Full opinion text
MEMORANDUM OPINION AND ORDER PARKER, Chief Judge. This is an appeal by Plaintiffs from administrative action by a federal agency, the United States Fish and Wildlife Service (“FWS”), in issuing a final Biological Opinion (“BO”) on two endangered species found in the middle Rio Grande basin, the Rio Grande Silvery Minnow (hybognathus amarus) and the Southwestern Willow Flycatcher (empidonax traillii extimus). In preparing to issue its Biological Opinion, FWS consulted with two federal “action” agencies, the Bureau of Reclamation (“BOR”), and the United States Army Corps of Engineers (“the Corps”), under the Endangered Species Act (“ESA”), 16 U.S.C. § 1531 et seq. Plaintiffs challenge the BO itself, as well as the procedure leading up it its issuance. The parties submitted detailed legal memoranda on the merits of Plaintiffs’ claims, and a hearing on the merits was held on November 19, 2001. At the hearing the various parties presented oral arguments. I have thoroughly considered the documents in the administrative record, the arguments of counsel, and the relevant case law. I conclude that the Biological Opinion should be affirmed. I. Introduction. At that time the original Complaint was filed, the Plaintiffs primarily challenged the failure by the two federal action agencies, BOR and the Corps, to consult fully with FWS on a broad range of allegedly discretionary water alternatives in order to save the Rio Grande Silvery Minnow and the Southwestern Willow Flycatcher from extinction. Plaintiffs are the two listed species and several non-profit environmental and conservation organizations acting on behalf of the listed species. In the original Complaint the Defendants were the two federal action agencies, BOR and the Corps, together with various agency officials sued in their official capacities. Several parties with interests in the water in the middle Rio Grande basin were allowed to intervene. These include the Middle Rio Grande Conservancy District (“MRGCD”), the City of Albuquerque, the State of New Mexico, and the Rio Chama Acequia Association. No claims are made against the Intervenors, but MRGCD has filed a cross-claim under the Quiet Title Act against the United States, asserting title to some of the property within the MRGCD. On June 29, 2001, shortly before a scheduled hearing on the merits before this Court, FWS issued its final BO. At the hearing on July 2, 2001, recognizing the legal significance of the issuance of the final BO, Plaintiffs requested and were granted leave to file a Second Amended Complaint to reflect the effects of the issuance of the June 29 final BO on their claims. The thrust of the Second Amended Complaint is to attack the final BO under the applicable standard of this Court’s review of final agency action, while continuing to maintain Plaintiffs’ original challenge to the scope of the consultation. The Second Amended Complaint adds as Defendants the United States, FWS, and the Secretary of Interior. The Second Amended Complaint alleges procedural and substantive violations of Section 7(a)(2) of the ESA (Count 1), and seeks review of the BO under the Administrative Procedures Act, 5 U.S.C. § 701 et seq. (“APA”) for these violations (Count 6). The alleged procedural violation is the failure of BOR and the Corps to consult fully with FWS over all their discretionary agency actions. The substantive violation is for allegedly causing jeopardy to the endangered species. Plaintiffs allege that FWS violated the APA when it issued the BO with a Reasonable and Prudent Alternative (“RPA”) that allows significant river drying, and that FWS failed to use the best available scientific data as required by the ESA. The Second Amended Complaint also contains a cause of action against the federal agencies, under Section 7(a)(1) of the ESA (Count 2), for failure to conserve the listed species. Plaintiffs dropped their claim under Section 7(d) of the ESA (Count 3), because that section applies only during ongoing consultation, and they also dropped their claim against BOR for failure to comply with NEPA (Count 5). The Second Amended Complaint alleges a violation of Section 9 of the ESA for “take” of the species (Count 4). This claim received some attention during the briefing, but Plaintiffs expressly acknowledged at the November 19, 2001 hearing that the take claim is not ripe without further factual development. Finally, Count 7 of the Second Amended Complaint involves Plaintiffs’ challenges to FWS’s issuance of the Incidental Take Statement (“ITS”). The ITS purports to shield all federal and non-federal water users in the middle Rio Grande basin from “take” liability under Section 9 of the ESA. II. Standard of Review. This Court sits as an appellate court with respect to reviewing Plaintiffs’ claims regarding final agency action. Olenhouse v. Commodity Credit Corp., 42 F.3d 1560, 1580 (10th Cir.1994). Biological opinions issued under section 7 of the ESA are reviewed under Administrative Procedures Act (“APA”) standards to determine whether the opinion is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). See Bennett v. Spear, 520 U.S. 154, 178, 117 S.Ct. 1154, 137 L.Ed.2d 281 (1997) (holding biological opinion was final agency action because it effectively stopped further proceedings by action agency); Wyoming Farm Bureau Federation v. Babbitt, 199 F.3d 1224, 1231 (10th Cir.2000) (review of ESA claims governed by APA); Biodiversity Legal Foundation v. Babbitt, 146 F.3d 1249, 1252 (10th Cir.1998) (in examining whether FWS’s actions violate ESA, court relies on standards of review provided in APA); Pacific Coast Federation of Fishermen’s Ass’ns, Inc. v. National Marine Fisheries Serv., 265 F.3d 1028, 1034 (9th Cir.2001) (same). This deferential standard is designed to ensure that the agency’s decision was based on relevant factors and that it contained no “clear error of judgment.” Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 378, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989); Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971). A BO is arbitrary and capricious under the APA if the agency has “entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Pacific Coast Federation of Fishermen’s Ass’ns, Inc. v. National Marine Fisheries Serv., 265 F.3d at 1034 (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983)). “A biological opinion may also be invalid if it fails to use the best available scientific information as required by 16 U.S.C. § 1536(a)(2).” Id., 265 F.3d at 1034. In deciding disputes that involve primarily issues of fact requiring a high level of technical expertise, the court “must defer to ‘the informed discretion of the responsible federal agencies.’ ” Marsh v. Oregon Natural Resources Council, 490 U.S. at 377, 109 S.Ct. 1851 (quoting Kleppe v. Sierra Club, 427 U.S. 390, 412, 96 S.Ct. 2718, 49 L.Ed.2d 576 (1976)). III. Background. A. Endangered Species Act (“ESA”). In 1973, Congress enacted the ESA in response to growing concern about the extinction of various species “as a consequence of economic growth and development untempered by adequate concern and conservation.” 