The effects of redistributed heat flux on ocean climate change in FAFMIP heat flux anomaly experiments

Abstract

The Flux–Anomaly-Forced Model Inter-comparison Project (FAFMIP) aims to study the response of the ocean to changes in prescribed surface flux (heat, freshwater, and wind) caused by atmospheric CO2 increase. Among these surface flux perturbations, the heat flux has the greatest effect on ocean heat uptake (OHU), the Atlantic Meridional Circulation (AMOC), and sea level changes. This occurs in part because of the redistributed heat flux (Qr’), which is the component of the heat flux driven by ocean circulation changes and almost nearly doubles the initial heat flux perturbation (F) prescribed by FAFMIP in the North Atlantic (NA). In this study, we used a standalone ocean model to quantitatively investigate the impact of Qr’ on OHU and the AMOC in prescribed heat anomaly experiments (faf-heat experiment in FAFMIP). We also showed that whether the other two FAFMIP heat flux anomaly experiments (faf-heat-NA50pct and faf-heat-NA0pct) can reasonably simulate the ocean’s response to prescribed surface heat flux, especially OHU and AMOC. Our findings show that faf-heat-NA50pct is better because the redistribution-feedback (Qr’) nearly equals the initial perturbation (50%F) in the NA and thus recovers the total perturbation F in faf-heat. Although direct comparisons cannot be drawn, both faf-heat-NA50pct and faf-heat experiments with no effect of Qr’ cause AMOC weakening that is roughly comparable with the 1pctCO 2 experiments. The magnitude of simulated AMOC weakening in faf-heat-NA50pct is 3.9Sv, accounting for about 29% of the AMOC intensity in unperturbed experiment, which is closer to the faf-heat experiment (about 31%) with no effect of Qr’ relative to the faf-heat experiment. Because of negative Qr’ in all regions except the NA region especially for the Central and Eastern Equatorial Pacific and Southern Ocean, the simulated global OHU in faf-heat-NA50pct is reduced by 10% as compared to the added OHU in the faf-heat experiment. The Qr’ in the faf-heat experiment increases OHU by approximately 13% of the total OHU and further reduces AMOC intensity by 3.4 Sv, accounting for 26% of the AMOC strength in an unperturbed experiment. Faf-heat-NA0pct results in a significant reduction in OHU (approximately 26%), and the magnitude of AMOC weakening is only approximately 1.9 Sv, despite no F being added in the NA.