The potential of remotely sensed soil moisture for operational flood forecasting
Speaker: Niko Wanders, Utrecht University, The Netherlands
Series: EEWR Brown Bag Seminars
Location: Engineering Quad E225
Date/Time: Friday, February 14, 2014, 12:00 p.m. - 1:00 p.m.
Nowadays, remotely sensed soil moisture is readily available from multiple space born sensors. The high temporal resolution and global coverage make these products very suitable for large-scale land-surface applications. Here we study the potential of these products for hydrological model calibration and improving real-time flood forecasting in large catchments.
Soil moisture observations from satellites provide the closest thing to a direct measurement of the state of the unsaturated zone at large spatial scales. They are potentially useful in calibrating unsaturated zone model parameters for large scale hydrological models. Additionally, we evaluate the added value of assimilated remotely sensed soil moisture for the European Flood Awareness System (EFAS) and its potential to improve the timing and height of the flood peak and low flows. EFAS is used for operational flood forecasting in Europe and uses a distributed hydrological model for flood predictions for lead times up to 10 days. EFAS is calibrated for the Upper Danube using remotely sensed soil moisture and discharge observations. Hereafter, satellitederived soil moisture from three sensors and discharge observations are assimilated into the EFAS system using a Ensemble Kalman Filter for flood forecasting. The added value of the soil moisture observations is evaluated and compared to the gain obtained by only using discharge observations in the calibration and assimilation. The performance is evaluated against observed discharge timeseries distributed throughout the catchment.
Results show that remotely sensed soil moisture holds potential for calibration of hydrological models and increases the performance of flood forecasting systems for large catchments, like the Upper Danube.