: Sections 4, 9, and 16. Immediately west of the southern half of Section 9, OFG occupies most of the southern half of Section 8. Immediately west of Section 16, OFG occupies Section 17, as well as, immediately south of Section 17, all of Section 20 and most of the northern half of Section 29. OFG also extends to parts of four other sections: Sections 10 and 15 east of Sections 9 and 16, respectively, and Sections 18 and 19, west of Sections 17 and 20, respectively. The existing surface waters and nearly all of the existing wetlands are on the two columns of sections running north and south: on the east, Sections 4, 9, and 16 and, on the west, Sections 17, 20, the south part of Section 8, and the north part of Section 29. The northernmost extent of OFG, which consists of Section 4 and the north half of Section 9, is known as the Panhandle. Horse Creek enters OFG at the southwest corner of the Panhandle, at a point midway along the west border of Section 9. The stream flows south through the approximate center of OFG for about 1 1/2 miles until it leaves OFG for a very short distance at the southwest corner of Section 16, as it crosses a corner of property owned by the Carlton-Smith family (Carlton cutout). Horse Creek re-enters OFG at the northeast corner of Section 20 and runs just inside the eastern border of Section 20 and the portion of Section 29 within OFG. Horse Creek leaves OFG near the midpoint of the east border of Section 29. Numerous tributary streams enter Horse Creek within OFG, from the east and west sides of the creek. IMC and DEP have assigned to each of these streams or stream segments a number, followed by a letter to indicate if the stream or stream segment enters Horse Creek from the east or west. To the west of Horse Creek, proceeding from south to north, the streams are 0w, 1w, 2w, 3w, 4w, 5w, 6w, 7w, 8w, and 9w. To the east of Horse Creek, proceeding from south to north, the streams are 12e, 11e, 10e, 5e, 9e, 4e, 8e, 7e, 6e, 2e, 3e, and the Stream 1e series, consisting of Streams (sometimes referred to as stream segments) 1ee, 1ed, 1ec, 1eb, and 1ef. All of the streams join Horse Creek on OFG except Stream 2e, which joins Horse Creek a few hundred feet upstream of the point at which Horse Creek enters OFG, and Stream 7w, which empties into a backwater swamp (G185/G186) that, in turn, empties into either Horse Creek or the lower end of Stream 6w immediately before it empties into Horse Creek. The alphanumeric designation of the backwater swamp in the preceding paragraph is based on the Map F-2 series, which assign such a designation to each existing wetland community and then identifies the wetland community. For example, the backwater swamp consists of a wet prairie (G185) surrounded by a mixed wetland hardwoods (G186). If a wetland consists of more than one wetland community, this Recommended Order will refer to it either as a wetland complex with its lowest-numbered wetland community--here, wetland complex G185--or the combination of wetland communities--here, G185/G186. Reclaimed wetlands are identified by Figure 13A5-1, which assigns each wetland an alphanumeric designation and identifies its community. The letter indicates if the reclaimed wetland is east ("E") or west ("W") of Horse Creek. Table 13A5-1 2AI identifies each reclaimed wetland by its alphanumeric designation, community, acreage, and status as connected, isolated, or isolated and ephemeral. Table 13A5-1 2AI identifies 110 wetlands to be reclaimed. The largest wetland is E003, which is a 23.8-acre mixed wetland hardwoods that constitutes the riparian wetland of the Stream 1e series. The next largest is W003, which is a 20.7-acre wet prairie at the headwaters of Stream 9w. Only three other reclaimed wetlands will be at least ten acres: E018, an 11.3-acre wet prairie fringe on the east side of Section 4; E020, an 11.5-acre freshwater marsh at the center of E018; and W039, an 11.2-acre bay swamp at the headwater of Stream 1w. Thirteen reclaimed wetlands are at least five acres, but less than ten acres, and 30 reclaimed wetlands are less than one acre. Table 13A5-1 2AI identifies 44 reclaimed ephemeral wetlands totaling 101 acres. Reclaimed uplands are identified by Map I-2. Although the scales of Map I-2 (one inch equals about 820.5 feet) and the Map F-2 series (one inch equals about 833.3 feet) are larger than the scales of nearly all of the other maps and figures in these cases, acreages derived from these maps for uplands and existing wetlands are very rough approximations and do not approach in accuracy the acreages derived from Table 13A5-1 2AI for reclaimed wetlands. These maps and figures omit one stream segment to be reclaimed. IMC and DEP restricted the designation scheme to streams and stream segments that had once been natural systems, thus excluding artificially created waterways, such as those created by agricultural ditches cut into swales to drain upslope wetlands and uplands. During the hearing, older aerial photographs revealed that, under this scheme, the parties had omitted one stream segment, which they designated Stream 3e?. Stream 3e? is northeast of Stream 3e, from which it is separated by a wetland (G133/G134/G135/G136). Besides the streams, two other areas within OFG require early identification due to their prominence in these cases. The northerly area is the Heart-Shaped Wetland (G138/G139/G140/G141/G143/G143A), which is the large wetland in Section 4 into which the Streams 1e series and Stream 3e empty. The other area of heightened importance is in the center of OFG in Sections 17 and 16 and is called the East Lobe, Central Lobe, and West Lobe or, collectively, the Lobes. Dominated by large bayhead headwaters (West Lobe--G197; Central Lobe--G179; East Lobe--G178), the Lobes and the streams connecting them to Horse Creek are entirely within the no-mine area. The West and Central Lobes connect to the west bank of Horse Creek by Streams 6w and 8w, respectively. The East Lobe connects to the east bank of Horse Creek by Stream 9e. The no-mine areas of the West and East Lobes are much larger than the no-mine area of the Central Lobe, and the East Lobe contains a large area of uplands extending east of, and supporting, the large bayhead. Most OFG wetlands are connected or contiguous, and many of these wetlands are riparian wetlands within the 100-year floodplain of Horse Creek or a floodplain of one of the tributaries of Horse Creek. (As used in this Recommended Order, the floodplain of Horse Creek runs roughly parallel to the banks of Horse Creek and excludes any portion of the floodplain more directly associated with Horse Creek's tributaries or their connected wetlands.) All or nearly all of the isolated wetlands on OFG are ephemeral and permanent, except in very low rainfall periods. The scale of mining is large. The phosphate matrix, which contains the phosphate rock, is overlaid by a layer of sand and clay overburden, which, with topsoil, is projected to range from 20-40 feet, averaging 27 feet, in thickness. The phosphate matrix is projected to range from 25-35 feet, averaging closer to 25 feet, in thickness, although as much as four feet of the matrix may consist of interburden, such as sand, clay, limerock, or gravelly materials. Thus, mining will remove, on average, 52 feet of the earth's surface. In no area will mining extend deeper than the top of the limey clay bed, which is the confining layer dividing the surficial aquifer from the intermediate aquifer, of which the limey clay bed is a part. (Technically, the matrix is part of the confining layer, but it provides so little confinement that it is easier to consider it part of the surficial aquifer. A consequence of this fact is that the removal of the matrix does not increase the rate of deep recharge, at least where the matrix is replaced with cast overburden.) At OFG, the thickness of the surficial aquifer varies from 65-70 feet at the basin divide to 50 feet or less at the riparian wetlands and averages 55 feet. Beneath the intermediate aquifer, which is about 300 feet thick at OFG, lies the Floridan Aquifer. IMC projects OFG to yield 24 million tons of phosphate rock, 26 million tons of clay tailings, and 68 million tons of sand tailings. IMC projects that the no-mine areas, which are discussed below, will result in five million tons of phosphate rock reserves remaining in the ground post-mining. The scale of the environmental impact of mining is correspondingly large. Mining removes all flora and fauna, all the topography, soils, and upper geology, in the path of the electric dragline, which, as long as a football field (including one end zone), removes the uplands, wetlands, streams, and soils covering the matrix. At the depths at which mining will take place, IMC will be removing the entire surficial aquifer. Applications, ERP, CRP Approval, and WRP Modification Preliminary Matters These cases involve permits and an approval of the phosphate mining and reclamation processes. These cases do not involve the processes by which IMC transforms phosphate into end products, mostly fertilizer. With one exception, these cases do not involve the processes by which IMC separates the phosphate ore from the sand and clay (i.e., the beneficiation process). (The exception is that IMC is seeking to extend by ten years the life of the Ft. Green beneficiation plant to separate the phosphate from the matrix slurried from OFG.) These other post- mining processes, which are separately permitted, are not directly involved in these cases because IMC will slurry the phosphate matrix mined from OFG to the existing Ft. Green beneficiation plant, which is already permitted and operating. Even though the WRP modification will authorize the relocating of already-permitted CSAs at the Ft. Green Mine, the WRP modification will not authorize the design or construction of the embankments that retain the water within these CSAs while they are essentially clay ponds. DEP will separately permit the construction and operation of CSAs O-1 and O-2. Application and Proposed Agency Action On April 24, 2000, IMC filed a Consolidated Development Application for an ERP to mine phosphate from the proposed 20,675-acre Ona Mine, approval of the CRP for the Ona Mine following the completion of mining, and modification to the existing WRP for the Ft. Green Mine to install three CSAs in the area of the Ft. Green Mine immediately west of the Ona Mine and extend the life of the Ft. Green beneficiation plant by ten years to process the matrix from the Ona Mine. On January 17, 2003, DEP issued an Intent to Issue an ERP and proposed approval of the CRP. Petitioners in several of the above-styled cases challenged this proposed agency action, and the parties embarked upon an energetic prehearing process of preparation, including extensive discovery and prehearing telephone conferences with the Administrative Law Judge, in anticipation of a final hearing in the fall of 2003. IMC and DEP entered into a Team Permitting Agreement, pursuant to 1996 legislation creating the concept of Ecosystem Management. The Team Permitting Agreement incorporates the concept of "net ecosystem benefit," but, on its face, is not binding on IMC. The obvious purpose of the Team Permitting Agreement was to induce the permitting agencies (i.e., DEP, Florida Fish and Wildlife Conservation Commission (FWC), Southwest Florida Water Management District (SWFWMD), two regional planning councils, the Florida Department of Community Affairs, the Florida Department of Transportation (DOT), Hardee County, DeSoto County, and the U.S. Army Corps of Engineers) to use a common development application and coordinate, to the greatest practical extent, their respective reviews of the proposed activities of IMC. Three weeks prior to the start of the final hearing, on September 15, 2003, DEP issued the Final Order in Charlotte County et al. v. IMC Phosphates Company and Department of Environmental Protection, 2003 WL 21801924, 4 ER FALR 42 (Altman Final Order). The Altman Final Order denies IMC's application for a WRP/ERP and disapproves IMC's proposed CRP for the Altman tract, which is a short distance northwest of OFG. Although the final and recommended orders are detailed and complex, the Altman Final Order essentially concludes that IMC's CRP was inconsistent with applicable law because its basic reclamation concept was "to replace an existing system of high-quality wetlands . . . with a deep freshwater marsh." On the same date of the Altman Final Order, DEP Deputy Secretary Allan Bedwell ordered DEP's Bureau of Mine Reclamation (BMR) to re-examine IMC's application for an ERP and request for approval of the CRP for the Ona Mine to assure consistency between the proposed agency action approving the ERP, CRP, and WRP modification and the Altman Final Order. The Bedwell memorandum specifically directs BMR to verify IMC's classification and characterization of the extent and quality of wetlands on the site; verify that IMC's proposed reclamation activities, including its proposed control of nuisance or exotic species, "maintain or improve the water quality and function" of the biological systems present at the site prior to mining; and verify that IMC meets the financial assurance requirements of law. The memorandum concludes by directing BMR to modify any proposed agency action, if necessary. By memorandum dated January 5, 2004, Richard Cantrell and Janet Llewellyn, Deputy Directors of DEP's Division of Water Management Resources, responded to the memorandum from Deputy Secretary Bedwell. With respect to IMC's classification and characterization of wetlands, the January 5 memorandum states that DEP staff had conducted additional review of available aerial photographs, reviewed field notes from previous field inspections, conducted new field inspections, and received comments from IMC and Charlotte County. To describe better onsite habitats and communities, DEP staff had also revised the DOT Florida Land Use, Cover, and Forms Classification System (FLUCFCS) for use at OFG. The FLUCFCS codes are a three-digit numbering system to classify and identify individual vegetative communities or land uses. With respect to the ability of the proposed reclamation to maintain or improve the water quality and function of biological systems, the January 5 memorandum states that Deputy Directors Cantrell and Llewellyn had recommended to IMC that it consider phasing the mining on Ona, so that it could apply its experience in reclaiming OFG to the remainder of the original Ona Mine; preserving additional onsite natural stream channels and proposing more detailed reclamation plans for mined streams; preserving additional onsite bay-dominated wetland systems; providing additional assurances that upgradient sand/scrub areas will continue to support hydrologically, through seepage, preserved and restored bayheads; providing a plan to control nuisance and exotic species in the uplands, which, if infested, would degrade adjacent wetlands post-mining; and providing assurances that groundwater flows to Horse Creek and its preserved tributaries will be maintained during mining and post-reclamation. With respect to financial responsibility, the January 5 memorandum states that Deputy Directors Cantrell and Llewellyn had advised IMC that it must provide its financial responsibility for the mitigation of all wetlands authorized to be mined, rather than providing its financial responsibility on a phased basis, as it had previously proposed. On January 30, 2004, IMC filed a voluminous amendment to the Consolidated Development Application in a package known as the January submittal. The most evident change made by the January submittal is the reduction of the Ona Mine to OFG, which was the westernmost one-fifth of the original Ona Mine. The introduction to the January submittal highlights the changes that IMC made to the original application. The introduction explains that IMC has employed a revised mapping protocol to ensure that all waters of the State, including wetlands delineated by Florida Administrative Code Rule 62-340.300 and other surface waters delineated by Florida Administrative Code Rule 62-340.600, are classified as wetlands or water, pursuant to the modified FLUCFCS codes. Rejecting the nomenclature of the January 5 memorandum regarding the phasing of mining at the Ona site, the introduction to the January submittal identifies OFG as a 4197- acre, "free-standing" mining tract, not in any way "coupled to or dependent on the development of the remainder of the Ona Tract," from which it was taken. The introduction explains that "free-standing" means that OFG is a "complete mining, reclamation, and mitigation proposal" and that the OFG ERP will be "for a single-phase project." The introduction to the January submittal notes that IMC has enlarged the no-mine area to include "nearly all of the natural stream channel tributaries to Horse Creek present in the portions of the Parcel that have not been converted to improved pasture." The amendments thus avoid disturbing four additional natural stream segments. The introduction explains that IMC considered a series of factors in determining whether to mine a stream segment: "stream segments length, the existing land cover adjacent to the stream and its watershed, the complexity of the channel geometry[,] and historical agricultural impacts." The introduction adds that IMC has added a "state-of-the-art" stream restoration plan for mined natural streams. The introduction to the January submittal states that IMC responded in two ways to the suggestions about bay swamps in the January 5 memorandum. First, IMC modified the conventional mapping protocol for bay swamps. Rather than require that the canopy of the subject community be dominated by loblolly bay, sweetbay, red bay, and swamp bay trees, as prescribed by the FLUCFCS codes, IMC designated as bayheads "depressional, seepage-driven forested headwater wetlands, surrounded, at least in part, by moderately to well drained upland soils, with a defined outlet connection to waterways such that the 'bay head' soils are perennially moist but infrequently inundated." This new mapping protocol did not require the presence of bay trees in the canopy. Second, IMC enlarged the no-mine areas to avoid disturbing all but nine percent of existing bay swamps at OFG, totaling less than ten acres. IMC based its mine/no-mine decisions for particular bayheads on analysis of the hydrological, water quality, and relative functional value provided by these communities to fish and wildlife. The introduction concludes that IMC has also developed detailed plans to mitigate for the few mined bayheads. The introduction to the January submittal states that IMC has added new protections for the sand/scrub areas upgradient from, and providing seepage into, the bayheads in the West and East Lobes. First, IMC will avoid mining certain of these areas, presumably adjacent to the East Lobe. Second, IMC will employ special mining techniques and schedules to reclaim these upland areas quickly and effectively. Additionally, the introduction notes that IMC is proposing to: align the dragline "cut patterns" such that the spoil piles will be aligned with the groundwater seepage path where feasible or, where not feasible, to grade the spoil piles prior to backfilling the mine voids with sand so as not to impede post- reclamation groundwater flow; accelerate the sand backfilling schedule of the mined voids adjacent to avoided "bay heads" to one year following mining disturbance; and create a reclaimed stratigraphy that results in post-reclamation seasonal high and normal water table elevations and hydraulic conductivities in the seepage slopes that will provide the hydrologic support required to sustain these communities. As explained in a later section of the introduction to the January submittal, "stratigraphy" refers to the soil layers or horizons, which are described in detail below. The introduction states: "The majority of the overburden will be placed at depths below the surface soil horizons. As a result, the surface soils will either be comprised of translocated surface soils or a loose mixture of 'green manure organics,' overburden, and sand that both resembles the native soils and provides a suitable growing medium for the targeted vegetative communities." The introduction adds that, at final grade, sand tailings will always overlie overburden by at least 15 inches. The introduction asserts that the overburden underlying the backfilled sand tailings will be "comprised of and have properties which are similar to B horizons (subsoils) and C horizons (substratums) of native Florida soils." The introduction to the January submittal identifies a Habitat Management Plan (also known as the Site Habitat Management Plan) that, with the Conservation Easement and Easement Management Plan discussed below, will guide the revegetation of upland natural systems, control nuisance and exotic species in uplands, and manage all potential listed species that may be present, whether or not observed, in areas to be mined. The introduction also mentions habitat enhancements "to relocate Florida mice" and to manage gopher tortoises. The introduction concludes with IMC's undertaking to ensure that exotic/nuisance cover does not exceed ten percent in all reclaimed wetlands and to provide a 300-foot buffer around wetlands where cogongrass--a highly invasive nuisance exotic described in more detail below--will not exceed five percent coverage. The introduction to the January submittal notes that the proposed activities will maintain groundwater flows to Horse Creek and tributaries in the no-mine areas during mining and post-reclamation. The introduction again mentions IMC's commitment, where feasible, to align spoil piles with groundwater flow and, where not feasible, grade spoil piles before backfilling so as to add a thicker band of sand to these areas. The introduction also cites the ditch and berm system as a means to maintain groundwater seepage during mining. The introduction to the January submittal states that IMC will meet its financial-responsibility requirements for the entire cost of wetland-mitigation at OFG. The January submittal contains a discussion of community-mapping protocol. IMC's methodology for mapping bay swamps is discussed above. The most common vegetative communities and land uses are described in the following paragraphs. Improved pasture is actively grazed pasture dominated by cultivated pasture grasses, such as bahiagrass, but may support native grasses. Improved pasture may contain sporadic shrubs and trees. Pine flatwoods occupy flat topography on relatively poorly drained, acidic soils low in nutrients. The overstory is discontinuous with areas of dense, species-rich undergrowth or groundcover. Longleaf pine and slash pine predominate. Pine flatwoods require frequent fires, which are carried by grasses, and the pines' thick bark helps prevent fire damage to the trees. At one time, about three-quarters of Florida was covered by pine flatwoods. Palmetto prairies typically represent the undergrowth of pine flatwoods. Once the trees are removed, such as by timbering, the resulting community is a palmetto prairie, which is characterized by an often-dense cover of saw palmettos with no or scattered pines or oaks. Occupying dry, sandy, well-drained sites, sand live oak communities feature a predominance of sand live oaks and often succeed in relatively well-drained pine flatwoods after the removal of the pines, conversion to palmetto prairie, and suppression of fire. Sand live oak may also occupy xeric oak communities. Moister soils may support live oak communities, which also may succeed pine flatwoods after the removal of the pines, conversion to palmetto prairie, and suppression of fire. Hardwood-conifer mixed is a blend of hardwoods and pines with trees of both categories forming one-third to two- thirds of the cover. Hardwoods are often laurel oak and live oak, and pines are often slash pine, longleaf pine, and sand pine. The midstory is typically occupied by younger individuals of the overstory communities and wax myrtle. If sufficient light reaches the ground, groundcover may exist. Temperate hardwoods are often a forested uplands transition to a wetland. Temperate hardwoods are usually dominated by laurel oak, but other canopy species may include cabbage palm, slash pine, live oak, and water oak. Mixed hardwoods is a similar community, except that water oak is predominant in the canopy. Two of the three most prevalent forested wetlands on OFG are bay swamps, which have been discussed, and hydric oak forest, which, because of their location in the Horse Creek floodplain, will not be mined. At DEP's request, IMC remapped some of the floodplain that was uplands (and already in the no- mine area) to hydric oak forest. The other prevalent forested wetlands on OFG is mixed wetland hardwoods, which consists of a variety of hardwood species, such as the canopy species of red maple, laurel oak, live oak, sweetbay, and American elm. Slash pines may occur, but may not constitute more than one-third of the canopy. Suitable shrubs include primrose willow, wax myrtle, and buttonbush. Ferns are often present as groundcover. Often immediately downgradient of bay swamps, mixed wetland hardwoods are typically in the hydric floodplains of small streams. Transitioning between uplands, such as palmetto prairies, and the wetter soils hosting bay swamps and mixed wetland hardwoods, wetland forested mixed communities (also known as wetland mixed hardwood-coniferous) often occupy wet prairies from which fire has been suppressed for at least 20 years and, as such, "are largely or entirely an artifact of land use practices during the past sixty years or so that have allowed the conversion of wet prairies . . . to this cover type." The canopy of wetland forested mixed is slash pine, laurel oaks, live oaks, and other hardwoods that tolerate or prefer wetter soils. Wet prairies are a dense, species-rich herbaceous wetland, usually dominated by grasses. Wet prairies occupy soil that is frequently wet, but only briefly and shallowly inundated. Similar to freshwater marshes, but with shorter hydroperiods, wet prairies often fringe marshes, and their border will shift in accordance with rainfall levels over several years. Freshwater marshes consist predominantly of emergent aquatic herbs growing in shallow ponds or sloughs. Typical marsh herbs include pickerelweed, maidencane, and beakrushes. Hydroperiod and water depth drive the presence of species in different locations within a freshwater marsh. Marshes may be isolated or may occupy a slough in which their water flow is unidirectional. Heavily grazed or drained marshes may suffer dominance of primrose willow. Abundant softweed may indicate ditching, and soft rush, which cattle avoid, may indicate heavy grazing. Shrub marshes succeed stillwater freshwater marshes from which fire has been excluded. Shrub marshes form after agricultural ditching or culverted fill-road building. Common shrub species include buttonbush, southern willow, and primrose willow. Hydric trees, such as red maple and swamp tupelo, may occupy the edges of shrub marshes. IMC supplemented the January submittal with submittals dated February 26 and 27, 2004. Collectively, these are known as the February submittal. The February submittal is much less- extensive than the January submittal, although it includes substantive changes. After examining the January and February submittals, on February 27, 2004, DEP issued a Revised Notice of Intent to Issue an ERP for OFG, approved a revised CRP for OFG, and issued a revised WRP modification for the Ft. Green Mine, which now authorizes two CSAs--O-1 and O-2--that have the effect of relocating the previously approved CSAs farther away from Horse Creek and reducing their size due to the reduced scale of OFG as compared to the original Ona Mine; reconfiguring certain mitigation wetlands, necessitated by the relocation of CSAs O-1 and O-2, with a net addition of 2.7 acres of herbaceous wetland area; and changing the reclamation schedule to conform to the already-approved CRP for the Ft. Green Mine. IMC supplemented the January and February submittals with submittals dated March 30, April 18, and April 21, 2004. These submittals, which are known as the Composite submittal, are much less-extensive than the February submittal. DEP expressly incorporated the February submittal into the ERP, CRP approval, and WRP modification dated February 27, 2004. DEP has impliedly incorporated the changes in the Composite submittal into the ERP, CRP approval, and WRP modification. Thus, this Recommended Order uses the latest version of these documents when discussing the relevant permit or approval. The March 30, 2004, submittal updates the following maps, figures, and tables: Map F-2 (to correct legend), Map I-2 (to correct the post-reclamation vegetation in the vicinity of Streams 3e, 1w, 2w, 3w, and 4w), Figures 13A5-1 and 13B-8 (to reflect changes to Map I-2), Tables 12A1-1 and 13A1-1 (revised land uses in several stream locations), and Tables 13A5-1, 345A-1, and 26O-1 (to reflect above changes). The March 30, 2004, submittal also includes the Draft Study Plan for Burrowing Owls and Amphibians and revised Tables A and B for the Financial Responsibility section of the ERP. No material revisions are included in the submittals after March 30, 2004. Submittals after March 30, 2004, include financial responsibility forms, including a draft escrow agreement, and updated information on the temporary wetland crossing at the point that Stream 2e forms at the downstream end of the Heart-Shaped Wetland. The last item, dated April 20, 2004, is a revision of Figure 13B-8, but solely for the purpose of showing that the Heart-Shaped Wetland remains connected to Stream 2e, despite the temporary presence of a crossing. This is the last revision to the CDA prior to the commencement of the hearing. During the hearing, IMC submitted modifications of the mining and reclamation activities, and DEP agreed to all of these modifications. During the hearing, DEP proposed modifications of the mining and reclamation activities, and IMC agreed to all of these modifications. These modifications, such as identifying the annual hydroperiod of bay swamps as 8-11 months and the final changes to post-reclamation topography, are identified in this Recommended Order and incorporated into all references to the ERP or CRP approval. In general, the ERP addresses wetlands, surface waters, and species dependent upon either, and the CRP addresses uplands and species dependent exclusively upon uplands. Later sections of the Recommended Order will discuss the ERP, the CRP approval, and the WRP modification. All of the maps, figures, and tables incorporated into the ERP, CRP approval, or WRP modification are contained in the CDA. Overview of Mined Areas, No-Mine Areas, and Reclaimed Areas The ERP permits IMC to mine 3477 acres and requires IMC to reclaim 3477 acres. The ERP recognizes that IMC will not mine 721 acres, which is about 17 percent of the 4197-acre site. (Most acreage figures are rounded-off in this Recommended Order, so totals may not always appear accurate.) Although various exhibits and witnesses sometimes refer to the no-mine area as the preserved area, this label is true only insofar as IMC will "preserve" the area from mining. However, post-reclamation, the area is not preserved. After the property reverts to the Carlton-Smith family, it will return to its historical agricultural uses, subject to a Conservation Easement that is discussed below. Table 12A1-1 is the Mine Wide Land Use Analysis. Table 12A1-1 identifies, by acreage, each use or community presently at OFG, such acreage proposed to be mined, and such acreage proposed to be reclaimed. When not listed separately, this Recommended Order combines all non-forested wetlands, including mostly herbaceous wetlands and shrub marshes, into the category of herbaceous wetlands. Shrub marshes presently account for only 4.7 acres at OFG and will account for only 10.3 acres, post-reclamation. Ignoring 35 acres that presently are barren or in transportation or urban uses, the present uses or communities of OFG are agricultural (2146 acres), upland forests (904 acres), rangeland (510 acres), forested wetlands (380 acres), herbaceous wetlands (208 acres), and open water (15 acres). Nearly all of the existing agricultural uses are improved pasture (1942 acres); the only other use of significance is 165 acres of citrus. Well over half of the area to be mined is agricultural. Over half of the area to be mined is improved pasture (1776 acres, or about 51 percent of the mined area). Adding the citrus groves, woodland pasture, and insignificant other agricultural uses to the area to be mined, the total of agricultural uses to be mined is 1976 acres, or 57 percent of the mined area. The two most prevalent upland forest communities presently at OFG are sand live oak and pine flatwoods; the next largest community, hardwood-conifer mixed, accounts for about half of the size of sand live oak or pine flatwoods. These upland forests contribute about one-fifth of the area to be mined (731 acres, or 21 percent of the mined area). Cumulatively, then, agricultural land and upland forests constitute 78 percent of the mined area. For all practical purposes, all of the rangeland presently at OFG is palmetto prairie. This unimproved rangeland contributes a little less to the mining area that do upland forests; mining will consume 475 acres of rangeland, which is 14 percent of the mined area. Cumulatively, then, agricultural land, upland forests, and native rangeland will constitute 92 percent of the mined area. The addition of the remaining upland uses--25 acres of roads, 5 acres of barren spoil areas, and one acre of residential--results in a total of 3213 acres, or still 92 percent, of the 3477 acres to be mined. This leaves eight percent of the mined area, or 264 acres, as wetlands and other surface waters. As noted above, the wetlands are divided into forested and herbaceous wetlands. Forested wetlands will contribute 82 acres, or about two percent, of the mined area. Nearly all of the forested wetlands presently at OFG are divided almost equally among mixed wetland hardwoods, hydric oak forests, and bay swamps. Bay swamps total 104 acres. In terms of the forested wetlands present at OFG, mining will consume mostly mixed wetland hardwoods, of which 43 acres, or 36 percent of those present at OFG, will be mined. Mining will eliminate only nine acres, or nine percent, of bay swamps and six acres, or six percent, or hydric oak forests. Mining will eliminate a large percentage-- 67 percent--of hydric pine flatwoods present at OFG, but this is 12 acres of the 18 existing acres of this wetland forest community. Herbaceous wetlands will contribute 168 acres, or about five percent, of the mined area. Nearly all of the herbaceous wetland communities are wet prairies (108 acres) and freshwater marshes (81 acres). Mining will eliminate 95 acres, or 88 percent, of the wet prairie present at OFG, and 67 acres, or 83 percent, of the freshwater marshes present at OFG. IMC will mine 13.5 acres of open water, which consists primarily of cattle ponds and ditches. The only natural water habitat is natural streams, which total 2.2 acres. IMC will mine 0.9 acres of natural streams. Also incorporated into the ERP, Table 13A1-5, provides another measure of the impact of mining upon natural streams. According to Table 13A1-5, IMC will mine 2.8 acres of the 25.6 acres of natural streams. As noted in Table 13A1-5, reclamation of streams, which is discussed in detail below, is based on length, not acreage, and, under the circumstances, a linear measure is superior to an areal measure. Table 12A1-1 also provides the acreage of reclaimed community that IMC will construct. These habitats or uses are listed in the order of the size of the area to be reclaimed, starting with the largest. For agriculture, IMC will reclaim 1769 acres after mining 1976 acres. Adding the 170 acres of agriculture in the no-mine area, agricultural uses will total, post-reclamation, 1939 acres. For upland forest, IMC will reclaim 1055 acres after mining 731 acres. Adding the 173 acres of upland forest in the no-mine area, upland forest habitat will total, post- reclamation, 1227 acres. For rangeland, IMC will reclaim 323 acres after mining 475 acres. Adding the 35 acres of rangeland in the no- mine area, rangeland will total, post-reclamation, 358 acres. For herbaceous wetlands, IMC will reclaim 217 acres after mining 168 acres. Adding the 39 acres of herbaceous wetlands in the no-mine area, herbaceous wetlands will total, post-reclamation, 256 acres. For forested wetlands, IMC will reclaim 106 acres after mining 82 acres. Adding the 298 acres of forested wetlands in the no-mine area, forested wetlands will total, post-reclamation, 404 acres. ERP ERP Specific Condition 3 requires IMC to provide to DEP for its approval the form of financial responsibility that IMC chooses to use to secure performance of its mitigation costs. IMC may not work in any wetland or surface water until DEP has approved the method by which IMC has demonstrated financial responsibility. DEP shall release the security for each individual wetland that has been released by BMR, pursuant to Specific Condition 17. The escrow agreement is a two-party contract between IMC and J.P. Morgan Trust Company, as escrow agent. The escrow agreement acknowledges that IMC will transfer cash or securities to the escrow agent in the stated amount, representing IMC's obligations to perform ERP mitigation plus the ten percent add- on noted in the Conclusions of Law. If IMC fails to comply with the ERP or Section 3.3.7 of the SWFWMD Basis of Review, the escrow agent is authorized to make payments to DEP, upon receipt of DEP's written certification of IMC's default. The escrow agreement may be amended only by an instrument signed by IMC, DEP, and the escrow agent. ERP Specific Condition 3 requires IMC to calculate the amount of the security based on Table B, which is the Wetland Mitigation Financial Summary. Table B lists each forested and wetland community from Table 12A1-1, the acreage for each community, and the unit costs per acre of mitigation. The acreage figures are the acreage figures on Table 12A1-1. The unit costs per acre are as follows with the FLUCFCS codes in parentheses: herbaceous (641, 643)--$7304; forested bay wetland (611)--$11,692; other forested wetland (613, 617, 619, 630)--$11,347; shrub (646)--$8780; hydric palmetto prairie (648)--$9231; and (hydric) pine flatwoods (625)--$10,568. Table B also shows 10,141 feet of streams to be reclaimed at a cost per foot of $37, stream macroinvertebrate sampling at a total cost of $48,100, and water quality/quantity monitoring at a cost of $293,000. Adding the costs of wetland and stream reclamation, sampling, and monitoring, plus ten percent, Table B calculates the mitigation liability of IMC as $3,865,569. IMC has agreed to increase this amount for the reclamation of Stream 3e?. ERP Specific Condition 4 requires IMC to submit to BMR annual narrative reports, including the actual or projected start date, a description of the work completed since the last annual report, a description of the work anticipated for the next year, and the results of any pre-mining surveys of wildlife and endangered or threatened species conducted during the preceding year. The reports must describe any problems encountered and solutions implemented. ERP Specific Condition 5 requires IMC to submit to BMR annual hydrology reports. Relative to initial planting, IMC shall submit to BMR vegetative statistic reports in year 1, year 2, year 3, year 5, and every two years after year 5, IMC must submit to BMR vegetation statistic reports. ERP Specific Condition 6 addresses water quality in wetlands or other surface waters adjacent to, or downstream of, any site preparation, mining, or reclamation activities. Specific Condition 6.a requires, prior to any clearing or mining, IMC to sever the areas to be disturbed from adjacent wetlands. IMC severs or isolates the mining area when it constructs the ditch and berm adjacent to, but upland of, the adjacent wetlands not to be mined. Figure 14E-1 portrays the elements of the ditch and berm system as all outside of the no-mine area (or OFG property line, where applicable). In the illustration, from the mine cut toward the no-mine area (or OFG property line), IMC will construct the ditch, the 15-foot wide berm, the monitoring wells, and the silt fence. ERP Specific Condition 6.b requires the ditch and berm system to remain in place until IMC has completed mining and reclamation, monitoring indicates that no violation of "State Water Quality Standards" are expected, and DEP has determined that "the restored wetlands are adequately stabilized and sufficiently acclimated to ambient hydrological conditions." DEP's decision to allow the removal of the ditch and berm system shall be based on a site inspection and water quality monitoring data. Upon removal of the ditch and berm system, the area that had been within the ditch and berm system shall be restored to grade and revegetated according to the methods and criteria set forth in Specific Condition 14. ERP Specific Condition 6.c requires IMC to use best management practices for turbidity and erosion control to prevent siltation and turbid discharges in excess of State water quality standards, under Chapter 62-302, Florida Administrative Code. Specific Condition 6.d requires IMC daily to inspect and maintain its turbidity-control devices. If the berm impounds water above grade, IMC must daily visually inspect the integrity and stability of the embankment. ERP Specific Condition 7 requires that IMC implement a baseline monitoring program for surface water and groundwater and continue the program through the end of the mine life. The data from this program shall be included in the annual narrative reports described in Specific Condition 4. The locations of the sampling sites are depicted on Map D-4. ERP Specific Condition 7.a identifies three monitoring stations, which are in Horse Creek just upstream of the stream's entrance onto OFG (and possibly just upstream of the offsite confluence of Stream 2e with Horse Creek), in Horse Creek at State Road 64, and in West Fork a short distance upstream of its confluence with Horse Creek. Before and during mining, IMC must monthly monitor 18 parameters, including temperature, pH, dissolved oxygen, total suspended solids, conductivity, turbidity, color, total phosphorous, ammonia, nitrate/nitrite, and chlorophyll a. During mining, IMC must semi-annually monitor 11 additional parameters, including alkalinity, biological oxygen demand, chloride, and iron. ERP Specific Condition 7.b identifies one monitoring station, which is at the junction of Stream 6w and Horse Creek. Before and during mining, IMC must monthly monitor ten parameters, including temperature, pH, dissolved oxygen, total suspended solids, conductivity, and color. During mining operations, IMC must semi-annually monitor the same 11 additional parameters described in Specific Condition 7.a. ERP Specific Condition 7.c identifies two clusters of monitoring wells, one located near the offsite confluence of Stream 2e with Horse Creek and one located near the collecting station on West Fork near its junction with Horse Creek. During mining operations, IMC must semi-annually monitor 23 parameters, including pH, temperature, conductivity, alkalinity, total phosphorous, color, turbidity, chloride, iron, and nitrate/nitrite. ERP Specific Condition 8 requires IMC immediately to cease all work contributing to turbidity violations of "State Water Quality Standards established pursuant to Chapter 62-302, F.A.C." Specific Condition 8 requires IMC to stabilize all exposed soils contributing to the violation, modify work procedures that were responsible for the violation, repair existing turbidity-control devices, and install more such devices. Specific Condition 8 requires IMC to notify BMR within 24 hours of the detection of any turbidity violation. ERP Specific Condition 9 requires IMC to report all unauthorized releases or spills of wastewater or stormwater in excess of 1000 gallons per incident to BMR, as soon as practicable, but not later than 24 hours after detection. ERP Specific Condition 10 addresses water levels and flows in wetlands and other surface waters adjacent to, and downstream of, any site preparation, mining, and reclamation activities. Prior to any clearing or mining activities adjacent to no-mine wetlands and other surface waters, Specific Condition 10.a requires IMC to install monitoring wells and staff gauges and commence monitoring water levels, as required by ERP Monitoring Required, which is a part of the ERP that is discussed below. IMC shall monitor water levels in each of the no-mine streams at the point that it intercepts the 100-year floodplain of Horse Creek. ERP Specific Condition 10.a provides: During mining, recharge ditches adjacent to no-mine areas shall be charged with water or recharge wells shall be installed to maintain base flows and/or minimize stress to the vegetation in the preservation areas. Water levels in the recharge ditches shall be maintained at levels sufficient to support the normal seasonal water level fluctuations in the wetlands as determined from the baseline monitoring included in Table MR-1. Under ERP Specific Condition 10.a, prior to any clearing or mine activities, IMC must install monitoring wells and staff gauges and monitor water levels, as specified in the ERP Monitoring Required. IMC must daily monitor water levels in each of the no-mine streams at the point of its interception with the 100-year floodplain of Horse Creek. During mining, IMC shall charge recharge ditches with water or install recharge wells to maintain base flows and minimize stress to vegetation in no-mine areas. IMC must maintain water levels in the recharge ditches at levels sufficient to support the normal seasonal water level fluctuations in the wetlands, as determined from the baseline monitoring included in Table MR-1, which is described below. IMC must daily check the water levels in the recharge ditches, record this information in logs, and make these logs available to BMR during its quarterly inspections. IMC shall monthly inspect the water levels in adjacent no-mine wetlands and notify BMR in writing if these wetlands show signs of stress. If adjacent no-mine wetlands become stressed, upon DEP's approval, IMC will take additional actions, such as altering mining and reclamation procedures, modifying the recharge ditch, providing additional sources of water, and conducting additional monitoring. During the hearing, IMC hydrologist and engineer Dr. John Garlanger testified: "[IMC] will install a recharge well system along the preserved areas." (Tr., p. 2800) The parties treated recharge wells as a part of the ditch and berm system, both at the hearing and in their proposed recommended orders (DEP, paragraph 75; Charlotte County, paragraph 575; and IMC, paragraph 339.) However, Specific Condition 10.a imposes no such obligation upon IMC, nor does any other provision in the ERP or the CDA. The above-quoted provision of Specific Condition 10.a identifies recharge wells as an alternative. The other option in Specific Condition 10.a is to charge the ditches with water. This condition is confusing because it poses, as alternative requirements, one option of a specific effect--i.e., recharged ditches--and the other option of a means of achieving that effect--i.e., recharge wells. The objective is sufficient water in the ditch. The means of charging the ditch would appear to be limited to direct rainfall, pumping water from the mine cuts, diverting water from the mine recirculation system, or pumping water from the intermediate or Floridan aquifer through recharge wells; at least the first two of these charging options are already incorporated into the OFG ditch and berm system. Confirming that recharge wells are optional is Figure 14E-1, which labels the recharge well depicted at the bottom of the ditch as "Alternate--Recharge Well." Figure 14E-1 illustrates a pump forcing the water from the bottom of the deeper mine cut to the bottom of the recharge ditch. (Figure 14E-1 also illustrates that--in order, running from the mine cut toward the no-mine area (or OFG property line)--the ditch, the 15-foot wide berm, the monitoring wells, and the silt fence will all be located outside of the no-mine area (or within OFG).) ERP Specific Condition 10.b prohibits reductions in downstream flows from the project area that will cause water quality violations in Horse Creek or the degradation of natural systems. IMC shall monitor surface water levels continuously at the above-described points at State Road 64 and West Fork and monthly near the above-described junction of Stream 2e and Horse Creek. IMC shall monitor monthly at the above-described clusters of monitoring well locations and at piezometers located across Section 9 from the no-mine area into the uplands to the east, in the West Lobe and the adjacent uplands to the west, in the East Lobe and the adjacent uplands to the east, and in Horse Creek about one-quarter mile from the southern border of OFG. IMC shall daily monitor rainfalls at a rain gauge near the junction of Stream 2e and Horse Creek. IMC shall report the results of the monitoring in the reports required in Specific Condition 4. ERP Specific Condition 11 requires IMC to obtain authorization from FWC before relocating gopher tortoises or disturbing their burrows. ERP Specific Condition 11 also requires IMC to relocate gopher frogs and other commensals to FWC-approved sites before clearing. At the time of the hearing, FWC had not yet approved IMC's plan to relocate gopher tortoises, but this approval was expected shortly. ERP Specific Condition 12 requires IMC to complete mining, filling, and reclamation activities generally in accordance with the schedule stated in this condition. Specific Condition 12.a prohibits IMC from commencing severance or site preparation more than six months prior to mining, except as approved by DEP for directly transferring topsoil or muck to a contoured mitigation site. IMC must complete final grading, including muck placement, not later than 18 months after the completion of mining operations, which include the backfilling of sand tailings. IMC must conduct its hydrological assessment in the first year after contouring. ERP Specific Condition 12.a provides a timetable for work in wetlands and other surface waters. IMC may not commence severance or site preparation more than six months prior to mining. IMC shall complete final grading, including muck placement, not more than 18 months after the completion of mining operations, including backfilling with sand tailings. IMC shall complete Phase A planting, which is of species that tolerate a wide range of water levels, not more than six months after final grading or 12 months after muck placement. IMC shall conduct the hydrological assessment in the initial year after coutouring. IMC shall complete Phase B planting, which is of species that tolerate a narrower range of water levels, within 12 months after the hydrological assessment and Phase C planting, which is shade-adapted groundcover and shrubs, as well as additional trees and shrubs required to meet the density requirements of ERP Specific Condition 21 [sic; probably should be ERP Specific Condition 16], at least two years prior to release of forested wetlands. ERP Specific Condition 12.b provides that IMC shall clear, contour, revegetate, and reconnect wetlands and watersheds as shown in Tables 3AI-6A and 3AI-10A, Maps H-1, H-9, and I-6, and Figures 13B-8, 13A5-1, and CL-1. Table 3AI-6A lists each reclaimed wetland by number, the last year in which it will be disturbed, the last year in which it will be mined, the year in which grading will be completed, the year in which revegetation will be completed, and the number of years between mining or disturbance and reclamation and revegetation. The span of years between mining or disturbance and reclamation ranges from three (two wetlands) to eight (six wetlands). Table 3AI-10A is the Reclamation Schedule Summary. The table identifies four reclamation units in the Horse Creek sub-basin, one reclamation unit in the West Fork sub-basin, and one reclamation unit in the Brushy Creek sub-basin. For each reclamation unit, Table 3AI-10A shows the period of mining, period of mine operations, period for contouring, and period for revegetation. These years are relative: mining runs four years, mine operations run seven or eight years (starting one year after mining starts), contouring runs seven or eight years (starting within one year of the end of mining), and revegetation runs five or six years (starting one year after the start of contouring). Map H-1 is the Mine Plan. Map H-1 assumes four draglines will operate in OFG for five years of active mining. IMC's tentative plan is first to mine the west side of OFG, which is nearer the Ft. Green Mine at which the draglines are presumably deployed at present, and then to mine adjacent mining blocks. For instance, IMC would mine the northwest corner of Section 4 in Year 1, the southwest corner of Section 4 in Year 2, the northeast corner of Section 4 in Year 3, and the southeast corner of Section 4 in Year 4 before removing the dragline south of Section 4 to mine an unmined area in Year 5. Map H-1 depicts the ditch and berm system running continuously along the edge of the no-mine area from the north end of OFG, south along the no-mine borders that trace the east and west edges of the 100-year floodplain of Horse Creek, to their southern termini. On the east floodplain, the ditch and berm system turns east at the northwest corner of Section 21, near the Carlton cutout, runs to the easternmost extent of OFG, turns north to the northeast corner of Section 4, and runs to the northwest corner of Section 4, where the ditch and berm system ends. On the west floodplain, the ditch and berm system runs to the southernmost extent of OFG near its confluence with West Fork, turns west and north, as it traces the border of OFG along Sections 29, 20, and 19, where it ends at a point about one-quarter mile from the northern boundary of Section 19. For the areas closest to the no-mine area, Map H-1 also depicts the direction of the mine cuts and, inferentially, the spoil piles. These cuts and piles are generally perpendicular to the direction of Horse Creek. Figure 2AI-24 displays the locations of the six reclamation units identified in Table 3AI-10A. The West Fork and Brushy Creek reclamation units occupy the sub-basins bearing their names, so they are at the western and eastern edges, respectively, of OFG. The HC(1) reclamation unit is almost all of Section 4. According to Table 3AI-10A, IMC will mine this reclamation unit from 2006-09, contour it from 2009-15, and revegetate it from 2010-15. Combining the information from Map H-1 for the Stream 1e series, all of it but Stream 1ee, which is the most-downstream stream, will be mined in the first year of the sequence, and Stream 1ee will be mined in the second year. However, Stream 1ee will be disrupted longer because a 200 foot- wide dragline access corridor runs across it, just upstream of the Heart-Shaped Wetland, as shown on Map H-1 and Figure RAI 514-1. Map H-9 is the Tailing Fill Schedule. The tailings are the sand tailings; the clay tailings, which are called waste clays, are deposited in the CSAs. Sand tailings are backfilled into mine cuts starting in year 3, and the process is completed in year 7. Map H-9 reproduces the blocks shown on Map H-1, except for one change in Section 20, and adds two years to each block. An explanatory note on Map H-9 states that IMC will backfill and grade the upland areas immediately west of the West Lobe and east of the East Lobe with sand tailings within one year of mining. Map I-6 is the Post-Reclamation Streams. This Recommended Order addresses streams in detail below. As already noted, at the hearing, DEP identified Stream 3e? as another stream eligible for restoration under the eligibility criterion used in these cases, and IMC has agreed to restore this stream and add it to Map I-6. Figure 13B-8 is the Post-Reclamation Connection Status of the reclaimed wetlands. A map, Figure 13B-8 depicts connected wetlands, isolated wetlands, isolated wetlands that are ephemeral, and cattle ponds. Figure 13A5-1 is the Identification of Created Wetlands. Also a map, Figure 13A5-1 assigns numbers to each reclaimed wetland and identifies the habitat to be reclaimed. These two figures provide a good basis for comparing the reclaimed wetlands to the existing wetlands by type, location, size, and proximity to streams. These two figures confirm the removal of cattle ponds to points considerable distances from Horse Creek, streams, riparian wetlands, or even most isolated wetlands. Thirteen cattle ponds totaling 7.6 acres will be reclaimed on OFG. Generally, these cattle ponds are located as far away as possible from the 100-year floodplain of Horse Creek. Except for the cattle ponds and three connected reclaimed wetlands that drain to the West Fork or Brushy Creek, all of the connected reclaimed wetlands will be connected to Horse Creek, usually by streams, but in several cases directly to the 100-year floodplain of Horse Creek. Connected reclaimed wetlands include the headwater and intermittent wetlands of the Stream 1e series (E003/E006/E007/E008/E009/E013/E015/E016), the headwater wetlands of Stream 3e (E022/E023/E024), and the headwater wetlands of Stream 3e? (E018/E019/E020). The decision at the hearing to reclaim Stream 3e? is not reflected on Figure 13A5-1 or 13B-8, which depicts as isolated the large wetland to the northeast of the headwater wetland of Stream 3e. The Stream 1e series reclaimed wetlands complex totals 44.9 acres. The Stream 1e series existing wetlands complex covers a smaller area, perhaps 10 fewer acres. However, the reclaimed wetlands will be somewhat simpler. IMC will reclaim one freshwater marsh (E006) where five presently exist (G108, G115, G125, G126, and G129). IMC will replace two gum swamps (G123 and G121) and two wetland forested mixed (G102 and G132) with the predominant mixed wetland hardwoods (E003). IMC will replace one of the freshwater marshes with hydric oak forest. Just west of the riparian corridor, IMC will replace a wet prairie (G119) with a little hydric flatwoods (G119A) with another freshwater marsh (E014) and will mine a small wet prairie (G028) to the east of the corridor and not replace it with any wetland. On the plus side, IMC will add two very small bayheads (E008--0.7 acres and E013--0.7 acres) to the west side of the corridor and will relocate and expand a large hydric flatwoods (G107) that is beside a small unreclaimed community--a hydric woodland pasture (G105). The reclamation of the headwater of Stream 3e better re-creates the existing wetlands, in size and type of community. The only change is the conversion of a shrub marsh (G134) in the center of the wetland to a freshwater marsh (E023), essentially enlarging the freshwater marsh (G135) presently in the center of this wetland. The size of the existing and reclaimed wetlands associated with the riparian corridor of Stream 3e and its headwater wetland appear to be the same. The reclamation of the headwater of Stream 3e? provides a more complicated complex of wetland communities than presently exists at that location. The ditch (G019) will be replaced with a natural stream, whose riparian corridor is not depicted due to the fact that IMC agreed to reclaim Stream 3e? at the hearing; however, the reclaimed wetland corridor undoubtedly will be more functional than the present ditch. Presently, the headwater wetland is a large freshwater marsh (G016) fringed by mixed wetland hardwoods (G014) and a wet prairie (G105). A cattle pond (G017) is in the wet prairie, and another cattle pond is at the point where Stream 3e? forms. The north side of this wetland is heavily ditched. The reclaimed headwater wetland, which will be about the same size as the present wetland, will consist of an interior shrub marsh (E019) and freshwater marsh (E020) and a wet prairie fringe (E018). A replacement cattle pond (E026) is moved farther away from the headwater wetland. Reclamation around the Heart-Shaped Wetland results in a more complicated array of wetlands than presently exists. Three ephemeral wet prairies (E021, E026, and E031) will be reclaimed north and west of the Heart-Shaped Wetland and Stream 2e where no wetland exists presently. An isolated freshwater marsh (E034) will be reclaimed south of the Heart-Shaped Wetland where no wetland exists today. Two ephemeral wet prairies (E026 and E037) totaling 4.5 acres will be reclaimed south and east of Stream 2e, close to the no-mine area surrounding Streams 6e and 7e, again where no wetland exists presently. However, IMC will not reclaim a hydric flatwoods (G157) connected to the south border of the headwater wetland of Stream 8e. Reclamation will relocate the headwater wet prairie of Stream 9w closer to Horse Creek. Mining two wet prairies (G047 and G048) and reclaiming them with a single wet prairie of at least the same size (W003--20.7 acres), IMC will also reclaim the downstream portion of Stream 9w with a mixed wetland hardwoods and add a gum swamp (W005--2.4 acres) at the end of Stream 9w, as it enters the no-mine corridor of Horse Creek. IMC will also reclaim an ephemeral wet prairie (W002) just north of the reclaimed segment of Stream 9w. Across Horse Creek from its junction with Stream 9w, IMC will mine the eastern half of a roughly five-acre bayhead (G166), reclaiming the mined part of the bayhead with a mixed wetland hardwoods (E048--6.0 acres). However, where no wetlands presently exist, IMC will reclaim an ephemeral wet prairie (E044) and a larger wetland consisting of a freshwater marsh (E047--9.0 acres) fringed by an ephemeral wet prairie (E046--7.1 acres). In RAI-173 in the CDA, IMC explains that no-mine lines initially ran through some wetlands due to the limited level of detail available in the small scale maps used at the time. IMC representatives have discussed each such bifurcation with DEP biologist Christine Keenan, and IMC made adjustments that satisfied DEP, obviously not eliminating all of the bifurcated wetlands. Alluding to the impracticability of eliminating all bifurcated wetlands, IMC notes in its response to the request for additional information: "A small feature protruding into a mining area is one of the more difficult features to effectively mine around. It requires significant extra distance of ditch and berm systems, which both increases costs and results in greater losses of phosphate ore recovery." Subject to two exceptions, the southernmost extent of reclaimed ephemeral wetlands will be close to the Lobes, especially the West and Central Lobes. Eight such wetlands (W021, W015, W017/W018, W019/W020, W012, W013, W016 and W011) will be west of Horse Creek, and three such wetlands will be east of Horse Creek (E057, E061, and E053). (Although the headwater wetland of Stream 7w, W012 is depicted as ephemeral in Figure 13B-8.) Most of these wetlands will be wet prairies. Three of these reclaimed ephemeral wetlands appear to be in the location of existing wetlands (G093/G094, G091/G092, and G090), and the existing wetlands are freshwater marshes fringed with wet prairies, except that the smallest, G090, is a wet prairie. The last reclaimed wetland on the east side of Horse Creek is just north of the Carlton cutout. In reclaiming Stream 5e, IMC will reclaim a small bayhead (E063--1.3 acres) in the middle of the stream's OFG segment. This replaces a wet prairie/hydric oak forest (G204/G205) in the same location and of the same size. On the other side of Horse Creek and to the south of Stream 5e, IMC will reclaim the headwater wetlands of Streams 5w, 4w, 3w, and 2w. The headwater wetland of Stream 5w is a long freshwater marsh (G210) with a small shrub marsh (G207) that drains an elaborate array of agricultural ditches to the west. These ditches shifted some of the drainage that historically entered Stream 4w into Stream 5w. Reclaiming the stream with a wider wetland forested mixed corridor, as it will do for Streams 4w, 3w, and 2w, IMC will expand the headwater wetland by reclaiming a long freshwater marsh (W024--7.9 acres) fringed on its upgradient side by a small wet prairie (W023--2.2 acres). IMC will also remove a cattle pond (G209) presently abutting the center of the freshwater marsh. IMC will reclaim an ephemeral wet prairie (W026) between Streams 5w and 4w, relatively close to the Horse Creek floodplain. Except for a very small ephemeral wet prairie just west of the headwater wetland of Stream 4w and an ephemeral, largely mixed wetland hardwoods reclaimed in the West Fork sub- basin (W041/W042/W043), W026 is the southernmost reclaimed ephemeral wetland on OFG. The pattern of the reclamation of Streams 4w, 3w, and 2w is otherwise identical: each reclaimed stream, in a reclaimed wetland forested mixed corridor, will receive water from reclaimed freshwater marshes of 3.5 to 5.1 acres in size. Presently, Stream 4w has no headwater marsh, instead receiving water from the elaborate ditching scheme described in connection with Stream 5w. Streams 3w and 2w presently receive water from small headwater wetlands, although Stream 2w also receives water from an agricultural ditch. The last major reclamation on the west side of Horse Creek relates to Stream 1w. Alone of all the streams, Stream 1w is an agricultural ditch throughout its length, except for a short segment just upstream from the no-mine area. However, alone of all the streams at OFG, Stream 1w drains a primarily seepage-supported wetland. This well-defined headwater wetland complex comprises, from upstream to downstream, a cattle pond (G505), freshwater marsh (G506), mixed wetland hardwoods (G507), bay swamp (G513), wetland forested mixed (G512), wet prairie (G514), hydric oak forest (G511), and ditch (G512A). Reclaimed, this headwater will be the largest reclaimed bay swamp (W0399-1.2 acres). In addition to the two small bay swamps