I am involved in a variety of research projects, most of which focus on the development and application of statistical methods for temporal and spatial data. I enjoy collaborating with many different people, but particularly students and post-docs because I tend to learn a lot from them.

Multivariate state-space models

Multivariate Autoregressive State-Space (MARSS) models. Eli Holmes, Eric Ward, and I have developed an R package with a detailed user’s guide for analyzing a variety of multivariate time series problems under a maximum likelihood framework. Examples include movement tracking, dynamic linear models (DLM), dynamic factor analysis (DFA), and estimating community interactions & stability.

Time-Varying Vector Autogressive State-Space (TVVARSS) models. Eric Ward, Steve Katz, and I have developed an R package to fit a Bayesian version of the model developed by Ives & Dakos (2012). We expanded on their model to include multiple different forms of observation models. We are using it to examine changing species interactions and community stability in kelp forest ecosystems.

Integrated populations models

Skagit River steelhead. I developed a Bayesian IPM for a population of steelhead from the Skagit River to help biologists from the Skagit River System Cooperative and the Washington Department of Fish & Wildlife examine the effects of hatchery releases and instream flows on steelhead production. You can find a detailed description of the methods here.

Snake River Chinook. Eric Buhle, Jim Thorson, and I have developed an R package to fit a Bayesian IPM to multiple populations of Pacific salmon. Our model also includes broodstock removals for hatchery operations and the effects of hatchery-origin fish spawning in the wild. We are using the model to conduct population viability assessments and evaluate possible effects of various harvest control rules.

Joint dynamic species density models

I have been working with Jim Thorson and several others on a variety of applications of joint dynamic species density models (JDSDM) to marine and terrestrial fauna. You can learn more about the models, applications, and R packages at the FishStats website.