Impacts of density, proximity, and spatial clustering of residential development on puma prey composition

Background/Question/Methods

Although diet composition has been documented to shift in modified habitats, many studies treat modification as a binary condition despite its diversity in scale and form. To investigate patterns between resource use and residential development, we examined the influence of different housing descriptors on prey composition in pumas at multiple spatial scales. We fitted 29 pumas with GPS collars and identified potential kills sites from GPS locations. We investigated 1111 clusters in the field, 379 of which were kill sites. To investigate the spatial extent of housing impacts on prey composition, we compared average distance to the nearest house, spatial clustering of houses within 1 km, and housing density within 150 m, 500 m, and 1 km at kill sites of large prey (greater than 20 kg) and small prey. We compared prey diversity at all three spatial scales using five housing density designations: no housing, rural, exurban, suburban, and urban. Finally, we used linear regression models to examine the relationship between housing density within each puma’s 95% LoCoH home range and both individual diversity indices and percent large prey consumed in order to determine if population patterns or individual adaptation best explained increasing consumption of alternative prey.

Results/Conclusions

We found that prey diversity and proportion of small prey consumed increases with housing density at all three scales, with the most significant jumps occurring at the suburban level. Diversity was lowest when no housing was present, suggesting that in a pristine state, alternative prey are likely rarely utilized. Large prey were more likely to be killed farther from housing and at lower housing densities at all scales than small prey. Housing density in individual puma home ranges was not correlated to diversity or proportion of small prey consumed, suggesting that high consumption of alterative prey in modified habitats may be a learned behavior. These findings indicate that impacts of development on puma prey composition are substantial and conserved even at large scales. Similarities across spatial scales are potentially due to extensive low density development in our study area and reflect widespread impacts of housing on fundamental feeding behaviors. In addition, our results suggest that pumas are able to behaviorally adapt to human development by consuming a diversity of prey species, but they vary considerably in the degree of their foraging responses. More work is needed to explore the energetic consequences of increasing alternative prey consumption in modified landscapes.