Study: Climate TRACE undercounts urban vehicle emissions
A Northern Arizona University climatologist says the Climate TRACE database, co-founded by Al Gore, underestimates carbon dioxide from cars and trucks in cities by about 70 percent on average.
- A study claims the Climate TRACE database underestimates urban vehicle emissions by about 70 percent on average.
- The analysis covered cars and trucks across 260 cities in the United States.
- In some cities the gap exceeded 90 percent.
- A co-founder of the platform disputes the findings, saying its own comparisons with official city data did not show the same result.
Professor Kevin Gurney, a climatologist at Northern Arizona University, has published a study claiming that the global emissions database Climate TRACE underestimates carbon dioxide from urban road traffic. The results appeared in the peer-reviewed journal Environmental Research Letters, and the analysis covered emissions from cars and trucks across 260 cities in the United States. According to the study, the undercount averages about 70 percent, while in some cities the gap exceeds 90 percent.
Climate TRACE is a platform that combines satellite imagery and artificial intelligence to estimate emissions from a range of sources, and its co-founders include former US Vice President Al Gore. The significance of the finding is not merely a technical detail. Governments, city administrations and researchers increasingly rely on databases like this when shaping climate policy, setting emission-reduction targets and measuring progress. If the input data are systematically undercounted by large margins, leaning on them could distort priorities, especially where traffic is one of the largest urban sources of carbon dioxide.
What does the study claim, and what does it not prove?
It is important to separate what the paper shows from what cannot be concluded from it. Gurney’s analysis compares Climate TRACE estimates with other ways of estimating vehicle emissions in US cities and records a systematic difference in one direction, toward undercounting. That is a finding about one sector, road traffic, in one country. The study does not claim that the entire database is wrong for all sectors or all regions of the world, nor that global emission estimates are unusable. It concerns a specific category of sources in a specific sample of cities.
A single peer-reviewed study is also the start of a debate, not its conclusion. Every emissions estimate carries assumptions, about traffic density, vehicle types, fuel consumption and how satellite signals are translated into tonnes of carbon dioxide. Differences between two approaches can stem from a genuine error in one model, but also from differing definitions of what is being measured. Replication on other datasets and in other countries will show how robust the finding is.
Why is the platform’s response part of the story?
Climate TRACE co-founder Gavin McCormick disputes the findings. He says the comparisons the platform ran against official city data did not produce the same results the study describes. This disagreement is not a side note, it is the heart of the scientific process. When two sides differ over the size of an error, the next step is a transparent comparison of methods, input data and assumptions, ideally with independent verification.
For readers who follow climate topics, the lesson is neither that satellite estimates are worthless nor that they are flawless. They are tools with known limits that improve over time. A debate like this is, in fact, a sign that the system is working, because estimates are being exposed to criticism and tested. The open question remains: if even the most advanced emissions databases rest on assumptions that need constant checking, how much caution do we owe the numbers we use to make long-term decisions about the cities we live in?
