Bumble and Bumble: what’s black and yellow and maybe more than one species?

During the dark afternoons of December in New England, I like to scroll through my old field photos and think of all the green, growing things I’ve measured in beautiful places during those long-ago long-lit seasons. Yesterday I flipped through a couple field photos from a friend — “Photos of younger Jon! :)” he wrote in the email — and the same sunny feelings flooded in.

As a master’s student*, Dr. Jon Koch and his insect net chased bumble bees all over the western United States. He was studying bumble bee decline, but hit weird hurdle: a messy species boundary between two bumble bees. Taxonomists and field guides were torn on whether Bombus fervidus was or was not Bombus californicus. These two “species” in the Bombus fervidus species complex were nearly morphologically identical, except for their color patterns: B. fervidus is noted as usually mostly yellow with a little black, while B. californicus sports mostly black with some yellow in variable detail. They were maybe different species, maybe hybridizing, or maybe the same thing with different color morphs. As Jon explained to me, “If we don’t know what the species are, how will we manage them? Bumble bees are differentially sensitive to land use change, disease, etc. The bumble bees in the Bombus fervidus species complex are found to be impacted by one disease, Nosema bombi, but perhaps differently. Therefore, it is important to recognize what the species boundaries are because estimates of infection prevalence might be not be done correctly due to the inability to tell the species apart.” 

Jon wanted to bring some clarity to the species complex by providing some new molecular evidence with broader taxa sampling. His new PLoS ONE paper, “Phylogeny and population genetic analyses reveals cryptic speciation in the Bombus fervidus species complex (Hymenoptera: Apidae)” delivers on the broader taxa sampling — 320 specimens from 53 sites — but the clarity is a bit of a cliff hanger. During the fieldwork, Jon and his coauthors keyed out identifications for their bees based on the setal color, and also took a tarsal clipping from the mid-leg for DNA extraction and microsatellite genotyping. When they compared field identification to the genotypes, they had an ID rate of just under 94%. Jon and I agree that that’s a pretty good record for fieldwork with cryptic species** but he adds, “it’s also cool to think that 6.2% of the time we were wrong! These bees are great at fooling us.” 

The bees that were fooling Jon were B. fervidus dressed as B californicus and vice versa. In Pinnacles and Yosemite National Parks there were ten mostly black bees (the typical B. californicus look) that turned out to belong to the genetic cluster that usually wears mostly yellow. The rest of the bees with black setal coloration belonged to another clade based on genotype, though this clade also included some bees in yellow. I asked Jon, “What is going on with the bee costume parties in Pinnacles and Yosemite?” His wild speculation is that little black dress is the dominant phenotype for bees in these parks, and the typically-yellow-genotype wears black here because everyone else is doing it: “bumble bees are notorious for converging on a local phenotype, which can even make it very hard to tell distantly related species apart.” However, in the sites where both genetic clusters of the B. fervidus species complex overlap, they usually do not look alike, so they aren’t mimicking each other. 

Ultimately, Jon’s team determined that the species complex comprised two lineages, but that both lineages exhibit the yellow and black phenotypes depending on geography. So while the B. fervidus species complex is not a single species, B. fervidus and B. californicus are not NOT conspecifics. Jon explains, “those names [B. fervidus and B. californicus] might not even be valid! The holotype of B. californicus happens to be where the genotype assigned to the “B. fervidus” was collected in the Sierra Nevada.” In short, the original bee that taxonomists knew as B. californicus may actually be genetically on the B. fervidus side of the lineage, and eventually one or both names might need to be thrown out.

This “it’s complicated” conclusion might be depressing news for someone who dedicated so much time and energy towards disentangling the species complex, but Jon closes his email to me with a happy emoji “nature has so many surprises, and science is an ongoing process :)” In the meantime, this paper points out that even if we don’t have the right names in place, we know enough to recommend that managers use Jon’s non-lethal method of clipping a bit of mid-leg for genotyping, and monitor the two clades of the B. fervidus complex separately. This is a great reminder for all of us in conservation research: we need to keep the ongoing process in perspective, while also delivering our findings, however not-quite-as-clear-as-we-hoped or maybe-unnamed as they may be, to our partners in management and policy. 

