Weak 21st-century AMOC response to Greenland meltwater in a strongly eddying ocean model

TL;DR

The paper quantifies how Greenland meltwater influences the AMOC in an eddy-resolving ocean model under SSP5-8.5 forcing, finding a modest additional weakening of about $0.6 \pm 0.2$ Sv by 2100 that is robust across high- and low-resolution configurations. A strong state dependence is revealed: under present-day forcing, the same runoff can produce a much larger AMOC decline, while under strong warming the response is dominated by background stratification and scale-depth changes rather than meridional density shifts. The study concludes that background ocean state, not resolution, governs the meltwater-AMOC interaction, and highlights the need to address mean-state biases and coupling in future projections.

Abstract

Climate models project that the Atlantic Meridional Overturning Circulation (AMOC) will weaken in the 21st century, but the magnitude is highly uncertain. Some of this uncertainty is structural, as most climate models neglect increasing meltwater from the Greenland ice sheet and do not explicitly capture mesoscale ocean eddies. Here, we quantify the impact of Greenland meltwater on the AMOC until 2100 under SSP5-8.5 forcing for the first time in a strongly eddying (1/10° horizontal resolution) ocean model. The meltwater-induced additional AMOC weakening is small (0.6 $\pm$ 0.2 Sv) compared to the weakening due to warming alone, and similar at high and low resolution. The same meltwater would cause a stronger AMOC weakening under present-day climate conditions. We link both resolution-independence and state-dependence to large-scale controls of the AMOC. Our results demonstrate that the background ocean state is more important than resolution in determining how Greenland meltwater affects the AMOC.

Figures (4)

• Figure 1: AMOC timeseries at 26.5°N under historical and SSP5-8.5 forcing in the (a) high-resolution (1/10°) POP and (b) low-resolution (1°) POP. Black lines are the "Reference" simulations with climate change forcing only and blue lines are the "Meltwater" simulations with climate change and Greenland meltwater forcing. The cyan line in panel b is with Greenland meltwater forcing only and otherwise fixed climate boundary conditions. The observed (2004--2024) AMOC time series from the RAPID array Moat2025 is also shown in panel a. For comparison, the main results from the high-resolution simulation are duplicated semi-transparently in panel b.
• Figure 2: Meltwater tracer concentration and transport at 26°N: (a-c) Concentration of the meltwater tracer (on a logarithmic scale) in the year 2100, (d-f) Meltwater transport across 26°N (black) decomposed into transport by the Gulf Stream, transport in the upper 1000m excluding the Gulf Stream ("gyre recirculation"), and by the deep western boundary current (DWBC). The transport below 1000m outside of the DWBC is very small and not shown.
• Figure 3: Mixed-layer depth in March for HR-POP (first row) and LR-POP (second row) for 1981--2010 (first column) and 2080--2100 (second column). For 2080--2100, mixed-layer depths are shown from the "Meltwater" simulations, but they do not differ strongly from the "Reference" simulations for the same period.
• Figure 4: AMOC changes attributed to density gradient and scale depth changes. For each simulation, the AMOC anomaly (filled bars) is shown along with the reconstructed AMOC anomaly from thermal wind balance (non-filled bars) and its partitioning in a density gradient ($\Delta_y \rho$) and a scale depth ($H$) contribution. The pairs (high and low resolution) of AMOC anomalies are color-coded as follows. AMOC weakening (2080--2100 minus 1981--2010) due to SSP5-8.5 CO$_2$ forcing only: black, AMOC anomaly due to Greenland meltwater under SSP5-8.5 forcing (2080-2100): blue, AMOC anomaly due to Greenland meltwater under fixed normal-year forcing: cyan.