Community Structure And Differential Response Of Turtles To Agriculturally Induced Fragmentation: A Basin-Wide Assessment
Carol E. Rizkalla, Department of Forestry and Natural Resources, Purdue University, 195 Marsteller St., West Lafayette, IN 47907; (765) 494-3599; FAX (765) 496-2422; crizkall@purdue.edu
Robert K. Swihart, Department of Forestry and Natural Resources, Purdue University, 195 Marsteller St., West Lafayette, IN 47907
Wetland loss throughout the United States has motivated an increasing number of studies which attempt to use landscape characteristics to explain the diversity and abundance of herpetofauna. Most studies to date have focused on amphibians rather than reptiles. As part of the Upper Wabash Ecosystem Project, spatially hierarchical sampling was conducted from 2001-2003 in 35 randomly selected 9-mi2 cells throughout the upper Wabash River basin in Indiana. In addition to terrestrial captures of eastern box turtles, hoop nets were used at wetlands to capture snapping turtles (Chelydra serpentina) (n = 258), painted turtles (Chrysemys picta picta) (151), spiny softshells (Apalone spinifera spinifera) (70), red-eared sliders (Trachemys scripta elegans) (59), map turtles (Graptemys geographica) (27), false map turtles (Graptemys pseudogeographica) (6), Blanding's turtles (Emydoidea blandingii) (3), and musk turtles (Sternotherus odoratus) (3). We use nestedness analysis to examine the degree to which aquatic species were nonrandomly distributed among the 35 landscapes. The occurrence and abundance of several species also were modeled to test hypotheses regarding the importance of local, patch, and landscape-level variables. Factors at multiple spatial scales can affect turtle distributions, suggesting differential responses to landscape fragmentation.