References:Koch JB, Rodriguez J, Pitts JP, Strange JP (2018) Phylogeny and population genetic analyses reveals cryptic speciation in the Bombus fervidus species complex (Hymenoptera: Apidae). PLoS ONE 13(11): e0207080. 

*Now, old Jon and old Caitlin are David H. Smith postdoctoral fellows together :)

**see McDonough MacKenzie et al. 2017 — When I was a master’s student working with volunteer-collected data I would have killed for a 93.8% identification rate. One my species, Labrador tea, was correctly identified 27.3% of the time. This is not a cryptic species; it doesn't sometimes dress up as Diapensia. 

Pikas Meet Cute: Two Subspecies, One National Park

The National Park Service is wrapping up celebrations on its 102nd anniversary this August. I’m unabashedly biased towards park science: my dissertation and my postdoc research are both Acadia-based, while cleaning out old papers last week I actually paused for a moment before recycling a torn up, coffee-stained copy of a National Park research permit from 2013. (Don't worry, the original pdf is safely stored on an external hard drive.)

I’d report on the hybridization of pikas in Rocky Mountain National Park even without the excuse of a belated happy birthday to the National Park Service, but clearly covering research on pikas and #poopscience is the perfect way to honor the stewards of our public lands. There are charismatic megafauna (mini-fauna?) and there are charismatic landscapes, and the scientists who study pikas in the western National Parks enviably have cornered the market on both. Dr. Jessica Castillo Vardaro just published new research on the population genetics of American pikas in PLoS ONE last month. In “Identification of a contact zone and hybridization for two subspecies of the American pika (Ochotona princeps) within a single protected area” Castillo Vardaro and coauthors analyze the DNA in pika poop to pinpoint where the northern and southern Rocky Mountain lineages of these rabbit relatives meet. Their pika #poopscience spanned samples from Grand Teton National Park, Great Sand Dunes National Park, and Rocky Mountain National Park.

Before Castillo Vardaro’s work, there was some evidence that the northern and southern Rocky Mountain pika subspecies had a historic contact zone somewhere near-ish Rocky Mountain National Park. However, Castillo Vardaro wasn’t looking for a contact zone or hybrid pikas when she began working on the Pikas in Peril (PIP) project — a team of National Park Service staff and academic researchers. Pikas are a bit of poster child for climate change vulnerability — “a climate indicator species” — because they cannot tolerate prolonged exposure to high temperatures. Castillo Vardaro’s initial genetic analyses of pika populations in western National Parks focused on signals of isolation by distance (IBD). She explains, “the further individuals are apart geographically, the less related they are genetically. Since pikas typically establish territories close to where they were born and mate with their neighbors, I expected to see strong signals of IBD. I did in all of my study sites except Rocky Mountain National Park (ROMO).” Comparisons of the pika samples and their sequences to Genbank showed that there were two genetic lineages represented in ROMO — Northern and Southern. Then, at a pika meeting (could there be a cuter meeting?) Castillo Vardaro met Preston Somers, a researcher who studied pika dialect in the Rockies in the 1970's. She notes, “His work suggested there might be a contact zone, but we were the first to actually show it and evidence of contemporary gene flow. So, we weren't initially interested in studying ROMO as a potential contact zone, but we are now.” The analyses in this research are steeped in #poopscience, or what the paper refers to as “fecal samples…through a combination of random, targeted, and opportunistic sampling.” I asked Castillo Vardaro about the trade offs of #poopscience versus tissue samples. As a plant ecologist, my Methods have never included gems like, “We avoided collecting old fecal pellets by preferentially collecting pellets with green plant material inside to avoid degraded DNA” — but I was curious to hear more. Castillo Vardaro expounded,

Fecal DNA is essentially the mucus and cells lining the digestive tract that then coat the fecal pellet as it passes through. There are very few cells compared to tissue (organ tissue or ear clips), there are other things present that can inhibit the PCR process like plant secondary compounds, and the feces has been sitting around outside for an unknown amount of time so the DNA can degrade. Each sample has to be genotyped multiple times to overcome the errors resulting from low quality/quantity DNA. My genotyping success rate was 50% - 75%, after removing samples that failed, contaminated samples, and multiple samples collected from the same individual unknowingly. That's a lot of work in the lab.

