Game of Wolverines, Part One: Is Winter Coming? And Why Does it Matter?

So, what’s going on with the wolverine listing decision and the comment period being reopened? What’s the debate that warrants further comment from scientists and the public? I don’t know what prompted the USFWS to reopen to the comment period, but it is clear that there were disagreements over the science during the peer review and the comment process earlier this year. I’ll make an attempt to summarize the debate as I understand it, starting with the major papers (this post) and then looking at the scientific details that generated discussion (later this week.) Again, I do not know what’s going on, but for people who are interested in commenting, it helps to have a clear understanding of the state of the science. Since the discussion that occurred in the review reflects a bigger set of questions within the science, which I’ve been meaning to address for a while, the reopening of the comments provides an opportunity to look closely at the current state of wolverine research.

For background, remember that wolverines are hard to study; they’re a naturally rare, difficult-to-detect species living, here in the contiguous US, in some of the most challenging terrain in the country. Serious research on the Rocky Mountain wolverine population began only within the past two decades; prior to that, the state of Montana carried out track surveys and tried to monitor the population by assessing trapping take, but, while these activities provided insight into areas of presence, they offered no understanding of demographics or ecology. The scientific literature on wolverines remains comparatively sparse, and we still don’t have good information on demographics in the Rockies. Advocates petitioned to put wolverines on the Endangered Species List in 1994 and 2000; both petitions were denied on the basis of lack of adequate information. A petition denied in 2008 stated that Rocky Mountain wolverines were not distinct from the population in Canada and were therefore not a distinct population segment (DPS) listable under the ESA, an assertion for which there was no published substantiation; there was widespread speculation that this decision was based on the Bush administration’s ideological objections to the ESA. The current proposed rule relies on information that is finally within the body of published science, particularly a series of papers out of the Rocky Mountain Research Station that draw connections between wolverine dependence on snow and cold, and the future threat of diminishing snowpack. A number of other relevant, supporting papers exist, but the RMRS papers are at the heart of the finding that wolverines are threatened by climate change and warrant protection under the ESA. Trapping was found to be a secondary threat, in light of synergistic interactions with climate change. There is very little published information on the effects of trapping on wolverine populations, so this is a tricky topic.

Seven individual researchers participated in the peer review, all of whom are authors on papers that were considered as part of the listing process, and/or who have expertise in other areas. Five of the reviewers (Copeland, Aubry, Schwartz, Squires, and Zielinski) work or worked for the Forest Service; two (Magoun and Inman) do not, having worked more closely with state agencies and with research organizations.  All of these individuals are respected scientists with years of experience working on wolverines or, in the case of Zielinski, other members of the mustelid family. The first five reviewers found that the summary of the science and the conclusions reached in the listing rule were accurate and comprehensive, with minor adjustments to technical language. Magoun and Inman, however, questioned the analysis of the science and the conclusion that wolverines should be listed. In addition, the state of Montana submitted comments heavily critical of the proposed rule, and Wyoming and Idaho joined in to oppose federal protection. The alignment of the states against the federal government will surprise no one familiar with the script of carnivore conservation in the West, particularly since Montana was seeking to protect its limited trapping season, which is currently suspended under court order, pending the outcome of the listing decision. The critiques by Magoun and Inman, however, delve into some scientific issues that once again reflect on our limited knowledge of the species.

Here’s the background.

