| Remote sensing data play a key role for assessing wildlife habitat distribution and conservation. However, most efforts have depended on passive remote sensing data that poorly characterize the three-dimensional (3-D) structure of vegetation, an important variable influencing animal-habitat associations. In this thesis, we evaluated the consequences of integrating novel data of ecosystem 3-D structure from LiDAR (i.e. light detection and ranging) into current assessments of wildlife habitat distribution and conservation, with the main goal of quantifying LiDAR value for biodiversity and wildlife management. Using data from temperate and tropical landscapes (i.e. Idaho and Puerto Rico), this research exhibited the value of LiDAR data in characterizing key forest structure components for wildlife species, such as snags and understory shrub distribution, as well as for improved assessments of wildlife habitat suitability. In this sense, LiDAR helped to refine species-habitat models in ways not attained using traditional remote sensing technologies, making it possible to delineate known species associations with forest structure. In addition, LiDAR significantly improved the accuracy of current Landsat-based forest type classifications, which represent the principal source of geospatial data used in wildlife habitat studies. Finally, this research showed that incorporating remotely sensed data of vegetation structure can improve the results of regional conservation efforts such as Gap Analysis. This thesis demonstrated that LiDAR remote sensing has a great value for improved wildlife habitat assessments, providing unique opportunities to advance the way we manage and conserve biodiversity and habitats. |
