This session will feature a range of remote sensing and GIS projects primarily from our student attendees.
10:30 AM – 10:45 AM – Assessing Socio-Economic Influences on the Spatial Distribution of Building Damage from Hurricanes Helene and Milton
This study underscores the differential impact of disasters on communities with varying socio-economic conditions, demonstrating that disparities in damage levels are influenced not only by the physical intensity of hazards but also by communities’ capacity to prepare for and respond to such events. Specifically, this research examines the spatial distribution of building damage resulting from Hurricane Milton in 2024 and explores the socio-economic disparities associated with this damage. Utilizing FEMA damage assessment data, the study conducts a spatial analysis of building damage patterns. A statistical approach is then employed to evaluate the relationships between building damage, hurricane intensity, and socio-economic factors at both the building and census tract levels. Furthermore, geographically weighted regression (GWR) is applied to assess the spatial heterogeneity of factors influencing damage severity. The findings offer critical insights into the key determinants of building damage from events such as hurricanes and their spatial variability. By identifying the most vulnerable communities, this study provides valuable information for policymakers, emergency management officials, and community leaders to develop targeted strategies aimed at mitigating the impacts of future hurricanes.
Najiba Rashid, Oregon State University
10:45 AM – 11:00 AM – Utilizing GeoAI for Rapid and Accurate Damage Assessment
Geographic Information System (GIS) Artificial Intelligence [GeoAI] is a tool for analyzing and interpreting large spatial data with many applications. GeoAI enhances interpretation, analytics, and scalability of spatial data processing. GeoAI can extract data that may be hidden that normally would be looked over. GeoAI can aid a geospatial analyst in various needs and expedite the process of analyzing datasets and complex functions, thus increasing productivity. Subsets of GeoAI such as machine learning [ML] and deep learning [DL], allow for the automation of complex tasks such as data classification, object detection, and pattern recognition. AI/ML/DL models re[1]quire a substantial amount of data to train a model. To reduce the amount of data needed to train a model, Pre-trained deep learning models were used. Pre-trained AI models are ML models that are trained on large datasets that can be implemented in a GIS environment to address problems such as hurricane damage assessment in south Louisiana. The objective of this presentation is to showcase the significance of GeoAI in damage assessment classification and detection. Using pre-trained DL AI models as well as developing our own DL models, post-hurricane damage assessment and other use case in South Louisiana were attempted. The results from this ongoing project are presented.
Samuel Landry, Nicholls State University
11:00 AM – 11:15 AM – Enhancing Early Water Stress Detection in Dryland and Irrigated Maize using Multi-Modal Remote Sensing
Accurate assessment of crop water status is vital for timely management interventions, maximizing productivity, and ensuring efficient use of water resources. This study investigates the efficacy of eight spectral and thermal indices in detecting both short and long-term water stress conditions (represented by stomatal conductance and soil moisture variability, respectively) across maize growth stages under dryland and irrigated conditions. Conventional vegetation indices derived from spectral reflectance were compared with multi-dimensional indices that integrate weather information. To further enhance the predictive capacity of the multi-dimensional indices, four regression models—including machine learning approaches—were applied to estimate hot and cold pixel values of the indices. Statistical analysis was done to identify which among the selected indices showed consistently stronger correlations with stomatal conductance and soil moisture variability across crucial growth stages. Notably, thermal-derived multi-dimensional indices demonstrated greater sensitivity to short term stomatal response, while multispectral-based indices showed greater sensitivity to critical long-term periods of soil moisture stress. Among the regression models tested, the random forest algorithm outperformed others, yielding the highest predictive accuracy and lowest estimation errors. The findings emphasize the need to monitor water stress periods, as well as account for weather-specific influences when developing stress indices. By leveraging the complementary strengths of multiple indices and incorporating environmental influences, this approach offers a robust framework for early detection of crop water stress, providing critical insights for advancing precision agriculture practices.
