Autumn arctic pacific sea ice dipole as a source of predictability for subsequent spring barents sea ice condition
Journal
Journal of Climate
Journal Volume
34
Journal Issue
2
Pages
787-804
Date Issued
2021
Author(s)
Abstract
This study uses observational and reanalysis datasets in 1980-2016 to show a close connection between a boreal autumn sea ice dipole in the Arctic Pacific sector and sea ice anomalies in the Barents Sea (BS) during the following spring. The September-October Arctic Pacific sea ice dipole variations are highly correlated with the subsequent April-May BS sea ice variations (r 5 0.71). The strong connection between the regional sea ice variabilities across the Arctic uncovers a new source of predictability for spring BS sea ice prediction at 7-month lead time. A cross-validated linear regression prediction model using the Arctic Pacific sea ice dipole with 7-month lead time is demonstrated to have significant prediction skills with 0.54-0.85 anomaly correlation coefficients. The autumn sea ice dipole, manifested as sea ice retreat in the Beaufort and Chukchi Seas and expansion in the East Siberian and Laptev Seas, is primarily forced by preceding atmospheric shortwave anomalies from late spring to early autumn. The spring BS sea ice increases are mostly driven by an ocean-to-sea ice heat flux reduction in preceding months, associated with reduced horizontal ocean heat transport into the BS. The dynamical linkage between the two regional sea ice anomalies is suggested to involve positive stratospheric polar cap anomalies during autumn and winter, with its center slowly moving toward Greenland. The migration of the stratospheric anomalies is followed in midwinter by a negative North Atlantic Oscillation-like pattern in the troposphere, leading to reduced ocean heat transport into the BS and sea ice extent increase. ? 2020 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).
Subjects
Arctic
Atmospheric circulation
Ocean circulation
Sea ice
Seasonal forecasting
Atmospheric pressure
Forecasting
Heat flux
Heat transfer
Oceanography
Predictive analytics
Anomaly correlations
Dipole variations
Highly-correlated
North Atlantic oscillations
Ocean heat transport
Regression predictions
Sea ice extent
Sea ice retreats
atmospheric circulation
autumn
climate prediction
heat transfer
oceanic circulation
sea ice
seasonal variation
Arctic Ocean
Barents Sea
Type
journal article