Spatiotemporal variation in natural selection

Forthcoming in Evolution

Weese, D.J., S.P. Gordon, A.P. Hendry, and M.T. Kinnison. Spatiotemporal variation in linear natural selection on body color in wild guppies (Poecilia reticulata)

Despite a number of good examples of the ecological consequences of ongoing evolutionary change (i.e., contemporary evolution), establishing the generality of these eco-evolutionary interactions remains an important challenge. The relevance of such effects in any given situation will depend, at least partly, on the temporal dynamism of the evolutionary mechanisms (selection, gene flow, mutation) that are functionally linked to ecological processes. Testing for contemporary evolution has been facilitated by the recent development of analytical methods that estimate the strength and pattern of natural selection in wild populations using multiple regressions to predict individual fitness from trait values (the slopes of these regressions are called selection coefficients). By replicating these measurements in time, we can reveal the temporal scale at which eco-evolutionary effects are likely to be important.

Along with my co-authors, I used a series of mark-recapture experiments to investigate variation in natural selection on body color in Trinidadian guppies. Among-population variability in male colour is predicted to reflect a balance between the benefits of colour (attracting females) and the costs of colour (attracting predators). consistent with this idea, our natural selection coefficients were negative, colourful guppies generally had lower survival than did drab guppies. However, we also found that selection in this system is highly variable both spatially (among populations) and temporally (within the same population). Surprisingly, this variation was not consistently associated with the presence of dangerous guppy predators. These findings suggest complicated and dynamic interactions among sexual selection, natural selection, predation, and phenotypic evolution in this system. These types of interactions operating on very fine spatiotemporal scales suggest that eco-evolutionary dynamics may be commonplace, and relevant to ecological processes occurring in contemporary time.

Phenotypic plasticity and adaptation

I am interested in the role of phenotypic plasticity in adaptation to different environments. Divergent selection can result in adaptive genetic divergence among populations. However, individuals can also adapt through a plastic response; i.e. the environment might have a direct impact on the phenotype without influencing genetic change. The relative contribution of each (genetic vs. plastic adaptation) should be related to the level of dispersal and gene flow among selective environments. Gene flow can constrain adaptive genetic divergence, and therefore plasticity might be favoured under high gene flow scenarios. If heritable variation for plasticity occurs in a meta-population, I predict that plasticity might evolve in response to gene flow. At the same time, plasticity might permit increased dispersal and gene flow among environments.

During my doctoral studies at McGill, I developed a conceptual framework for understanding the above pathways. I then tested some of the framework's predictions in an empirical system: an African cichlid fish from high-oxygen rivers and adjacent low-oxygen swamps. I found that morphological plasticity, in response to dissolved oxygen concentration, was high overall in this species, but that plasticity was higher at locations where dispersal between environments should also be higher. I infer that plasticity might have evolved in response to gene flow between oxygen environments, but more studies are needed to determine causality.

The conceptual framework is published in the following article. The empirical studies are currently in review.
Crispo, E. (2008) Modifying effects of phenotypic plasticity on interactions among natural selection, adaptation and gene flow. Journal of Evolutionary Biology 21:1460-1469

Latest news: I successfully defended my PhD dissertation yesterday!

What are eco-evolutionary dynamics?

Evolution is obviously driven by ecological differences: think of the adaptive radiation of Darwin’s finches. Just as obviously, ecological processes are influenced by evolution: ecosystems depend on the oxygen produced following the evolution of photosynthesis. Less obvious is how these interactions play out on the short time scales most relevant to conservation and management. Do ecological changes (e.g., invasive species, climate change) drive appreciably evolutionary change over years or decades (i.e., “contemporary evolution”)? Does any such evolution influence ecological variables (population dynamics, community composition, ecosystem function) on similar time scale? These potential interactions between ecology and evolution, as shown in the figure, represent the growing field of eco-evolutionary dynamics.

Many studies have shown that ecological changes cause phenotypic changes in natural populations (eco-to-evo). Examples include species introduced to new environments, native species responding to introduced species, populations exposed to harvesting or pollution, and populations facing climate change. Existing work has shown that the phenotypic changes can be substantial, particularly when humans are involved. What isn’t known generally is just how much of this phenotypic change is the result of evolutionary change versus phenotypic plasticity. Even less is known about how these contemporary phenotypic changes then influence ecological variables on similar time frames (evo-to-eco) – but some nice examples can be provided.

Populations: Phenotypic changes from one year to the next clearly influence population size in ungulates. The genetic contribution to this phenotypic change is not known, whereas a study of butterflies has documented effects of genetic change on population sizes. What remains to be determined is just how common these effects are, and how important they are relative to traditional “ecological” effects (e.g., rainfall or temperature). In addition, it isn’t clear under which conditions these population dynamical effects of evolution can actually save natural populations from extinction (i.e., evolutionary rescue).

Communities: Genetic and phenotypic differences between individual plants have been shown to have noteworthy effects on arthropod communities. Similarly, genetically-based phenotypic differences between fish populations have strong influences on aquatic macro-invertebrate communities. What remains to be determined is, again, how common these effects are and, also, how year-to-year changes in these genes and traits (as opposed to the currently-studied static differences) influence those communities.