16 U.S.C. § 1531(a)(1). The Supreme Court stated: “the plain intent of Congress in enacting this statute was to halt and reverse the trend toward species extinction, whatever the cost.” Tennessee Valley Authority v. Hill, 437 U.S. 153, 184, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978). Listing a species as endangered triggers a consultation procedure required by Section 7 of the ESA. Federal agencies like BOR and the Corps must consult with FWS to “insure that any action authorized, funded, or carried out by such agency ... is not likely to jeopardize the continued existence” of the species. 16 U.S.C. § 1536(a)(2). This subsection further provides that “each agency shall use the best scientific and commercial data available.” Id. After FWS reviews “all relevant information,” the consultation process culminates in the issuance of a Biological Opinion (“BO”). 50 C.F.R. § 402.14(g)(1) and (4). The BO contains FWS’s opinion as to whether an agency action is likely to jeopardize the continued existence of an endangered species. 50 C.F.R. §§ 402(g)(4) and 402.14(h)(3); see also 16 U.S.C. § 1536(b)(3)(A). When a final BO concludes that the agency action is likely to jeopardize a listed species, the BO must include reasonable and prudent alternatives (“RPAs”) that will not jeopardize the species, to the extent that the action agencies are able to implement the alternatives. 50 C.F.R. § 402.14(h)(3); see also 16 U.S.C. §§ 1536(b)(3)(A) and 1536(b)(4). If FWS determines that the agency action, or implementation of RPAs, will result only in an “incidental take” of a species, FWS must indicate “reasonable and prudent measures” which will minimize the impact on the species. 50 C.F.R. § 402.14(i); see also 16 U.S.C. § 1536(b)(4). In addition, FWS must issue an Incidental Take Statement (“ITS”) specifying the conditions under which incidental taking of the species may occur so as not to violate ESA Section 7(a)(2). 16 U.S.C. § 1536(b)(4). The ITS provides immunity from prosecution under Section 9 of the ESA for an incidental take that is made in accordance with the reasonable and prudent measures outlined in the BO. 16 U.S.C. § 1536(o)(2); 50 C.F.R. § 402.14(0(5). B. The silvery minnow. In 1994, the FWS listed the silvery minnow as an endangered species under 16 U.S.C. § 1533 of the ESA. 59 Fed.Reg. 36988 (July 20, 1994); FWS III.A.U- The FWS believed at the time of listing that the silvery minnow was found only in a 170 mile stretch of the Rio Grande from Cochi-ti Dam downstream to the headwaters of Elephant Butte Lake. FWS III.All. This portion of the river includes bank-to-bank diversion structures at Angostura (the first structure downstream from Cochiti Dam), Isleta (the structure immediately downstream from the Angostura Dam and the City of Albuquerque), and San Acacia (downstream from Socorro, New Mexico, and the final diversion structure above Elephant Butte Lake). By 1999, according to BOR, over 95% of the remaining wild silvery minnow population was concentrated in the San Acacia reach. BOR 19I at 71-72; BOR 196 at 52-3. A “reach” is that part of the Rio Grande extending downstream from the diversion dam for which the reach is named to the next significant physical feature of the river. The San Acacia reach spans from the San Acacia diversion dam to Elephant Butte Reservoir. It is only about 60 miles long, and represents a small fraction of the silvery minnow’s historic range. FWS III. A. II. It is also the section of the river that is most prone to drying. BOR 35 at 21; BOR 171 at [5033]. The silvery minnow is a small fish reaching 3.5 inches in length. It is a pelagic spawner, producing nonadhesive, semi-bouyant eggs. BOR 15 at ii; BOR 17 at 559. The eggs drift downstream until the larvae are sufficiently developed to move to the silvery minnow’s preferred shallow-water, low velocity habitats, lined with silt/ sand substrate. BOR 15 at 17; BOR 26 at 12-17; BOR 191 at 73. One source has attributed the concentration of silvery minnows in the San Acacia reach to the minnow’s spawning behavior and its apparent inability to move upstream beyond the San Acacia diversion structure. BOR at 15 at 18. Other sources attribute the relative scarcity of minnows throughout their former range in the Rio Grande to a variety of factors that include stream channelization, regulation of river flow, water diversion for agriculture, competition and predation, as well as dewatering. 59 Fed. Reg. 36988. Still others list as additional factors water quality degradation, introduction of non-native species, BOR 191 at 73, and entrainment of fish and larvae in irrigation facilities, BOR 196 at 88. C. The willow flycatcher. In 1995, FWS listed as endangered the Southwestern willow flycatcher, a bird found in several states of the American Southwest, including New Mexico, where it has been observed along the middle Rio Grande. 60 Fed.Reg. 10694 (Feb. 27, 1995); BOR 191 at 76. Drying of the Rio Grande is harmful to the willow flycatcher’s forage base and habitat. BOR 196 at 88. Overall numbers of flycatchers in areas surveyed have apparently increased in recent years, however, and the parties have confined their legal arguments to the plight of the silvery minnow. D. The administrative procedure regarding the silvery minnow. On July 6, 2000 FWS confirmed the initiation of formal consultation between FWS and the .action agencies, BOR and the Corps. Formal consultation culminated on June 29, 2001 when FWS issued the final BO, which found jeopardy to the silvery minnow from the proposed actions of BOR and the Corps. Both prior to initiation of formal consultation, and also during the intervening period before June 29, 2001, these agencies engaged in a number of preliminary steps that are detailed in the final BO. BO at 2. During this procedure, BOR changed its position as to ownership of the works of the Middle Rio Grande Project. Initially BOR had proposed undertaking no discretionary actions with respect to the three diversion dams on the ground that it did not hold title to them. BOR 167 at 21. BOR’s position regarding title changed, and on July 6, 2000, BOR advised FWS that it had concluded that it in fact owned the constructed facilities in the Middle Rio Grande Project. Ownership of these works is presently the subject of a pending cross-claim by Intervenor, Middle Rio Grande Conservancy District (“MRGCD”). Although BOR now asserts ownership of the diversion dams, BOR is not proposing to reduce water deliveries to MRGCD under the Middle Rio Grande Project or the San Juan-Chama Project. On July 20, 2000, I ordered the parties into mediation. Mediation efforts involving all the parties to this litigation ultimately were unsuccessful in resolving this dispute, and the federal agencies continued with their formal consultation under the ESA. Despite the failure of court-ordered mediation to resolve this case, further negotiations between the United States and the State of New Mexico led to an agreement by New Mexico to store 100,000 acre feet (“af”) of water over a period of three years, and to release up to 30,000 acre-feet per year (“afy”) to benefit the silvery minnow. See Conservation Water Agreement (“CWA”), FWS 2.XII.A.55 14 at 2-3. In turn, the United States agreed to provide incidental take coverage for all New Mexico water users in the Rio Grande system above Elephant Butte Reservoir. The terms of a Memorandum of Understanding (“MOU”) between the United States and the State of New Mexico that preceded the CWA were incorporated into a Biological Assessment (“BA”) submitted by BOR and the Corps to FWS on June 8, 2001. FWS 2.IX.A.95. The United States also negotiated a separate agreement with MRGCD in which