Prediction-Monitoring of Hydrological Variables based on Multivariate Assimilation of Ground and Satellite Data

Nowadays the capability of forecasting-monitoring of hydrological variables is one of the crucial issues in the hydrology context. Traditionally, calibrating and developing hydrological models is based on streamflow observation. This issue arises from the lack of the availability of in situ measurements of other hydrological variables. Moreover, the models that only rely on the streamflow observation measured at the subbasin’s outlet may incorrectly represent the internal watershed processes. Due to advances in remote-sensing techniques, a good opportunity is available for calibrating-developing hydrological models based on the remotely-sensed data. Due to the importance of the Urmia Lake basin and the lack of a hydrological system for forecasting hydrological variables in the basin, we applied Data Assimilation (DA) at the upstream part of the basin. Accordingly, we simultaneously used the in-situ measurement of streamflow and remotely-sensed MODIS Snow Cover Fraction (SCF) to forecast hydrological variables at the upstream part of the Mahabad basin. Moreover, we compared the result of EnKF with SUFI2. Results show the simultaneous utilization of SCF and streamflow can concurrently improve the simulation accuracy of both variables. However, both EnKF and SUFI2 access similar result concerning streamflow simulation, EnKF provide a better result for SCF compared to SUFI2.