Duke researchers provide building data contributing to clearest picture of global greenhouse emissions

With help from Duke’s Energy Data Analytics Lab, a new database is providing the clearest picture yet of global greenhouse gas emissions.

Dubbed Climate TRACE, the database maps human-produced emissions around the world to help policymakers craft more effective strategies for reducing global emissions.

According to its website, Climate TRACE was built by a coalition of over 100 universities, scientists and artificial intelligence experts. Researchers in Duke’s Energy Analytics Lab specifically contributed to mapping emissions from the buildings sector, which includes fuel combustion in residential, commercial and institutional buildings.

“Collectively, we’re hoping to make meaningful climate action faster and easier,” wrote Kyle Bradbury, director of the Energy Data Analytics Lab, in a March 7 email to The Chronicle.

Housed within the Nicholas Institute for Energy, Environment and Sustainability, the lab studies how remote data, such as satellite imagery, can support sustainable development and inform climate mitigation and adaptation strategies.

For Climate TRACE, the lab used a three-step process to create an emissions map for buildings. Researchers first identified the location of buildings around the world and sorted them by type, then gathered regional estimates of energy use intensity. They used that more detailed information to significantly enhance existing low-resolution data, including by accounting for seasonal variation in energy use.

Bradbury noted that existing EDGAR emissions data, which is provided by the European Commission, is reported in roughly 11-square-kilometer regions. By integrating the new satellite data, the Energy Data Analytics Lab was able to further sharpen estimates for Climate TRACE into one-kilometer regions.

The data gathered and analyzed by the Energy Data Analytics Lab for Climate TRACE is openly available to users in aggregates of 1-square-kilometer regions.

Bradbury noted that the sheer amount and diversity of buildings around the globe remains an obstacle. Currently, the database relies on estimations based on region and building type, such as whether they are commercial or residential. He also emphasized that the researchers continuously aim to “reduce the uncertainty of [their] estimates,” as well as “increase the frequency of updates” to improve the database’s accuracy.

“There are billions of buildings globally, each of which has its own occupants, energy uses and emissions,” Bradbury wrote. “It’s challenging to estimate emissions at that global scale while also maintaining a high enough spatial resolution and low enough latency for the data so that it can be effectively used for climate change mitigation and adaptation planning activities.”

The Climate TRACE website highlights key findings from its database, such as the underreporting of oil and gas emissions and the decline in deforestation-related emissions due to policy interventions. It also outlines numerous opportunities to reduce emissions further, alongside their projected quantitative impacts.

Bradbury added that assembling an emissions inventory is resource-intensive, which makes Climate TRACE especially valuable for communities lacking the ability to do so themselves, providing them with critical information to address greenhouse gas emissions and climate change.

“For our building team here at Duke, we’re continually working to improve all of the data we provide by gathering more information that we can use to reduce the uncertainty of our estimates, to increase the frequency of updates to the data and to continue to engage with the wider community around how these data can be made more impactful for end users,” he wrote.

---