In the coming years, I plan to advance the integration of geospatial data science (GDS), remote sensing, and machine learning to study the spatial and temporal dynamics of forest-based carbon systems. Working with a team of talented students, we are developing a novel, spatially explicit model of forest dependency that links forest carbon dynamics with supply chain trajectories across Canada, the U.S., and Brazil. This framework will incorporate life cycle assessment to evaluate the carbon footprint of timber production while accounting for ecological, climatic, and social risks. In parallel, we are designing a high-resolution biomass monitoring system for Michigan, combining time-series analysis with scenario-based simulations. To enhance accessibility and interactivity, we are also building an AI-driven tool using deep learning and large language models with conversational capabilities—enabling users to explore forest data and future projections through intuitive dialogue.
Together with colleagues and talented students, I am developing a research agenda that investigates the spatial impacts of the global transition to decarbonization. Our focus is on advancing geospatial data science techniques to model how emerging demands—such as those for critical minerals and renewable technologies—reshape land systems and infrastructure across scales. Using machine learning, remote sensing, and multi-scalar modeling, we aim to trace and predict complex spatial linkages between regions driving demand and those absorbing its environmental costs. This work seeks to push the boundaries of geocomputational approaches by integrating causal inference, automated detection, and large-scale spatial analysis to better understand the global footprints of decarbonization.