COS 15-1 - Variable effects of Arundo donax and prolonged drought on carnivore occupancy in southern California riparian ecosystems

Monday, August 7, 2017: 1:30 PM
D137, Oregon Convention Center
Molly Hardesty-Moore, Ecology, Evolution, and Marine Biology, University of California, Santa Barbara, Santa Barbara, CA and Douglas J. McCauley, Ecology, Evolution, and Marine Biology, University of California at Santa Barbara, Santa Barbara, CA

Very few regions of native riparian habitat still exist in southern California, and those that do are currently under threat by invasive species and prolonged drought. Arundo donax is thought to be the most impactful invasive plant in these ecosystems, with significant negative effects on birds and arthropods. Less is known about A. donax effects on California’s native carnivores, which rely on riparian ecosystems for water, prey, and temperature regulation. This study aimed to examine the combined effects of severe drought and A. donax abundance on native carnivore distributions in the Santa Clara River (SCR). We predicted the probability of carnivore occupancy would be negatively correlated with A. donax abundance but that there would be a stronger effect of A. donax during the wet season when overall resource availability is high, allowing avoidance of less preferred habitat. We assessed the variability of A. donax in four sites along the SCR, classifying patches into native (0-30% A. donax), mixed (30-70% A. donax), and Arundo (70-100% A. donax), and tested carnivore occupancy using 24 remote-activated cameras during two time periods: before and after the heavy rains of winter 2017. Occupancy for each species was modeled using single-season occupancy models in program PRESCENSE.


Preliminary results indicate a general trend of reduced occupancy in sites with high A. donax abundance: There were 68% more discrete (>1 min apart) sightings of carnivores in native patches than Arundo patches, but these results were not consistent at each site and for each species. The probability of coyote occupancy was 62% as likely in Arundo patches as native patches, supporting our hypothesis that A. donax decreases occupancy. However, the probability of bobcat occupancy was 352% as likely in Arundo patches as native patches, refuting our hypothesis. These results indicate an influence of factors not considered in this study, such as species-specific life history traits, habitat structure, and resource availability. It seems that A. donax does have the predicted effect of decreasing carnivore occupancy overall, but attention must be paid to variability in sites as well as species.