Climate TRACE to track real-time global carbon emissions

This is a re-post from Yale Climate Connections by Bruce Lieberman

Combating climate change faces big challenges even beyond those that many people think about much. For instance, carbon emissions are not tracked as they happen, so policy makers and the public don’t know who’s polluting, when, and how much. That makes international agreements, like the Paris Climate agreement, tough to verify.

That could soon change. A new international initiative called Climate TRACE (“Tracking Real-time Atmospheric Carbon Emissions”) intends to independently detect emissions around the globe as they’re emitted. The initiative will use a combination of satellite image processing, machine learning, and sensors stationed worldwide.

Key spokesmen for the project include former Vice President and climate activist Al Gore and Gavin McCormick, executive director of coalition member WattTime. That’s a subsidiary of the Colorado-based nonprofit Rocky Mountain Institute, which developed technology so smart devices can draw electrical power from renewable energy power plants.

Doctors need tools. So do climate scientists.

“The Earth is like a medical patient suffering from a condition called climate change,” McCormick said in announcing the project. “Trying to fix it with only years-late, self-reported emissions data is like asking a doctor to fix a serious disease with no more information than a list of symptoms the patient had years ago.”

“They’ll do their best,” McCormick continued. “But there’s a reason hospitals use blood pressure monitors, stethoscopes – maybe an X-ray or MRI – to check what’s wrong with you right now. If we’re serious about stopping climate change, it’s time we gave climate ‘doctors’ the same kind of tools.”

A vision of ‘unprecedented transparency, accountability’

Climate TRACE says it has so far signed on founding members across three continents. Tech companies include Blue Sky Analytics, which analyzes satellite and ground sensor data to assess climate change risks around the world, water quality, and other environmental changes; and Hypervine, which provides construction and mining industries with satellite imagery data and other digital intelligence to help them monitor and measure their carbon emissions.

‘New era of transparency, accountability’

The idea for Climate TRACE originated about a year ago, when WattTime, UK-based Carbon Tracker, a financial think tank that analyzes the impact of the global shift toward renewable energy on capital markets, and other nonprofits teamed up to apply for’s AI Impact Challenge. The group proposed a project to monitor all global power plant emissions from space. awarded the group a $1.7 million grant and offered experts in data engineering and machine learning to help. As word spread about the project, more organizations have signed on.

Gore, who had been interested in finding ways to more reliably account for global carbon emissions as countries try to meet commitments under the Paris Climate agreement, also became interested. By tracking where carbon pollution is coming from, Gore in mid-July pointed to the globe’s entering “a new era of unprecedented transparency and accountability.” Climate TRACE will essentially create “a massively distributed body cam for the planet,” Gore told Time magazine. Explaining that point, Time climate reporter Justin Worland wrote: “In other words, if a given country claims to have reduced, say power-plant emissions, other countries will soon be able to immediately tap into Climate Trace and get data to verify the claim.”

‘Ambitious vision, novel approach’

The program, when fully in operation, will allow countries to verify the claims of others, wrote Time climate reporter Justin Worland. “Accurately monitoring the world’s emissions from afar would be a significant feat that could reshape many key points of debate among those working to fight climate change,” he wrote. “Countries would be able to verify that their counterparts are following through on emissions reductions commitments, governments could crack down on companies that are covering up their true footprint, and environmental groups could trace illegal forestry practices that are reducing forest cover and emitting carbon in the process.”

“I’m thrilled that a big, smart, and well-resourced coalition is taking this on with an ambitious vision and a novel approach,” says Taryn Fransen of the World Resources Institute.

Gore and McCormick explained in a post that Climate TRACE will use artificial intelligence to recognize human-generated carbon pollution in different kinds of imagery (including visible light and infrared light) gathered from remote sensing networks around the world. Those analyses will then be cross-checked with other data-sets to make sure they’re reliable. They said that measuring power plant emissions in real time, for example, will require:

– imagery from the European Space Agency’s Sentinel 2 Earth observation satellites and from other orbiting satellite networks
– artificial intelligence algorithms from experts in computer vision at companies such as Pixel and Scientia Labs
– data pipeline engineering at
– power plant databases maintained by the World Resources Institute
– remote sensing capabilities from Valence Strategies
– power systems modeling from WattTime, and
– many other collaborations.

A boon to investors for monitoring, verifying

Investors are expected to be among users and beneficiaries of this new intelligence on carbon emissions, given their growing role in holding corporations accountable for their carbon emissions, explained Matt Gray, managing director of the Carbon Tracker Initiative. Climate TRACE “offers a potentially powerful application for investors to monitor and verify emissions from those asset owners who are unwilling or unable to disclose timely and accurate data,” he said.

Tracking carbon emissions in real time – and publicizing it – can be a powerful incentive for change. And there is precedent for this: For more than four months, from late October 2015 to mid-February 2016, invisible methane gas – 100,000 metric tons of it – billowed from a well blowout at the Aliso Canyon Natural Gas Storage Facility north of Los Angeles. It was the worst methane leak in U.S. history, and it was invisible – until outside observers began to study levels of pollution at the site and film the billowing gas with infrared imagery. (The photo above illustrates how this might work in the case of invisible methane emissions, but the photo itself is not applicable specifically to what Climate TRACE is doing.)

The film was especially startling, showing a plume of methane gas – appearing black in the infrared imagery – shooting from the mountains above the blown well. Nearby residents were outraged, and the visually dramatic leak attracted widespread national media coverage for months. Once capped in February 2016, the well was plugged up for good.

Monitoring and tracking the Aliso Canyon methane leak gave regulators and the wider public hour-by-hour insights into what was happening. But such granular tracking of emissions today is the exception, and not the rule. The Keeling Curve, which traces the rise of carbon dioxide in the global atmosphere in the 20th and 21st centuries, offers an iconic global view of rising carbon emissions, but tracking where those emissions are coming from, as they enter the atmosphere, is another matter entirely.

Climate TRACE plans to offer its first accounting of global emissions ahead of the next COP26 climate talks in the fall of 2021, Gore has said. In the Time story, he added that he thinks real-time and near real-time data on carbon emissions will create “a new reality” in the effort to reduce carbon emissions and combat climate change.

Only time will tell whether that “new reality” advances progress on global, sustained action to reduce carbon emissions. But it undoubtedly has the potential, as they say, to be one more important tool in the toolbox.

Posted by Guest Author on Monday, 31 August, 2020

Creative Commons License The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.