Potential predictability of meridional heat and volume transports in the North Atlantic ocean

Abstract

The North Atlantic meridional overturning circulation (AMOC) and the associated meridional heat transport (MHT) play a fundamental role in the North Atlantic climate variability on multi-decadal time scales and are therefore of high interest for potential climate predictions. While several potential predictability studies exist for the AMOC at individual latitudes, the potential predictability remains mostly unstudied for the MHT. In this thesis, I analyze both the potential predictability of the AMOC and of the MHT, with a dedicated focus on their latitude-dependence and their interrelation. To analyze the potential predictability on inter-annual to decadal times scales, I generate retrospective model simulations (hindcast ensembles) based on an oceanic state estimate (GECCO). I use two potential predictability measures (prognostic potential predictability and anomaly correlation) to assess the potential predictability structures for the North Atlantic. I initially analyze the potential predictability structures of the MHT and the AMOC separately, and then jointly. For the MHT, I find a latitude-dependent potential predictability structure that indicates a clear separation between the subpolar and the subtropical regime. Decomposing the MHT shows that the gyre component controls the subpolar potential predictability structure of the MHT, while the overturning component controls the subtropical potential predictability structure of the MHT. For the AMOC, I find a latitude-dependent potential predictability structure that is dynamically based on the potential predictability structure of the AMOC’s geostrophic part. The geostrophic part is estimated by removing the wind-driven Ekman transport and by using the thermal wind relation based on the zonal density gradients. A detailed analysis of the density gradient variability indicates the dominant role of the western boundary for the potential predictability of the AMOC. The joint analysis of the AMOC and the MHT shows that only the potential predictability of their geostrophic parts are related. The separate analyses of the MHT and the AMOC already suggest that the potential predictability structures of the MHT’s overturning component and the AMOC are comparable at subtropical latitudes. But the missing link can only be found if one relates the potential predictability structure of the thermal wind transport to the MHT’s overturning component from which the Ekman heat transport has been removed. The results from this study emphasize that the potential predictability of the MHT and the AMOC can neither be interchanged with each other nor be generalized from one latitude to the entire North Atlantic.