Impacts of Human Modification on Predator Behavior
Bobcats exposed to more human modification selected for more agriculture (Figure 5 upper row). This relationship was positive and quadratic in the mean of all temporal periods, quadratic in late winter, and linear during the spring/summer day temporal periods. There were also mostly positive quadratic relationships in the spring/summer day and night periods. Exurban selection and human modification exhibited a mean linear relationship and a negative linear relationship during the late winter season, indicating that bobcats avoided exurban habitat with increased human modification in late winter. Bobcats also exhibited negative linear trends in the relationship of human modification and distance to water in the mean, late winter, and spring/summer crepuscular periods, indicating selection closer to water with increased human modification.
Coyote selection for agriculture varied in response to human modification. Agriculture selection linearly increased in the mean and spring crepuscular and night periods and linearly decreased in the spring day period with increasing human modification (Figure 5 bottom row). There was no mean relationship between coyote selection for exurban habitat and human modification. However, there were two negative linear trends in the spring crepuscular and night periods, and a mostly positive quadratic relationship in the late winter crepuscular period. Coyote selection had a mean quadratic relationship between human modification and distance to water, which increased and then decreased, as well as slight linear negative trends in spring day, crepuscular, and night periods. There were stronger negative linear trends in the summer/fall crepuscular and night periods, and a negative quadratic trend in the summer/fall day period indicating that, in general, coyotes selected areas farther from water as human modification increased. Distance to road regressions yielded intercept-only top models for all temporal periods and means for both bobcats and coyotes, indicating no relationship between intensity of human modification and distance to road (Supporting Information).