But, the #poopscience lab work pays off if you need lots of samples across a broad geographic area:

In contrast, I just got back from a week in Montana where I was helping my coauthor Chris Ray trap pikas at a site she has been monitoring for 30 years. In four days of effort (two trap days, but it takes a day to set up traps and a day to check traps/process pikas) we trapped 5 pikas. One person can collect 10-25 quality fecal samples in a day, plus anyone can collect fecal samples for genetic analyses after about 10 minutes of training. So while I would have preferred to have worked with tissue, there is no way to sample the number of individual pikas necessary for 10 high resolution genetic analyses if you had to trap every animal.

The collaborative nature of Castillo Vardaro’s research and the Pikas in Peril reminded me of an earlier blog post I wrote about the Biological Conservation paper “The importance of non-academic coauthors in bridging the conservation genetics gap.”  I noticed that Castillo Vardaro’s PLoS coauthors were all academics, but she pointed out that her coauthor and grad mentor, Clint Epps, designed the PIP project alongside National Park Service personnel and other academic researchers. “The questions, goals, and desired products were explicit from the beginning. These included National Park Service reports, summaries, briefs (publications on the web and available at the parks themselves), spatial data, and research that could be utilized in each of the parks.”

While Castillo Vardaro was doing field work, she worked with National Park Service and US Fish and Wildlife Service biologists, interns, and volunteers. She noted, “we worked with interpretive staff to prepare the park specific resource briefs. We (myself, Clint Epps, and Doni Schwalm) also wrote a note on the potential effects of a proposed quarry site in Grand Teton National Park on the pika populations there, which was provided to resource managers there.” Basically, this work (one of Castillo Vardaro’s dissertation chapters) is the exception that proves the rule to the non-academic coauthors paper: here, the coauthor list belies the strong partnerships with non-academic scientists and managers, and if you didn’t know about Pikas in Peril, you might think wow, these academics really know how to put together explicit management implications single-handedly! 

Finally, in Castillo Vardaro’s research I saw a mirror of my own dissertation work. I had no pikas or fecal DNA, but we both finished our dissertation field work in National Parks before the 2016 election. Her work could inform whether pikas are listed as endangered or threatened under the Endangered Species Act; my research supported a climate change vulnerability assessment; and after our halcyon days as PhD students under the Obama administration, we are now watching an administration and Secretary of the Interior generally disregard the National Park Service expertise on these issues.

I told Castillo Vardaro that I feel an extra sense of urgency in publishing my Acadia papers now — especially in open access venues. I wondered if this was a personal quirk or if she felt a similar sense of responsibility for her field sites and study species. She agreed that highlighting the work that we are doing on public and federally managed lands is even more important in the current political climate. “One of the main reasons I chose to publish in PLOS ONE was because I wanted the manuscript to be accessible (open access).” She also noted that, “the PIP project was funded as part of the NPS Climate Change Response Program. I do worry about continued funding for similar projects and initiatives under Zinke and the Trump administration. Pikas tend to live in places that aren't as directly impacted by development as other ecosystems (it would be difficult to put a subdivision on the steep, rocky, side of a mountain), but the policies and proposed changes to the Endangered Species Act under the current administration to make it easier for development and resource extraction on public lands could definitely impact pikas.” 

The flipside of non-academic coauthors bridging a conservation gap is this: when the federal government is hostile towards non-extractive natural resource management, the academic coauthors in these partnerships will continue to publish our findings, piling up the evidence to support our field sites and our study species. For those of us in academia who completed National Park fieldwork in what seems like another era, getting the writing done can seem both daunting and futile. It's not. Traditionally, the first wedding anniversary is the “paper” anniversary, but for the National Park Service’s 102nd I think papers are still an appropriate — and important — gift. 

References:

Castillo Vardaro JA, Epps CW, Frable BW, Ray C (2018) Identification of a contact zone and hybridization for two subspecies of the American pika (Ochotona princeps) within a single protected area. PLoS ONE 13(7): e0199032.