State updating of root zone soil moisture estimates of an unsaturated zone metamodel for operational water resources management

Abstract

Combining metamodels with data assimilation schemes allows the incorporation of up-to-date information in metamodels, offering new opportunities for operational water resources management. We developed a data assimilation scheme for the unsaturated zone metamodel MetaSWAP using OpenDA, which is an open source data assimilation framework. A twin experiment showed the feasibility of applying an Ensemble Kalman filter as a data assimilation method for updating metamodels. Furthermore, we assessed the accuracy of root zone soil moisture model estimates when assimilating the regional SMAP L3 Enhanced surface soil moisture product. The model accuracy is assessed using in situ soil moisture measurements collected at 12 locations in the Twente region, the Netherlands. Although the accuracy of the model estimates does not improve in terms of correlation coefficient, the accuracy does improve in terms of Root Mean Square Error and bias. Therefore, the assimilation of surface soil moisture observations has value for updating root zone soil moisture model estimates. In addition, the accuracy of the model estimates improves on both regional and local spatial scales. The increasing availability of remotely sensed soil moisture data will lead to new possibilities for integrating metamodelling and data assimilation in operational water resources management. However, we expect that significant investments in computational capacities are necessary for effective implementation in decision-making. Keywords: Data assimilation, Ensemble Kalman filter, Hydrological modelling, Metamodelling, Remote sensing, SMAP, Soil moisture