CW-CESU: Analysis of Great Lakes Network Songbird data- GLKN

archived

The GLKN seeks a statistician with previous experience analyzing breeding songbird monitoring data to conduct trend analyses for species and guilds in each of the nine parks, power analyses of the overall project methods, and to use the data to identify priority species for each park. Data include distance estimates, and analyses must incorporate abundance estimates corrected for detection probabilities. We also seek to place parks in regional contexts by comparing our data with that collected by the North American Breeding Bird Survey and other datasets to be identified. The statistician will also assist the GLKN songbird biologist in summarizing the data for a technical report submitted to the National Park Service’s Natural Resource Report series.  Scientific community and researchers external to the NPS gain by new information information gleaned from monitoring in the national parks. This report and analysis will provide new techniques for analyzing bird abundance and trends, which can be used by other monitoring programs, both inside and outside of the NPS.  Using data collected from nine parks from approximately 2015 (earlier for some parks) through 2018, do the following: 1) Assess the power of the data to detect trends ("power analysis"), given an effect size, alpha level, current monitoring design, and recent abundance and occupancy data. We seek to answer this question both for key species of interest and for guilds --- birds grouped by breeding habitat, diet, and migratory status. 2) Test for trends in occupancy and abundance both for target species and guilds for each park. The outcome may be used to develop new conservation guidelines, refine existing guidelines, and modify the current monitoring protocol. The results of this project may also direct changes in park management and policies. 3) Assess park trends in state and regional contexts by comparing GLKN data with that collected by the North American Breeding Bird Survey and other datasets to be identified.

High