In 2010, Jeff Copeland and a number of his colleagues, including co-authors Audrey Magoun and Bob Inman, published a paper, The Bioclimatic envelope of the wolverine; Do climatic constraints limit its geographic distribution?, that tied global wolverine distribution to late spring snowpack. The paper correlated wolverine distribution with snow that remains on the ground until the period between April 24th and May 15th, proposing that the biological reason for this relationship lay in the fact that female wolverines den in the snow. The paper also suggested that the relationship is obligate – that is to say, wolverines depend absolutely on snow dens, are obligated to den in snow, and cannot successfully reproduce without snow. Despite some fluctuation in observed birth and weaning dates, May 15th represents an approximate date by which wolverine kits were weaned and out of the den, which was why this date was chosen as the upper limit on snowpack persistence. The researchers created a map of snow cover through the use of satellite imagery, incorporating seven years of real-world data on snow cover, and when den sites were overlaid on the map, the match was 97.9%, with 69% of the dens falling within areas that were snow covered for 6-7 years out of 7. Twelve dens (out of 562) occurred outside of the mapped snow cover; the authors stated that these sites were investigated and that they were snow dens, but that the satellites were unable to map the area as snow covered due to canopy cover and the percent of the area that was snow covered versus the resolution of the satellite images. The snow cover map encompassed 89% of telemetry points from wolverine studies worldwide.

The researchers also created a temperature map that looked at regions with maximum August temperatures of ≤ 22º C, and found that 90% of summer telemetry locations fit within this map. This suggested that wolverines could be limited by hot summer temperatures as well as by spring snow distribution. Since these two factors often overlap, especially at the southern edge of wolverine distribution, “….the relative importance of these factors for limiting wolverine range becomes increasingly difficult to assess.” The paper indicated, however, that cold and snow were definitely required for wolverine presence and persistence on a landscape. When I spoke to Jeff Copeland shortly after the publication of this paper, he emphasized that snow denning wasn’t meant to be a single explanatory factor for wolverine dependence on cold, but that it served as an umbrella for numerous other ways in which the species might be tied to frigid environments. In other words, he explained, you could add a bunch of other parameters to the habitat model, but since none of them offered any further refinement to the fit of the model and the data, it would simply clutter things up. As habitat models go, this one was streamlined and very elegant. It was also a clear link, the first in the published literature, between wolverines and climate change.

2011 saw a second paper on wolverines and climate change, McKelvey et al’s Climate change predicted to shift wolverine distributions, connectivity, and dispersal corridors. This paper relied on the 2010 paper to assess the possible effects of a warming climate on wolverine distribution by looking at what climate change would do to snow cover. It predicted that wolverine habitat and dispersal routes would shrink substantially through the 21st century, resulting in further isolation, both geographic and genetic, of wolverines in the Rockies. The bulk of this paper involves interpreting and downscaling global climate models and cross-fitting them to the MODIS data; I’m not going to detail the methods here, except to emphasize that this paper drew on the Copeland et al 2010 paper to reach its conclusions, and looked at the entire area of snowcover mapped in that 2010 paper.

In 2012, Bob Inman and Audrey Magoun, along with colleagues Jens Persson and Jenny Mattison from Sweden, published a paper, The wolverine’s niche: linking reproductive chronology, caching, competition, and climate, proposing several other ways in which wolverines might depend on cold environments to thrive. The paper details wolverine reproductive chronology, assessing cases from research conducted across wolverine range and showing that wolverines give birth across a range of dates, but usually from late January to mid-March, with weaning between April and May. The exploration of reproductive chronology draws connections between food availability and periods of high resource demand for females (pregnancy, lactation) and kits (first summer post-weaning) and suggests that wolverines give birth comparatively earlier than other non-hibernating mammals in order to take advantage of pulses of nutrition at critical times – cached ungulate carcasses see females through lactation and then the kits are weaned in time to take advantage of more varied and abundant summer food resources. The paper also puts forth some interesting hypotheses about competition and food caching behavior – namely that wolverines are able to thrive in a harsh niche because their physiology gives them advantages over other predators, and because they are able to cache and store scarce food resources, since low temperatures preserve meat for long periods of time. Taken together, all of these ideas provide interesting and compelling elaboration on the relationship between wolverines and cold environments – they add texture and intricacy to the broader strokes painted by Copeland et al in the snow map paper. The hypotheses about food storage, reproductive timing, and competition, however, have never been tested. They point to intriguing areas for future investigation, but there’s no evidence, yet, that these relationships are definite.

