Decadal prediction of the North Atlantic subpolar gyre in the HiGEM high-resolution climate model

Jon Robson,David P Stevens,Irene Polo,Dan L R Hodson,Len C Shaffrey

doi:10.1007/s00382-017-3649-2

Jon Robson, David P Stevens + Show 3 more

Open Access

https://doi.org/10.1007/s00382-017-3649-2

Copy DOI

Abstract

This paper presents an analysis of initialised decadal hindcasts of the North Atlantic subpolar gyre (SPG) using the HiGEM model, which has a nominal grid-spacing of 90 km in the atmosphere, and 1/3^{circ } in the ocean. HiGEM decadal predictions (HiGEM-DP) exhibit significant skill at capturing 0–500 m ocean heat content in the SPG, and outperform historically forced transient integrations and persistence for up to a decade ahead. An analysis of case-studies of North Atlantic decadal change, including the 1960s cooling, the mid-1990s warming, and the post-2005 cooling, show that changes in ocean circulation and heat transport dominate the predictions of the SPG. However, different processes are found to dominate heat content changes in different regions of the SPG. Specifically, ocean advection dominates in the east, but surface fluxes dominate in the west. Furthermore, compared to previous studies, we find a smaller role for ocean heat transport changes due to ocean circulation anomalies at the latitudes of the SPG, and, for the 1960s cooling, a greater role for surface fluxes. Finally, HiGEM-DP predicts the observed positive state of the North Atlantic Oscillation in the early 1990s. These results support an important role for the ocean in driving past changes in the North Atlantic region, and suggest that these changes were predictable.

Highlights

For clarity we only show the ensemble mean of the hindcasts, and we smooth the transient runs with a 1-year running-mean
Crosses and diamonds show the integrated change in heat content in corresponding section of the transient run for lead year 5 and 10, respectively. e, f Show the same as (d), but for the 1995 and 2005 hindcasts. g Shows the anomalous 3-year-mean heat fluxes in the hindcasts integrated over the subpolar gyre (SPG). h and i Show the same as (g) but for the 1995 and
As a strong Atlantic Meridional Overturning Circulation (AMOC) is initialised in HiGEM-DP in the 1990s (see Shaffrey et al (2016), their Fig. 9), these results suggest that the ocean plays an important role in maintaining the positive NAO

Summary

Introduction

There is substantial evidence that links changes in North Atlantic Sea Surface Temperature (SST) to important climate impacts, including the numbers of hurricanes (Goldenberg et al 2001; Zhang and Delworth 2006; Smith et al 2010), and the surface climate over the surrounding continents (Sutton and Hodson 2005; Zhang and Delworth 2006; Knight et al 2006; Sutton and Dong 2012). The North Atlantic, and the North Atlantic subpolar gyre (SPG), is a region where hindcasts (or retrospective climate predictions), which are initialised from observations, are found to have significant skill in capturing past variations of SST (Smith et al 2010; GarcíaSerrano et al 2012; Matei et al 2012; Doblas-Reyes et al 2013; Kirtman et al 2013; Shaffrey et al 2016). In order to build confidence that future predictions will remain skillful, it is important to understand why hindcasts are successful, and if the dominant mechanisms are robust. An important method for understanding the sources of skill in initialised climate predictions has been a focus on specific case studies of large decadal change.

Results

Discussion

Conclusion