This is the fourth round of judicial activity involving a six-inch long finch-billed bird called palila, found only on the slopes of Mauna Kea on the Island of Hawaii.

As an endangered species under the Endangered Species Act ("Act"), 16 U.S.C. Secs. 1531-43 (1982), the bird (Loxioides bailleui), a member of the Hawaiian honeycreeper family, also has legal status and wings its way into federal court as a plaintiff in its own right. The Palila (which has earned the right to be capitalized since it is a party to this proceeding) is represented by attorneys for the Sierra Club, Audubon Society, and other environmental parties who obtained an order directing the Hawaii Department of Land and Natural Resources ("Department") to remove mouflon sheep from its critical habitat. Sports hunters, represented by the Hawaii Rifle Association, among others, had intervened to dispute the contention that the Palila was "harmed" by the presence of mouflon sheep. Hence, these appeals. But, first, some history.

In 1978 the Sierra Club and others brought an action under the Act on behalf of the Palila, claiming that the Department's practice of maintaining feral goats and sheep (animals that originally were domesticated but were allowed to run wild) in the Palila's critical habitat1 constituted an unlawful "taking" under the Act. The district court agreed and ordered the Department to remove the animals because it found that the goats and sheep destroyed the mamane-naio woodlands upon which the Palila depend.2 Palila v. Hawaii Dept. of Land & Natural Resources ("Palila I"), 471 F.Supp.985 (D.Haw.1979). This court affirmed. Palila v. Hawaii Dept. Land & Natural Resources ("Palila II"), 639 F.2d 495 (9th Cir.1981).

In 1984 the Sierra Club reopened the 1978 proceeding by moving to amend its original complaint to add mouflon sheep as destructive animals to be removed from the Palila's habitat. The mouflon sheep had been introduced by the Department between 1962 and 1966 for the enjoyment of sport hunters. Apparently, they had not been the target of the original complaint because research into their effect upon the Palila's habitat had not been completed. The mouflon sheep, like the feral sheep and goats before them, feed on the mamane trees.

In November 1986 the district court ruled in favor of the Sierra Club. Palila v. Hawaii Dept. of Land & Natural Resources ("Palila III"), 649 F.Supp.1070 (D.Haw.1986). It found that presence of mouflon sheep "harmed" the Palila within the meaning of 50 C.F.R. Sec. 17.3's definition of "harm" in two ways:3 (1) the eating habits of the sheep destroyed the mamane woodland and thus caused habitat degradation that could result in extinction; (2) were the mouflon to continue eating the mamane, the woodland would not regenerate and the Palila population would not recover to a point where it could be removed from the Endangered Species list.

The Department and intervenors filed timely appeals. We granted the United States amicus curiae status to represent the view of the Secretary that Judge King's order should be affirmed, but for reasons different than those stated in his opinion.

* The Department argues that the district court construed the definition of "harm" in 50 C.F.R. Sec. 17.3 too broadly.4 The scope of the definition of harm is important because it in part sets the limit on what acts or omissions violate the Act's prohibition against "taking" an endangered species.5

In making this argument, the Department suggests dichotomy between "actual" and "potential" harm. The Department believes that actual harm only includes those acts which result in the immediate destruction of the Palila's food sources; all other acts are "potential" harm no matter how clear the causal link and beyond the reach of the Act. Thus, the Department challenges the district court's finding that habitat destruction which could drive the Palila to extinction constitutes "harm."

We inquire whether the district court's interpretation is consistent with the Secretary's construction of the statute since he is charged with enforcing the Act, and entitled to deference if his regulation is reasonable and not in conflict with the intent of Congress. See United States v. Riverside Bayview Homes, Inc., 474 U.S. 121, 131, 106 S. Ct. 455, 461, 88 L. Ed. 2d 419 (1985).

While promulgating a revised definition of harm, the Secretary noted that harm includes not only direct physical injury, but also injury caused by impairment of essential behavior patterns via habitat modification that can have significant and permanent effects on a listed species. 46 Fed.Reg. 54748, 54750 (1981) (codified at 50 C.F.R. Sec. 17.3). Moreover, in that same promulgation notice, the Secretary let stand the district court's construction of harm in Palila I. Id. at 54749-50. In Palila I, the district court construed harm to include habitat destruction that could result in the extinction of the Palila--exactly the same type of injury at issue here. See generally Palila I, 471 F. Supp. at 985. We conclude that the district court's inclusion within the definition of "harm" of habitat destruction that could drive the Palila to extinction falls within the Secretary's interpretation.

The Secretary's inclusion of habitat destruction that could result in extinction follows the plain language of the statute because it serves the overall purpose of the Act, which is "to provide a means whereby the ecosystems upon which endangered species and threatened species depend may be conserved...." 16 U.S.C. Sec. 1531(b). The definition serves the overall purpose of the Act since it conserves the Palila's threatened ecosystem (the mamane-naio woodland).

