All

Rapid and Proficient Sea Ice Predictions

2025-03-18 00:17:49

Artificial intelligence enables the creation of a highly efficient and proficient substitute for a coupled Arctic sea ice prediction model through the use of generative diffusion techniques.

Graphs and a map from the study. — (Left) The figures display the sea ice divergence (a, e, and i), shear rate (b, f, and j), sea-ice concentration (c, g, and k), and the semidiurnal variation in sea ice thickness (d, h, and l) as modeled by neXtSIM (a–d), the deterministic emulated forecast (e–h), and an ensemble member from the residual diffusion forecast (i–l). These data are pertinent to the date of December 30, 2017, at 03:00 UTC, with a forecast lead time of 50 days. (Right) A sample of the pan-Arctic validation sea ice thickness, as simulated by neXtSIM for a snapshot from January 2015, is presented on the right, alongside the 64 × 64 grid point surrogate domain (indicated by the red box) for frames (a) to (l). Credit: Finn et al. [2024], Figures 1 and 6.

Editors' Highlights provide concise overviews of the latest research articles published by the editors of AGU journals.

Source: Journal of Advances in Modeling Earth Systems

Sea ice model surrogates developed during the past decade are disrupting polar foreca

Recommend