Hierarchical modeling to identify habitat associations of secretive marsh birds in the Great Lakes

Session: 09. - Modeling, Detecting, and Managing Rarity

Lisa Elliott, Conservation Sciences Program, University of Minnesota, [email protected]
Annie Bracey, Natural Resources Research Inst., [email protected]
Gerald Niemi, University of Minnesota-Duluth, [email protected]
Douglas Johnson, Department of Fisheries, Wildlife and Conservation Biology, University of Minnesota, [email protected]
Thomas Gehring, Department of Biology and Institute for Great Lakes Research, [email protected]
Erin Giese, UW-Green Bay's Cofrin Center for Biodiversity, [email protected]
Greg Grabas, Environment and Climate Chanage Canada, [email protected]
Robert Howe, UW-Green Bay, [email protected]
Christopher Norment, The College at Brockport, State University of New York, [email protected]
Douglas Tozer, Bird Studies Canada, [email protected]

Abstract

Secretive marsh birds are notoriously difficult to census because they are both uncommon and cryptic. Thus it is a challenge to identify regionally specific habitat associations, distributions, and population trends for them. To better understand the habitat associations of rare and declining wetland birds in the Great Lakes basin, we are developing six single-species, single season occupancy models using seven years (2011-2017) of bird survey data from the Great Lakes Coastal Wetland Monitoring Program and remotely sensed landscape data. These hierarchical models account for separate processes of occurrence and, given occurrence, detection. Preliminary results indicate that the probability of detection for Least Bittern (Ixobrychus exilis) and American Bittern (Botaurus lentiginosus) are influenced by time of day, whereas detection of Pied-billed Grebe (Podilymbus podiceps), Virginia Rail (Rallus limicola), Sora (Porzana carolina), and Common Gallinule (Gallinula galeata) is unaffected by the time of surveys. Human development is a primary landscape variable negatively influencing the probability of occurrence of American Bittern whereas human population size negatively influences the probability of occurrence of Least Bittern. Resulting models quantify species-specific habitat associations and will provide basin-wide predictive models on the distribution of rare, obligate coastal wetland birds to prioritize areas for conservation or potential restoration.

1. Keyword
avian ecology

2. Keyword
coastal wetlands

3. Keyword
Great Lakes basin

4. Additional Keyword
modeling

5. Additional Keyword
spatial distribution

6. Additional Keyword
Great Lakes Restoration Initiative (GLRI)