Inference of ecological and environmental models

Ecological and environmental models are continually evolving. Within fisheries research, individual movement, environmental drivers, and interspecific interactions are key areas of interest stimulating the development of new and complex modelling efforts. Individual-based models (IBMs) are one example of such model in which individual animals interact with one another and the landscape in which they live, with population metrics emerging from the actions of collective individuals. When used in spatially-explicit landscapes IBMs can show how populations are expected to change over time in response to management actions, and have therefore been shown to be effective management tools in many systems. For instance, IBMs are being used to design strategies for the conservation and exploitation of fisheries, and for assessing the effects on populations of major construction projects and novel agricultural chemicals. However, good understanding of fits to data and associated uncertainty are needed before such models can be used to support decision making.

Hence, there is urgent need to improve methods of calibrating complex multiparameter models: existing methods are too slow, and not always accurate. This project aims to improve the best existing method: Approximate Bayesian Computation, ABC. ABC is currently being used for statistical inference in a diverse range of applications in ecology, evolution and more widely. In most cases the challenges of parameter estimation and model comparison are both of importance, but implementation can prove computationally expensive. This project aims to improve ABC methods and apply them in collaboration with environmental researchers, to help them in fitting models to data. Initial focus will be on IBMs developed for fisheries management by Cefas, the UK governments marine and freshwater science experts, https://www.cefas.co.uk/