Date of Graduation


Document Type


Degree Type



Eberly College of Arts and Sciences


Geology and Geography

Committee Chair

Timothy Warner

Committee Co-Chair

Nicholas Skowronski

Committee Member

Nicholas Skowronski

Committee Member

Aaron Maxwell

Committee Member

Cedric Landenberger

Committee Member

Neil Gifford


Keystone structure is the spatial structure required by a given species, at a scale that is determined by that species’ needs and mobility. The endangered Karner blue butterfly (Lycaeides melissa samuelis, hereafter KBB) has a keystone structure that incorporates trees and bushes to provide the mixture of sun and shade required to fulfil its life functions. Airborne light detection and ranging (lidar) is a potentially invaluable tool for characterizing keystone structures. However, lidar has yet to be utilized to evaluate structural suitability of KBB habitats. Therefore, I investigated the use of lidar for characterizing critical attributes of KBB habitat structure, and its use in the evaluation of management practices. Structural diversity was summarized from lidar using two approaches: one that attempted to test the canopy cover criteria used in the field-based Glacial Lake Albany habitat mapping (hereafter GLA heterogeneity), and a second based on the texture of the lidar-derived canopy cover imagery. These lidar-derived measures were calculated at five scales, using kernels (moving windows) with areas of 0.05 ha to 19.2 ha. The lidar heterogeneity measures derived at 0.9 ha or less were highly correlated with density of field observations of KBB presence, with the highest correlation at 0.2 ha. Larger kernels were poorly correlated with KBB presence. Notably, the 0.9 ha scale corresponds to more than 75% of KBB mobility range observations, as reported in a previous field study. GLA heterogeneity was also found to be consistently more correlated with KBB observations than the texture measure. The criteria used to establish the four GLA heterogeneity classes appear to be useful, based on rank correlation relationships with the classes were combined or evaluated individually. The 0.2 ha kernel GLA heterogeneity was used to evaluate the effects of prescribed burning on structural suitability, and was found to be significantly correlated with burn intensity.