IceDiff: High Resolution and High-Quality Sea Ice Forecasting with Generative Diffusion Prior

TL;DR

IceDiff is a novel two-stage deep learning framework that significantly improves high-resolution sea ice forecasting, enabling detailed 6.25km resolution predictions crucial for scientific and operational applications.

Contribution

The paper introduces IceDiff, the first model to achieve 6.25km resolution sea ice forecasting using a combination of vision transformers and diffusion models.

Findings

Achieves 6.25km resolution sea ice forecasts.

Outperforms previous methods at coarser resolutions.

Demonstrates practical utility for scientific and operational use.

Abstract

Variation of Arctic sea ice has significant impacts on polar ecosystems, transporting routes, coastal communities, and global climate. Tracing the change of sea ice at a finer scale is paramount for both operational applications and scientific studies. Recent pan-Arctic sea ice forecasting methods that leverage advances in artificial intelligence has made promising progress over numerical models. However, forecasting sea ice at higher resolutions is still under-explored. To bridge the gap, we propose a two-staged deep learning framework, IceDiff, to forecast sea ice concentration at finer scales. IceDiff first leverages an independently trained vision transformer to generate coarse yet superior forecasting over previous methods at a regular 25km x 25km grid. This high-quality sea ice forecasting can be utilized as reliable guidance for the next stage. Subsequently, an unconditional diffusion model pre-trained on sea ice concentration maps is utilized for sampling down-scaled sea ice forecasting via a zero-shot guided sampling strategy and a patch-based method. For the first time, IceDiff demonstrates sea ice forecasting with the 6.25km x 6.25km resolution. IceDiff extends the boundary of existing sea ice forecasting models and more importantly, its capability to generate high-resolution sea ice concentration data is vital for pragmatic usages and research.