Species Distribution Models and Abundance Estimates enhance Breeding Bird Atlas Data

Abstract

Breeding bird atlases play a crucial role in understanding bird species distribution and abundance during the breeding season. This information is essential for creating accurate species distribution maps, which are fundamental for understanding the geographic range of species and identifying areas of high conservation value. Our goals were to produce species distribution models and population estimates at the Minnesota state scale for as many breeding bird species as possible using data collected during the Minnesota Breeding Bird Atlas (MNBBA; 2009–2013). The MNBBA effort included volunteer atlas observations and systematic point-count bird surveys. We used three modelling strategies to maximize the number of species we modelled: 1) bootstrapped Poisson generalized linear models with a detectability offset to predict species’ density and population size, 2) bootstrapped Poisson generalized linear models to predict a species’ point count index of abundance, and 3) Maxent models to predict a species’ index of environmental suitability. The first two modelling strategies used point-count data from the MNBBA and Minnesota National Forest Breeding Bird Program. In addition to the point count data, we used georeferenced records from MNBBA volunteer atlas observations in our Maxent models. We applied the first strategy to 73 species, the second to 30, and the third to 33 species each (136 species in total). We also produced statewide population estimates for the 73 species using the first strategy. Here we present example results for three species in each group and discuss the merits of each method. We found that using multiple modelling approaches allowed us to model more species than if we had only used the most statistically rigorous method. We suggest testing these methods in other regions; however, as they provide hypotheses for further examination on predictions of species distributions and abundances for bird species over a large, statewide region using atlas data.