Acidification and transports of water masses and CO2 in the North Atlantic

Abstract

ABSTRACT: The rise in the atmospheric CO2 levels due to human activities (CANT) is softened by its oceanic uptake. But this absorption leads to a suite of chemical changes collectively known as ocean acidification. Although acidification occurs in the world ocean, its impacts tend to be stronger in the high latitude oceans. Moreover, in some regions where vertical movements are relatively fast, i.e., in regions of water mass formation such as the Subpolar North Atlantic, the timescale for deep penetration of CANT is on the order of decades, thus being faster exposed to the acidification effects. This thesis focuses on the acidification and transports of water masses and CO2 in the North Atlantic Subpolar Gyre (NASPG). To determine the effect of the circulation changes in the oceanic uptake and storage of CANT, the water mass distribution, transport and transformation in the NASPG are discussed for the first decade of the 2000s (2002-2010), as well as the inter-annual variability of the water mass structure from 1997 to 2010. The reduction of the magnitude of the upper limb of the Atlantic Meridional Overturning Circulation (AMOC) between 1997 and the 2000s is associated with the reduction in the northward transport of the Central Waters. This reduction is partially compensated by the reduction of the southward flow of the lower limb of the AMOC, associated with the decrease in the transports of the Polar Intermediate Water and the Subpolar Mode Water in the Irminger Basin. The box model analysis revealed that the Central Waters, Labrador Sea Water, Subarctic Intermediate Water and Iceland-Scotland Overflow Water from the East North Atlantic Basin cross over the Reykjanes Ridge and enter the Irminger Basin, where they are transformed and/or densified, passing from the upper and intermediate water domains to the deep water domain. The changes in CANT, pH, total alkalinity (AT) and aragonite saturation were evaluated in the main water masses of the Irminger and Iceland Basins for the period 1981-2014. The CANT uptake in both basins led to significant acidification rates in the whole water column, which drive the shoaling of the aragonite saturation horizon at about 10 m/yr. By separating the observed pH changes into an anthropogenic (derived from the CANT uptake, DpHCant) and a non-anthropogenic component (not directly related to the CANT uptake, DpHVar), an attribution to the underlying drivers is provided. At steady state, DpHVar would be constant and all the pH changes would be explained through DpHCant. However, in the upper layers the DpHVar action (driven by the advection of subtropical waters) counteracts the effect of DpHCant, while in the intermediate layers of the Irminger Basin, the effect of DpHVar (derived from the aging of the waters) reinforces the acidification derived from DpHCant. Based on the rates of changes of pH and aragonite saturation, a decrease in 0.31 pH units is inferred for the surface layers by 2065, at which time the entire water column of the Irminger and Iceland Basins will be undersaturated in aragonite. The data also showed significant increasing trends of alkalinity in the deep waters of the Irminger Basin that may be related to the increase in the discharge of the Arctic rivers. Since lateral advection of CANT from middle to high latitudes provides the main supply of CANT to the NASPG, knowing the way this CANT is transported is a crucial issue for understanding how the ocean is storing CANT. In this thesis the inter-annual to decadal variability in the transport of CANT (Tcant) across the Subpolar North Atlantic is investigated for the period 1997-2010. The Tcant was decomposed into its diapycnal and isopycnal components, being the former the main driver of the variability of the Tcant. The CANT concentration plays an important role in both components: the horizontal gradient of CANT is responsible for its isopycnic southward transport, mainly in the intermediate and deep waters of the Irminger Basin; while the CANT-laden waters flowing northwards are responsible for the large diapycnic northward transport. At inter-annual to decadal timescales, the variability of the AMOC dominates the Tcant variability, but the CANT increase seems to control the Tcant change on longer timescales, and it is very likely to cause an increase in the Tcant across the Subpolar North Atlantic.