We are seeking a highly motivated and technically driven Research Assistant (RA) to join a project focused on advancing forest disturbance mapping in Michigan. Forest disturbances are major drivers of ecosystem change, influencing carbon storage, forest structure, and long-term resilience. Accurately identifying and tracking these events through time is essential for sustainable forest management and climate modeling. 

The RA will contribute to the development of next-generation disturbance maps by integrating multi-temporal satellite imagery, geospatial embedding fields, and machine learning approaches.

•  Process and analyze large-scale, multi-temporal satellite datasets (e.g., Landsat, Sentinel-2) alongside advanced geospatial data products.
•  Create and manage robust training datasets to support environmental classification and mapping tasks.
•  Apply machine learning techniques to identify, classify, and analyze spatial patterns of forest disturbances.
•  Develop and streamline analytical workflows using Python and cloud-based geospatial platforms (such as Google Earth Engine).
•  Assist in evaluating model accuracy, interpreting results, and supporting the generation of high-resolution, regional-scale maps.

• Background in remote sensing, geospatial data science, environmental data science, or related fields, preferably with a Bachelor’s degree
• Experience working with Landsat and/or Sentinel-2 time series
• Programming experience in Python and Google Earth Engine
• A strong interest in machine learning, GeoAI, and foundation models
• Familiarity with classification algorithms (e.g., Random Forest, gradient boosting, neural networks)
• Strong analytical thinking and coding skills