The influences of patch shape and boundary contrast on insect response to fragmentation in California grasslands

Landscape ecologists typically identify boundaries to demarcate habitat patches. The boundary between two habitat types may be abrupt, such as the transition between a grassland and a parking lot, or more gradual, such as the shift between successional forest stages. Two key aspects of landscape boundaries, their shape and contrast, are predicted to influence movement of materials, plants, and animals. Ecological theory suggests that a patch's perimeter-to-area ratio should strongly influence animal emigration when patch boundaries are relatively permeable, but not when boundaries are more severe. We investigated the interactive effects of patch shape and boundary contrast on movement of ground-dwelling beetles (Carabidae and Tenebrionidae) in native grassland habitat at Jepson Prairie, Solano County, California, USA. We conducted a field experiment with two patch shape treatments, square and rectangle, that held patch area constant, and two boundary contrast treatments created by mowing grass surrounding each plot at two different heights. We monitored the number of beetles leaving each patch over a three-week period following treatment establishment. We observed a significant effect of boundary contrast on net movement of beetles, with low contrast boundaries exhibiting net immigration and high contrast boundaries experiencing net emigration. Moreover, the importance of patch shape appeared to be greater for low contrast versus high contrast boundaries, consistent with theoretical expectations. Our combined observations indicate that these ground-dwelling beetles were more likely to move into patches that were rectangular and surrounded by a low contrast matrix than patches that were square or surrounded by a high contrast matrix. We conclude that net movement of beetles across patch boundaries is strongly influenced by boundary contrast and may be affected by patch shape when boundary contrast is low.