29.07.2013 Views

2007 Summaries of Wildlife Research Findings - Minnesota State ...

2007 Summaries of Wildlife Research Findings - Minnesota State ...

2007 Summaries of Wildlife Research Findings - Minnesota State ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

VARIANCE OF STRATIFIED SURVEY ESTIMATORS WITH PROBABILITY OF<br />

DETECTION ADJUSTMENTS 1<br />

John Fieberg and John H. Giudice<br />

ABSTRACT<br />

511<br />

Estimates <strong>of</strong> wildlife population sizes are frequently constructed by combining counts <strong>of</strong><br />

observed animals from a stratified survey <strong>of</strong> aerial sampling units with an estimated probability<br />

<strong>of</strong> detecting animals. Unlike traditional stratified survey designs, stratum-specific estimates <strong>of</strong><br />

population size will be correlated if a common detection model is used to adjust counts for<br />

undetected animals in all strata. We illustrate this concept in the context <strong>of</strong> aerial surveys,<br />

considering 2 cases: 1) a single-detection parameter is estimated under the assumption <strong>of</strong><br />

constant detection probabilities; and 2) a logistic regression model is used to estimate<br />

heterogeneous detection probabilities. Naïve estimates <strong>of</strong> variance formed by summing<br />

stratum-specific estimates <strong>of</strong> variance may result in significant bias, particularly if there are a<br />

large number <strong>of</strong> strata, if detection probabilities are small, or if estimates <strong>of</strong> detection<br />

probabilities are imprecise.<br />

________________<br />

1 Journal <strong>of</strong> <strong>Wildlife</strong> Management (2008) 72(3):837-844.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!