Correcting the radar rainfall forcing of a hydrological model with data assimilation: application to flood forecasting in the Lez catchment in Southern France

Abstract

Abstract. The present study explores the application of a data assimilation (DA) procedure to correct the radar rainfall inputs of an event-based, distributed, parsimonious hydrological model. An extended Kalman filter algorithm was built on top of a rainfall-runoff model in order to assimilate discharge observations at the catchment outlet. This work focuses primarily on the uncertainty in the rainfall data and considers this as the principal source of error in the simulated discharges, neglecting simplifications in the hydrological model structure and poor knowledge of catchment physics. The study site is the 114 km 2 Lez catchment near Montpellier, France. This catchment is subject to heavy orographic rainfall and characterised by a karstic geology, leading to flash flooding events. The hydrological model uses a derived version of the SCS method, combined with a Lag and Route transfer function. Because the radar rainfall input to the model depends on geographical features and cloud structures, it is particularly uncertain and results in significant errors in the simulated discharges. This study seeks to demonstrate that a simple DA algorithm is capable of rendering radar rainfall suitable for hydrological forecasting. To test this hypothesis, the DA analysis was applied to estimate a constant hyetograph correction to each of 19 flood events. The analysis was carried in two different modes: by assimilating observations at all available time steps, referred to here as reanalysis mode, and by using only observations up to 3 h before the flood peak to mimic an operational environment, referred to as pseudo-forecast mode. In reanalysis mode, the resulting correction of the radar rainfall data was then compared to the mean field bias (MFB), a corrective coefficient determined using rain gauge measurements. It was shown that the radar rainfall corrected using DA leads to improved discharge simulations and Nash-Sutcliffe efficiency criteria compared to the MFB correction. In pseudo-forecast mode, the reduction of the uncertainty in the rainfall data leads to a reduction of the error in the simulated discharge, but uncertainty from the model parameterisation diminishes data assimilation efficiency. While the DA algorithm used is this study is effective in correcting uncertain radar rainfall, model uncertainty remains an important challenge for flood forecasting within the Lez catchment.

Benefits and limitations of data assimilation for discharge forecasting using an event-based rainfall–runoff model

“…Possible improvement could be expected by extending the data assimilation correction to other components of the hydrological model such as the value of the threshold that triggers the direct runoff, the parameters that control the shape of the falling limb during multi-peak events or the rainfall calculated over the catchment. This last correction is addressed in Harader et al (2012). The present study, which presents the benefits and the limitations of a data assimilation for hydrological forecast, must be extended to other lead times, other catchments and hydrological models in order to draw more general conclusions.…”

“…The fact that the assimilation is unable to improve both peaks reveals that model errors do not result solely from an inaccurate specification of S. Other sources of errors have to be considered and corrected by the data assimilation technique such as the threshold that triggers the runoff or the rainfall that force the model. The correction of this last component is within the scope of Harader et al (2012). In the case of the second or subsequent peaks in multi-peak events, such as December 2002 or September 2005, uncertainties in the parameters or variables which control the shape of the falling limb, such as ds and w or stoc(t), could be also taken into account.…”

A coupled atmospheric–hydrologic modeling system with variable grid sizes for rainfall–runoff simulation in semi-humid and semi-arid watersheds: how does the coupling scale affects the results?

“…In order to eliminate the discrepancy of the initial and boundary conditions with the driven data, another outermost domain is set beyond the WRF three nested domains to downscale the FNL data to a spatial resolution of 27 km. The integration time step of the model is set to 6 s, and the output data interval is set to 1 h (Hong and Lee, 2009). Forty layers are considered in the three nested domains, with a toplayer pressure of 50 hPa.…”

A GIS-Based Methodology to Combine Rain Gauge and Radar Rainfall Estimates of Precipitation Using the Conditional Merging Technique for High-Resolution Quantitative Precipitation Forecasts in Țibleș and Rodnei Mountains

“…All these tools have uncertainties related to their input data and parameter settings. Rainfall-runoff models are sensitive to rainfall quantities and their spatial distribution throughout the catchment; thus, errors in precipitation estimates have a significant impact on predictability and event reconstruction [1].…”