This research aims to validate a spatially explicit mechanistic modeling framework that integrates environmental changes with disease ecology and genomics. Using North American bats and the fungal pathogen causing white-nose syndrome as an example, the model can be generalized to similar applications. A flexible, compartmental Susceptible-Infected-Recovered-type module will be added to an established spatially explicit, individual-based, eco-evolutionary model of metapopulation dynamics extensively used in landscape genetics, developed by Prof. Erin Landguth and her research group at the University of Montana (UM). Bayesian statistical sampling techniques will be incorporated to validate the model against real data, extending the work in my Ph.D. thesis. The plan is to visit Erin and start a long-term collaboration with her and her group. Erin, who was my thesis opponent, is a well-known researcher in the field of wildlife ecology. My visit to the UM research team would ideally combine the expertise of LUT and UM: the wildlife know-how and simulation tools of the UM group, and the computational statistics methods of our side, from the Centre of Excellence in Inverse Problems. At LUT we have recently developed a new approach for Bayesian inference for ‘intractable’ or stochastic dynamical systems, based on empirical cumulative distribution functions (eCDF) to construct Gaussian likelihoods. The method has been successfully used in several applications. However, the application to the ABM framework is novel, and requires new algorithmic developments. Additionally, the generalizable framework for modeling eco-evolutionary processes and disease propagation can be applied to address other epidemiological challenges, such as the rising white-tailed deer population and the associated increase in tick-borne health risks in Finland, as well as similar global issues.

2025

2026