This research focuses on developing data-driven modeling frameworks and decision support tools to improve water quality, resource management, and sustainability in agricultural systems. We integrate field observations, geospatial data, and process-based models to better understand how management practices influence environmental and economic outcomes. A central component of this work involves watershed-scale modeling and analysis of agricultural landscapes, where we compile and harmonize diverse datasets (e.g., land use, soils, weather, and management practices) to support reliable simulation and assessment. These modeling efforts provide a foundation for evaluating best management practices (BMPs) and their impacts on water quality and system performance. We also emphasize strong connections between research and real-world application. Through collaboration with ranchers, extension agents, and stakeholders, our work is designed to support practical decision-making in working agricultural systems. Current efforts include the development of decision support systems (DSS) that integrate modeling outputs with field-based information to guide management strategies. In addition, our research explores emerging approaches for pasture and biomass assessment, combining field measurements with data-driven and imagery-based methods to support improved grazing management decisions. By linking modeling, field data, and stakeholder input, our goal is to provide actionable tools that enhance both environmental sustainability and agricultural productivity.Our lab advances the use of artificial intelligence and machine learning to improve environmental prediction, mapping, and process understanding. We integrate AI methods with traditional hydrologic knowledge to enhance model performance, interpretability, and scalability. Applications include: * Pattern recognition from large geospatial datasets * Hybrid modeling frameworks combining physics and data-driven approaches * Flood and environmental risk mapping A strong emphasis is placed on explainable and responsible AI, ensuring transparency and usability for scientific and applied decision-making. Our lab advances the use of artificial intelligence and machine learning to improve environmental prediction, mapping, and process understanding. We integrate AI methods with traditional hydrologic knowledge to enhance model performance, interpretability, and scalability. Applications include: * Pattern recognition from large geospatial datasets * Hybrid modeling frameworks combining physics and data-driven approaches * Flood and environmental risk mapping A strong emphasis is placed on explainable and responsible AI, ensuring transparency and usability for scientific and applied decision-making.

This research focuses on developing AI-driven geospatial frameworks to assess environmental risks and support resilient decision-making in complex natural systems. This work integrates machine learning, deep learning, and Earth observation data to map and understand spatial vulnerability across diverse environmental domains. A central component of this research is the development of flood susceptibility and risk mapping frameworks, where advanced deep learning models are used to capture nonlinear interactions among hydrologic, environmental, and land-surface factors. These models are coupled with explainable AI techniques to ensure that predictions are not only accurate but also interpretable and physically meaningful. Beyond flooding, this framework is extended to other environmental systems, including: Groundwater vulnerability assessment Drought susceptibility and water stress mapping Multi-hazard environmental risk analysis A key innovation of this work is the integration of Earth Observation (EO) data to bridge the gap between modeled risk and observed environmental dynamics. By linking predictive models with satellite-derived indicators such as surface water persistence, this research provides a more comprehensive understanding of how environmental risks manifest in space and time. Overall, this research advances a scalable and transferable approach for spatial risk assessment, supporting applications in water resources management, land-use planning, climate adaptation, and disaster resilience — particularly in vulnerable regions such as Florida.Our lab advances the use of artificial intelligence and machine learning to improve environmental prediction, mapping, and process understanding. We integrate AI methods with traditional hydrologic knowledge to enhance model performance, interpretability, and scalability. Applications include: * Pattern recognition from large geospatial datasets * Hybrid modeling frameworks combining physics and data-driven approaches * Flood and environmental risk mapping A strong emphasis is placed on explainable and responsible AI, ensuring transparency and usability for scientific and applied decision-making. Our lab advances the use of artificial intelligence and machine learning to improve environmental prediction, mapping, and process understanding. We integrate AI methods with traditional hydrologic knowledge to enhance model performance, interpretability, and scalability. Applications include: * Pattern recognition from large geospatial datasets * Hybrid modeling frameworks combining physics and data-driven approaches * Flood and environmental risk mapping A strong emphasis is placed on explainable and responsible AI, ensuring transparency and usability for scientific and applied decision-making.

We develop and apply physically based hydrologic and water quality models to quantify water flow, storage, and nutrient dynamics across agricultural landscapes, rangelands, wetlands, and watersheds. Our work focuses on linking field-scale processes to watershed-scale outcomes, with particular attention to nutrient transport, runoff generation, and hydrologic connectivity. These models support evaluation of best management practices (BMPs), assessment of environmental risk, and scenario-based analysis under changing climate, land use, and management conditions.Our lab advances the use of artificial intelligence and machine learning to improve environmental prediction, mapping, and process understanding. We integrate AI methods with traditional hydrologic knowledge to enhance model performance, interpretability, and scalability. Applications include: * Pattern recognition from large geospatial datasets * Hybrid modeling frameworks combining physics and data-driven approaches * Flood and environmental risk mapping A strong emphasis is placed on explainable and responsible AI, ensuring transparency and usability for scientific and applied decision-making. Our lab advances the use of artificial intelligence and machine learning to improve environmental prediction, mapping, and process understanding. We integrate AI methods with traditional hydrologic knowledge to enhance model performance, interpretability, and scalability. Applications include: * Pattern recognition from large geospatial datasets * Hybrid modeling frameworks combining physics and data-driven approaches * Flood and environmental risk mapping A strong emphasis is placed on explainable and responsible AI, ensuring transparency and usability for scientific and applied decision-making.