Researching how AI understands the spatial world

My research explores how AI, geospatial data, and Earth observation can be combined to better understand complex spatial and environmental systems; and how those methods can move beyond experiments into practical geospatial workflows. I also collaborate with universities across Canada on applied research projects spanning GeoAI, environmental monitoring, spatial analytics, and emerging AI methods.

View Photos from my Arctic Expeditions

Research Directions

  • Geospatial AI Systems: I study and develop ways of integrating modern AI methods into geospatial workflows. This includes geospatial foundation models, computer vision, deep learning for imagery, LLM-powered systems, retrieval-augmented generation, and agentic AI. A central question is how these technologies move from promising models into reliable systems that GIS professionals, researchers, and organizations can actually use.
  • Earth Observation & Environmental AI: A second strand of my research focuses on extracting useful information from satellite, aerial, and other spatial observations. My work has included land-cover analysis, carbon monitoring, change detection, remote sensing time series, environmental modelling, and machine learning approaches for interpreting large Earth observation datasets. This work connects advances in GeoAI with practical environmental monitoring and decision support.
  • Ocean & Climate Systems: My foundational research focused on Arctic marine carbon cycling and the role of Arctic waters in the global carbon system. Using field observations, satellite data, spatial analysis, and environmental modelling, I studied air-sea CO₂ exchange and variability of carbon fluxes across the Canadian Arctic Archipelago and Hudson Bay. That scientific background continues to shape my approach to AI: complex environmental systems rarely reduce to a single model, dataset, or metric.