Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 335 - Recent Advances in Spatial and Spatio-Temporal Modeling
Type: Contributed
Date/Time: Tuesday, August 9, 2022 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics and the Environment
Abstract #322141
Title: Multivariate Spatio-Temporal Models for Landscape Change Using Aerial Imagery
Author(s): Xinyi Lu*
Companies: Colorado State University
Keywords: Bayesian; spatio-temporal; data augmentation; shrubification; climate change
Abstract:

The Alaskan landscape has undergone substantial changes in recent decades, most notably the expansion of shrubs and trees across the Arctic. We developed a dynamic statistical model to quantify the impact of climate change on the structural transformation of ecosystems using remotely sensed imagery. Our model accommodates changes in temperature and precipitation to infer and predict rates of land cover transitions while accounting for spatio-temporal heterogeneity. Transition types are highly correlated at both plot and subplot levels in our study system, therefore we characterized multiscale spatial correlation using Gaussian processes. Imagery pairs were collected at irregular time intervals, therefore we modeled dynamic state probabilities that evolve annually using a hierarchical framework. We developed a PĆ³lya-Gamma representation of our model to improve computation. Our model facilitates inference on the response of ecosystem state probabilities to shifts in climate and can be used to predict future land cover transitions under various climate scenarios.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program