The Kalamazoo River Study Group ("KRSG"), the plaintiff, appeals the district court's grant of summary judgment in favor of Benteler Industries, Inc. ("Benteler"), the defendant, in this response cost recovery action brought pursuant to the Comprehensive Environmental Response, Compensation and Liability Act ("CERCLA"), 42 U.S.C. § 9601 et seq., and the Michigan Natural Resources and Environmental Protection Act ("NREPA"), M.C.L.A. § 324.20101, et seq.

In 1990, after nearly 20 years of investigating polychorinated biphenyl (PCB) contamination in the Kalamazoo River, the Michigan Department of Natural Resources ("MDNR")1 determined that a three-mile portion of Portage Creek from Cork Street to the Kalamazoo River, and a 35-mile portion of the Kalamazoo River from this confluence downstream to the Allegan City Dam (the "Site") were heavily concentrated with PCBs. Consequently, the MDNR listed the Site as an environmental contamination site under the Michigan Environmental Response Act, M.C.L.A §§ 299.601 et seq. (repealed and replaced by NREPA, M.C.L.A. § 324.20101, et. seq.) and the United States Environmental Protection Agency ("EPA") listed the Site on the National Priority List ("NPL") as a Superfund Site pursuant to CERCLA § 105, 42 U.S.C. § 9605.2 The MDNR and the EPA authorized the MDNR to conduct an Endangerment/Risk Assessment (E/RA) for the Site.

Following the E/RA, MDNR identified three paper mills--HM Holdings, Inc., Georgia-Pacific Corporation and Simpson Plainwell Paper Company--with facilities located on or near the Kalamazoo River as potentially responsible parties ("PRPs") for the PCB contamination. All three companies entered into an Administrative Order by Consent ("AOC") which required them to perform a Remedial Investigation and Feasibility Study ("RI/FS") at the Site. Subsequently, James River Company, which voluntarily agreed to pay a portion of the RI/FS costs, joined with the other three companies to form an unincorporated association called the Kalamazoo River Study Group ("KRSG").

Benteler is engaged in the manufacture of automotive parts. It purchased the manufacturing facility at issue in this case in 1989.3 The facility is located about 3200 feet north of the far southeastern boundary of Morrow Lake, which is an impoundment of the Kalamazoo River located upstream from the Site. Located on Benteler's property is a 3200-foot drainage ditch that runs north to south toward Morrow Lake. At the north end of the ditch next to the Benteler parking lot is a headwall with a drain that discharges storm water run-off and floor drainage from the plant into the ditch.4 This ditch is at the center of the controversy.

When Benteler purchased the facility, it contained transformers and capacitors that contained PCBs. In 1989, while Benteler was preparing the facility for use, one of the transformers was damaged and leaked PCBs. In the process of responding to this leak, Benteler discovered PCBs throughout the plant, in the drain lines leading toward the ditch, and in the ditch itself near the headwall. In response, Benteler, with the assistance of environmental consultants, undertook testing and investigating to determine the extent of PCB contamination in the ditch.

In 1993, pursuant to a Remedial Action Work Plan prepared by the environmental consultants and approved by MDNR, Benteler undertook remedial actions to remove PCBs from the ditch, which included the excavation of approximately 800 cubic yards of soil from the first 600 feet of the ditch and the remediation of the storm sewer lines that emptied into the ditch. All remediation activities were completed by October 11, 1993. Further verification samples were taken in 1994 and 1996. On October 16, 1996, the MDNR issued Benteler a clean closure letter.

In 1995, KRSG filed the underlying action against eight corporations, including Benteler, with facilities on or near the Kalamazoo River, alleging that they contributed to the PCB contamination at the Site and should be responsible for some of the cleanup costs. KRSG alleged that Benteler's facility discharged oil-containing wastewater, storm water and cooling water from its plant, through the plant's drainage pipes and into this drainage ditch. The discharged water, according to KRSG, then transported PCB-contaminated soil and sediment from the ditch and released it to Morrow Lake, then to the Kalamazoo River and downstream to the Site. Benteler moved for summary judgment, arguing that KRSG could not prove that Benteler had contributed PCBs to the Site. In its response, KRSG included the affidavit of Dr. Mark Brown who concluded that Benteler had in fact discharged water and PCBs to the Kalamazoo River and thus downstream to the Site. The district court granted summary judgment in favor of Benteler, finding that Dr. Brown's conclusion rested on an unreliable foundation.STANDARD OF REVIEW

This court reviews the district court's grant of summary judgment de novo. Morris v. Crete Carrier Corp., 105 F.3d 279, 280 (6th Cir.1997). Summary judgment is appropriate only when there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law. Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986). The initial burden is on the moving party to demonstrate that an essential element of the non-moving party's case is lacking. Id. The burden then shifts to the non-moving party to come forward with specific facts, supported by evidence in the record, upon which a reasonable jury could return a verdict for the non-moving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S. Ct. 2505, 91 L. Ed. 2d 202 (1986).

