Satellite data and algorithms reveal which ships are emitting too much nitrogen

Ships still emit large amounts of nitrogen oxides. Until now it has been impossible to measure this at sea, but that is about to change. Solomia Kulchava has developed an algorithm that combines satellite data to identify which ships are emitting too much. Ms. Kulchava received her Ph.D. June 11th.

While automobiles, power plants, and industry are reducing nitrogen oxide (NOx) emissions, the shipping industry is lagging behind. Even though exhaust gas regulations for new ships have become stricter, exhaust gases have hardly decreased. Nitrogen oxides cause smog and acid rain, which are harmful to humans and nature. The International Maritime Organization (IMO) aims to regulate emissions from ships, but this is difficult.

“In ports and along the coast we can measure emissions, but in the open ocean it’s not possible. We don’t know what type of fuel the ship is using,” said Solomia Kurchava, a Ukrainian who studied economics and theoretical physics in Poland. He explains. Are you using nitrogen oxides and how much nitrogen oxides are you emitting? ”

But now there is a satellite instrument, Tropomi, led by the Royal Netherlands Meteorological Institute (KNMI). Since 2017, it has been orbiting the Earth every 100 minutes, measuring various compounds in the lower atmosphere, including methane, ozone, carbon monoxide, and nitrogen dioxide (NO2).

“We will create a complete world map that will allow us to monitor air quality and climate-related compounds every day. With this instrument, we will be able to map air pollution country by country and even city by city,” Kulchava said. ”.

The data can be used to estimate emissions from activities such as biomass burning and coal mining, as well as from shipping. “When I started my research in 2019, it was just becoming clear that Tropomi could detect NO2 plumes from several individual ships,” Kulchava added.

This is a promising finding, but there are some caveats. “The data is noisy, and it’s difficult to separate emissions from ships from the rest of the data. Additionally, we can’t distinguish between plumes from all ships. Well, how far can we go?”

This challenge formed the basis of Kulchava’s doctoral research. “Based on machine learning, we need to develop a method that can not only automatically detect ship plumes, but also identify which ship the plume came from and whether that ship is emitting more than expected. did.”

Kulchava developed an algorithm that integrates Tropomi data with additional information about the ship’s movements and extracts relevant details. “We are proud of the results. We started from scratch, but now we have a solid foundation to move closer to monitoring ship emissions based on satellite data. How much can we improve? ” Perhaps other compounds could help “track the ship’s footprint.” ”

Once introduced, the system will be available to the Netherlands’ Human Environment and Transport Authority (ILT). “Inspectors will then be able to board the vessels identified as abnormal emitters and examine things such as fuel quality and engine settings,” Tropomi says.

Kulchava says it was sometimes very difficult to solve all the problems she encountered. But she is not tired of Tropomi. “At SRON Netherlands Space Research Institute, I am currently working on Tropomi data on methane. We are developing it.”