Abstract:
Successful climate prediction at seasonal-to-interannual time scales may depend on the optimal initialization of the land surface states, in particular soil moisture. Such optimal initialization can be achieved by assimilating soil moisture observations into the land model prior to the forecast. We assess the performance of the Extended Kalman filter (EKF) and the Ensemble Kalman filter (EnKF) for soil moisture estimation when used with the Catchment Land Surface Model (CLSM) of the NASA Seasonal-to-Interannual Prediction Project.

Overall, we find that the EKF and the EnKF are able to derive satisfactory estimates of soil moisture. In the case of the EnKF, just four ensemble members prove sufficient. The EKF and the EnKF (with four ensemble members) show comparable performance for comparable computational effort. For five or more ensemble members, the EnKF outperforms the EKF, albeit at greater computational expense. This is attributed to the EnKF's flexibility in representing non-additive model errors such as errors in certain forcing variables or errors in model parameters.

In summary we can say that the EnKF is more robust and offers more flexibility in covariance modeling. This leads to its superior performance in this study and makes the EnKF a promising approach for soil moisture initialization of seasonal climate forecasts.