This court must review the evidence in the light most favorable to the non-moving party. Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986). In addition, "[c]redibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of judge. The evidence of the non-movant is to be believed, and all justifiable inferences are to be drawn in his favor." Schaffer v. A.O. Smith Harvestore Products, Inc., 74 F.3d 722, 727 (6th Cir.1996) (quoting Anderson, 477 U.S. at 255, 106 S. Ct. 2505). Nevertheless, the non-moving party must present more than a mere scintilla of evidence in support of her position; the non-moving party must present evidence on which the jury could reasonably find for the plaintiff. Hartsel v. Keys, 87 F.3d 795, 799 (6th Cir.1996) (citing Anderson, 477 U.S. at 252, 106 S. Ct. 2505).

The purpose of CERCLA is "to facilitate the prompt cleanup of hazardous waste sites by placing the ultimate financial responsibility for cleanup on those responsible for hazardous wastes." United States v. R.W. Meyer, 889 F.2d 1497, 1500 (6th Cir.1989). In order to establish a prima facie case of CERCLA liability against Benteler, KRSG must establish that: (1) there was a release or threatened release of a hazardous substance; (2) the site of the release or threatened release is a "facility" as that term is defined the statute; (3) the release or threatened release has caused KRSG to incur response costs; and (4) Benteler is among a statutorily-defined group of persons, which includes the owner or operator of an owner or operator of a facility. See CERCLA § 107, 42 U.S.C. § 9607(a). Benteler only challenges KRSG's ability to establish the third element, causation. In a "two-site" case such as this, where hazardous substances are released at one site and allegedly travel to a second site, in order to make out a prima facie case, the plaintiff must establish a causal connection between the defendant's release of hazardous substances and the plaintiff's response costs incurred in cleaning them up. Thomas v. FAG Bearings Corp., 846 F. Supp. 1382, 1387 (W.D.Mo.1994).

Benteler does not deny that there has been PCB contamination in its facility and in the ditch during its ownership of the facility; however, Benteler contends that the conditions of the ditch are such that PCBs from its facility have not and could not have migrated all the way down the ditch to Morrow Lake and into the Kalamazoo River during its ownership of the facility. Benteler has offered a variety of evidence in support of this position.

Benteler has presented evidence that the PCBs in the ditch were confined to the northern part of the ditch near the headwall. In conjunction with the Remedial Action Work Plan prepared by Benteler's environmental consultants in 1993, the MDNR only required remediation activities in the ditch to a distance of approximately 600 feet from the headwall. MDNR did not require Benteler to remediate the remaining 2600 feet of the ditch because there were either no PCBs detected or no PCBs detected in excess of Michigan's most stringent clean-up criteria. The soil studies from 1989, 1990, 1991 and 1993 reveal that the highest concentrations of PCBs were within the first 325 feet of the ditch (ranging from 120 mg/kg to 6.18 mg/kg). Between 325-600 feet, the highest concentration was 6.6 mg/kg. Beyond 600 feet the concentration of PCBs declined significantly. At 625 feet the concentration was 0.9 mg/kg. From 720 to 1250 feet the highest concentration was 0.112 mg/kg. At 1400, 1425, 1450, and 1596 feet, there was no detection of PCBs.5

Benteler conducted further testing in August 1996. In addition to samples taken at 500 and 1000 feet, samples were taken further down the ditch at 1500, 2000, 2500 and 2700 feet from the headwall. No PCBs were detected at any of these locations with the exception of the 1500-foot point where there was a PCB reading of 0.43 mg/kg. When the same sample from the 1500-foot point was retested, it produced a non-detectable result consistent with the other samples. On the basis of this information, Benteler claims that the ditch contains no scientifically detectable PCB levels beyond 1500 feet, or stated another way, the remaining 1700 feet of the ditch do not contain any PCBs.

Additionally, Benteler has provided hydrogeological evidence that demonstrates that the ditch could not have acted as a conduit of PCBs to the Kalamazoo River. Due to the depth of the water table and the permeability of soil in this area, Benteler's experts contend that any water discharged to the ditch during Benteler's ownership would have been absorbed within 600-800 feet of the headwall. Specifically, in August 1996, one of Benteler's experts, Lawrence M. Austin, a Certified Professional Geologist (CPG), performed an analysis of whether, during Benteler's ownership of the facility, a sufficient amount of water was or could be discharged to allow PCBs to be carried down the ditch to Morrow Lake. The analysis of streamflow was completed using an approximation of 100-400 gallons per day of non-contact cooling water discharged by Benteler into the ditch since 1989, and a 3.5-inch rainfall event in a 24-hour period over the entire parking lot and building area. Assuming all of the water runs off of the paved and roofed surfaces, this would generate, according to Austin, a 200,000 cubic feet flow in a 24-hour period. The results of Austin's modeling show the stream flow in the ditch declines to a no-flow condition within 600 to 800 feet of the north headwall due to absorption into the soil.6 Consequently, Austin concluded that the ditch lacked sufficient waterflow to carry PCBs the length of the ditch and into Morrow Lake.

