Researchers combine AI with satellites to measure Arctic Sea ice thickness year round

LinkedIn +

An international team of scientists has announced that it is now able to measure the thickness of ice in the Arctic Sea 365 days a year using satellites.

The solution was developed by an international team that includes the UK’s National Oceanography Centre (NOC) and the Arctic University of Norway (UiT), along with partners from the USA, Canada and Germany.

Sea ice poses a critical hazard to ships, particularly small and medium-sized vessels, with accurate information about the location and thickness of ice needed in advance. The Norwegian Meteorological Institute provides sea ice forecasts for the Arctic but previously lacked reliable information concerning ice thickness for the summer months.

Jack Landy, team leader from the department of physics and technology at UiT, said, “The Arctic ice is melting faster than ever. We need information about the thickness of the sea ice, both to reduce safety risks for businesses and shipping in the Arctic, but also to make more accurate forecasts for the future climate.”

Satellites have been used to measure the thickness of ice in the Arctic since the 1980s. But the technique has only worked in winter, from October to March when the ice and snow are cold and dry, as the satellites were unable to distinguish between melting ice and sea water. To solve the problem, researchers adopted artificial intelligence using convolutional neural network methods to classify sea ice and open water leads, and examined the data from the satellites. They were able to identify when the satellite sees the ocean as opposed to melting ice. The advantage of this is that it improves the robustness of the satellite thickness measurement and extends the lead time of the sea ice forecasts by several months.

Landy added, “Using the new satellite data, we are finally able to make sea ice forecasts informed by the ice thickness all year round, thereby reducing risks for ships and fishing boats. We can also predict whether there will be ice or not at a given location in September by measuring the ice thickness in May, helping shippers to navigate much easier.”

In addition, the scientists constructed a new computer model of the satellite sensor, to ensure it is measuring the correct height and thickness. The research team has developed the first data set showing the thickness of sea ice across the entire Arctic and through a whole year.

Dr Michel Tsamados, associate professor in polar observation and modeling at University College London and one of the researchers behind the study, said, “When we use the new ice thickness data in advanced climate models, it will improve both our short-term forecasts for the weather at the mid-latitudes and the long-term forecasts that show what climate we will have in the future.”

Dr Yevgeny Aksenov, principal researcher at NOC, said, “Sea ice thickness is a key factor in designing operational forecasting and hazard warning systems for polar navigation, since it affects the safe speed of vessels in icy conditions. However, looking only at sea ice conditions can also be misleading. Other hazards, such as high ocean waves frequently occurring in the polar seas in the absence of sea ice, can damage ships and offshore structures. We also need awareness of other combined hazards, such as ship hull and offshore structures damaged by sea ice and waves in the polar environment as they have wide relevance to maritime industries’ safety.”

To read the complete study, click here.

Share this story:

About Author

, editor-in-chief

Dan first joined UKi Media & Events in 2014 having spent the early years of his career in the recruitment industry. As editor, he now produces content for Meteorological Technology International, unearthing the latest technological advances and research methods for the publication of each exciting new issue. When he’s not reporting on the latest meteorological news, Dan can be found on the golf course or apprehensively planning his next DIY project.

Comments are closed.