SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
ts01.docWord document61KOrbital parameters and greenhouse gas concentrations of the simulations for PI, 32 ka BP and LGM.
fs01.tifTIFF image1053KTemporal changes of the Atlantic Meridional Overturning Circulation (AMOC) in the different hosing experiments. The state before applying freshwater forcing is named by the ‘CLTM’ (i.e. climatology mean), and the ‘FWP’ (i.e. fresh water perturbation) stands for the mean state of the last 10 years of freshening. Then, the ‘RFWP’ (i.e. recovered from freshwater perturbation) is used to present the state with AMOC overshoot during the LGM and 32 ka BP, or the mean state between 380 and 400 year during the PI, during which the AMOC has completely recovered. Using the same version of COSMOS model, Zhang et al. [2012] recently has done a series of experiments to explain that the similarity of AMOC strength between the LGM and PI experiments are determined by the conditions of salinity stratifications in the deep Atlantic Ocean. Furthermore, as shown in Figure S5, there is a stronger heat losing from ocean to atmosphere under 32 ka BP conditions, which thermodynamically enhances surface ocean water sinking into deeper layer, and ultimately induces a stronger AMOC strength than in the LGM experiment.
fs02.tifTIFF image373KThe annul mean mixed layer depth (AMMLD) in the North Atlantic before applying the freshwater perturbation (left column, 21-100 model year mean), and twenty years after the end of freshwater perturbation (right column, 265-275 model year mean) under the conditions of (a, b) PI, (c, d) 32 ka BP and (e, f) LGM. In panel (e), three boxes indicate the areas of deep-water formation in the South Labrador Sea (red), the GIN Sea (black) and the South-Icelandic Sea (yellow), respectively.
fs03.tifTIFF image903KThe winter mean mixed layer depth in the North Atlantic before applying the freshwater perturbation and twenty years after the end of freshwater perturbation under the conditions of PI, 32 ka BP and LGM. The subfigures are shown in accordance with the order of Figure S2.
fs04.tifTIFF image313KThe anomaly of sea surface temperature between the states of ‘RFWP’ and ‘FWP’. From top to bottom, it is shown for the experiments PI, 32 ka BP and LGM, respectively.
fs05.tifTIFF image775KThe anomaly of temperature between different states in the freshening experiments. The left column shows the difference between the states of ‘CLTM’ and ‘FWP’, and the right exhibits the anomaly between the states of ‘RFWP’ and ‘FWP’. From top to bottom, it is shown for the experiments PI, 32 ka BP and LGM, respectively.
fs06.tifTIFF image613KThe anomaly of salinity between different states in the freshening experiments, which are shown in accordance with the order of Figure S5.
fs07.tifTIFF image98KThe anomaly of annual mean total heat flux between 32 ka BP and LGM experiment. The bluish colors indicate that there is more heat loss from ocean to atmosphere under 32 ka BP conditions, relative to LGM. In addition, the lines exhibit 15% sea ice compactness lines during 32 ka BP (black) and LGM (yellow), respectively (winter: thick, summer: thin). The region in grey means the land in the LGM run.
fs08.tifTIFF image86KTime series of the AMMLD in the South Labrador Sea in the experiments of (a) PI, (b) 32 ka and (c) LGM. The blue and orange columns indicate the time phases of AMOC overshoot in the experiment of 32 ka BP and LGM, respectively.
Readme.docWord document32KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.