Seasonal Arctic sea ice forecasting with probabilistic deep learning

Andersson, Tom R. ORCID: https://orcid.org/0000-0002-1556-9932; Hosking, J. Scott ORCID: https://orcid.org/0000-0002-3646-3504; Pérez-Ortiz, María; Paige, Brooks; Elliott, Andrew; Russell, Chris; Law, Stephen; Jones, Daniel C. ORCID: https://orcid.org/0000-0002-8701-4506; Wilkinson, Jeremy ORCID: https://orcid.org/0000-0002-7166-3042; Phillips, Tony ORCID: https://orcid.org/0000-0002-3058-9157; Byrne, James ORCID: https://orcid.org/0000-0003-3731-2377; Tietsche, Steffen; Sarojini, Beena Balan; Blanchard-Wrigglesworth, Eduardo; Aksenov, Yevgeny ORCID: https://orcid.org/0000-0001-6132-3434; Downie, Rod; Shuckburgh, Emily ORCID: https://orcid.org/0000-0001-9206-3444. 2021 Seasonal Arctic sea ice forecasting with probabilistic deep learning. Nature Communications, 12, 5124. 12, pp. 10.1038/s41467-021-25257-4

Before downloading, please read NORA policies.

Preview

Text (Open Access)
© 2021. The Authors.
s41467-021-25257-4.pdf - Published Version
Available under License Creative Commons Attribution 4.0.
Download (2MB) | Preview

Official URL: https://www.nature.com/articles/s41467-021-25257-4

Abstract/Summary

Anthropogenic warming has led to an unprecedented year-round reduction in Arctic sea ice extent. This has far-reaching consequences for indigenous and local communities, polar ecosystems, and global climate, motivating the need for accurate seasonal sea ice forecasts. While physics-based dynamical models can successfully forecast sea ice concentration several weeks ahead, they struggle to outperform simple statistical benchmarks at longer lead times. We present a probabilistic, deep learning sea ice forecasting system, IceNet. The system has been trained on climate simulations and observational data to forecast the next 6 months of monthly-averaged sea ice concentration maps. We show that IceNet advances the range of accurate sea ice forecasts, outperforming a state-of-the-art dynamical model in seasonal forecasts of summer sea ice, particularly for extreme sea ice events. This step-change in sea ice forecasting ability brings us closer to conservation tools that mitigate risks associated with rapid sea ice loss.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	10.1038/s41467-021-25257-4
ISSN:	20411723
Additional Keywords:	computer science, cryospheric science, environmental impact, statistics
Date made live:	26 Aug 2021 14:50 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/529437

Item Type:

Publication - Article

Digital Object Identifier (DOI):

10.1038/s41467-021-25257-4

ISSN:

20411723

Additional Keywords:

computer science, cryospheric science, environmental impact, statistics

Date made live:

26 Aug 2021 14:50 +0 (UTC)

URI:

https://nora.nerc.ac.uk/id/eprint/529437