MRGCD guarantees that 50% of the amount of water delivered by the United States at the Isleta diversion dam will reach the downstream diversion dam at San Acacia. BOR 32ss ^ 1..1(d) at 3. MRGCD also agrees to assist the federal agencies in developing a fish passage at San Acacia dam, and to assist in rescue of stranded silvery minnows. Id. ¶ J.i(c) and (e) at 3. The terms of these agreements coincide with the implementation period of the BO, which ends December 31, 2003. In its final BO, issued on June 29, 2001, FWS finds that the water operations and river maintenance actions of BOR and the Corps in the middle Rio Grande are “likely to jeopardize the continued existence of the silvery minnow.” BO at 107. The BO proposes a single RPA that FWS believes “will avoid jeopardy to the silvery minnow,” BO at 108, and an Incidental Take Statement (“ITS”) that allows a certain amount of “take” of silvery minnows under specified conditions. BO at 111-113. The ITS purports to shield from take liability under Section 9 of the ESA virtually all water users in the middle Rio Grande basin. BO at 113, and at J-J- IV. The Best Available Scientific Data Claim. Plaintiffs claim that the final BO is arbitrary and capricious because it is contrary to the best available scientific data in the administrative record. The ESA requires federal agencies to use the best scientific data available when consulting about whether federal actions will jeopardize endangered species: Each Federal agency shall, in consultation with and with the assistance of the Secretary, insure that any action authorized, funded, or carried out by such agency (hereinafter in this section referred to as an “agency action”) is not likely to jeopardize the continued existence of any endangered species or threatened species or result in the destruction or adverse modification of habitat of such species which is determined by the Secretary, after consultation as appropriate with affected States, to be critical, unless such agency has been granted an exemption for such action by the Committee pursuant to subsection (h) of this section. In fulfilling the requirements of this paragraph each agency shall use the best scientific and commercial data available. 16 U.S.C. § 1536(a)(2) (emphasis added). This requirement is also written into the applicable ESA regulations. See 50 C.F.R. § 402.14(d) (requiring federal agency requesting consultation to provide FWS with “best scientific ... data available or which can be obtained during the consultation”); 50 C.F.R. § 402.14(g)(8) (requiring FWS to use “best scientific ... data available” in formulating BO and RPAs). A violation of the best available scientific data requirement may be grounds to invalidate a BO as arbitrary and capricious. See Resources Ltd. v. Robertson, 35 F.3d 1300, 1304-05 (9th Cir.1993) (holding federal action agency acted arbitrarily and capriciously in relying on BO with no-jeopardy conclusion because FWS not provided with best available science). Although the ESA requires that FWS be provided with, and use, the best available scientific data, an RPA that is supported only by “weak” scientific evidence is not arbitrary and capricious. See Greenpeace Action v. Franklin, 14 F.3d 1324, 1336 (9th Cir.1992) (upholding action agency’s adoption of “no jeopardy” finding even though it was based on admittedly “weak” scientific evidence); Stop H-3 v. Dole, 740 F.2d 1442, 1460 (9th Cir.1984) (holding it was not clear error of judgment for agency to rely on admittedly weak scientific information). Further, while FWS must rely on the best scientific data available in formulating an RPA, the agency is not limited to scientific considerations; it may also consider political and pragmatic factors. Southwest Center for Biological Diversity v. U.S. Bur. of Reclamation, 143 F.3d 515, 523 and n. 5 (9th Cir.1998). An agency cannot ignore available scientific information. See Conner v. Burford, 848 F.2d 1441, 1454 (9th Cir.1988) (“In light of the ESA requirement that the agencies use the best scientific and commercial data available to insure that protected species are not jeopardized, ... the USFWS cannot ignore available biological information”) (internal cite omitted). Only if there is “no scientific evidence in the record to support” an RPA will it be rejected. Pacific Coast Federation of Fishermen’s Assns., Inc. v. National Marine Fisheries Serv., 265 F.3d 1028, 1037 (9th Cir.2001). The RPA that FWS has developed to avoid jeopardy to the silvery minnow appears at pages 108 through 111 of the BO. It consists of several elements, some of which are “water-related” (A through E, H, and M), others of which are “non-water-related” (F, G, I through L). The water-related features require continuous flow in the Rio Grande from Cochiti Dam to Isleta Diversion Dam year-round, and continuous flow from Cochiti all the way to Elephant Butte Reservoir from October 31 to June 15, with a spike in the flow between April 15 and June 15, if necessary, to induce spring spawning. From June 16 to July 1, the flow downstream of Isleta, as measured at San Acacia, will be “ramped down” gradually, and the RPA specifies a flow from July 1 to October 31 of only 50 cubic feet per second (“cfs”) over the San Acacia Diversion Dam. Other water-related elements include using supplemental water from a Conservation Water Agreement (“CWA”) between the United States and the State of New Mexico to help meet these flows, pumping from the Low Flow Conveyance Channel (“LFCC”) into the river downstream of San Acacia when river drying is likely, and developing better measuring and monitoring of flows. Finally, in years of above-average snowpack and excess water, the Corps will cause over-bank flooding to improve silvery minnow backwater habitat. The non-water-related elements of the RPA include propagation activities to augment captive populations of silvery minnows and facilitate repopulating the upper reaches of the river, efforts to establish silvery minnows upstream and in “off-channel refugial sites,” construction of a fish passage at San Acacia Diversion Dam to allow upstream movement of silvery minnows, and various habitat restoration projects. In addition, the San Marcial Railroad Bridge will be relocated to increase the river’s channel capacity. Plaintiffs challenge the RPA. They first suggest that the record reveals FWS based its judgment not on scientific data, but on pragmatic considerations. They maintain that FWS accepted the State of New Mexico’s proposal for the 30,000 afy of supplemental water that is the subject of the CWA only because BOR and the Corps insisted they had no discretion to do more to maintain a higher level of flow in the river. The Federal Defendants do not dispute that BOR and the Corps believed that they lacked discretion to do more, and that this belief contributed to FWS’ decision to change the water-related elements in the final RPA. In fact, in a June 20, 2001 memo Theresa Davidson, FWS’s chief fishery biologist, included this as a reason for changing the February 9, 2001 draft RPA requirement of continuous river flows year-round from Cochiti Dam to Elephant Butte Reservoir: But why eliminate the requirement for year round continuous flow from Cochiti Dam to Elephant Butte? After the issuance of the draft BO in February, 2001, the Service met with Reclamation to discuss the RPA. It was Reclamations position that they did not have