Landscape patterns and patch dynamics in Hamilton county over a forty year period: applicability to the conservation of the eastern box turtle

Exurban development is the fastest growing form of land use in the southeastern US and the primary driver of habitat and biodiversity loss. Species with long generation times such as the eastern box turtle (Terrapene carolina carolina) can persist in an urbanizing environment but have higher mortality than in forested habitat and show a response lag which delays detection of population decline. I quantify land use and land cover change over a forty year period by photo interpreting historic imagery that is orthorectified using a direct linear transformation model. A GIS database is created for three study sites and landscape pattern analyzed to determine the effects of historic land use on the eastern box turtle habitat using a suite of landscape metrics. A core habitat loss model is created using the core patch metric and box turtle life history traits, home range diameter, dispersal distance and re-generation. Spatial structure of fragmentation across time is characterized using global and local autocorrelation statistics and residual analysis of ordinary least squares (OLS) and geographically weighted regression (GWR) models using core as dependent variable and area, perimeter and mean slope as independent variables. Increasing fragmentation and road density over time is indicated by the landscape metrics for site 2 and 3. Regression model residual analysis suggests that the fragmentation trend at site 2 is clumped and scattered at site 3. All three sites lost forest and agriculture area and show an increase in urban and transportation area. Significantly 20% of the area of site 2 is being converted to urban land use since 1963. The rate per year of core forest loss at site 1 and 2 is decreasing and increasing at site 3 where the highest rate per year of core forest habitat loss was 8% between 1997 and 2007. Rate of core habitat loss per year is decreasing at site 1 and 3 and increasing at site 2 which lost 6% between 1997 and 2007. These rates of habitat loss suggest that site 1 could sustain three generations of box turtles until all core habitat has disappeared. However, sites 2 and 3 could not sustain one generation of box turtles until all core habitat is gone.