Mechanisms influencing seasonal to inter-annual prediction skill of sea ice extent in the Arctic Ocean in MIROC

Jun Ono,Masayoshi Ishii,Masato I Nodzu,Hiroaki Tatebe,Yoshiki Komuro

doi:10.5194/tc-12-675-2018

Abstract

Abstract. To assess the skill of seasonal to inter-annual predictions of the detrended sea ice extent in the Arctic Ocean (SIEAO) and to clarify the underlying physical processes, we conducted ensemble hindcasts, started on 1 January, 1 April, 1 July and 1 October for each year from 1980 to 2011, for lead times up to three years, using the Model for Interdisciplinary Research on Climate (MIROC) version 5 initialised with the observed atmosphere and ocean anomalies and sea ice concentration. Significant skill is found for the winter months: the December SIEAO can be predicted up to 11 months ahead (anomaly correlation coefficient is 0.42). This skill might be attributed to the subsurface ocean heat content originating in the North Atlantic. A plausible mechanism is as follows: the subsurface water flows into the Barents Sea from spring to fall and emerges at the surface in winter by vertical mixing, and eventually affects the sea ice variability there. Meanwhile, the September SIEAO predictions are skillful for lead times of up to two months, due to the persistence of sea ice in the Beaufort, Chukchi, and East Siberian seas initialised in July, as suggested by previous studies.

Highlights

The Arctic has warmed more than twice as much as the global average (e.g., Bekryaev et al, 2010; Cohen et al, 2014), this is referred to as Arctic amplification
We first examine the potential predictability of sea ice extent in the Arctic Ocean (SIEAO) (Fig. 1), based on the lagged auto-correlation coefficients, which is the skill of the persistence forecast
In the hindcasts started from 1 July, the anomaly correlation coefficient (ACC) for September is statistically significant and exceeds that of the persistence forecast, suggesting that September SIEAO can be dynamically predicted from the previous July (ACC = 0.79)

Summary

Introduction

The Arctic has warmed more than twice as much as the global average (e.g., Bekryaev et al, 2010; Cohen et al, 2014), this is referred to as Arctic amplification. A previous study based on two and five year perfect-model experiments from 1 January and 1 September has shown that the potential predictability for sea ice extent remains statistically significant at lead times up to 1–2 years This is primarily because of the persistence of ice thickness anomalies from summer to summer and the persistence of sea surface temperature anomalies from the melt to growth seasons (BlanchardWrigglesworth et al, 2011a; Guemas et al, 2014). Ono et al.: Mechanisms influencing the Arctic sea ice prediction et al, 2016; Sigmond et al, 2016) In these ensemble hindcasts, it is found that ice thickness and surface or subsurface water temperatures are closely related to the prediction skill, as suggested by idealised or perfect-model experiments with climate models (e.g., Blanchard-Wrigglesworth et al, 2011b; Chevallier and Salas y Mélia, 2012; Day et al, 2014a). The present study reveals that subsurface ocean heat content originating from the North Atlantic contributes to the predictability of winter sea ice through advection and vertical mixing processes, which is somewhat different from the re-emergence process of the local subsurface ocean temperature suggested by Bushuk et al (2017)

Experimental design

Predictability of Arctic sea ice extent

Possible mechanisms for prediction skill

Findings

Concluding remarks