Research

Spatial synchrony and regional stability

Populations separated by hundreds of kilometers often exhibit synchronized fluctuations. Moran's theorem provides a classic explanation for this phenomenon: if populations share the same (log-)linear dynamics and experience similar environmental variability, the correlation in their fluctuations will match that of the environment. By providing a baseline expectation for the degree of synchrony, Moran's theorem can be used to measure meaningful deviations and potentially identify their origins.

My research focuses on what happens to synchrony when Moran's baseline assumptions break. For instance, dispersal can couple populations occupying different patches directly rather than through weather or climate. Additionally, density dependence in nature is often nonlinear rather than (log-)linear. Finally, synecological processes such as trophic interactions between prey and predator species can often induce synchrony. Each of these processes often leaves a unique signature in the distance-decay correlation that can be exploited in order to identify the causes of synchrony.

The is not a purely academic exercise because synchrony is what converts local variability into regional risk: when all populations collapse simultaneously, no rescue effect can occur because all sources of recolonization have effectively disappeared. Understanding the mechanisms that govern synchrony can thus inform spatial management decisions by predicting the likelihood of success for popular conservation efforts such as increasing habitat connectivity.

Metapopulation models · long-term monitoring data · wavelet and coherence methods

Climate variability and population persistence

Risk assessments usually focus on changes in mean temperature but organisms experience sequences of fluctuating temperatures whose clustering and variability can be more important than their mean for predicting organismal performance.

Work with collaborators in climate science and engineering has shown that temperature is becoming more autocorrelated in space and time, and that feeding projected thermal regimes into dynamical models of ectotherm performance produces regionally idiosyncratic effects on stability but a near-universal increase in extinction risk, with temperate species faring worse than the mean-temperature framing would suggest.

Earth system model projections · thermal performance curves · extinction risk

Biodiversity-ecosystem functioning relationship

Additive partitioning methods have been the workhorse of the biodiversity-ecosystem functioning field for over two decades. Split a mixture's yield into a selection effect and a complementarity effect, and you have a story about why diverse communities produce more. Unfortunately, these types of methods are used far more often than they are interrogated.

I collaborate with Pradeep Pillai on the conceptual and mathematical foundations of these methods, which all suffer from major philosophical and technical issues. The main philosophical issue lies in an inherent circularity, namely that species coexistence in mixtures is a consequence of the niche differences that the partition is supposed to detect, so the design encodes its conclusion into its premise. The key technical issue is nonlinearity, which violates an important yet unrecognized assumption of these methods. Specifically, when the relationship between density and ecosystem functioning is nonlinear in monocultures, additive partitioning schemes fail due to the emergence of a non-identifiability problem that causes the terms to no longer map cleanly onto the broad classes of mechanisms they are named after.

Coexistence theory · partitioning schemes · theory-experiment mismatch