Dataset: OGCM/OCMIP-Sarmiento
Deployment: USJGOFS_SMP

A major achievement of the last decade of oceanic observations by the Joint Global Ocean Flux Study (JGOFS) and World Ocean Circulation Experiment (WOCE) has been the dramatic improvement in our ability to use observations to estimate the oceanic inventory of anthropogenic carbon. This improvement has been driven both by increased data and by the development of improved techniques for analyzing the observations. An important role has been played by the rapid development of ocean general circulation models that are now being used routinely to do tracer simulations and provide estimates of anthropogenic carbon uptake. However, despite the great progress that has been made, there remain many important issues that need to be resolved. The Ocean Carbon Model Intercomparison Project (OCMIP) shows very large differences between model simulations, particularly in the deep ocean tracer distributions, in the long term behavior of the models with respect to uptake of anthropogenic carbon, and in tracer-tracer relationships. Furthermore, there are very large disagreements with observations, particularly in the deep ocean. Some of the data analysis methods are based on assumptions that are difficult to verify and whose uncertainty is hard to estimate.

We propose to carry out model sensitivity studies to examine several hypotheses for model differences. We further propose to make use of measurements from the JGOFS and WOCE surveys to determine which model gives a more realistic simulation, and to improve our estimates of the oceanic uptake and storage of anthropogenic CO2, bomb radiocarbon, and CFCs. Our goal is to advance both the development of reliable ocean models and the estimation of oceanic uptake of anthropogenic CO2 from observations with the aim of improving estimates of the oceanic carbon sink and our understanding of the ocean carbon cycle.
First, we will examine model simulated tracer-tracer relationships of anthropogenic CO2, bomb ¹⁴C, and CFCs to evaluate four specific aspects of model physics that we believe are important in determining tracer distributions. They are;

1. the relative magnitude of advective return flows of the North Atlantic Deep Water through the thermocline and the Southern Ocean;
2. the magnitude of convective exchange between surface and deep waters in the Southern Ocean;
3. bottom boundary layer advective pathways of newly formed deep waters; and
4. the magnitude and spatial distribution air-sea gas exchange.

We will use the Princeton biogeochemical model to conduct a suite of simulations that isolate these mechanisms. Furthermore, model simulations will be compared with appropriate data from the JGOFS and WOCE surveys.

Second, we will improve the estimates of anthropogenic CO2 inventories by making use of the model simulations in conjunction with observations to critically examine the C* methodology of separating the anthropogenic CO2 from the background total CO2 [Gruber et al., 1996] and by exploring alternative methods using the distributions of CFCs, bomb ¹⁴C, and CCl4. The potential to use these measurements to estimate anthropogenic CO2 inventory rests on theoretical arguments and supporting preliminary observations that these transient tracers would be good analogs of anthropogenic CO2 distributions in the ocean, given the similarities in their growth history of CFCs and CCl4 in the atmosphere and the time scale since the injection of bomb ¹⁴C into the ocean. The robustness of the relationship of anthropogenic CO2 with CFCs, bomb ¹⁴C, and CCl4 will be investigated using our new simulations in conjunction with simulations from the OCMIP. We will then use the survey tracer data to improve estimates of the oceanic uptake of anthropogenic carbon.

We have gathered a team that has expertise in ocean observations and ocean biogeochemical modeling. A coordinated examination of these tracers that historically have been examined in relative isolation promises a better understanding anthropogenic CO2, bomb ¹⁴C, CFCs, and CCl4 uptake in ocean models and in the real ocean.