in the wetland corridor of Stream 1e series, the small bay swamp in Stream 5e, and the Stream 1w headwater bay swamp, the only other bay swamp to be reclaimed on OFG will be a part of a wetland (W037/W036) that will be in the center of Section 19 and drain into the West Fork. The bay swamp component of this wetland will be 4.4 acres and will replace a similarly sized wetland (H008/H009/H009A) with a smaller bay swamp core. Map CL-1 is the Reclamation Schedule. This map identifies the year in which specific areas within OFG will be reclaimed. With two exceptions, Map CL-1 tracks Map H-9, which is the Tailing Fill Schedule, by identifying the same blocks and adding two years to each of them. One exception may be due to the February 19, 2004, and February 26, 2004, revisions of Map H-9. The latter revision changed the year of backfilling part of northwestern Section 20 from year 7 to year 5. Map CL-1 tracks the older version of Map H-9 and provides for reclamation of this area within Section 20 for year 9, not year 7. This means that part of the northwestern Section 20 would remain backfilled, but not revegetated, for four years. This may be an oversight in Map CL-1 because it was last revised January 22, 2004. The other exception concerns the uplands immediately east of the East Lobe. Map H-9 provides for sand tailings for the northern half of this area in year 6 and for the southern half of this area in year 5, but Map CL-1 provides for both areas to be reclaimed in year 7, so the southern half would remain backfilled, but not revegetated, for two years. This may be intentional, as ERP Specific Condition 12.d requires that IMC backfill and contour the two areas upslope of the bayheads in the West and East Lobes within one year after the completion of mining, but nothing in the ERP requires expedited revegetation of these upland areas. ERP Specific Condition 12.b requires IMC to include mining and reclamation schedule updates in the annual reclamation report that it files, pursuant to Chapter 62C-16, Florida Administrative Code. Specific Condition 12.b warns that "significant changes" to these schedules may require a permit modification. ERP Specific Condition 12.c states, in its entirety: "Mine cuts shall be oriented in the direction of ground water flow, generally perpendicular to Horse Creek as shown on Map H-1." The introduction to the January submittal, witnesses, and parties agree that IMC is required to orient the spoil piles in the direction of groundwater only to the extent practicable, so the unconditional language of ERP Special Condition 12.c is inadvertent. ERP Specific Condition 12.d provides that sand tailings placement and final contouring shall be completed within one year after the completion of mining, as shown on Map H-9, in the two areas upslope from the unmined bayheads (G178 and G197), which are in the East and West Lobes. ERP Specific Condition 13 addresses the construction, removal, and revegetation of the pipeline corridor shown on Figure RAI 514-1. This figure depicts a narrow "Mine Access Corridor (Pipelines, Road, Powerlines)" passing at the point that Stream 2e forms at the downgradient end of the Heart-Shaped Wetland. Specific Condition 13 contains seven subsections governing the pipeline corridor to minimize its impact on the wetlands and other surface waters that it crosses. Figure RAI 514-1 also depicts a 200-foot wide "Dragline Walkpath Corridor" that crosses Stream 1ee and Stream 3e within 100 feet of the Heart-Shaped Wetland. No conditions attach to the construction, operation, removal, and reclamation of this area because, unlike the pipeline corridor as it crosses Stream 2e, all of this portion of the dragline corridor will be mined. ERP Specific Condition 14 states that IMC shall restore as mitigation 322 acres of wetlands, as shown in Maps I-1, I-2, I-3, and I-6; Figure 13A5-1; and the post-reclamation cross-sections. Map I-1 is the Post Reclamation Topo. IMC updated this map with several limited changes at the end of the hearing, and DEP accepted the new Map I-1. Comparing Map I-1 with Map C-1, which is the Existing Topography, the post-mining topography substantially replicates the pre-mining topography, although Table 26M-1 reveals a lowering of some of the highest pre-mining elevations, including the highest elevation by eight feet. Maps I-2 and I-3 are, respectively, Post Reclamation Vegetation and Post Reclamation Soils. As noted above, Specific Condition 14 references these maps, but only in connection with the restoration of 322 acres of wetlands. Maps I-2 and I-3 cover all of OFG, so they cover wetlands and other surface waters, which are properly the subject of an ERP, and uplands, which are properly the subject of a CRP approval. Naturally, the ERP does not incorporate the all of Maps I-2 and I-3 because they include all of the uplands. Unfortunately, as discussed in the next section, the CRP approval likewise fails to obligate IMC to reclaim the uplands in accordance with Map I-2 and the upland soils in accordance with Map I-3. This omission is inadvertent, so the Recommended Order will assume that IMC will reclaim the uplands as depicted in Map I-2 and the upland soils as depicted in Map I-3. Although the upland portions of Maps I-2 and I-3 should be discussed in the next section, they will be discussed in this section because the CRP approval fails to incorporate them and discussing both maps in one place allows for a more coherent presentation. Map I-2 is the Post Reclamation Vegetation. Map I-2 depicts the post-reclamation upland and wetland vegetation on OFG. This map reveals wide edges of roughly one-quarter to one- half mile of reclaimed improved pasture on the east and west edges of OFG. The core of OFG is Horse Creek and its 100-year floodplain, which are always within, but do not always define, the no-mine area. Between the no-mine area and the reclaimed improved pasture are the reclaimed wetlands described above and larger area of reclaimed uplands described below. Map I-2 and Map F-1, which is Pre Mining Vegetation, allow a comparison, by community, location, and area, of reclaimed uplands with existing uplands. In broad overview, IMC will reclaim everything in Section 4 outside the Heart-Shaped Wetland, which is the northernmost extent of the no-mine area, and Stream 2e. From the point that Horse Creek enters OFG, IMC will reclaim a broad area between the no-mine area and reclaimed improved pasture, south to the Carlton cutout. From this point, reclamation will be limited to the west side of Horse Creek, and the area between the no-mine area and reclaimed improved pasture will narrow progressively for the remaining 1 1/2 miles that Horse Creek runs in OFG. The width of the core, or no-mine area, is generally about 750 feet, but widens considerably at different points. Where Horse Creek enters OFG, the no-mine area is approximately 1750 feet wide, but narrows south of Stream 8e to about 750 feet. From the Central Lobe to the East Lobe, the no-mine area expands to nearly 4000 feet across. Except for another expansion at the West Lobe, the width of the no-mine area south of the Lobes remains at about 750 feet until Horse Creek exits OFG. The riparian wetlands of Horse Creek, which are within the no-mine area, are mixed wetland hardwoods for the first mile that Horse Creek flows in OFG and hydric oak forest for the remainder of Horse Creek's passage through OFG. The width of the non-pasture uplands adjacent to the no-mine area also varies. In describing the width of these upland areas between the no-mine area and the reclaimed improved pasture, this Recommended Order will include the reclaimed wetlands described above. These wetland areas are small, except for the headwater wet prairie of Stream 9w, the headwater freshwater marshes of Streams 5w, 4w, 3w, and 2w, and a few isolated wetlands. On both sides of Stream 2e, IMC will reclaim a band of hardwood conifer mixed of about one-half mile in width. At present, this area is occupied by a smaller area of hardwood conifer mixed and nearly a one-half mile wide band of pine flatwoods or, to the south, pine flatwoods and sand live oak. East of Streams 6e, 7e, and 8e, IMC will reclaim a band 1500-3000 feet wide of hardwood conifer mixed, shrub and brushland, and sand live oak, between the no-mine area and the reclaimed improved pasture. This replaces a broader area of pine flatwoods, sand live oak, palmetto prairie, and xeric oak. From Stream 8e south, IMC will reclaim uplands on both sides of Horse Creek. At this point, the reclaimed area between the no-mine area and the reclaimed improved pastures measures about 1750 feet wide on the west of Horse Creek and about 2000 feet wide on the east of Horse Creek. Including the no-mine area in the center, these reclaimed areas average about one-mile wide south to the Lobes. From Stream 8e south to the East Lobe, IMC will reclaim largely hardwood conifer mixed. This replaces a large citrus grove, a larger area of improved pasture, and three smaller areas of palmetto prairie. On the west side of Horse Creek, the vegetation is more varied, both at present and as reclaimed. North of Stream 9w, IMC will reclaim a large palmetto prairie, a sizeable area of sand live oak, and a small area of temperate hardwood. South of Stream 9w, IMC will reclaim a large area of hardwood conifer mixed, areas of pine flatwoods, sand live oak, and palmetto prairie, and a small area of temperate hardwood. The uplands surrounding Stream 9w presently consist of improved pasture along the downstream half of the conveyance and palmetto prairie and sand live oak along and near its upstream reach. South of Stream 9w are a large area of improved pasture, pine flatwoods, and sand live oak and two smaller areas of palmetto prairie. The combination of no-mine area and reclaimed area, exclusive of reclaimed improved pasture, attains its greatest width--about 10,000 feet--from the western edge of the West Lobe to the eastern edge of the East Lobe, although this includes a 1000-foot strip of improved pasture between the bayhead in the East Lobe and sand live oak east of the bayhead. This area narrows to less than 6000 feet, just north of the Carlton cutout. South of this point, at which the reclaimed upland habitat will be found only on the west side of Horse Creek, the total width of the no-mine area and reclaimed area east of the reclaimed improved pasture tapers down from a little over 3000 feet to less than 1500 feet at the south end of OFG. Map I-2 also discloses the communities or habitats that will exist, post-reclamation, on OFG. These communities or habitats include those that will be in the no-mine area and those that will be reclaimed. At present, the West Lobe is mostly bayhead, wet prairie, and wetland forested mixed with smaller areas of hydric woodland pasture and shrub marsh. The West Lobe also includes upland communities of palmetto prairie, temperate hardwoods, and pine flatwoods. A large wet prairie extends from the northwest corner of the West Lobe. IMC will reclaim this wet prairie as improved pasture with a small strip of hardwood-conifer mixed. To the west of the West Lobe is a small strip of improved pasture and a large area of hardwood-conifer mixed. IMC will reclaim the improved pasture with hardwood-conifer mixed and sand live oak and most of the hardwood-conifer mixed with sand live oak. The areas surrounding the no-mine area associated with Stream 6w are currently improved pasture; IMC will reclaim these areas as hardwood-conifer mixed. The Central Lobe is mostly bayhead with small areas of wetland forested mixed and wet prairie. Palmetto prairie is also within the Central Lobe, nearer to Horse Creek. IMC will reclaim the areas around the Central Lobe and Stream 7w with hardwood-conifer mixed and some palmetto prairie. At present, the Central Lobe and Stream 7w are surrounded by palmetto prairie and some pine flatwoods with an area of sand live oak to the northwest of the Central Lobe. Unlike the no-mine areas forming the West and Central Lobes, which incorporate insubstantial areas of uplands, the no- mine area forming the East Lobe, like the no-mine area around Streams 6e, 7e, and 8e, incorporates a substantial area of uplands. Upgradient of the large bayhead forming the western half of the East Lobe is the 1000-foot strip of improved pasture, and upgradient of the pasture is a large sand live oak area. IMC will mine the eastern half of this sand live oak area and reclaim it as xeric oak. IMC will mine a small wet prairie presently at the southern tip of the bayhead in the East Lobe and reclaim the area as hardwood-conifer mixed. From the East Lobe south to the Carlton cutout, the reclaimed uplands will consist of a long area of temperate hardwoods abutting the no-mine area and a wider area of hardwood-conifer mixed abutting the temperate hardwoods. This area is presently improved pasture. On the west side of Horse Creek, south of the Carlton cutout, the area outside the no-mine area is presently improved pasture, except for a large palmetto prairie around and south of the headwater wetland of Stream 1w. Between the no-mine area and reclaimed improved pasture, IMC will reclaim palmetto prairie and a small area of hardwood-conifer mixed between the headwater wetlands of Streams 5w and 3w. Map I-3 is the Post Reclamation Soils. The legend classifies the soils by "[moderately well-drained]--greater than 30"; "[poorly drained]--greater than 30"; "[poorly drained]-- less than 30"; "[poorly drained]--stream"; "[very poorly drained]--muck"; and "[very poorly drained--mineral depression]." The references to "30" are the thicknesses, in inches, of sand tailings over overburden. Maps E-1 and E-2 are, respectively, Detailed Existing Soils and General Existing Soils. Comparisons between these two maps, on the one hand, and Map I-3, on the other hand, reveal specifics of the soil-reclamation process. The most distinctive feature of soils present at OFG is the thin band of Felda Fine Sand, Frequently Flooded, that runs down the center of OFG. As always, this reinforces the most distinctive feature of OFG--Horse Creek. However, the Felda Fine Sand extends beyond the Horse Creek floodplains to Stream 2e, the Stream 1e series, and the headwater wetland of Stream 5w. All of these soils are in the no-mine area except at the Stream 1e series and headwater wetland of Stream 5w. A closely related soil underlies the floodplain of the lower end of Stream 6w, which is also in the no-mine area. These are the only locations on OFG with these soils. The Felda Fine Sand is a "poorly drained soil having layers of loamy and/or spodic materials underlying sandy surfaces at least 20 inches thick on streams terraces and floodplains." Exclusive of the loamy or spodic materials, Map I-3 shows that IMC will reclaim the drainage characteristics of this type of soil at the Stream 1e series, but not at the headwater wetland of Stream 5w. IMC will also reclaim this type of soil at Streams 9w, 5w, 4w, 3w, 2w, and 1w. Another distinctive soil, pre-mining, is "moderately well to excessively drained soils having layers of loamy and/or spodic materials underlying sandy surfaces greater than 30 inches thick on gentle upland slopes and rises." Except for a couple of areas at the eastern end of the East Lobe, these soils presently are all outside of the no-mine area. IMC will reclaim these soils, generally in the areas previously described as sand live oak or xeric oak, as well as in a long band along the southern border of the slough associated with Stream 9w and a large area on the west sides of Sections 29 and 20. These areas correspond reasonably well in area and location to the existing soils with the same drainage characteristics. The two most poorly drained soils, pre-mining, are "very poorly drained to poorly drained mineral soils in depressions" and "very poorly drained soils with organic surfaces on low gradient seepage slopes." The latter are exclusively mucky soils, and the former range from mucky fine sand to fine sand. Most of the mucky soils are in the no-mine area, such as in each of the Lobes and along Streams 6e and 7e. IMC will not reclaim with similar soils the three areas with these mucky soils that are outside the no-mine area. The mucky fine soils are more widely distributed outside the no-mine area. The only significant areas of fine mucky sand presently at OFG underlie the Heart-Shaped Wetland, the headwater wetland of Stream 8e, and parts of the West Lobe. IMC will reclaim these mucky fine soils generally in accordance with their present areas and locations. The most significant reductions in area are from the slough of Stream 9w and the northeast corner of Section 4. Except for another category of poorly drained soil and four small areas of a somewhat poorly drained soil--all within the no-mine area--the remaining soil is "poorly drained soils having layers of loamy and/or spodic materials underlying sandy surfaces predominantly greater than 30 inches thick primarily on gently sloping uplands." The reclaimed counterpart of this poorly drained soil occupies the largest part of OFG, post-reclamation. This represents a substantial expansion of coverage of this type of soil, mostly at the expense of "poorly drained soils having layers of loamy and/or spodic materials underlying sand surfaces less than 30 inches thick primarily on gently sloping uplands." Map I-6 is the Post Reclamation Streams. These are addressed below. Figure 13A5-1 is the Identification of Created Wetlands. These wetlands have already been discussed. ERP Specific Condition 14 states that IMC shall reclaim wetlands in accordance with the schedule contained in Table 3AI-6A, which has been discussed. Specific Condition 14 lists various requirements applicable to the wetlands that IMC will create. ERP Specific Condition 14.a requires IMC to remove "suitable topsoil" prior to mining wetlands. IMC must time the clearing of topsoil donor sites and reclaiming of other sites so that it optimizes the opportunities for the direct transfer of topsoil, without any intervening storage time. If IMC must remove wetland topsoil more than six months before it will be spread at a reclamation site, IMC must store the topsoil in such a way as to minimize oxidation and colonization by nuisance species. Specific Condition 14.a encourages IMC to relocate any endangered or threatened plant species to appropriate mitigation sites. ERP Specific Condition 14.b requires IMC to grade reclaimed forested wetland areas after backfilling them with sand tailings and/or overburden and cap them with "several inches of wetland topsoil." IMC shall use direct transfer of topsoil and live materials, such as stumps, shrubs, and small trees, where feasible. However, Specific Condition 14.b states in boldface: "All reclaimed bay swamps shall receive several inches of muck directly transferred from forested wetlands approved for mining." Specific Condition 14.b provides that wetland topsoil should be reasonably free of nuisance and exotic plant species before application to wetland mitigation areas. ERP Specific Condition 14.c requires IMC to grade reclaimed herbaceous and shrub marsh wetland areas after backfilling them with sand tailings and/or overburden and cap them with "several inches of wetland topsoil when available." Specific Condition 14.c provides that wetland topsoil should be reasonably free of nuisance and exotic plant species before application to wetland mitigation areas. ERP Specific Condition 14.d requires IMC to design marshes and wet prairies "to maintain the diversity of community types that existed prior to mining in order to support a wide range of wildlife species including birds, reptiles, and amphibians." Specific Condition 14.d requires IMC to reclaim marshes and wet prairies with variations in hydroperiod and slope "to provide the greatest diversity of available habitat," with marsh hydroperiods ranging from ephemeral through permanently flooded. Specifying a range of slope values, Specific Condition 14.d adds that most marshes shall have slopes gradual enough to support wide transition zones with a diversity of vegetation. ERP Specific Condition 14.d provides that IMC shall construct ephemeral marshes and wet prairies as identified in Figure 13B-8, which, discussed above, addresses the status of individual wetlands as connected, isolated, or isolated and ephemeral. Although not incorporated into the ERP, Table 13A1-4 indicates that IMC will mine 27 of the 29 ephemeral wetlands or 22 of the 27 acres of ephemeral wetlands, but will reclaim 44 ephemeral wetlands totaling 101 acres, as indicated on Table 13A5-1 2AI discussed above. ERP Specific Condition 14.e provides that at least half of all herbaceous and shrub marshes shall be rim mulched with several inches of wet prairie, pine flatwoods, or palmetto prairie topsoil, and IMC shall use direct transfer, where feasible. ERP Specific Condition 14.f requires IMC to use "several inches" of wet prairie, hydric pine flatwoods, or hydric palmetto prairie topsoil for all wet prairie and hydric palmetto prairie areas, and IMC shall use direct transfer, where feasible. However, instead of topsoiling, IMC may use "[o]ther innovative methods" that are likely to produce the same diversity of wet prairie forbs and grasses. ERP Specific Condition 14.g requires IMC to construct, in forested wetlands, hummocks several inches above the wet-season high water line. The hummocks shall be 8-12 feet long and 3-6 feet wide. To increase habitat heterogeneity, IMC shall place brushpiles, logs, and tree stumps in the reclaimed area, which it shall roughly grade in some areas. ERP Specific Condition 14.h requires IMC to construct streams in accordance with the Stream Restoration Plan. Specific Condition 14.h also requires IMC to employ an experienced stream restoration scientist, subject to BMR approval, to provide project oversight and conduct regular inspections during construction and planting. First appearing in the January submittal, the Stream Restoration Plan is a design document that specifies, in detail, the physical characteristics of each reclaimed stream. For each reclaimed stream or stream segment, the Stream Restoration Plan provides detailed information of physical structure; channel planform or shape; hydrologic characteristics in terms of such factors as storage, conveyance, and attenuation; geomorphic characteristics such as the substrate and floodplain soil types and the effects of flows upon these materials; vegetation along the stream corridor, including the addition of snags and debris dams to re-create natural microhabitats; construction supervision; and monitoring. The Stream Restoration Plan focuses upon the design of the basin, reach, and microhabitat of each reclaimed stream. For microhabitat, the Stream Restoration Plan promises that: the ecology of most of the reaches is expected to be improved through reclamation. For all reaches except 1e and 3e (which are wholly situated in generally native land cover), the forested riparian zone will be substantially increased since improved pasture adjacent to the stream channels will [be] replaced with forested canopy. Acknowledging the importance of small headwater streams to the overall integrity of a large watershed, the Stream Restoration Plan recognizes the hydrological and biological functions of the tributaries and their riparian wetlands--namely, flood conveyance, attenuation, and storage and aquatic and wetland habitat. Among other things, the Stream Restoration Plan repeatedly stresses the importance of achieving "rapid closure of the riparian canopies." In addition to providing habitat, a riparian canopy reduces solar heating of the stream, thus lowering the water temperature and minimizing weedy vegetation on the stream banks. Among the effects of lowering the water temperature is lowering the amount of water lost to evaporation. The installation of trees along and sometimes within the reclaimed channels will facilitate the rapid development of root systems to stabilize the substrate and provide submerged root structure, which is an important microhabitat for macroinvertebrates and fish. Mature trees in the floodplain also provide additional attenuation. In addition to serving as a design document to govern the reclamation of mined streams on OFG, the Stream Restoration Plan is also a descriptive document, detailing the relevant characteristics of the streams presently at OFG. The Stream Restoration Plan uses several classifications that are useful in analyzing streams and their functions. These classifications include the Rosgen classification of stream shape (the Rosgen classification of bottom sediment is irrelevant because all existing and reclaimed streams at OFG have sandy bottoms), the Strahler convention of stream orders, the duration of flow, and the channel morphology. The Rosgen classification of stream shape divides the streams at OFG into type E and type C. Type E streams are well- incised and hydraulically efficient; their width-to-depth ratios are less than 12:1. Shallower and wider than type E streams, as these values relate to each other, type C streams at OFG are often associated with small wetland riparian zones and depressions, which are absent from type E streams at OFG. The Strahler convention classifies streams based on their relative location in the upstream order of conveyances with the most-upstream streams classified as first-order streams. Except for Stream 2e and the Stream 1e series downstream of Streams 1eb and 1ef, all of the tributary streams on OFG are first-order streams, meaning essentially that they are the most upstream channelized conveyance receiving runoff or groundwater flow. Streams 2e, 1ec, 1ed, and 1ee are second- order streams, meaning that they receive flow from at least two first-order streams. In terms of flow, perennial streams receive groundwater flow throughout the year in most years, ephemeral streams flow sporadically in response to rain and typically lack groundwater inputs, and intermittent streams flow during the wet season in response to groundwater and rain inputs and during the dry season sporadically in response to rain inputs only. Most, if not all, of the tributary streams on OFG are intermittent. However, almost all of the streams cease to flow due to low rainfall and overflow their banks due to very high rainfall. Even Horse Creek dried up at State Road 64 during the low-rain conditions in 2000. In terms of morphology, all streams at OFG are either in uninterrupted channels or interrupted channels. Interrupted channels mean that the stream passes through flow-through marshes and swamps. Describing the existing streams in a slightly larger setting, the Stream Restoration Plan divides them into three groups, based on channel morphology and the vegetation and land uses adjacent to the channel. First, Streams 3e and 1e series are "surrounded by native habitat used for low-intensity cattle grazing. These are type C streams with a more diffuse riparian canopy and associated wetlands along the stream channel." Second, the portions of Streams 5e, 1w, 2w, 3w, 4w, 5w, 7w, and 9w within the floodplain forest of Horse Creek are type E streams with oaks and palmettos along, and often crowding, the channel. Third, the portions of the same eight streams that are outside of the floodplain forest of Horse Creek are type E streams, devoid of riparian vegetation and degraded by agricultural land uses, such as improved pasture and cattle grazing. The Stream Restoration Plan describes the Stream 1e series as follows: Reach 1e provides drainage for a series of interconnected flow-through wetlands punctuated by five relatively short stream segments. The segments represent a total of some 2,039 linear feet of channel. They have shallow, sandy banks with little vegetation in the stream channel. A wide riparian canopy of slash pine, laurel oak, dahoon holly and wax myrtle is present along most of this reach. The palmetto edge of the floodplain varies in width, but is generally more than 100 feet from either bank, suggesting frequent inundation. The channel substrate is sandy except where near a swamp, where it becomes increasingly more organic. Each flow-through wetland occurs in shallow depressions which overflow into C-type channels that are typically several hundred feet long. Key components of this conveyance type include the lip elevation at which wetland flow enters the channel and the elevation at which the streams dissipate their discharge to the downstream flow- through wetland. Most of the stream segments in this conveyance system appear to be in good geomorphic condition. Most of these channels typically have wetland and/or upland hardwood trees in the riparian zone with little understory. The Stream Restoration Plan reports that the channel of Stream 3e is in good geomorphic condition. The upper part of the channel flows through a scattered open canopy of trees with herbaceous cover in the riparian zone. The lower part of the channel mostly flows through treeless banks lined with palmettos. The channel has vegetation in it where it is exposed to sunlight. In other respects, Stream 3e is like Stream 1e series, except that the channel is uninterrupted and shorter. The length of Stream 3e is 611-630 feet. Stream 1eb is 486 feet, Stream 1ef is 223 feet, Stream 1ec is 315 feet, Stream 1ed is 283 feet, and Stream 1ee is 732 feet. The 2039-foot length of the Stream 1e series is exclusive of the system's headwater and flow-through wetlands. The Stream 1e series has the most linear feet of any tributary stream on OFG. In addition to the Stream 1e series and Stream 3e, the only other stream on the east side of Horse Creek to be mined is Stream 5e, which is an agriculturally disturbed stream with a narrow riparian canopy. The Stream Restoration Plan states that the lower portion of Stream 5e, which is within OFG, is in better condition than the upper portion, which is frequented by cattle and leads to a cattle pond and agriculturally altered wetland. However, in contrast to the Stream 1e series and Streams 6e, 7e, and 8e, Stream 5e is isolated in a vast monocommunity of improved pasture. The streams on the west side of Horse Creek have all been impacted by agricultural practices, mostly cattle ranching, ditching streams, sloughs, and other wetlands, and excavating cattle ponds in wetlands. The only streams entirely in the no- mine area on the west side of Horse Creek are Streams 8w and 6w, which are part of the Central and West Lobes, respectively. Relative to their surrounding communities, the streams on the west side of Horse Creek fall into three groups. Streams 6w and 8w are integrated into diverse communities of uplands and wetlands. Like Stream 5e, Streams 5w, 4w, 3w, and 2w are lonely departures from the monocommunity of improved pasture and, thus, attractors of thirsty or hot cattle. All of these streams have been impacted, to varying degrees, by ditching, which, with cattle disturbances, has led to unstable banks and erosion. Functionally, Streams 9w, 7w, and 1w are between these two groups. As a stream, Stream 9w is surrounded by improved pasture; however, it drains a large wet prairie surrounded by large areas of palmetto prairie to the south and west and sand live oak to the north and east. Prior to agricultural disturbance, Stream 9w was much higher functioning, at least with respect to flood conveyance, attenuation, and storage. At one time, this stream led upgradient to a long slough. After the slough was ditched to hasten drainage, the channel of Stream 9w suffered from excessive hydraulic forces, resulting in bank instability and a curious channel formation that fits the type E stream, even though the valley slope is consistent with other type C streams at OFG. Stream 9w is the second-shortest stream on OFG at 472 feet. Draining the smallest area of all tributaries on OFG (30 acres), Stream 7w lies between a large palmetto prairie to the north and improved pasture to the south. Stream 7w is the shortest stream on OFG at 456 feet. Stream 7w's upper section is characterized by unstable banks vegetated by pasture grasses. Stream 1w runs from Horse Creek through improved pasture, but enters a large palmetto prairie before draining a wetland that includes a relatively small bayhead. The upper half and extreme lower portions are in good condition with appropriate vegetation, but the channel is eroded in areas where it runs through pasture. IMC will reclaim the headwater wetland of Stream 1w with a large bayhead. ERP Specific Condition 14.i requires IMC to survey the final contours of each mitigation wetland to the precision of a one-foot contour. Within 60 days of final grading, IMC shall submit to BMR, for its approval, a topographic map and representative cross sections for each wetland and extending at least 200 feet into the adjacent uplands. IMC must also submit surveyed profiles and cross sections for all reclaimed streams. All topographic maps must meet the minimum technical standards of Chapter 472, Florida Statutes. ERP Specific Condition 14.j states that IMC shall assess the hydrology of the modeled wetlands through the installation of monitoring wells and staff gauges at mutually agreed-upon sites in these reclaimed wetlands. For at least two years after the final contouring of each wetland, IMC shall monitor the hydrology for the parameters listed in Table MR-2, which is described below. IMC shall submit the analysis to BMR within 30 days of its completion. If BMR does not approve the hydrology, IMC shall have 60 days to submit a remedial plan. ERP Specific Condition 14.k requires that freshwater marsh and ephemeral marsh vegetation shall develop from direct placement of donor topsoil or planting of herbaceous marsh species in the densities and numbers specified in the Freshwater Marsh and Wet Prairie/Ephemeral Marsh planting tables, so as to meet the requirements of ERP Specific Condition 16. Both tables require plantings on three-foot centers, or 4840 plants per acre, and specify suitable water levels for each species. The Freshwater Marsh planting table lists 22 approved species, and the Wet Prairie/Ephemeral Marsh planting table lists 35 approved species. ERP Specific Condition 14.l requires IMC to plant the uplands surrounding wet prairies with collected native grass seed, such as creeping bluestem, sand cordgrass, blue maidencane, bluestem, lovegrass, and eastern gamma grass, to prevent invasion by non-native or range grasses. ERP Specific Condition 14.m provides that IMC shall develop shrub marsh vegetation by directly placing donor topsoil at the location of the reclaimed shrub marsh and planting herbaceous and shrub marsh species in the densities and numbers specified in the Shrub Marsh planting table, so as to meet the requirements of ERP Specific Condition 16. The Shrub Marsh planting table requires IMC to plant herbaceous species on three-foot centers, or 4840 plants per acre, and shrub species at an average density of 900 plants per acre. The planting table lists 18 approved species and requires IMC to plant at least five different shrub species. The planting table also specifies suitable water levels. ERP Specific Condition 14.n provides that IMC shall plant forested wetlands in the densities, species richness, and dominance specified in the Bay swamp/Gumswamp/Hydric Oak Forest/Wet Pine Flatwoods/Mixed Wetland Hardwood/Mixed Forest Swamp, "as appropriate for each community type" to meet the requirements of ERP Specific Condition 16. IMC shall plant appropriate species based on the design elevations, hydrology monitoring, and mitigation goals. ERP Specific Condition 14.o provides that IMC shall plant shade-tolerant herbaceous species after establishing suitable shade, by year 7, in hardwood swamps, mixed forest swamps, and bay and gum swamps. Specific Condition 14.o states: "At least 5 of the species listed in the Tables in n above and others like goldenclub . . . and swamp lily . . . shall be planted." The items listed in Specific Condition 14.n, however, are communities, not species. ERP Specific Condition 15 requires IMC to implement a monitoring and maintenance program to promote the survivorship and growth of desirable species in all mitigation areas. ERP Specific Condition 15.a requires IMC to conduct "quarterly or semi-annual" inspections of wetlands for nuisance and exotic species. IMC shall control these species by herbicide, fire, hydrological, or mechanical means "to limit cover of nuisance species to less than ten (10) percent and to remove exotic species when present in each created wetland." IMC must annually use manual or chemical treatment of nuisance and exotic species when their cover in any area of at least one acre is greater than ten percent or any exotic species are present. IMC must use manual or chemical treatment if cogongrass covers more than five percent within 300 feet of any reclaimed wetland. ERP Specific Condition 15.b allows IMC to control water levels with outflow control structures and pumps, as needed to enhance the survivorship and growth of sensitive taxa. However, IMC must remove all water management structures at least two years prior to requesting release. ERP Specific Condition 15.c requires IMC to make supplemental tree and shrub plantings, pursuant to Specific Condition 14, when tree/shrub densities fall below those required in ERP Specific Condition 16. Specific Condition 15.d requires IMC to make supplemental herbaceous plantings, pursuant to ERP Specific Condition 14, when cover by a "diversity of non- nuisance, non-exotic wetland species as listed in Chapter 62-340.450, F.A.C.," falls below that required in ERP Specific Condition 16. ERP Specific Condition 16 provides the conditions for DEP to release IMC of further obligation for reclaimed wetlands. DEP shall release the 105 acres of reclaimed forested wetlands and 217 acres of herbaceous wetlands when IMC has constructed them in accordance with the ERP requirements; IMC has not intervened, for two consecutive years (absent BMR approval), by irrigating, dewatering, or replanting desirable vegetation; and the remaining requirements of ERP Specific Condition 16 have been met. IMC must indicate in its annual narrative, which is required by Specific Condition 5, the start date for the non- intervention period. ERP Specific Condition 16.A requires that the water quality meet Class III standards, as described in Florida Administrative Code Chapter 62-302. ERP Specific Condition 16.B addresses water quantity. ERP Specific Condition 16.B.1 requires each created wetland to have hydroperiods and inundation depths sufficient to support wetland vegetation and within the range of conditions occurring in the reference wetlands of the same community for the same period, based on the monitoring data developed in accordance with ERP Specific Condition 14.j. Tributary wetlands must have seasonal flow patterns similar to specified reference wetlands for the same period. ERP Specific Condition 16.B.2 states that IMC modeled 24 representative reclaimed wetlands that IMC has modeled during the application process to predict subsurface conditions after excavation and backfilling. Figure 13-3 depicts these modeled wetlands, which are within 13 wetland complexes, and the proposed transects. All of the modeling transects are aligned east-west, which is the direction of groundflow. As discussed in detail below, the primary hydrological model used by Dr. Garlanger requires an input for the length of the upland in terms of the distance from the basin divide to the riparian wetland. Therefore, the transects probably must run in the direction of groundwater flow. Absent an ability to model the hydroperiod and inundation depth of a wetland across a sand tailings valley and cast overburden plateau--i.e., in a north-south direction-- multiple east-west transects in wetlands with long north-south dimensions would better reveal whether the wetland design were adequately accounting for the alternating pattern of sand tailings valleys and cast overburden plateaus. For all the areas for which Map H-1 provides probable orientations of spoil piles--basically, for present purposes, everywhere but Section 4--the spoil piles are oriented in the same alignment as the transects, so the transects will not cross the sand tailing valleys/cast overburden peaks. In other words, each of the transects will run along the portion of each wetland for which the relative depths of sand tailings and cast overburden remain constant, avoiding the potentially more problematic situation of alternating rows of sand tailing valley and cast overburden peak. As noted below, the north-south dimension of W039 assures that one cast overburden spoil pile and part of another will underlie W039. The north-south dimensions of W003 and E046/E047 also are long enough to guarantee significant alterations in geology. ERP Specific Condition 16.B.2 requires that, prior to the construction of the modeled 24 wetlands, IMC shall reassess and, if necessary, modify their design. The modifications shall be based on the targeted hydroperiods and inundation depths set forth in Table 1, which is described below, and updated analysis from an "integrated surface and ground water model that has been calibrated to actual field conditions at the location of the wetland to be constructed." Lastly, ERP Specific Condition 16.B.2 requires IMC to use a similarly calibrated model to design the other reclaimed wetland, so that they achieve the targeted hydroperiods and inundation depths set forth in Table 1. For the 24 modeled wetlands, Table 1 identifies eight types of wetland community, prescribes hydroperiods and inundation depths for each wetland habitat, and projects a hydroperiod for each of the 24 modeled wetlands. As amended at the hearing for bay swamp hydroperiods, the hydroperiods and inundation depths for the wetland communities are: bay swamps-- 8-11 months with inundation depths of 0-6 inches; gum swamps-- 3-12 months with inundation depths of 0-12 inches; mixed wetland hardwoods and wetland forested mix--3-9 months with inundation depths of 0-6 inches; hydric pine flatwoods--1.5-4.5 months with inundation depths of 0-6 inches; freshwater marshes--7-12 months with inundation depths of 6-30 inches; wet prairies--2-8 months with inundation depths of 0-6 inches; and shrub marshes--7-12 months with inundation depths of 6-24 inches. The 24 reclaimed wetlands to be modeled include three bay swamps: W039, which is the headwater wetland of Stream 1w; E008, which is a small part of the wetland into which Streams 1eb and 1ef drain; and E063, which is a small bay swamp in the middle of Stream 5e. The only other bay swamps to be reclaimed are E007, which is a small part of the wetland into which Stream 1ec drains, and W036, which is in the center of Section 19 and drains offsite into West Fork. The only other modeled wetlands that are part of the riparian wetlands of Stream 1e series are E007 and E009, which are near E008 and are the only hydric pine flatwoods to be modeled. The only other hydric pine flatwoods to be reclaimed is E015, which is also part of the riparian wetlands of Stream 1e series. Other modeled wetlands of particular importance are W003, which will be a large wet prairie wetland serving as the headwater wetland of Stream 9w; W031, which will be the freshwater marsh serving as the headwater wetland of Stream 3w; E018, E046, and E057, which are wet prairie fringes; E018, E042, E046, and E057, which are ephemeral wetlands (E042 is the only modeled ephemeral wet prairie that is not a fringe wetland); and all of the connected wetlands of Streams 3e and 3e?: E024, which is a wetland forested mix that is the riparian wetland along Stream 3e; E023, which is a freshwater marsh immediately upstream of E024; E022, which is a mixed wetland hardwoods joining the upstream side of E023; E018, which is a wet prairie fringing the headwater wetland of Stream 3e?; E019, which is a shrub marsh (the only modeled shrub marsh) fringed by E018; and E020, which is a freshwater marsh joining E019 and also fringed by E018. ERP Specific Condition 16.B.3 states the IMC shall monitor the 24 modeled wetlands, as prescribed by ERP Monitoring Required Section D and Table MR-2, which are discussed below. ERP Specific Condition 16.B.4 requires that the ephemeral wetlands shall remain inundated no more than eight months per year during a normal water year, which is between the 20th and 80th percentiles of historical record in terms of total rainfall and major storm occurrence. ERP Specific Conditions 16.C.1 and 2 apply to all mitigation areas within the scope of the ERP. Specific Condition 16.C.1 requires that non-nuisance, non-exotic wetland species listed in Florida Administrative Code Rule 62-340.450 cover at least 80 percent of the groundcover or attain the range of values documented in specific reference wetlands of the target community. Desirable groundcover plant species must be reproducing naturally. ERP Specific Condition 16.C.2 provides that nuisance vegetation species, such as cattail, primrose willow, and climbing hemp vine, shall cover less than 10 percent of the total wetland area. Invasive exotic species, such as melaleuca, Chinese tallow, and Brazilian pepper, shall not be considered as an acceptable component of the vegetative community. For herbaceous marshes, ERP Specific Condition 16.C requires that native species typical of the reference marshes dominate the cover and that they be distributed in zonation patterns similar to reference marshes. Species richness and dominance regimes shall be within the range of values documented within the reference marshes. For wet prairies, ERP Specific Condition 16.C requires that native species typical of the reference wet prairies dominate the cover. Species richness and dominance regimes shall be within the range of values documented within the reference wet prairies. Range grasses, such as bahiagrass and Bermuda grass, shall cover, in total, less than 10 percent of the wet prairie. For shrub marshes, ERP Specific Condition 16.C requires that native species typical of the reference shrub marshes dominate the cover. Carolina willow and wax myrtle shall cover, in total, less than 30 percent of the marsh. For all forested wetlands, ERP Specific Condition provides that the forested canopy shall have an average of at least 400 live trees per acre that are at least 12 feet tall, except for cabbage palms, which shall have a leaf, including the stalk, that is at least three feet long. In the alternative, the forested canopy shall meet or exceed the range of canopy and sub-canopy tree densities in specified reference wetlands. No area greater than an acre shall have less than 200 trees per acre. Hydric pine flatwoods shall average 50 trees per acre. For all forested wetlands, ERP Specific Condition provides that the shrub layer shall average at least 100 shrubs per acre or shall meet or exceed the range of shrub densities in specified reference wetlands. Early successional species, such as Carolina willow, saltbush, and wax myrtle, do not count in meeting this density requirement, but the monitoring reports shall include such species. Hydric pine flatwoods shall have an average density of 350 shrubs per acre, and the primary species shall be typical of hydric pine flatwoods, such as saw palmetto, gallberry, and fetterbush. For all forested wetlands, ERP Specific Condition states that the canopy and shrub strata shall each have the species richness values and dominance regimes within the range of values in specified reference wetlands/floodplains of the target community. Canopy and shrub measurements are limited to those indigenous species that will contribute to the appropriate strata of the mature forested wetlands/floodplains. Up to half of the trees and shrubs in the upper transitional zone may consist of appropriate upland and facultative species, as found in specified reference wetlands. Desirable canopy and shrub species shall be reproducing naturally. For all forested wetlands, ERP Specific Condition provides that herbaceous vegetation shall have the species richness values and dominance regimes within the range of values in specified reference wetlands/floodplains of the target community. In making this evaluation, DEP shall consider the relative age of the mitigation site, as compared to specified reference wetlands. ERP Specific Condition 16.D.1 requires that all stream banks be stable, subject to normal erosion and deposition zones, as evidenced by the conformance of the stream with the applicable Rosgen type C or E, as described in the appropriate reference streams. ERP Specific Condition 16.D.2 requires that the physical characteristics of the reclaimed stream conform to its design. ERP Specific Condition 16.D.3 requires that tree roots, log jams, snags, and other instream structure shall be present at desirable intervals along the reclaimed stream. ERP Specific Condition 16.D.4 provides that species diversity and richness of the macroinvertebrate community shall be within the range of values documented in the reference streams or reported values of similar streams systems in central Florida. Also, all functional feeding guilds of macroinvertebrates found in the reference streams shall be present in the reclaimed streams. In the alternative, IMC may show that the reclaimed stream has met the minimum thresholds for the "good" classification in DEP's Stream Condition Index for macroinvertebrates and habitat quality. ERP Specific Condition 16.E provides that, throughout OFG, at least 105 acres of reclaimed forested wetlands and 217 acres of reclaimed herbaceous wetlands shall be determined to be wetlands or other surface waters. IMC shall achieve the minimum acreage for each wetland, as indicated on Map I-2 and associated figures and tables. However, IMC may make minor changes in the size, shape, or location of individual reclaimed wetlands, subject to BMR's approval. ERP Specific Condition 17 provides that DEP shall release IMC from further obligation regarding mitigation when ERP Specific Condition 16 has been met. IMC initiates the release procedure by notifying DEP that IMC believes the mitigation is ready for release, but this notice may not be earlier than two years after the completion of mitigation. DEP must respond within 120 days. ERP Specific Condition 17.d provides: "[DEP] may release the mitigation wetlands based on a visual evaluation, notwithstanding that all the requirements of Specific Condition 16 have not been met." ERP Specific Condition 18 applies to the surface water management system. The system must conform to the plans, specifications, and performance criteria approved by the ERP. ERP Specific Condition 19 requires IMC clearly to identify all no-mine areas in the field within two years of the issuance of the ERP. ERP Specific Condition 20 states that BMR will review the ERP at the end of the first five-year term after its issuance and at the end of each succeeding five-year term, if any. The purpose of the review is to determine compliance with general and specific conditions, including monitoring requirements. BMR staff shall quarterly inspect the mine for compliance with these requirements. ERP Specific Condition 21 requires IMC to provide a phased Conservation Easement, in favor of DEP, on 525 acres of OFG, as depicted on Figure F-6. Figure F-6 shows two easement areas. Phase A, which is 372 acres, corresponds to the 100-year floodplain of Horse Creek. Phase A is in the no-mine area. Phase B, which is 153 acres, is a wider band running along both banks of the northernmost 1 1/2 miles of Horse Creek and mostly on only the west bank for the southernmost 2 miles of Horse Creek. Phase B consists of part of the reclaimed area. The corridor covered by both phases of the Conservation Easement is generally not wider than 1000 feet and thus does not capture all of the non-improved pasture upland communities reclaimed on either side of Horse Creek and described above. IMC is required to grant the Conservation Easement on the Phase A lands within six months of the issuance of the ERP. IMC is required to grant the Conservation Easement on the Phase B lands within six months of the release by DEP of IMC from further obligations regarding reclamation and mitigation. ERP Specific Condition 21 incorporates the Conservation Easement and Easement Management Plan. The Conservation Easement implicitly acknowledges the fact that IMC is contractually obligated to convey OFG back to the Carlton- Smith family, after IMC has been released from further obligations regarding reclamation and mitigation. Thus, post- mining, OFG will return to its historic agricultural uses-- mostly, cattle ranching. The restrictions and encumbrances included in the Conservation Easement are designed to provide some protection to the wetlands, streams, and uplands within the Phase A and Phase B areas. Granted to the Board of Trustees of the Internal Improvement Trust Fund of the State of Florida, for which DEP serves as an agent, the Conservation Easement allows IMC and its successors, including the Carlton-Smith family, to use the encumbered property for cattle ranching, but only to the extent consistent with "sustainable native range management practices." These sustainable native range management practices require, among other things, the natural renewal of the grazing capacity of the land by allowing native grasses and other native forage species to regenerate. The Easement Management Plan contemplates prescribed burns of portions of the corridor. The Conservation Easement also allows IMC and its successors, upon obtaining the necessary permits, to construct a commodious 200-foot wide accessway across the encumbered property for a road, pipelines, draglines, and/or utilities. ERP Specific Condition 22 requires IMC to enhance 80 acres of existing pastureland within several areas of the Horse Creek floodplain, as indicated on Figure F-5, which is Habitat Enhancements. Most of the depicted enhancement areas are on OFG, but two of them are a short distance from OFG. ERP Specific Condition 22 requires IMC to plant 100 longleaf pines and/or oaks per acre within several sites, covering 80 acres of existing pastureland, adjacent to the 100-year floodplain of Horse Creek. Most of the sites are on the west bank of Horse Creek, mostly south of the Lobes, but a couple of sites are on the east bank in the vicinity of the East Lobe. ERP Specific Condition 23 requires that IMC plant these areas within one year of the issuance of the ERP and that the overall survival rate be at least 80 percent, as of the time of the release of the last mitigation parcel. ERP Specific Condition 23 requires IMC to enhance existing xeric and scrub habitats within areas designated as ACI (Area of Conservation Interest)-2, ACI-4, and ACI-6, as depicted on Figure F-5. Specific Condition 23 states that IMC shall enhance the wildlife habitat of these areas by performing controlled burns, cutting overgrown trees, planting desirable species, and controlling nuisance and exotic species. Specific Condition 23 obligates IMC to complete these enhancements within three years of the issuance of the ERP. ACI-2 is about 1 1/2 miles west-southwest of the southern end of OFG, between State Road 64 and the West Fork. ACI-2 consists of about 60 acres of overgrown xeric habitat, featuring 40 acres of sand scrub, predominantly sand live oak. Gopher tortoises occupy ACI-2 at a density of about 1.6 reptiles per acre. Florida mice occupy ACI-2 at a density of 0.4 rodents per acre, meaning that only 15-25 Florida mice may occupy ACI-2. By fence-posting overgrown sand pine and sand live oak and conducting a prescribed burn, IMC will reduce the heavy canopy existing on ACI-2 and enhance the suitability of ACI-2 for gopher tortoises and Florida mice. IMC will also apply herbicides to nuisance exotic species, such as bahiagrass, after which IMC will direct seed the flatwoods on the site with suitable vegetative species. Following this work, IMC may relocate Florida mice from OFG to ACI-2, upon approval from the FWC. ACI-6 is about one mile east of the southern end of OFG. ACI-6 consists of about 421 acres of a mixture of open land and overgrown oak scrub. Gopher tortoises occupy ACI-6 at densities ranging from 0.7 to 1.8 animals per acre. After fence-posting overgrown oaks and sand pine, conducting prescribed burns, installing fencing to exclude cattle and feral hogs, applying herbicide to kill exotic species, and direct seeding appropriate vegetation, IMC may relocate Florida mice from OFG to ACI-6, upon approval from FWC. ACI-4 consists of about 82 acres at the eastern end of the East Lobe and is within the no-mine area. The western end of ACI-4 slopes to the west through a bahia pasture before it enters a large bay swamp at the western end of the East Lobe. This area has been impacted by partial clearing and the depositing of animal carcasses--the latter practice yielding the name assigned to this area, the "boneyard" scrub. ACI-4 is dominated by mature scrub oaks. Gopher tortoises occupy ACI-4 at the rate of 0.85 terrestrial turtles per acre, and gopher frogs frequent the mouths of tortoise burrows at the site, although no signs of Florida mice exist. After conducting enhancement activities similar to those to be conducted on the other ACIs, IMC intends to create and maintain more suitable habitat for Florida mice. Specific Condition 23 states that IMC shall enhance 25 acres of pasture on ACI-4 by planting 100 longleaf pines and/or oak trees, and IMC shall manage these areas to achieve an overall survival rate of 80 percent through release of the final reclamation parcel. ERP Specific Condition 24 notes that IMC has committed to initiate the management and evaluation of amphibians, including the Florida gopher frog, and shall adhere to the management plans outlined in the IMC Minewide Gopher Tortoise and Burrow Conceptual Management Plan that FWC has examined, but not yet approved. IMC shall expend at least $30,000 to compare amphibian use of reclaimed and unmined wetlands. IMC shall include progress reports as to this study with its annual narrative reports required under Specific Condition 4. ERP Specific Condition 25 incorporates Tables 2AI-1 and 2AI-2 to provide assurance that IMC has sufficient sand tailings for the timely reclamation of wetlands contemplated in the ERP. Table 2AI-1 is the IMC Overall Sand Balance. Table 2AI-2 is the [OFG] Sand Balance. Table 2AI-1 shows the sand tailings production of IMC's Four Corners and Ft. Green mines from 2004-2014 and assumes an initial mining year of 2006 for OFG. For each of these 11 years, Four Corners produces 27,000,000 tons of sand tailings. For the first seven of these years, Ft. Green produces 17,000,000 tons of sand tailings. During these 11 years, IMC needs anywhere from 13,300,000 to 54,900,000 tons of sand tailings to meet all of its reclamation obligations. The closest that IMC will come to exhausting its sand tailings stockpile will be in year 6 of the OFG mining operation (2011, if OFG mining starts in 2006). For this and the following year, the sand tailings stockpile will total 300,000 tons. By this time, IMC's requirements for sand tailings begin to taper off, so that, by the final year on the schedule (2014), the sand tailings stockpile increases to 20,600,000 tons. Table 2AI-2 shows that IMC can meet its reclamation obligations for the Ft. Green Mine and OFG without using any stockpiled sand tailings. The next section of the ERP is Monitoring Required. The designations for this section start with a letter. As its name suggests, ERP Monitoring Required describes the monitoring program. The presence of monitoring does not imply the presence of standards or criteria applicable to what is monitored or the presence of a remedy or sanction for noncompliance with any standard or criterion. The existence of this section of the ERP does not mean that other sections of the ERP may impose monitoring requirements, applicable standards and criteria, and remedies or sanctions for noncompliance. ERP Monitoring Required A.1 requires IMC to submit annual narrative reports to BMR detailing the progress of the restoration program identified in ERP Specific Condition 4. As required in ERP Specific Condition 5, IMC shall submit to BMR hydrology reports annually and vegetation reports annually for the first three years and every other year thereafter, until release. At least 60 days prior to sampling, ERP Monitoring Required A.2 requires IMC to submit, for agency approval, vegetation, hydrology, and macroinvertebrate monitoring plans detailing sampling techniques and locations. ERP Monitoring Required A.3 requires IMC to include in its annual hydrology reports the daily rainfall amounts for the Ft. Green and OFG gauges shown on Map D-4. ERP Monitoring Required A.4 states that, if BMR determines that restoration efforts are not trending toward achievement of the release conditions set forth in ERP Specific Condition 16, IMC shall have 30 days from notification to submit proposed corrective actions. IMC shall implement corrective actions within 90 days of their approval. ERP Monitoring Required B states that data compiled in the CDA will be the primary source of reference wetland information. IMC shall then collect additional stage and hydroperiod data from the modeled wetlands. Within one year of the issuance of the ERP, IMC shall submit to BMR, for approval, a proposed sampling plan, including locations, frequencies, and vegetation, hydrology, and macroinvertebrate sampling methods. ERP Monitoring Required B provides that IMC shall select several wetlands of each community and submit them to BMR for approval. It appears that this process has already been completed, and DEP should updated ERP Monitoring Required B by incorporating into the ERP Figure RF-1, which, although not presently incorporated into the ERP, identifies 26 reference wetlands on OFG and nine reference wetlands on the original Ona Mine to the east of OFG. These reference wetlands include the most important components of the Lobes, the Heart-Shaped Wetland, Stream 2e's riparian wetlands, several wetlands in the Stream 1e series, the headwater wetland of Stream 3e, isolated wetlands south and east of the headwater wetland of Stream 3e, parts of the headwater wetland of Stream 1w, and the riparian and headwater wetlands of Stream 8e. As noted below, the riparian and headwater wetlands of Stream 8e, which are selected as reference wetlands, are moderate functioning, but the riparian and headwater wetlands of Stream 7e, which are not selected as reference wetlands, are high and very high functioning. ERP Monitoring Required C is Compliance Monitoring. Monitoring Required C.1 provides that IMC shall submit water quality data with the annual narrative reports submitted pursuant to ERP Specific Condition 7. All monitoring reports must include specified information, such as the dates of sampling and analysis and a map showing sampling locations. ERP Monitoring Required C.2 states that IMC shall submit hydrology data with its annual narrative reports. ERP Monitoring Required C.3 states that IMC shall monitor water levels in wetlands in no-mine areas in accordance with Table MR-1, which is described below. ERP Monitoring Required C.4 notes that IMC shall measure and report surface water flows in accordance with ERP Specific Condition 10. IMC must include in its reports to BMR all U.S. Geologic Service data collected at State Road 64 and State Road 72, which is south of State Road 64, and rainfall data collected by the U.S. Geologic Service, Southwest Florida Water Management District, and IMC. The annual hydrographs for Horse Creek at State Road 64 and State Road 72 "should" be similar. IMC must obtain and report hydrological data from 30 days after the issuance of the ERP until three years after the hydrological reconnection of the last reclaimed area upstream of a water level monitoring location. Within 60 days of the receipt of such data, BMR shall notify IMC of any changes to mining or reclamation that are necessary, and IMC shall have 60 days to respond to this notice. ERP Monitoring Required C.5 grants IMC a 50-meter temporary mixing zone adjacent to construction and in waters of the state; provided, however, this mixing zone is in effect only during the construction of the pipeline crossing just downstream of the Heart-Shaped Wetland. IMC must halt construction if monitoring reveals that turbidity at the site is more than 29 NTUs above upstream locations. ERP Monitoring Required C.6 states: "Compliance Monitoring Summary--See Table MR-1." Table MR-1 is discussed below, in connection with Table MR-2. ERP Monitoring Required D is Release Criteria Monitoring. Applying to vegetation, Monitoring Required D.1 provides that IMC shall conduct all monitoring of herbaceous vegetation during or immediately after the summer growing season. Monitoring Required D.1 requires the reports to include a description of collection methods and location maps. IMC must report data separately for individual wetlands. IMC must report separate density and cover information for trees, shrubs, and groundcover, as well as information about any supplemental planting. Applying to water quantity, ERP Monitoring Required D.2 provides that IMC shall submit water quantity data with its annual narrative reports, as required in ERP Specific Condition 4. IMC shall collect onsite daily rainfall data at OFG. ERP Monitoring Required D.3 requires: "Soils, macroinvertebrates and stream channel integrity/morphology shall be monitored as described in Table MR-2." ERP Monitoring Required D.4 states: "Release Monitoring Criteria Summary--See Table MR-2." Tables MR-1 and MR-2 refer to the monitoring required for compliance and release, respectively. The identification of these tables as "summaries" and the vague references to them in ERP Monitoring Required C.6 and D.4 suggest that the tables do not contain any performance standards and may imply that, except for the asterisked notes in Table MR-1, they summarize all of the performance standards and criteria contained in the ERP. If summaries, the tables should not introduce new elements, but they do just that with respect to the methods, sampling schemes, and frequency of monitoring. For water quantity monitoring, for instance, Table MR-2's promise of weekly readings of