The world’s pollution will finally be tracked in real time. That’s huge.
Like Skynet, but good! | Shutterstock The new Climate TRACE Coalition is assembling the data and running the AI. There’s an old truism in the business world: what gets measured gets managed. One of the challenges in managing the greenhouse gas emissions warming the atmosphere is that they aren’t measured very well. “Currently, most countries do not know where most of their emissions come from,” says Kelly Sims Gallagher, a professor of energy and environmental policy at Tufts University’s Fletcher School. “Even in advanced economies like the United States, emissions are estimated for many sectors.” Without this information “you cannot devise smart and effective policies to mitigate emissions,” she says, and “you cannot track them to see if you are making progress against your goals.” The lack of good data also complicates international climate negotiations. “It’s frustrating that nearly three decades after countries committed under the United Nations Framework Convention on Climate Change (UNFCCC) to publish national GHG emissions inventories, we still don’t have recent, comprehensive, and consistent inventories for all countries,” says Taryn Fransen of the World Resources Institute. The lack of reliable data leads to endless time spent haggling over monitoring, reporting, and verification, and a persistent background level of mutual suspicion. In late 2015, just before the 2016 international climate negotiations in Morocco, China revealed that it had underestimated the amount of coal it burned by 17 percent — one billion tons of additional greenhouse gas emissions, equal to the emissions of Germany’s entire energy sector. That kind of thing does not build confidence. OWID Getting a bit out of hand there at the end. The ultimate solution to this problem — the killer app, as it were — would be real-time tracking of all global greenhouse gases, verified by objective third parties, and available for free to the public. When countries began meeting under the UNFCCC in the mid-1990s, that vision was speculative science fiction. It was basically regarded as science fiction when the Paris Agreement was signed in 2015. But science moves quickly — in particular, artificial intelligence, the ability to rapidly integrate multiple data sources, has advanced rapidly in recent years. Now, a new alliance of climate research groups called the Climate TRACE (Tracking Real-Time Atmospheric Carbon Emissions) Coalition has launched an effort to make the vision a reality, and they’re aiming to have it ready for COP26, the climate meetings in Glasgow, Scotland, in November2021 (postponed from November 2020). If they pull it off, it could completely change the tenor and direction of international climate talks. It could also make it easier for the hundreds of companies, cities, counties, and states that have made ambitious climate commitments to reliably track their process. “Trying to build up such an infrastructure the old, bottom-up way won’t get us where we need to be in time,” says Fransen, “so I’m thrilled that a big, smart, and well-resourced coalition is taking this on with an ambitious vision and a novel approach.” Let’s have a closer look at that approach and then consider its implications for climate politics. Shutterstock It knows when you’ve been sleeping ... How tracking power plant emissions became tracking all emissions The story begins last year, when a group of nonprofits, including US-based WattTime and UK-based Carbon Tracker, applied for and won a $1.7 million grant from Google.org (Google’s philanthropic arm) to track global power plant emissions in real time using satellite data and AI algorithms. Google.org likes it so much they sent a team to help with the project. In May 2019, I wrote a story on the effort and its implications. “The next day our phone started ringing,” says WattTime founder and executive director Gavin McCormick. (WattTime is now a part of the clean energy nonprofit Rocky Mountain Institute.) Other research organizations told him, “this is really cool that you can do this in the power sector, but climate change is a lot bigger than the power sector,” McCormick says. They began offering up data helpful to tracking other sources of greenhouse gas emissions, with an eye toward eventually covering all of them. “One of my favorite examples is OceanMind,” he says. “They had been in the business of using entirely different types of sensors [the Automatic Identification System, a global network of onboard transponders] to detect illegal fishing. They were able to track every ship in the world in real time.” Based on ships’ movement and engine specifications, their GHGs can now be monitored in real time. One of the early phone calls came from Al Gore, who had been looking for more reliable ways to track emissions and also saw the Vox article. He and McCormick began collaborating and reaching out to other groups. The reception was better than they had hoped. “Every organization that I’ve approached has been enthusiastic,” Gore says. “So far, the response has been incredible.” “I’m trained as an economist, so I’m used to being real cynical,” says McCormick, “but over the last year, we found companies saying, ‘we just think this is important — we can’t keep [our data] siloed any longer.’” These various data sources and trackers have come together as Climate TRACE. The coalition is adding other partners and covering more and more emissions. The eventual result will be comprehensive, reliable, publicly accessible global emissions data, accompanied by periodic reports. Proud to introduce @ClimateTRACE! A new coalition using AI, satellites, machine learning, & other tech to build the world’s first tool to trace all human-caused pollution to the source in real time. Here’s my @Medium post w/ Gavin McCormick from @WattTime.https://t.co/RTmSsJ8kg1— Al Gore (@algore) July 15, 2020 The