Big Wind, Bird Kills and Uncertainty’s Burdens

A new judgment for wildlife plaintiffs shows that bird kills at turbines remains a divisive issue.

The U.S. Fish & Wildlife Service has had a hard time with wind turbines.  It has long been at work attempting to “facilitate the development of renewable energy,” a/k/a permit the siting and operation of wind turbines (and solar arrays) notwithstanding the attendant risks to resident wildlife populations.

In 2009 FWS issued rules authorizing the limited killing of bald and golden eagles under the Bald and Golden Eagle Protection Act (BGEPA) for “otherwise lawful activities” that involve a certain amount of eagle mortality “incidental” thereto.  The maximum duration of such a permit: 5 years.

Golden Eagle

Golden Eagle

At the time, FWS didn’t anticipate that it would be issuing a lot of these BGEPA permits to wind developers.  Indeed, the environmental assessment and finding of no significant impact (EA/FONSI) done for the 2009 “5 year rule” concluded that the effects would be minimal precisely because such permits would be rare.  But then the push for wind development hit and the developers came calling.  Their objection?  Five years is too short.  The average wind development project is amortized over 20 or 30 years and the prospect that 5 years into the project a needed permission could be revoked because evidence of significant bird kills had accumulated was too daunting.

So FWS proposed a new rule in April 2012, aiming to extend the maximum term for these permits to 30 years.  And the real maneuver: calling the rule “strictly administrative,” FWS decided that it was categorically exempt and, thus, that no EIS and not even an EA/FONSI was required of this new “30 year rule.”

When Is An Action “Strictly Administrative”?

Readers will recall that this issue will likely play some role in the lawsuits raising NEPA claims against the recent WOTUS rulemaking. {See our first post on it here.} The Justice Department has had some success arguing that NEPA does not require any “detailed statement” be prepared for actions that may end up significantly impacting the environment if the actions are preliminary or jurisdictional in nature.  Reasonable people disagree about the WOTUS rule’s fit in that category. {See this nice collection of resources by RegBlog} But the 30 year rule undertaken to “provide more certainty to [wind] project proponents and their funding sources,” 77 Fed. Reg. at 22,275, encountered fierce resistance on the NEPA CATX from the outset.

Not only did the environmental community almost unanimously oppose the application of the CATX.  Several other government bureaus did so and, ultimately, FWS’s own eagle program manager did so as well.  Loudly.

In her view, many of the comments ha[d] a great deal of merit . . . and it was therefore a “no-brainer that [FWS] needed to do a NEPA analysis.  According to [this official] “the [30-year] permit will be inherently less protective for eagles than 5-year permits that require, upon renewal, the project proponent to implement any known measures to reduce [killing] many of which would not have been known when the original permit was issued.

Shearwater et al . v. Ashe  (N.D. Cal. 2015), Slip Opinion at 12 (emphasis added).

Other FWS officials noted that there was no way this CATX or the underlying determination could be “tiered” to the 2009 EA/FONSI and that the agency had not, as CEQ and Department of Interior rules require, considered any “extraordinary circumstances” that might put this particular “purely administrative” move outside the scope of the agency’s CATX.  Slip Op. at 12.

Ultimately, the FWS Director Dan Ashe overruled the staff’s objections and ordered that the CATX be applied and the rulemaking go forward with no NEPA documents.  Id. at 13.  He also concluded that consultations pursuant to Endangered Species Act § 7 were unneeded.

The Line Between NEPA Compliance and Judicial Enforcement

For somewhat obvious reasons, a rule that simply increases the maximum possible permit term from 5 to 30 years isn’t the likeliest subject of a lawsuit in federal court.  Whose injury by that rule would be direct and documentable enough to give him or her standing under Article III to bring such a suit?  Without actual permits in particular locations, that is very hard to figure.