monitoring wells and piezometers for part of the year conflicts with the monthly reading required in ERP Specific Condition 10.b. If summaries of performance standards and criteria, the tables should capture all of the compliance and release criteria, but they do not. For water quality, for example, Table MR-2, which is limited to five parameters, potentially conflicts with ERP Specific Condition 16.A's broad assurance of compliance with Class III water quality standards, which encompass a broad range of parameters, including iron. For water quantity, Table MR-2 also omits the enforceable streamflow criteria of ERP Specific Condition 10.b. For soil, Table MR-2 includes one parameter--litter accumulation--for which no corresponding criterion exists and includes substrate-- for which important criteria exist as to the depths of sand tailings, topsoil, green manure, and muck--but omits any release criteria. Addressing two of the most important parts of the ERP--monitoring and performance criteria--these tables must be interpreted as subordinate to the remainder of the ERP, so that if they conflict with another ERP provision, the other ERP provision controls, but if they add a requirement not elsewhere found in the ERP, the requirement applies to the proposed activities. Table MR-1 is the Compliance Monitoring Criteria Summary. Table MR-1 identifies two monitoring parameters: water quality and water quantity. Asterisked notes state that the Table MR-1 requirements for water quality are in addition to those set forth in Specific Condition 7, which are discussed above, and the Table MR-1 requirements for water quantity are in addition to those set forth in Specific Condition 10.b, which are discussed above. For water quality, Table MR-1 addresses only turbidity. The compliance criterion is the Class III standard. The "proposed methods" are for IMC to monitor water, at mid- depth, 50 meters upstream and downstream from the point of severance and reconnection of each wetland. The frequency of monitoring is daily during severance or reconnection or during pipeline corridor construction or removal. The duration of monitoring is at least one wet season prior to mining, during mining, and through contouring. For water quantity, Table MR-1 addresses water levels, flow, hydrographs, soil moisture, and plant stress. The compliance criteria are soils sufficiently moist to support wetland vegetation and prevent oxidation and water levels in recharge ditches sufficient to simulate normal seasonal fluctuations of water in adjacent wetlands and other surface waters. The "proposed methods" are for IMC to install staff gauges, monitoring wells, piezometers, and flow meters in recharge ditches and wetlands in the no-mine area and at the point that the 100-year floodplain of Horse Creek intercepts the unmined portions of Streams 2e, 6e, 7e, 8e, 9e, 6w, and 8w. The frequency of monitoring is to check rainfall and recharge ditches daily, staff gauges in streams "continuously," and monitoring wells and piezometers weekly. The duration of monitoring is at least one wet season prior to mining, during mining, and through contouring. Table MR-2 is the Release Monitoring Criteria Summary. Table MR-2 identifies five monitoring parameters: water quality, water quantity, stream channel integrity and morphology, soils, and vegetation. For water quality, Table MR-2 addresses dissolved oxygen, turbidity, temperature, pH, conductivity, and, for all streams, all of the parameters in ERP Specific Condition 7.a. The compliance criteria are Class III standards. The locations are at or near the connection of wetlands in the no-mine area and at or near vegetative transects in streams and representative wetlands. The frequency is monthly from May to October prior to the reconnection to wetlands in the no-mine area and monthly from May through October of the year prior to the release request. The duration of monitoring is at least two years after the completion of contouring. For water quantity, Table MR-2 addresses water levels, flow, hydroperiod, rainfall, and hydrographs. The release criteria are values within the range of values documented in specified reference wetlands for each community type and, for hydroperiods and water levels, within the range of values predicted by modeling. The "proposed methods" are the same instruments identified for water quantity in Table MR-1. The locations for sampling are at or near the connection to wetlands in the no-mine area and at representative locations, including the deepest depths, of several representative wetlands of each community type. The frequency of monitoring is to check rainfall daily, staff gauges in streams "continuously," monitoring wells and piezometers weekly from May through October and monthly from November through April, and flow at sufficiently frequent intervals to generate rating curves for the streams. The duration of monitoring is at least two years after the completion of contouring. For stream channel integrity and morphology, Table MR-2 addresses channel stability and erosion, channel sinuosity channel profile, and cross sections. The release criteria are: "Stable channel and banks, no significant erosion, or bank undercutting, stream morphology within the range of values appropriate for the designed stream type (Rosgen C or E)." The location of sampling is over the entire channel length and representative cross sections. The frequency of monitoring for channel stability and erosion is after "significant" rain events for at least the first two years after contouring. The frequency of monitoring for channel sinuosity, channel profiles, and cross sections is years 2, 5, and 10. For soils, Table MR-2 addresses substrate description, litter accumulation, and compaction, but lists no release criteria. For vegetation, Table MR-2 addresses the species list and percent cover, FLUCFCS Level III map, percent bare ground and open water, nuisance species cover, upland species cover, tree density, shrub density, tree height, tree breast height diameter starting in year 5, and fruit and seedlings (starting in year 7). The release criteria are 400 trees per acre that are 12 feet tall, 100 shrubs per acre, species richness and diversity within the range of reference forested and herbaceous wetlands, 80 percent groundcover, and less than ten percent nuisance species. The location of sampling is randomly selected sites along several transects across each wetland, and the frequency of monitoring is years 1, 2, 3, 5, and every other year through the year prior to release. For macroinvertebrates, Table MR-2 addresses the number and identity of each taxon, diversity, functional feeding guilds, and the DEP Stream Condition Index. The release criteria are: "Species diversity, richness within range of reference wetlands, all functional feeding guilds or qualify as 'good' or better in the SCI." The location of sampling is in at least one representative 100-meter reach in each stream, and the frequency is at least twice yearly for at least the year prior to the release request for a stream. CRP The introductory CRP narrative describes IMC's plans to reclaim uplands, but does not impose any obligations upon IMC. Instead, the narrative introduces the reclamation project and summarizes the provisions of the general and specific conditions of the CRP. The failure to incorporate Map I-2, whose wetlands were incorporated by the ERP, and Map I-3 is material. CRP General Conditions 8, 9, and 10, discussed below, impose upon IMC certain requirements when reclaiming certain communities, but do not themselves impose the requirement of reclaiming these communities. The same is true for CRP Specific Condition 8. The only subcondition mentioning Map I-2 is Specific Condition 8.c, which alludes to Map I-2 while imposing upon IMC the reclamation technique of backfilling at least 15 inches of sand tailings upon those areas to be reclaimed as temperate hardwoods, live oak, and hardwood-conifer mixed. If this indirect reference imposes upon IMC the obligation of reclaiming these three upland forests pursuant to their depiction on Map I- 2, it is odd that Specific Conditions 8.a and 8.b fail even to mention Map I-2 in their discussion of the sand tailing and topsoil requirements for reclaimed pine flatwoods and sand live oak and xeric oak, especially when these three upland forest communities account for over 400 acres of reclaimed uplands, according to Table 12A1-1, which is also not incorporated into the CRP. The narrative portion of the CRP states that IMC's reclamation plan is to create 1769 acres of pasture, 50 acres of herbaceous, shrub, and mixed rangeland, 273 acres of palmetto prairie, 194 acres of pine flatwoods, 33 acres of xeric oak, 43 acres of temperate hardwood forest, 39 acres of live oak forest, 196 acres of sand live oak forest, and 550 acres of hardwood- conifer mixed forest. The CRP notes that most of the communities in the no-mine area, enhanced areas, and reclaimed communities will form part of a "larger mosaic of diverse upland and wetland habitat associated with Horse Creek and will serve as important wildlife corridors." The failure of the CRP approval to incorporate Map I-2 is an oversight. In the introduction to the January submittal, IMC proposed to reclaim the uplands, by community and area, as enumerated in Table 12A1-1, and, by community and location, as depicted on Map I-2. The failure to incorporate Map I-3 is probably an oversight, based on the second CRP narrative quoted below. The CRP narrative states that IMC has developed a Habitat Management Plan (HMP), which includes detailed pre- mining wildlife surveys and relocation programs. The narrative states that IMC will relocate, disturb the habitat of, and reclaim habitat for Florida mice, gopher tortoises, gopher frogs, and other commensals, pursuant to approvals from FWC. The narrative reports that IMC's Indigo Snake Management Plan has already received approval from the required agencies. Also, IMC will spend at least $30,000 to fund research on the potential of relocating burrowing owls onto reclaimed landscapes and at least $30,000 to analyze amphibian use of natural and reclaimed wetlands. However, the ERP and CRP approval incorporate only parts of the HMP. The CRP narrative adds: In addition to wetlands, a significant portion of the reclamation plan will focus on wildlife habitat through the creation of a diversity of upland habitat types adjacent to the Horse Creek corridor. This will provide a contiguous corridor averaging half a mile wide. IMC has committed to reclaim significant areas of pine flatwoods, palmetto prairie, sand live oak, and other upland habitats well beyond what is required by existing reclamation rules. This will be accomplished mainly through topsoiling and planting of a diversity of native species including shrubs and groundcover species. The use of exotic forage grasses will be minimized and native grass species will be emphasized in the groundcover of reclaimed upland habitat areas. A diversity of shrubs will also be planted in reclaimed upland forest areas. In addition, most of the mitigation wetlands will be created with diverse upland habitats surrounding them, resulting in enhanced wildlife and water quality functions. The CRP narrative addresses reclaimed soils: Special emphasis has also been placed on improving post reclamation soils. . . . Emphasis has been placed on restoring soils to more closely mimic native soils and existing soil horizons by making greater use of native topsoil and incorporating a greater percentage of sand at the surface. Green manure will be incorporated into surface soils where native topsoil is not used. In most cases, existing overburden spoil piles will be graded down and then capped with several feet of sand tailings. The thickness of the sand layer will be determined based on the targeted reclaimed land use with some wetlands requiring additional overburden to restore appropriate hydrology. The CRP narrative acknowledges that IMC has developed an Integrated Site Habitat Management Plan that includes plans for the reclamation of uplands, control of nuisance and exotic species in uplands, and management of all listed species. The CRP narrative asserts that IMC will reclaim and manage over 1378 acres of uplands, such as by removing cogongrass and maintaining it to less than 10 percent coverage, except less than 5 percent coverage within 300 feet of wetlands. The CRP narrative mentions that IMC has "volunteered" the Conservation Easement and Easement Management Plan to encumber not less than 525 acres associated with Horse Creek. CRP General Condition 7 states: "[IMC] is encouraged to implement the Integrated Habitat Network (IHN) concept (where possible) when establishing reclaimed upland and wetland forested areas." As overlaid on OFG, the IHN, which is developed by DEP, is depicted in Figure 12-5. The IHN covers almost all of the no-mine area; the floodplains and headwater wetlands of the Stream 1e series, Stream 3e, and Stream 3e?; much of the non-pasture reclaimed uplands; and a large area of reclaimed improved pasture south and west of the reclaimed sand live oak area immediately west of the West Lobe. The backbone of the IHN is the network of rivers and streams, with their floodplains, that provide multifunctional habitat for wildlife. As noted in the introduction to the January submittal, the HMP helps implement the portion of the IHN located at OFG. Although only selectively incorporated into the ERP and CRP approval, the HMP describes IMC's overall plan for reclaiming OFG. The stated goal of the HMP is "to maintain or improve the biological functions of the wetlands and uplands . . . as an integrated component of the mining and reclamation plans." The HMP adds: "By preserving and managing the highest quality habitats on [OFG], these reserves will serve as source populations to recolonize the remainder of the site following completion of reclamation." Overall, the reclamation plan and HMP try to restore a functional interrelationship of uplands, wetlands, and surface water to replace the reduced functions that result from the agricultural alterations to uplands, wetlands, and most of the surface water, leaving large areas of a patchwork fragmentation of habitats. The HMP covers habitat management prior to land clearing, species-specific management techniques immediately prior to land clearing, species-specific management techniques during mining, habitat management in no-mine areas, reclamation goals for habitat, reclaimed habitat management after release, and, in the second part of the HMP, specific actions for each listed wildlife and plant species. Prior to land clearing, IMC will engage in little active habitat management, apart from surveys, as the Carlton- Smith family continues its agricultural uses of the land, which it is entitled to do under its contract with IMC. Immediately prior to land clearing, IMC will relocate each species, after obtaining the necessary permits, either by capture or, for the more mobile species, controlled burns or directional clearing to encourage wildlife migration into an adjoining refuge area. For listed bird species, IMC will protect their nesting areas or restrict land clearing to non-nesting season. During mining, aquatic- and wetland-dependent species will continue to have access to Horse Creek and its riparian wetlands, which are never isolated by the ditch and berm system. The only permitted direct disturbance of the no-mine area is outside Horse Creek's direct floodplain. During mining, the vast water recirculation system will provide incidental, temporary habitat for many aquatic- or wetland-dependent species. The second part of the HMP identifies management techniques for specific listed species of vertebrates. The HMP states that no listed plants exist on OFG. The HMP addresses 15 listed species observed on OFG and nine listed species that could potentially use OFG. The HMP mistakenly lists the Florida panther in the latter category, rather than the former category, but the error is harmless given the limited use of OFG by the Florida panther and the apparent lack of a breeding population north of the Caloosahatchee River. The following paragraphs describe the HMP's treatment of several listed species using OFG. Noting that the American alligator, which is a species of special concern, occupies freshwater habitats throughout Florida, plenty of such habitats exist around the mining areas, and the alligator is mobile, IMC expects that the American alligator will move out of the way of mining activities, so no management measures will be used for alligators. Presumably well-served by former Land-and-Lakes reclamation and an opportunistic inhabitant of deep wetland reclamation, alligator management is of no importance in these cases. The HMP reports two possible observations on OFG of the Florida panther, which is an endangered species. There is no doubt about one of these observations. On the other hand, there is no doubt that OFG is far from prime panther habitat. Thus, IMC will check for panther signs during pre-clearing surveys and anticipates that the unmined floodplains that are part of the IHM will maintain suitable habitat--presumably, for travel. IMC has already mapped the distribution on OFG of the gopher tortoise, which prefers well-drained, sandy soils characteristic of xeric and mesic habitats. IMC has already prepared a management plan for gopher tortoises, which are a species of special concern, and, upon DEP approval, will engage in several measures to reduce mortality due to mining activities, including, upon receipt of an FWC permit, relocating gopher tortoises, as well as other commensal species found in or near the tortoises' burrows, to appropriate locations, including one or more of the above-described ACIs. The Sherman's fox squirrel, which is a species of special concern, prefers sandhill communities and woodland pastures, and many of these squirrels use suitable areas of OFG. They are mobile, and, during mining operations, they will move to the no-mine areas adjacent to Horse Creek. Prior to land clearing, IMC will survey each area, and, if it finds active nests, these areas will be avoided until the young squirrels have left the nests, pursuant to FWC requirements. The Florida Mouse, which is a species of special concern, inhabits sand pine scrub and other xeric communities and is a commensal of the gopher tortoise. Prior to land clearing of suitable Florida Mouse habitat, IMC will conduct live-trapping. If any such mice are captured, IMC will relocate them to a suitable relocation site, such as to ACI-2, ACI-4, or ACI-6 or to xeric or pine flatwoods/dry prairie habitat that will be reclaimed on OFG. IMC will employ similar procedures for the Florida gopher frog, which is another commensal of the gopher tortoise. A species of special concern, the Florida gopher frog will also be the subject, with other amphibians, of research regarding use of reclaimed habitats and funded by IMC with at least $30,000. The Audubon's crested caracara, which is a threatened species, prefers dry prairie with scattered marshes and improved pasture. They typically nest in cabbage palms or live oak trees. Observers have seen a pair of caracaras on OFG, but attempts to locate a nest onsite have been unsuccessful. Prior to clearing cabbage palms, IMC will again survey the area for nests. If IMC finds a nest onsite or within 1500 feet of OFG, it will develop an FWC-approved management plan. The post- reclamation palmetto prairie and pine flatwoods are good caracara habitat. One of the few listed species whose habitat needs have been well-served by agricultural conversions to improved pasture, the burrowing owl occupies numerous areas on OFG. IMC intends to schedule land clearing in areas with active burrows during non-nesting season, but, if this is impossible, IMC will attempt to empty the burrow prior to clearing the land. Additionally, IMC will spend at least $30,000 to fund research to improve the technology to relocate onto reclaimed land burrowing owls, which are a species of special concern. Although IMC found on OFG no nests of sandhill cranes, which are threatened, or little blue herons, which are a species of special concern, sandhill cranes nest in reclaimed wetlands on the Ft. Green Mine, and IMC expects sandhill cranes to nest in the reclaimed wetlands at OFG. Prior to mining, IMC will survey marshes for sandhill crane and little blue heron nests, and, if it finds any, it will disturb those areas in non- nesting season. Wood storks, which are endangered, use OFG for foraging, but IMC found no evidence of wood stork rookeries on or nearby OFG. The nearest known active rookery is 22 miles from OFG. Prior to landclearing during wood stork nesting season, IMC will survey each wetland with the potential to support stork nesting sites. If IMC finds any nests, it will follow the latest guidelines from FWC or U.S. Fish and Wildlife Service for protecting the site. For the white ibis, snowy egret, and tricolored heron, which are species of special concern, IMC will survey those wetlands that are suitable nesting site prior to landclearing. If any active nests are found, IMC will schedule landclearing during non-nesting season. CRP General Condition 8 provides that groundcover in all upland forests shall include one or more of the following native plants: fruit-bearing shrubs, low-growing legumes, native grasses, and sedges. CRP General Condition 9 provides that IMC shall use native grasses and shrubs when reclaiming grasslands and shrub and brushlands. CRP General Condition 10 provides that IMC shall incorporate clumps of trees in reclaimed improved pasture so that each ten acres has "some trees." CRP General Condition 11 states that IMC shall make "every effort" to control nuisance and exotic species within the mine. CRP Specific Condition 1 is ERP Specific Condition CRP Specific Condition 2 is ERP Specific Condition 23. CRP Specific Condition 3 is ERP Specific Condition 11. CRP Specific Condition 4 is for IMC to obtain authorization from the FWC to trap and relocate Florida mice. Specific Condition 4 requires the trapping and relocation of Florida mice prior to clearing areas inhabited by them. CRP Specific Condition 5 requires IMC to make "every effort" to relocate listed plant species to suitable reclamation sites when such species are encountered prior to or during land clearing. CRP Specific Condition 6 is ERP Specific Condition 12.c. CRP Specific Condition 7 is ERP Specific Condition 12.d. CRP Specific Condition 8.a provides: Areas designated as pine flatwoods . . . and palmetto prairie shall be reclaimed by placing a minimum layer of fifteen (15) inches of sand tailings over the overburden and topsoiling with three (3) to six (6) inches of direct transferred or stockpiled native topsoils from pine flatwoods or palmetto prairie areas as that topsoil is available and feasible to move. Feasible means of good quality, relatively free of nuisance/exotics species, and within 1.5 miles of the receiver site. If topsoil is not available or feasible to move, a green manure crop will be seeded and disked in after it has matured before applying a flatwoods or palmetto prairie native ground cover seed mix to this site. In flatwoods, longleaf pine . . . or slash pine . . . shall be planted in the appropriate areas to achieve densities between 25 and 75 trees per acre. In flatwoods and palmetto prairie, shrubs typical of central Florida flatwoods and palmetto prairies will be recruited from the topsoiling, planting, and/or seeding to achieve a minimum average density of 300 shrubs per acre. The total vegetation covered by hydric flatwoods will be greater than 80 percent, in mesic flatwoods and palmetto prairies will be greater than 60 percent, and in scrubby flatwoods, greater than 40 percent. CRP Specific Condition 8.b provides: Areas designated as sand live oak or xeric oak scrub . . . shall be reclaimed by placing several feet of sand tailings over the overburden and topsoiling with three (3) to six (6) inches of direct transferred or stockpiled native topsoil from scrubby flatwoods or scrub areas. Feasible means of good quality, relatively free of nuisance/exotics species, and within 1.5 miles of the receiver site. If topsoil is not available or feasible to move, a green manure crop will be seeded and disked in after it has matured before applying a scrubby flatwoods or scrubby native ground cover seed mix to this site. Trees and shrubs typical of central Florida scrubs will be recruited from the topsoil, planted, and/or seeded to achieve a minimum density of 600 plants per acre. Vegetative cover in these areas will be greater than 40 percent. CRP Specific Condition 8.c provides: Other upland forest areas, including [temperate hardwoods, live oak, and hardwood-conifer mixed], shall be reclaimed, as illustrated by Map I-2, by placing a minimum layer of fifteen (15) inches of sand tailings over the overburden, capping the area with approximately three (3) inches of overburden and disking the surface to reduce compaction of the upper soil layer prior to revegetation. Other uplands shall be revegetated with a native ground cover, planted with trees to achieve a density of 200 plants per acre, and planted with shrubs to achieve a density of 200 shrubs per acre. CRP Specific Condition 8.d provides that IMC shall incorporate native grass species into the groundcover of all reclaimed uplands. CRP Specific Condition 8.e allows IMC to use bahia grass, Bermuda grass, and exotic grass species as groundcover in native habitats only in "limited amounts" needed for "initial stabilization in areas highly prone to erosion." When using these grasses, IMC must maintain them to prevent their proliferation. CRP Specific Condition 9 is ERP Specific Condition CRP Specific Condition 10 is ERP Specific Condition 21. CRP Specific Condition 11 resembles ERP Specific Condition 11, but requires more of IMC. CRP Specific Condition 11 states that IMC "has committed" to initiate the management and evaluation of amphibians, including the Florida gopher frog, and shall adhere to the provisions of the IMC Minewide Gopher Tortoise and Burrow Conceptual Management Plan. IMC shall pay at least $30,000 to conduct a study of amphibian use of reclaimed and unmined wetlands. IMC shall report its progress in the annual narrative reports that it must file, pursuant to Florida Administrative Code Rule 62C-16.0091. CRP Specific Condition 12 contains similar provisions for the burrowing owl. Related to ERP Specific Condition 15.a, CRP Specific Condition 13 requires IMC to make "every effort" to control cogongrass by eradicating it prior to mining, removing it after it colonizes spoil piles during mining, inspecting donor topsoil sites to prevent infestation by it, and regularly treating it on reclaimed sites to maintain coverage below 10 percent, or 5 percent within 300 feet of any reclaimed wetland. WRP The WRP at issue is for the Ft. Green Mine, not OFG. The basic purpose of the WRP is to permit IMC to dispose of the clay tailings extracted from OFG in CSAs O-1 and O-2, which are located at the southern end of the Ft. Green Mine. In an unchallenged action, DEP, on March 20, 2001, approved a requested modification of the CRP approval for the Ft. Green Mine to permit the changes sought in these cases for the Ft. Green Mine WRP. Thus, the WRP modification sought in these cases is merely a conforming modification. Normally, a WRP/ERP would take precedence over a CRP approval because mining may not start without a WRP/ERP, but may start without a CRP approval. In the unusual situation at the Ft. Green Mine, where the mining has been completed, the analysis of the WRP modification is limited to, primarily, the sufficiency of the changes in mitigation to offset the already- completed mining and, secondarily, the relevant impacts of the mitigation itself. DEP issued the WRP on May 1, 1995. This permit allowed IMC to mine 524.6 acres of wetlands at the Ft. Green Mine. On February 3, 1997, DEP issued an ERP to allow IMC to disturb 1.39 acres of surface water for a utility corridor. Following the receipt of a request by IMC for a major modification of the WRP to permit the mining of 7.6 acres of wetlands, DEP consolidated this request, the utility-corridor ERP, and the original WRP into a new WRP issued July 28, 1999. After a modification to the new WRP in 2000 that is irrelevant to the present cases and other irrelevant permitting activity, IMC has requested the modification that is at issue in these cases. Because this WRP modification follows the completion of mining and the near-completion of backfilling of sand tailings into the mine cuts, a denial would not spare the wetlands and other surface waters from the impacts of mining. Rather, a denial would leave the Ft. Green Mine with greater impacts and less mitigation. In simplest terms, a denial would harm the water resources of the District. Strengthening the already-approved mitigation and diminishing the impacts of the already-approved CSAs, this WRP modification will authorize IMC to reduce the size of the two CSAs (O-1 and O-2) in the southern end of the Ft. Green Mine and relocate them farther from Horse Creek; to relocate several reclaimed wetlands in the vicinity of CSAs O-1 and O-2 and expand their area by 2.7 acres with minor changes to some sub- basin boundaries; and to modify the reclamation schedule to conform to a modification already approved without challenge for the Ft. Green Mine CRP. The record demonstrates that the reduction in size and relocation of the CSAs away from Horse Creek will reduce the hydrological and biological impacts from those already permitted. The record demonstrates that the expansion of the area of reclaimed wetlands will add mitigation to offset the hydrological and biological impacts from already-completed mining activities. The record demonstrates that the relocation of the reclaimed wetlands and modification of the reclamation schedule will not affect the impacts or mitigation. Other Mitigation/Reclamation Projects Introduction The formation of wetlands vegetation, according to IMC biologist Dr. Andre Clewell, is a function of topography, hydrology, soils, and physical environment--to which should be added time. The formation of soils, according to Charlotte County soil expert Lewis Carter, is a function of parent material, time, relief, vegetation, and climate. Hydrology is dependent upon, among other things, topography, soils, geology, vegetation, and climate. Successful reclamation must thus account for the complex interdependency of the dynamic processes involving vegetation, soil, and hydrology. Although actual reclamation follows a clear order-- geology, soils, contouring, and planting--the order of the design process is not so clear. Presumably, in designing a reclamation plan, the biologist, soil scientist, and hydrologist would each prefer to have the final--as in last and authoritative--word. In general, the comparison of older mitigation sites to newer mitigation sites requires caution due to two factors, which somewhat counterbalance each other. The vegetation of the older sites has had longer to establish itself. The importance of this factor varies based on the type of vegetation. Groundcover establishes more quickly than shrubs, and shrubs establish more quickly than trees, but groundcover that requires protection from the tree canopy may not be able to colonize an area until the trees are well-established. Soils take a longer time to recover, generally longer than the timeframes involved in phosphate mining reclamation in Florida. The soils present in Hardee County took 5000 to 10,000 years to form. The A horizon, or topsoil layer, at OFG formed over 300-500 years. However, if the soil and hydrology are suitable at a reclaimed site, an A horizon may start to reform in as little as 10 years, but, even under ideal conditions, it will take several hundred years to reform to the extent and condition in existence prior to mining. The mucky soils underlying bay swamps form at the rate of about one inch per 1000 years. Offsetting the advantage of age for vegetation and soils, the older reclamation sites may suffer from less advanced designs and construction techniques. Newer sites benefit from advances in science and technology that have enabled phosphate mining companies to design and implement reclamation projects that more successfully replace the functions of the natural systems and communities lost to mining. Some of these advances have resulted in dramatic, sudden improvements in reclamation. The assessment of past reclamation projects must account, not only for the age of each project, but also the willingness of the phosphate mining company at the time to employ the then-available science and technology. The ratio of the cost of reclamation to projected revenues depends on the variables of specific mitigation expenses, mining expenses, and the value of the phosphate rock. These economic factors operate against the backdrop of a dynamic regulatory environment. In these cases, for example, IMC's willingness to reduce its mining impacts and expand its mitigation was a direct result of the Altman Final Order and DEP's decision to revisit its earlier decision to permit the Ona Mine. Uplands The uplands at OFG are more amenable to successful reclamation than the wetlands or streams at OFG. Uplands provide crucial functions. Certain uplands, such as those that provide seepage to wetlands or prime recharge to deep aquifers, provide hydrological functions as complex as the hydrological functions of many wetlands. Certain uplands provide irreplaceable habitat. Certain uplands vegetation is as vulnerable to climactic or anthropogenic disturbance as any wetlands vegetation. However, for the most part, the functions of uplands are not as complex or important as the functions of wetlands and other surface waters, when examined from the perspective of the water resources of the District, and these functions are more easily reclaimed. Over 77 percent of OFG and over 90 percent of the uplands at OFG are agricultural (2146 acres) or pine flatwoods, palmetto prairie, or sand live oak (1120 acres). (As noted above, palmetto prairie and sand live oak share many attributes of pine flatwoods, which they often succeed.) In terms of function, tolerance to ranges of hydrology and soils, and robustness of post-reclamation vegetation, these 3266 acres of uplands communities will be easier to reclaim than all of the proposed streams and wetlands, except for deep marshes, although pine flatwoods and palmetto prairies present the greatest difficulties in uplands reclamation due to their soil and hydrological requirements, including access to the post- reclamation water table. Impacts to uplands include the disappearance--even temporarily--of critical habitat for listed species, the susceptibility of uplands to post-disturbance nuisance exotics, and, for upland forested communities, the relatively long period required for restoration of the canopy. However, these impacts can be offset in most cases. Management plans can mitigate the temporary or permanent loss of specific upland habitat, depending on the availability of habitat and the robustness and abundance of the species requiring the habitat. Absent the presence of rare uplands habitat and/or rare species requiring the habitat, a greater problem with uplands reclamation is controlling nuisance exotics. Various grass species, including Bahia, Bermuda, torpedo, centipede, Natal, and cogon, impede progress in the development of a healthy uplands community. One of the world's ten worst weeds, cogongrass is limited to uplands, although it may extend into the higher parts of wet prairies and drier areas within forested wetlands. Although nuisance and exotic species may invade undisturbed areas, the removal of existing upland vegetation exacerbates the problem by removing native competitors and stimulating unwanted germination. However, ongoing maintenance, through a combination of herbicides, manual removal, and fire, controls the nuisance exotics long enough that the native vegetation can colonize the disturbed area. Upland forested communities require protection from grazing and mowing to permit their establishment. Canopy development takes years for any upland forested community and, for slower-growing xeric systems, at least a decade. The timely restoration of an appropriate fire regime is also important for the health of many upland communities. Not surprisingly, the record demonstrates the successful reclamation of uplands at several mitigation sites. In recent years, reclamation scientists have restored uplands structure of uplands by restoring the understory and midstory. Uplands restoration has improved with the introduction of new, more effective reclamation techniques, such as topsoiling and seeding. Until 1987, for instance, restoration biologists did not know that wiregrass--a key component of the understory of pine flatwoods--produced seeds. This knowledge has assisted in the reclamation of a proper understory of pine flatwoods. The favorable prognosis of uplands reclamation means that extensive areas of OFG uplands may be mined. Their functions will be substantially replaced, in a reasonable period of time, upon the establishment of the reclaimed upland community, although the destruction of xeric communities means their absence for relatively long periods of time and the destruction of uplands providing seepage support to wetlands requires the close-tolerance hydrology and soils associated with the most difficult wetlands reclamation. Approved in 1989 and amended in 1994, constructed by 1986, and released in 1994, Best of the West (NP-SWB(1D)) was targeted for 15-18 acres of xeric habitat. Best of the West was constructed on sand tailings overlaying overburden, although this site exhibits some stunted vegetative growth where the sand tailings may not be very thick and the roots of trees may have encountered the hardened overburden. FWC assisted the phosphate mining company in designing the reclamation plan for this site, which has resulted in the successful reclamation of 10 acres of xeric habitat. The CDA provides some background on Best of the West. The West Noralyn Xeric Scrub Reclamation (N-5), which was constructed by 1986, contained "mulched overburden plots" and 60 acres of unmined scrub. Containing a total of 462 acres of reclaimed and unmined land, Noralyn was the first attempt to create a large-scale xeric community. About 120 acres of Noralyn received 12 inches of donor topsoil from a comparable xeric community. Due to a lack of representation in the donor site, supplemental plantings of longleaf pine, sand pine, and rosemary followed. The overall project has been "moderately successful," but the 18 acres that yielded "exceptional results" were dubbed "Best of the West." Best of the West thus illustrates a recurrent feature of much reclamation activity, in which successful projects are actually small parts of the original project area, the rest of which is substantially less successful. The CDA states that, in January 2000, IMC initiated a land management program for Noralyn that includes herbicide applications and prescribed burns. After herbicide was applied to kill cogongrass, IMC conducted the first burn in March 2001. Noralyn is now being managed for four to five families of Florida scrub jays, a listed species. Four Eastern Indigo snakes, 225 gopher tortoises, numerous gopher frogs, and 119 Florida mice have been relocated to Noralyn. Approved in 1988, constructed in 1991, and released in 1992, Hardee Lakes topsoil (FG-PC(1A)) has a 7.9-acre uplands component that was topsoiled with one inch over overburden. Despite receiving no maintenance, the site displays few weeds or nuisance exotics, although cogongrass has invaded the site. The reclaimed site displays saw palmetto, gallberry clumps, creeping bluestem grass, and, in topsoiled areas, flowering milkwood. The site includes an ecotone between pine flatwoods and a wet prairie, which developed due to the appropriate slope and soil. The CDA identifies two one-acre demonstration projects with Hardee Lakes topsoil. The Ft. Green-Hardee Lakes Pine Flatwoods Project, a topsoiled site, has achieved a lower ratio of saw palmetto to pines than is presently typically of fire-suppressed communities and is more typical of historic Florida pine flatwoods. The Ft. Green-Hardee Lakes Palmetto Prairie Site, also topsoiled, has been successfully revegetated with saw palmettos and other appropriate species. An interesting uplands reclamation site, for its different use of soils, is the Bald Mountain complex (KC-LB(2) and LB(4)), which is a 180-acre site. In a reclamation project approved in 1989 and 1996, constructed in 1993, and released in 1994 and 2002, IMC backfilled the Bald Mountain site with sand tailings down to 40 feet, capped the sand tailings with six inches of overburden, and then mixed the soils. Nearby, Little Bald Mountain received only sand tailings. Scrub were planted on both locations, but Bald Mountain also received sandhill plantings. Bald Mountain contains suitable sandhill species, such as sandhill buckwheat, although natal grass has been a problem. Natal grass is an invasive grass that colonizes quickly and often requires manual removal. Little Bald Mountain contains appropriate understory grasses, including short-leaved rosemary, an endangered species; Gopher apple, an important wildlife food; and Ashe's [savory] mint, a listed species. The rosemary and mint are reseeding themselves. The site also contains several large palmettos that were started from seed. Approved in 1996, constructed in 2000, and not yet released, Ft. Green/Horse Creek Xeric (FG-HC(3 & 5)) is a 99-acre uplands site reclaimed as xeric oak. IMC backfilled at least six feet of sand tailings over the overburden and then added topsoil over the sand. Already, this site, which is in the nearby Ft. Green Mine, has developed all levels of structure in the appropriate ecosystem, although, according to the CDA, it received irrigation "frequently" from an irrigation system at the start of the project. The site includes denser vegetation, such as shrub palmetto, grasses, and forbs. The direct transfer of topsoil has added species diversity, such as a Florida spruce and a listed orchid. The site also contains a small number of longleaf pines. IMC has hand-removed natal grass at this site, but has lately been using a new selective herbicide. According to the CDA, though, the presence of invasive exotics throughout the site is limited to 0.4 percent. One of the best upland reclamation sites is MU 15E Topsoil (FCL-LMR(6)), which was approved and constructed in 2002 and has not been released. This is a 30-acre topsoiled site in which IMC transferred topsoil carefully: if topsoil was taken from a depression on the donor site, the topsoil was placed in a depression in the receiving site. This site already displays a rich diverse plant palette with hardly any weedy or exotic species. In this site, palmetto and wet prairies slope down to a flatwoods marsh. This site also contains a reclaimed ephemeral wet prairie--possibly the only known ephemeral wet prairie ever reclaimed after phosphate mining. With modest efforts regarding soils and possibly more strenuous efforts regarding nuisance exotics, the reclamation of uplands is relatively easily attained, provided the sites can be protected for the longer timeframes necessary to establish upland forests and especially upland xeric communities and an appropriately shallow water table is reclaimed for pine flatwoods and palmetto prairies. Wetlands Wetlands reclamation is generally more difficult than uplands reclamation. Successful wetlands reclamation typically requires better command of post-reclamation topography, hydrology, soils, and physical environment. Material deviations in these parameters reduce, or eliminate, many wetlands functions, such as floodplain communication, nutrient sequestration, floodwater attenuation, ecotone transitions, and habitat diversification. The loss of such functions may result in immediate problems with water quality, water quantity, and habitat. Given the greater difficulty in successful wetlands reclamation, experience in wetlands reclamation is, not surprisingly, more mixed than the generally favorable experience in uplands reclamation. The greater difficulty in, and more guarded prognosis of, wetlands reclamation, as compared to uplands reclamation, means that the disturbance of wetlands demands closer analysis of the functions of the wetlands proposed to be mined, the functions of the wetlands proposed to be reclaimed, and the reclaimed soils, hydrology, topography, and physical environment on which the reclamation scientists will rely in reclaiming wetlands functions. The most important factor in wetlands reclamation is hydrology. Wetlands with less rigorous hydrological needs, especially if they also tolerate deeper water over longer periods of time, reclaim much more easily than wetlands with more precise hydrological needs, especially if they require shallower water over shorter periods of time. The phosphate mining industry has repeatedly reclaimed marshes and cypress swamps that are inundated deeply and for extended periods of time, but has had a much harder time reclaiming shallower wetlands requiring shorter hydroperiods or shallower water levels. The two most difficult wetlands of this type to reclaim are bay swamps and wet prairies. Among herbaceous wetlands, deep marshes are the easiest to reclaim. Often a target of Land-and-Lakes reclamation, deep marshes also are the result of reclamation projects that failed to create targeted shallower wetlands. Charlotte County ecologist Kevin Irwin noted that deep marshes are easier to reclaim than forested wetlands, for which the post-reclamation hydrology must be more precise. Similarly, a freshwater marsh, which tolerates 6-30 inches of water from 7-12 months annually, is easier to reclaim that a wet prairie, which tolerates 0-6 inches of water from 2-8 months annually. Among forested wetlands, bayheads or bay swamps, as defined in these cases as seepage forested wetlands, are harder to reclaim than mixed wetland hardwoods, as IMC biologist Dr. Douglas Durbin testified--likely, again, due to the requirement of more precise post-reclamation hydrology. Accordingly, the parties do not dispute the ability of the phosphate mining industry to reclaim deep marsh habitat, including freshwater marshes and shrub marshes, as well as deep swamps--principally cypress swamps. Like wet prairies, which sometimes fringe deep marshes, deep marshes provide habitat, supply food, attenuate floodwaters, and improve water quality. Deep marshes may host large numbers of different plant species. However, like lakes, deep marshes remove larger amounts of water from the watershed, as compared to shallower wetlands with shorter hydroperiods, due to evapotranspiration. The reclamation projects known as Morrow Swamp, Ag East, 8.4-acre Wetland, and 84(5) trace a short history of the reclamation of deep-marsh habitat. Permitted in 1980, constructed in 1982, and released in 1984, 150-acre Morrow Swamp represents a prototype, second- generation wetlands reclamation project. According to the CDA, Morrow Swamp is from an era in which reclamation did not attempt to restore topography: "This ecosystem included the reclamation of 150 acres of wetland (freshwater marsh, hardwood swamp, and open water) and 216 acres of contiguous uplands. The reclamation site was originally pine flatwoods and rangeland before it was mined in 1978 and 1979." Designed and built before reclamation scientists concentrated on soils, the hydrological connection between Morrow Swamp and Payne Creek, into which Morrow Swamp releases water, is a concrete structure in a berm that leads to a swale that empties into Payne Creek. Morrow Swamp reveals one obvious shortcoming of mechanical outflow devices, at least if they depend on ongoing maintenance, because vegetation and sedimentation in the infrequently maintained outflow device have blocked the flow of water and contributed to water levels deeper than designed. The reclamation scientists pushed the row-plantings of trees in Morrow Swamp in an effort to understand the relationship of vegetation and hydroperiod. In doing so, they killed thousands of trees, such as the cypress trees that Authority ecologist, Brian Winchester, found that grew to 6-8 inches in diameter and suddenly died. This tree mortality was likely due to problems with water depths and hydroperiods, as suggested by the healthier cypress trees lining the shallower fringe of the marsh. Morrow Swamp operates as a basin with a perched water table atop compacted, relatively impermeable overburden. Beneath the dry overburden is moist soil, so there is no groundwater connection between the marsh and the surficial aquifer. According to Mr. Carter, sand is 15 times more permeable than overburden. Morrow Swamp presents numerous shortcomings, but not to alligators, who find ample food and habitat in and about the deep marsh. More importantly, the emergent-zone vegetation within Morrow Swamp is sequestering nutrients and thus providing water-quality functions. Unfortunately, the deeper water supports only floating vegetation, which is much less efficient at sequestering nutrients, and less diverse than the shallower emergent vegetation, so the excessive depths of Morrow Swamp limit its water-quality functions. Although short of a model wetlands reclamation project, Morrow Swamp was an important milestone in the development of wetlands reclamation techniques and clearly functions as a deep shrub marsh today. Permitted in 1985, constructed in 1986, and released in 2002, 214-acre Ag East (PC-SP(1C)) was built on the knowledge acquired from Morrow Swamp. At Ag East, which is just northeast of Morrow Swamp, the reclamation scientists, planting a large variety of trees, focused on water levels and hydroperiods. The reclamation scientists engineered a wetland system with less open water than Morrow Swamp. They also inoculated the surface with a layer of organic mulch material 2-4 inches thick. However, the design of Ag East again incorporated mechanical devices to control water levels. A weir at one corner of Ag East contains boards; by removing or adding boards, reclamation scientists could control the water depths behind the weir. The deep marsh within Ag East is excessively deep with an excessively long hydroperiod. In certain respects, Ag East has functioned better than Morrow Swamp, although there is some question as to vegetative mix establishing the site and the associated functions that the vegetation will provide. Again, though, Ag East features a functioning deep marsh. One clear shortcoming of Ag East was the failure to create appropriate upland habitat, such as pine flatwoods, around the wetlands, so that wetland species could find appropriate uplands habitat for breeding, nesting, or feeding. The CDA notes the availability of quarterly water quality monitoring data, over a five-year period, for pH, dissolved oxygen, conductance, and total phosphorus, among other parameters, but the results are not contained in this record. Permitted in 1983, constructed by 1986, and released in 1995, 8.4-Acre Wetland (FG-83(1)), which was targeted for 8.4 acres of wetland forested mixed, represents an early use of topsoil, which was a good seed source for herbaceous species and helped increase the effective depth of overburden. As noted above, shallower overburden discourages tree growth past a certain stage. However, 8.4-Acre Wetland also uses a water- control weir to control water depths on the reclaimed wetland. Despite its smaller size than Morrow Swamp or Ag East, 8.4-Acre Wetland was a more ambitious project hydrologically, as it attempted to replace a seepage wetland with a seepage wetland that would receive water from the surrounding uplands. Unlike Morrow Swamp and Ag East, 8.4-Acre Wetland was designed to reclaim only forested wetlands, not forested wetlands and marsh wetlands. Unfortunately, 8.4-Acre Wetland did not re-create a seepage wetland due to excessively deep water and excessively long hydroperiods. Emphasizing instead the creation of microtopography, the reclamation scientists added sand-tailings hummocks within the deeper marsh, effectively lowering the water table under the mound, and planted wetland herbaceous and forested species that could not tolerate the wetter conditions around the hummock. The evidence is conflicting as to the success of these hummock plantings, but the idea was sound. Parts of 8.4-Acre Wetland are at least half infested with cattails, and sizeable areas within 8.4-Acre Wetland are reclaimed marsh, not swamp--despite the attempt of the reclamation scientists to reclaim forested wetlands only. Permitted in 1985, constructed by 1987, and released in 1998, 84(5) (FG-84(5)) was targeted for 17.1 acres of wetland forested mixed and 2.3 acres of freshwater marsh. This site is notable for its soil characteristics. After two soil borings, Mr. Carter could not find a water table in the first 80 inches beneath the surface. However, he found an A horizon, but the CDA notes that this site received 18 inches of donor topsoil. Even more recent reclamation projects have tended to yield deep marshes. Permitted in 1997, constructed in 2002, and not yet released, 198-acre P-20 (FG-HC(9)) exists behind the berm that remains from the ditch and berm system that existed during mining. The sole outlet of the marsh is a discharge pipe, which, presently clogged with vegetation, appears to be contributing to excessively high water depths and excessively long hydroperiods, resulting in an abrupt transition from marsh to uplands without the zonal wetlands associated with natural transitions from marsh to uplands. Water in the marsh spreads into the surrounding uplands, which are planted with upland trees. The berm also prevents natural communication between the marsh and the floodplain of Horse Creek, which is a short distance to the west of P-20. In the reclamation projects described above, more often than not, the reclamation scientists reclaimed deep marshes while targeting shallower wetland systems or at least shallower marshes or swamps. By the mid-1980s, wetlands reclamation scientists were addressing more closely hydrology, vegetation, topsoil, and surrounding upland design, and DEP was imposing post-reclamation monitoring requirements on the phosphate mining companies. One common feature of most of these deep-marsh reclamations is their reliance upon artificial drainage outlets. Inadequate or nonexistent maintenance of these outlets causes excessive water depths for excessive periods. Additionally, reliance on artificial drainage outlets betrays the choice not to attempt more sophisticated design and more precise contouring of the post-reclamation landscape. Improvements in the design and execution of contouring could produce relief from the deep- marsh tendencies of reclamation practices in at least three ways: by flattening the slopes of the edges of the marshes to encourage the formation of more emergent vegetation and wet prairie fringes; introducing a more irregular microtopography in the submerged bottom, including hummocks, to develop greater habitat diversity; and engineering and grading more closely the topographical outlets of marshes, instead of relying on manmade drainage devices that required more maintenance than they received, to better reproduce pre-mining drainage features and access effectively the reclaimed water table. After 8.4-Acre Wetland, reclamation scientists produced, in addition to the P-20s, other marshes with better fringes, so as to support wet prairie fringes, but the most, and evidently only, successful example of shallow-wetland reclamation over an extensive area is PC-SP(2D) (SP-2D). Permitted in 1988, constructed in 1992, and released in 1998 (wetlands), SP(2D) comprises 97 acres of forested and herbaceous wetlands. According to Mr. Winchester, SP-2D exhibits a more natural hydroperiod than the other reclaimed wetlands that he studied. Mr. Winchester visited SP-2D during the dry season, and the shallow wetland was appropriately dry, even though other reclaimed wetlands at the time were inappropriately wet. Mr. Winchester also found less than ten percent coverage by exotic vegetation. Wet prairie fringes deeper marsh at SP-2D, rather than forming larger areas of isolated or connected wet prairie, but this wetland achieves extensive shallow-water areas. According to Authority ecologist Charles Courtney, the marsh of SP-2D appears fairly healthy and contains appropriate vegetation. SP-2D contains sawgrass and forbs, including maidencane and duck potato. Crayfish occupy the wet prairie fringe and are eaten by white ibis and otter. The marsh zonation found at SP-2D is partly a result of appropriate soil reclamation. Mr. Carter found good communication between the shallow marsh at SP(2D) and the surficial aquifer. In the wet season, Mr. Carter found the water table at eight inches above grade, demonstrating that the dry conditions found by Mr. Winchester during the dry season did not extend inappropriately into the wet season. Mr. Carter determined that the first four inches of the wetland is mulched topsoil overlying at least four feet of sand tailings. The subsurface soils were appropriately saturated. Permitted in 2002, constructed in 2003, and not yet released, 1.3-acre FCL-NRM(1) (Regional Tract O, ACOE #362) also contains wet prairie vegetation, but the value of this site, for present purposes, is limited by two factors: its age and its use of a technique not proposed for OFG. Regional Tract O, ACOE #362, is a new site that showcases the success--one year after planting--of the technique of cutting wet prairie sod at a donor site and laying it at the recipient site. Sod-cutting is a good technique, earlier used at Morrow Swamp, but is more expensive than the topsoil transfer proposed for OFG. The reclamation of forested wetlands has improved in recent years. To some extent, the history of forested-wetlands reclamation tracks the path of herbaceous-wetlands reclamation: deeper water for longer periods followed by instances of shallower water for shorter periods. Early in the forested-wetlands reclamation process, reclamation scientists and phosphate mining companies favored cypress trees due to their tolerance of a wider range of water depths and hydroperiods than other wetland trees. However, cypress trees do not occur naturally in the forested wetlands being mined in this part of Florida. Over time, reclamation scientists deemphasized the number of species of wetland trees and emphasized instead species that corresponded to those in comparable forested wetlands. Herbaceous and forested wetlands present different reclamation challenges due to the time each type of wetland requires for revegetation. An herbaceous wetland takes 1-2 years to revegetate, but a forested wetland may take 1-2 decades to gain "really good structure," as Dr. Clewell testified. In addition to taking longer to establish than herbaceous wetlands, forested wetlands require two stages of plantings because the groundcover cannot be added until 4-5 years after planting the trees, so that the trees provide sufficient cover for the appropriate groundcover to grow. The hydrological requirements of different forested wetlands vary. IMC will be reclaiming mostly mixed wetland hardwoods (44 acres), bay swamps and wetland forested mix (each 18 acres), and hydric pine flatwoods (15 acres). All of these communities require water depths equal to those required by wet prairies. Hydric pine flatwoods have a very short hydroperiod-- shorter even than the wet prairie. Bay swamps have a long hydroperiod, comparable to that of the freshwater marsh. And mixed wetland hardwoods and wetland forested mix have hydroperiods roughly equal to that of the wet prairie. The dryness required by mixed wetland hardwoods, wetland forested mix, and especially hydric pine flatwoods make them difficult to reclaim. At first glance, the longer hydroperiod of the bay swamp would seem to make it easier to reclaim, among forested wetlands, but two factors make the bay swamp the most difficult of forested wetlands to reclaim. First, as defined in these cases, the bay swamp provides a critical seepage function, which is hard to create because of its reliance on a precise reclamation of topography, hydrology, and soils, at least with respect to the soil-drainage characteristics. Second, the mucky soils of the bay swamps are difficult to reclaim, given their slow rate of formation, as noted above. Thus, even without the requirement of the dominance of bay trees within the bay swamp, as defined in these cases, bay swamps are very difficult to reclaim, as reclamation experience bears out. An early reclaimed forested wetland is 4.9-acre Bay Swamp (BF-1), which was created on land that had been cleared, but at least large portions of it were never mined, so, except possibly for a disturbed A horizon, the pre-mining soils and site hydrology were intact. Permitted under a predecessor program in 1979, constructed by 1980, and released in 1982, Bay Swamp earned restrained praise from the Authority as, with Dogleg Branch, one of the two highest-functioning reclamation sites. This praise is quickly conditioned with the warning that Bay Swamp did not reclaim as a bay swamp, but as another type of forested wetland, albeit a relatively high functioning one. For all these reasons, Bay Swamp is of limited relevance in evaluating the success of forested wetlands reclamation projects. However, in commenting upon Bay Swamp, the CDA offers some insight into the evolution of reclamation design standards and objectives and the optimism of reclamation scientists when it notes the difficulty of establishing loblolly bay-dominated swamps, "apparent[ly because they require] perennially moist, or wet, soil that is not inundated. Heretofore, these moisture conditions have not been specified as an objective in reclamation design. If these moisture conditions were targeted for reclamation, loblolly bay swamp creation would likely become routine." Another candidate for a reclaimed bay swamp is Lake Branch Crossing (BF-ASP(2A)). Permitted in 1993 and modified in 1997, constructed in 1996, and not yet released, 13.4-acre Lake Branch Crossing contains numerous sweet bays, loblolly bays, and black gums. However, this site was replanted with 4000 trees in mid-2002, and over one-quarter of these trees are displaying signs of stress, so they may not survive. Lake Branch Crossing is bound by a berm with culverts, which may not share a common elevation. Lake Branch Crossing is another excessively deep wetland with an excessively long hydroperiod. Although Lake Branch Crossing exhibits some seepage, it derives its water from a nearby CSA with a much-higher elevation and thus does not compare to the seepage systems to be reclaimed at OFG. The final candidate for a reclaimed bay swamp is Hardee Lakes (FG-PC(1A)), which is a 76-acre wetland forested mixed at the top of the Payne Creek floodplain. Permitted in 1989 and modified in 1994, constructed by 1991, and released in 2000, Hardee Lakes (which is not Hardee Lakes topsoil--the uplands site described above) contains a narrow seepage slope between the berm along the edge of a reclaimed lake and the natural Payne Creek floodplain. Although Hardee Lakes contains some bay trees and operates as a seepage wetland, the setting is inapt for present purposes, given the narrow slope descending from the nearby reclaimed lake, which provides the water for the seepage system. Like Lake Branch Crossing, Hardee Lakes presents an unrealistically easy exercise in the reclamation of a seepage slope and is therefore irrelevant to these cases. At OFG, broader seepage slopes will receive much of their water from upgradient groundwater that is not derived from a lake or other surface water, so the reclamation scientists must reclaim more accurately the topography, hydrology, and soils, again, at least with respect to soil-drainage characteristics. Reclamation scientists monitored Hardee Lakes following reclamation. Besides the seepage slope described in the preceding paragraph, Hardee Lakes contains shallower wetlands, including productive wet prairie and mixed wetland hardwoods that are growing without the need of hummocks, but these areas appear to be more isolated than extensive. As IMC restoration ecologist John Kiefer noted, shallow swamps are better than deep swamps. Again, the tendency toward deeper reclaimed systems, even recently, has plagued reclaimed forested wetlands, such as Lake Branch Crossing, as it has plagued reclaimed herbaceous wetlands. Permitted in 1992 and modified in 1998, constructed in 2002, and not yet released, North Bradley (KC-HP(3) and PD-HP(1B)) was reclaimed for 12 acres of wetland hardwood forest, 21 acres of wetland conifer forest, and 87 acres of herbaceous marsh. North Bradley suffers from poor communication with its water table, as evidenced by Mr. Carter's discovery of a perched water table under the marshes and an excessively deep water table, at 48 inches, under the forested wetlands, as compared to a water table at 40 inches under the uplands. Although the marsh is present, the forested wetland is largely absent. The SP(2D) of forested reclamation projects is Dogleg Branch (L-SP(12A)). The 19.8-acre wetland component of Dogleg was targeted exclusively for wetland hardwood forest. Another 83 acres of Dogleg was reclaimed as upland hardwood forests. Permitted in 1983, constructed by 1984, and released in 1991 (uplands) and 1996 (wetlands), Dogleg's hydrology is better, as one reclaimed area reveals seepage from a mesic area sheetflowing into the stream channel, which was also reclaimed and is discussed in the following section. Due to its proximity to the reclaimed wetlands, this mesic area was probably part of the reclaimed uplands. According to the CDA, Dogleg received transfers of its own mulch and received several phases of tree plantings over several years. The CDA notes that Dogleg was the first forested wetland mitigation project under Florida's dredge and fill rules. Trees were established in part by the transplanting