Listen to this post

In a decision that included three opinions from the three-judge panel, the Ninth Circuit in San Luis & Delta-Mendota Water Authority v. Jewell, 2014 U.S. App. LEXIS 4781, reversed the district court’s decision and upheld a 2008 biological opinion prepared by the U.S. Fish & Wildlife Service (FWS) regarding the effect of the Central Valley Project (CVP) and State Water Project (SWP) on the endangered delta smelt.

The U.S. Bureau of Reclamation (Reclamation), which operates the CVP, requested the opinion in accordance with the Endangered Species Act (the ESA). The CVP and SWP supply water to more than 20 million people in southern California from the Sacramento-San Joaquin Delta (the Delta). The estuary at the confluence of the Delta and the San Francisco Bay forms the spawning habitat for the endangered, two-to-three inch delta smelt.

The 400-page biological opinion found that the continued operations of the CVP and SWP threatened the delta smelt and proposed several alternatives for Reclamation to follow to protect the smelt. The district court held that the biological opinion was arbitrary and capricious and remanded it back to FWS for another draft.

The court acknowledged the major implications of reversing the district court, but noted that it was not reviewing the case to balance the interests of the smelt against water users. The court’s review was limited to whether the biological opinion was arbitrary and capricious. While the ESA requires FWS to base the biological opinion on the best scientific evidence available, the court should defer to FWS to determine what evidence is the best and only find the biological opinion flawed if it completely disregards other available evidence.

The court considered each of the flaws the district court identified in the biological opinion and reversed the district on each one. First, the biological opinion did not err in relying on raw salvage numbers of smelt from the pumping stations’ screen traps to set the flow limits at pumping stations in the Delta. The court acknowledged that raw salvage figures were not the most precise figures to use. However, it was not arbitrary and capricious to do so because the raw figures resulted in more conservative flow limits and greater safeguards for the smelt. As the court would do throughout its decision, it emphasized that this was exactly the type of scientific determination that courts should defer to experts to make.

Second, the court held that FWS’s methodology for establishing the geographic range in the estuary with the suitable salinity for the smelt’s spawning habitat was not arbitrary and capricious. While the model FWS used had flaws, the biological opinion identified the flaws in the alternative models. As a result, it was improper for the district court to substitute its judgment for FWS’s judgment and select one flawed model over another.

Third, the court held that the biological opinion’s Incidental Take Statement was not flawed. The court found that it was reasonable for the biological opinion to use different data sets to estimate the incidental take of adult and juvenile smelt. Additionally, it was reasonable for the biological opinion to use an average cumulative salvage index to set the incidental take limits.

Fourth, the court upheld the biological opinion’s conclusions regarding the indirect effects of the CVP and SWP on the smelt’s food supply, pollution, predation, aquatic vegetation, and toxic bacteria. The biological opinion recognized that many factors impact the dynamic smelt population, but each of these factors is affected by the flow from CVP and SWP operations. The court considered each factor and while deferring to the scientific expertise of FWS, determined that the biological opinion’s conclusions were not arbitrary and capricious.

Fifth, the court held that FWS was not required to support the “nonjeopardy elements” of the “reasonably prudent alternatives” described in the biological opinion. FWS’s regulations define a “reasonably prudent alternative” as an action that 1) can be implemented in a manner consistent with the intended purpose; 2) can be implemented within the scope of the agency’s authority; 3) is economically and technologically feasible; and 4) would avoid jeopardizing the existence of the endangered species or its habitat. The first three elements are known as the “nonjeopardy elements,” and because FWS has a duty to protect the smelt “whatever the cost,” FWS was not required to consider these “nonjeopardy elements” in the biological opinion. The court noted that even if FWS was required to support the “nonjeopardy elements,” sufficient support existed in the record.

After reversing all of the bases on which the district court remanded the biological opinion, the court affirmed the district court’s holdings on a couple points.  First, the court held the agency was not required to separate discretionary from non-discretionary actions for purposes of establishing the environmental baseline. Second, the court held the National Environmental Policy Act (NEPA) does not require FWS to prepare an Environmental Impact Statement (EIS) in conjunction with the issuance of the biological opinion. Despite the “powerful coercive effect” that the biological opinion would have on Reclamation to follow the recommendations, merely issuing the opinion was not a “major federal action significantly affecting the quality of the human environment.” Reclamation’s adoption and implementation of the biological opinion however, did trigger an obligation to comply with NEPA.


Courts should be highly deferential to the scientific expertise of FWS when reviewing biological opinions prepared by FWS pursuant to the ESA. Although agencies must base their findings on the best scientific evidence available, agencies are not required to use the best evidence possible. As a result, courts should only find a biological opinion to be arbitrary and capricious if the agency ignores available scientific data that is superior to the evidence that the agency relies on.

Leave a Reply

Your email address will not be published. Required fields are marked *