the discretion to guarantee flows, because they could not guarantee that they could obtain enough supplemental water, they could not bypass stored water from El Vado, and though they claim ownership over the diversion dam, they could not force the MRGCD to bypass any necessary water to maintain continuous flows to the headwaters of Elephant Butte. FWS 2.IX.A.79 at 2. However, in evaluating a final BO for compliance with the scientific data standard, a court need not inquire into FWS’s reasons for changing its mind; rather the reviewing court should focus on the question of whether FWS was provided with, and used, the best scientific data. See Southwest Center for Biological Diversity v. U.S. Bur. of Reclamation, 143 F.3d at 523 (holding that if final RPA complied with jeopardy standard and could be implemented by agency, reasons why FWS “changed its mind” need not be considered). Plaintiffs next contend that FWS failed to evaluate meaningfully the scientific data in the record in formulating the RPA. Underlying Plaintiffs’ attack on the RPA is their belief that the minnow needs more water to survive than is provided in the RPA, that the RPA allows massive river drying, and that little or no river drying should be allowed in the San Acacia reach. They argue that FWS’s determination that the RPA will avoid jeopardy to the silvery minnow is directly contradicted by scientific studies in the administrative record, and that FWS has not explained away these studies. Plaintiffs claim that the administrative record shows that FWS has consistently interpreted the scientific data to mean that river drying in the San Acacia stretch, where the last remaining minnows are located, must be avoided to the maximum extent possible to avoid jeopardy. In addition, Plaintiffs assert that FWS’s position has become more emphatic in the years since the endangered listing in 1994, as the minnow population has continued to spiral downward. Plaintiffs go through a detailed analysis starting with the 1994 listing of the minnow, FWS III.A. 19, to show that FWS has emphasized the importance of river drying as a danger to the minnow’s survival. In the 1999 recovery plan developed by FWS for the silvery minnow with input from numerous outside scientists and stakeholders, FWS V.C.22, “providing sustained flow” was listed among the factors considered important to the minnow’s recovery. FWS V.C.22 at 12-13. Also, “river dewa-tering” is mentioned as a factor in drops in population levels of the minnow. Id. at kk-15. The recovery plan estimated that it would be necessary to bypass up to 200 cfs of water at San Acacia in order to maintain flowing water downstream to San Marcial. Id. at 50. Plaintiffs contend that in the January 1999 draft BO, FWS 2.IX.A.55, which was never finalized, FWS rejected BOR’s plan to lease 30,000 afy for the minnow, and instead required that supplemental water be sufficient to maintain continuous river flows throughout the San Acacia reach. Id. at 107. In the November 2000 proposed BO, FWS stressed the need to ensure no significant river drying below Isleta and San Acacia Dams, stating: “Continued conditions that dewater sections of the middle Rio Grande, resulting in reduced silvery minnow reproduction and recruitment throughout the species’ range, could lead to its extinction ... [and] dewatering [the San Acacia] reach in 2000 could have critical consequences for the species.” FWS 2.IX.A59 at 17-18, 21. In the February 2001 draft BO, BOR 196, FWS held that jeopardy to the minnow could only be avoided with continuous flows in the middle Rio Grande past San Marcial, the last gauging station before Elephant Butte Reservoir, and FWS set 50-150 cfs as the target range of those flows at San Marcial, depending on precipitation. Plaintiffs cite other scientific evidence in the record that supports the notion that continuous river flow is needed: Steven Platania, the leading silvery minnow researcher and expert for over 16 years, identifies river drying, especially within the San Acacia reach, as the most dire of three substantial factors adversely affecting the minnow. BOR 10, ¶¶ 32-35. James Wilber, BOR’s fishery biologist, confirms that with the dramatic decrease in abundance of minnows upstream, river drying in the lower reaches could jeopardize the continued existence of the species. BOR 37, ¶¶ 29-31. And Plaintiffs’ experts Dean Hendrickson and John Pittenger support the same conclusions about the effects of river drying below San Acacia diversion dam. Hendrickson stated, “Unless the last remaining native habitat of the Rio Grande silvery minnow is kept intact, with sufficient river flows to prevent further river drying, the species could suffer irreversible losses and become extinct.” BOR 36, ¶ 31p. Pittenger stated, “river drying events have a major influence on mortality ... of Rio Grande silvery minnow.” FWS 2.IX.A.69. Plaintiffs argue that in the June 2001 final BO, FWS failed to invoke any biological evidence or reasoning for allowing inadequate flows and extensive river drying, and that FWS has not proposed a way to replace the minnow habitat lost by river drying. Thus, Plaintiffs contend there is no “rational connection” between the “facts found in the BO” and the RPA’s conclusion that the minnow will not be jeopardized by the river drying expected and allowed under the RPA. Plaintiffs maintain that the RPA allows the destruction of about 75% of the minnow’s remaining habitat, the area below San Acacia Dam. Plaintiffs arrive at this figure from their interpretation of a March 19, 2001 internal memorandum by Theresa Davidson. FWS 2.IX.A.6h. This, Plaintiffs argue, combined with the unrefuted scientific evidence that the minnow is exceptionally vulnerable to river drying, means the RPA is not rationally related to the evidence in the administrative record, making it arbitrary and capricious. The Federal Defendants acknowledge that massive river drying would mean the death of many silvery minnows. They contend, however, that FWS formulated an RPA that meets the biological needs of the minnow in a manner different from the approaches set forth in the draft documents that preceded the final BO. They maintain that FWS balanced flow requirements with other water-related and non-water-related elements to avoid jeopardy. My review of the record indicates that the water-related elements in the RPA are probably adequate in a year of average precipitation to prevent the type of massive river drying that would mean the demise of the silvery minnow. Plaintiffs’ 75% figure is an exaggeration. I have studied the March 19, 2001 Davidson memo closely, because it is cited frequently by Plaintiffs, and compared it with a June 20, 2001 Davidson memo, FWS 2.IX.A. 79, and with other data in the record. My calculations result in an estimate that less than 33% of the San Acacia reach may be too dry to support the minnow. Still, the RPA appears to allow close to 30% of the San Acacia stretch to dry during some parts of the summer months when irrigation is at its peak. This drying in the San Acacia reach is offset by other factors such as providing enough water to result in continuous spring flows, and ramping down flows gradually after June 15, which will allow for better rescue of minnows stranded in isolated pools. The RPA also provides for a spring spawning spike to induce spawning. Plaintiffs’ expert, Pittenger, agrees that spring spawning flows