of rooted tree stumps. Forest herbs and shrubs and mature cabbage palms were transplanted from nearby donor sites. Despite these and other efforts, according to the CDA, "design flaws attributable to a lack of prior restoration experience required costly mid-course corrections." Due to high tree mortality, trees had to be replanted over 11 years. The CDA concludes that the problem was a depressed water table due to nearby ongoing mining operations--if Dogleg had a ditch and berm system, it certainly did not have recharge wells. Following mining, according to the July 1995 semi-annual report, over 30 acres of mine pits immediately east and north of the unmined headwaters of Dogleg were filled with sand tailings, which then released "[c]onsiderable in-bank storage of ground water from this sand[, which] has seeped ever since through Dogleg Preserve and into the replacement stream." Prior to the cessation of mining, though, Dogleg suffered dehydration. According to the CDA, due to the drawdown, the topsoil dried out, and the overburden, on which the topsoil had been placed, hardened in the dry season, retarding root extension. The actual soil conditions are described in greatest detail in the July 1995 semi-annual report, which states that 12 inches of topsoil overlaid the "overburden fill," which was "clayey sand." Repeated and persistent replanting of trees, seedlings, and saplings eventually succeeded in establishing an appropriate wetland forest, which, given the prevalence of hardwoods, would constitute the successful reclamation of a mixed wetland hardwoods community, given the negligible representation of cypress trees and other conifers at the site. As reclaimed, Dogleg hosts 24 different species of wetland trees, including all that occur on OFG. Dogleg's forested wetlands are functioning well, although the reclaimed uplands have a major cogongrass infestation. Permitted in 1985, constructed by 1987, and released in 1998, 19.4-acre FG-84(5) (84(5)) was targeted almost entirely for wetland forested mixed, and small areas within 84(5) have achieved this objective. However, reclamation scientists planted so many cypress trees that their dominance today precludes the application of the wetland forested mixed label to the overall wetland. Nonetheless, 84(5) is a relatively high- functioning forested wetland community today. Engineered to contain hummocks, 84(5) also featured the use of transferred topsoil overlying cast overburden to a depth of at least six feet. Despite the presence of the topsoil layer, the proximity of the cast overburden to the surface, without an intervening sand layer, may have discouraged the formation of an appropriate water table. Although drawing on a lake, 84(5) displayed, in one soil boring during the middle of the wet season, no water table--not even a perched one--through the first 80 inches below grade. A small strip of saturated soil existed at the surface, but the highly compacted and impermeable overburden prevented communication between the wetland and the surficial aquifer. The slopes of 84(5) are also excessively steep. Substantial efforts are required to reclaim the shallow herbaceous wetlands and forested wetlands to be reclaimed at OFG. Deeper marshes and swamps require less effort to reclaim, although they develop more often than targeted when the reclamation scientists overshoot the mark as to hydrology. For shallow wetland systems, which are more important to reclaim, the failures far outnumber the successes, even today, so considerable caution is required in mining high-functioning shallow wetland systems and considerable effort is required in their reclamation. No bay swamps have been reclaimed, except under atypical conditions. Streams The successful reclamation of streams has also proven elusive to reclamation scientists and the phosphate mining industry. Although only one reclamation of a high-functioning, extensive shallow herbaceous wetland exists, fringe and small- scale shallow wetlands have been reclaimed. The difference between the reclamation of shallow herbaceous wetlands and streams is that reclamation scientists have benefited from 25 years of trial and error in engineering shallow wetlands. No similar history exists in the engineering of streams. Only nine stream-reclamation sites are identified in these cases, and, as DEP contends, only one of these sites is successful: Dogleg Branch. And even Dogleg Branch fails to access its floodplain properly and probably never will. The biggest difference between shallow wetlands reclamation and stream reclamation is that, until OFG, the phosphate mining industry has not intensively designed stream-reclamation projects, so IMC and its reclamation scientists have little experience on which to draw. A wetlands-reclamation practice, as found in a Florida Institute of Phosphate Research study described by Mr. Irwin, has been to reclaim wetlands downslope from their pre-mining location. Concentrating reclaimed wetlands downslope facilitates the re-creation of supporting hydrology. For OFG, IMC proposes to relocate wetlands downslope--probably to good effect, given the reversion of OFG to cattle ranching, post- reclamation. However, an adverse aspect of this practice has been the mining of upslope, lower-order tributaries and their replacement with downslope deeper marshes. Although difficult to quantify, this and similar reclamation practices have resulted in the destruction, by phosphate mining, of many lower- order streams and their permanent loss to the watershed and ecosystem. When attempting to reclaim streams, rather than convert them to downslope marshes, the phosphate mining industry and reclamation scientists have enjoyed little success. Two reasons likely explain this poor record: the complexity of the functions of a lower-order stream system, including its riparian wetlands and floodplain, and an excessive reliance on the ability of streams, post-reclamation, to self-organize. The importance inherent in the stream, its riparian wetlands, and its floodplain, as a functional unit, is reflected in the decision of IMC to extend the no-mine area to Horse Creek and its 100-year floodplain. Dr. Durbin accurately observes that IMC and its 100-year floodplain are, respectively, the first and second most important natural resources present at OFG. Horse Creek's tributaries and their floodplains are important for many of the same reasons. Relying upon reclaimed systems to self-organize is an essential element of effective reclamation. Natural and anthropogenic forces shape all of the natural systems present at OFG, and these forces will shape the reclaimed systems. Good reclamation engineering accounts for the dynamic nature of these reclaimed systems by establishing initial conditions, such as natural outfalls instead of weirs and culverts, that can evolve productively in response to the forces to which they are subject and eventually become high functioning, self-sustaining ecosystems. On the continuum between intensively engineered reclamation projects and reclamation projects that rely on self- organization, stream-reclamation projects in the phosphate mining industry have so heavily emphasized the latter approach over the former that they may be said to have reclaimed streams incidentally. That is, reclamation scientists have reclaimed streams by contouring valleys so that the erosive process of flowing water would form a stream channel over time: often, a long time. At DEP's urging after the issuance of the Altman Final Order, IMC has introduced a much more intensively engineered stream-reclamation effort in its Stream Restoration Plan. The main problem in assessing the likelihood of the success of the highly engineered Stream Restoration Plan is its novelty. On the one hand, the incidental reclamation of streams typically has been so slow in restoring functions that a more intensively engineered plan could generate quick gains, at least in the replacement of the functions of low-functioning stream systems, such as those that have been substantially altered by agricultural uses. On the other hand, the Stream Restoration Plan has little success--and no engineered success--on which to build, and misdesigned elements could take longer to correct than the undesigned elements in an incidentally reclaimed stream. Thus, when the uncertainties of successful stream reclamation are combined with the complex functions of lower-order tributaries, their riparian wetlands, and their floodplains, the higher- functioning streams at OFG are less attractive candidates for mining and reclamation than even the shallow wetlands discussed above. Horse Creek's tributaries are not necessarily low- functioning due to their status as intermittently flowing, lower-order streams. Even intermittently flowing, lower-order streams, such as all of the tributaries of Horse Creek, restrict the erosion of sediment into higher-order streams, uptake nutrients, maintain appropriate pH levels, and provide useful habitat for macrobenthic communities, macroinvertebrates, amphibians, and small fish. Intermittently flowing lower-order streams attenuate floodwaters by diverting floodwaters into the streams' floodplains, thus reducing peak flows, extending the duration that floodwater is detained upstream, and increasing groundwater recharge and, thus, streamflow. Intermittently flowing lower-order streams also supply energy for higher-order streams and the organisms associated with these stream systems, as organic material from vegetation, algae, and fungi in the lower-order streams eventually is flushed downstream to serve as food sources to downstream organisms. The functions of streams, including intermittently flowing lower-order streams, become even more complex and difficult to replace when considered in relation to the functions of the riparian forested wetlands associated with many lower-order streams, such as the Stream 1e series. The riparian forested wetlands provide additional attenuation of floodwaters, as the trees impede the flow of floodwater more than would ground-hugging herbaceous vegetation. Mature trees lining the stream provide a canopy that can cool the waters in the warmer months (thus reducing water loss to evaporation), provide downstream food in the form of leaf litter in the seasonal loss of leaves, shield interior water and habitats from the effects of wind, provide habitat for feeding and hiding for wildlife, and protect the channel from the impact of cattle (thus reducing the damage from the production of waste and turbidity and destruction of the channel and vegetation). The riparian forested wetlands are important in the sequestration of nutrients. If accompanied by flow-through wetland systems, such as those present in the Stream 1e series, riparian forested wetlands display a complex interrelationship between the roots and soils that contributes to improved water quality, among other things. The riparian forested wetlands also provide microhabitats whose detail and design would defy the restoration efforts of even the most dedicated of stream- restoration specialists, of whom IMC's stream-restoration scientist, John Kiefer, is one. For some of the stream-restoration projects, DEP explicitly permitted or approved the reclamation of a stream. For other such projects, DEP, at best, implicitly permitted or approved the reclamation of a stream. Four of the projects are tributaries to the South Prong Alafia River and are in close proximity to each other. From upstream to downstream, they are Dogleg Branch, whose forested wetland component has been discussed above; Lizard Branch (IMC-L-SP(10)); Jamerson Junior (IMC-L-CFB(1)); and Hall's Branch (BP-L-SPA(1)). Hall's Branch is about 4-5 miles upstream from the confluence of the South Prong Alafia River and North Prong Alafia River. All four of these reclaimed streams are now part of the Alafia River State Park. As noted above, Dogleg, a 19.8-acre wetland hardwood forest and 83-acre upland hardwood forest, was constructed in 1984 and is the oldest of these four reclamation sites adjoining the South Prong Alafia River. Next oldest is Hall's Branch, which was permitted as a 3.8-acre wetland hardwood forest in 1982, constructed by 1985, and released in 1996. Next oldest is Jamerson Junior, which was permitted as a 4.3-acre wetland forested mixed in 1984, constructed in 1986, and released in 1996. Ten years younger than the others is Lizard Branch, which was permitted in 1983 and modified in 1991, constructed in 1994, and released in 1996; some question exists as to its target community, but it was probably a swamp. The reclaimed stream at Dogleg Branch is part of a second-order stream, although the CDA reports that Dogleg Branch was a first-order stream. Pre-mining, Dogleg Branch and Lizard Branch joined prior to emptying into South Prong Alafia River. Portions of the record suggest that the reclaimed stream lies between unmined stream segments upstream and downstream, although one exhibit, cited below, implies that the mining captured the point at which the stream started. The CDA and the July 1995 semi-annual report state that the headwaters of Dogleg were unmined or preserved. The CDA adds, with more detail than the other sources, that the headwater and first 600 feet of the stream were unmined, and the next 1000 feet, down to the forested riparian corridor of South Prong Alafia River, was mined. Due to its detail, the CDA version is credited, as is the July 1995 semi-annual report: the headwaters of Dogleg Branch are unmined. The July 1995 semi-annual report states that the stream-reclamation component of Dogleg Branch required persistence, as did its forested wetlands component. In 1987, one year after the filling of the mine cuts with sand tailings, as described above, it was necessary to cut a new channel, because the gradient of the old reclaimed channel was too shallow and forced water to back up in the unmined headwaters. Reflective of the age of the reclaimed stream, the understory vegetative species associated with Dogleg Branch are more successional, having replaced the lower-functioning pioneer vegetative species that first predominated after reclamation. As a stream-reclamation project, Dogleg Branch has achieved close to the same success that it has achieved as a reclaimed wetlands forest or that SP(2D) has achieved as an extensive herbaceous shallow water wetland. The slope of Dogleg Branch's reclaimed channel is steeper than the slopes of its unmined channels, and the reclaimed segment, which functions well vertically within the banks of the channel, does not access its floodplain properly, largely due to its entrenched nature. Due to the entrenchment underway, it is unlikely that the reclaimed segment of Dogleg Branch will ever communicate with its floodplain, as its unmined segments do. Entrenchment is a measure of channel incision-- specifically, the width of the floodprone area, at a water level at twice bankfull, divided by the bankfull width. Entrenchment may cause excessive erosion, which may result in adverse downstream conditions, such as turbidity and lost habitat. Proceeding perpendicular to the flow of the water, entrenchment extends the channel into the riparian wetlands or uplands alongside the stream, dewatering any nearby wetlands and disturbing the local hydrology. Especially if entrenchment is associated with head-cutting, which operates up the streambed, the resulting erosion deepens the channel sufficiently that the water in major storm events can no longer enter its floodplain, but rushes instead downstream. Although the failure of Dogleg Branch to access its floodplain would not affect macroinvertebrates, which do not use the floodplains, the failure of the reclaimed stream to access its floodplain harms fish, which cannot access the floodplain during high water levels to forage, spawn, and escape predators or high water volumes, and reduces valuable aquatic-upland ecotones. This failure also reduces the ability of the stream to attenuate floodwaters. By chance, Charlotte County's stream- restoration expert Frederick Koonce visited Dogleg Branch shortly after a June 2003 storm event and saw the water from the stream enter the floodplains adjacent to the unmined segments of Dogleg Branch, but not the reclaimed segment. The less-rigorous approach of incidental stream restoration, at least in the mid-1990s, is evident the summer 1994 semi-annual report on Dogleg Branch, in which Dr. Clewell provides a detailed discussion of the biological aspects of the reclamation of this site. Implying that the incidental stream element of the Dogleg reclamation project may be nine years younger than provided in the parties' stipulation, Dr. Clewell writes: The temporary land use area was abandoned and reclaimed during the autumn of 1993. The perimeter canal was filled and the access road removed between Dogleg marsh and the unmined tip of original Dogleg Branch. Within a few days of a site inspection on December 2, 1993, final grading and revegetation had been completed, and water was discharging from Dogleg marsh into unmined Dogleg Branch for the first time ever. The water was free of turbidity. The entire connection had been sodded with bahiagrass turf. Dogleg Branch enjoys good water quality. On the two days that Charlotte County water quality scientist William Dunson tested its waters, in October 2003 and March 2004, the reclaimed Dogleg Branch had dissolved oxygen of 6.8 and 8.6 mg/l, iron of 325 and 212 ug/l, manganese of 41 and 22 ug/l, and aluminum of 160 and 132 ug/l. The Class III water standard for dissolved oxygen is 5 mg/l, except that daily and seasonal fluctuations above 5 mg/l must be maintained. The Class III water standard for iron is no more than 1.0 mg/l (or 1000 ug/l). There are no Class III water standards for manganese and aluminum. Dogleg Branch also passed chronic toxicity testing for reproductivity and malformation. However, Dogleg Branch is distinguishable from at least one of the OFG streams. Dogleg Branch is a much less complex restoration project because reclamation scientists did not need to re-create headwaters, the first 600 feet of stream downstream of the headwaters, or flow-through wetlands. Also, the mined segment of Dogleg was much shorter than the mined segment of the Stream 1e series: 1000 feet versus 2039 feet for the Stream 1e series. Betraying an emphasis on forested wetlands to the exclusion of streams, Dr. Clewell places Hall's Branch a close second to Dogleg among stream-reclamation projects. However, DEP properly did not add a second stream to its list of successful stream-reclamation projects. Reclaimed Hall's Branch is not close to performing the functions of reclaimed Dogleg Branch, and, because of the large gap between Dogleg and all of the other reclaimed streams, it is irrelevant which of them occupies second place. The most visible shortcoming of the reclaimed stream at Hall's Branch is its color. Parts of the water in the reclaimed stream within Hall's Branch are highly discolored with iron flocculent leaching from the surrounding mesic forest and shrub communities. Mr. Dunson's water quality tests in reclaimed Hall's Branch, in October 2003 and March 2004, revealed iron levels of 117,000 ug/l and 4025 ug/l, which are 117 times and 4 times the Class III water standard. Dissolved oxygen was also well below Class III standards at 1.5 mg/l and 2.1 mg/l. Manganese was 1880 ug/l and 392 ug/l, and aluminum was 226 ug/l and 35 ug/l. Like Dogleg Branch, Hall's Branch also passed chronic toxicity tests for reproductivity and malformation. The hydrological connection between the surficial aquifer and the reclaimed stream at Hall's Branch is probably interrupted. Mr. Carter, who did not visit Dogleg Branch, inspected Hall's Branch and found the water table 12 inches below the surface. A soil sample reveals overburden with a layer of topsoil. The CDA seems to indicate that part of Hall's Branch was backfilled with sand tailings of an unspecified depth and part of it was merely contoured overburden--a pattern suggestive of that planned for OFG. The CDA states that trees were planted in mulched areas. The reclaimed forest is dominated by cypress, not the targeted wetland hardwoods. Jamerson Junior is a 4.3-acre reclamation site permitted as a wetland forested mixed community in 1984, constructed by late 1985, and released in early 1996. Part of the reclaimed stream is a second-order stream. Like Hall's Branch, Jamerson Junior also shows signs of orange-colored water leaching in to the stream from the nearby mesic zone. However, the water quality in Jamerson Junior is closer to the water quality in Dogleg Branch than Hall's Branch. Mr. Dunson's iron readings, in October 2003 and March 2004, were 583 ug/l and 195 ug/l, which are within Class III standards. Dissolved oxygen was slightly higher than at Dogleg Branch: 7.0 mg/l and 8.0 mg/l. Manganese was 136 ug/l and 21 ug/l, and aluminum was 391 ug/l and 101 ug/l. However, Jamerson Junior failed chronic toxicity testing for reproductivity, but passed for malformation. This is the only stream that IMC also tested for toxicity, and IMC obtained similar results, according to Dr. Durbin. Soil samples reveal a highly variable soil structure underlying Jamerson Junior. Subsequent reclamation work on the stream required the addition of material to change the elevation of the stream bed and possibly to change the drainage characteristics of the original backfilled material. On the day that Mr. Carter visited Jamerson Junior on August 14, 2003, he found the stream flowing. During the wet season, the water table should normally be expressed in the stream. Presenting a more interrupted relationship between the surficial aquifer and the stream than at Hall's Branch, Jamerson Junior displays no connection between the stream bed and water table, at least to a depth of 40 inches. A soil boring revealed water immediately underneath the stream bed, but, at about 15 inches beneath the bottom of the bed, the soil dried to moist; at 40 inches, Mr. Carter found the water table under the stream. Likewise, the Jamerson Junior channel was poorly integrated with the surrounding wetlands and uplands. At the banks of the stream, Mr. Carter did not find the water table within 80 inches of the surface, which is additional evidence of a discontinuity between the water table and the stream. Much of the reclaimed forested areas are mesic, not hydric. The reclaimed floodplains are narrower than the floodplains in the unmined adjacent area, and the slope of the reclaimed channel is steeper than the slope of the unmined channel. The reclaimed uplands are infested with cogongrass, although less than is present at Dogleg. Lizard Branch is a 6-acre reclamation site permitted as a swamp community in 1983 and modified in 1991, constructed by 1994, and released in 1996. Few of the planted gums and maples are surviving. The uplands surrounding the reclaimed area are infested with cogongrass, which has penetrated the shallower wetlands. Lizard Branch is one of the lowest- functioning forested wetlands. Lizard Branch joins Jamerson Junior as one of only two of six reclaimed stream sites to fail chronic toxicity testing for reproduction, although it passed for malformation. Lizard Branch had the highest two dissolved oxygen readings of all six sites tested by Mr. Dunson: 12.6 mg/l and 7.1 mg/l. Its iron levels were 547 ug/l and 352 ug/l. Manganese was second lowest, behind only Dogleg Branch, at 71 ug/l and 30 ug/l. Aluminum was second highest at 445 ug/l and 45 ug/l. Lizard Branch is an interesting, recent reclamation site for several reasons. Lizard Branch represents a relatively recent instance of the destruction of a stream without its re- creation and either the failure of the incidental reclamation of a stream or the subsequent permission by DEP to allow the permanent elimination of the stream. Mr. Winchester testified that he could not even find a stream at Lizard Branch. Charlotte County ichthyologist Thomas Fraser treated Lizard Branch as a stream, but grouped it with marshes in his analysis, apparently due to the lack of channel formation. The fact is that, despite any effort to reclaim a stream, little, if any, stream structure is present at Lizard Branch. However, a stream once flowed over the reclaimed portion of Lizard Branch. In the summer 1994 semi-annual report, Dr. Clewell notes that Brewster Phosphate received a dredge and fill permit in 1983 to dredge and fill the "headwaters of two streams, Dogleg Branch and Lizard Branch" in connection with the mining at Lonesome Mine. Dr. Clewell adds: The permit was issued with the stipulation that the streams and their attendant riverine forest would be restored on adjacent physically reclaimed lands, concomitant with mining. The permit further stipulated that restoration would be monitored and that semi-annual reports documenting progress in vegetational restoration would be submitted to [DEP.] In the report, Dr. Clewell notes that reporting on Lizard Branch has been "discontinued" and DEP issued a new permit in 1991. The 1991 permit modification is not part of this record, but the result was the elimination of a stream, or at least any signs of a stream ten years after construction. Three of the remaining reclaimed-stream projects were built at about the same time as Lizard Branch project. For only one of these projects did the reclamation scientists explicitly target a stream. Permitted in 1985 and subject to a consent order in 1996, constructed in 1991-92 and 1995, and not yet released, 9.6-acre Tadpole Wetland (H-SPA(1)) was targeted to be about one-third wetland forested mix and two-thirds freshwater marsh. Much cogongrass has infested Tadpole, whose stream enters the Alafia River floodplain and leads to a ditch that runs the remainder of the distance to a point close to the Alafia River. Tadpole's water passed chronic toxicity testing for reproductivity and malformation. However, its water violated Class III standards for dissolved oxygen, with readings of 2.8 mg/l and 2.1 mg/l, and for iron, with readings of 11,300 ug/l and 1100 ug/l. Manganese levels were 166 ug/l and 20 ug/l, and aluminum levels were 660 ug/l--the single highest reading among the four reclaimed streams tested--and 95 ug/l. Permitted in 1985, constructed by 1996, and not yet released, Pickle Wetland (H-SPA(1)) is a 34-acre site, 0.8 acres of which was to be reclaimed as stream. A deep marsh that requires treatment of its nuisance exotics, such as cattails and primrose willow, Pickle is just northeast of Tadpole and a few miles north of Morrow Swamp and Ag East. Pickle's stream is surrounded by uplands. Pickle is the only reclaimed stream of six tested to fail chronic toxicity testing for malformation, although it passed for reproductivity. Pickle has the lowest dissolved oxygen of the six reclaimed streams tested by Mr. Dunson: 0.8 mg/l and 1.2 mg/l. Its iron levels violated Class III standards in October 2003, with a level of 4230 ug/l, but passed in March 2004, with a level of 786 ug/l. Manganese was 127 ug/l and 72 ug/l, and aluminum was 107 ug/l and less than 5 ug/l. Permitted in 1991, constructed in 1995, and not yet released, Trib A ((BF-ASP(2A)) is a 120-acre site to be reclaimed as a wetland forested mix, but it includes a slough that empties into an unmined channel with streamflow. To the extent that a reclaimed stream channel is discernible on Trib A, nine years after the completion of its reclamation, the channel is much more steeply sloped than the adjacent unmined channel-- steeper than the two percent slope, beyond which sandy stream bottoms begin to erode. Not surprisingly, the reclaimed channel has begun to head cut and entrench. In an adjacent unmined area, a stream exists within a floodplain with a very flat slope. In the mined area, the reclaimed floodplain is steeper, suggestive of impeded communication between the reclaimed stream and its floodplain. The groundwater communication at Trib A is almost as interrupted as it was at Jamerson Junior. At Trib A, the uppermost 20 inches of soil was saturated, at the time of Mr. Carter's site inspection. Beneath a moist soil layer, the water table occurred at 40-50 inches deep. Parts of Trib A were topsoiled, but the next layer down was originally from an area below the C horizon. However, the soil-formation process is underway. Permitted in 1995, constructed by 1998, and not yet released, 17.6-acre File 20-2B and 70-3 Dinosaur Wetland (FG- GSB(7)) was reclaimed as a freshwater marsh. Dinosaur is due south of Morrow Swamp and is a headwater wetland. The site is still undergoing treatment for cattails. The record describes little, if anything, about the status of this stream. The last two stream-reclamation reclamations were built at least five years after the last pair. Again, DEP and the phosphate mining company identified a stream as a target for only one of the projects. Permitted in 1989, 1992, and 1998, constructed in 1999, and not yet released, South Bradley (KC-HP(1A) is a 171- acre site, 1.7 acres of which was to be reclaimed as stream. South Bradley is just north of Pickle. The channel is steeply incised and deep at points. The channel runs through forested and unforested areas. Charlotte County ichthyologist Thomas Fraser found iron flocculent in South Bradley and no fish within this area of the reclaimed stream, but three species of fish in a nearby area. Permitted in 1999, constructed by 2003, and not yet released, MU R Wetland H (KC-HB(1)) is a 4.8-acre site to be reclaimed as wetland hardwood forest. Monitoring has not yet begun for this site. Although a tailwater system receiving water from a ditch running to a lake, rather than a natural stream, the channel that has formed in MU R Wetland H does not join the existing downstream channel; the two channels are offset by 75-100 feet. Also, the reclaimed floodplain of MU R Wetland H is more steeply sloped than the floodplain of the adjacent unmined area. The slope of the reclaimed channel is steeper than the slope of the unmined channel, and, due to poor design parameters, the new channel is headcutting into the floodplain, which does not appear to be communicating appropriately with the stream. Combining a more steeply sloped reclaimed floodplain with a headcutting reclaimed stream means, among other things, substantially less communication between the stream and its floodplain. The hydrology of MU Wetland H appears to have been ineffectively reclaimed. In the forested wetland a short distance from the stream, the soil remained unsaturated until 80 inches deep. Closer to the stream, the soil was saturated at a depth of 18-20 inches, but the underlying overburden remained dry to a depth of 70 inches, indicating again a failure to reclaim the water table at appropriate depths. As with all of the almost countless reclamation sites on which the parties' expert witnesses copiously opined, MU R Wetland H is not well-developed in the record in terms of pre- mining conditions, design elements, construction techniques, and post-reclamation conditions. However, the dislocated stream that has formed within this reclaimed wetland stream reinforces the principle that even incidental stream reclamation requires some engineering. The excessive reliance upon a contoured valley to self-organize into a stream, as noted above, has impeded the progress of the science of stream restoration, as applied to mined land in Florida. This factor is unique to streams and does not apply to uplands and wetlands. However, another factor has impeded progress in reclaiming successful systems--whether uplands, wetlands, or streams. This factor is undue emphasis on the identity of post-reclamation vegetation, as compared to pre- mining or reference vegetation, at the expense of function. Charlotte County and the Authority stressed the process of the identification of vegetative species, at the expense of undertaking complex functional analysis and attempting to situate reclaimed systems in the process of energy consumption and production. In part, their cases relied on showing that past reclamation projects, as well as that proposed for OFG, do not replicate pre-mining or reference-site vegetation. An undue emphasis on species identity suffers from two major flaws. First, as Dr. Clewell and Ms. Keenan testified, reclaimed sites undergo stages of colonization, and, during early stages, less-desirable species, such as Carolina willow and wax myrtle, may predominate at more-desirable canopy-forming species succeed them. Ms. Keenan added that the life expectancy of Carolina willow, in this part of Florida, is about 25 years, and no reclaimed site older than 15 years is dominated by Carolina willow. Second, any measure of species identity risks the elevation of replication over function, as DEP has already recognized. A criterion of replication, for example, discredits a reclaimed site with a lower species-identity score because it has been colonized by a greater share of more-desirable species than occupy the reference site. DEP has wisely discontinued the practice of assessing reclamation success in partial reliance upon the Morisita's Index. This index measures the identity of species between two sites or the same site pre-mining and post-reclamation, as a criterion of successful wetlands reclamation. In a similar vein, DEP has recently recognized that vegetative analysis cannot preemption functional analysis, especially as to streams. This recognition is evidenced by a report entitled, "Riparian Wetland Mitigation: Development of Assessment Methods, Success Criteria and Mitigation Guidelines," which was managed by Ms. Keenan, revised May 10, 2001, and filed with the U.S. Environmental Protection Agency Grants Management Office (Riparian Wetland Mitigation). Riparian Wetland Mitigation notes the unsatisfactory history of stream reclamation projects with their emphasis on vegetation to the exclusion of stream hydrology and geomorphology. Riparian Wetland Mitigation states: The more recent methods [of stream restoration] recognize that streams are not simply water conveyance structures, but are complex systems dependent on a variety of hydrological, morphological, and biological characteristics. It is now recognized that in order to successfully restore or create a stream, hydrology, geology and morphology must be considered in the design. Noting the increasing extent to which the phosphate mining industry is applying for permits to mine more and larger stream systems and reclaim them on mined land, Riparian Wetland Mitigation frankly admits: The success criteria included in permits issued by the Department for these newly created streams have been based primarily on vegetational characteristics as is typical of most permits requiring wetland mitigation. However, vegetation alone is a poor indicator of stream function and community health. The results of regular permit compliance inspections of existing stream mitigation projects . . . have suggested that for several projects, although existing riparian vegetation was meeting or trending toward meeting permit requirements, problems existed with site hydrology and habitat quality of the stream channel itself. DEP thus adopted a rapid bioassessment method known as BioRecon, which tests macroinvertebrates, and added two other components: habitat assessment and physical/chemical characterization. DEP then performed "BioRecon, habitat assessment, and physical/chemical sampling" on eight reclaimed streams. Of the eight sites sampled, "only one passed the BioRecon and Habitat Assessment." (It is unclear whether Riparian Wetland Mitigation intends to imply that this site-- obviously, Dogleg Branch--also passed the physical/chemical composition, but it probably did.) DEP then tested smaller, unmined streams and confirmed that they, too, could pass BioRecon and Habitat Assessment. Riparian Wetland Mitigation states that DEP will collect data from comparable unmined streams and attempt to relate geomorphological, hydrological, and biological data to develop more refined criteria by which to assess proposed stream-reclamation projects. When DEP issues these criteria, the likelihood of success of a specific stream-reclamation project will be easier to assess. Until then, the assessment of a specific stream-reclamation project remains more difficult, in the context of past reclamation projects that have reduced or even eliminated important functions of streams. Although DEP's new guidelines for stream restoration will mark a transition from a predominantly vegetative to a multi-variable analysis of stream function, even a predominantly vegetative analysis of stream function is superior to IMC's analysis of streams predominantly from the perspective of flood control, as set forth in the CDA prior to the Altman Final Order. In a remarkably candid admission of the difficulty of reclaiming the many functions of unaltered stream systems, including their riparian wetlands and floodplains, IMC, in its response to RAI-102 in the CDA, states: Although it is impossible in a reasonable amount of time to expect to restore the functionality of the creek systems and associated uplands which historically occurred on the One site and are proposed for mining, it is reasonable to conclude that the reclamation plan restores the primary functions of the watershed[:] i.e. the capture, storage, distribution, and release of precipitation. IMC's subsequent discussion in RAI-102 emphasizes the efficacy of mitigation, from a biological perspective, but only as to stream systems whose pre-mining condition is substantially altered. For relatively unaltered systems, IMC's message remains that the reclamation of functions, besides water quantity, within a reasonable period of time is "impossible." Summary of Findings on Past Mitigation/Reclamation Any attempt at assessing past reclamation projects is impeded by the general lack of data presently available, for each reclamation site, describing pre-mining hydrological, topographical, soil, and geological conditions; the functions of pre-mining communities; reclamation techniques; post-reclamation hydrological, topographical, soil, and geological conditions; and the functions, as they have evolved over time, of reclaimed communities. For post-reclamation water tables, the auger and shovel work of one or two men substitutes for several years of weekly piezometer readings in the wet season and monthly piezometer readings in the dry season--correlated to daily rainfall data collected at the same site. For post-reclamation water quality, a few preliminary toxicity and a few dozen water quality readings--some under less than optimal conditions-- substitute for systematic water-quality testing of a broad range of parameters, again over years. For post-reclamation soils, one soil scientists finds an A horizon and concludes substantial formation has taken place within 10 years; another finds an A horizon--never the same one at the same place--and concludes topsoil transfer; and both are probably correct. Absent better data, reliable analysis is difficult because a wide variety of factors may have contributed to the successes of SP(2D) and Dogleg and the failures of too many other sites to list. Even so, a few facts emerge. IMC can reclaim extensive areas of uplands, deep marshes, and cypress swamps, although difficulties remain with each of these types of reclamation projects. With greater difficulty, IMC can reclaim pine flatwoods and palmetto prairies. With even greater difficulty, IMC can also reclaim forested wetlands, except bay swamps. Far more difficult to reclaim than the communities mentioned in the preceding paragraph are extensive shallow wetlands, seepage bayheads, and streams. Any finding of present ability to reclaim these systems must uneasily account for the numerous failures littering the landscape, the failure ever to reclaim successfully a bayhead as bay swamps typically occur in the landscape, and the unsettling fact that nearly all reclamation successes of shallow wetlands are small patches-- almost always far smaller than designed. Any finding of present ability to reclaim these systems must rely heavily on SP(2D) and Dogleg Branch and the design of the current reclamation plan. The probability of the successful reclamation of any community, but especially extensive shallow wetlands, seepage bayheads, and streams, requires careful analysis of each community proposed to be mined and each community proposed to be reclaimed. For each such community, it is necessary to assess its ultimate functions of consuming and producing energy within a robust, sustainable ecosystem. Additional Features of OFG, Mining, and Reclamation Introduction The preceding sections detail the ERP, CRP approval, and WRP modification and other mitigation sites involving the reclamation of uplands, wetlands, and streams. This section adds information concerning OFG in its pre-mining condition, the proposed mining operations, and the proposed reclamation. OFG IMC adequately mapped the vegetative communities at OFG. As Doreen Donovan, IMC's wetlands biologist testified, trained persons using the FLUCFCS system of classifying vegetative communities tend to fall into one of two categories: lumpers or splitters. Scale dictates FLUCFCS code in many cases. Where one biologist may designate a larger, more varied area with one code, another biologist may designate the same area with several codes. The purpose of FLUCFCS coding dictates the scale. Subordinating vegetative-identity analysis to functional analysis undermines the arguments of Charlotte County and the Authority for an unrealistic level of precision in this exercise. The discrepancies in vegetative mapping noted by Mr. Erwin were insignificant. Many were the product of scaling differences, as noted in the preceding paragraph. Some were the product of distinctions without much, or any, difference, given the context and extent of the proposed activities. For present purposes, absent demonstrated differences in wildlife utilization, groundwater movement, or soil, distinctions between, for example, xeric oak and sand live oak on ten acres are essentially irrelevant. In total area, as compared to the 4197 acres of OFG, the claimed discrepancies did not rise to the level of noteworthy. As for the wetlands at OFG, DEP's acknowledged expert in wetlands identification, Deputy Director Cantrell, personally visited OFG and confirmed the accuracy of the wetlands determinations made three years earlier in December 2000 when DEP issued a Binding Wetland Jurisdictional Determination, which remains valid through December 2005. Deputy Director Cantrell noted minor omissions that might total a couple of acres, but these are insignificant, again given the scale of the proposed activity. The sole material flaw in IMC's mapping of OFG is in the omission of floodplains of the tributaries from Map C-3, although Dr. Garlanger's hydrological analysis, described below, adequately considered the storage and conveyance characteristics of these floodplains. Proper analysis of the tributaries' functions, besides flood control, and proposals to reclaim them is impeded by IMC's failure to depict graphically the 2.3-, 25-, and 100-year floodplains. The record suggests that BMR may have waived any requirement for maps of the floodplains except for those of Horse Creek, but the record does not suggest that, if BMR actually waived this requirement, it thus insulated the CDA from scrutiny with respect to all the information that would have been contained in floodplain maps or assured IMC of favorable analysis of this missing information. Charlotte County hydrologist John Loper prepared floodplain maps, which are Charlotte County Exhibits 1762 (mean annual floodplain), 1763 (25-year floodplain), and 1764 (100- year floodplain). These are credited as accurate depictions of the floodplains of the tributaries of Horse Creek. Mr. Loper's maps reveal little difference between the 25- and 100-year floodplains over much of OFG, including the Panhandle. The two floodplains of Stream 3e are slightly different, but the two floodplains of the Stream 1e series are less noticeably different. Focusing on the 25-year floodplain, the only wide, lengthy floodplain outside of the no-mine area is the floodplain along the Stream 1e series, which is the widest band of floodplain outside the no-mine area. At places, the floodplain of the Stream 1e series is as wide as the corresponding floodplain of Horse Creek. Even at its narrowest, which is along Stream 1ee, the floodplain of the Stream 1e series is as wide as that of Stream 2e and wider than that of Stream 3e. No 25-year floodplain runs along ditched Stream 3e?. The only other portions of the 25-year floodplain contiguous to the floodplain of Horse Creek, but outside the no-mine area, are the large wet prairie at the head of Stream 9w, the large wet prairie at the head of Stream 5w, and the headwater wetlands of Streams 1w-4w. As already noted and discussed in more detail below, all of these wetland systems, including the headwaters of Streams 1w and 3e, are lower-functioning than the wetland system associated with the Stream 1e series. As noted above, over half of the area to be mined is agricultural and another quarter of the area to be mined is uplands consisting largely of sand live oak, pine flatwoods, and palmetto prairie. Accordingly, OFG is characterized by native flatwoods soils, which exhibit high infiltration rates, but restricted percolation due to underlying hardpan or loamy horizons. About one-fifth of the soils at OFG are xeric soils. The wet season water table in the wetter areas will be 0-2 feet below grade and in the uplands over 3 feet below grade. Nothing in the record suggests that IMC will have much difficulty in reclaiming agricultural land or sand live oak communities. Nothing in the record suggests that any of the sand live oak that will be mined is atypically valuable habitat. As noted above, the pine flatwoods and palmetto prairie are more difficult to reclaim, but the pine flatwoods and palmetto prairie at OFG are not atypical instances of these common upland habitats. Some of these communities have been stressed by the lack of fire, so that hardwoods, such as oaks, have become sufficiently established as to resist thinning by fire. Lack of fire has also resulted in overgrown vegetation in more xeric areas. Among forested wetlands, IMC will mine 43 acres of mixed wetland hardwoods, 12 acres of hydric pine flatwoods, 9 acres of bay swamps, and 6 acres of hydric oak forests. Among herbaceous wetlands, IMC will mine 95 acres of wet prairie and 67 acres of freshwater marsh. Map F-3 depicts these wetlands with color-coding for ranges of wetlands values, under the Wetland Rapid Assessment Procedure (WRAP), which is used by the U.S. Army Corps of Engineers. Following a weeklong investigation of wetlands at the Ona Mine, as well as other IMC mines in the vicinity, the U.S. Army Corps of Engineers expressly approved revisions to WRAP to accommodate local conditions at OFG. DEP used a different assessment procedure, but WRAP remains useful for general indications of wetlands function. The WRAP scoring scale runs from 0-1, with 1.0 a perfect score. For ease of reading, the following sections shall identify wetlands scoring below 0.31 as very low functioning, wetlands scoring from 0.31 to 0.5 as low functioning, wetlands scoring from 0.51 to 0.7 as moderate functioning, wetlands scoring from 0.71-0.8 as high functioning, wetlands scoring from 0.81-0.9 as very high functioning, and wetlands scoring from 0.91-1.0 as the highest functioning. The asymmetry of the labeling scheme is to allow differentiation among the wetlands in the highest three categories, which, at OFG, are disproportionately represented, as compared to the lowest three categories. The purpose of these descriptors is only to differentiate relative values. As already discussed, the Map F-2 series identifies existing wetlands alphanumerically and by community, and Map I-2 similarly identifies all post-reclamation communities. In contrast to all reclaimed wetlands, which, as already noted, start with an "E" or "W," all existing wetlands start with a "G" or "H." The ease with which freshwater marshes are reclaimed obviates the necessity of extensively analyzing the condition of marshes presently at OFG, absent evidence of atypical habitat value. In general, the wetland corridor of Horse Creek, as defined by the no-mine area, ranges in quality from very high functioning in Section 29, which is the southernmost end of Horse Creek in OFG, to high functioning north of Section 29. However, narrow fringes of this corridor north of Section 29 are low functioning. Starting from the south, in Section 29, three wetlands are outside of the no-mine area: H031/H032/H033/H034, the G005 wetland complex, and a fringe of the wetlands running adjacent to Horse Creek--the western edges of G262, G266, and G259A are outside of the no-mine area. H031 is the largest part of the H031 complex and is mixed wetland hardwoods. H032 is a small freshwater marsh, and H033 is a hydric oak forest of the same size. H034 is a slightly larger wet prairie. H033 is low functioning. The remainder are high functioning. IMC will reclaim the same communities, as an ephemeral wetland complex. Pre-mining and post-reclamation, this wetland drains into West Fork Horse Creek. Considerably larger than H031, the G505 wetland complex is the headwater wetland of Stream 1w. G512 is the largest component of the G505 wetland complex and is wetland forested mixed. G513 is the next largest component and is a bay swamp. G514 is a fringe wet prairie. Slightly larger than G514, G511 is hydric oak forest. G507 is mixed wetland hardwoods, G506 is a small freshwater marsh, and G505 is a cattle pond. The mixed wetland hardwoods and fringe wet prairie are very high functioning, the bay swamp is high functioning, and the remaining wetlands are moderate functioning. IMC will reclaim the G505 wetland complex as a single bay swamp. G262 and G266 are wet prairie and hydric rangeland, respectively. G259A is mixed wetland hardwoods. The wet prairie and hydric rangeland are moderate functioning, and the mixed wetland hardwoods is very high functioning. IMC will reclaim these wetlands as wet prairie. Section 20 contains the headwater wetlands of Streams 2w, 3w, 4w, and 5w. These are mostly marshes, and they are all low to moderate functioning. These systems have been heavily impacted by agricultural uses. IMC will reclaim these as headwater systems, mostly marshes. IMC will also create one small and one medium ephemeral wet prairie near the headwater wetland of Stream 4w. Section 19, which drains to West Fork Horse Creek, contains three wet prairies (H002, H005, and H006) and a complex consisting of a bayhead (H009A) surrounded by a mixed wetland hardwoods (H009), which is fringed by a small wet prairie (H008). These wetlands are all low to moderate functioning. IMC will reclaim the H008 complex with a bay swamp buffered by a temperate hardwood, and it will restore a cattle pond at the site of the H002 complex. The reclaimed bay swamp will drain to West Fork Horse Creek. Section 18 contains a very low functioning, small wet prairie (H056), which is the only wetland in one of the three lowest ranges of WRAP scores outside of the wetland corridor of Horse Creek. Section 18 also contains a small part of a large wetland that is mostly in Section 17. The latter wetland is addressed in the discussion of wetlands in Section 17. Section 17 contains the West and Central Lobes. The entire Central Lobe is in the no-mine area, but a large wet prairie (G188) abuts the wetlands in the no-mine area of the West Lobe. IMC will reclaim this wet prairie, which is low functioning, as improved pasture, with a strip of hardwood conifer mixed. Several wetlands unassociated with the West and Central Lobes are outside the no-mine area, but on either side of Stream 6w, which leads to the West Lobe. G183, which is the headwater wetland of Stream 7w, is a freshwater marsh, which is moderate functioning. IMC will not reclaim the existing portion of Stream 7w upstream of the no-mine area, so the connected headwater marsh will be reclaimed as an ephemeral wet prairie. South of Stream 7w is a group of four small wetlands: G089, G090, G091/G092, and G093/G094. G089 and G090 are very small wet prairies. G091 and G093 are freshwater marshes, and G092 and G094 are wet prairie fringes. G090 is low functioning, and G089 and G091 are moderate functioning. G093 is very high functioning, and G094 is high functioning. Even the maps on the February submittal CD are unclear, but it appears that G089 and G090 will be reclaimed as ephemeral wet prairies. IMC will reclaim G091 as a small freshwater marsh fringed by a large mixed wetland hardwood and G093 as a large freshwater marsh fringed on the east by a small mixed wetland hardwood. The last version of Figure 13B-8 depicts the small freshwater marsh as isolated, but the large freshwater marsh as ephemeral. IMC will also create two small ephemeral wet prairies due south of the West Lobe and one small ephemeral wet prairie just east of the north end of the West Lobe. About one mile west of Horse Creek is a large wet prairie surrounding a smaller freshwater marsh that has been ditched for agricultural purposes. Part of this wet prairie extends into Section 18. The portion of this system in Section 18 is low functioning; the rest of it is moderate functioning. IMC will reclaim this entire area as improved pasture, except for replacing a single cattle pond. Section 16 spans Horse Creek, but mostly covers an area east of the stream, including the East Lobe. The only wetland outside the no-mine area on the west side of Horse Creek is G076/G077, a freshwater marsh fringed by a wet prairie. This small wetland is moderate functioning, and IMC will reclaim it as an ephemeral wet prairie. East of Horse Creek lies Stream 5e and its flow- through wetland, G204/G205. Predominantly a wet prairie, G204 is low functioning. IMC will reclaim it as a bay swamp. A small fringe wet prairie (G177) lies at the south end of the East Lobe, outside of the no-mine area, but it is low functioning, and IMC will reclaim it as hardwood-conifer mixed. A mixed wetland hardwood (G096), which is moderate functioning, fringed by a wet prairie (G097), which is low functioning, lie just north of where the no-mine area of the East Lobe joins the main no-mine area along Horse Creek. IMC will reclaim this wetland as a freshwater marsh fringed on the east by a wet prairie, and this wetland will be connected to the wetlands of the Horse Creek corridor. A freshwater marsh (G058) lies outside the no-mine area just north of the northeast tip of the East Lobe. This wetland is moderate functioning. IMC will reclaim this site as improved pasture, but will create a small ephemeral wet prairie just to the west of G058 and a larger freshwater marsh to the west of the created wet prairie. Section 8 contains two large areas of wet prairie (G048 and G047) at the head of Stream 9w. These wet prairies are moderate functioning, as are a couple of small wet prairies in Section 8 at the western boundary of OFG. IMC will reclaim these areas mostly as improved pasture, although it will create a large, connected wet prairie over the southeastern part of G048, but extending farther to the south and east. This reclaimed wet prairie will form the headwater wetland of reclaimed Stream 9w, which, as already mentioned, will be shortened from its current length. The only other wetland in Section 8 and outside the no-mine area is a freshwater marsh (G052). This marsh is high functioning. IMC will reclaim this site with a marsh and wet prairie. Like Section 16, Section 9 spans both sides of Horse Creek. On the west side of Horse Creek is mixed wetland hardwoods (G055) fringed by hydric woodland pasture (G054). The mixed wetland hardwoods is high functioning, and the hydric woodland pasture is moderate functioning. IMC will reclaim this site with a gum swamp fringed by temperate hardwoods upland. On the east side of Horse Creek, a small wet prairie (G167) is outside the no-mine area. This very high functioning wet prairie is connected to a large bay swamp (G166) to the north. The bay swamp, which is high functioning, lies partly within and partly outside the no-mine area and is connected to the wetland corridor of Horse Creek. Although high functioning, G166 is overdrained by a tile drain system that drains the citrus grove immediately upland and east of G166. Two mixed wetland hardwoods, which are outside the no-mine area, fringe the bay swamp; they are high functioning. IMC will reclaim a gum swamp for the wet prairie and all mixed wetland hardwoods for the east side of the bay swamp. Just north of the bay swamp that straddles the no- mine boundary is a much smaller bay swamp (G163) fringed by mixed wetland hardwoods (G164) that also straddle the no-mine boundary. Also connected to the wetland corridor of Horse Creek, these wetlands are very high functioning, and IMC will reclaim them with pine flatwoods. Between these two bay swamps straddling the no-mine boundary and the headwater wetland of Stream 8e is a small wet prairie (G041), which is moderate functioning and outside the no-mine area. IMC will reclaim this site with another ephemeral wet prairie. At the southern tip of the headwater wetland of Stream 8e is hydric flatwoods (G157), which is moderate functioning. IMC will reclaim this connected wetland with sand pine flatwoods. A smaller hydric woodland pasture (G154) also connects to another section of hydric flatwoods, which is in the no-mine area between the headwater wetlands of Streams 8e and 7e. The hydric woodland pasture is moderate functioning, and IMC will replace it with hardwood-conifer mixed, although IMC will reclaim a somewhat larger area of mixed wetland hardwoods just north of the present site of the hydric woodland pasture, where no wetland presently exists. The remaining wetlands outside the no-mine area in Section 9 are six isolated wet prairies. They are small wetlands, except for G039/G040, which is a wet prairie fringing a cattle pond, and G039, which is at the eastern boundary of OFG. However, they are all high functioning, even the wet prairie fringing the cattle pond. In this general area, IMC reclaims three ephemeral wet prairies, much closer to the no- mine area than the sites of the six isolated wet prairies, and a small freshwater marsh fringed by a community that is not listed in the legend in Map I-2. Interestingly, IMC also reclaims a large area of shrub and brushland and larger area of sand live oak, again closer to the no-mine area than the sites of some of the six isolated wet prairies. The remainder of the area will be reclaimed as improved pasture. Section 4 contains no-mine area in its southeast corner: Stream 2e and the Heart-Shaped Wetland. Almost all of the wetlands outside the no-mine area in Section 4 are in the top three scoring categories of functioning. Of the six wetlands complexes on OFG that are, in whole or in part, highest functioning, four of them are in Section 4. The two highest functioning wetlands outside Section 4 are in the no-mine area, and one of the highest functioning wetlands in Section 4 is in the Heart-Shaped Wetland. Three of the highest functioning wetlands are thus to be mined. Outside of Section 4, there are 14 wetlands or wetlands complexes outside the no-mine area that are in the second- and third-highest scoring categories. These are the mixed wetland hardwoods (H031) in Section 29; a small piece of mixed wetland hardwoods (G259A) straddling the no-mine boundary in Section 29; the bay swamp and mixed wetland hardwoods to the north in the headwater wetland of Stream 1w, which straddles Sections 29 and 20; the freshwater marsh partly fringed by wet prairie (G093) south of Stream 6w in Section 17; the freshwater marsh (G052) connected to Stream 9w and straddling Sections 17 and 8; the mixed wetland hardwoods flow-through wetland (G055) in Stream 9w and straddling Sections 8 and 9; the two bisected bay swamps (G166 and G163) and their mixed wetland hardwoods fringes in Section 9; and the six isolated wet prairies in the northeast corner of Section 9. In Section 4, there are only nine wetlands or wetlands complexes outside the no-mine area that are not in the second- or third-highest scoring categories, and all but two of them--a very small wet prairie fringe (G006) and half of a larger hydric woodland pasture (G105)--are at least moderate functioning. The wetlands in Section 4 fall into three categories: connected to the Stream 1e series, connected to Streams 3e and 3e?, and isolated. The long connected wetland of Stream 1e is mixed wetland hardwoods (G110). This wetland is high functioning, except for the headwater wetland of Stream 1ef, which is highest functioning. A narrow strip of wetland forested mixed (G132) runs along Stream 1ee. This wetland is moderate functioning. Proceeding from south to north, upstream the Stream 1e series, a freshwater marsh (G129) immediately upstream of Stream 1ee is high functioning, as is a smaller freshwater marsh (G125) immediately upstream of Stream 1ed. Two gum swamps (G123 and G121) in the flow-through wetland at the head of Stream 1ed are very high functioning, as is a freshwater marsh (G126) in the same wetland complex. Just downstream of Stream 1ef is a small freshwater marsh (G115) that is high functioning. Part of the mixed wetland hardwoods abutting this marsh to the east is very high functioning. Just upstream of Stream 1eb is the largest wetland complex of the Stream 1e series wetlands system. The largest communities forming this complex are hydric flatwoods (G107) and mixed wetland hardwoods (G110). The mixed wetland hardwoods envelope a small freshwater marsh (G108) and are fringed on the north by a strip of wetland forested mixed (G102). At the northernmost end of this complex is hydric woodland pasture. All of these communities are high functioning except the hydric woodland pasture, which is moderate functioning, and the hydric flatwoods and half of the marsh, which are very high functioning. Working back downstream, IMC will reclaim the mixed wetland hardwoods of the stream corridor, neglecting to replace the complexity provided by the three of the four flow-through marshes (G108, G125, and G129), the larger headwater marsh (G126), and the two gum swamps. IMC will also neglect to replace even the wetland function of the large hydric flatwoods (G107) and smaller hydric woodland pasture, as these sites are reclaimed as upland communities: pine flatwoods and temperate hardwoods, respectively. However, IMC will add complexity by adding a small marsh abutting the temperate hardwoods, two small bay swamps along the west side of the upper end of the Stream 1e series, a band of hydric flatwoods on both sides of part of the upper stream and a thicker area of hydric flatwoods east of Stream 1ed, a moderately sized area of hydric palmetto prairie within the thicker area of hydric flatwoods, and a thickened wetland corridor--mixed wetland hardwood--along Stream 1ee. The long connected wetland of Stream 3e (G137), which is wetland forested mixed, connects to a headwater or flow- through wetland, whose southern component (G136) is also wetland forested mixed. These wetlands are moderate functioning. The remainder of the wetland upstream of Stream 3e is marsh (G135), wet prairie (G134), and mixed wetland hardwoods (G133); they are all high functioning. The narrow wetland corridor of Stream 3e? is high functioning. The headwater wetland of Stream 3e? is a freshwater marsh (G016) fringed on the south by wet prairie (G015) and the north by mixed wetland hardwoods (G014). The mixed wetland hardwoods is moderate functioning; the marsh and wet prairie are high functioning. Working downstream along Streams 3e and 3e?, IMC will reclaim a large freshwater marsh/shrub marsh complex, fringed by wet prairie, at the site of the large headwater wetland of Stream 3e?. In place of the ditch, where IMC will restore Stream 3e?, IMC will probably reclaim mixed wetland hardwoods. (At present, Map I-2 shows improved pasture, but that was before IMC agreed to reclaim Stream 3e?.) IMC will reclaim the wetland complex between Stream 3e? and 3e with the same vegetative communities, except that it will eliminate some of the present system's complexity by replacing the wet prairie with freshwater marsh. Although Map I-2 inadvertently omits any reclaimed wetland community along Stream 3e, Figure 13A5-1 shows reclaimed wetland forested mixed. There are four isolated wetlands in the vicinity of Stream 1e series. At the northern boundary of OFG is a small wet prairie (G027), which is high functioning. Just west of Stream 1ec is a small hydric flatwoods (G118), which is moderate functioning. Just south of this hydric flatwoods is a larger wet prairie (G119) with a small area of hydric flatwoods (G119A), which are both high functioning. Just east of Stream 1ec is a small wet prairie (G028), which is high functioning, even though it is ditched. IMC will reclaim the high-functioning wet prairie (G027) with a freshwater marsh, the small, moderate-functioning hydric flatwoods (G118) with hydric flatwoods and possibly part of one of the bay swamps, the high-functioning wet prairie/hydric flatwoods (G119) with rangeland abutting a freshwater marsh, and the small, high functioning wet prairie (G028) also with the upland community of rangeland. There are four isolated wetlands south and east of Streams 3e and 3e?. The two largest are freshwater marshes (G024 and G021) fringed by wet prairies (G023 and G022, respectively). These are all highest functioning, except that G023 is high functioning. The two smaller wetlands are wet prairies (G025 and G026), which are both very high functioning. IMC will reclaim all four of these wetlands at their present sites with the same communities, except that IMC will replace one very high functioning wet prairie (G026) with improved pasture. North of the headwater wetland of Stream 3e? are five isolated wetlands. The largest is a large freshwater marsh (G004) at the northeast corner of OFG. A wet prairie (G005) fringes the southern edge of this wetland complex, which is ditched. The marsh is high functioning, but the wet prairie is moderate functioning. Two smaller ditched marshes (G008 and G010) lie southwest of this large complex; they are moderate functioning. A small mixed wetland hardwoods (G007) fringed by a narrow wet prairie (G006), which are north of the two marshes, are moderate and low functioning, respectively. The final isolated wetland is a freshwater marsh (G012) fringed by wet prairie (G011) and connected by ditch to the G014 wetland complex. The marsh is high functioning, and the wet prairie fringe is moderate functioning. IMC will reclaim improved pasture at the sites of four of these five wetlands. At the site of the large freshwater marsh (G004), IMC will reclaim a freshwater marsh, which will be fringed by wetland forested mixed. The wetland forested mixed will be fringed by hydric oak forest, which will be fringed by palmetto prairie. IMC will mine 10,566 linear feet of streams, reclaiming 10,919 linear feet. The current condition of these streams has already been adequately addressed, largely by Mr. Kiefer's assessment in the Stream Reclamation Plan, described above. All the tributaries are Class III waters, although, as Deputy Director Cantrell testified, they might not meet all Class III water standards. In fact, it is unlikely, given the level of agricultural alteration, for these tributaries, both within and without the no-mine area, to meet all Class III standards. As Deputy Director Cantrell testified, the unditched streams are the Stream 1e series, Stream 3e, and Stream 5e, although upstream of OFG, Stream 5e and its headwater wetlands have suffered extensive agricultural impacts. With the exception of the Stream 1e series and probably Stream 3e, elevated levels of turbidity and nutrients and reduced levels of dissolved oxygen are to be expected in the water of the tributaries on OFG due to the extensive ensuing erosion and low- flowing characteristics of these streams. Mining Ditch and Berm System Six months prior to the commencement of mining of each block, IMC will construct a ditch and berm system between the block and the adjoining no-mine area. The ditch and berm system captures the stormwater runoff that would otherwise leave the mine site and releases the groundwater that would otherwise remain at the mine site. The phosphate mining industry began using ditch and berm systems during mining in the late 1980s and early 1990s. IMC has designed the ditch and berm system to capture the water from the 25-year, 24-hour storm event with several feet of freeboard. For storms not in excess of the design storm, the ditch, which runs between the berm and the mine cut, will carry water around the perimeter of the mining block. During periods of high rainfall, IMC will pump the water in the ditch into the mine recirculation system to prevent unintended discharges. When the mine recirculation system reaches its capacity, it releases excess water into Horse Creek upstream of OFG at two outfalls that have already received National Pollutant Discharge Elimination System (NPDES) permits for use with the Ft. Green beneficiation plant. Maintained during all phases of mining operations, ditch and berm systems have effectively protected water quality during mining operations. The only indication in this record of a breach of a ditch and berm system has been one designed to meet older, more relaxed standards. The other function of the ditch and berm system is to dewater the mine site and restore the water table to nearby wetlands in the no-mine area. The removal of the water from the surficial aquifer at the mine cut effectively lowers the water table by, typically, 52 feet, which is the average depth of the excavation at OFG. Lowering the water table in the mine cut by any sizeable amount creates a powerful gradient, which draws more water from the unmined, adjacent surficial aquifer to fill the void of the removed water. Unchecked, this process would fill the mine cut with water so as to prevent mining operations and empty nearby wetlands of water so as to deprive them of their normal water levels and hydroperiods. To prevent these diversions of the unmined surficial aquifer from taking place, pumps send the groundwater entering the mine cut into the mine recirculation system and ditch. To maintain adequate groundwater flow from the ditch into unmined wetlands, the ditch must maintain adequate water levels. While constructing the ditch and berm system, IMC will construct monitoring wells between the ditch and the wetland or surface water, which will indicate when groundwater flows are less than the pre-mining flows, for which IMC will have already collected the data. Varying permeabilities of adjacent soils or inadequate maintenance of the ditch may cause the system to fail to maintain the proper hydration of nearby unmined wetlands. Due to failures of its ditch and berm system, IMC has several times dewatered nearby wetlands. Recent failures occurred at the East Fork Manatee River in November or December 1999, the North Fork of the Manatee River in March 2000, and two more recent failures at the Ft. Green Mine. To maintain the ditch and berm system, an inspector will daily drive a vehicle along the top of the berm to check the berm and the water level in ditch. However, recharge wells are also necessary to ensure that the ditch and berm system prevents the dehydration of unmined wetlands is recharge wells. Recharge wells would reduce the frequency and extent of wetland drawdowns. Strategically located throughout the length of the ditch, recharge wells would be drilled into the bottom of the ditch to the intermediate or Floridan aquifer. By this means, recharge wells actively maintain appropriate water levels in the ditches and prevent drawdowns. IMC has several alternative sources for the water for these recharge wells: the water pumped from the surficial aquifer during the dewatering of the mine, the groundwater that has returned to areas already backfilled with sand tailings, or the water from the mine recirculation system, provided it is filtered. Notwithstanding testimony to the contrary, neither the CRP approval nor the ERP requires IMC to install recharge wells. These documents fail to impose upon IMC any specific action, if the monitoring wells reveal reduced or eliminated groundwater flows into the wetlands and surface waters. Both documents acknowledge the possibility that IMC may need to install recharge wells to recharge the ditch. In his testimony, Dr. Garlanger recommended the installation of floats on the top of each recharge well to allow the inspector visually checking the ditch and berm readily to check each recharge well at the same time. Clearly, the presence of floats atop recharge wells would allow early identification and repair of malfunctioning recharge wells, prior to the loss of water from the ditch and the dehydration of nearby unmined wetlands. 