What I do
My work addresses fundamental questions in population, community, and evolutionary ecology. I am most interested in questions to do with system dynamics (changes over time) and the processes that drive them. Accordingly, much of my work uses a combination of mathematical modeling and experiments in tractable model systems such as laboratory microcosms. I use mathematical models because ecological and evolutionary systems are complex and nonlinear. It’s hard to develop and test rigorous hypotheses without mathematical help. I use microcosms because their high control and replicability allows powerful, rigorous experiments that would otherwise be impossible.
My lab is currently funded by an NSERC Discovery Grant.
Below are the main lines of research currently going on in the lab.
Spatially-separated populations of the same species often fluctuate synchronously, even though they’re hundreds or even thousands of km apart. The result is that, across vast areas, all populations increase (or decrease) simultaneously. Coexisting populations of different species also often fluctuate synchronously, even though there are strong reasons to think they wouldn’t. For instance, you’d think competing species would exhibit antisynchrony, since if one increases that should cause the other to decrease. Synchrony is pretty amazing when you think about it, and so cries out for an explanation. We have theories of why synchrony happens, but those theories are hard to test in nature because it’s impossible to do experiments at the right spatial and temporal scales. You can’t, say, manipulate the weather across all of Canada and then wait a century to see what happens to the spatial synchrony of lynx-hare cycles. The solution is to scale nature down and do experiments in protist microcosms. I’m currently seeking grad students interested in pursuing synchrony-related experimental and modeling projects.
Much of my work on synchrony is in collaboration with my former postdoc David Vasseur of Yale University.
Publications: Vasseur & Fox 2007 Ecol. Lett., Vasseur & Fox 2009 Nature, Fox et al. 2011 Ecol. Lett., Fox et al. 2013 PLoS One, Vasseur et al. 2014 Proc Roy Soc B
Local adaptation in time and space
Environmental conditions vary in space, imposing contrasting selection pressures that favor different phenotypes and genotypes. That can cause local adaptation, with each environment being dominated by the organisms that are fittest locally. An analogous process can occur in ecology, with different species competitively dominating in different environments. Local adaptation is a powerful way to maintain diversity. But the environment also varies in time. Can you have local adaptation in time, with different locally-adapted genotypes or species dominating at different times and thereby coexisting? The simplest theoretical models say no. As Graham Bell elegantly put it, spatial variation creates opportunities, but temporal variation imposes obligations. All else being equal, a trait that makes you well-adapted to today’s conditions at the cost of making you poorly-adapted to tomorrow’s conditions will not be favored by selection, because you’re obliged to experience tomorrow’s conditions. Of course, the world might be more complicated than the simplest theory assumes. But theories of local adaptation in time are hard to test; how do you tell how fit an organism would be in past or future environments? I’m testing for local adaptation in space and time using lake bacteria. Water chemistry varies a lot among lakes, and over time. By freezing bacterial isolates and water samples at -80 C, we can use the freezer as a ‘time machine’. That is, we can reciprocally transplant bacteria back into water from the past, and forward into water from the future. Pilot data suggest geographic variation in the strength of local adaptation (and maladaptation) in space and time. I’m currently seeking a graduate student interested in expanding this work to many more sites and times.
Publications: Fox & Harder 2015 Evolution
Local adaptation, species interactions, and range limits in mountain plants
Mountain slopes exhibit steep environmental gradients. These gradients affect adaptive evolution (e.g., by selecting for local adaptation), species interactions (neighboring plants generally compete with one another low down, but facilitate one another up high), and elevational range limits (different species are found at different elevations). But there’s been little work on the interplay of these. For instance, locally-adapted plants should find their local environment less “stressful”, and so be less likely to experience facilitation by neighbors. Conversely, insofar as neighbors facilitate one another by ameliorating harsh abiotic conditions, they should weaken or even reverse selection for adaptation to local abiotic conditions. If neighbors facilitate one another, does that mean they extend rather than limit one another’s elevational ranges? And how does the distribution of competition and facilitation along elevational gradients both affect, and reflect, the distribution of species along those gradients? My lab addresses these questions by reciprocally transplanting species within and beyond their elevational range limits, and by combining these reciprocal transplants with neighbor removal experiments.
Quantifying macroevolutionary forces using the Price equation
Fossil data record the macroevolutionary history of life on earth. Those data often record dramatic changes in major clades, such as the rapid dwarfing that N. American mammals underwent during the Paleocene/Ecocene Thermal Maximum (PETM) 55 MYa. Using an elegant theoretical tool called the Price equation, we can tease apart and quantify the causes of macroevolutionary change in any group of species at any site: species selection (non-random speciation and/or extinction with respect to species’ phenotypes), anagenesis (within-lineage evolution), and immigration of species to the site from elsewhere. In collaboration with my paleontologist colleague Jessica Theodor and her group, I’ve used this approach to discover that mammalian dwarfing during the PETM occurred despite species selection favoring larger mammals. We’re currently seeking a graduate student to apply the same approach to other macroevolutionary datasets.
Publications: Rankin et al. 2015 Proc Roy Soc B.
I have various other ongoing projects. Directly testing the role of drift (demographic stochasticity) in community dynamics via manipulations of community size. Using the Price equation to analyze the biodiversity-ecosystem function relationship. Others…