EN ENThe planet is heating up more than previously expected, according to the latest report from the IPCC, and the Arctic glaciers are particularly affected by the consequences.
Since the 1990s, in the Arctic, near-surface air temperatures have risen more than twice the global average, while the extent of the Arctic ice sheet has halved compared to the first satellite images in our possession, dating back to 1979. The melting of the ice sheets has devastating effects on the polar ecosystems but also on the climate system of the entire planet. For this reason, it is essential to monitor its state and possibly predict its trend even in the short to medium term.
Monitoring has been carried out for decades through deterministic predictive models, based on the application of physical laws and some variables, regarding the atmosphere, ice, and seawater. The continuous instability and interdependence of these natural factors make the work of forecasters particularly complex, who are unable to achieve the necessary effectiveness to make valuable predictions beyond a few weeks. Climatologists have been trying for years to refine the criteria considered and integrate different models to improve their performance, but with little success.
A few days ago, researchers from the British Antarctic Survey and the Alan Turing Institute presented their new predictive system, based on AI. IceNet, as they called it, is already able to assess the seasonal melting of the ice sheets with 95% accuracy and could improve further. Its system, in fact, thanks to a huge amount of data, “learns” to understand how the ice varies and continues to improve with the acquisition of new information.
According to the head of the AI LAB at the British Antarctic Survey, Scott Hosking, this demonstrates how AI could be successfully applied in much of the research on climate and the environment. The continuous improvement capability of AI systems is indeed perfect for interpreting contexts in continuous evolution such as climate, financial markets, or the production process in its entirety.
If you also want to experience the power of artificial intelligence to interpret the complex facets of the market, contact us at info@aidia.it or fill out the form in the contact us section and schedule a free consultation with us.
_Source: Andersson, Hosking, Pérez-Ortiz et al., Seasonal Arctic sea ice forecasting with probabilistic deep learning. Nat Commun 12, 5124 (2021) _
Marketing Specialist at AIDIA, graduated in International Studies in Florence, passionate about history, economics, and the bizarre things of the world.
At Aidia, we develop AI-based software solutions, NLP solutions, Big Data Analytics, and Data Science. Innovative solutions to optimize processes and streamline workflows. To learn more, contact us or send an email to info@aidia.it.
