A comprehensive analysis of the U.S. west coast ocean circulation and California Current System (CCS) will be performed using the Regional Ocean Modeling System (ROMS) and 4-Dimensional Variational (4D-Var) data assimilation. Ocean observations from the period 1981-2011 will be assimilated into ROMS, in conjunction with the best available estimates of ocean surface forcing and open boundary conditions. The goals of this research project are: (1) To perform retrospective analyses of the ocean circulation during the last 3 decades; (2) Use the analyses to document and understand changes in spring-summer stratification and how these changes, along with variations in wind-stress and wind-stress curl, influence upwelling intensity and upwelling source water depth; (3) Identify the dominant space-time modes of circulation variability on intraseasonal-to-interannual and decadal timescales and how these modes are related to climate modes such as the Pacific decadal Oscillation and North Pacific Gyre Oscillation; (4) Quantify the impact on the analyses of the in situ and satellite observing networks; (5) Make the circulation analyses available via a data server to the community at large for scientific and marine resource management applications.
Intellectual merit This research will yield the first analyses of the California Current System circulation generated by a regional ocean model forced by the best available ocean surface fluxes from regional atmospheric models. Also, this assimilative ocean model is unique in that it provides formal estimates of the expected error in the analyses, and provides a quantitative assessment of the impact of each individual observation on any aspect of the resulting circulation estimates. Analysis error estimates are critically important for establishing confidence in the analyses, and observation impact and observation sensitivity studies can be used in the future to design optimal observing arrays. A by-product of the proposed analyses are fields of surface forcing and open boundary condition corrections that provide information about potential errors in these fields and errors in the model. The investigators are a uniquely qualified and experienced team of ocean modelers, all with considerable experience in both modeling the CCS and 4D-Var data assimilation, and the lead investigator is one of the lead developers of the ROMS 4D-Var system.
Broader impacts The proposed retrospective ocean analyses are significant for a number of reasons: (i) They can be used to quantify variability on weekly to decadal timescales, and trends in the dominant circulation features, particularly in subsurface regions that are not well observed; (ii) They provide complete realizations of the circulation at high temporal resolution on the full grid of the model that are largely free of the sampling and aliasing issues that often plague observations; (iii) They can be used to test hypotheses about ocean dynamics, climate variability, and climate change. The resulting analyses will form a valuable community resource in the form of a data archive that can be used for numerous applications, including: process studies of the CCS; input for driving offline ecosystem and individual based models; a source of open boundary conditions for higher resolution models nested within the native ROMS CCS grid; and initial conditions for the development of ocean hindcasting/forecasting systems, to name but a few. This project will also provide training for a graduate student and post-doctoral researchers in advanced 4D-Var data assimilation methods, since it is important that we continue to train the next generation of ocean data assimilators. The proposed research will also use the adjoint of 4D-Var which is an original and transformative aspect of this work. This project will also complement the NSF Ocean Observatories Initiative, and will support synthesis activities, development of user products, and array assessment within the U.S. Integrated Ocean Observing Systems along the U.S. west coast.
