Abstract:
|
Animal abundance and density are often of particular interest in ecology. Due to recent advancements in both technology and statistical methodology, wildlife surveys deploying remote detectors are becoming increasingly common. Examples of such detectors include microphones and cameras, which detect animals without making any direct contact with individuals themselves. One difficulty arising from the analysis of the resulting data is that animal locations are not known with certainty. A family of methods known as spatially explicit capture-recapture (SECR) overcomes this by describing probabilistically where animals are likely to be. Further, some detectors may collect auxiliary data informative on these locations, such as estimated distances and bearings to animals, or the strengths and arrival times of acoustic signals. These data allow for animal location estimates that are more precise, and this increased precision propagates through to the overall abundance and density estimates. In this talk I present the new R package admbsecr. This is the only software available capable of fitting SECR models incorporating such data, overcoming the computational obstacles these provide.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.