The Secretary's construction of harm is also consistent with the policy of Congress evidenced by the legislative history. For example, in the Senate Report on the Act: " 'Take' is defined in ... the broadest possible manner to include every conceivable way in which a person can 'take' or attempt to 'take' any fish or wildlife." S.Rep. No. 307, 93d Cong., 1st Sess. (1973), reprinted in 1973 U.S. Code Cong. & Admin. News 2989, 2995. The House Report said that the "harassment" form of taking would "allow, for example, the Secretary to regulate or prohibit the activities of birdwatchers where the effect of those activities might disturb the birds and make it difficult for them to hatch or raise their young." H.R.Rep. No. 412, 93d Cong., 1st Sess. (1973), 1973 U.S.Code Cong. & Admin.News 2989, reprinted in 4 House Miscellaneous Reports on Public Bills, 93d Cong., 1st Sess. 11 (1973). If the "harassment" form of taking includes activities so remote from actual injury to the bird as birdwatching, then the "harm" form of taking should include more direct activities, such as the mouflon sheep preventing any mamane from growing to maturity.6

The Department contends that the district court erred when it found an unlawful "taking" within the meaning of section 9 of the Act. (Section 9--codified as 16 U.S.C. Sec. 1538--lists the conduct prohibited by the Act). The Department argues that no taking exists because the evidence shows that (1) a huntable number of sheep (a flock large enough to sustain sports hunting) could co-exist with the Palila; and (2) the Palila are doing poorly because of the recently removed feral sheep and goats, not the mouflon sheep. Our review is for clear error. Oregon Envtl. Council v. Kunzman, 817 F.2d 484, 493 (9th Cir.1987).

The Department's witnesses conceded that a large number of mouflon sheep in one area could significantly damage the mamane-naio woodlands and thereby drive the Palila to extinction. However, these witnesses maintained that a huntable number of mouflon sheep could co-exist with the Palila. In support of its co-existence thesis, the Department makes four arguments. First, since the removal of the feral sheep and goats, the mamane-naio woodland has regenerated. This regeneration will support both the mouflon sheep and the Palila. Second, the Department has begun a number of regeneration projects (replanting, fertilizing, etc.). Third, the mouflon sheep would not cause significant degradation if the Department controlled their density. Fourth, the population of the Palila has increased since January 1985.

The Sierra Club's witnesses controverted the Department's thesis of co-existence. First, although regeneration (new mamane seedlings and sprouts) has occurred in many areas, it takes twenty-five years for the mamane seedlings and sprouts to become mature trees capable of providing food and shelter for the Palila. However, for the first ten to fifteen years of this growth period, the mouflon sheep can kill the mamane trees and no significant regeneration would occur, at least not sufficient to sustain the Palila unless the trees survive to twenty-five years of age. Second, the Sierra Club's witnesses showed that the Department's additional programs as an alternative to removal of the mouflon sheep would not work. Third, they disagreed with the premise that the mouflon sheep population could co-exist with the Palila if the Department controlled their density. Fourth, the Sierra Club witnesses stated that the Palila's population, despite short-term fluctuations, has been static over the long term.

The Sierra Club witnesses put forth their own thesis: Because the grazing and browsing habits of the mouflon sheep destroy the mamane woodland upon which the Palila depend entirely for their existence, the sheep must be removed. This thesis received the support of one of the state's witnesses. This witness conceded that he believes that the mouflon sheep must be removed to ensure the survival of the Palila.

The Sierra Club's witnesses are not contradicted by the documentary evidence (i.e., studies of the Palila, mouflon sheep, etc.), and the Sierra Club witnesses advanced a coherent and plausible thesis. On the issue of co-existence, then, the district court's decision to accept the Sierra Club's witnesses' testimony as more credible cannot be clearly erroneous.

B. Feral Sheep and Goats Versus Mouflon Sheep

The Department's witnesses asserted that there had been significant regeneration wherever the feral animals had been removed. The Sierra Club's witnesses agreed, but they went on to argue that where mouflon sheep have appeared, no significant regeneration has occurred.

On the question of which animals--the feral sheep and goats or the mouflon--damage the mamane, the district court again gave more credibility to the Sierra Club's witnesses; this preference cannot be clearly erroneous where the Sierra Club's witnesses were not contradicted by documentary evidence. Indeed, the testimony given by the Sierra Club witnesses--noticeable regeneration has occurred only where the feral animals have been removed and no mouflon sheep have appeared--is both plausible and consistent.

We affirm the district court's finding that the Department's permitting mouflon sheep in the area constitutes a "taking" of the Palila's habitat. The district court made its findings based on the testimony of the Sierra Club witnesses, which was not contradicted by extrinsic evidence. Therefore, the district court's findings should not be held clearly erroneous. See Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 575, 105 S. Ct. 1504, 1512, 84 L. Ed. 2d 518 (1985)("When a trial judge's finding is based on his decision to credit the testimony of one of two or more witnesses, each of whom has told a coherent and facially plausible story that is not contradicted by extrinsic evidence, that finding, if not internally inconsistent, can virtually never be clear error").

Under this resolution of the appeal, we do not reach the issue of whether harm includes habitat degradation that merely retards recovery. The district court's (and the Secretary's) interpretation of harm as including habitat destruction that could result in extinction, and findings to that effect are enough to sustain an order for the removal of the mouflon sheep.7

The district court's finding of habitat degradation that could result in extinction constitutes "harm." The district court's finding of a "taking" was not clearly erroneous. We do not reach the issue of whether the district court properly found that harm included habitat degradation that prevents recovery of an endangered species.

AFFIRMED.