Satellites are our eyes in the sky or, er, in space. Now, a coalition of groups is turning to satellites and AI technology to help monitor emissions from all the world’s power plants.
This week, Google announced it had awarded $US1.7 million ($2.4 million) to the team, which include WattTime, a nonprofit that develops tech to help decrease dirty energy use; Carbon Tracker, a think tank that focuses on the financial part of transitioning away from fossil fuels; and the World Resources Institute (WRI), which focuses on how the world can live sustainably.
They all want to learn more about the greenhouse gases and other pollutants that pour out of the world’s power plants, with the aim of holding companies and polluters accountable. They’ll be relying on satellites currently up in space to gather data for this endeavour.
Google awarded this grant as part of its Google AI Impact Challenge, which gave out $US25 million ($36 million) to organisations around the world. All applicants had to figure out how they would use artificial intelligence to help solve some world problem. In this case, the issue is both pollution and climate change.
While the US and many countries throughout Europe have stringent regulations that compel industries to monitor emissions from power plants, the same isn’t true for countries such as Vietnam or Indonesia.
This program, which relies on the global network of satellites to work in tandem with an AI-powered algorithm that can spot emissions, aims to keep tabs on every single power plant in the world. However, the program will prioritise those power plants that currently fly beneath the radar.
“The idea is to expand this to have global coverage and every power plant,” Durand D’souza, a data scientist with Carbon Tracker’s Power and Utilities Team, told Earther.
Carbon Tracker has already proven the effectiveness of this sort of satellite spying, using it to estimate the use of coal plants in China year. As far as the organisation knows, it was the first to use this technology to play Big Brother on the coal industry. So WattTime and WRI approached the team at Carbon Tracker about expanding this methodology to every single power plant in the world.
They hope to eventually publish this information in real-time online where anyone can access it: Individuals, nonprofits, government entities. The groups can’t effect change directly just by monitoring, but they can give others the tools and information they need to do so.
For instance, this data could give someone ammo to point out if a power plant was running when it shouldn’t have been. Otherwise, that data may not come out until a year or two years after the fact, explained D’souza.
“Another really important goal is to hold companies directly accountable in real time,” he told us. “We can create this extra pressure on the companies.”
Carbon Tracker hopes this project could help put an end to the building of new coal plants and the operation of existing ones by 2040 so that the world can get on track to meet the climate targets put forth by the Paris Agreement.
WattTime, however, is more interested in how this could help inform our understanding of when a power grid is relying on dirty energy versus clean energy, explained Grace Mitchell, a data analyst with the organisation. The group also hopes this project will help bring an end to the pollution many communities around the world face as a result of power plants running recklessly.
Still, this tech has limitations. It relies on images that clouds can mess with, so the quality of the data gathered varies. The satellites can’t measure what kind of pollutants are being emitted, or how much exactly. They can provide estimates, though, and that’s better than nothing in places where the data just doesn’t exist.
“Our main constraint is basically just satellite data,” D’souza said. “And this is constantly improving.”
The groups are still figuring timelines, but the next couple of years may be tough on fossil fuel companies hoping to go about business as usual. The eye in the sky will be watching, and it will be judging.