Stat Cafe - Dr. Toryn Schafer
Scalable Bayesian Spatial Modeling and Agent-Based Simulation for Ecological Decision Support
- Time: Wednesday, 4/8/2026 from 11:30AM to 1:00PM
- Location: BLOC 503
Description
Ecological management often requires linking uncertain data to scenario-based predictions of how human activity may affect wildlife. This is difficult when observations are sparse, collected across multiple observation windows, and decision-relevant quantities are not direct outputs of standard statistical models. I will present a framework that connects Bayesian spatial inference with simulation. We use a computationally efficient recursive Bayesian approach to fit spatial point process models to point pattern data from long-term monitoring and aerial imagery with non-overlapping sampling frames. Posterior predictive spatial fields initialize a modular agent-based model that represents environment, human activity, and animal responses, summarized through disturbance metrics.
Our Speaker
Toryn Schafer is an Assistant Professor in the Department of Statistics at Texas A&M University. She was previously a postdoctoral associate in the Department of Statistics and Data Science at Cornell University and earned her PhD in Statistics from the University of Missouri in 2020. Her research interests include Bayesian spatio-temporal modeling, statistical computing, and applied work in ecological and environmental sciences.