Improving land surface model soil moisture estimates through joint state-parameter assimilation of in situ soil moisture observations

Accurate soil moisture estimates are essential for effective management and operational planning of flood and drought events, both of which are intensifying under a changing climate. However, soil moisture values derived from land surface models often exhibit significant deviations from in situ observations. Data assimilation combines model information with observations to enhance prediction accuracy. Previous studies have typically focused on either estimating the initial soil moisture state or optimising Pedotransfer Function (PTF) constants (or parameters), which link soil texture to the hydraulic properties of land surface models. In contrast, this study introduces a novel extension by performing joint state‐parameter assimilation for the JULES model, simultaneously optimising both PTF constants and initial soil moisture states within an augmented framework. Using an Ensemble–Variational hybrid data assimilation scheme, we ingested field‐scale soil moisture observations from the Cosmic‐ray Soil Moisture Monitoring Network across 16 diverse sites in the UK. The joint state‐parameter assimilation improved average Kling–Gupta Efficiency by approximately 113% relative to the open‐loop, 105% relative to state‐only and 9.6% relative to parameter‐only experiments. These results highlight the strong potential of the proposed framework for enhancing land‐surface reanalysis, numerical weather prediction and hydrometeorological forecasting of floods and droughts.