Assessing ecological risks at US Department of Energy facilities using methods borrowed from landscape ecology and habitat suitability analysis. Part I. Analysis of historical aerial photography and maps

Ecological risks are inherently spatial, and the consideration of scale is critical to all aspects of the ecological risk assessment process. Analysis of alterations in land use and habitat quality and quantity in relation to historical, continuing, and future US Department of Energy (USDOE) operations may be as important as assessing impacts resulting from the release of chemical contaminants into the environment. We examined several contributions from basic research and development in landscape ecology and habitat suitability analysis and their potential application in assessing ecological risks at USDOE facilities. Methods from landscape ecology are presented in this manuscript, while habitat suitability analysis methods and landscape-level ecological models are presented in a second manuscript. The Oak Ridge Reservation (ORR) in eastern Tennessee and the Savannah River Site (SRS) in western South Carolina were selected as representative USDOE facilities for this study. Changes in land use were analysed for sites within the ORR and the SRS using historical aerial photography, which was digitized in order to quantify the change over time using indices that describe landscape structure. A time series of changes in land use and habitat at a larger spatial scale was analysed by digitizing US Geological Survey (USGS) quadrangle maps (124,000) using two scales of resolution for an area of East Tennessee that included the ORR; landscape ecological indices were calculated, and USGS quadrangle maps that included areas inside and outside USDOE boundaries were compared. Historical alterations in land use and habitat type were successfully reconstructed for sites within the USDOE facilities at a small scale using aerial photography and at a large spatial scale for a USDOE facility and the surrounding landscape using a time series of USGS quadrangle maps; both data sources provided useful information. The percentage of the landscape occupied by a particular patch type, the number of patches of a particular type, the total number of patch types, the total number of patches, the total edge, the meanpatch fractal dimesion, and the contagion index were landscape ecological indices that adequately described the change over time that has occurred on the ORR and SRS. The methods presented in this manuscript could effectively be used in ecological risk assessments both at USDOE facilities and non-USDOE lands.