Seasonal Arctic sea ice forecasting with probabilistic deep learning

Abstract

This is a non-peer reviewed preprint submitted to EarthArXiv. If published in a journal, a link to the final version of the manuscript will be available via the 'Peer-reviewed Publication DOI' link on the right-hand side of this webpage. Anthropogenic warming has led to an unprecedented year-round reduction in Arctic sea ice extent 1,2 . 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 models at longer lead times 3,4 and calibrating their forecasts can be challenging 5 . We present a probabilistic, deep learning 6 sea ice forecasting system, IceNet. The system has been trained on climate simulations covering 1850-2100 and observational data from 1979-2011 to forecast the next 6 months of monthly-averaged sea ice concentration maps. IceNet advances the range of accurate sea ice forecasts, outperforming a state-of-the-art dynamical model 7 in seasonal forecasts of summer sea ice. It also demonstrates a greater ability to predict anomalous pan-Arctic sea ice extents than the models submitted to the Sea Ice Outlook programme 8 . In addition, IceNet's well-calibrated probabilistic forecasts mean it can reliably bound the ice edge between two contours. IceNet's accuracy and reliability represent a step-change in sea ice forecasting, providing a robust framework to build early-warning systems and conservation tools that mitigate risks associated with rapid sea ice loss.Near-surface air temperatures in the Arctic have increased at roughly twice the rate of the global average, a phenomenon known as 'Arctic amplification', caused by a number of positive feedbacks 1,2,9 . Rising temperatures have played a key role in reducing Arctic sea ice, with September sea ice extent now around half that of 1979 when satellite measurements of the Arctic began 10 . This downward trend will continue, even in optimistic greenhouse gas emission reduction scenarios 11 . Climate simulations project the Arctic to be ice free in the summer by 2050 12 . Other studies put this date as early as the 2030s 13 . Such unprecedented sea ice loss has profound local and regional consequences: it is the greatest threat to polar bear populations 14 ; it has increased the intensity and frequency of algal blooms that propagate toxins throughout the food web 15 ; and it poses significant challenges for Indigenous Peoples, with impacts ranging from food security 15 to loss of culture 16 .

An Evaluation of the Performance of Sea Ice Thickness Forecasts to Support Arctic Marine Transport

“…This research is looking for potentially useful correspondence between the forecasts and observations. A consistent finding across this research is that currently operational physics-based forecast models provide skilful seasonal forecasts of large-scale sea ice presence for some regions and seasons [8][9][10][11][12]. For instance, Dirkson, Denis and Merryfield (2019) report that the C3S multi-model seasonal forecast for Arctic sea ice cover during the annual minimum in September is substantially better than simple statistical forecasts derived from past observations [13].…”

A deep-learning estimate of the decadal trends in the Southern Ocean carbon storage

“…Our deep-learning model 58 is a type of neural network that we adapted from the U-net model introduced in a previous study aimed to predict atmospheric ozone concentrations 17 . Similar architectures are also applied in other earth science studies 59 . The schematic diagram of the U-net model is shown in Fig.…”

Benchmarking Probabilistic Machine Learning Models for Arctic Sea Ice Forecasting

“…These records include monthly mean values of sea-ice extent (SIE) from Nimbus-7 SSMR and DMSP SSM/I-SSMIS passive microwave data version 2 and 9 meteorological data variables obtained from ERA-5 global reanalysis product. 3 The choice and details of these variables is presented in our previous causal discovery study conducted on Arctic sea-ice [6]. To conduct our experiments, we first combined all the raw variable datasets to have single temporal and spatial resolution, i.e.…”

“…Probabilistic machine learning models, particularly Bayesian models, provide a principled approach for quantifying uncertainty. Though recent data-driven approaches have shown promising results in sea-ice forecasting, they still struggle with sub-seasonal forecasting at longer lead times [2,3]. We therefore present a rigorous study of four probabilistic models and two baseline models to forecast sea ice extent at multiple lead times, further proposing directions to quantify epistemic uncertainty in model predictions.…”