Friday, September 20, 2013

The Proposed Removal of Gray Wolves from Endangered Species Status in the U.S. is Arbitrary, Capricious, and Inconsistent

21 July 2013


Gray-Wolf-picOn June 13, 2013, the U.S. Fish and Wildlife Service (FWS) proposed removing the gray wolf, Canis lupus, from the List of Endangered and Threatened Wildlife. Under this proposal, the species of gray wolves – which was at one time protected across the 48 contiguous U.S. states – would no longer be so protected. Only one subspecies of gray wolf would be granted protection: the Mexican wolf (Canis lupus baileyi). This would, in effect, protect only 75 of the wolves in the U.S. (the current size of the wild population of Mexican wolves). The FWS is soliciting comments on this proposed ruling here, where you can also find the text of the proposal. If you find this proposed ruling as problematic as I do, I would urge you to submit a comment.

Philosophers of biology such as myself, many of whom are well-versed in the challenges of defining the terms "species," "subspecies," and "population," not to mention skills in evaluating arguments, are particularly well placed to see the flaws in the proposed ruling. I was first alerted to potential problems in the proposal upon reading this editorial, which blamed the ruling on pressure from "a loose coalition of hunters' groups, outfitters, and ranchers." While I don't have any evidence for this assertion, after reading through the ill-defended proposal one has to wonder, "Why this? Why now? And why has the fact-finding process of the Endangered Species Act been corrupted with a 'government-manufactured scientific consensus'?" (thanks to Eric Schliesser for the felicitous phrase).

On the basis for the proposal
  One interesting feature of the Endangered Species Act, on which all such proposals are based, is that whereas the Act's implementing regulations define "distinct population segment" (more on this later), they do not, as far as I can tell, define "species" or "subspecies." Instead, decisions are to be made on the basis of "the best available scientific information." The problem is, as the authors of the proposal acknowledge, that there is a lack of consensus among scientists on what species are, what subspecies are, and how many species and subspecies of wolves there are. Nonetheless, they declare that one paper, Chambers et. al (2012), "is the only peer-reviewed synthesis of its kind conducted for North American wolves and summarizes and synthesizes the best available scientific information on the issue." The paper takes "an integrative approach to species that encompasses the concordance principles of Avise and Ball (1990)... an eclectic approach that seeks to identify species as separate lineages supported by concordant data from various classes of genetic markers, morphometric analysis, behavior, and ecology." This is a respectable approach to delimiting species. However, it is only one contender among many; by one count, there are 26 species concepts in the literature.

Chambers et al. (2012) appears in the journal North American Fauna, a publication of the FWS itself; it is unclear why the paper wasn't sent to a more recognized peer-reviewed journal in the field, such as Conservation Biology. According to the website for North American Fauna, Chambers et al. (2012) is the only publication since 1991; it appears as though the journal was reborn specifically to publish the wolf study only to languish again afterward. Moreover, the authors all work for the FWS: S.M. Chambers (Albuquerque, New Mexico), S.R. Fain (Ashland, Oregon), B. Fazio (Albuquerque, New Mexico), and M. Amaral (Concord, NH). The proposal itself does not name authors, but includes FWS employees from places such as (hmmm) Albuquerque, New Mexico. In short, the FWS service has taken one of the most well-studied animals, about which there is great controversy (see, e.g., a letter of opposition signed by sixteen scientists with expertise in carnivore taxonomy and conservation biology), and benighted one paper, authored by itself (perhaps by some of the same people who wrote the proposal) as "the best available scientific information."

Wolves are hard to classify because all of the purported species (much less the purported subspecies) can and do breed with one another, as do wolves and coyotes. Furthermore, they disperse great distances; 65 to 154 km (40 to 96 miles), with dispersal distances of several hundred kilometers occasionally reported. Again, one wonders, what is going on? What is the urgency – so urgent that the FWS must hastily designate itself as the source of the best available science – to make policy when the science is so unsettled? One would think that, given the uncertainty and controversy over species in general and wolves in particular, it would be better to err on the side of protection. The FWS might at least have waited a few years to see if there was any critical response to Chambers et al. (2012). Instead, in a rather smelly self-citing act, the FWS has implied that the science is settled enough for us to lift protections that have been in place since 1978. This is capricious.

