Progress report advances small-data machine learning to predict soil organic carbon using satellite, management, and environment data.

This FY24 BIOAg progress report describes a seed grant developing a machine learning approach to estimate soil organic carbon (SOC) in the top 0–30 cm for carbon incentive programs that aim to reward measured benefits rather than practice adoption. The team is assembling SOC response data from the Washington Soil Health Initiative and integrating multispectral satellite imagery (Sentinel-1, Sentinel-2, Landsat) with environmental and management covariates, including climate variables from gridMET and crop information from WSDA and USDA Crop Data Layers. The unit of analysis will follow the 40-hectare site definition used in the State of the Soils Assessment. Work to date includes recruiting two graduate students, setting up weekly meetings, and progressing on data collation and early model development, with a target paper submission in January 2025. A social dimension component plans a hands-on workshop on remote sensing and data science with partners at Heritage University and Wenatchee Valley College.
This publication is part of an archive and may not meet current digital accessibility standards. CSANR is working to improve digital accessibility of all materials. If you need this content in an alternative format, please contact csanr@wsu.edu.
Authors
Doppa, J., Griffin LaHue, D., Gelardi, D., Gharsallaoui, M., Kandelati, A., Rajagopalan, K., and Jobe, J.
Related Product
Related Project
Year Published
2024
Areas of Focus
Agricultural Technology, Climate & Environment, and Research Engagement & Communication
Topics
Climate Change, Community Engaged Research, Production Systems, and Soils & Fertility
Collaborator
Funding Source