{"title":"Deep Learning Shows Promise for Seasonal Prediction of Antarctic Sea Ice in a Rapid Decline Scenario","authors":"","doi":"10.1007/s00376-024-3380-y","DOIUrl":null,"url":null,"abstract":"<h3>Abstract</h3> <p>The rapidly changing Antarctic sea ice has garnered significant interest. To enhance the prediction skill for sea ice and respond to the Sea Ice Prediction Network-South’s latest call, this study presents the reforecast results of Antarctic sea-ice area and extent from December to June of the coming year with a Convolutional Long Short-Term Memory (ConvLSTM) Network. The reforecast experiments demonstrate that ConvLSTM captures the interannual and interseasonal variability of Antarctic sea ice successfully, and performs better than the European Centre for Medium-Range Weather Forecasts. Based on this, we present the prediction from December 2023 to June 2024, indicating that the Antarctic sea ice will remain at lows, but may not create a new record low. This research highlights the promising application of deep learning in Antarctic sea-ice prediction.</p>","PeriodicalId":7249,"journal":{"name":"Advances in Atmospheric Sciences","volume":"10 1","pages":""},"PeriodicalIF":6.5000,"publicationDate":"2024-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advances in Atmospheric Sciences","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s00376-024-3380-y","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"METEOROLOGY & ATMOSPHERIC SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
The rapidly changing Antarctic sea ice has garnered significant interest. To enhance the prediction skill for sea ice and respond to the Sea Ice Prediction Network-South’s latest call, this study presents the reforecast results of Antarctic sea-ice area and extent from December to June of the coming year with a Convolutional Long Short-Term Memory (ConvLSTM) Network. The reforecast experiments demonstrate that ConvLSTM captures the interannual and interseasonal variability of Antarctic sea ice successfully, and performs better than the European Centre for Medium-Range Weather Forecasts. Based on this, we present the prediction from December 2023 to June 2024, indicating that the Antarctic sea ice will remain at lows, but may not create a new record low. This research highlights the promising application of deep learning in Antarctic sea-ice prediction.
期刊介绍:
Advances in Atmospheric Sciences, launched in 1984, aims to rapidly publish original scientific papers on the dynamics, physics and chemistry of the atmosphere and ocean. It covers the latest achievements and developments in the atmospheric sciences, including marine meteorology and meteorology-associated geophysics, as well as the theoretical and practical aspects of these disciplines.
Papers on weather systems, numerical weather prediction, climate dynamics and variability, satellite meteorology, remote sensing, air chemistry and the boundary layer, clouds and weather modification, can be found in the journal. Papers describing the application of new mathematics or new instruments are also collected here.