`
On the one hand, spatially or temporally variable environmental conditions can create new niches. By enabling species to specialize in conditions at particular places or times, environmental variation can help individuals reduce their exposure to between-species competition, sometimes allowing species to persist which would otherwise have been competitively excluded. Exploiting environmental variation in this way requires species to evolve traits that heighten their sensitivity to environmental variation. On the other hand, being sensitive to environmental variation can cause growth rates to vary in time. For organisms that reproduce in synchronized pulses, such as plants, year to year variation leads to lower long-run growth/fitness. The risks of environmental variation should lead species to evolve traits that will buffer them against variation.
Past work focused on ways that environmental variation could promote species coexistence. Current work leans toward evolutionary responses to environmental variation. For example, "Leaving home ain't easy: non-local seed dispersal is only evolutionarily stable in highly unpredictable environments" (Snyder, Proc. R. Soc. B, 2011, 278(1706), 739--744) explores the ability of environmental variation to drive the evolution of seed dispersal. Unless the environment is highly unpredictable, evolutionarily stable dispersal distances leave seeds close to their mother, in places whose environmental conditions are correlated with that of their natal location instead of in novel environments. This is an example of species increasing their sensitivity to environmental variation. "Coexistence and coevolution in fluctuating environments: Can the storage effect evolve?" (R. Snyder and P. Adler, in review) considers whether two competitors can co-evolve variable germination rates and thereby benefit from a storage effect. We find that it is very difficult to coevolve variable germination if germination is not predictive (more seeds germinate in ``good'' years), though it is possible to achieve if one species has a fixed strategy, perhaps due to reduced genetic variation. Here species evolve buffering strategies unless there is some other benefit (predictive germination) to increasing their sensitivity to year-to-year variation.
Many biological processes can be thought of as flows on networks. In ecology alone, networks have been used to represent foodwebs, dispersal networks, pollinator relationships, and insect social networks, while networks also play important conceptual roles in epidemiology, genomics, and proteomics. Networks allow us to recognize that some genes/protein residues/individuals/populations/habitat patches are more highly interconnected than others, but this increase in realism comes with an increase in model complexity. It is rarely possible to develop an intuitive understanding of the network dynamics. Recognizing the key role that network structure plays in network dynamics, many have tried to simplify networks by identifying non-random structures. However, this emphasis on topological features has often come at the cost of attention to dynamics. Much effort has been spent on trying to establish relationships between topological features and stability or robustness to species removal but relatively little attention has been paid to other aspects of dynamics.
Peter Thomas, Patrick Wintrode, and I have recently been awarded an Advancing Theory in Biology grant (NSF) to develop new approaches to studying biological networks. Right now, our work is focusing on finding ways to compress/coarse-grain networks in ways that preserve quantities of interest and on information theoretic approaches to network dynamics. For example, some proteins have binding sites that affect remote active sites. The network of amino acids making up the protein then becomes a communication channel between the binding site and the active site, and we can consider the channel capacity of this channel.