Native palms, such as the Sabal palmetto, play an important role in maintaining the ecological balance in Florida. As a side-effect of modern globalization, new phytopathogens like Texas Phoenix Palm Decline have been introduced into forest systems that threaten native palms. This presents new challenges for forestry managers and geographers. Advances in remote sensing has assisted the practice of forestry by providing spatial metrics regarding the type, quantity, location, and the state of heath for trees for many years. This study provides spatial details regarding the general palm decline in Florida by taking advantage of the new developments in deep learning constructs coupled with high resolution WorldView-2 multispectral/temporal satellite imagery and LiDAR point cloud data. A novel approach using TensorFlow deep learning classification, multiband spatial statistics and indices, data reduction, and step-wise refinement masking yielded a significant improvement over Random Forest classification in a comparison analysis. The results from the TensorFlow deep learning were then used to develop an Empirical Bayesian Kriging continuous raster as an informative map regarding palm decline zones using Normalized Difference Vegetation Index Change. The significance from this research showed a large portion of the study area exhibiting palm decline and provides a new methodology for deploying TensorFlow learning for multispectral satellite imagery.

The implementation and application of Geographic Information Systems (GIS) and spatial analyses have become standard practice in many archaeological projects. In this study, we demonstrate how GIS can play a crucial role in the study of taphonomy, i.e., understanding the processes that underpinned the creation of archaeological deposits, in this case the distribution of artifacts across an archeological site. The Stelida Naxos Archeological Project (SNAP) is focused on the exploration of a Paleolithic-Mesolithic stone tool quarry site located on the island of Naxos, Greece. An extensive pedestrian survey was conducted during the 2013 and 2014 archeological field seasons. An abundance of lithic material was collected across the surface, with some diagnostic pieces dating to more than 250 Kya. Spatial statistical analysis (Empirical Bayesian Kriging) was conducted on the survey data to generate predictive distribution maps for the site. This study then determined the contextual integrity of the surface artifact distributions through a study of geomorphic processes. A digital surface model (DSM) of the site was produced using Unmanned Aerial Vehicle (UAV) aerial photography and Structure from Motion (SfM) terrain modeling. The DSM employed to develop a Revised Universal Soil Loss Equation (RUSLE) model and hydrological flow models. The model results provide important insights into the site geomorphological processes and allow categorization of the diagnostic surface material locations based on their contextual integrity. The GIS analysis demonstrates that the surface artifact distribution has been significantly altered by post-depositional geomorphic processes, resulting in an overall low contextual integrity of surface artifacts. Conversely, the study identified a few areas with high contextual integrity, loci that represent prime locations for excavation. The results from this study will not only be used to inform and guide further development of the archeological project (as well as representing significant new data in its own right), but also contributes to current debates in survey archaeology, and in mapping and prospection more generally. This project demonstrates the benefit of using spatial analysis as a tool for planning of pedestrian surveys and for predictive mapping of artifact distributions prior to archaeological excavations. / Thesis / Master of Science (MSc)