are “a major determinant of spawning success,” FWS 2.IX.A.69 at 3. Pittenger lists affording “spring flows to promote successful spawning” as the first in the strategy for reducing risk of extinction in 2001 because “a successful spawn is critical to survival of the species”, id. at 10. The RPA protects that feature. Moreover, there is evidence of some recovery in silvery minnow population during 2001. Recent monitoring reports that predated the final BO found large numbers of silvery minnows in reaches of the river upstream of San Acacia in 2001, BOR 3s (Platania Population Monitoring Data, at 6). This indicates an apparent improved status of the minnow in 2001 in the Isleta and Angostura reaches, upstream of the San Acacia reach, which could result in some natural repopulation in the San Acacia reach because the minnows’ eggs drift freely downstream after the spawn. See also Dudley, R and S. Platania, Summary of Population Monitoring of Rio Grande Silvery Minnoiu (26-28 Jime 2001), dated 6 August 2001, BOR 38ss at 2, and 39ss at 9 (report finding large number (188) of silvery minnows in Angostura reach, upstream from Isleta diversion dam). Therefore, even though no minnows were found in the upper reaches during 2000, by June 29, 2001 there was some scientific information indicating that the minnow population was recovering in those reaches. The significance of finding minnows upstream of the San Acacia reach is illustrated in an August 14, 2000 memo of Hoagstrom and Brooks, of the FWS New Mexico Fishery Resources Office, Rio Grande silvery minnow recovery priorities, FWS 2.V.A.54,. In this memo, the authors state that while flows and habitat conditions in the San Acacia Reach are “important to long-term survival of the species, it is equally, if not more important to insure that the Rio Grande upstream of San Acacia enter into considerations for high priority recovery actions.” Id. at 1. The authors also explain that reduction in minnow population in the San Acacia Reach is caused not only by river channel drying, but also by a reduced population upstream. Id. at J. If spawning does not occur upstream, eggs will not flow downstream and hatch. Conversely, if more minnows are found upstream, the effect of downstream drying will be lessened. The RPA sets forth ways to assist upstream repopulation of minnows. The RPA’s non-water-related elements include habitat restoration, repopulation of the minnow in upstream reaches, and construction of a fish passage at San Acacia Diversion Dam. These features of the RPA, designed to enhance the water-related elements, aim to protect the minnow’s survival and to ensure its recovery by avoiding jeopardy to the species. Plaintiffs counter that until efforts to reestablish the minnow upstream succeed, the evidence clearly establishes that the river must be kept flowing in the San Acacia reach year-round, especially now when the minnow is in a near-term crisis and at risk of extinction. As mentioned earlier, however, the record does contain some data, albeit weak, that indicate upstream recovery has begun. Additionally, the RPA includes artificial propagation of minnows and reintroduction of minnows into upstream reaches. See FWS 2.VII.A.16 and .17 (e-mail memos regarding efforts at artificial propagation and release in May and June 2000). Finally, the consultation over the silvery minnow is a dynamic, ever-evolving process, and the BO is not intended to be the final solution to protecting the minnow from extinction. If there is a drought year, as it presently appears 2002 will be, FWS will reinitiate consultation. Even if that does not occur, consultation must begin again within the 20 remaining months in the term of the June 2001 BO, which was shortened from five years to 30 months. This shortened period probably will reduce many existing biological uncertainties and data gaps. My ruling on the scope of discretion, which follows, will guide the agencies and should help expedite the process when further consultation occurs. The ESA requires FWS to design an RPA that does not jeopardize the minnow, and in so doing to use the best available scientific data, but it does not require FWS to choose the best or most effective RPA that will avoid jeopardy. The best available scientific data was provided to FWS, which considered that data in determining that the actions proposed in the RPA avoid jeopardy. Plaintiffs make strong arguments emphasizing that water is important to the survival of the silvery minnow, and that extensive river drying is harmful. But Plaintiffs have not established that FWS ignored scientific data in the record, and they have pointed to no scientific data that were not considered by FWS. Although it is a close question, I find that FWS’s determination that it can avoid jeopardy by instituting the RPA, even though it allows for some intermittent drying of the river in the San Acacia reach, is not arbitrary and capricious, or contrary to law. Y. The Consultation Claim. From the beginning of this litigation, Plaintiffs have asserted as their main point that BOR and the Corps have failed to consult fully with FWS under the ESA over all the actions in regard to which BOR and the Corps have discretion (Count 1). Plaintiffs continue to challenge the outcome of the final BO, as well as the procedure used by the federal agencies, based on this assertion (Count 6). Plaintiffs maintain that the BO fails to examine the full scope of possible remedial actions that FWS could have prescribed in an adequate RPA, thereby leaving less water available to protect the endangered silvery minnow. The Federal Defendants respond, first, that the consultation claim is moot and, second, that the agencies have fulfilled the consultation requirements imposed by the ESA. I first rule that the scope of consultation issue is not moot. Next, I hold that the BOR retains sufficient discretion over its river management and operations in the middle Rio Grande, specifically water deliveries under the Middle Rio Grande Project and under the San Juan-Chama Project, to require BOR to consult over those actions under Section 7(a)(2) of the ESA. Finally, I examine whether the lack of full consultation is a ground for invalidating the final BO, and conclude that it is not. A. Mootness. Defendants assert that the consultation claim is moot because of the issuance of the final BO. An actual controversy must exist at all stages of the litigation. Arizonans for Official English v. Arizona, 520 U.S. 43, 67, 117 S.Ct. 1055, 137 L.Ed.2d 170 (1997). “A case is moot when the issues it presents are no longer live, or the parties lack a legally cognizable interest in the outcome.” Los Angeles County v. Davis, 440 U.S. 625, 631, 99 S.Ct. 1379, 59 L.Ed.2d 642 (1979) (quoting Powell v. McCormack, 395 U.S. 486, 496, 89 S.Ct. 1944, 23 L.Ed.2d 491 (1969)). A case may be moot if there is no reasonable expectation that the alleged violation will recur, and “interim relief or events have completely and irrevocably eradicated the effects of the alleged violation.” City of Albuquerque v. Browner, 97 F.3d 415, 420 (10th Cir.1996). In Southern Utah Wilderness Alliance v. Smith, 110 F.3d 724 (10th Cir.1997), a case in which the only relief sought by plaintiffs was to force ESA consultation, the Tenth Circuit held that the completion of consultation by the issuance of a final BO mooted that claim. Id. at 729-30. Here, however, Plaintiffs