Most corporate carbon footprints rest on a small set of shared public datasets. Two spend-based databases, USEEIO and CEDA, are used to calculate roughly 65% of corporate scope 3 measurements globally. eGRID is used by 85% of U.S. companies that report their scope 2 emission factor sources to CDP. When a federal dataset stops updating on schedule, as eGRID's 2024 release did, every company that depends on it is affected.
Watershed's research on open data keeps these foundations current, free, and reproducible. Through the Cornerstone Sustainability Data Initiative with Stanford and ERG, our scientists maintain USEEIO and Open CEDA, republish federal datasets, and document the methods behind them. We also release open benchmarks so anyone can test how well AI performs carbon accounting tasks. You can see a summary of related research and publications below. We'll add to this brief as new research becomes available.
Open data
Open CEDA Watershed, May 2025. A free version of CEDA, Watershed's global spend-based emissions database. It covers 400 industries in 148 countries, about 60,000 emission factors, and 95% of global GDP. It is licensed CC BY-SA 4.0 and updated annually. Developed by Watershed's climate and data science team, building on research Sangwon Suh began at Leiden University.
USEEIO: USEEIO is the United States Environmentally Extended Input/Output Family of Models that underpins the supply chain emissions factors originally developed by the US EPA and maintained by Cornerstone moving forward.This dataset covers 411 sectors for the United States.
eGRID: The Emissions and Generation Resource Integrated Database (eGrid) is the most up-to-date, regionally specific, and comprehensive view of electricity generation in the United States. The last version of eGrid (2023) was released in January 2025 by the US EPA. The Cornerstone data initiative released eGrid (2024) and will be continuing to update it moving forward.
Emissions Factor Hub: The EF Hub consolidates key United States data sources, including eGrid and the Waste Reduction Model (WARM), into a single easy-to-use set of emissions factors. The last version of the EF Hub was released in January 2025 by the US EPA. The Cornerstone data initiative will continue maintenance and development moving forward.
Land Use Change (LUC) A new methodology and data pipeline for estimating land use change related emissions of agricultural commodities. It is primarily intended for use in corporate GHG inventories following the new GHGP land sector and removal standard. This is maintained by the Cornerstone Data Initiative.
PCFBench: A diagnostic benchmark for product carbon footprint estimation arXiv preprint, August 2026. An open dataset for testing how well AI models build product carbon footprints, step by step. It includes 614 expert-labeled items across six tasks, from listing a product's materials to estimating its total emissions. The dataset and evaluation code are free to use. Written by Watershed's science team.
Methodology: How open models are built and maintained
Methodology for Cornerstone U.S. Model v0.2 Cornerstone working paper, GitHub, 2026. Documents how the U.S. part of Cornerstone's merged global model is built, step by step. It draws on both USEEIO and CEDA, targets 2023 emissions, and uses EPA's national greenhouse gas inventory. Every design choice links to a public discussion. Co-authored with Stanford and ERG.
Cornerstone: Technical architecture vision (v1) Cornerstone, Zenodo, 2026. Sets out the technical plan for Cornerstone's open data infrastructure. Co-authored with Stanford and ERG.
Scaling climate data like software Watershed blog, December 2025. Explains how the team applied software engineering practices to CEDA's pipeline, which combines more than 20 national and international datasets. Tests, versioning, and a staging environment halved the annual update effort. Written by Watershed's CEDA team.
Trends in the data
The data in this collection contains insights into the field of sustainability and our progress on climate change.
Reach: A few datasets underpin most footprints
- Two spend-based databases support most scope 3 measurement. USEEIO and CEDA are used to calculate roughly 65% of corporate scope 3 measurements globally. (Watershed, 2025b)
- U.S. scope 1 and 2 reporting leans on two EPA-origin datasets. Roughly 80% of American companies use the GHG Emission Factors Hub for scope 1. Of U.S. companies that report scope 2 factor sources to CDP, 85% use eGRID. (Nolan et al., 2026; Ingwersen, Morgan & Maldonado, 2026)
- Open CEDA covers nearly the whole global economy. It spans 400 industries in 148 countries, representing 95% of global GDP. (Watershed, 2025a)
Trends: What the updated data shows
- Emissions per dollar of spend fell about 2% in the latest update. Applied to a test group of companies, the CEDA update cut median footprints by 2.0% and USEEIO by 1.8%, after inflation. Paris-aligned targets typically require at least 6% a year. (Watershed, 2025b)
- Cleaner electricity drove the largest share of that change. Electricity accounted for 31% of the change in the models, and oil and gas production for 7%. (Watershed, 2025b)
- U.S. grid emission factors fell an average of 3% from 2023 to 2024. Changes varied by region, from a 13% increase in Upstate New York to a 10% decrease in SPP North. (Ingwersen, Morgan & Maldonado, 2026)
Note: Database trends and grid factors are measured from public data. Figures come from different datasets and methods and aren't comparable across sources.
What this means for sustainability teams
These datasets are accessible to all and updated regularly. Watershed maintains Open CEDA and USEEIO through the Cornerstone Sustainability Data Initiative, with Stanford's Sustainable Solutions Lab and ERG. Each is refreshed annually using the most recent complete year of data. Cornerstone also publishes eGRID and the GHG Emission Factors Hub, so U.S. scope 1 and 2 factors stay current.
New versions can change your footprint. The latest USEEIO and CEDA updates lowered median spend-based footprints by about 2%. When you adopt a new version, recalculate prior years on the same version. That separates data updates from reductions your team achieved.
Sign up for updates. To hear when new versions and datasets are released, sign up for Cornerstone updates. On the Cornerstone site you'll also find contact information where you can send questions.
About the researchers and project leads
Dr. Mo Li: Climate Science Lead at Watershed. She leads CEDA and works on the Cornerstone Sustainability Data Initiative, an open collaboration with Stanford and ERG. She earlier helped develop open-source data infrastructure for the U.S. EPA's USEEIO modeling framework. She holds a Ph.D. and M.S. in Natural Resources Science and Management from the University of Minnesota.
Dr. Steve Davis: Professor of Earth System Science in the Stanford Doerr School of Sustainability and leads the Sustainable Solutions Lab, a research group dedicated to quantifying how different human activities are affecting climate and air quality, how those environmental changes in turn jeopardize human wellbeing, and the relative priority of solutions.
Sarah Cashman: ERG vice presiden and director of ERG’s life cycle assessment team. She also specializes in applying LCA for indirect greenhouse gas emissions accounting, including scope 3 and embodied carbon accounting.
Dr. Wesley Ingwersen: Research Engineer at Stanford University's Sustainable Solutions Lab and the Technical Director of the Cornerstone Sustainability Data Initiative.
Yohanna Maldonado: Director of the Cornerstone Sustainability Data Initiative and Head of Climate Data at Watershed.
Ben Young: Vice president and senior life cycle analyst specializing in developing life cycle inventories for energy systems and performing input-output life cycle assessment at Eastern Research Group (ERG).
