Comparison of Regional Performance between EnKF and EnOI in the North Pacific

Seung-tae Lee, Seoul National University, School of Earth and Environmental Sciences, Seoul, Korea, Republic of (South), Yang-Ki Cho, Seoul National University, School of Earth and Environmental Sciences, Seoul, South Korea, Jihun Jung, Seoul National University, School of Earth and Environmental sciences, Seoul, Korea, Republic of (South), Byoung-Ju Choi, Chonnam National University, Department of Oceanography, South Korea, Young Ho Kim, Pukyong National University, Busan, South Korea and Sangil Kim, Pusan National University, South Korea
Abstract:
The North Pacific is a crucial region in decadal climate changes like the El-Ninõ Southern Oscillation (ENSO) and the Pacific decadal oscillation (PDO), which influence critically not only oceans but also atmosphere. Despite of the importance of ocean data in the North Pacific, available data have been rare due to limitation of accessibility. Data assimilation have been used to produce more available ocean data. Because the characteristics of currents and water masses in the North Pacific show distinctively spatial difference, it should select appropriate data assimilation method according to the regional characteristics based on the understanding each data assimilation method. In this study, we compared the regional performance difference between Ensemble Optimal Interpolation (EnOI), a simplified version of Ensemble Kalman Filter (EnKF), and EnKF in each region. Assimilation of sea surface temperature for 2015 using EnKF and EnOI methods with satellite data from the Operational Sea Surface Temperature and Sea Ice Analysis(OSTIA) was applied respectively. The RMSE calculated with satellite Sea Surface Temperature(SST) and Sea Surface Height(SSH) of EnKF was lower than that of EnOI in the Northwest Pacific where the spatial SST and u-v velocity correlation structure was complex. But, the RMSE of EnOI was lower than EnKF in the Equator area with simple SST and u-v correlation structure. The efficiency of each method in terms of accuracy, resources and time was also compared. EnOI was faster by about 7 times than EnKF.
Back to: Advances in Ocean Data Assimilation, Forecasting, and Reanalysis V Posters