Habitat Differences, Niche Overlap, and Niche Partitioning between Two Sympatric Southern California

Erbas-White, Kevin

25085 Portsmouth

Mission Viejo, CA USA

Wu, Jindong

Department of Geography

California State University

Fullerton, CA USA

Nick, Kevin E.

Department of Earth and Biological Sciences

Loma Linda University

Loma Linda, CA USA

Differences in resource utilization by closely related sympatric animals can result from competitive (niche partitioning) or non-competitive processes. The current view for snakes, including vipers, is pervasive niche partitioning. However, two sympatric southern California rattlesnakes (Crotalus helleri and C. ruber) at a single study site exhibited differential niche use but negligible partitioning among the four major niche axes (space, temperature, time, diet). We therefore examined habitat differences and niche (space) overlap between the two species at a larger scale across their broad region of sympatry in southern California. We obtained snake location records (N = 800 and 443, respectively) from the North American Field Herping Association. We used ArcGIS to identify pertinent environmental variables for each location. Binomial logistic regression using seven predictors (rock category, mean NDVI, standard deviation of NDVI, elevation, mean temperature, temperature range, and precipitation) showed that all, except for SD-NDVI and precipitation, significantly discriminated species locations (combined 84.2% success). Of six rock substrate categories, C. helleri records were more numerous than expected in argillaceous-sandstone and sandstone-conglomerate rock, whereas C. ruber records were over-represented on phaneritic metamorphic/igneous rock. Crotalus helleri also occurred in areas of higher vegetation greenness (mean NDVI), lower elevation, cooler temperature, and less temperature variation than C. ruber. These findings confirm differential habitat use across the two species’ ranges in southern California. More importantly, null models using Monte Carlo simulations revealed that niche overlap indices were significantly less than expected of randomly-assembled pseudo-communities for two niche features: mean NDVI and temperature range. The latter results suggest niche partitioning of space use, and provide the strongest such evidence for any New World pitvipers.