A New Method to Produce Sea Surface Temperature Using Satellite Data Assimilation into an Atmosphere–Ocean Mixed Layer Coupled Model

Abstract

new method of producing sea surface temperature (SST) data for numerical weather prediction is suggested, which is obtained from the assimilation of satellite-derived SST into an atmosphere?ocean mixed layer coupled model. The Weather Research and Forecasting (WRF) Model and the Noh mixed layer model are used for the atmosphere and ocean mixed layer models, respectively. Data assimilation (DA) is carried out in two steps, based on the estimation from the covariance matching method that the daily mean SST of satellite data is more accurate than the model data, if the number of data in a grid per day is sufficiently large?that is, the daily mean SST bias correction in the first DA and the sequential SST anomaly correction in the second DA. For the second DA, the model restarts from the initial condition corrected by the first DA, and DA is applied every 30 min using the nudging method. The daily mean and the diurnal variation of satellite SST are assimilated to the bulk and skin SST, respectively. The modeled results with the new data assimilation scheme are validated by statistical comparison with independent satellite and buoy data such as correlation coefficient, root-mean-square difference, and bias. Furthermore, the sensitivity and seasonal variation of the weighting factor in the second DA are examined. The new approach illustrates the possibility of applying the atmosphere?ocean mixed layer coupled model for the production of SST data combined with the assimilation of satellite data.