Activity Number:
|
24
- Assisting Natural Resource Agencies with Improved Inferences on Ecological Processes
|
Type:
|
Topic Contributed
|
Date/Time:
|
Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Statistics and the Environment
|
Abstract #327250
|
|
Title:
|
A Multiseason Site Occupancy Model for Use When Sites Are Not Revisited Among Seasons
|
Author(s):
|
Brian R Gray* and Darryl I MacKenzie and Richard A Erickson
|
Companies:
|
US Geological Survey and Proteus Wildlife Research Consultants and US Geological Survey
|
Keywords:
|
site occupancy;
mixture models;
classification errors
|
Abstract:
|
Models of Bernoulli processes with classification errors have been developed for sampling protocols that do and do not specify sampling across multiple seasons. These models are typically known by ecologists as multiseason (or dynamic) and single season models, respectively. Multiseason site occupancy models have not to our knowledge been developed for use with detection/nondetection data when sampling units (sites) are not revisited across seasons. We explore the estimation of site occupancy and detection parameters under such designs using what is effectively a concatenation of the single season site occupancy model. Concerns with this elaborated model include the increased complexity associated with addressing latent season effects on occupancy and detection means, latent site effects on detection, and temporal dependence among latent season effects on occupancy.
|
Authors who are presenting talks have a * after their name.