Developing Models for the Spatial Distribution of Halibut in the Gulf of Alaska

• Activity Number: 428
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Daniel C. Fortin*+ and Petrutza Caragea
• Companies: Iowa State University and Iowa State University

Abstract

Our work concerns the development of models for the spatial distribution of Halibut in the Gulf of Alaska. To date, it appears that the only methodologies of spatial statistics applied to such problems are those that fall under the umbrella of geostatistical methods. An alternative approach is presented by viewing the responses as forming a discrete-index random field, which is defined to exist at only a finite number of specific locations in a bounded geographic region. This second approach is adopted and the responses are modeled as a Markov random field. We discuss conceptual differences between geostatistical methods to a Markov random field methods, and compare the predictive ability between two models for the Halibut data.