Development of Soil Moisture Assimilation based on KIM-LETKF System

A land data assimilation system (LDAS) under the Korean Integrated Model (KIM) framework has been newly developed and a soil moisture data assimilation (DA) is tested at the preliminary state. KIM-LDAS is based on KIM ensemble prediction system (KIM-EPS) and ensemble DA system, which is a localization ensemble transform Kalman filter (LETKF), referred to as KIM-LETKF. A state-perturbation method with a horizontal length of 10 km and a standard deviation of 0.01 m
³
m
⁻³
is employed to represent uncertainties of model for soil moisture in KIM-EPS. Using the resultant background ensemble from KIM-EPS, the newly developed KIM-LETKF-based soil moisture DA system is tested over East Asia. It is found that proper analysis increments compared with innovations are achieved when a cut-off value for a horizontal localization factor is set to 100 km.