have always complained not only about completing consultation, but also about the breadth of the consultation, i.e. that the ongoing consultation was too narrow in scope. In Count 6 of their second amended complaint, Plaintiffs seek to invalidate the BO because of alleged improper narrowness of the consultation’s scope. Plaintiffs contend that a procedural violation of Section 7(a)(2) of this sort is sufficient to overturn the BO. There is some support for this contention, as discussed later in this opinion. Thus, in order to determine whether the BO is arbitrary and capricious on this basis, I must look at the merits of the consultation claim to decide whether the scope of consultation was too narrow, and if so go on to determine whether that is a ground for invalidating the BO. I find the reasoning in NRDC v. Houston, 146 F.3d 1118 (9th Cir.1998), where the court confronted the issue of mootness under very similar circumstances, to be persuasive here. The Plaintiffs in NRDC v. Houston had challenged the scope of consultation under Section 7(a)(2) of the ESA, and before the case was decided, a final BO was issued. The court stated: “Procedural violations of the ESA are not necessarily mooted by a finding by the FWS that a substantive violation of the ESA had not occurred.” Id. at 1128. Cases in which courts have concluded that the issuance of a BO renders a consultation claim moot have involved procedural circumstances different from those involved in this case. For example, in EPIC v. Pacific Lumber Co., 67 F.Supp.2d 1113 (N.D.Cal.1999), the only relief sought was “to forestall during the consultation period the irreversible or irretrievable commitment of resources,” id. at 1124. Similarly, in Southern Utah Wilderness Alliance v. Smith, the only relief sought by plaintiffs was to force ESA consultation. 110 F.3d at 729-30. In support of their mootness argument, Defendants cite Southwest Center for Biological Diversity v. U.S. Bur. of Reclamation, 143 F.3d 515 (9th Cir.1998), which held that BOR’s assertion that it lacked discretion to lower the water level in Lake Mead was immaterial to the court’s review of a final BO under the APA. Southwest Center is inapposite because there all claims against BOR had been dismissed by the district court for lack of notice under the citizen suit provisions of the ESA; the only remaining claim was a challenge to the final BO by FWS under the APA. That case did not squarely present the question of whether a procedural violation of Section 7(a)(2) of the ESA had occurred, nor did it discuss mootness. The Federal Defendants argue that the agencies expanded the consultation to encompass all depletions in the middle Rio Grande, thus mooting the claim that the consultation was too narrow. This is a matter to be determined on the merits of the consultation claim, however, not when determining mootness. Another reason why this issue is not moot is that the BO itself contemplates reinitiation of consultation, and so there is a reasonable expectation that the controversy about the scope of consultation will recur within the meaning of City of Albuquerque v. Browner, 97 F.3d at 420. This will happen for certain before December 31, 2003, the end of the 30-month period covered by the BO. It is likely to happen sooner, even as early as this summer, because current weather conditions, of which I take judicial notice, indicate the Rio Grande watershed is experiencing a very dry year that may cause significant river drying beyond that contemplated by the BO. If 2002 continues to be a dry year and the limits in the BO’s ITS are likely to be exceeded, FWS will reinitiate consultation, and it may need to require BOR and/or the Corps to release more water to protect the silvery minnow. To do that, the parties need a judicial determination of the extent of the federal action agencies’ discretion and the corresponding proper scope of consultation when it is reinitiated. Defendants have failed to carry their “heavy burden” of demonstrating mootness. Los Angeles County v. Davis, 440 U.S. at 631, 99 S.Ct. 1379. B. Consultation — Legal Requirements. Section 7(a)(2) sets forth the consultation requirement of the ESA. The consultation requirement is broad and significant in the overall scheme of the ESA. Once the silvery minnow was listed as endangered, BOR and the Corps were required to consult with FWS to “insure that any action authorized, funded, or carried out by such agency ... is not likely to jeopardize the continued existence of any endangered species....” 16 U.S.C. § 1536(a)(2). The regulations implementing Section 7 define agency “action” as “all activities or programs of any kind authorized, funded, or carried out, in whole or in part, by Federal agencies.... ” 50 C.F.R. § 402.02. The regulations also indicate that Section 7 applies to “all actions in which there is discretionary Federal involvement or control.” 50 C.F.R. § 402.03. Thé regulations also require each federal agency to “review its actions at the earliest possible time to determine whether any action may affect listed species-” 50 C.F.R. § 402.14(a). And if a new species is listed in the midst of an agency action, or if an action changes to the detriment of a species, or if the amount of taking of a species exceeds that specified in the Incidental Take Statement, or if significant new information is revealed, the agency must reinitiate consultation “where discretionary Federal involvement or control over the action has been retained or is authorized by law....” 50 C.F.R. § 402.16. Thus, the Court’s inquiry for each “agency action” is whether the agency retained “discretionary Federal involvement or control” within the meaning of 50 C.F.R. § 402.03. This inquiry is viewed against the backdrop of TVA v. Hill, in which the Supreme Court observed that the ESA commands that section 7 applies to “all ... actions authorized, funded, or carried out” by federal agencies and that “[t]his language admits of no exception.” 437 U.S. at 173, 98 S.Ct. 2279. Again, “Congress has spoken in the plainest of words, making it abundantly clear that the balance has been struck in favor of affording endangered species the highest of priorities, thereby adopting a policy which it described as ‘institutionalized caution.’ ” Id. at 194, 98 S.Ct. 2279. In some cases, it is apparent that a federal agency, sua sponte or at least without a court order, has consulted over agency actions. See, e.g., O’Neill v. United States, 50 F.3d 677 (9th Cir.1995) (ongoing water delivery contracts); Klamath Water Users Protective Association v. Patterson, 204 F.3d 1206 (9th Cir.1999) (water deliveries under reclamation project); Kandra v. United States, 145 F.Supp.2d 1192 (D.Or.2001) (same). In others, courts have ordered federal agencies to comply with their ESA consultation duties. Pacific Rivers Council v. Thomas, 30 F.3d 1050 (9th Cir.1994); NRDC v. Houston, 146 F.3d 1118 (9th Cir.1998) (renewal of water delivery contracts); Pacific Coast Federation of Fishermen’s Ass’ns v. U.S. Bureau of Reclamation, 138 F.Supp.2d 1228 (N.D.Cal.2001); WaterWatch of Oregon v. U.S. Army Corps of Engineers, 2000 WL 1100059 (D.Or. June 2000) (permits for irrigation pumps). In some, an agency’s failure to consult has been upheld by a court which found the agency lacked discretion over an action. Environmental Protection Info. Center v. Simpson Timber Co., 255 F.3d 1073 (9th Cir.2001) (incidental