Does Southern Ocean Surface Forcing Shape the Global Ocean Overturning Circulation?

Abstract

Paleoclimate proxy data suggest that the Atlantic Meridional Overturning Circulation (AMOC) was shallower at the Last Glacial Maximum (LGM) than its preindustrial (PI) depth. Previous studies have suggested that this shoaling necessarily accompanies Antarctic sea ice expansion at the LGM. Here the influence of Southern Ocean surface forcing on the AMOC depth is investigated using ocean‐only simulations from a state‐of‐the‐art climate model with surface forcing specified from the output of previous coupled PI and LGM simulations. In contrast to previous expectations, we find that applying LGM surface forcing in the Southern Ocean and PI surface forcing elsewhere causes the AMOC to shoal only about half as much as when LGM surface forcing is applied globally. We show that this occurs because diapycnal mixing renders the Southern Ocean overturning circulation more diabatic than previously assumed, which diminishes the influence of Southern Ocean surface buoyancy forcing on the depth of the AMOC.

Linking Glacial‐Interglacial States to Multiple Equilibria of Climate

“…The reorganization of the global overturning is consistent with conceptual models highlighting the role of the SH surface buoyancy fluxes in controlling the global MOC, and the depth of the interface between the upper and lower overturning cells (Burke et al, ; Ferrari et al, ; Marzocchi & Jansen, ; Sun et al, ; Watson et al, ). In steady state, poleward (equatorward) flowing surface waters must lose (gain) buoyancy to (from) the atmosphere and sea ice (Marshall, ; Marshall & Radko, ).…”

Reassessing the Role of the Indo‐Pacific in the Ocean's Global Overturning Circulation

“…However, abyssal mixing is considered the singular influential Indo-Pacific process in many idealized ocean models, while Indo-Pacific surface fluxes and shallow ocean dynamics are expected to adjust passively to the requirements of abyssal dynamics (e.g., Ferrari et al, 2014;Nikurashin & Vallis, 2011Radko & Kamenkovich, 2011;Thompson et al, 2016). As a consequence, significant emphasis has been placed on the role of Southern Ocean surface forcing in mediating the overturning circulation to its north (e.g., Ferrari et al, 2014;Sun et al, 2018;Thompson et al, 2016), particularly so in frameworks that consider the Southern Ocean and global circulation to be zonally symmetric (Bell, 2015;Gnanadesikan, 1999;Jansen, 2017;Jansen & Nadeau, 2016;Klinger & Marotzke, 1999;Marotzke & Klinger, 2000;Nikurashin & Vallis, 2011Radko & Kamenkovich, 2011;Samelson, 2009;Shakespeare & Hogg, 2012;Wolfe & Cessi, 2011).…”

“…The three-dimensional complexity of the GOC has been appreciated for decades; however, its governing dynamics have largely been explored in simplified ocean models (e.g., Gnanadesikan, 1999;Munk, 1966;Stommel, 1961, and many others). In general, such frameworks strive to remain simple enough that the response to a perturbation can be mechanistically understood while still including all fundamental ocean processes.…”

Transient Overturning Compensation between Atlantic and Indo-Pacific Basins

“…The meridional overturning circulation streamfunction, defined as a function of latitude and buoyancy, is calculated as where b is buoyancy [b 5 a(u 2 u ref )] with a the thermal expansion coefficient of seawater and u ref 5 08C, t 1 and t 2 define the averaging period, z bot represents the ocean bottom, y r is the total meridional velocity that includes both the Eulerian-mean flow (y) and the eddy bolus contribution due to the parameterized eddies, b 0 is the buoyancy field calculated by the model at each location and time step, and H (b 2 b 0 ) is the Heaviside step function of b 2 b 0 such that c i (y, b) represents the northward transport above the isopycnal denoted by b. The streamfunction c i (y, b) is mapped to depth coordinates using the mean depth of each isopycnal (e.g., Sun et al 2018) (Fig. 4).…”