To further demonstrate that the ditch did not act as a passageway for PCBs, Benteler has presented evidence, through the affidavit of Dr. Lawrence Halfen, that the soil and vegetation in the ditch is not consistent with a continuous flowing stream, an intermittent stream, or even with occasional discharges of sufficient magnitude to carry water down the ditch to Morrow Lake.

On the basis of this evidence, Benteler has adequately demonstrated that any PCBs from its property could not have migrated down the ditch to Lake Morrow and the Kalamazoo River. Therefore, the burden shifts to KRSG to come forward with sufficient facts that would allow a reasonable jury to find that Benteler caused PCB contamination at the Site. Anderson, 477 U.S. at 248, 106 S. Ct. 2505.

As previously stated, KRSG relies, in large part, on the affidavit of its expert, Dr. Brown, who opines that "[a] logical interpretation of the facts leads to the conclusion that the ditch has discharged water and PCBs to the Kalamazoo River and, therefore, PCBs from Benteler's facility are within the borders of the National Priorities List site." His conclusion is based upon four categories of information: a U.S. Geological Society Survey ("USGS") map, soil samples showing detectable PCB levels at 18 feet and 3 feet from Morrow Lake, stormwater and wastewater discharge documents prepared by Benteler's own environmental consultants showing Morrow Lake as the discharge point for the ditch, and challenges to the conclusions of Benteler's experts.

The USGS map shows a solid blue line running from Benteler's facility through the drainage ditch to Morrow Lake. The line, according to Dr. Brown, indicates the ditch has been a continuously running waterway flowing to Morrow Lake. He opined that the USGS map established that water from Benteler's facility flowed to Morrow Lake via the ditch.7

Dr. Brown also relies heavily on analytical test results to support his opinion that PCBs were discharged from Benteler's facility to Morrow Lake. He points to two soil samples taken by KRSG's consultants in 1994 which identified PCBs in the ditch eighteen feet and three feet from the shore of Morrow Lake, with concentration levels of 0.93 mg/kg and 0.43 mg/kg, respectively. Dr. Brown opined that the higher concentration of PCBs detected at 18 feet and the lower PCB concentration at 3 feet, coupled with the detection of PCB levels at 1700 feet from Morrow Lake,8 supported his conclusion that the ditch was the source of PCBs in Morrow Lake.9

Further evidence relied upon by Brown include documents in which Benteler's environmental consultants allegedly acknowledge that wastewater and stormwater runoff from Benteler's facility discharges to the drainage ditch and flows into Morrow Lake. One such document is a 1989 report prepared by an environmental consultant at Benteler's request. That report states that the drainage ditch "ultimately discharges into the Kalamazoo River." Of particular importance is a draft 1993 Stormwater Discharge Permit Application ("Application") prepared by an environmental consulting firm at Benteler's request. In the Application, the consultant stated that "Morrow Lake (part of the Kalamazoo River) is the discharge point for the drainage ditch leaving Benteler property." Moreover, the same consulting firm that prepared the 1993 Application stated in a 1995 report that "the drainage ditch during periods of high flow can discharge to Morrow Lake...."10

Brown also emphasizes two letters sent by MDNR to Benteler in the early 1990's. Both letters address discharges from the ditch to the Kalamazoo River. In a September 1990 letter the MDNR demanded that "[a]ll discharges to the ditch, including stormwater, must be ceased. This is due to the possibility that large stormwater flows could transport PCBs in the ditch to the Kalamazoo River."11 In a June 1991 letter, MDNR again demanded that Benteler "cease discharges to the ditch due to the possibility that contaminants could spread to the Kalamazoo River...." KRSG claims that Dr. Brown reasonably relied on these documents, and representations contained therein, to justify his opinion that wastewater discharged to Morrow Lake and the Kalamazoo River via the drainage ditch.12

In the course of reaching his conclusion that Benteler was a source of PCB contamination, Dr. Brown challenges the conclusions of Benteler's experts. Of particular importance is his criticism of Austin's streamflow study in which Austin concluded that any water flow in the ditch would be absorbed into the soil within 600-800 feet of the headwall.13 Brown's critique is based on the "Application" discussed above. In addition to the representations contained therein, it also contained an analysis of non-storm water discharges to the ditch from the Benteler facility. The analysis was based on visual inspection at the time the consultant visited the facility. It identified two non-storm water flows:

(1) Non-contact cooling waters from two unit man coolers which discharge approximately 100 gpm (gallons per minute) during cooling months; and

(2) Non-contact cooling water from an air compressor which discharges approximately 50 gpm year round.

