Scaling Data-Driven Movement Models to Spatial Population Dynamics

Overview

Understanding how individual movement behavior scales up to population-level patterns is a central challenge in ecology, epidemiology, and the study of spatially structured systems. The rapid growth of high-resolution tracking data has enabled the development of sophisticated stochastic-process models describing individual trajectories. However, translating these data-driven movement models into predictive frameworks for population dynamics remains a major theoretical and computational gap.

Despite progress at the individual level, a fundamental challenge remains: how do these data-driven movement models scale up to shape population-level outcomes such as spatial distributions, persistence, spread, and coexistence? Classical population models often assume mass action and simple diffusion or phenomenological dispersal terms, which may not accurately reflect the underlying behavioral processes observed in tracking data. Bridging this gap requires new mathematical tools to systematically derive and validate population-level models from empirically fitted individual-level processes.

This workshop will focus on advancing mathematical approaches for scaling up generally applicable individual-based movement models—particularly hierarchical continuous-times stochastic-process models—to population-level descriptions and inferences such as partial differential equations, network models, and density-based frame-works. By bringing together experts in applied mathematics, statistics, and ecology, the workshop aims to develop unified methodologies that connect empirical movement data with large-scale population processes. Addressing this scaling problem is essential not only for theoretical consistency, but also for improving the predictive power of models used in conservation planning, species management, and spatial population forecasting.