An historical analysis of the California Current circulation using ROMS 4D-Var: System configuration and diagnostics

• Published in: Ocean modelling (Oxford) Vol. 99; pp. 133 - 151
• Main Authors: Neveu, Emilie, Moore, Andrew M., Edwards, Christopher A., Fiechter, Jérôme, Drake, Patrick, Crawford, William J., Jacox, Michael G., Nuss, Emma
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2016
• Subjects: 4D-Var; Atmospherics; California Current; Circulation; Cost function; Data assimilation; Historic; Marine; Oceans; Sequences; Weather forecasting; INE, Pacific, California Current; California Current; 4D-Var; Data assimilation
• DOI: 10.1016/j.ocemod.2015.11.012
• ISSN: 1463-5003

•Two overlapping sequences of California Current circulation estimates were computed.•All available ocean observations were assimilated into ROMS using 4D-Var.•Data assimilation substantially alters the circulation.•A quantitative assessment of the efficacy of the data assimilation system was performed.•The analyses are a useful resource available to the community. The Regional Ocean Modeling System (ROMS) 4-dimensional variational (4D-Var) data assimilation tool has been used to compute two sequences of circulation analyses for the U.S. west coast. One sequence of analyses spans the period 1980–2010 and is subject to surface forcing derived from relatively low resolution atmospheric products from the Cross-Calibrated Multi-Platform wind product (CCMP) and the European Centre for Medium Range Weather Forecasts (ECMWF) reanalysis project. The second sequence spans the shorter period 1999–2012 and is subject to forcing derived from a high resolution product from the Naval Research Laboratory Coupled Ocean Atmosphere Mesoscale Prediction System (COAMPS). The two analysis periods are divided into eight day windows, and all available satellite observations of sea surface temperature and sea surface height, as well as in situhydrographic profiles are assimilated into ROMS using 4D-Var. The performance of the system is monitored in terms of the cost function and the statistics of the innovations, and the impact of data assimilated on the circulation is assessed by comparing the posterior circulation estimates with the prior circulation and the circulation from a run of the model without data assimilation, with particular emphasis on eddy kinetic energy. This is part I of a two part series, and the circulation variability of the 4D-Var analyses will be documented in part II.