Using these figures, Dr. Brown calculated that Benteler was discharging non-storm water between 72,000 and 210,000 gallons per day to the ditch. These figures stand in stark contrast to the figures used by Mr. Austin in his streamflow analysis in which he assumed non-storm water discharge between 100 and 400 gallons per day.14 Moreover, in addition to insufficient water flow, Brown contends that Austin underestimated the discharge rate, the rainfall, and the drainage area served by the ditch. Brown also contends that Austin improperly assumed that the runoff would be evenly distributed over a 24-hour period and that he failed to consider that the groundwater table will rise during periods of high runoff.

Finally, Brown challenges Dr. Halfen's opinions on the basis that his soil samples were taken after the remediation of the ditch; he failed to consider the level of non-storm water discharges; he improperly attempted to write off the PCB finding at the 1500-foot mark as anomalous; he ignored the underestimating bias associated with his "composite" samples; and he failed to recognize that the lack of obligate hydrophotic plant species is not inconsistent with the function of the ditch as an intermittent stream and conveyance for stormwater from the facility to Morrow Lake.

KRSG asserts that Dr. Brown's Affidavit has a strong factual basis and creates, at a minimum, a classic "battle of the experts" concerning whether PCBs from Benteler's property contributed to PCB contamination at the Site. Thus, KRSG contends that the district court overstepped its discretionary bounds on a motion for summary judgment by resolving credibility judgements in favor of Benteler's experts. KRSG is correct when it argues that credibility determinations, the weighing of evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of the judge, see Anderson, 477 U.S. at 255, 106 S. Ct. 2505; however, contrary to KRSG's contention, the district court did not "choose sides" to resolve a battle of the experts. Instead, the court focused on the factual underpinnings of Dr. Brown's conclusions. Citing Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993), the court stated that it "may, indeed must, look beyond the conclusions [of the experts] to determine whether the expert testimony rests on a reliable foundation." When the court did so, it found that Dr. Brown's conclusion was based on "speculation, conjecture, and possibility" and that "the inadequate factual basis makes Dr. Brown's affidavit scientifically unreliable." Therefore, the court held:

KRSG has created, at most, a question of fact as to whether or not there was a possibility that water flowed all the way down the ditch to Morrow Lake.... KRSG's entire theory of liability on behalf of Benteler is based upon the assumption that water flowed down the ditch to Morrow Lake. This assumption, however, is based solely on speculation and possibility. The existence of a possibility does not create a material issue of fact for trial because KRSG bears the burden of proof to show that Benteler did contribute to PCBs in the Kalamazoo River, not that it is possible that it might have contributed to the PCBs.

We agree. This court has stated that "although judges should respect scientific opinion and recognize their own limited scientific knowledge, nevertheless courts have a duty to inspect the reasoning of qualified scientific experts to determine whether a case should go to the jury." Turpin v. Merrell Dow Pharmaceuticals, Inc., 959 F.2d 1349, 1350 (6th Cir.1992). Taken in the light most favorable to KRSG, the scientific evidence that provides the foundation for Dr. Brown's opinion on causation is not sufficient to allow a jury to find that it is more probable than not that Benteler caused PCB contamination at the Site.

In particular, KRSG alleges that there was sufficient waterflow to carry PCBs through the entire 3200-foot length of the ditch and into Morrow Lake. But this allegation is based on nothing more than Dr. Brown's critique of Austin's streamflow analysis, in which he criticized Austin's model for underestimating the number of gallons of non-contact cooling water per day flowing into the ditch. Yet he fails to produce any affirmative evidence that this increased amount of water is sufficient to carry PCBs down the ditch and into the lake. Moreover, assuming KRSG's contention regarding waterflow to be true, one would reasonably expect to find detectable PCB levels throughout the length of the ditch. Yet, it is undisputed that there exists a nearly 1700-foot gap where no PCBs have been detected. KRSG relies on the two samples containing detectable PCB levels at 18 feet and 3 feet from the shore of Morrow Lake without producing any affirmative evidence to explain the absence of PCBs in the preceding 1700 feet of the ditch. In Turpin, this court stated that "[t]he analytical gap between the evidence presented and the inferences to be drawn ... is too wide. Under such circumstances, a jury should not be asked to speculate on the issue of causation." Id. at 1360-61 (emphasis added). Similarly, in this case, the evidence presented leaves a "gap" that is simply too wide to allow a jury to speculate on the ultimate issue of causation. We therefore affirm the district court's grant of summary judgment in favor of Benteler.

AFFIRMED.