Of course whether a legal challenge could be sustained in federal court is completely separate and apart from the legality of the underlying move pursuant to NEPA.  Putting aside the court’s standing analysis in Shearwater for purposes of argument, thus, it is the latter question on which the court’s opinion sinks or swims.  And it certainly doesn’t sink easily.  The opinion is a careful exploration of CATXs and the appropriateness of a CATX in the 30 year rule proceeding.

When Increasing the Duration of a Permit Changes Its Expected Effects in the Environment

One qualitative difference between a 30 year and a 5 year duration on permits of this kind is obvious: permits that must be renewed must be opened up to public scrutiny, however minimal, and every 5 years is a (relatively) short interval.  A less obvious but, in my view, more serious difference is how a permit allocates the burden of uncertainty.  As Judge Koh concluded in her opinion,

under the old five-year regime, a permittee would be required to satisfy the criteria for permit renewal every five years.  Under the new thirty-year regime, however, the burden shifts to FWS to determine [at the 5 year “review” intervals FWS substituted for actual permit expirations] whether any changes to the permit are necessary.

Slip Op. at 31.  Indeed, if this wasn’t the effect, and if it wasn’t the all important change the rule was made to make, it is hard to imagine why the wind industry pled as insistently as it did that the permit durations be changed.  But what will this reallocation of uncertainty’s burdens actually cause in the environment?  That is the key question: NEPA § 102(2)(C) requires projection and analysis of effects in the environment. Will realigning uncertainty’s burdens provoke such an impact?

Judge Koh finessed the issue by naming the defect in FWS’s rulemaking one of “explanation.”  FWS failed adequately to explain its conclusions.  See Slip Op. at 34.  Plenty of Ninth Circuit and SCOTUS precedent under the Administrative Procedure Act back up that maneuver.

But the question is worth asking in NEPA’s interests alone, unalloyed by that aspect of our legal system.  Suppose FWS grants a 30 year permit after what seems like an exhaustive, comprehensive analysis of the risk combined with a careful field test of turbines at a site, finding no evidence whatsoever of bird mortality within the confines of the permit’s parameters.  All we ever have in such cases is an absence of evidence.  That is, as statisticians are fond of reminding, not the same thing as evidence of absence.

Indeed, according to Pagel et al., considering only those cases where hard, physical evidence of mortality was collected, they were able to locate a “minimum of 85 eagle mortalities at 32 wind energy facilities in 10 states during 1997 through 30 June 2012Pagel et al. table (excluding the outlier Altamont Pass in CA).  See Joel E. Pagel et al., Bald Eagle and Golden Eagle Mortalities at Wind Energy Facilities in the Contiguous United States, 47 J. Raptor Res. 311, 312 (2013).  See their Table 1 at right.

So how to estimate effects?  As Huso and colleagues observe in their piece on a suggested statistical analysis for dealing with such uncertainties,

[w]hen probability of detection is high, observing no carcasses can be construed as evidence that no or few animals have been killed, i.e., evidence of absence.  When probability of detection is low, however, finding zero carcasses does not credibly rule out the possibility if a large take.  We are left with only absence of evidence.

Manuela M.P. Huso et al., Estimating Wind-Turbine-Caused Bird and Bat Fatality When Zero Carcasses are Observed, 25 Ecol. Apps. 1213, 1223 (2015).  Differentiating between those locations and/or applicant/operators is a matter for FWS’s expertise, its many partners’ data-gathering capacities, and the public.

But this is where trusting applicants and their search for evidence of wildlife mortality may be quite inferior to the continuous updating that comes with short-lived permits.  Often the tests applicant/operators employ are either designed to fail or somehow just accidentally find no carcasses—and the statistically silly inference therefrom (no effect) is too often the result.

Unless they appeal, it seems that FWS is going back to the NEPA books on this one.

I teach environmental, natural resources, and administrative law at Penn State Law. Before teaching I was an enforcement lawyer at U.S. EPA. Along the way I've done work for environmental nonprofits and written a fair bit about NEPA.
No Comment

Leave a Reply