take permit for logging); Sierra Club v. Babbitt, 65 F.3d 1502 (9th Cir.1995) (agreement for logging road); Center for Biological Diversity v. Nat’l Marine Fisheries Serv., 2001 WL 1602707 (N.D.Cal.2001) (longline fishing permits issued to fishing vessels). C. Scope of Consultation. Plaintiffs contend that BOR and the Corps, in their consultations with FWS, should have considered protecting the endangered silvery minnow more broadly: (1) by BOR limiting water deliveries to the MRGCD under the Middle Rio Grande Project; (2) by BOR altering water deliveries and operations to federal contractors under the San Juan-Chama Project; and (3) by the Corps deviating from normal operation of several of its reservoir facilities on the middle Rio Grande. “Agency action” under the ESA includes a broad array of actions, including annual water deliveries under federal projects, NRDC v. Houston, 146 F.3d 1118, 1130 (9th Cir.1998), renewal of long-term water delivery contracts, Id. at 1125, and ongoing long-range forest management plans, Pacific Rivers Council v. Thomas, 30 F.3d 1050, 1056 (9th Cir.1994). I conclude that the expansive definition of “agency action” in the ESA regulations, 50 C.F.R. § 402.02, encompasses the annual water deliveries by BOR under the Middle Rio Grande Project and the San Juan-Chama Project, and the Corps’ operation of reservoirs and dams in the middle Rio Grande. Thus, I turn to the more hotly contested issue, federal agency discretion. I begin the discretion analysis by examining the applicable statutes, regulations, and contracts for each of the actions at issue. 1. BOR’s water deliveries to MRGCD under the Middle Rio Grande Project. The Middle Rio Grande Conservancy District (“MRGCD”) was organized as a political subdivision of the State of New Mexico under the Conservancy Act of New Mexico in 1925. MRGCD constructed the major irrigation facilities on the middle Rio Grande in the 1930s, with partial funding from the United States related to Pueblo Indian water deliveries. Its irrigation and flood control works deteriorated seriously in the 1940s, and by then MRGCD had become financially insolvent. This led to the federal Middle Rio Grande Project, which dates back to 1947-48 when BOR and the Corps prepared a project plan. According to the plan, BOR would undertake channel rectification and rehabilitation of MRGCD irrigation and drainage works, and the Corps would construct reservoirs and levees for flood protection. BOR 807. Congress approved those proposals in the 1948 and 1950 Flood Control Acts. Pub.L. 80-858 (June 30, 1948) and Pub.L. 81-516 (May 17, 1950), BOR 795, 616. Under the Congressionally-approved plan for the Middle Rio Grande Project, BOR was directed to pay off the Middle Rio Grande Conservancy District’s outstanding debt, and to rehabilitate and expand irrigation, storage and diversion facilities along the middle Rio Grande. In return, MRGCD would transfer its assets and management of the system to the federal government, and repay a portion of project costs over time. The United States would thus assume ownership of all MRGCD diversion and storage facilities until project costs were repaid and Congress ordered a transfer back to MRGCD. Unlike most federal Reclamation projects, BOR did not appropriate water rights for the Middle Rio Grande Project. Under a 1951 contract, MRGCD was to assign its water rights to the United States, but ultimately MRGCD in 1963 transferred to BOR only the right to store water in El Vado Reservoir. MRGCD has since paid its debt to the United States, and asserts title to the works of the Project. MRGCD contends that the Court cannot decide the consultation claim until after the title question is resolved. I disagree. The United States claims ownership of these works, and under the Quiet Title Act, “possession or control” remains in the United States until 60 days after final resolution of the title issue. 28 U.S.C. § 2409a(b). BOR owns El Vado Dam and Reservoir and is authorized by federal law and state permit to store native Rio Grande water there for MRGCD. BOR releases that water at MRGCD’s call as agreed by contract and authorized by the state permit. MRGCD operates the United States’ diversion dams to divert these native flows for MRGCD and Pueblo Indian use during the irrigation season. BOR has consulted with FWS over some aspects of its operation of El Vado Reservoir and of diversions at the diversion dams, but it has not consulted about reducing water deliveries to MRGCD because it has taken the position that it lacks discretion to do so. Plaintiffs have identified two sources for BOR’s discretion to limit water deliveries to MRGCD under the Project. The first is the shortage provision in the applicable contracts. The 1951 Contract between the United States and MRGCD provides: United States Not Liable for Water Shortage 23. Should there ever occur a shortage in the quantity of water which normally would be available through and by means of said project works constructed in connection therewith, in no event shall any liability accrue therefor against the United States.... BOR 306 ¶ 23. A subsequent amendment (leaving paragraph 23 intact) called for the United States to “operate and maintain any works, structures, and improvements, including levees and flood control structures, as requested by [MRGCD].... ” BOR 306 (Art. 6, Contract Am. No. 2, dated Jan. I, 1955, as amended by Contract Am. No. 3.). A 1963 “Amendatory Contract,” which amended the 1951 contract, states in paragraph 12: b. Water Shortages — On account of drouth and other causes, there may occur at times during any year a shortage in the quantity of water available from the reservoir storage complex for use by the District pursuant to this contract. In no event shall any liability accrue against the United States ... for any damage, direct or indirect, arising out of any such shortage. BOR 521 § 12. b. The second source of BOR’s alleged discretion to restrict water deliveries to MRGCD stems from the Reclamation Act, 43 U.S.C. § 372, which limits use of federal project water to that reasonably needed for beneficial use. In addition, the Act requires that federal water projects conform with state water law, unless doing so would interfere with other federal requirements or interests. 43 U.S.C. § 388. Plaintiffs claim that both of these statutes are being violated if the MRGCD is using more water than reasonably needed for beneficial use. There is some evidence that this is happening. State Engineer Thomas C. Turney asserted in a letter to MRGCD that from 1989 to 1999, MRGCD diverted 11 afy per acre irrigated, while 7.2 afy is “sufficient and non-wasteful.” BOR í86. MRGCD disputes that it has exceeded reasonable beneficial use, and it is not my role in this litigation to determine that it has or has not. The point, however, is that BOR has a statutory duty to determine whether overuse is occurring. If excessive water is being diverted, this may provide BOR the requisite discretion under the ESA to consult with FWS about the quantity of water BOR delivers to MRGCD under the Middle Rio Grande Project. Diversions in excess of what is reasonably needed for beneficial use may be a source of water that could be used to help preserve the endangered minnow. In Klamath Water Users Protective Association v. Patterson, 204 F.3d 1206 (9th Cir.1999), after BOR consulted with FWS over water releases under