Mechanistic population modelling

Stage-structured pop. models for two riparian plant species (from Lytle et al., 2017)

Modelling approaches based on past correlations do poorly looking into the future for how species may respond to unprecedented changes. By explicitly modelling the biology underpinning the distribution of species rather than correlating the presence of species with particular conditions, mechanistic models offer a greater capacity to forecast species responses to novel combinations of environmental conditions. We develop mechanistic (or process-based) population models that link population dynamics with environmental variability, including matrix population models and time-varying logistic growth models. We also develop approaches to link population models together in a community-wide framework that enables an understanding of community dynamics under various future conditions. These are typically focused on river ecosystems by linking river flow regimes with invertebrates, fish and riparian plants, but we are expanding into new ecosystems, including the Antarctic.

Avatar
Jonathan Tonkin
Senior Lecturer & Rutherford Discovery Fellow

Publications

Increasing drought favors nonnative fishes in a dryland river: Evidence from a multispecies demographic model

Understanding how novel biological assemblages are structured in relation to dynamic environmental regimes remains a central challenge …

High mortality and enhanced recovery: Modelling the countervailing effects of disturbance on population dynamics

Disturbances cause high mortality in populations while simultaneously enhancing population growth by improving habitats. These …

Linking river flow regimes to riparian plant guilds: A community-wide modeling approach

Modeling riparian plant dynamics along rivers is complicated by the fact that plants have different edaphic and hydrologic requirements …