Kelechi Igwe, Kansas State University
11:15 AM – 11:30 AM – Mapping Seagrass-Coral Reef Mosaics Using Satellite Imagery to Quantify Seascape Structure in the Florida Keys
Seagrass meadows are vital benthic habitats that support biodiversity, facilitate biogeochemical cycling, and provide nursery and foraging grounds for a range of ecologically and economically important species. In the Florida Keys, seagrasses form dynamic mosaics with coral reef habitats, shaping the structure of benthic ecosystems across the seascape. Understanding the spatial structure of these mosaics (i.e., how seagrass and reef patches are distributed, shaped, and arranged) is critical for informing management decisions. However, these structural patterns remain under-characterized, especially at fine spatial resolutions. This study aims to quantify the seascape structure of seagrass–coral reef systems across three areas of interest along the Florida Keys Reef Tract: the Upper, Middle, and Lower Keys. These regions span a gradient of environmental conditions and disturbance regimes and were selected to reflect spatial heterogeneity within the broader reef tract. By integrating high-resolution PlanetScope satellite imagery (3-meter spatial resolution) with in situ ground-truth data, we developed benthic habitat maps tailored to the ecological characteristics of each area. The mapping approach combined supervised classification with a machine learning Random Forest algorithm, trained using field-based observations of benthic cover collected by the project team. The resulting maps varied in thematic resolution across the three regions, depending on local complexity and spectral properties of both habitats and data. Habitat classes include seagrass meadows, coral reef structures, bare sand, and hardbottom features. Classification schemes were customized per region to ensure ecological coherence and maximize classification accuracy. From the classified maps, we derived a suite of spatial pattern metrics to characterize the structural properties of seagrass beds as well as their spatial relationship to adjacent coral reef features. These metrics provide quantitative insight into habitat configuration, heterogeneity, and connectivity, which are critical drivers of ecological interactions such as species movement, trophic linkages, and resilience to disturbance. By integrating satellite remote sensing with field-based validation and landscape ecology techniques, this study advances our capacity to assess benthic habitat structure in complex seascapes. The resulting spatial data products and metrics offer valuable tools for resource managers aiming to protect and sustain the ecological integrity of the Florida Keys coastal seascapes.
Marianna Coppola, Florida International University
11:30 AM – 11:45 AM – Mapping Tomorrow: Igniting Geospatial Futures
A discussion for college students and folks that are early career geospatial. Focuses on talking about what skillsets are the most important for early career geospatial folks to succeed and advance in the industry. The discussion concludes with what ideas do early careers have to improve and revolutionize the industry, and how can we all get involved to support these.
Kimberly Mantey, U.S. Geological Survey: National Geospatial Technical Operations Center
11:45 AM – 12:00 PM – Integrated Geospatial Analysis of Burn Severity and Vegetation Recovery in the August Complex Fire (2020)
The August Complex Fire of 2020 was the largest wildfire in California history, burning over one million acres across the Mendocino National Forest. This study evaluates vegetation recovery over a 10-year period (2015–2025) integrating Landsat-derived Normalized Burn Ratio (NBR), differenced NBR (dNBR), hotspot analysis, and lidar-derived elevation and vegetation structure metrics. NBR change was assessed for pre-fire, fire-year, and post-fire conditions to quantify burn severity and recovery percentage. Hotspot (Gi*) analysis revealed spatial clusters of high-severity burn patterns, while lidar data provided insights into canopy loss and early regrowth patterns. Results indicate that vegetation has recovered to roughly 80–85% of pre-fire conditions by 2025, with the strongest regrowth occurring in low- and moderate-severity areas. High-severity zones show slower recovery, supported by both dNBR and lidar-derived canopy height. This project demonstrates how combining advanced GIS, remote sensing, and lidar techniques can provide a comprehensive assessment of post-fire landscape dynamics.
Dharm Barot, Embry-Riddle Aeronautical University
