new icn messageflickr-free-ic3d pan white
Predictive Modeling of Cave Entrances Utilizing Hyperspectral Imagery and Digital Elevation Models | by Chuck Sutherland
Back to photostream

Predictive Modeling of Cave Entrances Utilizing Hyperspectral Imagery and Digital Elevation Models

Tennessee is home to 10,161 known caves, more than any other state. Estimates suggest that there may be up to 2,000 more caves yet to be discovered. In this study, we attempt to answer how well cave entrance locations can be predicatively modeled using GIS. Landsat 8 imagery, and DEM based surfaces were used to assess predictive capability of the model with the goal of finding new, undescribed caves. Training data to run the model was compared against testing caves, and randomized points to determine significant analysis.

2,213 views
5 faves
1 comment
Uploaded on April 5, 2016