Live Streaming
Program Highlight
Company Profile
Zona Integritas
Thursday, 11 April 2024 12:46

Artificial Intelligence Monitors Emissions of Environmentally Destructive Methane Gas

Written by 
Rate this item
(1 Vote)

 

VOInews, Jakarta: Paris-based environmental engineering firm Kayrros is using satellite data and artificial intelligence (AI) models to accurately monitor leaks of methane, a greenhouse gas produced by fossil fuels. The invisible plumes of odourless gas in the air appear on the company's maps as coloured clouds, in satellite imagery.

 

Since 2019, Kayrros has detected nearly 10,000 methane concentration events around the world, from the United States to India to Turkmenistan and Algeria, due to harmful practices not being properly managed in fossil industry-based infrastructure.
The leaks occur due to pressure problems in gas pipelines and the opening of valves to avoid the risk of explosion," explains Kayrros data scientist Alexis Groshneri.

 

Methane has a shorter lifespan than carbon dioxide but greater warming potential, and is a factor in 30 per cent of global warming since the Industrial Revolution. According to the International Energy Agency (IEA), 40 per cent of methane emissions from the production of fossil energy sources (coal, oil and gas) can be avoided at little cost. However, these emissions must be properly monitored and measured.

 


Kayrros uses nine satellites operated by governments or official space agencies such as the European and American Space Agencies, which send their images back to Earth at different frequencies, several times a week or every 15 minutes, with greater or lesser resolution, depending on the model.

 

"We are working with the different sources available to try to improve the observations," Groshenry says.
The downside for mathematicians in green technology is that there is a huge amount of data to process, which is impossible to observe in a single image. That's where artificial intelligence comes in. (Daniel).

 

Source: AFP

Read 474 times Last modified on Thursday, 11 April 2024 13:32