Friday, April 8, 2011

Life After Death At Yellowstone: An Interview with Josh Miller

ResearchBlogging.orgIn my last post, I introduced a ground-breaking study recently published in PLoS ONE, that shows how we can infer long-term trends in animal populations just from their bones. This work has big implications for ecology, conservation, and public policy, and is also a really neat piece of science. For this post, I talked to the author of the study, Josh Miller, about his work and some of the tidbits that didn't make it into the paper.

Yellowstone NP gets a lot of visitors, and you surely must have had some interactions with them during your fieldwork. How did they react to what you were doing?
JM: I work in areas that are generally well off trail and in places most Yellowstone visitors just don't see. Over the years, there are have been very few times when tourists actually ever saw my teams conducting our bone work. Most of the time, conversation with the public occur in the evenings back at camp. We generally use the public campgrounds for our homebases and my research will often come up in conversation with tourists. When folks learn what my teams and I are up to, they are always very interested and ask lots of questions. Our National Parks are an important resource, and I think people like to be reminded of their biological and scientific value. At the same time, I think it gives folks a way of looking at Yellowstone in a new and exciting way. I know lots of people who talk to us one day and keep an eye out for bones the next.

Miller studying bone survey data sheets on Northern Range, Yellowstone National Park. Photo by Scott Rose.

You looked at hundreds of bones during your survey. Was there any particular specimen that stuck out in your mind? What about it was interesting?
JM: I looked at over 20,000 bones during my work in Yellowstone. And you are right, there are a few that really stand out. Some of the most memorable bones are those of animals with severe bone maladies. In some individuals we found severe arthritis or broken bones that didn’t heal properly. Other memorable bones include rare and unusual species. One of the most exciting finds was the skull of a mountain lion. We just stumbled upon on it one afternoon walking from one transect to another. This beautiful rounded huge cat skull just lying in the grass staring up at us –a rare and amazing site.

This paper focused on bones from large animals, but surely there are a lot of small animal bones out there too - rodents, bats, rabbits, etc. Do you think they would show a similar correlation over time between abundance in life and death? Or are the taphonomic effects too different between large and small animals to expect the same pattern?
JM: Stay tuned! I kept careful attention to the bones of the small mammals we found. My bone survey teams were amazingly good at finding bones of all shapes and sizes (from bison skulls to limb bones of squirrels). One of the challenges, unfortunately, is the lack of high-quality data on the living populations in Yellowstone. One thing I'll say at the moment, however, is that the record of small-bones is surprisingly rich and diverse on the Yellowstone landscapes.

I see that you used the open source stats program R to do your data analysis. Was this something you picked up just for your dissertation work? Why did you choose R over some of the other commercial packages that are out there?
JM: I was introduced to R during the early days of my graduate work. R is a very powerful statistics language, in part, because of the large community of scientists and academics that use R and contribute to its ever-expanding utility. Another reason I use R is that I can completely control all aspects of the analysis. In canned programs, much of the analysis sits under a black box and uncovering exactly how the data were analyzed can be very difficult. But most of all, R just fits how I do science.

Thank you for your time, Josh!

Miller, J. (2011). Ghosts of Yellowstone: Multi-Decadal Histories of Wildlife Populations Captured by Bones on a Modern Landscape PLoS ONE, 6 (3) DOI: 10.1371/journal.pone.0018057

Note: I'm an academic editor at PLoS ONE, but had no role in the handling of this paper.

Sunday, April 3, 2011

Life After Death at Yellowstone

ResearchBlogging.orgTaphonomy - the study of what happens to an organism after it dies - is integral to reconstructing the past. Perhaps the most important lessons come in inferring ecological interactions. Did that group of animals live and die together, or were they jumbled long after death? Were all of those shark teeth with the plesiosaur bones from a feeding frenzy, or just a fluke of currents? How closely does a set of fossils represent the relative abundance of the different species during their lifetime? Such examples are numerous, and thus we commonly think of taphonomy as a study in deep time. This is certainly true, but also certainly incomplete. In fact, some of the most ground-breaking taphonomic work has been done in contemporary ecosystems. Kay Behrensmeyer, for instance, has spent decades studying bone accumulations in Kenya, and a 1927 work by Johannes Weigelt (complete with photos of dead cattle) is still considered a classic.

A new study by paleontologist and taphonomist Josh Miller, just published in PLoS ONE, shows some of the great insights that can arise from looking at taphonomy in modern settings. Josh and his field assistants trekked through Yellowstone National Park (one of the western USA's oldest and best-known parks), cataloging the identity and physical condition of every animal bone sitting out on the surface (an elk skeleton from the project is shown at right; photo courtesy of and copyright Josh Miller). Using these data, Josh found that you can actually infer the major ups and downs of animal populations from their old bones. This is quite exciting, not just from a gee-whiz factor, but because it may be possible to infer population trends for areas where historical surveys are absent or spotty. Such data are important not only for ecologists, but for informed public policy. It sounds magical, so how was the study done?

Based on other studies (in combination with radiometric dating), it's known that bones in excellent condition usually came from animals that died only recently, whereas bones in crummy condition are from animals that died longer ago. By using the condition of the bones as a proxy for time since death, Josh estimated how long the various bones of various animals had been around. Then, based on the bone ages, he estimated the relative population of each type of animal a given number of years ago. We have very good wildlife census data for Yellowstone, and it turns out that estimates from the bones match the "real" values quite nicely. Boom years for animals (such as elk) mean lots of bones going into the system, bust years mean few bones, and these trends shows up in bone surveys.

You can read all about it at PLoS ONE, or here, here, and here. I recently talked to Josh to get a few behind-the-scenes tidbits. Stay tuned for the interview later today! [update: now posted here]

Miller, J. (2011). Ghosts of Yellowstone: Multi-Decadal Histories of Wildlife Populations Captured by Bones on a Modern Landscape PLoS ONE, 6 (3) DOI: 10.1371/journal.pone.0018057

Note: I'm an academic editor at PLoS ONE, but had no role in the handling of this paper.