2. Mine Recirculation System In addition to recycling the water used in mining operations, the mine recirculation system draws on sources deeper than the surficial aquifer, as well as rain. Water leaves the mine recirculation system through evapotranspiration and surface runoff. When water leaves the system as runoff, during or after major storm events, it does so through NPDES outfalls, and the high water volumes associated with the storm generally assure that any contaminants in the discharged water are sufficiently diluted. 3. Sand Tailings Budget For OFG, IMC has presented a reasonable sand tailings budget. Dr. Garlanger, whose expertise in geotechnical matters finds no match on the opposing side, has opined that the supply is ample. Charlotte County and the Authority have challenged the adequacy of the sand tailings budget. In part, Charlotte County and the Authority base their challenge to the sand tailings budget in part on an earlier comment by Dr. Garlanger concerning changing volumes of sand tailings, but he adequately explained that their reliance was misplaced. As noted above, the sand tailings budget at OFG requires sand from the Four Corners and Ft. Green mines. Conjuring up images of a sand Ponzi scheme, Charlotte County and the Authority seem to argue, in part, that there are not enough sand tailings, and DEP has allowed phosphate mining companies that have run out of nearby sand to substitute a Land-and-Lakes reclamation for the more sand-intensive reclamation that had originally been permitted and approved. OFG is early enough in the post Land-and-Lakes reclamation era that, if sand tailings from post-reclamation excavations are being moved around, OFG will get them. The obligation imposed upon IMC to obtain sand tailings backfill is not contingent upon feasibility; IMC must backfill the mine cuts with sand. The possibility that DEP would allow OFG to abandon one of the central tenets of this reclamation project by substituting Land-and-Lakes reclamation for topographic replication is inconceivable. Reclamation BMR Reclamation Guidelines BMR program administrator James (Bud) Cates supervises reclamation by the phosphate mining industry. Mr. Cates and Janine L. Callahan, also of BMR, prepared a document entitled, "Guidelines for the Reclamation, Management, and Disposition of Lands within the Southern Phosphate District of Florida" (Reclamation Guidelines). The document is dated August 2002. Although it is marked, "draft," Reclamation Guidelines is a revision of the first draft, which was prepared in 1993. The Administrative Law Judge commends the authors and DEP for the close attention to detail that has resisted finalization for nine years, but it would be imprudent to disregard the second draft while awaiting the next novennial revision, especially when DEP offered it as an exhibit (DEP Exhibit 37). Consistent with an emphasis on functional analysis and the creation of vegetative, hydrologic, and soils conditions that facilitate self-organization, Reclamation Guidelines defines "reclamation" as: the attempt to identify and replace those components/parameters of a community, resulting in the creation of a functional natural community analog. Emphasis is placed on the creation of functional soil, hydrology, and floral precursors that serve as the basis for food-web development. Because of the ecological need for fully functional communities, analogs are typically designed on a whole habitat basis rather than being designed around the specific needs of one or two species. These analogs are designed to incorporate a maximum initial diversity potential, based upon the premise that with proper management, the initial input will yield, over time, maximum ultimate diversity. Reclamation plans for and the activities used to create these replacement communities will be guided by existing knowledge of earthmoving, soils, hydrology, vegetation, general ecology, and wildlife management. Data in every applicable field should be constantly collected and used to increase knowledge and improve the results of the reclamation of natural community analogs. Focusing on specific reclamation techniques for soils, Reclamation Guidelines adds: The use of Topsoil/Vegetative Inoculum (T/VI) is extremely important to the introduction of organic matter, soil microbes, mycorrhizae, and plant propagules. These factors are critical to the creation of a living soil precursor. The T/VI is also the best known source of plant propagules that will provide the diversity inherent in a given community. Therefore, to the extent of material availability and economic feasibility, T/VI is recommended for use in the replacement of natural community analogs. The goal should be a three to six inch average depth with a minimum depth of no less than one inch over the base of sand, overburden, or sand/overburden mixture. Where T/VI availability problems occur, an artificially created topsoil precursor may be used in combination with all available T/VI or as a replacement for T/VI. Topsoil precursor may be created by incorporating a mixture of overburden, clay, and organics (hay mulch, wood chips, manure, green manure, or combinations thereof). All artificially created topsoil precursors should contain an organic portion and should be treated with microbial and mycorrhizal inoculum. For Sandhill, which has the least burdensome requirements among the three habitats most analogous to sand live oak (sand pine scrub, xeric oak scrub, and sandhill), Reclamation Guidelines notes that the objective is to concentrate a "deep layer of well-drained sands around/upon a topographic high to prove an area of rapid, positive infiltration and positive down-gradient seepage." The reclaimed sandhill habitat is adapted to excessively drained sands and requires "substantial depth to water table (although not as excessive or deep as scrub)." For soils, Reclamation Guidelines offers two options: six to eight feet of sand tailings covered with a layer of T/VI from a suitable donor scrub or eight to ten feet of sand tailings covered with a minimum four inch layer of artificially created topsoil precursor. For sand pine scrub and xeric oak scrub, the soil requirements are the same, except that the first option is for sand tailings eight to ten feet deep, not six to eight feet deep. As already noted, CRP Specific Condition 8.b requires IMC to reclaim sand live oak and xeric oak scrub with "several feet" of sand tailings and three to six inches of topsoiling from donor scrub or, if topsoiling is not feasible, the seeding and disking of a green manure crop. (Although omitted, the feasibility condition presumably qualifies the topsoiling requirement because Specific Condition 8.b defines "feasible.") For Pine Flatwoods and Dry Prairie, Reclamation Guidelines notes that the objective is to locate these communities on moderately to poorly drained soils, so that the depth to the water table is moderate to shallow. Most vegetation of these two communities is adapted to predominantly sand soils. For soils, Reclamation Guidelines offers two options: two to four feet of sand tailings covered with a layer of T/VI from a suitable donor flatwoods/dry prairie area or two to four feet of sand tailings covered with a minimum four inch layer of artificially created topsoil precursor. As already noted, CRP Specific Condition 8.a requires IMC to reclaim pine flatwoods and dry prairie with a minimum of 15 inches of sand tailings and three to six inches of transferred or stockpiled topsoil, if feasible, or, if not, the seeding and disking of a green manure crop. For Wetland Mixed Forest, Reclamation Guidelines notes that this community will occupy the outer limit of the floodplain down to the stream channel and the forested edge of deeper marshes. Likely to receive runoff from major storm events, Wetland Mixed Forest should be designed to contain and slow runoff while maintaining sufficient water for wetland viability. For soils, Reclamation Guidelines offers three options: decompacted overburden to a depth below the dry season water table overlying by a layer of T/VI from an appropriate donor site, two to three feet of sand tailings under a layer of T/VI, or either overburden or two to three feet of sand tailings covered by a minimum of four inches of artificially created topsoil precursor. As already noted, ERP Specific Condition 14.b requires IMC to reclaim all forested wetlands by backfilling with sand tailings or overburden to an unspecified depth under "several inches of wetland topsoil," if feasible. However, for bay swamps, Specific Condition 14.b adds in boldface: "All reclaimed bay swamps shall receive several inches of muck directly transferred from forested wetland approved for mining." Reclamation Guidelines treats Bay Swamp (and Cypress Swamp) separately from other forested wetlands. Noting that Bay Swamps are in areas of significant surficial seepage or high average groundwater elevation, Reclamation Guidelines states that Bay Swamps require sufficient seepage to remain saturated or a deep organic profile at and below the average water table elevation. For soils, Reclamation Guidelines states: "Bay swamps require the placement of one to three feet of organic muck as a depressed lens. The muck should be obtained from a suitable donor wetland." For Non-Forested Wetland, which includes wet prairies and freshwater marshes, Reclamation Guidelines is of value more to identify why the phosphate mining industry and DEP have overseen the routine reclamation of deeper wetlands, but not shallower wetlands. Treating these two very different communities under the same category, Reclamation Guidelines states: "All of the sub-categories may be constructed on overburden, with the exception of sand pond." Although the overburden option for reclaimed forested wetlands seems a stretch, given repeated problems of mature tree growth into overburden relatively close to grade, the overburden option for reclaimed wet prairie, other than fringing deeper marshes when properly sloped, can no longer merit serious consideration, given only one successful, extensive shallow-wetland reclamation site--SP(2D), whose reclaimed soil is four inches of mulched topsoil overlying four feet of sand tailings. However, consistent with its Reclamation Guidelines, DEP did not differentiate between wet prairies and deep marshes in the soil-reclamation requirements contained in the ERP. ERP Specific Condition 14.c allows backfilling with sand tailings or overburden and requires only "several inches of wetlands topsoils when available." Tellingly, Reclamation Guidelines divides aquatic systems into two categories: shallow (less than six feet deep) and deep. Shallow systems comprise swamps, marshes, sloughs, and ponds, but not streams. Nowhere does Reclamation Guidelines explicitly address the reclamation of streams. Comparing the soil-reclamation requirements that DEP has imposed on IMC in the CRP approval and ERP to the soil- reclamation specifications stated in BMR's Reclamation Guidelines, material discrepancies emerge as to the depth of sand tailings underlying four upland communities. If IMC transfers topsoil, sand live oak communities require at least six feet of sand tailings, not "several" feet; if IMC uses green manure, sand live oak communities require at least eight feet of sand tailings. Regardless whether topsoiled or green manured, xeric oak scrub communities require at least eight feet of sand tailings, not "several" feet. Regardless whether topsoiled or green manured, pine flatwoods and palmetto prairie require at least two feet of sand tailings, not 15 inches. There is a material discrepancy between the ERP and Reclamation Guidelines as to bay swamps. Reclamation Guidelines specifies one to three feet of organic muck for reclaimed Bay Swamps. ERP Specific Condition 14.b requires only "several inches of muck." Given the poor record reclaiming bay swamps, DEP, in forming this condition, is not relying on any experience-based knowledge that it has acquired, or, if it is, it did not add this information to the present record. There is no discrepancy as to wet prairies, but this is clearly due to a shortcoming in Reclamation Guidelines, at least as to non-fringe wet prairies. Under Reclamation Guidelines, wet prairies, at best, will continue to reclaim only as fringes, and only then if the edges of deeper wetlands have shallow slopes. Given the otherwise-uniform failure to reclaim extensive shallow wetlands, the actual soil regime at SP(2D) of four feet of sand tailings under four inches of topsoil must set the minimum soil criteria for wet prairie. 2. Geology and Soils For purposes of this Recommended Order, soils occur predominantly in the first two meters of the earth's surface. Below that depth, geologic characteristics predominate, so this Recommended Order refers to these deeper structures as geology. Post-reclamation, all of the soil and the top 45-50 feet of the geology are a product of IMC's reclamation activities. The post-reclamation geologic characteristics follow from the mining process, which deposits overburden within the mine cut in two locations. Most of the overburden is deposited in spoil piles within the cut. Some of the overburden is piled against the sides of the mine cut to reduce the seepage of water from the surrounding surficial aquifer into the cut. Both types of overburden are sometimes called "cast overburden." At OFG, prior to backfilling, the creation of cast overburden spoil piles will either leave alternating bands of sand tailings valleys and cast overburden spoil piles, each 330 feet wide, or each 165 feet wide; the record is not entirely clear on this point. The scenario with the greater hydrological impact is that each valley and the base of each spoil pile is 330 feet wide, but, even under this scenario, relatively little backfilled area would have less than five feet of sand tailings. If each sand tailings valley is 330 feet and each cast overburden spoil pile is also 330 feet at its base, the profile of each cast overburden spoil pile would appear to be a two- dimensional pyramid with its top cut off just below midpoint along its two slopes. The sides of the spoil piles of cast overburden are not perpendicular to the surface, but are sloped at about 1.5:1, according to Dr. Garlanger. Rounding off the depth of the mine cut to 50 feet, this 33-degree slope would travel 50 feet vertically at the point at which it had traveled 75 feet horizontally. Matching this slope with another on the other side of the spoil pile, 150 feet of the 330-foot wide overburden spoil pile would be consumed by the sloped sides, and 180 feet would be a plateau, at a constant elevation of 50 feet above the bottom of the mine pit. Adding 7.5 feet on either side of the plateau gains a depth of 5 feet, so the width of overburden under less than five feet of sand tailings would be 195 feet. Under the less-favorable scenario, for a 660-foot wide band of reclaimed geology, without regard to topsoil additions, the sand tailings, for the above-described 660-foot slice, will be at least 10 feet deep for a distance of 450 feet, or 68 percent of the reclaimed area, and will be at least 5 feet deep for a distance of 475 feet, or 72 percent of the reclaimed area. Adding the U-turns at the end of the rows would add only a little more area to the 28 percent of the reclaimed area with an overburden plateau within five feet of the surface. If the cast overburden spoil piles fill only half of each 330-foot wide cut, then the overburden plateaus would be much narrower. Each sand valley of 165 feet would abut a 33-degree slope that would again run 75 feet horizontal while climbing 50 feet vertical. Two of these slopes would consume 150 feet horizontal, leaving an overburden plateau of only 15 feet, leaving much less land with an overburden plateau within five feet of the surface. The shaping of the overburden that precedes the backfilling, the backfilling of sand tailings, and the transfer of topsoil are aided by substantial technological improvements in earthmoving equipment in recent years. Most importantly, earthmoving equipment has incorporated global positioning systems, so that they can now grade material to a tolerance of two centimeters, as compared to tolerances of six inches and one foot not long ago. This achievement permits the reclamation scientists to supervise backfilling more closely so as to replicate the design topography, which is a necessary, although not sufficient, condition of successful establishment of targeted hydroperiods and inundation levels. IMC soil scientist Joseph Schuster and Mr. Carter both presented detailed, well-documented testimony and are both competent soil scientists. They start from the same point, which is that pedogenesis, or soil formation, is a function of five factors: parent material, relief, climate, vegetation, and time. From there, they travel separate paths in their analysis and conclusions concerning the soil aspects of IMC's reclamation plan. In the successful reclamation of soils, Mr. Schuster highlights the creation of appropriate drainage characteristics, and Mr. Carter highlights the creation of appropriate soil horizons, although both experts acknowledge the importance of both these factors, and others, in soil formation and function. Their reasoning seemed mostly to be a question of differing emphases, although their conclusions were mutually exclusive. As already noted, the A horizon is the topsoil layer. (A mucky wetland may have an O horizon.) There is some variability among horizons--for example, the C horizon, which is described below, may occur immediately beneath the A horizon, especially in sandy material. But, for this part of Florida, typically, the E horizon forms under the A horizon. The E horizon is a leaching zone, through which rainwater transmits substances from the A horizon down to the B horizon, which is the accumulation zone beneath the E horizon. Florida typically has two types of B horizons: the Bh (or spodic) horizon, which is composed of loamy or spodic materials, and the Bt (or argyllic) horizon, which is composed of clayey materials. The spodic horizon is a mineral soil horizon containing aluminum and organic carbon, and possibly iron, which formed in a much colder climate, probably at least 10,000 years ago. Spodic horizons typically occur in the top two feet of the soil profile. Although spodic horizons may occur as deep as 40 feet, they occur at OFG within 20 inches of the surface, sometimes within only 10 inches. Beneath the B horizons is the C horizon, which is the parent material for pedogenesis. For the most part, Mr. Schuster's emphasis on reclaiming appropriate drainage is credited as the single most important factor in reclamation, and his seven drainage categories are ample for guiding the reclamation of the drainage characteristics of soils. More reclamation failures may necessitate the implementation of one of Mr. Carter's suggestions to carefully restore the soil horizons within the top two meters of the mine cut, as it is backfilled, or to use more clayey soils, such as those from drained CSAs, to add more nutrient-retaining capacity to the B and C horizons than nutrient-poor sand tailings provide. Mr. Carter's soil cores from reclamation sites, which reveal overburden close to the surface, presented stark contrasts to soil cores of native soils in the area, although drainage concerns outweigh pedogenic concerns. Mr. Carter correctly points out that, from a soils perspective, pre-mining overburden is not post-reclamation overburden. From a mining perspective, what lies above the unmined phosphate ore is overburden, and what lies in the ground, post-reclamation, is also overburden, which, to a certain depth, is dominated by characteristics of the B horizon and underlying C horizon. However, in a 52-foot deep phosphate mine, as opposed to typical road construction, which Mr. Schuster unpersuasively offered as a comparable, the overburden is ultimately dominated by geologic material from below the C horizon. From a soils perspective, what lies in the unmined ground are soil horizons that took many years to form, and what lies in the ground, post- reclamation, is nothing but an admixture of former soil horizons and geologic material that normally resides a little deeper in the earth's crust. As Mr. Carter notes, the result, post- reclamation, is less like soil and more like unconsolidated soil material with little horizonization even several years after reclamation, and, if an overburden layer is present close to the surface, it typically is tightly compacted. Soil horizons are not an incidental or random characteristic of undisturbed soils; soil horizons are an important component in the formation and functioning of soil. Mr. Schuster himself disclaims reliance upon overburden epipedons--which are organically influenced horizons typically above the B horizon--in the restoration of native ecosystems, although he does not object to the presence of such epipedons in agricultural restoration. If sand were displaced by overburden in the area of the E horizon, the E horizon will be unable to contribute to the formation of the B horizon, as it must, especially after the comprehensive disturbance of all soil horizons contemplated at OFG. Mr. Schuster's disclaimer bodes ill for the ERP provisions allowing overburden as an alternative to sand tailings for forested and herbaceous wetlands. However, Mr. Schuster's disdain for cast overburden near the surface is well-founded. His emphasis on drainage over soil horizons, including even overburden epipedons, may find support at Dogleg, which, according to the CDA, suffered the loss of its 12-inch topsoil layer due to oxidization and was left with overburden of a "clayey sand" texture that may have been more permeable than typical, less permeable overburden. This loss appears to have taken place over sufficient time that other conditions may have commenced to form an A horizon. However, when adjacent mining ended and the water table re-established itself, the reclaimed trees began to survive. Mr. Schuster accounts for the importance of pedogenesis, in addition to drainage characteristics, by identifying the topsoil/green manure, sand, and overburden as analogs of soil horizons. Certainly, the topsoil/green manure is a functional analog, and its thickness is not much of a variable. Sand tailings provide an appropriate texture for an A horizon. But the variability of the depths of sand tailings limits the force of Mr. Schuster's argument for functional analogs. For all wetland communities, overburden may occur at depths of only several inches, and, for pine flatwoods and palmetto prairies, overburden may occur at depths of 15 inches. Or sand tailings may be over 50 feet deep, atop a clay confining layer, not overburden. Setting aside the problem with the variability of depths of sand tailings, it is possible to treat sand tailings as a functional E horizon, through which materials will leach from the A horizon and into the B horizon, which is the zone of accumulation. However difficult it may be to cast the sand tailings in the role of a B horizon, it is impossible to cast them in the role of a C horizon. Ignoring the considerable amount of geologic material contained in cast overburden and possible textural issues, Mr. Schuster plausibly offers overburden as good B and C horizon material because of its higher clay or spodic content. Thus, the apparent impairment of pedogenesis may not be as extensive as first appears, provided overburden remains below the A and E horizons. Still, mining and reclamation, at least as designed for OFG, mean the loss of some soil functions for extensive periods of time, but proper reclamation of drainage characteristics and hydrology sufficiently mitigate these losses of function. Even Mr. Schuster's emphasis on drainage is not unconditional, as he relies on the application of topsoil or the implementation of a green-manure process to provide an immediate A horizon and accelerate the process by which the A horizon continues to form. Endorsed by Mr. Carter as a good idea to increase organic material and loosen the structure of the topsoil, green manure is the process by which a quick-growing cover crop is planted on the finished surface, post-reclamation. The crop is then disked into the soil to provide a quick infusion of nitrogen and organic matter. This approach has not previously been used in reclamation following phosphate mining, but it has been used in other applications and is effective. Post-reclamation, fire too will pump nutrients into the A horizon. Herbaceous wetlands, with their shallower roots, ought to be adequately served by Mr. Schuster's focus on the drainage characteristics of reclaimed soils. Forested wetlands present a different challenge due to their deeper root systems. Past reclamation of forested wetlands has experienced tree loss after several years of growth, possibly indicative of a problem with root development beyond a certain depth. Perhaps the roots cannot penetrate the overburden or cannot find the necessary nourishment, after penetrating the overburden; however, it is at least as likely, given the record of reclamation, that the mitigation site suffered from a poorly reclaimed water table, so that, for example, the water table was too high for too long, perched, or even too low for too long. Given the repeated problems with establishing appropriate water tables, post-reclamation, this factor looms as a likely explanation for tree die-off. However, Mr. Schuster's emphasis on drainage characteristics over pedogenic conditions carries more weight as to herbaceous wetlands and xeric habitats, where sandy soils predominate to relative great depths, and somewhat less weight as to forested wetlands. Mr. Schuster's emphasis on drainage over pedogenesis carries even less weight as to pine flatwoods and palmetto prairies, which are less tolerant to the disturbance of the spodic horizon in reclaimed soils. Obviously, overburden presents different textures and drainage characteristics than do native flatwoods soils. However, pine flatwoods and palmetto prairies are more dependent upon higher water tables than more xeric upland communities, so, again, past problems in reclaiming these upland communities again likely involve the failure to create an appropriate water table, post-reclamation. Differences between Mr. Schuster and Mr. Carter were harder to reconcile regarding the role of pH in soil. Mr. Schuster and Mr. Carter reached different results in field tests of soil pH. However, Mr. Schuster's testimony is credited that most ecosystems tolerate a wide range of pH, and the most important soil characteristic remains its drainage characteristics. Hydrology Introduction Removing and replacing the topography, soils, and geology, including the surficial aquifer, to a depth of 52 feet, under nearly 3500 acres of land necessitates hydrological analysis. Hydrological analysis is necessary to support three sets of projections: the streamflows of Horse Creek, downstream of OFG, during mining and after reclamation; hydroperiods and inundation depths of reclaimed wetlands, as the wetlands created in the reclaimed topography and soils fill and empty with water based on inputs and outputs from runoff and groundwater, inputs from rainfall, and outputs from evapotranspiration; and peak discharges from OFG, during mining and after reclamation. All hydrological analysis must account for the water budget, which balances the inputs and outputs of water. The elements of the water budget are rainfall, runoff, percolation (or infiltration), evapotranspiration, deep recharge (the recharge of the deeper aquifers), and groundwater outflow. Rainfall is the most important factor because it is the sole means by which water enters the system. Equal to the total of the outputs, annual rainfall is a large number, typically measuring in this part of Florida in excess of 50 inches. Rainfall is also a variable number in two respects. It varies from year to year. For the Peace River basin, annual rainfall from 1933 to 2002 has ranged from 35.89 inches to 74.5 inches with an average of 52.4 inches. However, rainfall in the Peace River basin has varied over eras. From 1933 to 1962, average annual rainfall was 55.48 inches. From 1962 to 2002, average annual rainfall was 51.02 inches. For the Peace River basin, the average annual rainfall has decreased about 4 1/2 inches in the past four decades when compared to the preceding three decades. Especially over shorter time intervals, rainfall also varies considerably from location to location within a relatively small area. Subject to these variabilities, especially the distance of the rainfall gauge to the location for which the water budget is constructed, rainfall is easily measured by rainfall gauges. Measurement means straightforward collection of data without elaborate modeling, calculation, or simulation. After rainfall, the most important element in the water budget is evapotranspiration, which is the combined effect of evaporation of water from soil, plant surfaces, wetlands, and open water and transpiration of water through vegetative processes. In this part of Florida, evapotranspiration releases about 75 percent of the rainfall back into the atmosphere, which, by convention, counts as a loss to the system. Unlike rainfall, evapotranspiration typically cannot be measured, except that the maximum evaporation, which is a pan containing water in the direct sun, is subject to direct measurement. Hydrologists have measured evapotranspiration from irrigated golf courses at 58-62 inches annually, and Dr. Garlanger has measured evapotranspiration from reclaimed CSAs at 39-41 inches annually, although both of these measurements may have been somewhat indirect. However, hydrologists widely recognize ranges of evapotranspiration for this part of Florida for different land uses. Annual rates of evapotranspiration for open water is 49-1 inches, for riparian wetlands is 47-49 inches, and for isolated wetlands is 43-44 inches. The annual evapotranspiration for pine flatwoods is 37-39 inches and for xeric uplands is 34-36 inches. Impervious surface, such as pavement or a roof, produces only 8-10 inches annually--absent weeds, all evaporation. In addition to land use, the amount of water available controls the amount of evapotranspiration. Elevations of the water table will affect evapotranspiration. Thus, hydrologists often measure potential and actual evapotranspiration. Anthropogenic impacts may increase or decrease evapotranspiration. Net additions of impervious surface, such as parking lots, roads, and rooftops, increase runoff and decrease evapotranspiration. Net additions of open water, such as lakes, ponds, and streams, decrease runoff and increase evapotranspiration. At the other end of the spectrum, deep recharge removes very little water at OFG. Even during mining, when the impacts would be greatest due to high withdrawals, the increase to deep recharge is 30-60 gallons per minute--insignificant as compared to the average recharge rate in the Peace River basin of 190,000 gallons per minute. In fact, according to RAI-192 in the CDA, rainfall, not deepwell water, is the primary source of water for the mine recirculation system. Deep recharge is typically one inch annually, although Charlotte County hydrologist Phillip Davis, in one of his scenarios, claimed that 2.5 inches of water annually would enter the intermediate aquifer from the surficial aquifer. This range of values for deep recharge is within the specified ranges for most types of evapotranspiration. Deep recharge cannot be directly measured. The record does not suggest much variability in deep recharge, which is controlled by the elevation of the water table and potentiometric surface of the Florida Aquifer, in undisturbed geologic systems in this part of Florida. Although the replacement of part of the confining layer between the surficial and intermediate aquifers could affect deep recharge, the potential impact at OFG appears to be very small due to the permeability of the matrix layer and impermeability of the clay bed beneath it. However, historic anthropogenic disturbances may have increased deep recharge. All groundwater withdrawals induce recharge, at least of the surficial aquifer. Withdrawals from the deeper aquifers, such as those taken by the phosphate mining industry prior to expanded recycling, could have caused increased rates of deep recharge, depending on the confining layers above the Floridan Aquifer within the area influenced by the withdrawals. To the extent that the effect of these deep withdrawals extended to the surficial aquifer, evapotranspiration and streamflow would have been reduced. Groundwater outflow has been measured in this area by Bill Lewelling of the U.S. Geologic Service. (Mr. Lewelling seems to have measured groundwater outflow indirectly by measuring chloride concentrations at different locations.) He found a range of 1.7-17.9 inches annually with an average of 9.2 inches annually. An important component of groundwater outflow, infiltration depends on soil type and antecedent saturation, so it is variable in terms of location and climate. However, it appears to vary within a relatively narrow range at OFG, pre- mining. One combination of water-budget elements that may be measured easily is streamflow, which, as noted above, is a combination of the runoff and groundwater outflow reaching the stream. Streamflow equals rainfall minus evapotranspiration minus deep recharge minus the change in uplands storage. For the purposes of Dr. Garlanger's analysis, uplands are everything, including wetlands, above riparian wetlands, and riparian wetlands are the area adjacent to a stream channel that remain perennially wet and are typically within the 25-year floodplain. Streamflow is not variable like rainfall as to location because the river or stream is fixed and so is the location of the gauge, but streamflow is highly variable as to volume, even from year to year. For Horse Creek at State Road 64, for example, annual streamflow from 1977 to 2001 has averaged 9.7 inches, but has ranged from one inch to 17 inches. For the Peace River at Arcadia, annual streamflow from 1950-1962 was 13.25 inches or 1334 cfs. From 1963 to 2002, average streamflow at the same location was 8.78 inches or 884 cfs. The SWFWMD has not yet set minimum flows and levels for the Peace River, but is presently in the process of setting these values. In these cases, streamflow is most often calculated to compare a model's output in streamflow to measured values for the same period of time, to determine streamflow for locations without a streamflow gauge, or to determine streamflow for locations with a streamflow gauge, but after changes in land use, such as the construction of a ditch and berm system or post-mining reclamation. Another combination of water-budget elements that can be measured, although with more difficulty than streamflow, is the water table. Most water table data are fairly recent, dating from the early 1990s. Mr. Davis testified that the water table data available for OFG were the most limited that he had ever encountered. Varying daily, the water table is the top of the surficial aquifer. The elevation of a non-perched water table, at any given time, is ultimately driven by all of the elements of the water budget, but is immediately reflective of surficial aquifer inputs and outputs and hydraulic conductivity. Hydraulic conductivity is the ability of a porous medium to transmit a specific fluid under a unit hydraulic gradient, so it is highly dependent on the physical properties of the medium through which the fluid is transmitted. Although hydraulic conductivity exists in the horizontal and vertical planes, this Recommended Order considers only horizontal hydraulic conductivity. Hydraulic conductivity is an important hydrological factor that can be measured, at least horizontally, although with difficulty. Hydraulic conductivity varies by location due to the variations in permeability of the geological structure through which the groundwater is passing. The hydraulic conductivity of sand tailings is about 38 feet per day, and the hydraulic conductivity of cast overburden is about one foot per day. Native soils are typically somewhere in between these two extremes. In one area, the matrix, pre-mining, had a permeability of 5-15 feet per day. IMC's assurances concerning streamflow, wetlands hydroperiod and inundation depths, and peak discharges must be assessed against three different backdrops. At one extreme, at least based on the present record, phosphate mining and reclamation, as distinguished from other phases of phosphate processing, have not caused adverse flooding; the sole example of flooding from a failed ditch and berm system--designed to meet more relaxed standards--occurred at the Kingsford Mine on January 1, 2003, and no serious environmental damage occurred. At the other extreme, reclamation after phosphate mining has routinely failed to reclaim targeted hydroperiods and inundation depths for shallower wetlands and many forested wetlands. In between these two extremes, although closer, at least recently, to the industry's flooding experience, is streamflow. Historic impacts to the Peace River are considered below, but an example of the minimal impact on streamflow of recent mining is found in the last 15 years' mining of the upper reaches Horse Creek. During this period, the streamflow of Horse Creek at State Road 64 has remained unchanged. The record does not support Mr. Davis's suggestion that high volumes of groundwater pumping and high volumes of NPDES discharges artificially added streamflow during this period. Resolution of the hydrological evidence in these cases requires close examination of the testimony of Dr. Garlanger, who addressed all three areas for IMC; Mr. Davis, who addressed streamflow and wetland hydroperiods and inundation depths for Charlotte County; and Mr. Loper, who addressed peak discharges for Charlotte County. All three of these witnesses are highly competent and patiently and thoroughly explained their hydrological analyses. Mr. Loper proved adept at finding flaws in IMC's analyses of peak discharges. Dr. Garlanger and his staff several times refined their work, even during the hearing, to incorporate Mr. Loper's findings. Differences remained between Mr. Loper and Dr. Garlanger, and, although it is possible that Mr. Loper is correct on these remaining points, Dr. Garlanger successfully discounted the importance of Mr. Loper's objections in projecting peak discharges. Examining the evidence in the backdrop of a record almost devoid of failures that have resulted in flooding, it proved impossible not to credit Dr. Garlanger's assurances about peak discharges. Mr. Davis was less successful in finding flaws in IMC's analysis of streamflow, or at least in finding material flaws. As detailed below, his theory attributing to phosphate mining a greater share of historic reductions in the streamflow of the Peace River seems less likely than Dr. Garlanger's theory attributing a lesser share of these historic reductions to phosphate mining. Mr. Davis substituted an integrated simulation model for Dr. Garlanger's uplands model and spreadsheet. The advantages of Mr. Davis's model emerged to a greater extent in simulating wetlands hydroperiods and inundation depths, not in simulating streamflows. This is discussed in detail below. The conflict between Mr. Davis and Dr. Garlanger over the ability to reclaim targeted hydroperiods and inundation depths has proved very difficult to resolve. Dr. Garlanger has vast experience in the phosphate mining industry and thus a clear advantage in projecting, as he has since 1974 at several hundreds of projects, peak discharges and streamflow. But this experience is no advantage as to projecting wetland hydroperiods and inundation depths. Dr. Garlanger did not state that he has projected hydroperiods and inundation depths for 30 years at several hundreds of projects. If he has done so, he has contributed to the numerous failures, described above, of reclaiming shallow wetlands. More likely, the phosphate mining industry has infrequently targeted shallow wetlands for reclamation, so Dr. Garlanger does not have extensive experience in creating the necessary hydroperiods and inundation depths for shallow wetlands. The reclamation of specific hydroperiods and inundation depths for shallow wetlands is likely a fairly recent development, perhaps due to the relaxed restoration expectations of earlier eras or the inability of earthmoving equipment to execute fine specifications in finished topography. In the CDA discussion of Bay Swamp, noted above, the author admits that reclamation historically has not attempted to reclaim the kind of interface necessary between shallow wetlands and the water table to support bay swamps. The parties' understandable, but unrealistic, pursuit of findings that all previous shallow-wetland reclamations of any size have failed or succeeded may have discouraged testimony candidly analyzing what hydrologists have learned from the limited successes and the many failures. Especially unfortunate is the omission of any discussion of the success of Dogleg, where, according to the CDA material, persistent replanting of trees over many years in soils with prominent, but perhaps atypically permeable, cast overburden profiles eventually succeeded, after the completion of nearby mining allowed the water table to reestablish itself. The record does not even indicate if Dogleg mining took place behind a ditch and berm system, nor does it adequately describe the texture of the overburden on which the topsoil rested. In addition to different levels of confidence attaching to the demonstrated ability of the phosphate mining industry to avoid adverse flooding and significant reductions in streamflow, on the one hand, and the routine inability of the phosphate mining industry to re-create the hydroperiods and inundation depths required for shallow wetlands, another point of differentiation exists between Dr. Garlanger's streamflow projections and his hydroperiod and inundation depth projections. Although he uses the same uplands model and similar wetlands models for both tasks, certain characteristics of his relatively simple modeling do not work as well in projecting hydroperiods and inundation depths as they do in projecting streamflows. Accurate projections of streamflow, at a discrete point downstream of the 4197 acres constituting OFG, are amenable to averaging, smoothing out input values, and substituting assumed values for calculated values. Accurate projections of hydroperiods and inundation depths require precise analysis of reclaimed wetlands--few over 10 acres, most less than a couple of acres--distributed over the 3477 acres of OFG to be mined. For each wetland, precision means daily accuracy to within a few inches of elevation of topography and water table and no more than a few feet of hydraulic conductivity. Streamflow projections, which have worked in the recent past, will continue to work, whether each projection within an area is accurate or any errors within an analyzed area offset errors in other areas, so that, notwithstanding flow discharge curves, small discrepancies in projected streamflow average out over longer periods of time. Hydroperiod and inundation depth projections, which may have been attempted, if at all, only rarely in the past, must be accurate over very small areas for very specific time intervals. Also, streamflow projections are less sensitive to misallocations between runoff and groundwater flow than are projections of shallow wetland hydroperiod and inundation depth. The record suggests that reclaiming short wetland hydroperiods and shallow inundation depths places new and more difficult demands upon the phosphate mining industry and its reclamation scientists. Although long accustomed to producing projects that did not flood and at least recently accustomed to producing projects that did not reduce streamflow, the phosphate mining industry and its reclamation scientists are only now acclimating to newer regulatory expectations that they produce projects that reliably reclaim shallow wetlands by re-creating functional relationships between these wetland systems and surface runoff and groundwater flow. Streamflow Streamflow in Horse Creek downstream of OFG and the Peace River is reduced during mining because the ditch and berm system captures all of the runoff, at least up to the capacity of the ditch and berm system. The ditch and berm system is designed to handle the 25-year, 24-hour storm event, although additional, unspecified freeboard is built into the system. The capacity of the ditch and berm system may be exceeded by more intense storms or perhaps even lesser storms, unless the 25-year storm design accounts for antecedent water levels, which may be higher in systems with recharge wells than in systems without the recharge wells. In any event in which the capacity of the ditch and berm system is exceeded, IMC pumps the water through the mine recirculation system and releases it through one of two NPDES outfalls upstream at Horse Creek. Because the ditch and berm system captures all of the runoff, under normal conditions, the reduction in streamflow after reclamation is generally less than the reduction in streamflow during mining. The removal of the ditch and berm system allows runoff again to contribute to streamflow. To analyze the impacts upon streamflow, Dr. Garlanger first performed a simplified water budget analysis at three locations: Horse Creek at State Road 72 (near Arcadia), the Peace River at Ft. Ogden (where the Authority withdraws its raw water--downstream of the confluence of Horse Creek and the Peace River), and the point at which the Peace River empties into Charlotte Harbor. Although Dr. Garlanger used uplands exclusively for this simplified exercise in constructing a conceptual water budget, adding the riparian wetlands would not substantially change the result because the wetlands runoff and evapotranspiration would be higher, but the wetlands groundwater outflow would be lower. Either way, Dr. Garlanger's analysis, which is sometimes called an analytic model, was merely a prelude to more sophisticated modeling. For his during-mining analysis, Dr. Garlanger assumed that the ditch and berm system would capture all the runoff from the 5.4 square miles of the Horse Creek sub-basin behind the ditch and berm system. In sequential mining, the ditch and berm system would not capture all of the 5.4 square miles at once. But, assuming the worst-case scenario, Dr. Garlanger assumed the capture of the runoff from entire sub-basin for a period of 25 years. Initially, Dr. Garlanger also assumed that the ditch and berm system would likewise not release any base flow. This is an unrealistic scenario because, as noted above, one of the two purposes of the ditch and berm system is to permit base flow into wetlands and streams. Later, Dr. Garlanger alternatively assumed that the ditch and berm system would release all of the base flow. If the ditch and berm system is equipped with recharge wells, it is reasonable to expect that the system will release all of the base flow. Calculating that the Horse Creek sub-basin upstream of State Road 64 is 39.5 square miles, Dr. Garlanger divided the average streamflow of 29.1 cfs at State Road 64 by the area of the sub-basin and determined that each square mile contributed 0.74 cfs of streamflow. Multiplying this number by the 5.4 miles captured by the ditch and berm system, Dr. Garlanger determined that, during mining, the ditch and berm system would reduce streamflow by 4 cfs, if it removed all base flow (and runoff). This very worst-case scenario would generate the following reductions in streamflow: in Horse Creek at State Road 72, 2.3 percent; in the Peace River at Ft. Ogden, 0.3 percent; and in the Peace River at Charlotte Harbor, 0.2 percent. Dr. Garlanger then calculated the reduction in streamflow in the probable scenario in which the ditch and berm system, with recharge wells, operates properly and releases the base flow, while still retaining all the runoff. Relying principally upon Mr. Lewelling's report on groundwater outflow in various locations within the Horse Creek sub-basin, Dr. Garlanger calculated that the capture rate would decrease from 0.74 cfs per square mile to 0.28 cfs per square mile. Applying a capture rate of 0.28 cfs per square mile times 5.4 miles, the reduction in streamflow, during mining, is more realistically 1.5 cfs. This means that, under the simplified analytic model, the ditch and berm system would reduce streamflow in Horse Creek at State Road 72 by less than one percent, in the Peace River at Ft. Ogden by .13 percent, and in the Peace River at Charlotte Harbor by .09 percent. These figures would represent the same reduction in streamflow caused by a decrease in average annual rainfall of 0.01 inches. Although, as discussed below, Dr. Garlanger also undertook more sophisticated modeling of streamflow during mining, this is a good point at which to address three of Mr. Davis's objections to Dr. Garlanger's during-mining analysis because these objections are more conceptual in nature and are not directed to Dr. Garlanger's model. Mr. Davis contended that the unmined wetlands would become dehydrated because: 1) the ditch and berm system would deprive them of surface flow or runoff from the areas behind the ditch and berm system; 2) the ditch and berm system would deprive them of adequate base flow or groundwater; and 3) water in the ditch would be lost to evapotranspiration. These objections are more applicable to a ditch and berm system without recharge wells. If the only source of water to rehydrate the wetlands is the groundwater running into the mine and rainfall directly on the area behind the berm, the loss of runoff into the area behind the berm and the loss of water to increased evaporation would require additional analysis to assure that adequate water remained to recharge the downstream wetlands through groundwater inputs. However, the recharge wells add additional water, probably from the deeper aquifers, so that adequate water can be supplied the downstream wetlands through groundwater inputs. To the extent that intercepted surface flow reduces water levels in the unmined wetlands, IMC can offset this loss by pumping more water into the ditch and increasing groundwater inputs into these wetlands. Mr. Davis's additional objection about additional evapotranspiration from the riparian wetlands assumes the condition that he claims will not occur--adequate hydration of the riparian wetlands--so it is impossible to credit this concern. Dr. Garlanger next analyzed streamflow by applying a simulation model. More sophisticated than the analytic model discussed in the preceding paragraphs, the uplands portion of this modeling also aided Dr. Garlanger's analysis of the hydroperiods and inundation depths of the wetlands in the no- mine area and the reclaimed wetlands, which are discussed in the next subsection. Dr. Garlanger's simulation model calculates site-specific groundwater outflows based on day-to-day hydrological conditions. Unlike the analytic model, which examined the effect on streamflow only during mining, the simulation model determines streamflow contributions from OFG without any mining disturbance for a 25-year period into the future, during mining, and after reclamation for the same 25- year period used in the no-mining analysis. The modeling proceeded in two stages. First, Dr. Garlanger modeled uplands. Then, inserting the groundwater and runoff outputs from the uplands model into a streamflow model, Dr. Garlanger modeled the riparian system to determine its contributions to streamflow at a point just downstream of OFG. Thus, rainfall is the only addition of water into the uplands system, but rainfall, groundwater outflow from the uplands into the riparian wetlands, and runoff from the uplands into the riparian wetlands are the additions of water into the riparian system. The uplands model is the Hydrological Evaluation of Landfill Performance (HELP) model. Developed for use in analyzing groundwater movement in landfills, HELP generally calculates groundwater outflow based on the hydraulic conductivity of the surficial aquifer divided by the square of the distance from the riparian wetland to the basin divide. In 2001, Dr. Garlanger modified the HELP model (HELPm). The modification multiplies the output from HELP by the square of the maximum height of the water table above the confining layer at the basin divide minus the square of the minimum height of the water table above the confining layer at the riparian wetlands. The only variable in HELPm is the maximum height of the water table above the confining layer; all other values, including those set forth above for HELP, are fixed. The modification improved the HELP model by allowing Dr. Garlanger, among other things, to reduce the extent to which the model is constrained by enabling him to input more realistic hydraulic conductivities. Using HELP, unmodified, Dr. Garlanger had had to input unrealistically high values for hydraulic conductivity. Hydraulic conductivity is either measured in the field or assumed. To simulate OFG without any mining for 25 years into the future, Dr. Garlanger had to obtain an input for hydraulic conductivity. Based on collected data from near the Panhandle as to daily fluctuations in the water table over a two-year period and sub-surface soil composition, as well as other information, Dr. Garlanger determined an average weighted hydraulic conductivity for OFG, pre-mining, of 19 feet per day with a low of 10 feet per day. Dr. Garlanger settled on an initial average weighted hydraulic conductivity of 15 feet per day for the surficial aquifer, but also identified a low-end average of 10 feet per day. As noted above, the contribution of an area of land to streamflow is dependent upon rainfall, evapotranspiration, deep recharge, and the change in storage, which is driven by the elevation of the water table (i.e., the top of the surficial aquifer) as it changes from day to day. Focusing on the vertical components of the water budget, HELPm calculates daily changes in storage, based on water table levels, so as to permit projections of runoff and groundwater outflow from the uplands. For rainfall, Dr. Garlanger relied upon the records of the Wauchula gauge, which is about 10 miles northeast from OFG. Rainfall data for this gauge go back to 1933, although to supplement some missing months, Dr. Garlanger relied on the Ft. Green gauge, which is closer to OFG, but does not go as far back as the Wauchula gauge. To supplement this information on the volume of rainfall, Dr. Garlanger added inputs on the frequency and rate of rainfall. For this calculation, Dr. Garlanger only used rainfall data for the period from 1978 to 2002 because the U.S. Geologic Service has collected streamflow data for Horse Creek at State Road 64 only as far back as 1978. Similar streamflow data for Horse Creek downstream at State Road 72 and for the Peace River go further back. Dr. Garlanger selected this timeframe so he could compare the model output of predicted streamflow to actual streamflow. HELPm calculates evapotranspiration, typically the largest source of water loss, on a daily basis. Dr. Garlanger calibrated evapotranspiration in his simulation by comparing HELPm calculations against average annual values for evapotranspiration for riparian wetlands, uplands, and wetlands in uplands, so as to permit the calculation of an average value of evapotranspiration for the Horse Creek basin above State Road Calibration is the process by which a hydrologist modifies the data inputs to the model based on measured data in order to produce a better match between observed and predicted data. Using generally accepted evapotranspiration values and the standard water-budget formula, Dr. Garlanger calculated average annual evapotranspiration for the Horse Creek basin above State Road 64 of 40.3 inches. He determined the following annual average evapotranspiration rates: riparian wetlands-- 47.5 inches; depressional wetlands--44 inches; seepage wetlands- -47.5 inches; well-drained uplands--34.5 inches; and other uplands--39 inches. Using this information, Dr. Garlanger then found the appropriate average annual evapotranspiration for the OFG uplands that he was modeling, and he reran the model five or six times until it produced outputs for uplands evapotranspiration consistent with this value. For uplands runoff, Dr. Garlanger turned to a well- recognized methodology for estimating the storage available in the uppermost foot of soil, as infiltration is an important factor in determining runoff. For groundwater outflow, Dr. Garlanger uses the one available equation, which is derived from Darcy's Law. Dr. Garlanger then ran his model for the no-mining, during-mining, and after-reclamation options, and he validated the model. In validation, the hydrologist confirms the model's outputs to measured data. In these exercises, Dr. Garlanger compared the predicted groundwater outflows with the empirical values published by Mr. Lewelling and predicted groundwater levels with those measured by IMC near the Panhandle. Dr. Garlanger ran the model with hydraulic conductivities of 10-15 feet per day and drainage times of 5-12 days. He eventually settled on an average hydraulic conductivity of 10 feet per day and an average drainage time of 12 days. Using these values, Dr. Garlanger validated his output by projecting streamflow from the entire 39.5-square mile area upstream of State Road 64, for which data exist. He found that the model produced a reasonable prediction of the flow duration curve. Dr. Garlanger then validated the output by comparing predicted and measured cumulative streamflow from 1978 through 1987, during which time mining in the Horse Creek basin was insignificant. He found a very good matchup between actual data and his model's predictions. Validating the output for average daily and average annual streamflow against actual data, Dr. Garlanger again found that the model performed acceptably. Dr. Garlanger then was prepared to model the 5.4 square-mile area for impact on Horse Creek streamflow at State Road 64 for 25 years without mining, during mining, and for 25 years after reclamation. For during-mining conditions, Dr. Garlanger assumed that the ditch and berm system would capture all of the runoff and none of the groundwater. For post-reclamation conditions, Dr. Garlanger assumed that the cast overburden spoil piles would be parallel to the flow of groundwater or, where that is not practicable, that the top of the spoil piles would be shaved by progressive amounts, ranging from five feet at the groundwater (or basin) divide progressively to 15 feet at the riparian wetland. This is vital to his calculations because of the vast difference in hydraulic conductivity of cast overburden spoil piles as compared to sand tailings. When oriented perpendicular to groundwater flow and unshaved, these spoil piles would act as underground dams, blocking the flow of groundwater. Dr. Garlanger modeled streamflow, in Horse Creek at State Road 64, which is just downstream of the confluence of Horse Creek and West Fork Horse Creek, under two scenarios: hydraulic conductivity of ten feet per day and drainage time of 12 days and hydraulic conductivity of