Applying the local ensemble transform Kalman filter to the Nonhydrostatic Icosahedral Atmospheric Model (NICAM)

Abstract

In this study, we apply the local ensemble transform Kalman filter (LETKF) to the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) to develop the NICAMLETKF. In addition, an algorithm to adaptively estimate the inflation parameter and the observational errors is introduced to the LETKF. The feasibility and stability of the NICAM-LETKF are investigated under the perfect model scenario. According to the results, we confirm that the converged analysis errors of the NICAM-LETKF are smaller than the observational errors, and the magnitude and distribution of the root mean square errors (RMSEs) are comparable to those of the ensemble spreads. In our experiments, we find that the inflation parameter is optimally tuned and the observational errors are close to the true value. It is concluded that the NICAM-LETKF works appropriately and stably under the perfect model scenario even if the inflation parameter and the observational errors are adaptively estimated within the LETKF.