magic sauce: satellite data + other data sources + artificial intelligence In addition to Gore and WattTime, the coalition now contains: Carbon Tracker uses machine learning and satellite data to predict the utilization of every power plant in the world; Earthrise Alliance aggregates and organizes publicly available environmental data into a format meaningful to journalists and researchers; Carbon Plan uses satellite data to track changes in aboveground biomass (especially forests) and the associated carbon emissions, down to a spatial resolution of 300 meters; Hudson Carbon uses satellite data to track changes in agricultural cover, cropping, and tilling, down to the level of the individual field, and compares that data against ground-level sensors (“They can tell you the inherent productivity of every field on every farm on the entire planet,” marvels Gore); OceanMind uses onboard sensors to track the global movement of ships in real time and combines that with engine specs to extrapolate carbon emissions; Rocky Mountain Institute combines multiple sources of data to quantify methane emissions from oil and gas infrastructure; Hypervine uses spectroscopic imagery to track vehicle usage and blasting at quarries, Blue Sky Analytics uses near-infrared and shortwave infrared imagery from satellites to track fires. The coalition will also be gathering data from a variety of other sources, from power grid data to fuel sales, sensor networks, and drones. “The trick has been combining different data sources,” says McCormick. All the data will be fed into the AI, which for now is a basic prototype. It will improve over time as the algorithms become more sophisticated and more sources come online. Fransen cautions that the satellite data available today is not a “silver bullet for all the holes in our GHG data.” She points to “limitations in terms of the resolution and frequency of emissions-monitoring satellites that are currently operating,” though that will improve as governments continue to invest in the technology. She also notes that “some of the low-hanging fruit that these novel approaches can help with are the same sources for which we already have relatively robust estimates from conventional sources,” like energy-related emissions. She is encouraged that TRACE is also looking to “activity data,” i.e., “data on activities that generate emissions, like driving, farming, etc.” Photo by Scott Olson/Getty Images John Duffy, planting corn on a farm he farms with his father, on April 23, 2020, near Dwight, Illinois. Gore acknowledges that “this is a work in progress,” but says the coalition is aiming big: “everything that can be known about where greenhouse gas emissions are coming from will be known, in near-real time.” The politics of radical emissions transparency If the TRACE coalition is successfully able to gather comprehensive data, build the AI to make sense of it, and make its results accessible to the public (which for now remains an open question), it will be like flipping on a light switch in a room that has been dimly lit with candles. More or less overnight, the world’s real-time flow of greenhouse gas emissions, previously estimated, will become visible. Gallagher’s first reaction to TRACE’s mission was, “Wow!” “Such an initiative is long past due,” she says, “and it is absolutely essential to improving our management of greenhouse gas emissions around the world.” Among other things, she says, “having real-time data will help parties to the UNFCCC better understand where and why they are falling short, and conversely, where they are doing better than expected and might be able to increase ambition in their next round of [nationally determined contributions].” Most international disputes over monitoring, reporting, and verification would effectively be mooted. “A system where everyone can see the same numbers, those numbers are grounded in real physics, and everyone can know that everyone else is telling the truth,” McCormick says, “may have a significant effect on the tone, taking a little of the negativity out of some of those negotiations.” What began as an initiative to measure emissions from the world's power plants has grown into something bigger than we could have imagined. We're thrilled to unveil @climatetrace. Soon GHG emissions tracking for every sector and country will be a reality. https://t.co/vuhMjvn7Rh— WattTime (@wattTime) July 15, 2020 It’s not just that the system would help identify errors and scofflaws. “It will empower the people who really are interested in reducing their emissions,” Gore says. Voters can hold politicians accountable for their climate promises. Workers, investors, and regulators can hold companies accountable. Countries can hold one another accountable. Transparency will bring accountability, as long as the coalition can maintain the trust of all the parties involved. “It is extremely important for this effort to be independent and reliable,” Gore says, “and for it to constantly improve.” What gets measured gets managed. If the Climate TRACE coalition accomplishes what it has set out to do, global greenhouse gases will finally be measured — and there will be no excuses left not to manage them. Support Vox’s explanatory journalism Every day at Vox, we aim to answer your most important questions and provide you, and our audience around the world, with information that has the power to save lives. Our mission has never been more vital than it is in this moment: to empower you through understanding. Vox’s work is reaching more people than ever, but our distinctive brand of explanatory journalism takes resources — particularly during a pandemic and an economic downturn. Your financial contribution will not constitute a donation, but it will enable our staff to continue to offer free articles, videos, and podcasts at the quality and volume that this moment requires. Please consider making a contribution to Vox today.