Daniel Dalthorp
Manuela Huso
Andy Aderman
Lisa Madsen
2020
<div class="abstract-group"><div class="article-section__content en main"><p>We develop a novel method of estimating population size from imperfectly detected counts of individuals and a separate estimate of detection probability. Observed counts are separated into classes within which detection probability is assumed constant. Within a detection class, counts are modeled as a single binomial observation<span> </span><i>X</i><span> </span>with success probability<span> </span><i>p</i><span> </span>where the goal is to estimate index<span> </span><i>N</i>. We use a Horvitz–Thompson‐like estimator for<span> </span><i>N</i><span> </span>and account for uncertainty in both sample data and estimated success probability via a parametric bootstrap. Unlike capture–recapture methods, our model does not require repeated sampling of the population. Our method is able to achieve good results, even with small<span> </span><i>X</i>. We show in a factorial simulation study that the median of the bootstrapped sample has small bias relative to<span> </span><i>N</i><span> </span>and that coverage probabilities of confidence intervals for<span> </span><i>N</i><span> </span>are near nominal under a wide array of scenarios. Our methodology begins to break down when<span> </span><i>P</i>(<i>X</i>=0)>0.1 but is still capable of obtaining reasonable confidence coverage. We illustrate the proposed technique by estimating (1) the size of a moose population in Alaska and (2) the number of bat fatalities at a wind power facility, both from samples with imperfect detection probabilities, estimated independently.</p></div></div>
application/pdf
10.1002/env.2603
en
Wiley
Estimating population size with imperfect detection using a parametric bootstrap
article