Mammalian Behavioral Ecology in Southern California Habitat Fragments Statement of Problem Results Discussion This project assessed the likelihood and possible consequences of human-wildlife conflict in habitat fragments within a suburban matrix. The effects of human presence on the mesopredator release hypothesis and the behavioral ecology of nine mammal species were also analyzed, focused on Canis latrans, Procyon lotor, and Felis catus. This study has shown that other factors within habitat fragments can create a significant effect on mammalian behavioral ecology. There was a significant difference in the number of captures between the control and impacted canyon, suggesting a reduction of animal movement due to human presence. This trend has been shown to exist globally and is a possible exclusionary factor for larger mammalian predators such as cougars and bobcats. Species richness was reduced in the impacted canyon, which is a trend observed in previous literature, and warrants the conservation of fragments where human activity is restricted. Species that were only present in the control canyon were Didelphis virgiana, Neatoma fuscipes, and Vulpes vulpes. As these mammals vary greatly in mass, body size most likely is not a strong contributing factor to the absence of a mammal species in a fragment with human presence. Mephitis mephitis, D. virgiana, and N. fuscipes were the only species observed solely at night. These behaviors do not show change from natural patterns. P. lotor and C. latrans captures were recorded during the day only in the control canyon, suggesting that human presence may shift activity of mammals toward nocturnality, which is supported by previous literature and aligns with literature that observed coyotes primarily rest in chaparral fragments during the day. The coyotes that were captured on video in this study exhibited no observable fear of the cameras, even staring directly into the camera during the day without postural changes, in one instance. This conflicts with previous literature showing that coyotes are wary of camera installations. However, the aforementioned literature involved coyotes living within an unfragmented landscape and increased fragmentation may lead to a reduction in coyote wariness of camera installations. Furthermore, if conflict does occur in these fragments, it is unlikely to end in serious human injury. This is due to 98% of wildlife captures representing either P. lotor or a species of smaller body size. These species present little to no danger to humans, and since 99% of human activity was recorded during the day, conflict is likely of little consequence. Table 1. All captures sorted by species and location type. Introduction The Palos Verdes Peninsula is a suburban matrix, with habitat fragments of varying quality lying within suburban development. This landscape poses two main issues in terms of wildlife management. Firstly, the proximity of human development to wildlife habitat leads to human-wildlife conflict. On the Palos Verdes Peninsula, coyotes are the only large mammalian predator that has not been extirpated. Coyotes have shifted activity towards nocturnality in urban areas and have shown to avoid areas most associated with human activity. Coyote occurrence has even been shown to be positively correlated to urban proximity. These abilities have lead to confrontations between the coyote and humans. Coyotes have been shown to forage in human areas at night while resting in chaparral during the day in Southern California. However, the canyon fragments in Palos Verdes experience the greatest human traffic during the day, and coyotes may be exhibiting an altered behavior to reduce conflict. Additionally, these habitat fragments do not ecologically function identically to larger habitat areas, creating a novel set of pressures for the wildlife species still persisting. As these natural places become smaller and more isolated from one another, the quality of habitat for certain species is reduced and for others has been shown to increase. P. lotor, Mephitis mephitis, and F. catus, have been shown to positively benefit from fragmentation, while predators such as C. latrans have reduced occurrence as fragments become smaller and more isolated. As fragmentation negatively affects large mammal predators and positively affects mesopredators, fragments are areas where a mesopredator release could occur. In another Southern California study, a high presence of larger mammal predators, especially C. latrans, occurred with a low occurrence of raccoons and feral cats. Figure 1. Captures of each species by location type Figure 2. Location type captures per species Figure 4. Daytime capture of C. latrans along trail in control canyon. Figure 3. Domestic captures per location type Figure 5. P. lotor captures by location type Figure 6. F. catus captures by location type Figure 7. C. latrans captures by location type Future Management Implications Materials and Methods This study has demonstrated that factors within fragments are ecologically significant and should be taken into account when managing wildlife within a fragmented landscape. The presence of humans and domestic dogs was shown to reduce species richness, and thus conservation plans should incorporate areas where human presence is limited. As human development continues, further research into coyote conflict is warranted as they have shown reduced wariness to manmade structures within fragments and have been able to successfully adapt to living within urban and suburban habitat matrices. As habitat continues to be fragmented, further research is required as significant differences in behavioral ecology of mammals have occurred in fragmented landscapes. Trail cameras, trail camera security boxes, security cables, memory cards, batteries, T-stakes, and tree straps were used in the field. The two canyons that were studied were Margate Canyon (control) and Agua Amarga Canyon (human presence). Cameras were installed in two pairs per canyon, with a creekbed and trail camera along the same perpendicular axis to the canyon’s length, for a total of 8 camera installations. The cameras were in place from 5-3-17 to 11-18-17 for a total time of 6 months and 16 days. A capture is when one animal is present in a video, and multiple individuals in the same video count as multiple captures. Chi-Square analysis was used to determine if differences in captures between location types were significant. Figure 8. S. niger captures by location type Figure 9. D. virgiana captures by location type Figure 10. V. vulpes captures by location type Figure 11. N. fuscipes captures by location type Figure 12. M. mephitis captures by location type Figure 13. S. audubonii captures by location type All photos taken by author. Special Acknowledgements to Dr. Stankowich, Rita Collins, Thomas Jankowski and Josh Weinik for support.