Anthropogenic and weather-related habitat fragmentation on barrier islands interact with factors that elevate perceived risk of predation to affect amount of habitat available for use by dune-dependent species, such as beach mice (Peromyscus polionotus spp). On two barrier islands along the Florida Gulf Coast, we conducted manipulative experiments using foraging trays to relate amount of food eaten (surrogate for perceived risk) to microhabitat and landscape structure (e.g., patch size, connectedness, distance to source dune) under ambient and heightened risk conditions (light or predator cues present). We planted broomsedge (Andropogon virginicus) to manipulate and test the influence of patch size (1, 2.5, and 5 m2) and corridor length (10 v. 20 m) on foraging, and tested effects of predator cues (cat, fox, and owl) and light (incandescent bulbs) as well. We also measured foraging in trays distributed throughout 1-ha grids (10-m spacing). Finally, we evaluated results of experiments in light of changes in beach mouse habitat predicted from models of sea level rise over the next century.
Results/Conclusions
Our findings indicate that owl cues, light, and distance to vegetation can increase risk perceived—and thus decrease activity—by beach mice. At the microhabitat scale, mice foraged less in the presence of owl cues, compared to other cues (least squares means (LSM) differences, p<0.05). At the patch level, mice foraged less when lights were present (LSM: t=-3.27, adj. p=0.05) and when lights and owls were present (LSM: t=-4.2, adj. p=0.02), compared to controls, but owls alone did not influence foraging. Patch size and corridor length did not affect foraging, but corridor presence was significant (more foraging with corridor) in summer (F=10.95, p=0.0004), but not winter. In the 1-ha grids, perceived risk increased as distance to vegetation increased (F=36.49, p<0.0001). With sea-level rise and extreme weather events, such as hurricanes, the best burrowing habitat for mice will occur on tall dunes with woody vegetation, but low un-vegetated areas will be important for movement. Our results suggest that management activities that reduce perceived risk and facilitate movement by mice, e.g., reducing artificial light near dunes and restoring vegetation following hurricanes, may be increasingly important as beach mouse populations are affected by climate change.