Adjusting for Imperfect Detection in Population Size Prediction by Matthew Higham, Oregon State University, October 31, Statistics Colloquium

Adjusting for Imperfect Detection in Population Size Prediction by Matthew Higham, Oregon State University, October 31, Statistics Colloquium, 1 – 1:45 pm, Stetson Court Classroom 101

Abstract:  A finite population prediction method that assumes all animals are detected during sampling is routinely used to predict moose abundance in Alaska and Canada. Because moose are less readily detected on ground without snow, declining snowfall in recent years makes the assumption of perfect detection less reasonable. In response, government agency biologists have started to collect additional data on detection. We consider a Bayesian model to incorporate the possibility of imperfect detection. Unlike a frequentist approach, the Bayesian model does not require a large count assumption. However, fitting the model takes much longer and also demands that the user be familiar with Bayesian statistics and Markov Chain Monte Carlo (MCMC) methods. The predictor is applied to a moose survey in Togiak National Wildlife Refuge in Alaska to predict the moose population total in the region.