Verdy, Ariane: Scripps Institution of Oceanography (Project Lead)
Key, Robert (Bob): Princeton University (Co-Investigator)
Mazloff, Matthew: Scripps Institution of Oceanography (Co-Investigator)
Sarmiento, Jorge: Princeton University (Co-Investigator)
Carter, Brendan: NOAA (Participant)
Project Funding:
2012 - 2014
NRA: 2011 NASA: Carbon Monitoring System
Funded by NASA
Abstract:
The challenge- Any Carbon Monitoring System (CMS) must account for fluxes of carbon between the atmosphere and the oceans, the world s largest reservoir of carbon dioxide (CO2). Currently, air-sea CO2 flux estimates are produced by sophisticated physical-biogeochemical models. However, these models still fail to represent significant patterns in the observed fluxes, and these discrepancies are thought to be largely due to errors in the simulation of biogeochemical processes.
Our goal- This proposal capitalizes on two recent developments in oceanography to lay the groundwork for a global ocean CMS with improved biogeochemistry. Satellite measurements of the surface ocean and sensor-based measurements of the interior ocean are rapidly increasing the temporal and spatial coverage of biogeochemical data. Simultaneously, the development of four-dimensional variational assimilation (4D-Var) modeling has combined the forward modeling and traditional static inversion approaches to overcome the primary limitations of both: forward models estimate what could have happened in the ocean rather than what actually happened, and inversions cannot yield predictions. The 4D-Var approach automates the process of adjusting initial conditions and model parameters to produce an optimal fit of the model to physical constraints and all available observations. Our vision is of a state-of-the-art global physical-biogeochemical ocean model that incorporates data from the growing global network of satellites, sensors, and shipboard measurements to improve its estimates of air-sea CO2 fluxes.
Our contribution- We will provide the missing components for 4D-Var physical-biogeochemical assimilation. As we build toward our goal of a global model-observation synthesis, each step of the proposed research will generate independently valuable scientific products:
1. We will test the efficacy of extending the 4D-Var approach to biogeochemistry by using it to optimize both the idealized biogeochemical and physical state of an eddy-resolving model of the California Current Ecosystem (CCE) for 2007-2011. The model will be optimized by adjusting the initial conditions, boundary conditions, external forcing, and parameter values to reduce the misfit between the model and the dense and diverse observations (including in situ measurements of carbon, oxygen, phosphate, pH, and alkalinity) available of the CCE during this time period.
2. We will further develop the biogeochemical component of the model to allow assimilation of satellite-based chlorophyll estimates and to improve the representation of other constraints, and optimize this new implementation of the physical-biogeochemical model to improve our estimate of air-sea CO2 fluxes in the CCE.
3. We will extend the data-processing of hydrographic observations to produce a self-consistent dataset of the quality, richness of properties, and temporal extent that will be required to constrain a global 4D-Var biogeochemical model. GLODAPv2 (GLobal Ocean Data Analysis Project version 2) will be a calibrated unification of existing biogeochemical data products and new data over the period 1972-2011.
As more observations become available, state estimation is undoubtedly the way forward for addressing the objectives of NASA's CMS by bringing together observations and modeling tools to generate accurate high-resolution and time-varying maps of air-sea CO2 fluxes. Together, the development of 4D-Var methods and the observational dataset will enable global model-observation syntheses of the ocean carbon cycle over climate-relevant time scales.
Publications:
Verdy, A., Mazloff, M. R., Cornuelle, B. D., Kim, S. Y. 2014. Wind-Driven Sea Level Variability on the California Coast: An Adjoint Sensitivity Analysis. Journal of Physical Oceanography. 44(1), 297-318. DOI: 10.1175/JPO-D-13-018.1
More details may be found in the following project profile(s):