Ecosystems: In the same plants and fish studied for community effects (above), genetic and phenotypic differences have been shown to influence ecosystem variables such as decomposition rates, dissolved organic material, light attenuation, and primary productivity. Since the study systems are the same as above, what remains to be discovered is also the same. It will also be interesting to know how often these ecosystem effects of evolution fall into the category of “ecosystem services” that have become so integral to conservation efforts.

The above listing highlights a few specific examples of how evolutionary change might influence ecological variables on short time scales. In addition to the specific uncertainties listed above, some additional general ones come to mind. How often do evolutionary effects on communities and ecosystems flow through the effects of evolution on population dynamics (indirect effects – red to black arrows in the figure) versus changes in the traits themselves (direct effects – red arrows only)? Do the effects of evolutionary change on ecological processes decrease from population to community to ecosystem variables? How often do true feedbacks occur – that is an ecological process drives evolution (green arrows) that then alters that same ecological process (red arrows) and so on? Eco-evolutionary dynamics is an area ripe for future work. A special journal issue on eco-evolutionary dynamics: http://rstb.royalsocietypublishing.org/content/364/1523

Thesis submission

On February first Xavier submitted his thesis "The driving factors of ecological speciation and their interactions" in which he looked at the effects of gene flow on local adaptation, and reciprocally, the effects of local adaptation on gene flow.

Evolution revealed by resurrecting historical resting eggs

Recently published in the January issue of Evolutionary Ecology!

Derry, A.M., S.E. Arnott, and P.T. Boag. 2010. Evolutionary shifts in copepod acid tolerance in an acid-recovering lake indicated by resurrected resting eggs. Evolutionary Ecology 24: 133-145.
http://www.springerlink.com/content/668651t1247q3v18/fulltext.pdf

Long-lived resting eggs of zooplankton are an ecological and evolutionary reservoir that can impact population and community responses to environmental change in lake ecosystems.
I investigated if adaptive shifts in copepod acid tolerance had arisen over a century of environmental change in an acid-recovering lake by resurrecting resting eggs dating from the late 1800s to present. My results suggest that copepods underwent shifts in acid tolerance following both anthropogenic acidification and pH recovery. Further, I found evidence to suggest that maternal effects (the effects of female body condition on the fitness of their offspring) can have an important role for the fitness of older zooplankton genotypes, irrespective of egg age, when historic resting eggs are given an opportunity to hatch into contemporary communities. Adaptive responses through time are important to consider because of their potential to influence community-level interactions in ecosystems recovering from anthropogenic disturbance.

Ben and I are planning to do some individual-based modelling to learn further about the effects of stochastic hatching from historical resting egg banks on contemporary evolution to environmental change.

How does evolution work?

I'd like to kick this blog off with a very broad question: how does evolution work?  The photo above was taken at Antelope Canyon, in Arizona.  The sandstone walls of the canyon have been carved by erosion over many millenia: uncountable grains of sand, myriad molecules of water, all rubbing against the rock on their journey downstream, gradually shaping the canyon into what it is today.  And what the canyon looks like depends very much on when you visit it; the floor of the canyon, which is just sand left behind by the last flash flood, rises and falls by many feet over short periods of time, or so I am told.  Formations that are hidden by sand one month may be many feet above your head the next month.  Regardless of when you visit it, however, the canyon is stunningly beautiful.

Antelope Canyon is an example of the marriage of the deterministic and the contingent.  On the one hand, the shapes carved in the canyon walls are not predetermined by any mathematical formula.  No computer could predict the particular form that the canyon has today.  A soft spot here and a hard spot there, an eddy here and a fast-flowing current there, have made it the way it is now; but slightly different conditions might have led to completely different contours.  In any case the canyon is not a static entity; if you could visit it a few thousand years from now, it might look quite different.  On the other hand, its shape is not completely random; there are many shapes that will never be formed by its particular dynamics.  Wait as long as you like, but you will never find a tiny replica of the Empire State building cut into the rock.  Sandstone and erosion have their own particular laws, which they always obey.

The question I am interested in, and the one that attracted me to work in the Hendry lab, is: how are the products of evolution — fish and frogs, birds and beavers, you and I — like or unlike Antelope Canyon?  To what extent are the products of evolution predictable and deterministic, a repeatable outcome given environmental conditions?  To what extent are they unique, the irreproducible results of a grand experiment that can never be replicated?  What are the laws that evolution obeys?  What sorts of products will it tend to create over and over, and what sorts of products, like the Empire State Building, are forever beyond its reach?  How quickly can its products change — how long does it take for a new species to form, or for a new trait to evolve?  And — a question that particularly interests me, but that seems completely unapproachable — why are the products of systems at the interface between determinism and contingency, like living organisms or Antelope Canyon, so incredibly beautiful?

Really answering such questions is far beyond the reach of a PhD thesis, or even an entire career; but they are the kinds of questions that inspire me.  All this could be framed in much more academic, jargon-laden ways: gene flow and selection gradients and additive genetic variance.  But it's important, I think, to keep in mind the ultimate reasons we have embarked on this journey: our curiosity about nature, and our joy upon beholding its beauty.  Let's keep the destination in mind, but enjoy the journey, too.