|Title||A simplified Extended Kalman Filter for the global operational soil moisture analysis at ECMWF|
|Year of Publication||2012|
|Authors||De Rosnay, P, Drusch, M, Vasiljevic, D, Balsamo, G, Albergel, C, Isaksen, L|
|Secondary Title||Technical Memorandum|
A new land surface analysis system based on a simplified point-wise Extended Kalman Filter (EKF) was implemented at ECMWF in the global operational Integrated Forecasting System (IFS) in November 2010. This system will allow consistent and optimal analyses of land surface parameters like soil moisture, soil temperatures, snow and vegetation properties. As part of the system implementation the surface analysis structure has been revised to permit an independent and parallel computation with the upper-air 4D-Var analysis. The new analysis system is used for the soil moisture analysis, replacing the previous Optimum Interpolation (OI) scheme. Similar to the OI system, the simplified EKF uses 2-metre air temperature and relative humidity observations from the SYNOP (land surface synoptic report) ground based networks to analyse soil moisture. This paper describes the new land surface analysis, its application for analysing soil moisture, and initial verification results that supported its operational implementation at ECMWF. The performance is evaluated based on a set of one-year analysis experiments. The simplified EKF is compared to the OI, on soil moisture, 2m temperature and relative humidity, showing a consistent improvement on screen level parameters and soil moisture forecasts. To demonstrate the potential of the new analysis scheme soil moisture derived from ASCAT (Advanced Scatterometer) has been assimilated through the simplified EKF.