fifteen feet per day and drainage time of five days. For post-reclamation hydraulic conductivity, Dr. Garlanger used 12 feet per day. With the higher streamflow reductions resulting from the lower hydraulic conductivities, Dr. Garlanger projected streamflow reductions, during mining, from 1.07-2.41 cfs and, after reclamation, from 0.10-0.14 cfs. These are average annual values. Generating a flow duration curve for Horse Creek at State Road 64 and using the more adverse data from the lower hydraulic conductivity value, Dr. Garlanger found a slight decrease, during mining, in flow during low-flow conditions, reflecting the mining of the Panhandle tributaries that contributed to groundwater outflow. Generating a stage duration curve, to depict the elevation of the water in the stream during the low-flow condition, Dr. Garlanger demonstrated that the difference is about three inches. After reclamation, as compared to pre-mining conditions, Dr. Garlanger determined that the average flow is decreased by 0.1 cfs, probably due to increased evapotranspiration from the additional reclaimed wetlands. This generates no discernible difference in the two flow duration curves for Horse Creek at State Road 64. Dr. Garlanger thus reasonably concluded that mining would not adversely affect the flow of Horse Creek at State Road 64 or dehydrate wetlands in the no-mine area. He concluded that, after reclamation, the impact would be de minimis as a decrease of 0.1 cfs is beyond the ability to measure flows. Farther downstream, at State Road 72, which is downstream of the confluence of Brushy Creek and Horse Creek, Dr. Garlanger calculated projected streamflow reductions, during mining, from 1.2-2.8 cfs and, after reclamation, from 0.12-0.16 cfs, which are too small to measure. Likewise, there are no discernible differences in the flow duration curves at State Road 72. Downstream of the confluence of Horse Creek and the Peace River, at Ft. Ogden, Dr. Garlanger calculated that the reduction in streamflow caused by mining at OFG would be equivalent to the reduction caused by a decrease of 0.01 inches of rainfall in the Peace River basin. Mr. Davis voiced many objections to Dr. Garlanger's streamflow calculations based on his reliance on HELPm. These objections are addressed at the end of the next section. Mr. Davis also voiced objections to Dr. Garlanger's calculations based on his understatement of the impact of phosphate mining on streamflow. As already noted, Dr. Garlanger made the better case on this issue. Distinguishing between the two rainfall eras in the Peace River basin--1933-1962 and 1969-1998--Dr. Garlanger reported that the measured average streamflow of the Peace River in the latter era was about 4.33 inches lower than the average streamflow of the Peace River in the former era. Finding that decreased average rainfall reduced streamflow by 3.75 inches per year, Dr. Garlanger calculates that the remaining 0.58 inches per year reduction in streamflow was largely due to an increase in deep recharge from 3.37 inches annually in the earlier era to 6.3 inches annually in the latter era. Anthropogenic changes in the Peace River basin have had opposing effects on streamflow. Urbanization, which causes increases in impervious surface, have increased runoff at the expense of evapotranspiration, thus increasing streamflow-- although certain demands of urbanization, such as groundwater pumping for potable water and industrial uses, will increase deep recharge, thus decreasing streamflow. Groundwater withdrawals by agriculture, industrial, utilities, and phosphate mining, net of the returns of these waters, have increased deep recharge, which, as just noted, decreases streamflow. Historically, phosphate mining's profligate use of deep groundwater also released much of the water back to streamflow, although the industry's historic predilection for Land-and-Lakes reclamation increased evapotranspiration and thus reduced streamflow. Converting inches of streamflow to cfs, Dr. Garlanger makes a good case that the streamflow of the Peace River is down about 500 cfs, mostly due to reduced rainfall amounts. About 50 cfs of that reduction is due to anthropogenic effects, and 5-15 cfs of man-caused reductions in the streamflow of the Peace River are due to phosphate mining. By contrast, Mr. Davis unconvincingly attributed a three-inch reduction in streamflow at the South Prong Alafia River to phosphate mining. This reduction in streamflow may be explained by Mr. Davis's failure to apply a lower and more reasonable streamflow assumption, absent mining; a lower and more likely rainfall amount; and a higher and more likely evapotranspiration rate. Wetland Hydroperiods and Inundation Depths 694. In making his groundwater calculations, Dr. Garlanger attempted to predict the behavior of the surficial aquifer, post-reclamation, and the ability of runoff and the water table to support the hydroperiods and inundation depths of the wetlands in the no-mine area and reclaimed wetlands. For this phase of his hydrological work, Dr. Garlanger again used the HELPm for the uplands and a long-term simulation model for the depressional wetlands in the uplands. The long-term simulation model is very similar to the streamflow model used for the riparian-wetland component of the streamflow modeling. Notwithstanding the replacement of the present geology with its more limited vertical permeability with wide bands of sand tailings down to the clay confining layer, Dr. Garlanger believes that deep recharge will remain unchanged by mining and reclamation because groundwater levels will return to their pre-mining elevations. To analyze the ability of the post-reclamation water table to support the reclaimed wetlands, Dr. Garlanger took 12 wetland cross-sections and projected fluctuations in water table and hydroperiod. These are presumably the 13 wetland complexes identified in Figure 13-3, described above. Dr. Garlanger testified about one modeled reclaimed wetland in detail--a freshwater marsh fringed by a wet prairie. This is E046/E047, which is a combined 16.1-acre wetland that is upgradient from E048, which is six-acre mixed wetland hardwoods that will replace the east half of a bay swamp (G166) and mixed wetland hardwoods fringes (G166B and G166C). Dr. Garlanger performs an iterative process based on a post-reclamation topographic map that starts with substantially pre-mining topography. Identifying the HELPm inputs, Dr. Garlanger takes the length of the upland to the riparian system and the assumed hydraulic conductivity based on the relative depths of sand tailings and cast overburden, and he then runs HELPm to determine the daily upland runoff and groundwater outflow. Dr. Garlanger then calculates the maximum height of the water table above the confining layer at any point downgradient from the basin divide to the riparian wetland. To input hydraulic conductivity, Dr. Garlanger testified that he obtains a value "based on the spoil piles and the depth that the spoil pile will be cut down to adjacent to the preserved area." (Tr, p. 2993) Applying the output to a wetlands model that is similar to the streamflow model, Dr. Garlanger then engages in an iterative process in which he adjusts and readjusts the post- reclamation topography to produce the proper elevation of the bottom of each modeled wetland for the hydroperiod that is stipulated for the vegetative community to be created in that location. Besides changing the bottom slope of each seepage wetland, the major adjustments for each wetland are narrowing its outlet or lowering its bottom elevation to extend its hydroperiod and deepen its inundation depth or broadening its outlet or raising its bottom elevation to shorten its hydroperiod and make its bottom elevation more shallow. Dr. Garlanger modeled the iterative process by continuing it late into the hearing, as he and IMC surveyor, Ted Smith, produced a "final" post-reclamation topographic map at the end of the hearing. Actually, even this map is not final, as Dr. Garlanger testified that he and Mr. Smith will produce the final topographic map, for wetlands, after the area is mined, photographed, backfilled, and graded, at which time they will know the location and direction of the cast overburden spoil piles. Dr. Garlanger will then use a calibrated model to account for actual in situ conditions. Due to the flatness of OFG, it is possible, even at this late stage, to regrade the sand tailings, if necessary for hydrological purposes. Monitoring wells will produce substantial data on the hydraulic conductivity of the no-mine area, as well as the hydroperiods of existing wetlands and the frequency with which seepage wetlands release water. Dr. Garlanger and IMC employees will also measure the hydraulic conductivity of the sand tailings and overburden in the reclaimed areas, also to assist their preparation of the final topographic map. As noted above, ERP Specific Condition 16.B.2 requires IMC to model 24 reclaimed wetlands to demonstrate successful water table re-creation and hydroperiod and inundation depth reclamation. Dr. Garlanger applied his models to confirm that, for each of the 24 modeled wetlands, the design topography and hydrology would produce the targeted hydroperiod and inundation depth. Mr. Davis modeled three reclaimed bay swamps. Bay swamps are the hardest wetlands for which to reclaim an appropriate water table due to their long hydroperiod, shallow inundation depths, and seepage characteristics. As noted above, no successful reclamation of bay swamps has ever taken place, except under circumstances inapplicable to OFG. The three reclaimed bay swamps are: E008, a 0.7-acre bay swamp abutting the west side of the Stream 1e series; E063, a 1.3-acre flow-through bay swamp in Stream 5e; and W039, an 11.2-acre bayhead from which Stream 1w will flow. W039 is a very large reclaimed wetland. After the 20.7-acre wet prairie (W003) to be reclaimed at the headwaters of Stream 9w and the 23.8-acre mixed wetland hardwoods (E003) lining the Stream 1e series, W039 is the largest reclaimed wetland at OFG, along with E018/E020, which are the isolated wet prairie fringe and freshwater marsh on the east side of Section 4. Mr. Davis testified as a witness in surrebuttal, which was necessitated by a late change by IMC in post- reclamation topography for these three bay swamps. Mr. Davis implied that he understood these three bay swamps better than he did the other reclaimed wetland systems. The fact is that he did understand these three reclaimed bay swamps better than he did any other reclaimed wetlands. Prior to testifying, at the order of the Administrative Law Judge, Mr. Davis and Dr. Garlanger conferred so that Mr. Davis, in preparing to respond to the "final" post-reclamation topography, would clarify any uncertainty about how Dr. Garlanger was modeling these wetlands and projecting their hydroperiods and inundation depths. Mr. Davis identified Dr. Garlanger's topographical changes to these three bay swamps. For E008, Dr. Garlanger lowered the west end of the wetland by 0.5 feet, extended a 114-foot contour up the channel, just east of an existing 115- foot contour, and possibly adjusted the slope. For E063, Dr. Garlanger lowered the bottom elevation by one foot, so that it can now store 0.3 feet of water, given its overflow popoff elevation. And for W039, Dr. Garlanger removed a slope and flattened the bottom, so that it can store 0.3 feet of water. From Dr. Garlanger's spreadsheets, Mr. Davis found the values for runoff, groundwater, and rainfall entering each wetland. Mr. Davis found that E008 received only 10 percent of its water from runoff, more of its water from rainfall, but most of its water from groundwater inflow. Noting that E008 abuts a reclaimed xeric area, Mr. Davis recalled a 6:1 ratio of groundwater inflow to runoff inflow. Mr. Davis explained that E008 loses most of its water to runoff. Mr. Davis found that the groundwater input for this wetland was consistent with the testimony of biologists, such as Deputy Director Cantrell, that bay swamps are primarily groundwater-driven systems, but questioned the absence of groundwater outflow to the adjacent, down-gradient riparian wetland (E003). For E063, however, Mr. Davis found that inputs from runoff, a more important source of water for this wetland, were about the same as inputs from groundwater. Although he did not testify to this fact, E063 is an unusual reclaimed bay swamp because it is the only one that will serve as a flow-through wetland, situated, as it is, in the middle of Stream 5e. This would seem to explain the larger role of surface water inputs than is typical of bay swamps adjacent to uplands. For W039, Mr. Davis found a small percentage of surface water and larger percentages of groundwater and rainfall as water sources for this wetland. Rainfall inputs would be greater due to the large area of the wetland, according to Mr. Davis. As a headwater wetland abutting uplands, W039 would be expected to have a higher input ratio, than E063, of water from groundwater versus runoff. Mr. Davis noted that W039 lost about half of its water to evapotranspiration, which would also make sense given its large surface area, and half to runoff, which would make sense given its status as a headwater wetland for Stream 1w. Mr. Davis then ran his MIKE SHE model to predict the hydroperiod for each wetland. This model is described in more detail at the end of this subsection. In simulating the hydrology of the reclaimed OFG, Mr. Davis assumed that the overburden spoil piles would be parallel to the direction of groundwater flow and eliminated any differential depressional storage, but he continued to assume two inches of depressional storage. (These assumptions are also discussed in connection with the MIKE SHE model.) Mr. Davis found that the 11.2-acre W039 will have a perfect hydroperiod. Its inundation hydroperiod will range from 8.6 months to 11.0 months, from bottom to top. Its saturation hydroperiod, which is water measured to a depth of 0.5 foot below the bottom of the wetland, will range from 8.8 months to 11.1 months, from bottom to top. Mr. Davis found that the 1.3-acre E063 will have a hydroperiod of 11.9 months, which is 0.9 months too long. Mr. Davis found that the 0.7-acre E008 will have a hydroperiod of 2.7 months for inundation and 4.6 months for saturation, which is about four months too short. 714. Crediting Mr. Davis's testimony, IMC's successful reclamation of an 11.2-acre bay swamp, dependent upon upland surface water and groundwater inputs, would be an unprecedented success. As discussed below, Mr. Davis's depressional assumption is not credited, so the hydroperiod of E063 would be shorter than the 11.9 months that he has calculated. Also, this reclaimed system will be a seepage system that would not permit the build-up of much standing water, so, even crediting Mr. Davis's calculations, Dr. Garlanger has achieved the proper hydrology for its reclamation too. It is more difficult to resolve the conflict in simulated hydroperiods for E008. E008 is a more complicated wetland to model because it is part of a reclaimed complex consisting of nine reclaimed wetlands. No other wetland complex to be reclaimed at OFG approaches this number of different communities in a single complex. Except for E018, which, although 30.7 acres, is a much simpler wetland system because it is an isolated complex of three wetlands, no other wetland complex to be reclaimed at OFG comes close to the area of the Stream 1e series' wetlands complex, which totals 35.1 acres, or over 10 percent of the wetlands to be reclaimed at OFG. Mr. Davis's unjustified depressional assumption generates excessively wet conditions, but, for E008, he found its hydroperiod to be too short by at least 3.4 months. And, of course, E008 is the difficult-to-reclaim bay swamp. The two models invite comparisons at this point. Mr. Davis's model, MIKE SHE, enjoys wide usage for calculating streamflows, hydroperiods, and inundation depths, as it has been used in these cases. MIKE SHE has been used successfully in large-scale settings. On the other hand, HELP was designed for calculating water levels in landfills. For calculating the uplands component of streamflow and hydroperiod, HELPm is used by Dr. Garlanger alone. The author of HELP's routine for lateral drainage and the subroutine for unsaturated vertical flow, Bruce McEnroe, pointed out that this model could accommodate only a regular, homogenous drainage layer, as would be found in a landfill, and could not accommodate the irregular, heterogeneous aquifer layer, which Dr. Garlanger was modeling. Mr. McEnroe also explained that the downstream boundary condition of HELP, which is free drainage, does not resemble the actual downstream boundary condition, in which groundwater cannot typically drain freely, and this limitation applies equally to the pre-mining and post-reclamation scenarios. Mr. McEnroe also found a mathematical error, but Dr. Garlanger later showed that it would alter results inconsequentially. Complaining about Dr. Garlanger's failure to provide comment lines in his source code, where he modified HELP, Mr. McEnroe emphasized that the model, as modified and used by Dr. Garlanger, really was no longer the HELP model. Counterposed to Mr. McEnroe's testimony was the testimony of Mark Ross, an associate professor of civil and environmental engineering at the University of South Florida College of Engineering. Professor Ross has 20 years' experience in hydrological modeling and has worked with the Florida Institute of Phosphate Research model that Mr. Davis helped develop, but which no longer is supported or in much use. Professor Ross conducted a peer review of the HELPm model, spending 20-30 hours in the process, exclusive of time spent discussing the model with Dr. Garlanger. Professor Ross endorsed Dr. Garlanger's use of a single value of .75 for evapotranspiration in riparian wetlands and his use of a weighted hydraulic conductivity. Professor Ross acknowledged that more complex models were available, but correctly opined that the simplest model was best if it could accommodate all of the available data. Although the emphasis in his testimony was on streamflow, Professor Ross addressed wetlands and their hydroperiods sufficiently to assure that his opinion of the sufficiency of the HELPm model covered both tasks. The interplay between the complexity of the model and availability of data emerged more clearly with the testimony of Authority hydrologist Henrik Sorensen, who developed code for the MIKE SHE model. Successful applications of this model range from the Danube River to Kuala Lampur to South Florida. The Danube River project was the construction of a dam, and hydrologists ran MIKE SHE to project the impact of the diverted streamflow on riparian wetlands. The Kuala Lampur project was the construction of a new city, and hydrologists ran MIKE SHE to project the impact of vastly changing land uses on the water level in the peat wetlands. South Florida projects have included a number of analyses of wetlands impacts of proposed activities. At Lake Tohopekaliga, hydrologists used MIKE SHE to project the effects on the water table and nearby wetlands of a 6-7 foot drawdown of the lake to remove muck. Unlike HELPm, MIKE SHE is an integrated model, meaning that all of its components are contained in a single model. Significant for present purposes, MIKE SHE integrates surface water and groundwater analysis in a single model, so as to facilitate the modeling of the interaction between a stream and surficial aquifer. This is especially important for simulating interactions between the surface and shallow water tables. MIKE SHE is a physically based model, meaning that it is based on equations derived from the laws of nature. In using HELPm and the spreadsheet models for streamflow and hydroperiod, Dr. Garlanger of course relies on laws of nature, but also relies on conceptualizations to link equation-driven outputs. As Mr. Sorensen explained, MIKE SHE is based on differential equations, so that it is dynamic as to time and space, but Dr. Garlanger's models are based on analytic equations, so they are limited to state-to-state solutions. The conceptualizations that link outputs and essentially integrate Dr. Garlanger's pairs of models are only as good as the conceptualizer, who, in the case of Dr. Garlanger, is very good, but conceptualizations can become so pervasive that the model loses its reliability and adds little or nothing to a conceptual exercise using an analytic model. Unlike MIKE SHE, HELPm is a lump-parameter model, which necessitates the input of average hydraulic conductivities, evapotranspiration rates, and leaf area indexes over relatively large areas and, in the case of evapotranspiration rates, sometimes at the expense of their calculation. Constraining a model, by inputting, rather than calculating, values to force results within an expected range, may resemble validation, but when the inputs become unrealistic, as Dr. Garlanger's hydraulic conductivity values were before he modified HELP, the model's credibility is impaired, not enhanced, by the process. Conceptualizations can eventually constrain modeled simulations so as to undermine confidence in the model's outputs. Unlike HELPm, MIKE SHE is spatially distributed, so that different land use types may be distributed throughout the model. HELPm may input different land uses for different basins, but MIKE SHE allows the user to input different land uses for different cells, each of the user's choice as to size. As noted by Mr. McEnroe, HELP was developed to simulate a shallow system running to a drain, and it remains well-suited for this task. In tracking the water table, HELPm assumes a constant thickness of the drainage layer, which reflects the design of landfills, not natural systems. As IMC contends, the post-reclamation geology will be far simpler than the pre-mining geology at OFG, but even the post-reclamation hydrology is far more complex than that of a landfill. With a 35:1 ratio of hydraulic conductivities, the surficial aquifer must negotiate the 330-foot wide valleys of sand tailings separated from 180-foot wide plateaus by 33-degree overburden slopes. Overburden peaks would have been simpler than overburden plateaus because the effective depth of sand tailings would have been at least five feet over nearly all of the mined area; as already noted, these overburden plateaus mean that, exclusive of shavings and toppings, overburden at less than five feet finished depth occupies about 28 percent of the surface of the mined area. This geology is much more complicated than the uniform geology of a landfill, especially when trying to project the surface water and groundwater inputs and outputs of shallow wetlands and streams, some of which will span several phases of this unusual geology. Unlike HELPm, MIKE SHE is used for its designed purpose when used for projecting streamflow and wetlands hydroperiods and inundation depths. It is widely used, peer- reviewed and supported with two or three updates annually. Mr. Sorensen made an interesting point when he opined that HELPm does a good job with average flows. This explains HELPm's reliability in calculating streamflows. Notwithstanding the calculation of peak discharge curves, accurate streamflow calculations--at least in this part of Florida--tolerate calculations based on average conditions and approximations much better than do accurate calculations of hydroperiod and inundation depths, especially concerning shallow wetlands in wetland complexes. MIKE SHE is not without its shortcomings, at least as applied in these cases. For his MIKE SHE simulation, Mr. Davis did not simulate first- and second-order streams, perched groundwater flow (i.e., interflow), or shallow concentrated overland flow, and, despite the model's sophistication, he still had to perform conceptualizations, such as of drainage. Mr. Davis's first two post-reclamation runs, prior to his final run of the three bay swamps, suffered from faulty assumptions. First, he assumed depressions and differential depressions based on a settling that Dr. Garlanger, with geotechnical engineering experience that Mr. Davis lacks, testified convincingly would not occur. Second, Mr. Davis assumed that the spoil piles would be oriented perpendicular to the direction of groundwater flow. Mr. Davis likely knew that IMC had agreed on December 23, 2003, to orient the mine cuts parallel to the direction of groundwater flow, to the extent practicable. Mr. Davis modeled the perpendicular scenario presumably due to the vagueness of the assurance, set forth only in the introduction to the January submittal, and thus unenforceable, that IMC would grade or shave the tops of overburden plateaus of spoil piles running perpendicular to groundflow. When performing his modeling, Mr. Davis could not have known of Dr. Garlanger's recommendation, as contained in a letter dated April 29, 2004--less than two weeks prior to the start of the final hearing--that IMC shave 5-15 feet off any perpendicular cast overburden spoil piles or that IMC would accept Dr. Garlanger's recommendation during the final hearing. As agreed to by IMC during the hearing, it will bulldoze any spoil piles oriented perpendicular to the direction of groundwater flow from 5-15 feet: the cut would allow five feet of sand tailings nearest the groundwater divide and would progressively deepen to allow 15 feet of sand tailings nearest the stream. For an average width of overburden of 195 feet with five feet thickness of sand tailings, which is the width calculated above under the less-favorable hydrological scenario with regard to the bases of the sand tailings valleys and cast overburden plateaus, Dr. Garlanger calculated a hydraulic conductivity of seven feet per day. Mr. Davis assumed that IMC would not be able to orient the spoil piles parallel to groundwater flow, but nothing indicates that the proper orientation of these piles will be impracticable over significant areas of land. If a turn of the dragline near Horse Creek leaves a relatively short area of spoil perpendicular to groundwater flow and if IMC will shave this area as it does rows, shaving the pile down 15 feet would substantially improve water table/shallow wetland interaction over the portion of the mined area that is left with an overburden plateau. Conceptualizing the contingency of a spoil pile blocking groundwater flow close to Horse Creek, such as from the U-turn of the dragline at the end of a row, the bulldozing of that spoil pile down to an effective 15-foot depth would leave a depth of at least 15 feet of sand tailings running 1095 feet, as measured alongside of Horse Creek out to a point at which the spoil piles would again run parallel to groundwater flow. If all of the spoil piles turned at Horse Creek and assuming that IMC will cut down the cast overburden piled against the sides of the mine cuts, for the distance equal to the distance between the edge of the no-mine area to the start of the curve, sand tailings would be at least 15 feet deep. The real problem with MIKE SHE, as applied at OFG, is its sophistication. Mr. Sorensen admitted that he had not reviewed the data available for this part of Florida, but claimed that he knew, based on his work in South Florida, that sufficient data existed to run the MIKE SHE model. This is highly unlikely. In addition to Mr. Davis's observation about the lack of data, the record reveals a slimmer universe of data than Mr. Sorensen imagined to exist. Measured values for the hydraulic conductivity of pre-mined or post-reclaimed areas are largely unavailable. For specific reclamation sites, little data exist of pre-mining and post-reclamation soil textures, water tables, and wetland hydroperiods and stage elevations. By volume, the two most critical inputs are rainfall and evapotranspiration, which must be calculated or assumed because, for practical purposes, it cannot be directly measured. A major determinant of evapotranspiration is the water table elevation. The critical inputs of rainfall and water table elevations illustrate the shortcomings of the data for these cases. Rainfall records in the general area cover a long period of time, except that collection points are usually far enough away from the site to be analyzed as to raise the probability of significant daily fluctuations, which average out over time. MIKE SHE inputs rainfall spatially and hourly while HELPm inputs a single daily value. Without regard to any particular application, MIKE SHE is the superior model on this point, but its superiority is wasted when the data of hourly rainfall for individual cells are unavailable and values, often based on much longer intervals at much greater distances, must be interpolated. Records for most surficial aquifer monitoring wells in the area date back only to the early 1990s and are fairly spotty as to locations. MIKE SHE inputs spatially distributed groundwater elevations, while HELPm inputs a single value. If, as Mr. Davis testified, multiple inputs of water table elevations, for which direct OFG data are unavailable, must rely on a hydrologist's knowledge of surficial aquifer responses, MIKE SHE would share the same tendency of HELPm--at least for this variable--of relying on external guidance to produce its output. By contrast, the scientists studying the Danube River had lacked the resources for many years to do much more than collect data, so the data for the Danube MIKE SHE simulation was much richer than the data available at OFG. In such data-rich environments, MIKE SHE is the superior model for wetland hydroperiods and inundation depths. The question in these cases is whether, given the limitations of the OFG data and HELPm in simulating hydroperiods and inundation depths, IMC has still provided reasonable assurance of the reclamation of functional hydroperiods and inundation depths for reclaimed wetlands. IMC's case as to reclaimed hydroperiods and inundation depths is undermined by certain aspects of the use of HELPm in these cases. The scientific method, which lends confidence to analysis-driven conclusions to the extent that others can reproduce the analytic process, is poorly served by computer code that is modified without notation and modeling results that no one can reproduce due to the repeated intervention of the modeler, applying his touch and feel to the simulation. Only at the end of nearly eight weeks of hearing and a conference between Dr. Garlanger and Mr. Davis could Mr. Davis finally gain sufficient understanding of Dr. Garlanger's modeling process to make a meaningful comparison between his conclusions and Dr. Garlanger's conclusions for the hydroperiods and inundation depths of three wetlands. When applied to project streamflow, with its relative amenability to average inputs, and when applied to projecting the hydroperiods and inundation depths of deeper and more isolated wetlands, HELPm, as used by Dr. Garlanger, who, as an experienced and highly competent hydrologist, can adjust and re- adjust inputs and outputs, produces reasonable assurance. However, Mr. Davis's analysis of Dr. Garlanger's work and other factors preclude a finding that Dr. Garlanger has provided reasonable assurance that IMC will reclaim a functional hydroperiod and inundation depths for E008. The finding in the preceding paragraph implies no similar rejection of Dr. Garlanger's modeling of the other wetlands. Most of the modeled reclaimed wetlands are isolated and do not present the challenge of simulating complex interactions among them, where an error in modeling an upgradient wetland will cause an error in modeling a downgradient wetland. A couple of the modeled reclaimed wetlands are headwater wetlands, which Dr. Garlanger has demonstrated his ability to model in W039. Outside of the Stream 1e series, the only wetlands similar in location to E008, as attached to a riparian system, will be E040, E048, E054 complex, and W044, of which only E048 is to be modeled. Mr. Davis also addressed E048 in surrebuttal. A wetland forested mixed, E048 will replace a high-functioning bay swamp abutting, or a part of, the riparian wetlands of Horse Creek. Mr. Davis admitted that he could agree with Dr. Garlanger's analysis of inputs into E048 from isolated reclaimed wetlands upgradient of E048, so that he could agree with Dr. Garlanger's projected hydroperiod for this reclaimed wetland. However, Mr. Davis explained that E008 is located in the flatter Panhandle, but that E048, as well as the other reclaimed wetlands listed in the preceding paragraph, are located in areas characterized by steeper grades and more xeric conditions, which support Dr. Garlanger's emphasis on groundwater inputs over surface water inputs. Peak Discharges During mining, the ditch and berm system prevents adverse flooding. If it operates as intended, the ditch and berm system delays the release of runoff from OFG by re-routing it through one of the NPDES outfalls. This decreases peak discharge downstream of OFG. Presumably, IMC will operate the recharge wells in anticipation of storm events--allowing the water levels to lower in advance of storms and maintaining higher water levels in advance of drier periods--so as not to raise the possibility of flooding by way of accelerated discharges through the NPDES outfalls. Failure of the ditch and berm system is highly improbable. The sole failure reported in this record did not involve a system as engineered as the one proposed for OFG, according to Dr. Garlanger. Another possible source of flooding during mining arises from the designed blockage of flow from unmined areas. IMC plans a single, elevated pipeline crossing across Stream 2e, and Dr. Garlanger explained that the design of the culvert, as part of this temporary crossing, will not result in adverse flooding during mining. Similar design work by Dr. Garlanger will be necessitated, if DEP issues a Final Order incorporating the recommendation below that the Stream 1e series and its 25-year floodplain also be placed in the no-mine area. The riparian wetlands for the Stream 1e series are narrowest along Stream 1ee, so this may be the location that DEP determines for the dragline walkpath corridor, if DEP determines that IMC may maintain a dragline crossing anywhere along the Stream 1e series. The sole issue, during mining, involving peak discharges is a legal question, which is whether IMC's ditch and berm system has the capacity to accommodate the design storm. As noted below, the design storm is the 25-year storm, if the ditch and berm system is an open drainage system, and the design storm is the 100-year storm, if the ditch and berm system is a closed drainage system. The capacity of the proposed ditch and berm system is designed to accommodate the 25-year storm, but not the 100-year storm. The facts necessary to determine if the ditch and berm system is open or closed are set forth above. In its Final Order, DEP must characterize a system that is closed in the sense of the availability of a passive discharge outfall, but open in the sense that, with the intervention of pumps--assuming the availability of electricity during a major storm or alternative sources of power--excessive volumes of water may be moved to an NPDES outfall. This is a minor issue because, even if DEP determines that the ditch and berm is a closed system, IMC may easily heighten the berm as necessary to accommodate the 100-year storm. Post-reclamation, many of the changes that IMC will make to OFG will reduce peak discharges. The agricultural alterations that ditched and drained wetlands accelerated drainage and increased peak discharges downstream, as compared to pre-existing natural drainage rates and peak discharge volumes. The removal of these ditches, the net addition of 24 acres of forested wetlands and 48 acres of herbaceous wetlands, the addition of sinuosity and in-stream structure to the reclaimed streams, and the redesigning of the banks of the reclaimed streams so as to permit communication between the reclaimed streams and their floodplains will attenuate floodwaters, slow the rate of runoff, increase temporary storage, and ultimately reduce peak discharges from their present values. Dr. Garlanger modeled peak discharges using the Channel Hydrologic Analysis Networking (CHAN) model, which is a widely accepted model to simulate peak discharges. As already noted, Mr. Loper found several inconsistencies and flaws in earlier modeling, but Dr. Garlanger, undeterred, re-ran the CHAN simulations, incorporating Mr. Loper's findings, as Dr. Garlanger deemed necessary. The bottom line is that, post-reclamation, very small increases in peak discharges will occur at the Carlton cutout and would occur at some property immediately downstream of the point at which Horse Creek leaves OFG. The owners of the Carlton cutout consented to the very minor flooding of their pasture land, and IMC, of course, has no objection to the very minor flooding of its downstream property. Even absent these consents, the very limited extent and frequency of flooding, given the prevailing agricultural uses in the area, could not be characterized as adverse. Among the points raised by Mr. Loper was the absence of mapping of any floodplain besides the 100-year floodplain of Horse Creek. The omission of other floodplains is of environmental or biological importance, but not direct hydrological importance. If for no other reason than that IMC will replicate pre-mining topography, especially at the lower elevations, there will be no loss of floodplain storage. 4. Water Quality Water quality violations characterize past efforts to reclaim streams, other than Dogleg Branch, but the good water quality at Dogleg Branch means that the phosphate mining industry can reclaim streams and maintain water quality, post- reclamation. The intensive engineering in IMC's Stream Restoration Plan raises the prospect of successfully reclaimed water quality, especially among the simpler, more altered stream systems to be reclaimed. There is little doubt that, during mining, few impacts to water quality take place. The ditch and berm systems in place during the upstream mining in the Horse Creek sub-basin have permitted no degradation of water quality. Given the present condition of most of the tributaries and extensive agricultural alterations of most of OFG, successful reclamation may be expected to result in certain changes to water quality, among already-altered tributaries, at least once the reclaimed communities have established themselves. Successful reclamation of these streams and their channels should lower turbidity, by replacing their incised, unstable stream channels and banks with stable channels and banks. The addition of riffles and structure to the stream bed should raise dissolved oxygen levels in these streams. Excluding cattle from these streams, by placing cattle ponds away from Horse Creek and vegetatively screening Horse Creek and the tributaries, should lower adverse impacts, such as turbidity, due to cattle damage to the banks, and nutrient loading, due to cattle waste discharges. Phosphorus is sometimes temporarily higher after mining, but this may be merely a trophic surge. Water temperature will cool with the addition of forested riparian wetlands, once the canopy develops, where none presently exists. However, none of these effects can be anticipated with the reclamation of the relatively pristine Stream 1e series. Other reclamation activities may also be anticipated to improve water quality. These activities include adding net wetlands area, replacing low-functioning wetlands with wetlands with the potential to achieve high-functioning levels, concentrating wetlands more around streams, adding supportive uplands, and otherwise increasing storage and slowing runoff. These activities will raise the level of natural filtration, compared to the natural filtration presently performed at OFG. Wildlife Management and Habitat The wildlife management plans are reasonable accommodations of wildlife that presently use OFG, based on the frequency of the usage by each species and the degree of protection afforded certain species. It is important that IMC update wildlife utilization information for the period that elapses between the site visits and the commencement of mining; wildlife usage by some species, especially the Audubon crested caracara, was discovered shortly before the hearing and, if later found to be more intense, will require more intensive wildlife management plans. Likewise, DEP will need confirmation of FWC's approval of IMC's gopher tortoise relocation plan. Always of especial concern is the Florida panther. Obviously, the accommodations necessary for one or two male Florida panthers visiting OFG are far less intensive than those necessary if a breeding pair had established themselves at the site. Ms. Keenan testified that the ERP/CRP approval should have incorporated the entire Habitat Management Plan. Although the ERP and CRP approval would be strengthened by the incorporation of the Habitat Management Plan, and DEP may elect to do so in its Final Order, the provisions actually incorporated adequately address wildlife management concerns. The evidence fails to establish that OFG, which has been logged over the years, presently supports red cockaded woodpeckers. Clearly, as is the case with the Audubon's crested caracara, IMC is committed to develop, prior to mining, appropriate management plans that meet the needs of whatever species are found using OFG between the hearing and the start of mining. In general, the reclamation of OFG will improve the value of the area for wildlife habitat. The concentration of reclaimed wetlands reduces induced edge by 36 miles. Induced edges artificially increase predation and decrease the function of the upland/wetland interface for those aquatic- or wetland- dependent species that rely on adjacent uplands during parts of their life cycle. The increased breadth of the riparian wetlands, which has been detailed above, also improves wildlife utilization and habitat values by discourage cattle from using the streams and adjacent wetlands. IMC's reclamation plan slightly increases the area of cattle ponds and locates them farther away from sensitive wetlands and streams. IMC's reclamation plan also serves the often- overlooked needs of amphibians. The creation of isolated and ephemeral wetlands, which will not receive floodwaters from Horse Creek or its tributaries in most storm events, will enable these amphibians to develop sustainable populations and flourish. At present, two factors have led to artificially high levels of predation of these amphibians by small fish. Ditching of formerly isolated wetlands and the proximity of still- isolated wetlands to tributaries and their connected wetlands-- so as to allow runoff to connect the two systems during storm events--allow small fish to enter the habitat of the amphibians and prey upon them at artificially high rates. Mitigation/Reclamation--Financial Responsibility IMC has never defaulted on any of its reclamation or mitigation responsibilities. Its mitigation cost estimates are ample to cover the listed expenses of the proposed wetlands mitigation, with two exceptions. For reasons set forth in the Conclusions of Law, IMC is not required to post financial security at this time for any CRP reclamation, such as the reclamation of uplands not relied upon by aquatic- and wetlands- dependent species, that is not also ERP mitigation. However, the listed expenses omit two important items of ERP mitigation. First, the listed expenses omit Dr. Garlanger's fees for final engineering work on wetlands hydroperiods and inundation depths after backfilling has been completed. This is an expense covered under reclamation, as well as mitigation, pursuant to Chapter 378, Part III, and Chapter 373, Part IV, Florida Statutes, respectively. Second, the listed expenses omit the cost of acquiring sand tailings, transporting them to the mine cut, and contouring them. For the reasons discussed in the Conclusions of Law, the cost of obtaining and transporting the sand tailings is not required under reclamation, pursuant to Chapter 378, Part III, Florida Statutes, but is required under mitigation under Chapter 373, Part IV, Florida Statutes. Charlotte County contends that the cost of obtaining, transporting, and contouring sand tailings is $35,588 per acre, according to Mr. Irwin. This represents $10,588 per acre, as Mr. Irwin's "best guesstimate" for earthmoving, which seems to include the stripping and preserving of the A and B horizons, and $25,000 per acre for the shaping of wetland reclamation units. This testimony includes items for which financial security is not required, such as preserving the A and B horizons, and excludes the third-party cost of acquiring sufficient sand tailings to backfill the OFG mine cuts to the post-reclamation topography and transporting these sand tailings to OFG. The record supplies no information on these costs.
Recommendation It is RECOMMENDED that the Department of Environmental Protection issue a Final Order: Granting the ERP with the conditions set forth in paragraph 884 above. Approving the CRP with the conditions set forth in paragraph 919 above. Approving the WRP modification when the ERP and CRP approval become final and the time for appeal has passed or, if an appeal is taken, all appellate review has been completed. Dismissing the petition for hearing of Petitioner Peace River/Manasota Regional Water Supply Authority for lack of standing. DONE AND ENTERED this 9th day of May, 2005, in Tallahassee, Leon County, Florida. S ROBERT E. MEALE Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 9th day of May, 2005. COPIES FURNISHED: Kathy C. Carter, Agency Clerk Department of Environmental Protection Office of General Counsel Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399-3000 Greg Munson, General Counsel Department of Environmental Protection Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399-3000 Douglas P. Manson Carey, O'Malley, Whitaker & Manson, P.A. 712 South Oregon Avenue Tampa, Florida 33606-2543 John R. Thomas Thomas & Associates, P.A. 233 3rd Street North, Suite 101 St. Petersburg, Florida 33701-3818 Edward P. de la Parte, Jr. de la Parte & Gilbert, P.A. Post Office Box 2350 Tampa, Florida 33601-2350 Renee Francis Lee Charlotte County Attorney's Office 18500 Murdock Circle Port Charlotte, Florida 33948 Alan R. Behrens Desoto Citizezs Against Pollution 8335 State Road 674 Wimauma, Florida 33598 Alan R. Behrens 4070 Southwest Armadillo Trail Arcadia, Florida 34266 Gary K. Oldehoff Sarasota County Attorney's Office 1660 Ringling Boulevard, Second Floor Sarasota, Florida 34236 Thomas L. Wright Lee County Attorney's Office 2115 Second Street Post Office Box 398 Ft. Myers, Florida 33902 Rory C. Ryan Holland & Knight, LLP Post Office Box 1526 Orlando, Florida 32802-1526 Frank Matthews Hopping, Green & Sams, P.A. 123 South Calhoun Street Post Office Box 6526 Tallahassee, Florida 32314 Susan L. Stephens Holland & Knight, LLP Post Office Box 810 Tallahassee, Florida 32302-0810 Francine M. Ffolkes Department of Environmental Protection 3900 Commonwealth Boulevard The Douglas Building, Mail Station 35 Tallahassee, Florida 32399-3000
The Issue The issue to be determined in this case is whether the Florida Fish and Wildlife Conservation Commission (“Commission”) is entitled to the requested minor modification of its existing Environmental Resource Permit and Sovereign Submerged Lands Authorization, which would authorize the backfilling of a portion of Fisheating Creek as part of a restoration project.
Findings Of Fact The Parties The Department is the state agency responsible for regulating construction activities in waters of the State. The Department has also been delegated authority to process and act on applications for authorization from the Board of Trustees for activities on sovereignty submerged lands. The Commission is the state wildlife management agency. The Commission is the applicant for the minor modification at issue in this proceeding. Petitioner, Save Our Creeks, Inc., is a non-profit Florida corporation with its offices in Lake Place, Florida. Save Our Creeks’ members are interested citizens and groups devoted to the conservation of natural resources, especially creeks and small waterways. Save Our Creeks owns property on Fisheating Creek in Glades County, approximately nine miles upstream of Cowbone Marsh. Petitioner, Environmental Confederation of Southwest Florida, Inc. (ECOSWF), is a non-profit Florida corporation with its offices in Sarasota, Florida. A substantial number of the members of Save Our Creeks and ECOSWF use and enjoy the waters of Fisheating Creek for a variety of purposes, including canoeing, boating, fishing, and wildlife observation. Their interests would be affected by the proposed project. Fisheating Creek and Cowbone Marsh Fisheating Creek flows from Highlands and Desoto Counties south and east through Glades County. The Creek runs in a northeastern direction through Cowbone Marsh before draining into Lake Okeechobee. The Creek contributes approximately nine percent of the flow into Lake Okeechobee. Fisheating Creek is designated as Class III waters. Cowbone Marsh is located about eight miles west of Lake Okeechobee. It is a mile and a half long and two miles wide, covering about 2,500 acres. Fisheating Creek and Cowbone Marsh are within the Fisheating Creek Wildlife Management Area. In 1929, the United States Army Corps of Engineers ("USACOE") prepared a survey map which shows Fisheating Creek as an open water route from Lake Okeechobee through Cowbone Marsh and continuing beyond. The accuracy of the course of the Creek as it is depicted in the 1929 map is not disputed by the parties. The 1929 map does not describe the depth or width of the Creek. Some evidence about historical widths and depths was presented, but it was incomplete. There was credible evidence showing that some segments of Fisheating Creek were four to five feet deep and 20 to 30 feet wide. There was also credible evidence that other segments of the Creek were shallower and narrower. The record shows only that canoes, kayaks, and other vessels drawing twelve inches of water or less have been used on the Creek. For a number of years, much of Fisheating Creek has been choked by vegetation and “tussocks.” Tussocks are floating mats of vegetation. Carolina willow now dominates Cowbone Marsh, having replaced areas that were previously open water or covered with herbaceous marsh communities. The vegetation in the Creek made navigation difficult or impossible through Cowbone Marsh. The 1998 Judgment and 1999 Settlement Agreement In 1989, Lykes Bros., Inc., asserted ownership of Fisheating Creek and tried to prevent public access to the Creek. The Board of Trustees responded with a civil action against Lykes Bros., seeking a determination that Fisheating Creek throughout Glades County is navigable and, consequently, the title to its bottom is held by the Board of Trustees as sovereignty submerged lands. Petitioners in this administrative proceeding intervened in the circuit court case on the side of the Board of Trustees. The jury found Fisheating Creek navigable throughout Glades County and the court entered a judgment in 1998 determining that the Creek is sovereignty land held in trust by the Board of Trustees. The judgment did not include any findings about the widths and depths of Fisheating Creek. The court retained jurisdiction to determine the boundaries of the Creek, but the boundaries were never determined. The circuit court case was appealed, but in May 1999, the parties entered into a settlement agreement pursuant to which Lykes Bros. agreed to sell to the Board of Trustees a conservation easement on upland areas adjacent to Fisheating Creek, to be held and managed for the benefit of the public. The conservation area is known as the Fisheating Creek Expanded Corridor. The settlement agreement also called for the Board of Trustees to lease the Fisheating Creek Expanded Corridor to the Commission, who the Board of Trustees designated as the managing agency. The settlement agreement acknowledges the public's "right to boat and canoe on Fisheating Creek throughout the entire Expanded Corridor.” With respect to navigation, the settlement agreement provides: Protection of Navigation. The navigability of Fisheating Creek throughout the entire Expanded Corridor shall be maintained and enhanced through a navigation maintenance program which includes aquatic weed control and removal of fallen logs and similar obstructions. This section does not authorize dredging. The Cookie-Cutter Project In January 2009, the Commission aerially applied an herbicide to kill the vegetation along the course of the Creek. In April 2010, the Commission contracted with A & L Aquatic Weed Control (“A & L”) to “[m]echanically dismantle floating tussocks.” The Commission directed A & L to perform the project by “shredding vegetation and accumulated organic material to re-open the navigation across Cowbone Marsh.” The Commission instructed A & L to re-open a channel "approximately 2.2 miles long and 18-20 feet wide,” and to clear some areas of the Creek “as wide as 35-feet wide occasionally as necessary to turn shredding equipment during the shredding process.” The Commission did not direct A & L to dredge a deeper channel. The vessel used by A & L to perform the work is known as a “cookie-cutter.” The cookie-cutter has two cutting wheels at the front of the vessel to shred and side-cast vegetation. The cutting wheels also act as propellers to propel the cookie- cutter forward. The cookie-cutter can clear woody vegetation up to four inches in diameter. The two cutting wheels can be lowered or raised in order to cut vegetation at various depths in the water. Evidence was presented to show how the cutting wheels could be lowered two to three feet, but it was not made clear whether the cutting wheels could be lowered even more. No evidence was presented to establish how deep the cookie-cutter blades were lowered into Fisheating Creek during the work performed by A & L. No evidence was presented to establish what depth of soil the cookie-cutter was capable of dredging through if the cutting wheels cut into the Creek bottom. The cookie-cutter began on the eastern side of Cowbone Marsh and moved upstream. The parties disputed the point of beginning. Petitioners contend it was farther upstream, but the more persuasive evidence for the point of beginning was presented by the Commission. The cookie-cutter generally followed the course of Fisheating Creek as depicted on the 1929 USACOE map. However, there are three areas where the cookie-cutter deviated from the 1929 map. One deviation is about 100 feet off-line. The other two deviations are 25 to 30 feet off-line. No explanation was given for the deviations, but the cookie-cutter operator generally followed the path of dead vegetation killed by the aerial spraying of herbicide and the line may have deviated from the true course of the Creek in these three areas. During the cookie-cutter project, water levels within the Creek and Marsh fluctuated. At some point, the project was postponed due to low water conditions. A sandbag dam was placed in the channel to artificially raise the water level so the cookie-cutter could continue. In July 2010, the Department and USACOE ordered the Commission to stop the project due to its adverse environmental impacts, including the draining of Cowbone Marsh. Before the cookie-cutter stopped, it had cleared about two miles of Fisheating Creek. Where the cookie-cutter stopped there is a discernible channel continuing west, but it is shallower and narrower than the channel created by the cookie-cutter. At this terminus, the cookie-cutter was dredging a deeper and wider channel than existed naturally. Additional evidence of dredging along the Creek channel is the soil cast up on the banks, and the removal of peat soils in the bottom of the Creek and exposure of underlying mineralized soil. The cookie-cutter altered the natural conditions of the Fisheating Creek in some areas by dredging the sides and bottom of the Creek. The dredging by the cookie-cutter altered the hydrology of the Creek and Marsh. The Marsh drained rapidly to Lake Okeechobee. In addition, large quantities of soil, muck, silt, and debris disturbed by the cookie-cutter were carried downstream toward Lake Okeechobee. Some of the soil and debris settled out at the mouth of the Creek, causing shoaling. The sides of the channel in many areas is continuing to erode. The Department’s Emergency Final Order In July 2010, the Department issued an Emergency Final Order, which directed the Commission to: (a) remove the cookie- cutter and immediately stop all activities associated with the cookie-cutter; (b) place temporary emergency flow restrictors in the channel to reduce flow velocities and minimize downstream sediment transport, as well as raise the water level to minimize surface and groundwater flow from the adjacent marsh into the channel; and (c) develop a long-term remedial plan to return water levels within the Marsh to pre-impact conditions and apply to the Department for an Environmental Resource Permit to implement the plan. In August 2010, pursuant to the Emergency Final Order, the Commission constructed an aluminum weir in the Creek to decrease flow velocities, reduce erosion, and maintain the hydration of the Marsh. The weir was placed approximately half a mile downstream from where the cookie-cutter stopped. During the wet season of 2010, the aluminum weir was completely submerged. Erosion and shoaling occurred immediately downstream. The Commission determined that the weir was ineffective and removed it. The EPA Compliance Orders In March 2011, the EPA issued an Administrative Compliance Order in which it alleged the Commission had engaged in "unauthorized activities associated with the excavation and construction of a channel within Cowbone Marsh.” The Commission was ordered to construct an initial check dam in the upper reaches of the Marsh to minimize the loss of groundwater and prevent further adverse impacts. In April 2011, EPA issued a second Administrative Compliance Order, directing the Commission to construct five additional check dams. The order describes the check dams as "initial corrective measures" and states that the “final restoration plan will include measures for backfilling the unauthorized cut through Cowbone Marsh.” The Initial Permits In May 2011, the Department issued to the Commission an Environmental Resource Permit and Sovereign Submerged Lands Authorization, which authorized the construction of six earthen check-dams within the portion of Fisheating Creek where the cookie-cutter had operated. The purpose of the check dams was to improve the hydrology of Cowbone Marsh and promote the accumulation of sediments within the channel to restore the natural depth and width of Fisheating Creek. The check dams were constructed using sand bags, marine plywood, coconut matting, and pressure-treated posts. The check dams have ten-foot wing walls which extend into the surrounding marsh. The wing walls are to prevent erosion around the dams and to direct water into the marsh. The installation of the check dams was completed in July 2011. Since that time, some repair efforts have been required to replace lost sandbags and to address erosion that has occurred around the check dams. The check dams have been somewhat successful in maintaining higher water levels in the Marsh. However, they have not restored natural hydrologic conditions, or prevented erosion along the channel. The Proposed Modification In June 2012, the Commission applied for a "minor modification" to the existing permits, which the Department granted. The modified permits authorize the Commission to backfill the channel cleared by the cookie-cutter with approximately 27,000 cubic yards of sand. The check dams would not be removed. The sand for the backfilling would be excavated from a "borrow" area located about a mile away. Petitioners contend that the borrow area is in wetlands, but the more persuasive evidence is that it is uplands. A 1.164-mile temporary access road would be constructed from the borrow area through uplands and wetlands to a 100-square-foot staging area adjacent to Fisheating Creek where the backfilling would begin. Wetland impacts would be minimized by constructing the temporary access road and staging area with interlocking mats. Petitioners did not show that the route or manner in which the temporary road would be constructed and used would have unacceptable adverse impacts to the environment or otherwise fail to comply with applicable criteria. The sand would be dumped into the Creek and then compacted. As the Creek was filled, the compacted sand would be used as a roadway for the trucks to transport sand to the end of the filled area to dump more sand, until the backfilling was completed. The proposed backfilling would not restore a typical stream profile, deepest in the middle and becoming more and more shallow moving toward the banks. That kind of profile can be seen in the photographs of Fisheating Creek taken before the cookie-cutter project. The proposed modification calls for filling the cut channel from "bank to bank": Final Grade: Fill must be compacted and ground surface elevations must be the same as the adjacent marsh ground surface elevations (within a tolerance of +6/-6 inches) The filled channel would be seeded and fertilized to grow native vegetation. The proposed seed mixture is mostly water grasses, but has some willow included. Compliance with Criteria Florida Administrative Code Rule 62-343.100 provides that a modification is treated as either minor or major depending on the magnitude of the changes and the potential for environmental impacts that differ from those addressed in the original permit: modification shall be considered to be minor only where the modification does not: Require a new site inspection by the Department in order to evaluate the request; or Substantially: Alter permit conditions; Increase the authorized discharge; Have substantially different or increased impacts on wetlands and other surface waters. . . ; Decrease the retention/detention specified by the original permit; Decrease any flood control elevations for roads or buildings specified by the original permit; or Increase the project area. At the final hearing, it was not shown how the modification meets the criteria for a minor modification. The proposed modification does not meet the criteria because it required new site visits, substantially alters the original permit conditions, and has a substantially different impact on wetlands. The criteria applicable to an application for a major modification were not identified, nor was it shown how the evidence presented at the final hearing satisfies the requirements for such an application. The proposed backfilling plan would not restore the natural conditions that existed in Fisheating Creek. The Commission did not show that it made a reasonable effort to determine the pre-disturbance conditions throughout the disturbed area. The proposed modification would not restore the natural depths in the Creek. The backfilling plan calls for a finished grade of plus or minus six inches above the level of the adjacent marsh. A final grade of zero to plus six inches would essentially eliminate Fisheating Creek. The maximum allowed depth of minus six inches below the level of the adjacent marsh would be shallower than the natural depths in portions of the Creek. Even the Department described the Creek was "one to two feet deep" before the cookie-cutter project. Adequate measures are not included in the permits to ensure that after backfilling and planting, the Creek would have the ordinary attributes of a creek. The proposed modification would not restore the pre- existing hydrologic conditions of the Creek. The modified Environmental Resource Permit requires strict compliance with the terms of the 1999 settlement agreement. The modification would not be consistent with the 1999 settlement agreement because the backfilling and planting would destroy the navigability of the Creek. Petitioners want to preserve the current depths of Fisheating Creek, but some of those depths are unnatural, being the result of dredging by the cookie-cutter. However, the proposed backfilling would not restore the natural depths in some parts of the Creek and would not maintain the navigability of the Creek, even for shallow draft vessels such as canoes and kayaks.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law it is RECOMMENDED that the Department deny the requested modification to the Commission's Environmental Resource Permit and Sovereignty Submerged Lands Authorization. DONE AND ENTERED this 3rd day of July, 2013, in Tallahassee, Leon County, Florida. S BRAM D. E. CANTER Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 3rd day of July, 2013. COPIES FURNISHED: W. Douglas Beason, Esquire Department of Environmental Protection Douglas Building, Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399-3000 Alisa A. Coe, Esquire Joshua D. Smith, Esquire Bradley I. B. Marshall, Esquire Earthjustice 111 South Martin Luther King, Jr., Boulevard Tallahassee, Florida 32301 Harold "Bud" Viehauer, General Counsel Ryan Osborne, Esquire Florida Fish and Wildlife Conservation Commission Bryant Building 620 South Meridian Street Tallahassee, Florida 32399-1050 Herschel T. Vinyard, Jr., Secretary Department of Environmental Protection Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399-3000 Matthew Z. Leopold, General Counsel Department of Environmental Protection Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399-3000 Lea Crandall, Agency Clerk Department of Environmental Protection Mail Station 35 3900 Commonwealth Boulevard Tallahassee, Florida 32399-3000
The Issue The issue in this case is whether, and under what conditions, the Respondent, St. Johns River Water Management District (District), should grant Environmental Resource Permit (ERP) No. 40-109-81153-1 authorizing Respondents, Jay and Linda Ginn (Ginns or Applicants), to construct a 136-unit single-family residential development with associated surface water management system.
