A model to predict the likelihood of cliff swallow nesting on highway structures in northern California

Abstract

Cliff swallows (Petrochelidon pyrrhonota) are colonially breeding migratory birds that frequently nest on highway structures. Under the Migratory Bird Treaty Act of 1918, people cannot harm swallows or their active nests. This restriction causes problems and delays for construction and maintenance divisions of many departments of transportation. In planning future projects, it would be useful for these divisions to have a habitat selection model that can predict the likelihood of cliff swallow nesting on a particular highway structure. We used logistic regression on data collected from 206 highway structures and 2 different land cover data sets to develop habitat selection models for northern California. The models indicated that low urban development and structure undersurfaces with multiple junctures were the 2 most important predictors of cliff swallow occupancy. Both the presence of water under a structure and a large underpass opening were also factors included in the models. The models correctly predicted 59% of sites occupied by cliff swallows and 88% of sites not occupied. The occupancy classification rate may offer departments of transportation useful insight into the nesting behavior of cliff swallows.

Keywords

• bridge
• classification
• cliff swallow
• habitat selection model
• highway structure
• human–wildlife confl icts
• logistic regression
• nest
• occupancy
• petrochelidon pyrrhonota
• prediction
• transportation