Sensitivity of hydrological ensemble forecasts to different sources and temporal resolutions of probabilistic quantitative precipitation forecasts: flash flood case studies in the Cévennes‐Vivarais region (Southern France)

Abstract

This article analyses the performance of an integrated hydrological ensemble prediction system adapted to small to mid-sized catchments (100-600 km 2 ) situated in the Cévennes-Vivarais region (Southern France) and characterized by short lag times (3-12 h). In this framework, flood forecasts need hourly Probabilistic Quantitative Precipitation Forecasts (PQPF) so as to provide early warning with 24-72 h of anticipation. Here, two sources of PQPF at daily and subdaily (6 h) meteorological time steps are considered: Ensemble Prediction Systems from the European Centre for Medium-range Weather Forecast (ECMWF) and analogy-based PQPF provided in real-time at the Laboratoire d'étude des Transferts en Hydrologie et Environnement. The two PQPF are firstly disaggregated to respect the required hydrological hourly time step, through either the use of a stochastic rainfall generator or the application of a multimodel approach. Then, disaggregated PQPF are used as input to a hydrological model, which is called TOPSIMPL, to provide hourly ensemble discharge forecasts up to 48 h ahead. Illustration and evaluation of ensemble discharge forecasts issued in near real-time conditions are given for some recently observed flash flood events. It is shown that hourly discharge forecasts are first-order conditioned by the accuracy of PQPF at daily or subdaily time step. Six-hour ensemble prediction systems correctly reproduce the rainfall temporal dynamics, whereas daily analogy-based PQPF are less underdispersive in terms of rainfall amounts. As a result, the merging of the two sources of PQPF substantially increases the performance of discharge forecasts, the contribution of a more sophisticated hourly rainfall generator becoming marginal. Figure 3. Ratio of RMSE to spread as a function of spread (1/10 mm/h) related to hourly precipitation distributions issued during the 2007 flood event. The horizontal black line corresponds to RMSE equal to spread 38 R. MARTY, I. ZIN AND CH. OBLED

Probabilistic flood forecasting on the Rhone River: evaluation with ensemble and analogue-based precipitation forecasts

“…On the other hand, analogue-based forecasts were found more reliable thanks to a larger spread, despite their tendency to be over-dispersive. These findings are in accordance with results of [15], although these were obtained at daily time step. A negative bias was also observed, weaker though than the one of ensemble-based forecasts.…”

“…This approach has been already tackled by two previous studies. Authors in [15] disaggregated daily analoguebased precipitation forecasts with the chronology of 6-hour ensemble forecasts, with the idea that the analogue method provides more reliable forecasts of precipitation amounts, while ensemble forecasts provide a useful information about the temporal dynamic of rainfall events. This approach, however, uses ensemble forecasts only for their temporal dynamic.…”

“…In a nutshell, ensemble forecasts represent mainly the uncertainty due to the dynamical evolution of the synoptic situation, while analogue-based forecasts represent the uncertainty of the thermodynamics related to the precipitation generation processes from a given synoptic situation. To date, very few studies [15] have evaluated and compared these two approaches in a flood forecasting context. Evaluation aims to assess (by means of scores) the quality of forecasts given the observations.…”

Optimising predictor domains for spatially coherent precipitation downscaling

“…Both PP and MOS methods are also applied for operational precipitation forecast (e.g. Marty et al, 2008Marty et al, , 2012Marty et al, , 2013Voisin et al, 2010;Nam et al, 2011;Liu and Coulibaly, 2011;Muluye, 2011).…”

Impact of improved meteorological forcing, profile of soil hydraulic conductivity and data assimilation on an operational Hydrological Ensemble Forecast System over France

“…HEFSs have also been developed at the European scale with the European Flood Forecasting System (De Roo et al, 2003), which subsequently led to the European Flood Alert System (EFAS) (Thielen et al, 2009a,b) and more recently, at the global scale, the Global Flood Awareness System (GloFAS), which provides forecasts for very large river basins (Alfieri et al, 2013). HEFSs now cover a large range of lead times from shortrange (1-2 days) (Alfieri and Thielen, 2012;Marty et al, 2013;Vincendon et al, 2011) to medium-range (3-10 days) (Thirel et al, 2010a,b;Addor et al, 2011) to longer range, with monthly or seasonal water resources management (Yuan et al, 2013;Singla et al, 2012).…”