Abstract:
|
In this talk we develop models for closed-population wildlife surveys that incorporate time-to-detection data. We show that many of the most widely-used wildlife survey models, including removal method models, distance sampling models, capture-recapture models and some occupancy models, can be viewed as survival or recurrent event models. We investigate what is gained by using time-to-detection data in cases in which it is traditionally neglected. In the case of distance sampling, something close to time-to-detection data was used until the early 1980s; we look at the reasons its use was discontinued, and suggest that it is time to reconsider its use. In the case of capture-recapture surveys, the advent of continuously-sampling detectors like camera traps and spatially explicit capture-recapture methods motivates the use of time-to-detection data, while time-to-detection data are also shown to be useful in occupancy modelling.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.