Research

The classical answer to the question of how competing species coexist in nature is that species coexist "stably" (i.e. each can invade an equilibrium population of the other from low abundance) by differing from each other in some way that lessens lessens their competitive influence on each other compared to their influence on themselves . The classical "niche theory" arising from these ideas also suggests that, among a set of coexisting competing species, not only must there be differences between species, but these differences must be large enough if stable coexistence is to occur--i.e. there are "limits to similarity”.

However, there is an important additional consideration that is essential to the concept of niche differences--robustness. Most competition models can actually produce stable coexistence of arbitrarily similar species if parameters are tuned. When we consider that environments may vary, or that parameters may not be tuned, we again see the need for species differences.

In past work, the Ostling lab further developed tools for calculating robustness or conversely “sensitivity” of coexistence, to generalize this idea and show the importance of differences in interaction with regulating factors for stable and robust coexistence.

We have also shown that "tight-packing" solutions to competition models, where arbitrarily similar species coexist, are not robust so long as the models are formulated so they are absent of unrealistic discontinuities.

Much of the theory developed in the Ostling lab is broadly applicable across communities of many different types of organisms. However, broad theory can only take us so far in determining what the dynamics of coexistence in nature really are. Fully convincing evidence of niche differences requires developing and testing system-specific niche theory.

Plant species coexistence has always been a fascinating puzzle in ecology because coexisting plant species seem to all require the same limiting resources—light, water, nutrients. Tree species coexistence is particular important given the role of forests in the global carbon cycle. Forests are also a data-rich system due to the Forest GEO global network of forest census sites all using the same sampling protocol. 

Particularly dominant hypotheses of coexistence of tropical tree species are: 1) successional niche differentiation, i.e. variation across species in strategy ranging from gap-specialization to shade-tolerance; and 2) differences in enemies and the tendency for more abundant species to be easier for enemies to find, i.e. the Janzen-Connell mechanism. Though there is evidence pointing to each, it is far from clear how big of a role they play. How much does each increase diversity compared with a neutral model? Or could these mechanisms actually lessen expected richness, as work from my lab on stochastic niche models mentioned above suggests can happen? To answer these questions, we need a deeper understanding of these mechanisms, and better tools to quantify their role using forest census data.

The Ostling lab is currently studying partial-differential equation models capturing the size-structured and patch-age structured nature of competition among tree species, to gain a more fundamental understanding of successional niche differentiation. We are also studying Janzen-Connell mechanisms and the patterns they produce, and collaborating with Brian Sedio's lab at UT Austin to use new chemical defense data they are amassing for the Forest GEO plots to quantify the role that differences in enemies and other mechamisms may play in the latitudinal diversity gradient in forests. That work is funded by a current NSF grant from the Division of Environmental Biology.