|
Activity Number:
|
319
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #304988 |
|
Title:
|
Bayesian Analysis of Animal Community Structure
|
|
Author(s):
|
Jeffrey A. Royle*+
|
|
Companies:
|
U.S. Geological Survey/Patuxent Wildlife Research Center
|
|
Address:
|
12100 Beech Forest Road, Laurel, MD, 20708,
|
|
Keywords:
|
species richness ; avian surveys ; biodiversity ; capture-recapture ; animal sampling ; hierarchical models of occurrence and abundance
|
|
Abstract:
|
Models of animal community structure are based on surveys of species' presence or absence on a sample of spatial units. One consideration in conducting inference about community structure is that species are detected imperfectly. This leads to fewer species observed in the sample than exist in the community. Classical methods for modeling community structure do not preserve species identity, and thus don't allow the development of predictive models of community composition. We describe a strategy based on species-specific models of occurrence, from which estimates of important summaries of community structure are derived by aggregating indicators of occurrence for all species observed or estimated to be in the community. We use a data augmentation approach to develop an efficient Bayesian procedure for estimation and prediction under this model using MCMC.
|
