Get access

A Spatial Modeling Approach to Identify Potential Butternut Restoration Sites in Mammoth Cave National Park

Authors

  • Laura M. Thompson,

    Corresponding author
    1. Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, U.S.A.
    Search for more papers by this author
  • Frank T. Van Manen,

    1. U.S. Geological Survey, Southern Appalachian Field Branch, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, U.S.A.
    Search for more papers by this author
  • Scott E. Schlarbaum,

    1. Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, U.S.A.
    Search for more papers by this author
  • Mark DePoy

    1. National Park Service, Division of Science and Resource Management, Mammoth Cave National Park, Mammoth Cave, KY 42259, U.S.A.
    Search for more papers by this author

Address correspondence to L. M. Thompson, email lthomp11@utk.edu

Abstract

Incorporation of disease resistance is nearly complete for several important North American hardwood species threatened by exotic fungal diseases. The next important step toward species restoration would be to develop reliable tools to delineate ideal restoration sites on a landscape scale. We integrated spatial modeling and remote sensing techniques to delineate potential restoration sites for Butternut (Juglans cinerea L.) trees, a hardwood species being decimated by an exotic fungus, in Mammoth Cave National Park (MCNP), Kentucky. We first developed a multivariate habitat model to determine optimum Butternut habitats within MCNP. Habitat characteristics of 54 known Butternut locations were used in combination with eight topographic and land use data layers to calculate an index of habitat suitability based on Mahalanobis distance (D2). We used a bootstrapping technique to test the reliability of model predictions. Based on a threshold value for the D2 statistic, 75.9% of the Butternut locations were correctly classified, indicating that the habitat model performed well. Because Butternut seedlings require extensive amounts of sunlight to become established, we used canopy cover data to refine our delineation of favorable areas for Butternut restoration. Areas with the most favorable conditions to establish Butternut seedlings were limited to 291.6 ha. Our study provides a useful reference on the amount and location of favorable Butternut habitat in MCNP and can be used to identify priority areas for future Butternut restoration. Given the availability of relevant habitat layers and accurate location records, our approach can be applied to other tree species and areas.

Ancillary