Projects
Active Research Projects
Machine learning-based flood susceptibility and risk assessment:
The goal of this project is to map flood susceptibility and assess compound flood risk in the Big and Little Sandy and Tygart Creek Basin of Eastern Kentucky, a mountainous, mining-impacted region with a long history of destructive flooding. The study integrates historical flood records, high-resolution LiDAR terrain data, land cover, soil, and climate data to train and compare multiple machine learning models, combined through an ensemble approach. Model outputs are validated against FEMA's National Flood Hazard Layer, satellite imagery, and GPS-based field observations. The study also examines how social vulnerability and surface mining disturbance relate to flood-prone areas. Findings will help identify high-priority communities and support flood risk management, land-use planning, and mine reclamation decisions in Appalachia.
Soil and water quality research:
The goal of this project is to establish a benchmark database of soil and water quality parameters for long-term monitoring of reclaimed mine sites in Eastern Kentucky. It integrates satellite, airborne, and in-situ data to evaluate soil and water quality variations across disturbed and control agroecosystems. The study utilizes multispectral and LiDAR imagery, weather and climate data, and local soil sampling to assess key parameters such as nutrient content, soil texture, and water characteristics. Findings will support the development of a comprehensive geospatial database for assessing ecosystem recovery and guiding sustainable land management practices.
Aerial monitoring of vegetation health dynamics:
This project integrates satellite imageries, multispectral and hyperspectral drone imageries, and LiDAR data to monitor vegetation health status in reclaimed mine sites in Eastern Kentucky. Field-based observations are time-consuming, labor-intensive, and costly. Using aerial monitoring is necessary for large-scale research on vegetation dynamics. The goal is to monitor landscape dynamics and guide post-mining land use.
Soil loss and vegetation growth assessment:
This study focuses on estimating soil loss using the Universal Soil Loss Equation (USLE) model and the SWAT model and assessing reclamation effort through vegetation recovery using LiDAR and field-based observation. The goal is to provide a comprehensive assessment of current reclamation practices and their effectiveness in mitigating soil erosion and facilitating vegetation growth in reclaimed mine sites.
Microclimate observations and monitoring
With reclamation activities actively ongoing over mining regions in eastern Kentucky, restoring these sites to their pre-mined status takes several years. However, there is a need to monitor how these activities improve the ecosystem in real-time. The observation and monitoring of microclimates such as CO₂ and water vapor exchange, precipitation, soil moisture, radiation, wind, and humidity will enable long-term studies of CO₂ exchange characterization, regrowth phenology, biomass productivity, water, vegetation, and carbon balance, and remotely sensed data validation.
Perceptions, practices, and preparedness for extreme weather events: Insights from communities in Kentucky
This study will examine how individuals and communities perceive, respond to, and prepare for extreme weather events such as droughts, floods, and temperature fluctuations. Kentucky serves as a key case study due to its growing vulnerability to these events, which have disrupted agriculture, soil health, and water management. The findings aim to guide extension services and policymakers in designing locally relevant, inclusive, and effective climate adaptation strategies. Ultimately, the research seeks to bridge the gap between scientific forecasts and public understanding, highlighting the need for better science communication and proactive, community-based responses to extreme weather risks.