a long-term contract designed to govern the management of a reclamation project dam, irrigators challenged BOR’s decision to reduce releases from the dam to provide minimum lake levels for fish species listed as threatened or endangered under the ESA. The irrigators claimed this would interfere with their rights under a contract. The contract had been entered into under “the Reclamation Act and acts of Congress relating to the preservation and development of fish and wildlife resources.” Id. at 1209 (internal quotations omitted). Although the irrigators were not parties to the contract, which was between BOR and a power company, they asserted their rights as third-party beneficiaries. The dam had been built to help the United States satisfy its contractual obligations to water users in the basin, including the irrigators, and for other purposes such as providing water to flood adjacent wildlife refuges. The court denied the irrigators’ requested relief, holding that they were not third-party beneficiaries under the contract. It also found that BOR had retained authority to manage the dam, “and because it remains the owner in fee simple of the Dam, it has responsibilities under the ESA as a federal agency. These responsibilities include taking control of the Dam when necessary to meet the requirements of the ESA, requirements that override the water rights of the Irrigators.” Id. at 1213. O’Neill v. United States, 50 F.3d 677 (9th Cir.1995) involved a long-term water service contract between the United States and a water district relating to the Central Valley Project in California. The contract contained a clause limiting the United States’ liability in the event of shortage caused by “errors in operation, drought, or any other causes.” Id. at 680. The water district sued soon after BOR announced that the district’s consumers would receive only half of their contractual supply of water. This occurred after consultation under the ESA resulted in a BO which concluded that BOR’s continued operations of the CVP “was likely to jeopardize the continued existence” of a listed salmon in the Sacramento River. Id. at 681. Citing Merrion v. Jicarilla Apache Tribe, 455 U.S. 130, 148, 102 S.Ct. 894, 71 L.Ed.2d 21 (1982), the Ninth Circuit observed that contracts to which the sovereign is a party are subject to subsequent legislation by the sovereign. O’Neill, 50 F.3d at 686. The court upheld BOR’s action taken to comply with the ESA. In WaterWatch of Oregon v. U.S. Army Corps of Engineers, 2000 WL 1100059 (D.Or. June 2000), the court confronted the issue of whether a federal agency maintained discretionary control over long-term permits for pumping irrigation water sufficient to require the agency to enter into Section 7(a)(2) consultations under the ESA. The federal agency contended that it did not need to consult about having issued the permits on the ground that it had no discretion concerning the permits during the permitted pumping. Id. at *8. The court disagreed, and found that the federal agency “retained ongoing discretionary involvement or control.” Id. In doing so, the court looked to the language of the permits and noted two critical features. First, the permits contained conditions expressly allowing the Corps to suspend, modify, or revoke them if doing so was “in the public interest,” the definition of which included “fish and wildlife concerns.” Id. at *7. Second, an amendment to one of the permits indicated that changes in water use could result in revocation of the permit. Id. at *8. Not all ongoing federal contracts involve the “discretionary Federal involvement or control” required by 50 C.F.R. § 402.03. See, e.g., Sierra Club v. Babbitt, 65 F.3d 1502 (9th Cir.1995), where the Ninth Circuit held that BLM lacked sufficient discretion under an ongoing right-of-way agreement for construction of a logging road. The court rejected plaintiffs blanket suggestion that the ESA “abrogates preexisting agreements.” Id. at 1510. It also rejected the notion that the ESA represents “Congress’s intent to void preexisting agreements.” Id. Thus, the ESA does not mandate that a federal agency “abrogate” or “void” existing contracts. Rather, the ESA only requires consultation if the terms of the contract give the agency discretion to perform the contract in a way to benefit listed species. See also Environmental Protection Info. Center v. Simpson Timber, 255 F.3d 1073 (9th Cir.2001), where the Ninth Circuit found no “agency action” and a lack of agency discretion under an ongoing permit issued to a timber company. The permit allowed the company incidental take of one species, but the permit did not retain federal discretion to alter it for new species listed after issuance. I do not read Environmental Protection Info. Center v. Simpson Timber to mean that there can never be federal agency discretion under existing permits or contracts to take action to protect endangered species. Rather, where there is discretion under permits or contracts, that discretion must be exercised. The present case is closer to the situation in NRDC v. Houston, 146 F.3d 1118 (9th Cir.1998), where the Ninth Circuit found the requisite agency discretion to alter the terms of renewal contracts based on the federal reclamation laws. Id. at 1126. The reclamation laws provided the right to renewal of long-term water delivery contracts on “mutually agreeable” terms. Id. When the contracts were up for renewal, BOR renewed them without altering the quantity of water to be delivered, even though some fish species affected by the project by then had been listed as threatened under the ESA. The court held that the reclamation laws gave BOR discretion to decrease the quantity of water to be delivered when renewing the contracts, even though those laws did not expressly state there was discretion to decrease the quantity of water. The reclamation laws gave the contractors “a first right” to a “share or quantity of the project’s available water supply ...” id,., but the ESA gave BOR the authority to reduce the total amount of available project water to be delivered to the contractors. This supports the position that the reclamation laws can be a source of federal agency discretion. In accordance with the extensive protective purposes of Section 7 of the ESA, “[t]he term ‘agency action’ has been defined broadly.” NRDC v. Houston, 146 F.3d at 1125; see also Pacific Rivers Council v. Thomas, 30 F.3d 1050,1054 (9th Cir.1994) (“there is little doubt that Congress intended to enact a broad definition of agency action in the ESA”). Conversely, what may be called the “discretionary exception” contained in 50 C.F.R. § 402.03 should be construed narrowly, consistent with the Supreme Court’s exhortation in TVA v. Hill that the language of Section 7 “admits of no exception.” 437 U.S. at 173, 98 S.Ct. 2279. Although it is a close question, I believe that BOR does have discretion over its actions resulting in the delivery of water to MRGCD under the Middle Rio Grande Project, and that it should have consulted with FWS regarding those water deliveries. The term “drouth and other causes” in the contract should be interpreted broadly in this case because drought and dryness are what affect the continued existence of the silvery minnow. Viewed together with BOR’s statutory duty to limit MRGCD’s diversions .to amounts reasonably needed for beneficial use, BOR’s discretion becomes even m