ABSTRACT

Ventilation rates quantify how the ocean communicates with the rest of the climate system on timescales ranging from month to millennia, and determine the ocean's ability to buffer the atmosphere from climate anomalies and to take up atmospheric trace gases, including anthropogenic carbon dioxide. Based on recent research that surface fluxes of newly-ventilated water must be partitioned according to the water's residence time in the ocean interior in order to describe correctly how inventories of transient tracers evolve with time, ventilation rate is therefore a distribution and not, as had been thought sufficient until now, a single number representing a bulk flux. This fundamentally changes the estimation problem to a deconvolution for a ventilation-rate distribution that propagates known mixed-layer concentrations to measured interior values.

Oceanographers from Columbia University and University of California at Irvine propose to use CFC, tritium, and radiocarbon transient tracer data, together with gridded temperature, salinity, nutrient, and oxygen data, to estimate the ventilation-rate distributions of the ocean. The ventilation-rate distributions will systematically be estimated for a range of density classes whose outcrops cover most of the global ocean surface. A major part of the proposed work will be a rigorous quantification of the uncertainties due to errors in the data and due to the underdetermined nature of the deconvolutions.

Two deconvolution methods will be employed: A tested parametric approach and a novel application
of the maximum-entropy method. The deconvolution of the ventilation-rate distribution will be constrained on decadal timescales by CFCs, tritium and bomb radiocarbon and by the background radiocarbon on longer timescales. Steady tracers will constrain the ventilation-rate distribution's spatial and seasonal dependence. The maximum-entropy inversions will use a state-of-the-art data-assimilation model to produce a prior guess for the ventilation-rate distribution. This model will also be used to generate realistic synthetic tracer data to quantify the systematic errors of both the parametric and the maximum-entropy deconvolutions. The research will provide a novel comprehensive picture of how the ocean communicates with the atmosphere on timescales of months to millennia and help reconcile disparate previous estimates of ventilation based on the incomplete bulk-flux picture.

The proposed research will provide the first global estimate of the ventilation-rate distribution for the current state of the ocean and a baseline estimate of its uncertainty so that future estimates of variability and climate change in ventilation can meaningfully be assessed. The proposed work will be synergistic with climate research on constraining the oceanic uptake of anthropogenic carbon. A MATLAB toolbox for performing generalized water-mass analysis using the maximum entropy method will be made available to the community. In addition to the primary scientific contributions, the proposed work will provide funding to support the career development of a postdoctoral scholar, and education and training for graduate students in ocean transport diagnostics, data analysis techniques, ocean modeling, and in analyzing the ocean's role in the climate system.

Please report errors in award information by writing to: awardsearch@nsf.gov.