Species Distribution Models and Abundance Estimates Enhance Breeding Bird Atlas Data.

Area of Science:

• Ecology
• Conservation Biology
• Ornithology

Background:

• Breeding bird atlases are vital for understanding bird distribution and abundance.
• Accurate species distribution maps are crucial for conservation efforts.
• Data from the Minnesota Breeding Bird Atlas (MNBBA) presented variability in structure and quality.

Purpose of the Study:

• To develop species distribution models (SDMs) for as many Minnesota breeding bird species as possible.
• To maximize species coverage despite heterogeneous atlas data.
• To produce statewide population estimates for modeled species.

Main Methods:

• Applied three distinct modeling strategies tailored to data characteristics: bootstrapped Poisson GLMs with detectability offset, bootstrapped Poisson GLMs, and Maxent models.
• Utilized data from the Minnesota Breeding Bird Atlas (2009-2013), combining volunteer observations and point-count surveys.
• Modeled 136 species in total, with 73 receiving density and population estimates.

Main Results:

• Successfully modeled 136 Minnesota breeding bird species by employing multiple modeling approaches.
• Demonstrated that linking model choice to data structure significantly enhances species coverage compared to single-model approaches.
• Generated statewide population estimates for 73 species.

Conclusions:

• The adaptable methodology effectively maximizes the number of species modeled from diverse atlas datasets.
• The developed models provide a foundation for broad-scale distribution and abundance hypotheses.
• This framework can be tested in other regions to enhance the utility of various atlas datasets for conservation and research.