Holocene Labrador Current strength reconstructed from sediment sortable silt proxy

English Abstract:

Variability in the Atlantic meridional overturning circulation results from changes in the sea-surface conditions. During the early to mid-Holocene period, weakening of deep water formation in the North Atlantic has been partly attributed to the fluctuation of freshwater discharge to the Labrador Sea but direct evidence for this linkage is lacking. Furthermore, a recent study has shown no evidence of only temperature and salinity link between Labrador and Nordic seas for the mid to late Holocene period, implying that the subpolar gyres were not influenced by the Labrador Sea input but rather by a decrease in wind stress (Thornalley et al., 2009). Previous studies have been hampered by the lack of a high-resolution proxy for the flux of cold, low saline water from the Labrador Current. Here we present such a high resolution sediment proxy for variations in the strength of the Labrador Current for the Holocene from two northwestern Labrador Shelf cores. We compare variations in paleo-current strength inferred from the sortable-silt in conjunction with other paleoclimate proxies. Our data suggest that there were two prominent weakening of the Labrador Current strength centered at 8.4 ka and 7.8 ka which were accompanied by high detrital carbonate and notable changes in sediment geochemical tracers. The Labrador Current was strong between 7.6 ka and 3.5 ka with minor fluctuation in its strength. Furthermore, an enigmatic increase in the Labrador Current strength, which was not accompanied by any changes in the sediment geochemistry, was also observed at 3.4 ka. High sortable-silt values can result either by reworking off the adjacent sediments on the shallow shelf or by a very strong Labrador Current. If there was a large sediment input from the Labrador Shelf, the concentration of ice-rafted detritus would be diluted, which is not the case as the concentration of ice-rafted detritus increases from ~5.4 ka to the present. We therefore suggest that the sortable-silt data appear to represent variations in the Labrador Current strength.