On the assimilation set-up of ASCAT soil moisture data for improving streamflow catchment simulation
Assimilation of remotely sensed surface soil moisture (SSM) data into hydrological catchment models has been identified as a means to improve stream flow simulations, but reported results vary markedly depending on the particular model, catchment and assimilation procedure used. In this study, the i...
| Autores: | , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión aceptada para publicación |
| Fecha de publicación: | 2018 |
| País: | España |
| Institución: | Universidad Pública de Navarra |
| Repositorio: | Academica-e. Repositorio Institucional de la Universidad Pública de Navarra |
| OAI Identifier: | oai:academica-e.unavarra.es:2454/32203 |
| Acceso en línea: | https://hdl.handle.net/2454/32203 |
| Access Level: | acceso abierto |
| Palabra clave: | Data assimilation ASCAT Surface soil moisture Ensemble Kalman Filter Stream flow simulation Hydrological catchment models |
| Sumario: | Assimilation of remotely sensed surface soil moisture (SSM) data into hydrological catchment models has been identified as a means to improve stream flow simulations, but reported results vary markedly depending on the particular model, catchment and assimilation procedure used. In this study, the in fluence of key aspects, such as the type of model, re-scaling technique and SSM observation error considered, were evaluated. For this aim, Advanced SCATterometer ASCAT-SSM observations were assimilated through the ensemble Kalman filter into two hydrological models of different complexity namely MISDc and TOPLATS) run on two Mediterranean catchments of similar size (750 km2). Three different re-scaling techniques were evaluated (linear re-scaling, variance matching and cumulative distribution function matching), and SSM observation error values ranging from 0.01% to 20% were considered. Four different efficiency measures were used for evaluating the results. Increases in Nash-Sutcliffe efficiency (0.03–0.15) and efficiency indices (10–45%) were obtained, especially when linear re-scaling and observation errors within 4-6% were considered. This study found out that there is a potential to improve stream flow prediction through data assimilation of remotely sensed SSM in catchments of different characteristics and with hydrological models of different conceptualizations schemes, but for that, a careful evaluation of the observation error and re-scaling technique set-up utilized is required. |
|---|