skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: MT-IceNet - A Spatial and Multi-Temporal Deep Learning Model for Arctic Sea Ice Forecasting
Arctic amplification has altered the climate patterns both regionally and globally, resulting in more frequent and more intense extreme weather events in the past few decades. The essential part of Arctic amplification is the unprecedented sea ice loss as demonstrated by satellite observations. Accurately forecasting Arctic sea ice from sub-seasonal to seasonal scales has been a major research question with fundamental challenges at play. In addition to physics-based Earth system models, researchers have been applying multiple statistical and machine learning models for sea ice forecasting. Looking at the potential of data-driven approaches to study sea ice variations, we propose MT-IceNet – a UNet-based spatial and multi-temporal (MT) deep learning model for forecasting Arctic sea ice concentration (SIC). The model uses an encoder-decoder architecture with skip connections and processes multi-temporal input streams to regenerate spatial maps at future timesteps. Using bi-monthly and monthly satellite retrieved sea ice data from NSIDC as well as atmospheric and oceanic variables from ERA5 reanalysis product during 1979-2021, we show that our proposed model provides promising predictive performance for per-pixel SIC forecasting with up to 60% decrease in prediction error for a lead time of 6 months as compared to its state-of-the-art counterparts.  more » « less
Award ID(s):
1942714 2118285
PAR ID:
10400897
Author(s) / Creator(s):
;
Date Published:
Journal Name:
9th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT 2022)
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; (, Proceedings of Tackling Climate Change with Machine Learning Workshop at International Conference on Machine Learning (ICML 2021))
    Accurately forecasting Arctic sea ice from sub- seasonal to seasonal scales has been a major scientific effort with fundamental challenges at play. In addition to physics-based earth system models, researchers have been applying multiple statistical and machine learning models for sea ice forecast- ing. Looking at the potential of data-driven sea ice forecasting, we propose an attention-based Long Short Term Memory (LSTM) ensemble method to predict monthly sea ice extent up to 1 month ahead. Using daily and monthly satellite retrieved sea ice data from NSIDC and atmospheric and oceanic variables from ERA5 reanalysis product for 39 years, we show that our multi-temporal ensemble method outperforms several baseline and recently proposed deep learning models. This will substantially improve our ability in predicting future Arctic sea ice changes, which is fundamental for forecasting transporting routes, resource development, coastal erosion, threats to Arctic coastal communities and wildlife. 
    more » « less
  2. ; ; (, Environmental Research: Climate)
    Abstract The recent Arctic sea ice loss is a key driver of the amplified surface warming in the northern high latitudes, and simultaneously a major source of uncertainty in model projections of Arctic climate change. Previous work has shown that the spread in model predictions of future Arctic amplification (AA) can be traced back to the inter-model spread in simulated long-term sea ice loss. We demonstrate that the strength of future AA is further linked to the current climate’s, observable sea ice state across the multi-model ensemble of the 6th Coupled Model Intercomparison Project (CMIP6). The implication is that the sea-ice climatology sets the stage for long-term changes through the 21st century, which mediate the degree by which Arctic warming is amplified with respect to global warming. We determine that a lower base-climate sea ice extent and sea ice concentration (SIC) in CMIP6 models enable stronger ice melt in both future climate and during the seasonal cycle. In particular, models with lower Arctic-mean SIC project stronger future ice loss and a more intense seasonal cycle in ice melt and growth. Both processes systemically link to a larger future AA across climate models. These results are manifested by the role of climate feedbacks that have been widely identified as major drivers of AA. We show in particular that models with low base-climate SIC predict a systematically stronger warming contribution through both sea-ice albedo feedback and temperature feedbacks in the future, as compared to models with high SIC. From our derived linear regressions in conjunction with observations, we estimate a 21st-century AA over sea ice of 2.47–3.34 with respect to global warming. Lastly, from the tight relationship between base-climate SIC and the projected timing of an ice-free September, we predict a seasonally ice-free Arctic by mid-century under a high-emission scenario. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ; et al (, Nature Communications)
    Abstract 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. 
    more » « less
  4. ; ; ; ; ; ; ; ; ; ; et al (, Bulletin of the American Meteorological Society)
    Abstract This study quantifies the state of the art in the rapidly growing field of seasonal Arctic sea ice prediction. A novel multimodel dataset of retrospective seasonal predictions of September Arctic sea ice is created and analyzed, consisting of community contributions from 17 statistical models and 17 dynamical models. Prediction skill is compared over the period 2001–20 for predictions of pan-Arctic sea ice extent (SIE), regional SIE, and local sea ice concentration (SIC) initialized on 1 June, 1 July, 1 August, and 1 September. This diverse set of statistical and dynamical models can individually predict linearly detrended pan-Arctic SIE anomalies with skill, and a multimodel median prediction has correlation coefficients of 0.79, 0.86, 0.92, and 0.99 at these respective initialization times. Regional SIE predictions have similar skill to pan-Arctic predictions in the Alaskan and Siberian regions, whereas regional skill is lower in the Canadian, Atlantic, and central Arctic sectors. The skill of dynamical and statistical models is generally comparable for pan-Arctic SIE, whereas dynamical models outperform their statistical counterparts for regional and local predictions. The prediction systems are found to provide the most value added relative to basic reference forecasts in the extreme SIE years of 1996, 2007, and 2012. SIE prediction errors do not show clear trends over time, suggesting that there has been minimal change in inherent sea ice predictability over the satellite era. Overall, this study demonstrates that there are bright prospects for skillful operational predictions of September sea ice at least 3 months in advance. 
    more » « less
  5. ; ; ; ; ; (, Journal of Glaciology)
    Abstract The accuracy of sea-ice motion products provided by the National Snow and Ice Data Center (NSIDC) and the Ocean and Sea Ice Satellite Application Facility (OSI-SAF) was validated with data collected by ice drifters that were deployed in the western Arctic Ocean in 2014 and 2016. Data from both NSIDC and OSI-SAF products exhibited statistically significant ( p < 0.001) correlation with drifter data. The OSI-SAF product tended to overestimate ice speed, while underestimation was demonstrated for the NSIDC product, especially for the melt season and the marginal ice zone. Monthly Lagrangian trajectories of ice floes were reconstructed using the products. Larger spatial variability in the deviation between NSIDC and drifter trajectories was observed than that of OSI-SAF, and seasonal variability in the deviation for NSIDC was observed. Furthermore, trajectories reconstructed using the NSIDC product were sensitive to variations in sea-ice concentration. The feasibility of using remote-sensing products to characterize sea-ice deformation was assessed by evaluating the distance between two arbitrary positions as estimated by the products. Compared with the OSI-SAF product, relative errors are lower (<11.6%), and spatial-temporal resolutions are higher in the NSIDC product, which makes it more suitable for estimating sea-ice deformation. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email