The sensitivity of the climate response to the magnitude and location of freshwater forcing: last glacial maximum experiments

Abstract

Proxy records indicate that the locations and magnitudes of freshwater forcing to the Atlantic Ocean basin as iceberg discharges into the high-latitude North Atlantic, Laurentide meltwater input to the Gulf of Mexico, or meltwater diversion to the North Atlantic via the St. Lawrence River and other eastern outlets may have influenced the North Atlantic thermohaline circulation and global climate. We have performed Last Glacial Maximum (LGM) simulations with the NCAR Community Climate System Model (CCSM3) in which the magnitude of the freshwater forcing has been varied from 0.1 to 1 Sv and inserted either into the subpolar North Atlantic Ocean or the Gulf of Mexico. In these glacial freshening experiments, the less dense freshwater provides a lid on the ocean water below, suppressing ocean convection and interaction with the atmosphere above and reducing the Atlantic Meridional Overturning Circulation (AMOC). This is the case whether the freshwater is added directly to the area of convection south of Greenland or transported there by the subtropical and subpolar gyres when added to the Gulf of Mexico. The AMOC reduction is less for the smaller freshwater forcings, but is not linear with the size of the freshwater perturbation. The recovery of the AMOC from a “slow” state is ∼200 years for the 0.1 Sv experiment and ∼500 years for the 1 Sv experiment. For glacial climates, with large Northern Hemisphere ice sheets and reduced greenhouse gases, the cold subpolar North Atlantic is primed to respond rapidly and dramatically to freshwater that is either directly dumped into this region or after being advected from the Gulf of Mexico. Greenland temperatures cool by 6–8 °C in all the experiments, with little sensitivity to the magnitude, location or duration of the freshwater forcing, but exhibiting large seasonality. Sea ice is important for explaining the responses. The Northern Hemisphere high latitudes are slow to recover. Antarctica and the Southern Ocean show a bipolar response, with warming and reduced sea ice. This warming continues after the cessation of the freshwater forcing and shows a dependence on the duration of the freshwater forcing. Equatorward of the expanded sea ice, the simulated temperature and salinity anomalies are sensitive to the amount of colder and fresher waters that are advected out of the subpolar North Atlantic. In the tropical Atlantic, the recovery of the Intertropical Convergence Zone (ITCZ) from its more southerly position during the freshwater forcing is much more rapid than the recovery of the AMOC, and is more related to the recovery of low-latitude surface temperatures than Greenland temperature or sea ice. These results have implications for using proxy records as indirect measures of the AMOC.