NAU Study Finds Climate TRACE Misses 70% of City Vehicle CO2 Emissions

TL;DR: A Northern Arizona University analysis shows Climate TRACE’s city vehicle CO2 estimates are roughly 70% lower than independent calculations, exceeding 90% shortfall in Indianapolis and Nashville.

Context Climate TRACE, a global emissions database co‑founded by former Vice President Al Gore, uses artificial‑intelligence models to map greenhouse‑gas sources. Accurate emissions data guide policy, investment and public understanding of climate risk.

Key Facts Researchers from NAU’s School of Informatics, Computing, and Cyber Systems compared Climate TRACE’s U.S. urban vehicle emissions with the Vulcan onroad database, which aligns traffic counts and fuel use with official statistics. Vulcan’s uncertainty sits near 14%, far tighter than the gap the team uncovered. Across 260 U.S. cities, Climate TRACE’s vehicle CO2 numbers were, on average, 70% lower than Vulcan’s. In Indianapolis and Nashville, the discrepancy exceeded 90%, meaning Climate TRACE recorded less than one‑tenth of the emissions that Vulcan calculated for those locales. The study, published in *Environmental Research Letters*, builds on earlier NAU work that identified similar undercounts for power‑plant emissions in the same database.

What It Means If Climate TRACE’s urban vehicle figures are systematically low, policymakers may underestimate the climate impact of transportation and misallocate resources for mitigation. The authors caution that the bias likely extends beyond the United States, raising questions about the global reliability of AI‑driven emissions inventories. They recommend tighter validation against ground‑truth data, transparent modeling assumptions and independent expert review before the data inform climate strategies. While AI can accelerate emissions tracking, scientific rigor remains essential to maintain trust and effectiveness.

Looking Ahead Future work will test Climate TRACE’s estimates against additional on‑the‑ground measurements and explore adjustments for the identified shortfall. Monitoring how the consortium responds will be critical for anyone relying on its data for climate policy or investment decisions.