The meridional overturning circulation (MOC) in the ocean is a vital component of the global climate system, impacting heat and freshwater transport on regional and global scales. The Atlantic MOC (AMOC), characterized by deep convection and deep-water formation in the Labrador Sea and Greenland-Iceland-Nordic (GIN) seas, is currently the dominant MOC, with a strength of approximately 18 Sv. In contrast, a comparable Pacific MOC (PMOC) is absent in the North Pacific. The evolutionary history of the AMOC and PMOC remains debated, with geological evidence suggesting a potential shift in the primary deep-water formation region from the Pacific to the Atlantic. While the asymmetry of net surface freshwater flux and ocean basin geometry are often cited as contributing factors to the Atlantic-Pacific differences, the role of continental topography remains a significant area of investigation. The uplift of major mountain ranges, including the Rocky Mountains, the Transantarctic and Gamburtsev Mountains, the Andes Mountains, Greenland, and the Tibetan Plateau (TP), has occurred over geological timescales, potentially influencing the global MOC (GMOC). The timing of these uplifts is often debated, but the TP's uplift coincides with the onset of NADW formation, suggesting a possible link. This study aims to numerically investigate the individual and cumulative effects of these major mountain ranges on the GMOC, focusing on isolating the topographic effects from other climate-influencing factors.

Previous research has explored the influence of orography on ocean circulation. Some studies suggested a strong North Pacific deep-water (NPDW) formation during the Paleocene, while North Atlantic deep-water (NADW) formation was weak and developed later. However, more recent research from the DeepMIP project challenges these findings, indicating inconclusive evidence for NPDW formation during the early Eocene and NADW formation during the early Eocene. The timing of the AMOC's full establishment also remains uncertain, with estimates ranging from the late Miocene to the late Pliocene to early Pleistocene. The asymmetry between the Atlantic and Pacific in overturning modes has been attributed to factors such as higher sea-surface salinity (SSS) in the North Atlantic (a net evaporation basin), ocean basin geometry (narrow basins favoring deep overturning), and ocean gateways (e.g., the opening of the Drake Passage/Tasman Seaway). Studies have also examined the impact of the uplift of large continental mountains on climate, suggesting a potential link between the evolution of continental terrain and large-scale transitions in the GMOC. Prior research has pointed to the importance of the TP as a significant factor affecting changes in the GMOC, although the precise chronology of its uplift remains a subject of debate.

This study uses the National Center for Atmospheric Research Community Earth System Model version 1.0 (NCAR CESM 1.0) with a T31_gx3v7 grid. The model includes atmospheric (CAM4), ocean (POP2), land (CLM4), and sea-ice (CICE) components. A series of topography experiments were conducted, all starting from a "Flat" scenario with globally flat topography at 50 m above sea level. This "Flat" scenario was integrated for 1600 years under pre-industrial conditions. Five major mountain ranges—Rocky Mountains (RM), Antarctic (AT), Andes Mountains (AM), Greenland (GL), and Tibetan Plateau (TP)—were sequentially added to the "Flat" scenario in experiments to isolate their individual and combined effects on the GMOC. The order of mountain addition in the "Flat2Real" experiment mirrored estimated uplift times. Additional experiments examined different combinations of mountain ranges. All other boundary conditions (bathymetry, continental configuration, greenhouse gas concentrations, solar radiation, orbital parameters) were kept consistent with modern conditions, except for the topography. The model outputs were then analyzed to assess changes in AMOC, PMOC, GMOC, sea surface density (SSD), SSS, Ekman pumping, wind stress, water mass transport, sea ice extent, and mixed layer depth (MLD). Statistical significance was assessed using student-t tests.

The sequential addition of mountains in the "Flat2Real" experiment showed that the TP uplift was pivotal in initiating the AMOC and shutting down the PMOC. Before the TP uplift, a weak AMOC and a strong PMOC were present, consistent with previous findings. Adding the TP resulted in a rapid increase in AMOC strength and a collapse of the PMOC. None of the other individual mountain uplifts could initiate the AMOC, while the TP alone caused the PMOC collapse. However, the AMOC only reached its realistic strength when coupled with the Antarctic continent. Experiments with individual mountains showed that the AT resulted in a stronger PMOC compared to the "Flat" scenario, while the RM had negligible effects. Combining the TP with the AT was the minimal topographic requirement for a fully developed AMOC similar to the "Real" scenario. The analysis of MOC patterns, SSD, SSS, and Ekman pumping revealed that the TP altered the global hydrological cycle, leading to increased salinity in the North Atlantic and decreased salinity in the North Pacific. This shift, coupled with enhanced Ekman pumping due to the AT, facilitated the NADW formation and the AMOC's establishment. Changes in atmospheric circulation and moisture transport associated with the TP uplift played crucial roles in altering the surface buoyancy and triggering the shutdown of the PMOC and the development of AMOC. A positive feedback loop was observed between AMOC strengthening and sea ice retreat in the subpolar Atlantic, accelerating the AMOC establishment. The Antarctic topography's role in enhancing Ekman pumping in the Southern Ocean was crucial, driving a stronger Deacon cell and contributing to the overall GMOC.

The findings demonstrate the profound influence of continental topography, particularly the TP and AT, on the GMOC. The TP acted as a freshwater attractor in the NH, while the AT drove strong Ekman pumping in the Southern Ocean. The TP's impact on the global hydrological cycle was key in initiating the AMOC and inhibiting the PMOC. The positive feedback between AMOC and sea ice changes also highlights the complex interactions within the climate system. This study underscores the importance of considering topographic effects when studying past and future climate changes, particularly in the context of the AMOC’s stability and the global ocean conveyor belt.

This study shows the critical roles of the Tibetan Plateau and Antarctic continent in shaping the modern-day global meridional overturning circulation. The Tibetan Plateau's influence on the global hydrological cycle, coupled with the Antarctic's contribution to Ekman pumping, was crucial in establishing the Atlantic MOC and shutting down the Pacific MOC. While the model reveals insights into the interactions of topography and ocean circulation, future research should incorporate factors like continental drift, changes in oceanic gateways, dynamic treatment of ice sheets and the carbon cycle for a more comprehensive understanding of long-term climate change.

The study's conclusions may have limitations due to the model resolution and the simplified treatment of certain factors. The model resolution may not perfectly capture the effects of complex topography, and the lack of dynamic treatment of continental drift, oceanic gateway changes, and ice sheets might influence results. The modern land-sea mask and ocean gateways might exaggerate the TP's impact on the GMOC. Additionally, the study's use of pre-industrial CO2 levels and the exclusion of chemical erosion effects on atmospheric CO2 could limit the generalizability of the results. Future studies incorporating more sophisticated models and considering the various factors involved are needed.