I’ve never seen these two papers as being in competition or contradiction; they address wolverines’ links to cold climates at different scales, but I’ve always perceived them as complimentary. The Inman et al paper doesn’t explicitly critique Copeland, it puts forth a food-based hypothesis for correlation of wolverine distribution with cold and spring snow. Also perhaps of note in light of the current situation, the first sentence of the abstract of Inman et al reads: “Wolverines are demographically vulnerable and susceptible to impacts from climate change.” There is no quarrel in this paper with the notion that snow is an important characteristic of wolverine habitat, or of the notion that climate change will be a problem for the species. Nor is there a hint of the idea that there must be only one biological reason for wolverines to be tied to cold and snow; food caching at cold sites, snow denning, and physiological adaptations that allow wolverines to outcompete other predators in cold, snowy situations are all part of the same story.

These are the three papers at the core of the discussion. All of them come out of a scientific process that begins with observation and intuition: we know that wolverines are creatures of the north, they’re only found in places that are extremely cold and snowy. Why? The two most important things for survival of any species are resources and reproduction, so the distribution pattern must have something to do with one or both of those factors. How? This is where hypotheses come in, as different scientists come up with possible explanations and test them. It’s also where one of the limitations of scientific inquiry is highlighted: experiments can only test one hypothesis at a time, and two published papers exploring different hypotheses about the same relationship can be perceived as being in competition or offering alternative, rather than complimentary or synergistic, explanations for a phenomenon. Ecology is especially tricky because it’s hard to design experiments and control variables at a scale that will offer conclusive evidence one way or another.

All of this is fine as long as you stay within the realm of dialogue about knowledge, but once you try to pull that knowledge into a decision-making situation that demands facts or truth, you start to run into problems. My father, who is a scholar of the history of science and technology, defines science as a process of human inquiry that makes natural phenomena incrementally more visible to our limited human faculties. It is a pursuit that is fundamentally tied to deep uncertainty and subject to constant revision. It represents our best current understanding of a given situation or process, but it’s never absolute. Unfortunately, in the popular mind science is seen as a way of uncovering Truth, and expressions of uncertainty are seen as negating the validity of an entire hypothesis. Most of the social issues around science – notably the campaigns against evolution and climate science, but also trends in some branches of academia that dislike and wish to discredit science – exploit this misunderstanding. The issue of how far science can generate unassailable Truth or Fact creates tension when we appeal to science to help create policy. That’s where we are right now.

Stay tuned for the next episode, Game of Wolverines, Part Two: A Clash of Analyses

(With apologies to wolverine fans who don’t watch or read fantasy epics, and to any Game of Thrones fans who may have ended up here thinking that this has to do with the actual series. I started watching the show while stuck in Ulaanbaatar for a week with nothing else to do, and I persisted in watching it – truly – because I was trying to figure out whether some of the characters were wearing wolverine pelts as they traipsed around fighting battles. Because mindless books are a great way to spend 24 hours of travel time between the US and Asia, I ended up reading the series to find out what happened, and discovered some wolverine references in there, too. My major preoccupation with these books is wondering how I would design a wolverine research project in a world where winter and summer last for years on end – is it even reasonable to assume that wolverines would persist in such a world, given that a years-long summer would disrupt their breeding cycles and melt out their favorite food-caching spots? I really wish George RR Martin and HBO would address this pressing question.

None of this is meant to draw parallels between specific people and specific characters (since almost everyone in the series is reprehensible), but simply to allude to the complicated human dynamics and loyalties that are at play in any wildlife conservation situation. And also because it affords a fun opportunity to play word games with some of the titles and key phrases from the books. Readers of the books will get the references. Others should just shrug and focus on the content.)

Advertisements

2 thoughts on “Game of Wolverines, Part One: Is Winter Coming? And Why Does it Matter?

  1. Nice summary of a complex and potentially confusing situation. Thanks very much for writing and posting. In addition to the wolverine info, apparently the majority of Americans could benefit by hearing your brief explanation of science, as illuminated by your father. Looking forward to Part 2.

  2. Pingback: Game of Wolverines, Part Two: A Clash of Analyses | The Wolverine Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s