The following images are from the final presentation I gave to the Remote Sensing II course at the University of Southern Mississippi in May 2017. It was the culmination of a semester long research project using LiDAR data to extract features on the ground that lie hidden beneath 2 meters of marsh grass.
Using Remote Sensing to Locate Shell Middens in the Grand Bay National Estuarine Research Reserve. Completed as a requirement for GHY412 Remote Sensing II and presented on May 2, 2017. This project sought to determine if LiDAR can be used to locate Indian middens in a salt marsh. Native Americans inhabited the study area for approximately 12,000 years harvesting shellfish such are clams and oysters. The shell piles, or middens, left behind by the previous inhabitants lie hidden beneath a cover of tall marsh grass. The strategy was to take raw LiDAR data to locate manmade features in a salt marsh. The data was sourced from Mississippi Automated Resource Information System (MARIS). Using ARCGIS, the data was used to generate a digital elevation model (DEM) of the study area. Model builder was used to create a repeatable process of the analysis so that parameters (minimum area, minimum circularity, minimum elevation) could be adjusted for a comparative analysis. Spatial analysis of point density, mean center, standard distance, and directional distribution. During a later course in Python (GHY468 Spatial Automation And Programming), I created an ARCGIS tool with Python that allows a user to provide a raster image and a single variable, circularity, to generate a shapefile containing possible middens.
​
Click below to view the slides...
