At fine scales, urban land covers have species-specific effects on local bat distributions. For example, Mexican free-tailed (Tadarida brasiliensis) and big brown (Eptesicus fuscus) bats are urban adapters. In contrast, the evening bat (Nycticeius humeralis) prefers non-urban land cover types. Tree dwelling species such as red (Lasiurus borealis) and tri-colored (Perimyotis subflavus) bats, on the other hand, are limited, by urban land covers to small forest remnants. However, it is still unclear at a broad scale, to what extent urban land covers collectively affect bats or whether the spatial configuration of urban land cover types matter.
Our objective was to investigate the effect of urbanization on bat distribution with respect to land cover composition and configuration at a broad landscape scale. Using the North American Bat Monitoring Program (NABat) acoustic mobile transect survey protocol, we sampled 32 NAbat grids (10km by 10km) throughout the state of North Carolina in 2015. We collected 3322 acoustic recordings. We used the National Land Cover Database 2011 and FRAGSTATS to extract land cover information. We constructed generalized additive models to explore the relationships between bat activities and indices of land cover proportion, diversity, evenness, fragmentation, and shape.
Results/Conclusions
Preliminary analyses showed that open space urban development (<20% impervious surfaces) correlated positively with activity of big brown (coefficient 0.250, p = 0.013), silver-haired (Lasionycteris noctivagans, coefficient 0.155, p = 0.006), Mexican free-tailed (coefficient 0.110, p = 0.005) bats, and negatively with tri-colored bat (coefficient -0.218, p = 0.023). Mexican free-tailed bat activity positively correlated with high intensity urban development (>80% impervious surfaces, coefficient 0.848, p = 0.032). No association with urban land cover was found for red, hoary (Lasiurus cinereus), or evening bats. However, evening bat activity positively correlated with overall land cover type evenness (coefficient 7.753, p = 0.008), suggesting adding anthropogenic land cover types might impact this species even though it prefers no particular land cover type. Fragmentation of open space urban development increased Mexican free-tailed bat activity (coefficient 0.022, p = 0.027), but decreased tri-colored bat activity (coefficient -0.056, p = 0.017). Shape index of low intensity development (20-50% impervious surfaces) positively affected big brown bat (coefficient 0.005, p = 0.013) activity. This study suggests both land cover composition and configuration are important at a landscape scale for the distribution in activity of multiple bat species.