Now let's turn to the details of the proposed ruling itself. I'll focus on two main issues, first, the determination that gray wolves in the Pacific Northwest are not a "distinct population segment" and second, the delisting of Canis lupus as a whole.

On not listing the Pacific Northwest gray wolves as endangered
According to the proposed ruling, there have been sightings of at least 3 separate packs of adult wolves in Washington state:
  • Lookout pack: at least four adults/yearlings and six pups. The pack produced another litter of at least four pups in 2009, as well as a probable litter in 2007. Since then, however, the pack appears to have suffered significant human-caused mortality from illegal killing, which effectively dispersed the rest of the pack.
  • Teanaway pack: three adults and four pups. Reproduction was confirmed in the Teanaway pack in June 2012.
  • Wenatchee pack: two wolves, one thought to be a dispersing animal from the Teanaway pack. Reproduction has not yet been documented in this pack
Wolves have also dispersed to Oregon and California, but without documented reproduction.
Quoting from the proposal:
The Act's implementing regulations define a "population" as a "group of fish or common spatial arrangement that interbreed when mature" 50 CFR 17.3). That group may consist of a single collection of organisms, or multiple loosely bounded, regionally distributed collections of organisms all of the same species or subspecies.
We have refined that definition in experimental gray wolf reintroduction rules to mean "at least two breeding pairs of gray wolves that each successfully raise at least two young" annually for 2 consecutive years (59 FR 60252 and 60266, November 22, 1994 [unfortunately, not online, or I would link to them and discuss]). This definition represents what we believe are the minimum standards for a gray wolf population.
Given that the above described wolves do not meet the "refined" definition, the FWS asserts that the Pacific Northwest wolves do not constitute a "distinct population segment" (DPS).
However, this "refined" definition is highly problematic.

First, it dramatically narrows the original Endangered Species Act (ESA) understanding of a population, which allowed for a population to consist of "multiple loosely bounded, regionally distributed collections of organisms all of the same species or subspecies," with no reproduction requirement. More precisely, the ESA seems to have had in mind a metapopulation, i.e., a "population of populations" (Levins 1969). In a metapopulation, some organisms may be dispersing from one population to another or be in the process of forming a population without, at a particular moment in time, belonging to any one population. The Pacific Northwest wolves seem to clearly fit the ESA's characterization.

Second, it's not clear what the scientific basis is for the "refined" definition. Again, I unfortunately do not have access to the relevant documents, but as I have argued elsewhere, populations and metapopulations ought to be characterized in terms of the interactions between the organisms. These interactions include, but importantly are not limited to, reproductive ones: they include survival interactions, and, for metapopulations, migration from one population to another. Previous rulings have determined that "by definition lone dispersers do not constitute a population or even part of a population, since they are not 'in common spatial arrangement' sufficient to interbreed with other members of a population." However, it makes little sense to characterize a widely-ranging species like wolves in geographic terms; a process-based concept, such as the one I have defended, is more appropriate. (I would argue that it is more appropriate in general, but I need not make that case here

Third, and perhaps most importantly, the "refined" definition creates a Catch-22-like situation. Note that the Lookout pack might have easily met the DPS criterion if five of its members had not been killed. This creates a situation where new populations cannot be established because they are not protected and thus subject to being killed by those who do not want such protections to be in place.

(Note that gray wolf populations are so diminished in the U.S. because they were targeted for complete eradication, a goal that some still harbor). What is the biological significance of two breeding pairs of gray wolves that each successfully raise at least two young annually for two consecutive years? This is an arbitrary and self-defeating dividing line that undermines the very purpose of the ESA.

