Use of restored grasslands by mammals in a dynamic agroecosystem: insights from camera trapping

Habitat loss is one of the leading causes of endangerment for terrestrial vertebrates. For instance, 99.9% of the tallgrass prairie has been lost in Illinois. Restoration programs such as the Conservation Reserve Program (CRP) and State Acres for Wildlife Enhancement (SAFE) were created in part to ameliorate grassland loss. Previous assessments of responses of mammals to grassland restoration efforts have not focused on medium to large sized species because of sampling difficulties. More generally, few assessments of restoration outcomes consider effects of landscape context. I integrated camera trapping with occupancy modeling for two seasons (summer and winter) to assess mammal responses on 30 restored grassland sites in a dynamic agroecosystem in Illinois from 2014 to 2015. I tested hypotheses about the effects of local habitat conditions and landscape context on use of restored grasslands by four focal species: raccoons (Procyon lotor), eastern cottontails (Sylvilagus floridanus), coyotes (Canis latrans), and white-tailed deer (Odocoileus virginianus). Most species showed seasonal differences in grassland use that reflected the dynamic nature of the surrounding agricultural matrix (i.e., loss of hiding cover and supplemental food due to crop harvesting). Overall, landscape context was important in determining use of created grasslands. For instance, distance to nearest forest was the main predictor of site occupancy for raccoons, which has management implications regarding future site enrollment to reduce predation on grassland songbird nests. The red fox (Vulpes vulpes) was not detected on any site despite >5,000 camera nights of sampling, which points to the need for further monitoring to determine the status of this small canid in non-urban areas of the Midwest. Grasslands created by the CRP and SAFE programs provide habitat for medium and large mammals, but use of these habitats strongly depends on the temporally dynamic matrix and landscape context