The benefits of using remotely sensed soil moisture in parameter identification of large‐scale hydrological models

Abstract

Large-scale hydrological models are nowadays mostly calibrated using observed discharge. As a result, a large part of the hydrological system, in particular the unsaturated zone, remains uncalibrated. Soil moisture observations from satellites have the potential to fill this gap. Here we evaluate the added value of remotely sensed soil moisture in calibration of large-scale hydrological models by addressing two research questions: (1) Which parameters of hydrological models can be identified by calibration with remotely sensed soil moisture? (2) Does calibration with remotely sensed soil moisture lead to an improved calibration of hydrological models compared to calibration based only on discharge observations, such that this leads to improved simulations of soil moisture content and discharge? A dual state and parameter Ensemble Kalman Filter is used to calibrate the hydrological model LISFLOOD for the Upper Danube. Calibration is done using discharge and remotely sensed soil moisture acquired by AMSR-E, SMOS, and ASCAT. Calibration with discharge data improves the estimation of groundwater and routing parameters. Calibration with only remotely sensed soil moisture results in an accurate identification of parameters related to land-surface processes. For the Upper Danube upstream area up to 40,000 km 2 , calibration on both discharge and soil moisture results in a reduction by 10-30% in the RMSE for discharge simulations, compared to calibration on discharge alone. The conclusion is that remotely sensed soil moisture holds potential for calibration of hydrological models, leading to a better simulation of soil moisture content throughout the catchment and a better simulation of discharge in upstream areas.

Application of Remote Sensing Data to Constrain Operational Rainfall-Driven Flood Forecasting: A Review

“…Operational flood forecasting models, including lumped, semi-distributed and distributed models, contain a number of parameters that can be highly conceptualized and are not directly measurable [136]. These parameters are traditionally estimated through minimizing the differences between observed and simulated streamflow in order to generate accurate and reliable future flow forecasts.…”

“…In spite of the challenge in improving streamflow prediction by calibration using soil moisture, several explorations have been carried out recently [136,[144][145][146][147] to examine the potential of improving short-term streamflow forecasting by calibrating model parameters using soil moisture remote sensing data (Table 3). A relatively early study conducted by Parajka et al [146] found that joint-calibration of a semi-distributed model using both streamflow and SAR/ERS derived soil moisture data improved the accuracy of the soil moisture predictions without degradation of the streamflow predictions.…”

“…Nevertheless it was found that the joint calibration results in more robust calibration than the calibration using streamflow only case [145,146]. While studies by Parajka et al [145,146] used low temporal resolution SAR data in a semi-distributed system, further attention has been paid to calibrate fully distributed hydrologic models using higher temporal resolution microwave data [136,144,147]. Sutanudjaja et al [144] used SCAT/ERS derived soil moisture data for a coupled groundwater-land surface model and found that the joint calibration using flow and soil moisture can reproduce good soil moisture and streamflow, as well as ground water head predictions.…”

Multi-variable SWAT model calibration with remotely sensed evapotranspiration and observed flow

“…However, model calibration is commonly performed with watershed outlet streamflow data only. Consequently, part of the processes, specially from the unsaturated zone, can remain uncalibrated (WANDERS et al, 2014).…”

“…Recently, some studies have integrated remote sensing data into the calibration of distributed hydrological models (GITHUI; SELLE; THAYALAKUMARAN, 2012;MUTHUWATTA;BOOIJ;RIENTJES, 2009) obtaining streamflow prediction performance improvement (KUNNATH-POOVAKKA et al, 2016;WANDERS et al, 2014;ZHANG et al, 2009). Rajib, Merwade and Yu (2016) calibrated the SWAT model using streamflow and soil moisture simultaneously, and compared it with the conventional calibration (streamflow only).…”