On not listing the Canis lupis species as endangered
Turning to the issue of the delisting of Canis lupus as a whole, it is important to understand the background rulings. In 1978, gray wolves were listed as endangered across the entire 48 contiguous U.S. states, under the understanding that this constituted their historic range. Since that time, gray wolf populations in the Northern Rocky Mountain and Western Great Lakes areas were deemed "recovered," thus removing protections for wolves in those areas, allowing their "regulated" killing (e.g., Wyoming protects wolves in only 15% of the state, allowing their eradication elsewhere). But protections remained for gray wolves across the rest of the U.S. The current proposal, in line with the previously discussed Chambers et al. (2012), asserts that, whereas we previously thought the Eastern wolf was a subspecies of gray wolf (Canis lupus lycaon), we should now understand that the Eastern wolf is a separate species (Canis lycaon). Thus, with the assertion of one FWS article, all of the wolves in the eastern U.S. are deemed "not listable" as part of Canis lupus. Note that there was no formal taxonomic revision, and again, note that this re-classification is highly controversial and contestable. As for the status of this new Canis lycaon species, this is unaddressed by the document; it is left in limbo. This seems inconsistent with the goals of the ESA, which are to protect species that are endangered, not to lift protections via a mere reclassification.

Dark gray areas show current locations of C. lupus. Hatched areas represent the boundaries of the only populations recognized as DPSs under the proposal. Gray areas represent the purported historic range of gray wolves, with the eastern states left unhighlighted given the proposal's claim that those areas are the historic range of a different wolf species.  Only the Mexican Wolf Experimental Population Area would retain protections under the proposal.

And what about the rest of the U.S. gray wolves? According to the proposal, they are either in one of the two "recovered" populations, or they are dispersers from those populations (and dispersers don't count), or they are in the Pacific Northwest (so again they don't count), or, they are one of the special 75 Canis lupus baileyi, which are the only wolves to receive protection under the proposal. Thus, there is no need to list Canis lupus, or any Canis lupus subspecies other than Canis lupus baileyi, or any DPSs. Why is that, given that there are areas where wolves lived historically where they are not present now? Bizarrely, the proposal seems to suggest that there should only be protections in areas where there are now Canis lupus species, subspecies, or DPSs. But this is inconsistent with past practice. To cite one well-known instance, wolves were reintroduced to Yellowstone under the auspices of the ESA in 1995 after they had been completely extirpated there by the early-mid-1900s. The touted "recoveries" would never had occurred had listings only been limited to areas where wolves were present. As this article points out:
The Endangered Species Act requires federal agencies to conserve threatened and endangered species until they are no longer in danger of becoming extinct "throughout all or a significant portion of" their range. Wolves' historic range included nearly all of the contiguous United States. While much of the American landscape has changed and wolves will never occupy their entire former range, significant chunks of suitable habitat remain, including in California, the Pacific Northwest, the southern Rockies, and the Northeast. Federal agencies have not fulfilled their congressional mandate until gray wolves are recovered in these areas.
The proposal admits that "Our understanding of the historical interactions between subspecies or genetically different populations (e.g., Leonard et al. 2005) is that they are dynamic processes and boundaries are in constant (and continuing) flux." Given what we know about wolves – they interbreed across subspecies, they can disperse great distances to quickly expand and colonize new areas – it makes sense to consider the current range to be the historic range. It does not make sense to grant that processes are in flux while only considering current range as protectable, especially given that the actions of hunters are largely what has prevented and is preventing expansion of the current range.
In short, the proposed removal of gray wolves from endangered species status in the U.S. is arbitrary, capricious, and inconsistent. It should be rejected.

Acknowledgements: Thanks to Mike Ghiselin's Pizza Munch Group at the California Academy of Sciences and the Griesemer/Millstein Philosophy of Biology Lab for helpful discussion about the proposed ruling. Some members also sent me valuable comments on a draft of this posting.