Prof. Maxwell

I am currently an Associate Professor in the Department of Geology and Geography at West Virginia University. I teach geospatial science courses for both undergraduate and graduate students. I am also the director of West Virginia View, a consortium of public, private, and non-profit remote sensing organizations in West Virginia, and a faculty director of the West Virginia GIS Technical Center. Prior to coming to West Virginia University, I was an Assistant Professor at Alderson Broaddus University. Prior to teaching, I worked as a Remote Sensing Analyst at the Natural Resource Analysis Center (NRAC) at West Virginia University.

I am a graduate of Alderson Broaddus where I received bachelor degrees in Biology, Chemistry, and Environmental Science. I then attended West Virginia University where I earned a master degree in Geology followed by a PhD in Geology. I also hold a Geographic Information Systems Professional (GISP) certification from the GIS Certification Institute.

The primary objectives of my work are to investigate computational methods to extract useful information from geospatial data to make informed decisions and to train students to be effective and thoughtful geospatial scientists and professionals.

Prof. Maxwell

CV

Google Scholar

WVU Geol. and Geog.

WVGISTC

Projects

  • Extracting geomorphic features from LiDAR data using deep learning
  • Sinkhole extraction from digital terrain data using deep learning-based semantic segmentation and digital terrain data
  • Best practices for assessing deep learning output in remote sensing
  • Forest fuel load estimation with terrestrial LiDAR and machine learning regression
  • Generation of synthetic forest plots for use in predictive modeling
  • Community flood resiliency in West Virginia
  • Development of an R package for deep learning-based semantic segmentation applied to geospatial data (geodl)

Teaching

  • Geography 350/550: Introduction to GIScience
  • Geography 455/655: Introduction to Remote Sensing
  • Geography 456: Remote Sensing Applications
  • Geography 457/657: Open-Source Spatial Analytics
  • Geography 461/663: Client-Side Web GIS
  • Geography 462/662: Digital Cartography
  • Geography 520: Methods in Open Science
  • Geography 551: Open-Source GIScience
  • Geography 493/693: Geospatial Deep Learning

Research Interest

  • Spatial predictive modeling (classification and probabilistic prediction)
  • Accuracy assessment of remote sensing products
  • Application of machine learning, deep learning, convolutional neural networks (CNNs), and Mamba in the geospatial sciences
  • UNet-like architectures for geospatial semantic segmentation
  • Deep learning semantic segmentation and object detection
  • Digital terrain analysis and LiDAR
  • Geographic object-based image analysis (GEOBIA)
  • High spatial resolution land cover mapping and assessment
  • Geomorphic mapping and modeling
  • Forest fire fuel mapping and prescribed fire treatment
  • Wetland and forest type classification
  • Synthetic data generation

Graduate Students

  • Samira Rifat Prova (Geography MA): Water extent change analysis in Bangladesh using Sentinel-2 MSI data
  • Whitney Belcher (Geography MA): Creating flood fatality geospatial dashboards for West Virginia
  • Behnam Solouki (Geography MA): Comparing UNet and HRNet for geomorphic feature extraction from digital terrain model-derived land surface parameters
  • Muntasir Tabasum (Geography PhD): TBD
  • Matt Wozniak (Geography PhD): Persistent forests: topological data analysis of lidar-derived structure in forest ecosystems
  • Sarah Farhadpour (Geography PhD): Exploring generalized UNet-based deep learning semantic segmentation architectures for anthropogenic geomorphic feature extraction using LiDAR-derived land surface parameters
  • Sara Lusher (Geography MA, completed): Impact of flood zone boundary uncertainty on building-level risk assessment
  • Shannon Maynard (Geology MS, completed): Sinkhold extraction using geodl in R
  • Muhammad Ali (Geology MS, completed): Slope stability below historic pre-law mine benches in the northern coalfields of West Virginia
  • Shobha Yadav (Geography PhD, completed): Linkages between atmospheric circulation, weather, climate, land cover and social dynamics of the Tibetan Plateau
  • Faith Hartley (Geography MA, completed): Using Landsat-based phenology metrics, terrain variables, and machine learning for mapping and probabilistic prediction of forest community types in West Virginia
  • Jaimee Pyron (Geography MA, completed): Wetland mapping using machine learning, Sentinel-1, and digital terrain data
  • Caleb Malay (Geography MA, completed): Comparison of slope failure predictive models in different physiographic regions
  • Hartford Johnson (Geography MA, completed): Assessing fire recovery in California using time series analysis and the Landsat data archive