Agricultural and urban sprawl can fragment terrestrial habitats, modifying biodiversity and community structure. Fragmentation alters the richness and composition of plant communities, but to what degree this affects the small mammal community is less clear. One way to assess how fragmentation affects the small mammal community is to assay seed predation as a proxy for foraging behavior. We asked whether fragment size influences small mammal seed predation in an experimentally fragmented landscape. Seed predation was quantified by deploying a total of 247 seed depots in large (n=6), small (n=5), or matrix (n=5) treatments in forested patches situated in a grassland landscape at the University of Kansas Field Station. Depots were Ziploc® containers with lid for excluding birds, and two 1'' holes for the entry and exit of granivores. We placed 25 heat-killed Helianthus anuus seeds in each sand-filled depot prior to field deployment. After six nights, we collected the depots and counted the remaining seeds.
Results/Conclusions
The average proportion of seeds consumed was significantly higher in large contiguous patches compared to fragmented habitats (clusters of small patches) (Wilcoxon rank sum, W=2, p=0.017) and compared to depots deployed in the matrix (W=22, p<0.001). Larger intact foraging area on large fragments could accommodate larger populations of small mammals or large-bodied species that require more space and consume more seeds. Evidence of larger-bodied mammals on large patches is supported by the presence of 24 Neotoma floridana middens found exclusively in large patches. However, midden density did not correlate with seed predation (Spearman’s rank correlation, S = 56, p=0.24). Continuous habitat and woody cover has been shown to increase movement on large patches in our landscape. More movement would, in turn, increase the chance of encountering a depot, increasing seed predation relative to small patches. In contrast, movement in small patches may be limited due to perceived predation risk in exposed areas between fragments, thereby reducing average granivory. We cannot distinguish whether seed predation is higher on large patches due to larger mammal populations, more movement, or larger-bodied species. Nonetheless, our results show that patch size influences seed predation. If rodents selectively forage certain species’ of seeds, higher seed predation on large patches may influence plant community composition.