Findings Of Fact The Parties and Proposed Project Respondent, the District, is a special taxing district created by Chapter 373, Florida Statutes, charged with the duty to prevent harm to the water resources of the District, and to administer and enforce the cited statutes and Florida Administrative Code Rules promulgated by the District under the authority of those statutes. (Unless otherwise stated, all Florida Statutes refer to the 2003 codification, and all Florida Administrative Code Rules refer to the current codification.) Respondents, Jay and Linda Ginn, are the owners of 47 acres of land located just west of the City of St. Augustine in St. Johns County, Florida. They are seeking ERP Permit No. 40- 109-81153-1 from the District to construct a 136-acre residential community and associated surface water management facilities on the property, to be known as Ravenswood Forest. The 47-acre project site is predominantly uplands, with a large (10.98-acre) wetland (Wetland 1) located on the eastern boundary and completely separating the uplands on the project site from adjacent properties to the east. While the central portion of the site is mostly a sand pine vegetated community, and the western portion is largely a pine flatwood community, there are six other smaller wetlands scattered within the upland areas lying west of Wetland 1, each numbered separately, 2 through 7. The site is currently undeveloped except for some cleared areas that are used as dirt road trails and a borrow pit or pond excavated in the central part of the site. This clearing and excavation was accomplished in the 1980’s for a project that was never completed. The project site is bordered on the north by Ravenswood Drive. On the east lies an existing residential development probably constructed in the 1970’s; to the west of the project site is a power-line easement; and to the south is a Time Warner cable facility. The land elevations at the project site are generally higher on the west and slope off to Wetland 1 on the east. Under current conditions, water generally drains from west to east into Wetland 1. Some water from the site, as well as some water entering the site from off-site properties to the west, flows into the existing pond or borrow pit located in the central portion of the site. Under extreme rainfall conditions, the borrow pit/pond can reach a stage that allows it to overflow and discharge into Wetland 1. Some off-site water also enters Wetland 1 at its north end. Water that originates from properties to the west of the Ravenswood site is conveyed through ditches to the roadside ditch that runs along the south side of Ravenswood Drive. Water in this roadside ditch ultimately enters Wetland 1 at its north end and flows south. Once in Wetland 1, water moves north to south. Water leaves the part of Wetland 1 that is located on the Ravenswood site and continues to flow south through ditches and culverts ultimately to the San Sebastian River. The Wetland 1 system is contiguous with wetlands located on property owned by Petitioner, Marilyn McMulkin. Mrs. McMulkin lives on Hibiscus Street to the east of the project. Mrs. McMulkin is disabled and enjoys observing wildlife from her home. Mrs. McMulkin has observed woodstorks, kites, deer, cardinals, birds, otter, indigo snake, flying squirrels, gopher tortoises, and (more recently) bald eagles on her property or around the neighborhood. Mrs. McMulkin informed the District of the presence of the bald eagle in 2002, but it was not discovered until November of 2003 that there was an eagle nest on the Ginns property in Wetland 1. Petitioner, Diane Mills, owns a house and property on Hibiscus Street to the east of the Project. The proposed stormwater discharge for the Project is to a wetland system that is contiguous with a wetland system that is in close proximity to Mrs. Mills' property. Petitioners' property is not located in a flood plain identified by FEMA. Nevertheless, Petitioners' property experiences flooding. At times, the flooding has come through Mrs. McMulkin's house and exited out the front door. The flood water, which can be 18-24 inches high in some places on Mrs. McMulkin's property, comes across her backyard, goes through or around her house, enters Hibiscus Street and turns north. The flooding started in the late 1980's and comes from the north and west, from the Ginns' property. The flooding started after Mr. Clyatt Powell, a previous co-owner of the Ravenswood property, started clearing and creating fill roads on the property using dirt excavated from the property. The flooding now occurs every year and has increased in duration and frequency; the flooding gets worse after the rain stops and hours pass. The evidence, including Petitioners' Exhibit 1, indicated that there are numerous other possible reasons, besides activities on the Ginns' property in the late 1980's, for the onset and exacerbation of Petitioners' flooding problems, including: failure to properly maintain existing drainage facilities; other development in the area; and failure to improve drainage facilities as development proceeds. The parties have stipulated that Petitioners have standing to object to ERP Permit No. 40-109-81153-1. Project Description As indicated, water that originates west of the project site currently enters the project site in two ways: (1) it moves across the western project boundary; and (2) it travels north to a ditch located on the south side of Ravenswood Drive and is conveyed to Wetland 1. The offsite water that moves across the western project boundary comes from a 16-acre area identified as Basin C (called Basin 4 post-development). The offsite water that moves north to the ditch and enters Wetland 1 comes from a 106.87-acre area identified as Basin D (called Basin 5 post-development). The project’s stormwater conveyance and treatment facilities include two connected wet detention ponds with an outfall to a wetland on the eastern portion of the project site. Stormwater from most of the project site will be conveyed to a pond, or detention area (DA) DA-1, which will be located near (and partially coinciding with the location of) the existing pond or borrow pit. The water elevation in DA-1 will be controlled at a level of 26 feet. Water from DA-1 will spill over through a control structure into a pipe that will convey the spill-over to DA-2. In addition to the spill-over from DA-1, offsite water that currently enters the project site across the western boundary will be conveyed to a wetland area at the southwest corner of the project site. At that point, some of the water will be taken into DA-2 through an inlet structure. The water elevation in DA-2 will be controlled at level 21. Water from DA-2 will be released by a control structure to a spreader swale in Wetland 1. While some of the water conveyed to the wetland area at the southwest corner of the project site will enter DA-2, as described, some will discharge over an irregular weir (a low area that holds water until it stages up and flows out) and move around the southern boundary of the project site and flow east into Wetland 1. Wetland 1 is a 10.98-acre onsite portion of a larger offsite wetland area extending to the south and east (which includes the wetlands on Mrs. McMulkin's property). For purposes of an Overall Watershed Study performed by the Ginns' engineering consultant, the combined onsite and offsite wetlands was designated Node 98 (pre-development) and Node 99 (post- development). From those areas, water drains south to ditches and culverts and eventually to the San Sebastian River. Best management practices will be used during project construction to address erosion and sediment control. Such measures will include silt fences around the construction site, hay bales in ditches and inlets, and maintenance of construction equipment to prevent release of pollutants, and may include staked sod on banks and turbidity barriers, if needed. In addition, the District's TSR imposed permit conditions that require erosion and sediment control measures to be implemented. The District's TSR also imposed a permit condition that requires District approval of a dewatering plan within 30 days of permit issuance and prior to construction. The Ginns intend to retain the dewatering from construction on the project site. Wetland Impacts Onsite Wetlands Wetland 1 is a 10.98-acre mixed-forested wetland system. Its overall condition is good. It has a variety of vegetative strata, a mature canopy, dense understory and groundcover, open water areas, and permanent water of varying levels over the course of a year. These attributes allow for species diversity. Although surrounded by development, the wetland is a good source for a variety of species to forage, breed, nest, and roost. In terms of vegetation, the wetland is not unique to northeast Florida, but in November 2003 an eagle nest was discovered in it. A second wetland area onsite (Wetland 2) is a 0.29-acre coniferous depression located near the western boundary of the site. The overall value of the functions provided by Wetland 2 is minimal or low. It has a fairly sparse pine canopy and scattered ferns provide for little refuge and nesting. Water does stand in it, but not for extended periods of time, which does not allow for breeding of most amphibians. The vegetation and inundation do not foster lower trophic animals. For that reason, although the semi-open canopy would be conducive to use by woodstorks, birds and small mammals do not forage there. A third wetland area onsite (Wetland 3) is a 0.28-acre mixed-forested wetland on the northern portion of the site. The quality of Wetland 3 is low. A 24-inch culvert drains the area into a 600-foot long drainage ditch along the south side of Ravenswood Drive leading to Wetland 1. As a result, its hydroperiod is reduced and, although it has a healthy pine and cypress canopy, it also has invasive Chinese tallow and upland species, along with some maple. The mature canopy and its proximity to Ravenswood Drive would allow for nesting, but no use of the wetland by listed species has been observed. In order to return Wetland 3 to being productive, its hydroperiod would have to be restored by eliminating the connection to the Ravenswood Drive ditch. A fourth wetland area onsite (Wetland 4) is a 0.01- acre portion of a mixed-forested wetland on the western boundary of the site that extends offsite to the west. Its value is poor because: a power line easement runs through it; it has been used as a trail road, so it is void of vegetation; and it is such a small fringe of an offsite wetland that it does not provide much habitat value. A fifth wetland area onsite (Wetland 5) is a 0.01-acre portion of the same offsite mixed-forested wetland that Wetland 4 is part of. Wetland 5 has a cleared trail road through its upland fringe. Wetland 5 has moderate value. It is vegetated except on its upland side (although its vegetation is not unique to northeast Florida), has a nice canopy, and provides fish and wildlife value (although not as much as the interior of the offsite wetland). A sixth wetland area onsite (Wetland 6) is a 0.28-acre wetland located in the western portion of the site. It is a depression with a coniferous-dominated canopy with some bays and a sparse understory of ferns and cord grass that is of moderate value overall. It does not connect with any other wetlands by standing or flowing water and is not unique. It has water in it sufficient to allow breeding, so there would be foraging in it. Although not discovered by the Ginns' consultants initially, a great blue heron has been observed utilizing the wetland. No listed species have been observed using it. Wetland 6 could be good gopher frog habitat due to its isolation near uplands and its intermittent inundation, limiting predation by fish. In addition, four gopher tortoise burrows have been identified in uplands on the project site, and gopher frogs use gopher tortoise burrows. The gopher frog is not a listed species; the gopher tortoise is listed by the State of Florida as a species of special concern but is not aquatic or wetland-dependent. Woodstorks are listed as endangered. Although no woodstorks were observed using Wetland 6, they rely on isolated wetlands drying down to concentrate fish and prey in the isolated wetlands. With its semi-open canopy, Wetland 6 could be used by woodstorks, which have a wingspan similar to great blue herons, which were seen using Wetland 6. However, Wetland would not provide a significant food source for wading birds such as woodstorks. The other surface water area onsite (Wetland 7) is the existing 0.97-acre pond or borrow pit in the southwest portion of the project site. The pond is man-made with a narrow littoral shelf dominated by torpedo grass; levels appears to fluctuate as groundwater does; and it is not unique. It connects to Wetland 1 during seasonal high water. It has some fish, but the steep slope to its littoral shelf minimizes the shelf's value for fish, tadpoles, and larvae stage for amphibians because fish can forage easily on the shelf. The Ginns propose to fill Wetlands 2, 3, 4, and 6; to not impact Wetland 5; and to fill a 0.45-acre portion of Wetland and dredge the remaining part into DA-1. Also, 0.18 acre of Wetland 1 (0.03 acre is offsite) will be temporarily disturbed during installation of the utility lines to provide service to the project. Individually and cumulatively, the wetlands that are less than 0.5-acre--Wetlands 3, 6, 2, 4, and 5--are low quality and not more than minimal value to fish and wildlife except for Wetland 5, because it is a viable part of an offsite wetland with value. While the Ginns have sought a permit to fill Wetland 4, they actually do not intend to fill it. Instead, they will simply treat the wetland as filled for the purpose of avoiding a County requirement of providing a wetland buffer and setback, which would inhibit the development of three lots. Offsite Wetlands The proposed project would not be expected to have an impact on offsite wetlands. Neither DA-1 nor DA-2, especially with the special conditions imposed by the District, will draw down offsite wetlands. The seasonal high water (SHW) table in the area of DA- 1 is estimated at elevation 26 to 29. With a SHW table of 26, DA-1 will not influence groundwater. Even with a SHW table of 29, DA-1 will not influence the groundwater beyond the project's western boundary. DA-1 will not adversely affect offsite wetlands. A MODFLOW model was run to demonstrate the influence of DA-1 on nearby wetlands assuming that DA-1 would be controlled at elevation 21, that the groundwater elevation was 29, and that no cutoff wall or liner would be present. The model results demonstrated that the influence of DA-1 on groundwater would barely extend offsite. The current proposed elevation for DA-1 is 26, which is higher than the elevation used in the model and which would result in less influence on groundwater. The seasonal high water table in the area of DA-2 is 28.5 to 29.5. A cutoff wall is proposed to be installed around the western portion of DA-2 to prevent it from drawing down the water levels in the adjacent wetlands such that the wetlands would be adversely affected. The vertical cutoff wall will be constructed of clay and will extend from the land surface down to an existing horizontal layer of relatively impermeable soil called hardpan. The cutoff wall tied into the hardpan would act as a barrier to vertical and horizontal groundwater flow, essentially severing the flow. A MODFLOW model demonstrated that DA-2 with the cutoff wall will not draw down the adjacent wetlands. The blow counts shown on the boring logs and the permeability rates of soils at the proposed location of DA-2 indicate the presence of hardpan. The hardpan is present in the area of DA-2 at approximately 10 to 15 feet below the land surface. The thickness of the hardpan layer is at least 5 feet. The Ginns measured the permeability of hardpan in various locations on the project site. The cutoff wall design is based on tying into a hardpan layer with a permeability of 0.052 feet per day. Because permeability may vary across the project site, the District recommended a permit condition that would require a professional engineer to test for the presence and permeability of the hardpan along the length of the cutoff wall. If the hardpan is not continuous, or if its permeability is higher than 0.052 feet per day, then a liner will be required to be installed instead of a cutoff wall. The liner would be installed under the western third of DA-2, west of a north-south line connecting the easterly ends of the cutoff wall. (The location of the liner is indicated in yellow on Applicants' Exhibit 5B, sheet 8, and is described in District Exhibit 10.) The liner would be 2 feet thick and constructed of clay with a permeability of no more than 1 x 10-6 centimeters per second. A liner on a portion of the bottom of pond DA-2 will horizontally sever a portion of the pond bottom from the groundwater to negate the influence of DA-2 on groundwater in the area. A clay liner would function to prevent adverse drawdown impacts to adjacent wetlands. The project, with either a cutoff wall or a clay liner, will not result in a drawdown of the groundwater table such that adjacent wetlands would be adversely affected. Reduction and Elimination of Impacts The Ginns evaluated practicable design alternatives for eliminating the temporary impact to 0.18-acre of Wetland 1. The analysis indicated that routing the proposed utility services around the project site was possible but would require a lift station that would cost approximately $80,000 to $100,000. The impact avoided is a temporary impact; it is likely that the area to be impacted can be successfully reestablished and restored; and preservation of Wetland 1 is proposed to address lag-time for reestablishment. It was determined by the Ginns and District staff that the costs of avoidance outweigh the environmental benefits of avoidance. Petitioners put on evidence to question the validity of the Wetland 1 reduction/elimination analysis. First, Mr. Mills, who has experience installing sewer/water pipes, testified to his belief that a lift station would cost only approximately $50,000 to $60,000. He also pointed out that using a lift station and forced main method would make it approximately a third less expensive per linear foot to install the pipe line itself. This is because a gravity sewer, which would be required if a lift station and forced main is not used, must be laid at precise grades, making it is more difficult and costly to lay. However, Mr. Mills acknowledged that, due to the relatively narrow width of the right-of-way along Ravenswood Drive, it would be necessary to obtain a waiver of the usual requirement to separate the sewer and water lines by at least 10 feet. He thought that a five-foot separation waiver would be possible for his proposed alternative route if the "horizontal" separation was at least 18 inches. (It is not clear what Mr. Mills meant by "horizontal.") In addition, he did not analyze how the per-linear-foot cost savings from use of the lift station and forced main sewer would compare to the additional cost of the lift station, even if it is just $50,000 to $60,000, as he thinks. However, it would appear that his proposed alternative route is approximately three times as long as the route proposed by the Ginns, so that the total cost of laying the sewer pipeline itself would be approximately equal under either proposal. Mr. Mills's testimony also suggested that the Ginns did not account for the possible disturbance to the Ravenswood eagles if an emergency repair to the water/sewer is necessary during nesting season. While this is a possibility, it is speculative. There is no reason to think such emergency repairs will be necessary, at least during the approximately 20-year life expectancy of the water/sewer line. Practicable design modifications to avoid filling Wetland 4 also were evaluated. Not filling Wetland 4 would trigger St. Johns County wetland setback requirements that would eliminate three building lots, at a cost of $4,684 per lot. Meanwhile, the impacted wetland is small and of poor quality, and the filling of Wetland 4 can be offset by proposed mitigation. As a result, the costs of avoidance outweigh the environmental benefits of avoidance. Relying on ERP-A.H. 12.2.2.1 the Ginns did not perform reduction/elimination analyses for Wetlands 2 and 6, and the District did not require them. As explained in testimony, the District interprets ERP-A.H. 12.2.1.1 to require a reduction/elimination analysis only when a project will result in adverse impacts such that it does not meet the requirements of ERP-A.H. 12.2.2 through 12.2.3.7 and 12.2.5 through 12.3.8. But ERP-A.H. 12.2.2.1 does not require compliance with those sections for regulated activities in isolated wetlands less than one-half acre in size except in circumstances not applicable to this case: if they are used by threatened or endangered species; if they are located in an area of critical state concern; if they are connected at seasonal high water level to other wetlands; and if they are "more than minimal value," singularly or cumulatively, to fish and wildlife. See ERP-A.H. 12.2.2.1(a) through (d). Under the District's interpretation of ERP-A.H. 12.2.1.1, since ERP-A.H. 12.2.2.1 does not require compliance with the very sections that determine whether a reduction/elimination analysis is necessary under ERP-A.H. 12.2.1.1, such an analysis is not required for Wetlands 2 and 6. Relying on ERP-A.H. 12.2.1.2, a., the Ginns did not perform reduction/elimination analyses for Wetlands 3 and 7, and the District did not require them, because the functions provided by Wetlands 3 and 7 are "low" and the proposed mitigation to offset the impacts to these wetlands provides greater long-term value. Petitioners' environmental expert opined that an reduction/elimination analysis should have been performed for all of the wetlands on the project site, even if isolated and less than half an acre size, because all of the wetlands on the project site have ecological value. For example, small and isolated wetlands can be have value for amphibians, including the gopher frog. But his position does not square with the ERP- A.H., as reasonably interpreted by the District. Specifically, the tests are "more than minimal value" under ERP-A.H. 12.2.2.1(d) and "low value" under ERP-A.H. 12.2.1.2, a. Secondary Impacts The impacts to the wetlands and other surface waters are not expected to result in adverse secondary impacts to the water resources, including endangered or threatened listed species or their habitats. In accordance with ERP-A.H. 12.2.7(a), the design incorporates upland preserved buffers with minimum widths of 15 feet and an average width of 25 feet around the wetlands that will not be impacted. Sediment and erosion control measures will assure that the construction will not have an adverse secondary impact on water quality. The proposed development will be served by central water and sewer provided by the City of St. Augustine, eliminating a potential for secondary impacts to water quality from residential septic tanks or septic drainfields. In order to provide additional measures to avoid secondary impacts to Wetland 1, which is the location of the bald eagles’ nest, the Applicants proposed additional protections in a Bald Eagle Management Plan (BEMP) (App. Ex. 14). Under the terms of the BEMP, all land clearing, infrastructure installation, and exterior construction on homes located within in the primary zone (a distance within 750 feet of the nest tree) is restricted to the non-nesting season (generally May 15 through September 30). In the secondary zone (area between 750 feet and 1500 feet from the nest tree), exterior construction, infrastructure installation, and land clearing may take place during the nesting season with appropriate monitoring as described in the BEMP. Proposed Mitigation The Ginns have proposed mitigation for the purpose of offsetting adverse impacts to wetland functions. They have proposed to provide mitigation for: the 0.18-acre temporary impact to Wetland 1 during installation of a water/sewer line extending from existing City of St. Augustine service to the east (at Theodore Street); the impacts to Wetlands 3, 4 and 7; and the secondary impacts to the offsite portion of Wetland 4. The Ginns propose to grade the 0.18-acre temporary impact area in Wetland 1 to pre-construction elevations, plant 72 trees, and monitor annually for 5 years to document success. Although the easement is 30 feet in width, work will be confined to 20 feet where vegetation will be cleared, the top 1 foot of soil removed and stored for replacing, the trench excavated, the utility lines installed, the trench refilled, the top foot replaced, the area replanted with native vegetation, and re- vegetation monitored. To facilitate success, the historic water regime and historic seed source will give the re-vegetation effort a jump-start. The Ginns propose to restore and enhance a 0.12-acre portion of Wetland 1 that has been degraded by a trail road. They will grade the area to match the elevations of adjacent wetland, plant 48 trees, and monitor annually for 5 years to document success. This is proposed to offset the impacts to Wetland 4. The proposed grading, replanting, and monitoring will allow the area to be enhanced causing an environmental benefit. The Ginns propose to preserve 10.58 acres of wetlands and 3.99 acres of uplands in Wetland 1, 1 acre of upland buffers adjacent to Wetlands 1 and 5, and the 0.01 acre wetland in Wetland 5. The upland buffer will be a minimum of 15 feet wide with an average of 25 feet wide for Wetland 1 and 25 feet wide for Wetland 5. A conservation easement will be conveyed to the District to preserve Wetlands 1 and 5, the upland buffers, and the wetland restoration and enhancement areas. The preservation of wetlands provides mitigation value because it provides perpetual protection by ensuring that development will not occur in those areas, as well as preventing activities that are unregulated from occurring there. This will allow the conserved lands to mature and provide more forage and habitat for the wildlife that would utilize those areas. Mitigation for Wetlands 2 and 6 was not provided because they are isolated wetlands less than 0.5-acre in size that are not used by threatened or endangered species; are not located in an area of critical state concern; are not connected at seasonal high water level to other wetlands; and are not more than minimal value, singularly or cumulatively, to fish and wildlife. As previously referenced in the explanation of why no reduction/elimination analysis was required for these wetlands, ERP-A.H. 12.2.2.1(d) does not require compliance with under ERP- A.H. 12.3 through 12.3.8 (mitigation requirements) for regulated activities in isolated wetlands less than one-half acre in size except in circumstances found not to be present in this case. See Finding 44, supra. The cost of the proposed mitigation will be approximately $15,000. Operation and Maintenance A non-profit corporation that is a homeowners association (HOA) will be responsible for the operation, maintenance, and repair of the surface water management system. An HOA is a typical operation and maintenance entity for a subdivision and is an acceptable entity under District rules. See ERP-A.H. 7.1.1(e) and 7.1.2; Fla. Admin. Code R. 40C- 42.027(3) and (4). The Articles of Incorporation for the HOA and the Declaration of Covenants, Conditions, and Restrictions contain the language required by District rules. Water Quantity To address water quantity criteria, the Applicants' engineers ran a model (AdICPR, Version 1.4) to compare the peak rate discharge from the project in the pre-project state versus the peak rate discharge after the project is put in place. The pre-project data input into the model were defined by those conditions that existed in 1985 or 1986, prior to the partial work that was conducted, but not completed, on the site in the late 1980's. The project’s 1985/1986 site condition included a feature called Depression A that attenuated some onsite as well as offsite stormwater. Because of work that was done on the project site after 1985/1986 (i.e., the excavation of the borrow pit and road-clearing activities in the late 1980's), the peak rate of discharge for the 1985/1986 project site condition was lower than the peak rate of discharge for today’s project site condition. (Flooding at Mrs. McMulkin's house began after the work was performed on the project site in the late 1980's.) Because this partial work conducted in the late 1980's increased peak rate discharge from the site, by taking the pre-project conditions back to the time prior to that work, the peak rate of discharge in the 1985-86 pre-project condition was lower than it would be under today's conditions. The model results indicated that for the 25-year, 24- hour storm event, the pre-project peak rate discharge is 61.44 cubic feet per second (cfs). The post-project peak rate discharge is 28.16 cfs. Because the completed project reduces the pre-project peak rate discharges, the project will not cause any adverse flooding impacts off the property downstream. A similar analysis of the peak rate discharges under pre-project conditions that exist today (rather than in 1986) was compared to peak rate discharges for the post-project conditions. This analysis also showed post-project peak rate discharges to be less than the peak rate discharges from the site using today’s conditions as pre-project conditions. As further support to demonstrate that the project would not cause additional flooding downstream, a second modeling analysis was conducted, which is referred to as the Ravenswood Overall Watershed Model (OWM). The Applicants' engineer identified water flowing into the system from the entire watershed basin, including the project site under both the pre- and post-project conditions. The water regime was evaluated to determine what effect the proposed project will have on the overall peak rate discharges, the overall staging, and the duration of the staging within the basin that ultimately receives the water from the overall watershed. This receiving basin area was defined as the "wetland node" (Node 98 pre- project, and Node 99 post-project). As previously stated, the area within this "wetland node" includes more than just the portion of Wetland 1 that is located on the Ravenswood site. It also includes the areas to the south and east of the on-site Wetland 1 (including properties owned by the Petitioners) and extends down to an east-west ditch located just north of Josiah Street. The project’s surface water management system will not discharge to a landlocked basin. The project is not located in a floodway or floodplain. The project is not located downstream of a point on a watercourse where the drainage is five square miles or more. The project is impounding water only for temporary storage purposes. Based on testimony from their experts, Petitioners contend that reasonable assurances have not been given as to water quantity criteria due to various alleged problems regarding the modeling performed by the Ginns' engineer. Tailwater Elevations First, they raise what they call "the tailwater problem." According to Petitioners, the Ginns' modeling was flawed because it did not use a 19.27-foot SHW elevation in Wetland 1 as the tailwater elevation. The 19.27-foot SHW was identified by the Ginns' biologist in the Wetland 1 near the location of the proposed utility line crossing the wetland and was used as the pre-development tailwater in the analysis of the project site. The post-development tailwater condition was different because constructing the project would change the discharge point, and "tailwater" refers to the water elevation at the final discharge of the stormwater management system. (SW- A.H., Section 9.7) The post-development tailwater was 21 feet, which reflects the elevation of the top of the spreader swale that will be constructed, and it rose to 21.3 feet at peak flow over that berm. For the OWM, the final discharge point of the system being modeled was the east-west ditch located just north of Josiah Street, where the tailwater elevation was approximately 18.1 feet, not the 19.27 feet SHW mark to the north in Wetland 1. The tailwater condition used in the modeling was correct. Petitioners also mention in their PRO that "the Applicants' analysis shows that, at certain times after the 25 year, 24 hour storm event, in the post development state, Wetland 1 will have higher staging than in the predevelopment state." But those stages are after peak flows have occurred and are below flood stages. This is not an expected result of post- development peak-flow attenuation. Watershed Criticism The second major criticism Petitioners level at the Applicants' modeling is that parts of the applicable watershed basins were omitted. These include basins to the west of the project site, as well as basins to the north of the site, which Petitioners lumped into the so-called "tailwater problem." Petitioners sought to show that the basins identified by the Ginns as draining onto the project site from the west were undersized, thus underestimating the amount of offsite water flowing onto the project site. With respect to Basin C, Petitioners' witness testified that the basin should be 60 acres instead of 30 acres in size, and that consequently more water would flow into pond DA-2 and thus reduce the residence time of the permanent pool volume. In fact, Basin C is 16 acres in size, not 30 acres. The water from Basin C moves onto the project site over the western project boundary. A portion of the water from Basin C will be directed to pond DA-2 through an inlet structure, and the rest will move over an irregular weir and around the project site. With respect to Basin D, Petitioners' witness testified that the basin should encompass an additional 20 acres to the west and north. West of Basin D, there are ditches routing water flow away from the watershed, so it is unclear how water from an additional 20 acres would enter the watershed. The western boundary of the OWM is consistent with the western boundaries delineated in two studies performed for St. Johns County. Petitioners' witness testified that all of the water from the western offsite basins currently travels across the project site's western boundary, and that in post-development all of that water will enter pond DA-2 through the inlet structure. In fact, currently only the water from Basin C flows across the project site's western boundary. Post-development, only a portion of water from Basin C will enter pond DA-2. Currently and post-development, the water in Basin D travels north to a ditch south of Ravenswood Drive and discharges into Wetland 1. Petitioners also sought to show that a 50-acre area north of the project site should have been included in the OWM. Petitioners' witness testified that there is a "strong possibility" that the northern area drains into the project site by means of overtopping Ravenswood Drive. The witness' estimate of 50 acres was based on review of topographical maps; the witness has not seen water flowing over Ravenswood Drive. The Ginns' engineer testified that the area north of Ravenswood Drive does not enter the project site, based on his review of two reports prepared by different engineering firms for St. Johns County, conversations with one of those engineering firms, conversations with the St. Johns County engineer, reviews of aerials and contour maps, and site observations. Based on site observations, the area north of the project site drains north and then east. One report prepared for St. Johns County did not include the northern area in the watershed, and the other report included an area to the north consisting of 12 acres. The Ginns' engineer added the 12-acre area to the OWM and assumed the existence of an unobstructed culvert through which this additional water could enter Wetland 1, but the model results showed no effect of the project on stages or duration in the wetland. Even if a 50-acre area were included in the OWM, the result would be an increase in both pre-development and post- development peak rates of discharge. So long as the post- development peak rate of discharge is lower than the pre- development peak rate of discharge, then the conveyance system downstream will experience a rate of water flow that is the same or lower than before the project, and the project will not cause adverse flooding impacts offsite. Petitioners' witness did not have any documents to support his version of the delineations of Basins C and D and the area north of Ravenswood Drive. Time of Concentration Time of concentration (TC) is the time that it takes a drop of water to travel from the hydraulically most distant point in a watershed. Petitioners sought to show that the TC used for Basin C was incorrect. Part of Petitioners' rationale is related to their criticism of the watersheds used in the Ginns' modeling. Petitioners' witness testified that the TC was too low because the distance traveled in Basin C should be longer because Basin C should be larger. The appropriateness of the Basin C delineation already has been addressed. See Finding 71, supra. Petitioners' witness also testified that the TC used for the post-development analysis was too high because water will travel faster after development. However, the project will not develop Basins C and D, and thus using the same TC in pre- development and post-development is appropriate. The project will develop Basins A and B (called Basins 1, 2, and 3 post- development), and the post-development TC for those basins were, in fact, lower than those used in the pre-development analysis. Groundwater Infiltration in DA-2 One witness for Petitioners opined that groundwater would move up through the bottom of DA-2 as a result of upwelling (also referred to as infiltration or seepage), such that 1,941 gallons per day (gpd) would enter DA-2. That witness agreed that if a liner were installed in a portion of DA-2, the liner would reduce upwelling in a portion of the pond. Another witness for Petitioners opined that 200 gpd of groundwater would enter the eastern part and 20,000 gpd would enter the western part of DA-2. Although that witness stated that upwelling of 200 gpd is not a significant input and that upwelling of 20,000 gpd is a significant input, he had not performed calculations to determine the significance. Even if more than 20,000 gpd of groundwater entered DA-2, DA-2 will provide sufficient permanent pool residence time without any change to the currently designed permanent pool size or the orifice size. Although part of one system, even if DA-2 is considered separate from DA-1, DA-2 is designed to provide an additional permanent pool volume of 6.57 acre-feet (in addition to the 20.5 acre/feet provided by DA-1). This 6.57 acre-feet provided by DA-2, is more than the 4.889 acre-feet of permanent pool volume that would be necessary to achieve a 21-day residence time for the 24+ acres that discharge directly into DA-2, as well as background seepage into DA-2 at a rate of 0.0403 cfs, which is more upwelling than estimated by Petitioners' two witnesses. There is adequate permanent pool volume in DA-2 to accommodate the entire flow from Basin C and for water entering through the pond bottom and pond sides and provide at least 21 days of residence time. Water Quality Criteria Presumptive Water Quality The stormwater system proposed by the Ginns is designed in accordance with Florida Administrative Code Rules 40C-42.024, 40C-42.025, and 40C-42.026(4). Wet detention ponds must be designed for a permanent pool residence time of 14 days with a littoral zone, or for a residence time of 21 days without a littoral zone, which is the case for this project. See Fla. Admin. Code R. 40C-42.026(4)(c) and (d). DA-1 and DA-2 contain sufficient permanent pool volume to provide a residence time of 31.5 days, which is the amount of time required for projects that discharge to Class II Outstanding Florida Waters, even though the receiving waterbody for this project is classified as Class III Waters. See Fla. Admin. Code R. 40C-42.026(4)(k)1. Best management practices will be used during project construction to address erosion and sediment control. Such measures will include silt fences around the construction site, hay bales in ditches and inlets, and maintenance of construction equipment to prevent release of pollutants, and may include staked sod on banks and turbidity barriers if needed. In addition, the District proposed permit conditions that require erosion and sediment control measures to be implemented. (Dist. Ex. 1, pp. 8-9, #4; Dist. Ex. 2, p. 1, ##3, 4, and 5, and p. 6, #10). ERP/MSSW/Stormwater Special Conditions incorporated into the proposed permit require that all wetland areas or water bodies outside the specific limits of construction must be protected from erosion, siltation, scouring or excess turbidity, and dewatering. (Dist. Ex. 2). The District also proposed a permit condition that requires District approval of a dewatering plan for construction, including DA-1 and DA-2, within 30 days of permit issuance and prior to construction. The Ginns intend to retain the dewatering from construction on the project site. As previously described, Petitioners' engineering witness sought to show that DA-2 will not provide the required permanent pool residence time because Basin C should be 60 acres in size. Petitioners' environmental witness also expressed concern about the capacity of the ponds to provide the water quality treatment required to meet the presumptive water quality criteria in the rules, but those concerns were based on information he obtained from Petitioners' engineering witness. Those issues already have been addressed. See Findings 77-78, supra. Groundwater Contamination Besides those issues, Petitioners raised the issue that groundwater contamination from a former landfill nearby and from some onsite sludge and trash disposal could be drawn into the proposed stormwater management system and cause water quality violations in the receiving waters. If groundwater is contaminated, the surface water management system could allow groundwater to become surface water in proposed DA-1. St. Johns County operated a landfill from the mid-1950s to 1977 in an area northwest of the project site. The landfill accepted household and industrial waste, which was buried in groundwater, which in turn could greatly enhance the creation of leachate and impacted water. Groundwater flows from west to east in the vicinity of the landfill and the project site but there was conflicting evidence as to a minor portion of the property. The Ginns' witness testified that if the landfill extended far enough south, a small part of the project site could be downgradient from the landfill. But there was no evidence that the landfill extended that far south. Petitioners' witness testified that the groundwater flow varies on the south side of the landfill so that groundwater might flow southeast toward the site. Even if Petitioners' witness is correct, the surface water management system was designed, as Petitioners' other witness agreed, so that DA-1 would have minimal influence on groundwater near the pond. In 1989, sewage sludge and garbage were placed in a pit in the central part of the project site, north of the existing pond, which also is the area for proposed DA-1; and at various times refuse--including a couple of batteries, a few sealed buckets, and concrete--has been placed on the surface of the site. In 1989, to determine the amount of sewage and garbage on the project site, the St. Johns County Health Department chose several locations evidencing recent excavation south of Ravenswood Drive, had the areas re-excavated, and found one bag of garbage and debris such as tree stumps and palmettos. In 2001, an empty 55-gallon drum was on the site; there was no evidence what it once contained or what it contained when deposited onsite, if anything. In addition, trespassers dumped solid waste on the property from time to time. Petitioners' witness searched the site with a magnetometer and found nothing significant. On the same day, another of Petitioners’ witnesses sampled with an auger but the auger did not bore for core or any other type sample; it merely measured groundwater level. In 1985, 1999, and 2000, groundwater offsite of the project near the landfill was sampled at various times and places by various consultants to determine whether groundwater was being contaminated by the landfill. The groundwater sampling did not detect any violations of water quality standards. Consultants for the Ginns twice sampled groundwater beneath the project site and also modeled contaminant migration. The first time, in 2001, they used three wells to sample the site in the northwest for potential impacts to the property from the landfill. The second time, they sampled the site through cluster wells in the northwest, middle, and south. (Each cluster well samples in a shallow and in a deeper location.) The well locations were closest to the offsite landfill and within an area where refuse may have been buried in the north- central part of the site. Due to natural processes since 1989, no sewage sludge deposited onsite then would be expected to remain on the surface or be found in the groundwater. The evidence was that the sewage sludge and garbage were excavated. Although samples taken near the center of the property contained substances that are water quality parameters, they were not found in sufficient concentration to be water quality violations. There is an iron stain in the sand north of the existing pond in the area where pond DA-1 is to be located. Based on dissolved oxygen levels in the groundwater, Petitioners' witness suggested that the stain is due to buried sewage, but the oxygen levels are not in violation of water quality standards and, while toward the low end of not being a violation, the levels could be due to natural causes. No evidence was presented establishing that the presence of the iron stain will lead to a violation of water quality standards. Petitioners' witness, Mr. Boyes, testified that iron was a health concern. But iron itself is a secondary drinking water standard, which is not a health-based standard but pertains to odor and appearance of drinking water. See § 403.852(12) and (13), Fla. Stat. Petitioners argued that the Phase I study was defective because historical activity on the project site was not adequately addressed. But the Phase I study was only part of the evidence considered during this de novo hearing. Following up on the Phase I study, the 2001 sampling analyzed for 68 volatile organics and 72 semi-volatile organics, which would have picked up solvents, some pesticides, petroleum hydrocarbons, and polynuclear aromatic hydrocarbons--the full range of semi-volatile and volatile organics. The sampling in August 2003 occurred because some of the semi-volatile parameters sampled earlier needed to be more precisely measured, and it was a much broader analysis that included 63 semi-volatiles, 73 volatile organic compounds, 23 polynuclear aromatic hydrocarbons, 25 organic phosphate pesticides, 13 chlorinated herbicides, 13 metals, and ammonia and phosphorus. The parameters for which sampling and analyses were done included parameters that were representative of contaminants in landfills that would have now spread to the project site. They also would have detected any contamination due to historical activity on the project site. Yet groundwater testing demonstrated that existing groundwater at the project site meets state water quality standards. Based on the lack of contaminants found in these samples taken from groundwater at the project site 50 years after the landfill began operation, the logical conclusion is that either groundwater does not flow from the landfill toward the project site or that the groundwater moving away from the landfill is not contaminated. Groundwater that may enter the stormwater ponds will not contain contaminants that will exceed surface water quality standards or groundwater quality standards. Taken together, the evidence was adequate to give reasonable assurances that groundwater entering the stormwater ponds will not contain contaminants that exceed surface water quality standards or groundwater quality standards and that water quality violations would not occur from contaminated water groundwater drawn into the proposed stormwater management system, whether from the old landfill or from onsite waste disposal. The greater weight of the evidence was that there are no violations of water quality standards in groundwater beneath the project site and that nothing has happened on the site that would cause violations to occur in the future. Contrary to Petitioners' suggestion, a permit condition requiring continued monitoring for onsite contamination is not warranted. J. Fish and Wildlife Except for the bald eagle nest, all issues regarding fish and wildlife, listed species, and their habitat as they relate to ERP-A.H. 12.2.2 through 12.2.2.4 already have been addressed. When the Ginns were made aware in November 2003 that there was an eagle nest in Wetland 1, they retained the services of Tony Steffer, an eagle expert with over 25 years of experience working specifically with eagles and eagle management issues, including extensive hands-on experience with eagles and the conduct of field studies, aerial surveys, and behavioral observations as well as numerous research projects on the bald eagle. Mr. Steffer visited the Ravenswood site on numerous occasions since the discovery of the nest, made observations, and was integral in the drafting of the Ravenswood BEMP. It is Mr. Steffer’s opinion that the proposed project, with the implementation of the BEMP, will not adversely affect the eagles. This opinion was based on Mr. Steffer's extensive knowledge and experience with eagle behavior and human interactions. In addition, Mr. Steffer considered the physical characteristics of the Ravenswood site and the nest tree, the dense vegetation in Wetland 1 surrounding the nest site, and the existing surrounding land uses, including the existing residential community that lies a distance of about 310 feet from the nest site, the existing roadways and associated traffic, and the school (with attendant playground noise) that is to north of the site. In Mr. Steffer's opinion, the eagles are deriving their security from the buffering effects provided by the surrounding wetland. He observed that the nesting and incubating eagles were not disturbed when he set up his scope at about 300-320 feet from the tree. The BEMP requires that Wetland 1, and the upland islands located within it, be preserved and limits the work associated with the water/sewer line to the non-nesting season. With the BEMP implemented, Mr. Steffer expressed confidence that the Ravenswood eagles would be able to tolerate the proposed activities allowed under the BEMP. The Ravenswood project plans and the BEMP were reviewed by the U.S. Fish and Wildlife Service (USFWS). The USFWS analyzed information in their files relating to projects which proposed activities within the primary zone of an eagle nest and reported abandoned nests. None of the reported abandoned nests could be attributed to human activities in and around the nest tree. Based on the project plans, the terms of the BEMP, and this analysis, the USFWS concluded that the Ravenswood project "is not likely to adversely affect" the bald eagles at the Ravenswood site. According to the coordination procedures agreed to and employed by the USFWS and the Florida Fish and Wildlife Conservation Commission (FFWCC), the USFWS takes the lead in reviewing bald eagle issues associated with development projects. In accordance with these procedures, for the Ravenswood project, the USFWS coordinated their review and their draft comments with the FFWCC. The FFWCC concurred with the USFWS’s position that the project, with the implementation of the BEMP, will not adversely affect the Ravenswood eagles or their nest. This position by both agencies is consistent with the expert testimony of Mr. Don Palmer, which was based on his 29 years of experience with the USFWS in bald eagle and human interactions. Petitioners and their witnesses raised several valid concerns regarding the continued viability of the Ravenswood eagle nest during and after implementation of the proposed project. One concern expressed was that parts of the Habitat Management Guidelines for the Bald Eagle in the Southeast Region (Eagle Management Guidelines) seem inconsistent with the proposed project. For example, the Eagle Management Guidelines state: "The emphasis [of the guidelines] is to avoid or minimize detrimental human-related impacts on bald eagles, particularly during the nesting season." They also state that the primary zone, which in this case is the area within a 750 foot radius of the nest tree, is "the most critical area and must be maintained to promote acceptable conditions for eagles." They recommend no residential development within the primary zone "at any time." (Emphasis in original.) They also recommend no major activities such as land clearing and construction in the secondary zone during the nesting season because "[e]ven intermittent use or activities [of that kind] of short duration during nesting are likely to constitute disturbance." But the eagle experts explained that the Eagle Management Guidelines have not been updated since 1987, and it has been learned since then that eagles can tolerate more disturbance than was thought at that time. Another concern was that the Ravenswood eagles may have chosen the nest site in Wetland 1 not only for its insulation from existing development to the north and east but also for the relatively sparse development to the west. Along those lines, it was not clear from the evidence that the eagles are used to flying over developed land to forage on the San Sebastian River and its estuaries to the east, as the eagle experts seemed to believe. Mr. Mills testified that eagles have been seen foraging around stocked fish ponds to the west, which also could be the source of catfish bones found beneath the Ravenswood nest. But it is believed that the confident testimony of the eagle experts must be accepted and credited notwithstanding Petitioners' unspecific concerns along these lines. Finally, Petitioners expressed concern about the effectiveness of the monitoring during the nesting required under the BEMP. Some of Petitioners' witnesses related less-than-perfect experiences with eagle monitoring, including malfeasance (monitors sleeping instead of monitoring), unresponsive developers (ignoring monitors' requests to stop work because of signs of eagle disturbance, or delaying work stoppage), and indications that some eagle monitors may lack independence from the hiring developer (giving rise, in a worst case, to the question whether an illegal conspiracy exists between them to ignore signs of disturbance when no independent observer is around). Notwithstanding these concerns, Petitioners' witnesses conceded that eagle monitoring can be and is sometimes effective. If Mr. Steffer is retained as the eagle monitor for this project, or to recruit and train eagle monitors to work under his supervision, there is no reason to think that eagle monitoring in this case will not be conducted in good faith and effectively. Even if the Ginns do not retain Mr. Steffer for those purposes, the evidence did not suggest a valid reason to assume that the Ginns' proposed eagle monitoring will not be conducted in good faith and effectively. K. Other 40C-4.301 Criteria – 40C-4.301(1)(g)-(k) 40C-4.301.301(1)(g) - No minimum surface or groundwater levels or surface water flows have been established pursuant to Florida Administrative Code Rules Chapter 40C-8 in the area of the project. 40C-4.301.301(1)(h) - There are no works of the District in the area of the project. 40C-4.301.301(1)(i) - The proposed wet detention system is typical and is based on accepted engineering practices. Wet detention systems are one of the most easily maintained stormwater management systems and require very little maintenance, just periodically checking the outfall structure for clogging. 40C-4.301.301(1)(j) - The Ginns own the property where the project is located free from mortgages and liens. As previously indicated, they will establish an operation and maintenance entity. The cost of mitigation is less than $25,000 so that financial responsibility for mitigation was not required to be established. (Costs associated with the proposed BEMP are not included as part of the Ginns' mitigation proposal.) 40C-4.301.301(1)(k) - The project is not located in a basin subject to special criteria. Public Interest Test in 40C-4.302 The seven-factor public interest test is a balancing test. The test applies to the parts of the project that are in, on, or over wetlands, and those parts must not be contrary to the public interest unless they are located in, on, or over an Outstanding Florida Water (OFW) or significantly degrade an OFW, in which case the project must be clearly in the public interest. No part of the project is located within an OFW. Balancing the public interest test factors, the project will not be contrary to the public interest. 40C-4.302(1)(a)1. - The project will not adversely affect the public health, safety, or welfare or the property of others because the surface water management system is designed in accordance with District criteria, the post-development peak rate of discharge is less than the pre-development peak rate of discharge, and the project will not cause flooding to offsite properties. 40C-4.302(1)(a)2. - Mitigation will offset any adverse impacts of the project to the conservation of fish and wildlife or their habitats, and the BEMP is designed to prevent adverse effects on the Ravenswood eagles. Although active gopher tortoise burrows were observed on the site, the impacts to these burrows are addressed by the FFWCC’s incidental take permit. The mitigation that is required as part of that permit will adequately offset the impacts to this species. 40C-4.302(1)(a)3. - The project will not adversely affect navigation or cause harmful shoaling. The project will not adversely affect the flow of water or cause harmful erosion. The project's design includes erosion and sediment control measures. The project's design minimizes flow velocities by including flat slopes for pipes. The stormwater will be discharged through an upsized pipe, which will reduce the velocity of the water. The stormwater will discharge into a spreader swale (also called a velocity attenuation pond), which will further reduce the velocity and will prevent erosion in Wetland 1. The other findings of fact relevant to this criterion are in the section entitled "Water Quantity." See Findings 61-67, supra. 40C-4.302(1)(a)4. – Development of the project will not adversely affect the legal recreational use of the project site. (Illegal use by trespassers should not be considered under this criterion.) There also will not be any adverse impact on recreational use in the vicinity of the project site. Wetlands 1 and 5 may provide benefit to marine productivity by supplying detritus to the marine habitat, and these wetlands will remain. 40C-4.302(1)(a)5. - The project will be of a permanent nature except for the temporary impacts to Wetland 1. Mitigation will offset the temporary adverse impacts. 40C-4.302(1)(a)6. - The District found no archeological or historical resources on the site, and the District received information from the Division of Historical Resources indicating there would be no adverse impacts from this project to significant historical or archeological resources. 40C-4.302(1)(a)7. - Considering the mitigation proposal, and the proposed BEMP, there will be no adverse effects on the current condition and relative value of functions being performed by areas affected by the proposed project. The proposed project is no worse than neutral measured against any one of these criteria, individually. For that reason, it must be determined that, on balance, consideration these factors indicates that the project is not contrary to the public interest. Other 40C-4.302 Criteria The proposed mitigation is located within the same drainage basin as the project and offsets the adverse impacts so the project would not cause an unacceptable cumulative impact. The project is not located in or near Class II waters. The project does not contain seawalls and is not located in an estuary or lagoon. The District reviewed a dredge and fill violation that occurred on the project site and was handled by the Department of Environmental Regulation (DER) in 1989. The Ginns owned the property with others in 1989. Although they did not conduct the activity that caused the violation, they took responsibility for resolving the matter in a timely manner through entry of a Consent Order. The evidence was that they complied with the terms of the Consent Order. Applicants' Exhibit 30K was a letter from DER dated February 13, 1991, verifying compliance based on a site inspection. Inexplicably, the file reference number did not match the number on the Consent Order. But Mr. Ginn testified that he has heard nothing since concerning the matter either from DER, or its successor agency (the Department of Environmental Protection), or from the District. The evidence was that the Ginns have not violated any rules described in Florida Administrative Code Rule 40C- 4.302(2). There also was no evidence of any other DER or DEP violations after 1989.
