Current Research Themes

Our research is centered on developing innovative methods and technologies to address key challenges in the fields of agriculture, water resources, and human system integration:

• Continental Scale Crop Land Cover Land Use Classification (CLCLU): We are developing new techniques for large-scale classification of crop land cover and land use, which is crucial for understanding agricultural patterns, managing resources, and supporting food security. This project, pronounced as "clu"... short for CLCLU, leverages advanced deep learning algorithms and high-resolution satellite data to accurately classify and monitor diverse crop types and land uses across vast geographic areas.
• Developing Global Soil Carbon Models: We are working on advancing global carbon models to better understand the soil carbon and its impact on climate change. This research focuses on integrating data from various sources, such as satellite observations and terrestrial measurements to create comprehensive models of carbon stocks globally. Our aim is to improve estimates of current and attainable soil carbon stocks and sequestration potential.
• Hyperspectral Imaging for Crop Stress: This research theme focuses on using hyperspectral imaging to detect early signs of crop stress due to factors like drought, pests, diseases, or nutrient deficiencies. By capturing a broad spectrum of light beyond the visible range, hyperspectral sensors can identify subtle changes in plant physiology that are invisible to the naked eye. The insights gained from this imaging technique can help farmers and decision-makers take timely actions to mitigate crop loss and improve agricultural productivity. Hopefully, this will lead to better guidance for future satellite missions too!
• AI Models for Streamflow Predictions: We are developing more robust and accurate artificial intelligence models for predicting streamflow, which is essential for managing water resources, flood forecasting, and drought preparedness. Our models use state-of-the-art machine learning techniques to analyze a wide range of hydrological, meteorological, and geospatial data, providing improved predictions of river and stream behaviors under varying environmental conditions. We are looking for ways to extend this to water quality too, and integrated with existing models.
• Assimilation of Aerial and SpaceBorne Earth Observations: This theme aims to enhance water and crop system models by integrating data from aerial and satellite-based Earth observations (EO). By assimilating EO data, such as soil moisture, evapotranspiration, vegetation indices, and surface temperature, into predictive models, we improve the accuracy and reliability of forecasts related to crop growth, water usage, salinity, and ecosystem health. This approach supports more informed decision-making in agriculture and natural resource management.
• Identifying Stakeholder Mental Models to Improve System Modeling: Our research investigates how to better understand stakeholder mental models—individuals' and groups' perceptions, beliefs, and assumptions about complex systems. We are working on developing a new method to get By identifying and integrating this valuable information into our models, we aim to create more accurate and representative system models that reflect the perspectives and priorities of diverse stakeholders. This approach enhances the relevance and applicability of our models to real-world decision-making.