You are here:
Estimating occupancy rates with imperfect detection under complex survey designs
Citation:
Olsen, Tony. Estimating occupancy rates with imperfect detection under complex survey designs. International Indian Statistical Association Conference on Statistics, Corvallis, OR, August 18 - 21, 2016.
Impact/Purpose:
Monitoring to determine the presence of threatened or endangered animal species is a common task that federal agencies and other organizations must complete. Typically, the monitoring design is a complex design that involves stratification and unequal probability of selection. When conducting field visits to the selected sites, a common problem is that during a single visit to a site, it is possible not to detect the species even when it is present, that is, the probability of detection is less than one. The objective of the survey is to estimate the proportion, or area, of the study region where the species is present. Estimation of site occupancy rates when detection probabilities are less than one requires that the analysis incorporates information on the survey design. This presentation generalizes existing methods to do this and gives an example based on monitoring of two amphibian species by the US Forest Service. This is an extra product for SSWR 3.01A.
Description:
Monitoring the occurrence of specific amphibian species is of interest. Typically, the monitoring design is a complex design that involves stratification and unequal probability of selection. When conducting field visits to selected sites, a common problem is that during a single visit to a site, it is possible not to detect an amphibian species even when it is present, that is, the probability of detection is less than one. The objective of the survey is to estimate the proportion, or area, of the study region where the species is present. Estimation of site occupancy rates when detection probabilities are less than one have been developed by MacKenzie et al. (2006) under the assumption of a simple random sample. In this paper, using the notion of generalized estimating functions, their procedures are generalized to cover more general complex survey designs.MacKenzie, D. I., J. D. Nichols, J. A. Royle, K. H. Pollock, L. L. Bailey and J. E. Hines (2006). Occupancy estimation and modeling: inferring patterns and dynamics of species occurrence. San Diego, California, Academic Press.