Speaker: Dr. Pavel Chernyavskiy, Department of Mathematics and Statistics, University of Wyoming

Title: Covariate-driven non-stationary spatial models with applications to water quality in North American lakes

Abstract:
Freshwater ecosystems are vulnerable to anthropogenic changes that occur at disparate spatial scales, leading to nutrient pollution and poor water quality. Recent non-stationary spatial models that admit covariates into the mean, variance, and covariance functions can improve our understanding of the mechanisms by which environmental stressors impact freshwater resources. This project has two main goals: 1) develop and implement novel covariate-driven non-stationary models via No-U-Turn Hamiltonian Monte Carlo (NUTS HMC) in Stan; and 2) investigate environmental correlates of non-stationary spatial dependence for water quality in North America. NUTS HMC was chosen over Gibbs and Adaptive Metropolis-Hastings because it does not require conjugate priors and it produces considerably more effective samples/iteration. I will discuss issues related to: 1) model specification and implementation, 2) covariate selection, and 3) results for phosphorus pollution in Iowa, Illinois, and Wisconsin freshwater lakes (N=315) collected from the LAGOS-NE database.