Transport across 48°N in the Atlantic Ocean

Abstract

Transports across 48°N in the Atlantic Ocean are estimated from five repeat World Ocean Circulation Experiment (WOCE) hydrographic lines collected in this region in 1993?2000, from time-varying air?sea heat and freshwater fluxes north of 48°N, and from a synthesis of these two data sources using inverse box model methods. Results from hydrography and air?sea fluxes treated separately are analogous to recently published transport variation studies and demonstrate the sensitivity of the results to either the choice of reference level and reference velocities for thermal wind calculations or the specific flux dataset chosen. In addition, flux-based calculations do not include the effects of subsurface mixing on overturning and transports of specific water masses. The inverse model approach was used to find unknown depth-independent velocities, interior diapycnal fluxes, and adjustments to air?sea fluxes subject to various constraints on the system. Various model choices were made to focus on annually averaged results, as opposed to instantaneous values during the occupation of the hydrographic lines. The results reflect the constraints and choices made in the construction of the model. The inverse model solutions show only marginal, not significantly different temporal changes in the net overturning cell strength and heat transport across 48°N. These small changes are similar to seasonally or annually averaged numerical model simulations of overturning. Significant variability is found for deep transports and air?sea flux quantities in density layers. Put another way, if one ignores the details of layer exchanges, the model can be constrained to produce the same net overturning for each repeat line; however, constraining individual layers to have the same transport for each line fails. Diapycnal fluxes are found to be important in the mean but are relatively constant from one repeat line to the next. Mean air?sea fluxes are modified slightly but are still essentially consistent with either the NCEP data or the National Oceanography Centre, Southampton (NOC) Comprehensive Ocean?Atmosphere Data Set (COADS) within error. Modest reductions in air?sea flux uncertainties would give these constraints a much greater impact. Direct transport estimates over broader regions than the western boundary North Atlantic Current are needed to help resolve circulation structure that is important for variability in net overturning.