Having your cake and eating it: A modelling framework to combine process-based population dynamics and dispersal simulation

Abstract

This paper will introduce and demonstrate the capabilities of the newly extended 'Spatial DYMEX' model software. Spatial DYMEX allows modellers to develop process-based population dynamics simulations that include spatially-explicit dispersal. Traditionally, models of species dynamics have focused on either population demographics or dispersal processes. Where trade-offs in model detail between processes describing population dynamics and dispersal have frequently been required, Spatial DYMEX utilises a computationally-efficient cohort-based approach that allows the model to capture both temporal and spatial population variability. The computational efficiency results from the professionally developed underlying code structure and re-usable modules that allow the modeller to rapidly develop models without knowledge of computer code. Cohorts are a single entity that is used to represent a well-defined group of individuals who share the same or similar characteristics (e.g. of the same age and sex) who are studied over a period of time. They are commonly used in ecological modelling. DYMEX includes a graphical user interface that employs conceptual modules that are familiar to ecologists (e.g. lifecycles) and combines them with a library of customisable mathematical functions (e.g. logistic). Multiple lifecycles can be included in a model, and they can be linked to simulate a variety of ecological interactions (e.g. predation or competition). Deme structures can be used to include the effects of genetic diversity, for exploring topics such as pesticide resistance. Spatial DYMEX includes a range of graphical and spatial visualisation tools to analyse model outputs. This novel modelling software allows the modeller to capture simultaneously the dynamics of spread with a consideration of the ecological, environmental and behavioural factors that precede a migration or dispersal event and the consequences of dispersal. We demonstrate the capabilities of the enhanced Spatial DYMEX package using a case study of the population dynamics and dispersal of an aphid-virus-wheat system. The mechanistic, cohort-based approach offered by DYMEX allows us to simulate how various environmental drivers may impact upon processes that drive wheat growth and aphid (Rhopalosiphum padi) population dynamics, allowing us to estimate potential outcomes under future climate scenarios. A driving question for this research is whether Yellow Dwarf Virus, which is spread by aphids, will become a greater threat to wheat yields under climate change. The spatial modelling capabilities of the latest DYMEX now allow us to examine the spatial implications, where shifts in the timing of migration events at the start of the season may alter the timing of arrival and subsequent population dynamics of the aphids and virus in the wheat crop.