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the St. Johns River Water Management District enter a final order issuing to Jay and Linda Ginn ERP number 40-109-81153-1, subject to the conditions set forth in District Exhibits 1, 2, and 10. DONE AND ENTERED this 16th day of April, 2004, in Tallahassee, Leon County, Florida. S J. LAWRENCE JOHNSTON Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 16th day of April, 2004.
Findings Of Fact Background On January 29, 1981 respondent/applicant, George H. Hodges, Jr. (applicant or Hodges), filed application number 16 39644 with respondent, Department of Environmenta1 Regulation_ (DER), seeking a dredge and fill permit to generally authorize the excavation of 26,000 cubic yards of material from a 3,700 foot portion of an existing channel (Old Pablo Creek) just west of the Intracoastal Waterway (ICW) in Jacksonville, Florida. The channel then proposed was a straight channel along the northern boundary of his property. Hodges also sought to construct two boat slips, three floating docks, an 850 foot vertical bulkhead adjacent to the docks, and to dispose of all dredged material in a diked upland site. Thereafter, DER informally advised applicant that it intended to deny the application for various reasons, including the fact that the dredging would eliminate .75 acres of marsh and wetlands. After receiving this advice, Hodges proposed a series of amendments to his application in 1984 and 1985 in an effort to counter and satisfy DER's objections. The final amendment was made on September 10, 1985. As finally amended, Hodges proposed to confine all dredging to existing salt channels, thereby eliminating the objection that adjacent marshes would be destroyed. Applicant also proposed to restrict his dredging to only 2,250 feet along the northern portion of Old Pablo Creek and to remove 29,250 cubic yards of fill (silt) and sand and place the same in a 12.5 acre upland spoil site. By proposed agency action issued on February 28, 1985, DER announced it intended to issue the requested permit. This prompted a protest and request for hearing from petitioner, Jacksonville Shipyards, Inc. (JSI), which owns and operates a ship repair facility on the ICW just south of the proposed project. In its petition, JSI generally alleged that (a) Hodges had failed to give reasonable assurances that water quality standards would not be violated, (b) the project would adversely affect its property, (c) the project would have an adverse effect on the conservation of fish and wildlife, (d) the project would cause harmful erosion or shoaling, (e) DER failed to consider the long-term effect of the project on marine productivity and the cumulative impact of the project, and (f) the proposed vertical bulkhead did not meet statutory requirements. The Project The project site is a shallow horseshoe shaped creek approximately 3,700 feet in length which meanders through a vegetated salt marsh just west of the ICW in Duval County, Florida. Both ends of the creek connect into the ICW. The site is approximately one-half mile north of the bridge on Atlantic Boulevard which crosses the ICW. The ICW is a man-made channel constructed by the U. S. Corps of Engineers which runs in a north-south direction just east of the project site. It is commonly referred to as Pablo Creek. The channel or creek in which the dredging will occur is known as Old Pablo Creek (creek). An excellent aerial view of the entire area is shown in petitioner's exhibit 4 received in evidence. The creek is a predominately marine water classified as a Class III water of the State. Accordingly, it is subject to DER's regulatory jurisdiction. For purposes of this hearing, the parties have referred to the upper and lower portions of the creek as the northern and southern portions, respectively. Hodges intends to dredge the northern portion of the creek, which measures approximately 2250 feet in length from the ICW to a bend at its western end which crosses Hodges' property and where a residential site is located. According to Hodge's affidavit of ownership, he is the "fee interest owner of adjoining lands except for the dredge channel which is owned by the State of Florida". He acknowledged, however, that the residential site is owned by his superintendent, and that the marshes adjoining the most southern bend in the northern portion of the creek, and the southern portion of the creek, are owned by JSI. Except for the cleared residential site at its western end, the creek is surrounded by vegetation and salt marshes. The vegetated portion of the marsh is marked by a clearly delineated edge which separates it from the creek bottom. The dominant species of vegetation in the marsh are Juncus and Spartina. The marsh serves as a habitat and breeding ground for numerous species, including fiddler crabs, mussels, barnacles, mollusks, faunal communities and gastropods. In addition, the marsh is beneficial because of its biotic productivity and entrapment of nutrients and sediments. For this reason, the habitat should be maintained. Some forty years ago, the portion of the creek that Hodges intends to dredge was eight to twelve feet deep. However, dredging of the ICW by the Corps of Engineers and the placement of fill at the site of the Atlantic Boulevard Bridge have contributed to the shallowing of the creek over time. Today, portions of the creek are exposed and impassable under low tide conditions. Indeed, many parts of the creek are dry during the low tide phase of the ICW. At high tide, the creek is flooded to an approximate depth of four feet. Hodges proposes to dredge the creek channel to a uniform depth of five feet below mean low water (MLW) with side slopes at a 3:1 ratio to restore navigational access from his upland property to the ICW. He has represented that his use of the channel will be restricted to one, or possibly two, small boats for personal use and enjoyment. When completed, the creek channel will have a depth of nine feet at high tide, or an average depth of seven feet over a diurnal cycle. In his amended application, Hodges proposed to confine his dredging to existing creek channels, and to not disturb the actual body of the salt marsh or the vegetation bordering the creek. It is noted that there is no vegetation growing in the existing creek bottom. However, at hearing he conceded that dredging "may include some minor removal of isolated patches of grass growing in the creek channel". One such patch of grass lies in the elbow of the canal which reaches south of Hodges' property, a patch separated from the main body of the marsh by a five foot wide slough deep enough to be navigated at high tide. Hodges estimates this patch of grass to be less than 1/100 of an acre in size (10' x 40') and maintains the effect of its removal would be negligible. The excavation will be effected by means of a Mud Cat hydraulic dredge which operates by suctioning the sediment and water into a pipe. The dredge material (sediment/water mixture) will then be pumped into a series of containment cells on a 12.5 acre upland spoil site that lies approximately one-half mile northeast of the project. Any discharge from the spoil site will be to Greenfield Creek, a tidally influenced creek connected to the St. Johns River. The natural grade of the existing creek bottom is at or below the mean low water datum. At high tide the existing creek is 4.3 feet deep at its deepest point and gradually slopes upward to a depth of 2.4 feet near the marsh. The elevation of the creek where it meets the marsh is close to mean high water. Even so, the channel width does not always correspond with the mean high water line boundaries of the creek, and creek waters sometimes inundate and extend back into the marsh at high tide. Because Old Pablo Creek is tidally influenced, any water quality violations in the northern portion of the creek can be expected to also have an adverse effect in the southern portion as well. Creek Width Petitioner has raised the issue of whether the creek is as wide as Hodges represents it to be on the drawings attached to the amended application. This is significant since (a) the engineering plans are based upon the assumption that the measurements in the application are correct, (b) the proposed dimensions (depth and side slopes) of the new channel are dependent upon the existing creek having a minimum width of from sixty to eighty feet, as represented by Hodges, and (c) any excavation outside of the existing channel will result in the removal (destruction) of vegetation and marsh. In his application, Hodges reflects the top width of the creek to be sixty to eighty feet, which width will enable him to dredge the channel to an average depth of five feet below MLW, and maintain a side slope ratio of 3:1. This ratio is necessary because of the composition of the sediment in the creek. The minimum top width required to excavate a channel with 3:1 side slopes to a depth of five feet below MLW is fifty- four feet. Petitioner's exhibit 4 identifies five points along the eastern half of the northern portion which have been measured by the parties to determine the actual width of the creek. Although only five points were measured, it may be inferred that these distances are representative of the creek's width throughout its eastern half. At points five through eight, the widths are forty-nine, thirty-five, fifty and fifty feet, respectively, which are less than the measurements contained in the application. If the channel is constructed with the minimum top width (54 feet) required to have 3:1 side slopes, it will result in the elimination of 6 feet of marsh at point 5, 19.5 feet of marsh at point 6, and 4.1 feet of marsh at both points 7 and 8. This equates to the elimination of approximately .33 acres of marsh. Since the above measurements are representative of the eastern half of the northern portion, other areas of vegetation, albeit in unknown proportions, would also have to eliminated. If, for example, applicant attempts to construct a channel within the confines of the portion of the creek that has a top width of only thirty-five feet (point 6), the maximum channel that could be constructed would be V-shaped with a depth of one foot at low tide. Assuming the remaining part of the channel was excavated to -5' MLW, a stagnant area would develop in this portion of the channel and adversely affect water quality. However, to counter the problem at point 6, Hodges intends to remove one patch of grass 10' by 40' in size to achieve the desired width. Any adverse effects on the adjacent marsh at that particular point would be negligible. Because the estimated creek width is not accurate, even the agency now concedes the engineering plans are no longer useful. As a condition to the issuance of a permit, DER has suggested that Hodges be required to submit new certified engineering drawings depicting the proposed cross-section of the channel. It also suggests that the proposed cross-section comply with the top-widths depicted in applicant's exhibit 53, and depict side-slopes of three to one. It further suggests that a condition for the issuance of any permit be a requirement that the 3:1 ratio be maintained, and that other than point 6, no other grass be removed. Finally, the agency proposes that if the new plans and conditions do not permit a -5 MLW depth, the proposed depth be reduced accordingly. However, the evidence supports a finding that either vegetation must be removed at various points along the eastern half of the creek in order to maintain a 3:1 ratio for side slopes, or the depth must be reduced. By reducing the depth at certain points, stagnant areas in the creek will develop, thereby adversely affecting the quality of the water. Further, as noted hereinafter, the validity of the flushing analysis performed by applicant's experts rests upon the assumption that a -5' MLW uniform depth will be used. Finally, the applicant has not given reasonable assurance that the marsh and habitat will not be adversely affected by the elimination of the vegetation which is necessary to achieve the desired depth and concomitant 3:1 ratio. Therefore, the alternative conditions suggested by DER are neither reasonable or appropriate. The Spoil Area The spoil area to be used by applicant is a 12.5 acre upland disposal site approximately one-half mile northeast of Hodges' property. Applicant does not own the upland spoil site but has obtained easements from the owner which expire in March, 1987. In other words, he must complete all work on the project by that date or lose access to the property. The proposed spoil site is completely diked, and is sectioned off into three sections by interior dikes with overflow pipes. Internal baffles and silt fences are also designed into the area. Uncontradicted testimony established that the spoil area is "unusually well designed". Any discharge from the spoil area will be to Greenfield Creek, a tidally influenced creek connected to the St. Johns River. Discharge, if any, will be outfall from an overflow structure in the third section of the spoil area to a dump area land then by sheet flow to salt marshes adjacent to Greenfield Creek. The vegetation in Greenfield Creek consists of a salt marsh expanse of Spartina alterniTlora and Juncus roemerianus. Both species survive in and are indicative of regular introduction of saline waters, and show high tolerance to varying salinity levels. If saline waters from Old Pablo Creek were introduced into Greenfield Creek, it would have no adverse impact on the Greenfield Creek ecosystem. The size of the site was originally designed for a project of 100,000 cubic yards. The site will retain all |effluent from the dredging. The expected total effluent, both sediment and water, is roughly 5.3 million cubic feet of material, assuming a ratio of 6.7 cubic feet of water for each |cubic foot of sediment dredged. This is slightly lower than the 5.4 million cubic feet total capacity of the site. The supernatant from the discharge being deposited into the first cell of the spoil area will only flow into the next cell when the first cell fills and the level of the supernatant rises above the top of the vertical drain pipe overflow structure. If rainfall events cause the cells to fill with water during dredging and discharge operations, the discharge to the next cell or to Greenfield Creek will be primarily fresh water. This will occur because introduction of fresh rainwater into the brackish water from the dredge area will cause stratification, and the fresh rainwater will form a layer on top that will flow into the overflow structure. Turbidity Effects In removing the mud bottom from the creek to a depth of -5' MLW, some turbidity will occur. This is a natural by- product of using the hydraulic dredge. However, the amount of turbidity, and its effect on the waters at the dredge site and discharge point, are in issue. State water quality standards prohibit the discharge of water with a turbidity level greater than twenty-nine nephelometric units (NTU's) above the background levels of the receiving waters. The evidence indicates that the background turbidity levels at the creek are now in the range of ten to twenty NTV's. Excessive levels can result in adverse effects on local biota such as decreasing productivity by reducing light penetration. Excessive turbidity can also be expected to suffocate organisms. The area to be dredged contains sediment deposited from the surrounding salt marsh and carried in from the ICW. The sediment is composed of 14% clay, with the remainder being sand and silt. This was confirmed by a laboratory analysis conducted by JSI. As a general rule, the coarser the material, the faster it tends to settle out thereby creating less turbidity problems. Therefore, sand, which is of a grain size, can be expected to settle out quickly while silt takes somewhat longer. However, clay size particles are much smaller than silt and do not settle out as easily. Applicant made no laboratory analysis of sediment and consequently he erroneously assumed the mud to be sand and silt, and did not take the clay particles into account. The dredging in the creek will cause the turbidity levels to rise to 150 NTU's. However, the placement of a turbidity screen at the entrance to the ICW will prevent the release of this turbidity into that water body. Therefore, if a permit is issued, such screens should be used by Hodges at the dredge site. At the spoil site, clay size particles will also be included in the matter pumped for discharge. If these particles do not settle out, or are not treated, their discharge into Greenfield Creek (a jurisdictional water) will cause violations of the turbidity standards. To counter their effects, flocculants (chemicals) should be added when necessary to the confined material to aid the particles in settling. If a permit is issued, this should be made a condition in the permit. Dissolved Oxygen Impacts The dissolved oxygen (DO) levels in the creek fluctuate on a daily and seasonal basis. As a general rule, DO levels tend to be lower in warmer weather and during the early morning hours. Therefore, a "worst case" situation will generally occur in the summer months in the early part of the day. State water quality standards contained in Rule 17- 3.121(13), F.A.C., provide that in predominately marine waters, the concentrations of DO "shall not average less than 5 milligrams per liter in a 24-hour period and shall never be less than 4 milligrams per liter." Sampling conducted by petitioner at 5:00 a.m. in early July, 1986 during high tide revealed readings ranging from 3.06 mg/1 in the western portion of the creek to 4.59 mg/1 at the mouth of the creek. Dissolved oxygen levels in the ICW ranged from 3.94 to 4.68 mg/1. Hodges also sampled the creek and ICW in the late morning or early afternoon on August 6,1986 and determined DO levels to be 4.8 mg/1 in the creek and 5.8 mg/1 in the ICW. Testing at that hour of the day produced higher values than those found by JSI. The readings collectively confirm that DO levels in the creek are approximately 1.0 mg/1 less than the DO levels in the ICW. This deficit is primarily caused by the high oxygen demand exerted by the adjacent marsh and muds in the creek. This situation will not be changed by the dredging. The flushing time of the creek channel is an important factor in predicting post-dredging impacts on water quality. Flushing time determines how rapidly waters of the ICW will exchange and mix with the water in the creek channel. Both Hodges and JSI conducted tidal prism studies to determine how many tidal cycles would be required to flush a hypothetical pollutant to 10% of its initial concentration. Under worst case conditions, the channel is expected after dredging to flush every 3 to 4 tidal cycles or 1.6 days. Under more favorable conditions, the creek is expected to flush every 2 to 3 tidal cycles. This compares with the current system which flushes almost 100% every tidal cycle or once every twelve hours. The increased flushing time is due to the significantly greater volume of water that will enter the creek channel after dredging. Because of increased channel depths, the water will move at a slower velocity. Therefore, the oxygen consuming components have a longer period of time to react in the water column. This in turn will cause reductions in DO levels of between .7 mg/l and 1.5 mg/l in the creek. This was confirmed through tidal prism modeling performed by JSI. In this regard, it is noted that JSI's modeling was more sophisticated, better calibrated, and its assumptions were more accurate and reasonable. Consequently, its testing results are considered to be more reliable and persuasive than that of applicant. It must also be recognized that the deepening of those areas that are currently exposed at low tide will allow water to move more easily through the channel and remove some oxygen demanding sediments that now draw from a shallow water column. This will tend to have a beneficial effect on water quality. However, the overall impact of these beneficial effects is unknown, and it was not demonstrated that the otherwise adverse effect on DO will be offset or minimized by the unmeasured impact of deepening the shallow areas. Therefore, applicant has not given reasonable assurance that water quality standards will not be violated by the project. At the same time, it must be further noted that a reduction in the channel depth due to the smaller width of the creek will alter the results of the tidal prism studies, as well as negate some of the beneficial effects caused by deepening the shallow portions of the channel. To what extent the studies are changed, or benefits will be reduced, is not of record. Other Effects of Project As noted earlier, Hodges intends to use one or two boats on the deepened channel. The use of the boats will not introduce pollutants in any significant quantity. Hodges proposes to construct his docks and place rip- rap on the northern side of the widest portion of the creek channel. Little, if any, vegetation will be eliminated by these activities. The use of rip-rap for the construction of the bulkhead is the most environmentally sound means of bulkheading, and will stabilize the shoreline as well as provide habitat for aquatic organisms. The dredging of the creek channel will improve the navigability of the creek, and permit the use of boats in areas where access is now impossible under low-tide conditions. In addition, the sharp bends in the creek will prevent the operation of boats at high speeds. JSI's concern that boats may run aground once they leave the northern portion and enter the southern portion is not meritorious since few, if any, are expected to use the latter part of the creek, and the sharp bends will force boaters to operate at low speeds. Shoaling or erosion of the southern portion will not result from the proposed activities. Indeed, an increased flushing and introduction of new flow into the system may benefit the northern portion. Any situation occurring in that part of the creek should not exceed the rate of siltation occurring under current conditions. The benthic organisms which populate the bottom of Old Pablo Creek include crabs, mussels, barnacles and other species normally associated with estuarine systems. The removal of the mud bottom in the dredging operation may remove some of these organisms. However, this should not significantly change the habitat of these benthic organisms. Rapid recolonization by these species would be expected with recolonization substantially underway within forty-eight hours
Recommendation Based on the foregoing findings of fact and conclusions of law, it is RECOMMENDED that application number 16-39644 of George H. Hodges, Jr. for a dredge and fill permit be DENIED. DONE and ORDERED this 2nd day of December, 1986 in Tallahassee, Florida. DONALD R. ALEXANDER, Hearing Officer Division of Administrative Hearings The Oakland Building 2009 Apalachee Parkway Tallahassee, Florida 32399 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 2nd day of December, 1986.
Findings Of Fact The Parties. The Petitioners, Joseph and Lena Smith, Eugene and Anna Colwell, and Jerry and Brenda Harris, are littoral owners and operators of sports fishing facilities on Orange Lake, a freshwater body of approximately 7,000 acres of open water and 15,000 acres of associated wetlands, whose southern margin constitutes the boundary between Alachua and Marion Counties in north central Florida. Respondent, the St. Johns River Water Management District (hereinafter referred to as the "District"), is a special taxing district created by Chapter 373, Florida Statutes, charged with the statutory responsibility for the management of water and related land resources; the promotion of conservation, development, and proper utilization of surface and ground water; and the preservation of natural resources, fish and wildlife, pursuant to Chapter 373, Florida Statutes. Intervenor, the Sierra Club, Inc. (hereinafter referred to as "Sierra"), is a not-for-profit California corporation registered to do business within the State of Florida. Sierra is an international corporation whose purpose is to explore, enjoy and protect the natural resources of the earth. Intervenor, Florida Defenders of the Environment, Inc. (hereinafter referred to as "Florida Defenders"), is a not-for-profit Florida corporation whose purpose is to preserve and restore Florida's natural resources. Intervenor, the Florida Department of Environmental Protection (hereinafter referred to as "DEP"), is an agency of the State of Florida charged with the responsibility of controlling and prohibiting pollution of the air and water of the State of Florida. See Chapter 403, Florida Statutes. DEP is also charged with responsibility for management of the Paynes Prairie State Preserve. Section 373.026, Florida Statutes. Intervenor, the Attorney General of the State of Florida (hereinafter referred to as the "Attorney General"), sits as a Trustee of the sovereignty submerged lands of the State and as one of the legal owners of the State's property including the Paynes Prairie State Preserve. The Challenged Rules. The District issued an order on November 7, 1993, authorizing the publication of a notice of intent to amend Chapter 40C-2, Florida Administrative Code, by adopting proposed Rule 40C-2.302, Florida Administrative Code, and amending Rule 40C-2.051(6), Florida Administrative Code (hereinafter jointly referred to as the "Challenged Rules"). Proposed Rule 40C-2.302, Florida Administrative Code (hereinafter individually referred to as the "Reservation Rule"), provides: 40C-2.302 Reservation of Water From Use. The Governing Board finds that reserving a certain portion of the surface water flow through Prairie Creek and Camps Canal south of Newnans Lake in Alachua County, Florida, is necessary in order to protect the fish and wildlife which utilize the Paynes Prairie State Preserve, in Alachua County, Florida. The Board therefore reserves from use by permit applicants that portion of surface water flow in Prairie Creek and Camps Canal that drains by gravity through an existing multiple culvert structure into Paynes Prairie. This reservation is for an average flow of [35] cubic feet per second (23 million gallons per day) representing approximately forty five per cent (45 percent) of the calculated historic flow of surface water through Prairie Creek and Camps Canal. The specific authority for the Reservation Rule is Sections 373.044, 373.113, 373.171, 373.216 and 373.219, Florida Statutes. The law implemented by the Reservation Rule is Sections 373.219 and 373.223, Florida Statutes. The proposed amendment to Rule 40C-2.051, Florida Administrative Code (hereinafter individually referred to as the "Exemption Rule"), provides, in pertinent part: 40C-5.2.051 Exemptions. No permit shall be required under the provisions of this rule for the following water uses: through (5) No change (6) Water, whether withdrawn or diverted, when used for purposes of protection of fish and wildlife or the public health and safety when and where the Governing Board has, by regulation, reserved said water from use by permit applicant pursuant to Subsection 373.223(3), F.S. The specific authority for the Exemption Rule is Sections 373.044, 373.113 and 373.171, Florida Statutes. The law implemented by the Exemption Rule is Sections 373.103, 373.171, 373.216, 373.219, 403.501 et seq. and 288.501 et seq., Florida Statutes. Orange Creek Basin. Orange Creek Basin is the name given to the hydrological features of approximately 400 square miles of Alachua, Putnam and Marion Counties, Florida. Orange Creek Basin is a major sub-basin of the Lower Ocklawaha River Basin. Surface water in the Orange Creek Basin flows generally in a north to south direction. Orange Creek Basin is made up of several sub-basins, including Newnans Lake, Paynes Prairie, Orange Lake and Lochloosa Lake sub-basins. Surface water within the approximately 100 square miles of Newnans Lake sub-basin drains into Newnans Lake. When sufficiently high, water in Newnans Lake discharges over a weir structure from the southern end of the lake into Prairie Creek. The weir structure at the southern end of Newnans Lake may be adjusted to control the amount of water flowing into Prairie Creek. The weir was installed in 1966. It was adjusted by the Florida Game and Freshwater Fish Commission in 1976. Water flows south into Prairie Creek, the south and southwest through Prairie Creek to two man-made structures. The first is a gated culvert structure consisting of 3 Culverts (the "Camps Canal Culverts"), through which some of the Prairie Creek water enters Paynes Prairie. The second man-made feature is a levee and a canal named Camps Canal. The levee diverts water in Prairie Creek, which does not flow into Paynes Prairie by gravity, through Camps Canal to the south to the River Styx, which flows into Orange Lake. If the elevation of surface water in Prairie Creek exceeds 58.91 feet National Geodetic Vertical Datum (hereinafter referred to as "NGVD"), a portion of the volume of Prairie Creek will flow, by gravity, into Paynes Prairie through the Camps Canal Culverts. The Paynes Prairie sub-basin covers an area of approximately 49 square miles. Surface water in this sub-basin drains into a natural geological feature known as Alachua Sink. Surface water in the approximately 56 square mile Orange Lake sub- basin flows into Orange Lake. Surface water flows out of Orange Lake through Orange Creek. Outflow is controlled by Orange Lake Dam. The Orange Lake Dam has a fixed crest elevation of 58 feet NGVD. Water levels in Orange Lake must exceed 58 feet NGVD before there is surface water outflow from Orange Lake. Surface water within the approximately 75 square mile Lochloosa Lake sub-basin drains into Lochloosa Lake. Lochloosa Lake has two outlets: Lochloosa Slough in the east and Cross Creek in the south. Cross Creek connects Lochloosa Lake to Orange Lake. Paynes Prairie State Preserve. Prior to the construction of the weir at the outlet from Newnans Lake to Prairie Creek, all surface water from Newnans Lake flowed from Newnans Lake to Prairie Creek unimpeded. Prior to 1927 all surface water in Prairie Creek flowed south into an area known as Paynes Prairie. Paynes Prairie is located in Alachua County. All water in Prairie Creek entered Paynes Prairie and flowed across Paynes Prairie to Alachua Sink. Alachua Sink is a natural geological feature located in the north- central portion of Paynes Prairie. At Alachua Sink surface water enters the Florida aquifer. In 1927 a levee was constructed around the eastern boundary of Paynes Prairie, and Camps Canal was excavated in order to divert water from Paynes Prairie. Due to the levee, water in Prairie Creek was diverted into Camps Canal beginning in approximately 1927. The water flowed into the River Styx and then into Orange Lake. Canals and levees were also constructed within Paynes Prairie to convey surface water in Paynes Prairie into Alachua Sink and Camps Canal. The modifications to Paynes Prairie made in 1927 were intended to drain Paynes Prairie so that the land could be utilized for agricultural purposes, including the raising of cattle. Paynes Prairie continued to be used primarily for the raising of cattle between 1927 and early 1970. In 1970, the State of Florida began acquiring parts of Paynes Prairie. Property acquired by the State was used to create the Paynes Prairie State Preserve (hereinafter referred to as the "Preserve"). Land is still being acquired by the State. The Preserve currently consists of approximately 20,600 acres. Approximately 18,000 acres of the Preserve were acquired within the first 4 years after acquisitions by the State began. Approximately 12,000 acres are considered wetlands. Two major highways, U.S. Highway 441 and Interstate 75 run north-south across the middle and western portion of Paynes Prairie. U.S. 441 was constructed in 1927 and I-75 was constructed in 1964. In 1975 the State of Florida's Department of Natural Resources (which is now DEP) breached the levee at Camps Canal in order to restore part of the water flow from Prairie Creek to the Preserve. In 1979 flashboard riser Culverts were placed in the breach in the Camps Canal levee. In 1988 the Camps Canal Culverts were constructed. The Preserve, a unique land feature, was designated a National Natural Landmark in 1974 by the United States Department of the Interior. No consumptive use permit concerning water that flows into Paynes Prairie or the Preserve has been issued by the District. No consumptive use permits have been issued by the District for surface water withdrawals from Newnans Lake, Prairie Creek or Orange Creek. The Current General Hydrologic Condition of the Preserve. The Preserve is one of the largest continuous wetland systems in Florida and the Southeastern United States. The Preserve and Paynes Prairie constitute one of the largest wetland areas formed by the collapse of a sinkhole, Alachua Sink. Since 1975, at least some water has flowed into the Preserve from Prairie Creek through the Camps Canal Culverts and its predecessors. The "inverts" of the Prairie Creek-Camps Canal Culverts are above the creek-canal bottom. This means that if water in Prairie Creek does not reach a certain level, no water will flow through the Camps Canal Culverts into the Preserve. Under these conditions, all water in Prairie Creek will flow through Camps Canal and eventually to Orange Lake. The amount of water flowing through the Camps Canal Culverts is also limited to a maximum amount due to the size of the Culverts. The exact amount of water that may flow through the Camps Canal Culverts into the Preserve depends on the amount of water in Prairie Creek coming from Newnans Lake and the capacity of the Culverts to move the water. Water flowing into the Preserve through the Camps Canal Culverts constitutes approximately 50 percent of the surface water entering the Preserve. After water flows into the Preserve through the Camps Canal Culverts it flows in a broad, shallow path, referred to as "sheetflow," over the eastern portion of the Preserve. The sheetflow from Camps Canal Culverts creates approximately 550 to 600 acres of shallow marsh community. The water eventually flows into an area known as Alachua Lake in the central portion of the Preserve. Water discharging from Alachua Lake flows through a water control structure consisting of four gated Culverts, known as the Main Structure, into Alachua Sink. Water also enters the Preserve from the north through a tributary known as Sweetwater Branch. Water flows through Sweetwater Branch into Alachua Sink. Sweetwater Branch is channelized over its entire length, preventing water from reaching into the Preserve or Alachua Lake. The District's Purpose in Adopting, and the District's Interpretation of, the Challenged Rules. The District's intent in adopting the Challenge Rules was to reserve water which the District had concluded is required for the protection of fish and wildlife in Paynes Prairie. The District is attempting to carry out its intent by providing in the Reservation Rule that whatever amount of water that may flow through the Camps Canal Culverts by gravity into the Preserve may not be used for other purposes. The District is further attempting to carry out its intent by providing in the Exemption Rule that any amount of water that has been reserved by the District because it is required for the protection of fish and wildlife pursuant to Section 373.223(3), Florida Statutes, exempt from the consumptive use permit process. The Reservation Rule is not intended to reserve a specific quantity of water for the Preserve. Rather, the Reservation Rule reserves only that amount of water that flows through the Camps Canal Culverts by force of gravity. The intent is to allow the natural existing hydrologic regime of the Preserve to continue. The quantity of the water reserved by the Reservation Rule is identified, in part, as follows: The Governing Board finds that reserving a certain portion of the surface water flow through Prairie Creek and Camps Canal south of Newnans Lake in Alachua County, Florida, is necessary in order to protect the fish and wildlife which utilize the Paynes Prairie State Preserve, in Alachua County, Florida. The Board therefore reserves from use by permit applicants that portion of surface water flow in Prairie Creek and Camps Canal that drains by gravity through an existing multiple culvert structure into Paynes Prairie. . . . [Emphasis added]. The last sentence of the Reservation Rule goes on to prove: This reservation is for an average flow of [35] cubic feet per second (23 million gallons per day) representing approximately forty five per cent (45 percent) of the calculated historic flow of surface water through Prairie Creek and Camps Canal. This portion of the Reservation Rule was not included by the District to establish a minimum and/or maximum quantity of water that is being reserved for the protection of fish and wildlife in the Preserve. This portion of the Reservation Rule represents a very condensed summary of the historical hydrologic data relied upon by the District in deciding to reserve water for the Preserve's fish and wildlife. The Exemption Rule was intended to make clear that any time the District reserves water which it determines is required to protect fish and wildlife or the public safety, that no consumptive use permit is necessary. The District's Determination that Water is Necessary for the Protection of Fish and Wildlife in Paynes Prairie. In reaching its decision that the quantity of water flowing through the Camps Canal Culverts by force of gravity into the Preserve is required for the protection of the fish and wildlife of the Preserve, the District relied upon a study of the Orange Creek Basin which District staff had begun in the 1980s. There were three objectives for the Orange Creek Basin study: (a) the first objective of the study was to develop a predictive hydrologic model that could be used to predict water levels throughout the basin and the water courses that connect the various major lakes and prairie systems; (b) the second objective of the Orange Creek Basin study was to develop environmental and hydrologic criteria that could be used to evaluate the environmental impacts of different water management alternatives in the basin; and (c) the third objective was to look at alternatives for management of water within the District. Substantial evidence concerning the manner in which the Orange Creek Basin study was conducted, the results of the study and the rationale for the District's conclusion that the quantity of water flowing through the Camps Canal Culverts by force of gravity is required to protect the fish and wildlife of the Preserve was presented during the final hearing of this case by the District. The evidence presented by the District to support a finding that the quantity of water flowing through the Camps Canal Culverts by force of gravity is required to protect the fish and wildlife of the Preserve was not rebutted by competent subs by the Petitioners. The only witness called by the Petitioners was an expert in hydrology. The Petitioners' expert only suggested that he had questions about the District's hydrologic study. He was unable, however, to testify that the hydrologic study relied on by the District was unreasonable or inaccurate. The Petitioners also offered no evidence to counter the testimony of the District's expert on the environment of Paynes Prairie. The testimony of the District's expert proved that, even without the results of the hydrologic study conducted by the District, the evidence concerning the Preserve's environment supports a finding that the water reserved by the Reservation Rule is required for the protection of fish and wildlife. Generally, the evidence proved that, if the water being reserved is not continued to allow to flow naturally into the Preserve, the range of water fluctuations and the resulting natural impact of the environment of the Preserve will not be achieved. There exists in the Preserve currently, a range of plant communities and fish and wildlife. The nature of those communities, fish and wildlife depends on the amount of water in the communities. The communities range from those existing in upland areas, which have the lowest levels of water, down to deep marshes, where water levels are the greatest. In between are emergent marsh (also called "shallow marsh"), cypress swamps, mixed scrub-shrub wetland, wet prairie, old field, hudric forest, mesic forest and xeric community. The various types of communities are in a state of fluctuation depending on the levels of water flowing into the Preserve. The evidence presented by the District, and was uncontroverted by the Petitioners, proved that these fluctuations are environmentally desirable; that natural fluctuations of water levels in the Preserve are required for the protection of fish and wildlife. It is for this reason, therefore, that the District decided to reserve the amount of water flowing by gravity through the Camps Canal Culverts, and not some specified volume. The Rationale for the District's Finding that Water is Required for the Protection of Fish and Wildlife. Although the District and some of the Intervenors have proposed several findings of fact that support the ultimate finding of fact that the water reserved by the Reservation Rule is required to protect fish and wildlife. Those findings of fact are subordinate to the ultimate relevant fact in this case. Therefore, rather than rewrite all of those subordinate facts, the District's subordinate findings of fact (which cover those subordinate findings suggested by the Intervenors) will be quoted and adopted in this Final Order. The findings of fact of the District quoted and adopted herein which relate to the hydrologic portion of the District's study are as follows. The findings have been modified to reflect terms used throughout this Final Order. The findings of the District adopted are District findings of fact 44 through 74: Surface water hydrologic models are a tool used by water resource professionals to enable them to simulate or calculate certain characteristics of a hydrologic system from data that relates to or is collected from within that system. T. 65, 66, 90, 91, 779. In this basin, the staff of the District developed a surface water model in order to calculate anticipated water levels and discharge volumes at various points throughout the basin expected to be associated with several alternative water management strategies. T. 90, 91, SJ Ex 1 p 27. The specific model used by the District is the Streamflow Synthesis and Reservoir Regulation (SSARR) mathematical model, developed by the U.S. Army Corps of Engineers. This particular model is generally accepted and used in the field of hydrology for the purposes for which it was used here by the District staff. T. 90, 91, SJ Ex 1 p 27. The model combines two types of data, the first of which are "fixed basin parameters" such as drainage area, soil moisture run-off relationships, and storage capacity of the water bodies in the basin. Fixed basin parameters do not change over time. T. 98, 99, SJ Ex 1 pp 32-37. The second type of data used by the model is "time series" data such as rainfall, evaporation, lake elevations and discharges at several points throughout the basin. Time series data does change over time. T. 98, 99, SJ Ex 1 pp 38-40. Rainfall data for the basin is the most important input element for the model because rainfall drives the system from a hydrologic perspective. T. 95. Rainfall data from 5 recording stations scattered over the basin were utilized, with one station located at the University of Florida in Gainesville yielding data for more than 50 years, although only data for the 50 year period from 1942-1991 was used in the model. T. 96, 97, SJ Ex 1 pp 38, 39, 62, 175. The other 4 rainfall recording stations used in the model have recorded rainfall for periods ranging from 11 years to 37 years. SJ Ex 1 p 39. In a basin the size of the Orange Creek Basin, day to day rainfall amounts may vary from one recording station to another, however, on an annualized basis, rainfall amounts are relatively consistent between the rainfall recording stations utilized in the District's model. T. 97, 98, 184, 727. Both the number and location of rainfall recording stations used for the model are adequate to characterize rainfall for the basin. T. 97, 98, 184. Fifty years of hydrologic data were utilized by the District in the model, because corresponding records existed for rainfall, lake levels, and discharge for this period of time. In addition, a 50 year period is more likely to exhibit a full range of hydrologic conditions, such as droughts and floods, than a shorter increment of time would. T. 104. The model utilizes both the fixed basin parameters and the time series data to calculate an associated lake level for any of the lakes in the basin or a discharge measurement at one of several points in the basin for any particular day during the 50 year period represented by the hydrologic data on which the model is based. T. 98-100. The model was initially run to calculate several hydrologic values with existing conditions in place. Existing conditions, for purposes of comparison with other alternatives, assumes the Newnans Lake weir to be in place, the gates to the Camps Canal Culverts to be in an open position and the gates to the main structure Culverts in the Preserve to be in an open position. T. 99, SJ Ex 1 p 83. For all scenarios examined, the model assumes existing land uses to be in place, in all years simulated, in order to allow consistent comparisons of hydrologic conditions over the 50 years for which data was available. T. 134, 135. In the "existing conditions" scenario the model calculates the volume of water discharging from Newnans Lake southward into Prairie Creek for each day during the 50 year period from 1942-1991. T. 100. Discharge measurements were made by District staff at the downstream end of the Camps Canal Culverts from which a rating curve was developed for the structure. T. 101, 102, SJ Ex 1 pp 33, 36. A rating curve is a means by which the flow capacity of a water control structure such as a culvert may be calculated. T. 101, 102. Using the rating curve developed by District staff for the Camps Canal Culverts, the model, having calculated the volume of water moving from Newnans Lake into Prairie Creek, can then calculate the volume of water passing through the Culverts at the Camps Canal Culverts into the Preserve versus the volume moving on southward through Camps Canal to Orange Lake for each day or year during the 50 year period from 1942-1991. T. 101, 102, SJ Ex 1 p 84, Appendix Table E-45. Having calculated the annual volume of surface water entering the Preserve and the annual volume moving into and through Camps Canal to Orange Lake for each of the 50 years between 1942- 1991, District staff then divided the 50 year totals for each by 50 to arrive at a yearly average volume of water going to the Preserve versus a yearly average volume going through Camps Canal to Orange Lake, under existing conditions. T. 101-104, SJ Ex 1 p 84, Appendix Table E-45. Based on the volumes calculated for the 50 year period between 1942-1991, on average, 45 percent of Prairie Creek flow enters Preserve through the Camps Canal Culverts under existing conditions. This equates to 35 cubic feet per second (cfs), or 23 million gallons per day (mgd). T. 103, 605, 606, SJ Ex 1 p 84, Appendix Table E-45. Also based on the volumes calculated for the 50 year period between 1942-1991, on average, 55 percent of Prairie Creek flow goes into Camps Canal and moves on southward to the River Styx and then to Orange Lake under existing conditions. T. 103, SJ Ex 1 Appendix Table E-45. Making a calculation of flow based on 50 years of historic hydrologic data does not guarantee that the next 50 years will be identical to the period during which the calculation was developed, however, it is reasonable to assume that the next 50 years will be statistically similar to the previous 50 years and that hydrologic conditions, on average, will be the same. T. 104, 143. Both the general methodology and the specific model used by the District to quantify the average volume of flow entering the Preserve under existing conditions, which also represents the volume of flow which the rule would reserve for fish and wildlife which use the Preserve, are based on logic and accepted scientific principles. T. 90, 91, 97, 102, 128, 729. The rule in issue does not reserve a specific amount of water for the protection of fish and wildlife using the Preserve, rather, it reserves the amount which will flow by gravity through the existing Camps Canal Culverts with the gates in an open position, which will in essence, maintain the existing volume of flow into the Preserve. T. 604, 605, 624. Thirty-five cfs does not necessarily represent the specific volume of water that will flow into Preserve on a given day, rather, the specific volume would be dependent on hydrologic conditions on that given day. T. 105, 106. Nevertheless, 45 percent of flow, or 35 cfs, or 23 mgd, represents a reasonably accurate calculation, based on the data available, of the average volume of Prairie Creek flow which will enter the Preserve by gravity pursuant to the Reservation Rule. T. 101- 104, 638, SJ Ex 1. With the existing conditions hydrologic regime which the Reservation Rule would continue in place, the model calculates that the mean elevation of Orange Lake would be 57.26 feet NGVD. T. 121, 122, SJ Ex 8 (arithmetic mean). If no Prairie Creek flow were allowed to enter the Preserve and all of its flow went to Orange Lake, the model calculates the mean elevation of Orange Lake to be 57.51 feet NGVD. T. 121, 122, SJ Ex 8 (arithmetic mean). Thus, the mean elevation of Orange Lake rises by only 0.25 feet when all of the Prairie Creek flow is diverted to Orange Lake. SJ Ex 8. The impact of a 0.25 feet change in the mean elevation of Orange Lake from a hydrologic perspective is small given the 11 feet fluctuation in elevations that has occurred naturally over time in the lake. T. 125. By contrast, if no Prairie Creek flow were allowed to enter the Preserve and all of its flow went to Orange Lake, the mean elevation of water levels within the Preserve, as calculated by the model, would decline by 0.65 feet. SJ Ex 7. Eliminating all Prairie Creek flow from the Preserve would decrease the amount of wetted acreage in the central portion of the prairie by up to 2400 acres. T. 203, SJ Ex 1 p 131, SJ Ex 6. In addition, the acreage wetted in the eastern lobe of the Preserve by the sheetflow of Prairie Creek water as it moves from the Camps Canal Culverts to Alachua Lake would also be eliminated. T. 116, SJ Ex 1 p 131. The findings of fact of the District quoted and adopted herein which relate to the environment of, and the alternative course of action considered for, the Preserve are as follows. The findings have been modified to reflect terms used throughout this Final Order. The findings of the District adopted are District findings of fact 79 through 127: The eastern and western lobes of the Preserve are approximately the same elevation and have similar gradients; however, the plant communities within the eastern lobe differ from the plant communities in the western lobe. The plant community within the eastern lobe is predominantly a shallow marsh community while the plant community within the western lobe varies from wet prairie to old field. T. 262, 263; SJ Exs 3, 10B, 10H. For the western lobe of the Preserve, consisting of the area west of U.S. Highway 441, rainfall is the only source of water except when extremely high water levels occur in Alachua Lake. T. 263, 272. When extremely high water levels occur on Alachua Lake water can backflow through the culverts under U.S. Highway 441 and Interstate Highway 75 and inundate the western lobe. T. 272. The eastern lobe of the Preserve is dependent upon sheetflow from Prairie Creek for its source of water. T. 263. Prior to the construction of Cones Levee the sheetflow from Prairie Creek inundated approximately 1,200 acres of the eastern lobe. Today, however, sheetflow inundates directly 600 acres and indirectly another 600 acres in the eastern lobe. T. 264, 265; SJ Ex 10B. Without the Prairie Creek sheetflow, the biological character of the eastern lobe would change to resemble the more terrestrial nature of the western lobe. T. 263, 272, 518. The fish and wildlife inhabiting the Preserve are totally dependent upon its surface water hydrology. T. 276. Of the 21 species of plants living within the Preserve that are listed by the federal government or the State of Florida as endangered, threatened or species of special concern, four species are wetland species. T. 268, 358, 359, 360. Twenty species of animals living on the Preserve are listed by the federal government or the State of Florida as endangered, threatened or species of special concern. Seventeen of these species are wetland dependent. T. 269. Birds, including a number of species listed as endangered or threatened such as great blue herons, woodstorks, anhingas, limpkins, sandhill cranes and ospreys, use the shrub communities around Alachua Lake, the cypress swamp in the eastern lobe and other areas of the eastern lobe for breeding, nesting, and foraging. T. 269, 270, 271, 277, 364, 365. Several species of migratory ducks overwinter in the central area of the Preserve, particularly in the shrub wetland communities around Alachua Lake. Without the flow of water from Prairie Creek the open water in Alachua Lake would be lost and consequently, the overwintering habitat for the ducks would be lost. T. 240, 270, 518. Immature bald eagles use the eastern lobe wetlands for foraging. T. 270. Additionally, the northern harrier, American kestrel and peragrine falcon use wetlands within the Preserve as foraging habitat. T. 364, 365. Mammals, such as river otters, brown water rat, bobcats, bats and long-tailed weasels, use the wetlands within the Preserve, and the eastern lobe particularly, as breeding, nesting, and/or foraging habitat. Reptiles, such as the American alligator, live in the Preserve. T. 270-271, 375, 377-378; SJ Ex 14. The diversity and abundance of animals living in or using the Preserve is greater in the eastern lobe and central area than the western lobe. T. 273, 274. Different species of birds frequent the western lobe. Typically, species more indicative of a drier terrestrial environment are found in the western lobe. T. 272. If the Prairie Creek flow is diverted from the Preserve, the eastern lobe would be driven towards a drier, terrestrial habitat and the functions of the eastern lobe wetlands would be totally lost. T. 277. The sheetflow across the eastern lobe is a unique feature of the Preserve, and without this sheetflow animals such as the endangered brown water rat would not live there. T. 277. Without the Prairie Creek sheetflow, animals dependent on Alachua Lake and the wetlands, such as the brown water rat and the woodstork, would have to find other areas to live, forage, breed and nest due to the loss of wetlands and open water habitat. T. 277, 518. When the water levels in the Preserve are low and wetlands are lost, the birds that depend on the wetlands for nesting will not nest in the Preserve nor elsewhere. T. 532. The wetland communities within the Preserve require a range of water level fluctuations which includes periods of high water levels, average water levels and low water levels. Wetlands must remain wet long enough to exclude upland plants and to conserve hydric soils, yet sufficiently dry often enough to allow germination of wetland plants and the compaction and oxidation of flocculent sediments. T. 293, 294, 298, 299, 310, 311; SJ Ex 1 pp. 23-25. Periods of high water levels maintain lower swamp and shallow marsh habitats, facilitate the dispersal of the seeds of wetland plants, allow wetland species that normally occur at lower elevations to move up into the forested communities, prevent the encroachment of upland species into the upper wetland area, and advance the transportation of organic matter from uplands to wetlands. Inundation of the floodplain and forested communities provide nesting, spawning, refugia, and foraging habitat for fish and other aquatic organisms. T. 294, 296, 310, 311; SJ Ex 1 pp 23- 25. The frequency, timing and duration of high water levels influence the composition and survival of wetland forests. T. 310, 311; SJ Ex 1 p 23. Periods of average water levels create and maintain organic soils and maintain wetland habitat for wetland dependent wildlife. T. 293, 297; SJ Ex 1 p 25. Periods of low water levels rejuvenate floodplain wetlands by allowing seed germination and growth of wetland plants. Seeds of many wetland plant species require saturated soils without standing water in order to germinate. T. 291, 293, 298, 299; SJ Ex 1 pp 24, 25. Periods of low water levels increase the rate of aerobic microbial breakdown and decomposition of organic sediments, and allows the consolidation and compaction of flocculent organic sediments. The consolidation, compaction and decomposition of flocculent organic sediments improves substrates for fish nesting and seed germination. T. 298, 299; SJ Ex 1 pp 24-25. Upland animals use the wetlands during periods of low water levels for foraging and breeding. T. 298, 299. Three elevation transects were used by District staff to identify the elevations of plant communities on the Preserve and develop environmental criteria for the Preserve floodplain. T. 302, 305-306; SJ Ex 1 pp 26, 27, 31, 60. Ecological criteria were developed by District staff to accommodate the hydroperiod requirements of lake and wetland biota. The ecological criteria consisted of hydrologic duration, i.e. how long an area is flooded; and recurrence intervals, i.e. how often an area is flooded. T. 304, 309; SJ Ex 1 pp 23, 61. Maintaining appropriate hydrologic durations and recurrence intervals for plant communities enables the plant communities to support populations of fish and wildlife. T. 307, 312. The District identified the following five significant water management levels: infrequent high water level, frequent high water level, minimum average water level, frequent low water level, and infrequent low water level. The water management levels characterize zones along the elevation gradient of the Preserve. T. 307, 308; SJ Ex 1 p 61. The five different recurrence intervals and the associated hydrologic durations became the hydrologic criteria used by District staff for the water management levels. T. 312. The District evaluated six water management alternatives for the Preserve: the "existing conditions" alternative which simulated the current morphometry of the Paynes Prairie sub-basin; the "total restoration" alternative, under which all the Prairie Creek flow is restored to Paynes Prairie; the "50/50 management" alternative, under which the inflow capacity at the Camps Canal Culvert is reduced by 50 percent and the outflow capacity at the main structure at Alachua Lake is reduced by 50 percent; the "elevation threshold" alternative, under which when the water level at Newnans lake is at 66 feet NGVD or above and the water level at Orange Lake is at 56 feet NGVD or below, then the inflow structure at Camps Canal Culvert is reduced by 50 percent while the outflow capacity at the main structure is maintained at 100 percent; the "Sweetwater Branch" alternative, under which flow from Prairie Creek is replaced by Sweetwater Branch flow; and the "no restoration" alternative, under which the entire flow from Prairie Creek is diverted to Orange Lake. T. 313, 314; SJ Ex 1 p 119. Based upon the hydrologic durations and recurrence intervals defined by the ecologic criteria, the District determined five water management levels for each water management alternative. SJ Ex 1 p 61. The five water management levels and the associated recurrence intervals and hydrologic durations form a fluctuation management regime. The fluctuation management regime for each water management alternative was evaluated with respect to the existing biological features of the aquatic and wetland communities of the Paynes Prairie sub-basin. SJ Ex 1 pp 61, 124, 125. Under the total restoration alternative the water levels on the Preserve would rise thereby improving the hydrologic regime on the prairie, but the possibility of flooding and damaging U.S. Highway 441 would also increase. The minimum average water level of Orange Lake would decrease by 0.67 feet. T. 331, 333; SJ Ex 1 pp 125-130; SJ Ex 8. The no restoration alternative would not satisfy all the hydrologic criteria. The minimum average water level on the Preserve would decrease by 1.01 feet under this alternative. Under this alternative the acreage inundated by the minimum average water level is reduced by approximately 2,400 acres. Additional wetland acres are lost due to the absence of the Prairie Creek sheetflow across the eastern lobe. The minimum average water level in Orange Lake would increase by 0.16 of a foot. T. 324, 334-336; SJ Ex 1 pp 124, 125, 131; SJ Ex 8. Eliminating the flow of Prairie Creek into Paynes Prairie would be detrimental to the current and future biological conditions on the Preserve. SJ Ex 1 p 131. Under the 50/50 management alternative the average flow from Prairie Creek would be reduced from 45 percent to 22.5 percent and the outflow to Alachua Sink would be reduced by 26 percent. T. 337; SJ Ex 1 p 131. The high water levels and the low water levels increase slightly within the Preserve and Orange Lake under the 50/50 management alternative; however, the residence time of water and the concentration of nutrients, including phosphorous and nitrogen, would increase thereby degrading water quality within the Preserve. T. 338, 340, 341; SJ Ex 1 pp 124, 125, 127, 128, 131, 132; SJ Exs 7 and 8. The reduction of sheetflow from Prairie Creek under the 50/50 management alternative would adversely affect the wetlands in the eastern lobe. SJ Ex 1 p 132. Under the elevation threshold management alternative water levels within the Preserve would decrease. The Preserve would receive less water during some periods of naturally high flows reducing the duration and frequency of inundation in the eastern lobe wetlands and, therefore, negatively impacting wildlife dependent upon seasonal high flows. T. 344; SJ Ex 1 p 133; SJ Ex 7. The flow provided by Sweetwater Branch provides approximately 15 percent of the Preserve's average inflow, whereas Prairie Creek provides approximately 50 percent of the Preserve's average inflow. T. 346. Sweetwater Branch is more or less confined to a channel and discharges into Alachua Sink bypassing the Preserve and its eastern lobe. T. 347. Under the Sweetwater Branch alternative the eastern lobe would be deprived of the sheetflow essential to the maintenance of wetlands and the wildlife in the eastern lobe. The eastern lobe would dry out and the plant communities would change to old field or wet prairie. The functions of the plant communities to wildlife would also change under this alternative. T. 347. The Sweetwater Branch alternative would not support fish and wildlife in the eastern lobe of the Preserve. T. 347. The water quality of Sweetwater Branch is poor. Sweetwater Branch has higher concentrations of nitrogen and phosphorous than Prairie Creek. If the nutrient-rich Sweetwater Branch water was diverted onto the Preserve the types and abundances of vegetative communities would change from native vegetation to monocultures of nuisance vegetation that thrive in nutrient-rich environments. T. 346-349; SJ Ex 1 pp 133-134. The existing conditions alternative provides over the long term an average of approximately 45 percent of the Prairie Creek flow by gravity flow through the Camps Canal Culvert to the Preserve. T. 355, 356; SJ Ex 1 p 121. Under the existing conditions alternative, the five hydrologic criteria for both the Preserve and Orange Lake are met and the water level elevations meet the desired recurrence intervals and hydrologic durations. T. 324, 350, 351. The fluctuation management regime provided by the existing conditions alternative partially restores sheetflow from Prairie Creek to the Preserve in sufficient, but fluctuating, water quantities necessary to maintain habitat for fish and wildlife within the eastern lobe. T. 350, 351. It is essential for the protection of the fish and wildlife that utilize and depend upon the Preserve to maintain the flow of Prairie Creek into the Preserve. T. 351, 517. The Preserve needs flow from Prairie Creek in volumes reserved by the proposed rule to protect its fish and wildlife. T. 351. The management levels established by the environmental criteria used for each of the water bodies in the basin will continue to be met in Orange Lake with an average of 45 percent of Prairie Creek flow going to the Preserve and 55 percent going to Orange Lake. T. 432, SJ Ex 1 pp 127, 134, 146. Based upon the substantial and uncontroverted evidence in this case, it is concluded that the water reserved by the Reservation Rule is required for the protection of fish and wildlife of the Preserve.
