Abstract: The high diversity and variability of the natural world drives questions in ecology, but also makes them challenging to answer. When everything is changing, how can we make generalizations and draw conclusions? Statistics and probability, integrated with dynamical models, have provided ways forward over the last century, leading to theory and empirical tools ever more in use today. But variation was traditionally assumed to be stationary: means and variances were fixed and did not change with time. The stationary assumption for climate meant that it could be described by a single frequency distribution for any location, with parameters that did not change with time, yet allowed temporal environmental fluctuations to be assessed for their roles in population fluctuations, in community outcomes, such as species coexistence, and life-history evolution. Long-term climate change provides a new challenge. Means and variances of environmental fluctuations change with time, which means that environmental fluctuations are nonstationary. What then applies consistently within populations and communities, and how can we make generalizations and predictions, and draw conclusions? In recent years, a number of solutions have been emerging. Here I focus on a solution based on tracking populations and communities over complex landscapes, as the climate changes. I define the experienced environment as the environment measured according to where a population is concentrated. Most important, I show in models that experienced environments can emerge as stationary or nearly stationary even though the climate at any given location changes. If we can predict this experienced environment, we can use standard methods to model populations and communities and draw conclusions. I show how in certain models, the nature of the experienced environment, with and without climate change, can be characterized. I introduce a new class of models, called finact models, which are highly tractable for simulation, but in simple cases allow analytical solutions. I use the results of these models to generalize quantitative community theory (aka “modern coexistence theory”) to the case of long-term climate change. Within this generalization, I show how it is possible to assess climate change effects on average fitness differences and stabilizing mechanisms, and to determine when specific stabilizing mechanisms such as the storage effect, and when specific equalizing mechanisms such as fitness tradeoffs, are strengthened or weakened under climate change. In sum, I show how a powerful generalization of quantitative community theory, accounting for climate change, is emerging.
.jpg "Peter Chesson photo")