Researchers Use Artificial Intelligence to Predict Arctic Ocean Ice Loss

TEMPO.CO, Jakarta – Tim researcher from the British Antarctic Survey (BAS) and the Alan Turing Institute utilize artificial intelligence (AI) technology to more accurately estimate the condition of ice in the Arctic sea. These predictions could support new early warning systems that protect Arctic wildlife and coastal communities from the effects of sea ice loss.

Diterbitkan minggu ini di jurnal Nature Communications, tim peneliti internasional itu menjelaskan bagaimana sistem kecerdasan buatan, IceNet, mengatasi tantangan dalam menghasilkan prakiraan es laut Arktik yang akurat untuk musim mendatang. “Sesuatu yang telah dihindari ilmuwan selama beberapa dekade,” tertulis dalam penelitian, seperti dikutip Phys, Minggu, 29 Agustus 2021.

Es laut, lapisan luas air laut beku yang muncul di Kutub Utara dan Selatan, sangat sulit diprediksi karena hubungannya yang kompleks dengan atmosfer di atas dan laut di bawah. Sensitivitas es laut terhadap peningkatan suhu telah menyebabkan area es laut Arktik musim panas berkurang setengahnya selama empat dekade terakhir, setara dengan hilangnya area sekitar 25 kali ukuran Inggris Raya.

Perubahan yang semakin cepat ini memiliki konsekuensi dramatis bagi iklim bumi, bagi ekosistem Arktik, serta masyarakat adat dan lokal yang mata pencahariannya terkait dengan siklus es laut musiman. “IceNet, hampir 95 persen akurat dalam memprediksi apakah es laut itu akan hadir dua bulan ke depan, lebih baik daripada model berbasis fisika terkemuka,” katanya.

Lead author Tom Andersson, Data Scientist at the Alan Turing Institute-funded BAS AI Lab, explained, “The Arctic is a region at the forefront of climate change and has experienced substantial warming over the past 40 years. “IceNet has the potential to fill an urgent gap in estimating sea ice for Arctic sustainability efforts and runs thousands of times faster than traditional methods,” he said.

Scott Hosking, Principal Investigator, Co-leader of the BAS AI Lab and Senior Research Fellow at the Alan Turing Institute, said he was pleased to see how AI made his team rethink how to do environmental research. “Our new ocean ice forecasting framework combines data from satellite sensors with climate model output in a way that traditional systems cannot.”

Unlike conventional forecasting systems that try to model the laws of physics directly, the authors designed IceNet based on a concept called deep learning. Through this approach, the model ‘learns’ how sea ice is changing from thousands of years of climate simulation data, along with decades of observational data to predict the extent of future Arctic sea ice moons.

Andersson concludes that they have now demonstrated that AI can accurately predict sea ice. The next goal is to develop a daily version of the model and run it publicly in real-time, such as a weather forecast. “It can operate as an early warning system for risks associated with rapid sea ice loss,” Anderson said.

PHYS | NATURE COMMUNICATION

Read:
It is said that Indonesia is entering the Covid-19 Hyperendemic Period, what does it mean?

.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.