Generating Coherent Ensemble Forecasts After Hydrological Postprocessing: Adaptations of ECC‐Based Methods

Abstract

Hydrological ensemble forecasts are frequently miscalibrated, and need to be statistically postprocessed in order to account for the total predictive uncertainty. Very often, this step relies on parametric, univariate techniques that ignore the between‐basins and between‐lead times dependencies. This calls for a procedure referred to as sampling‐reordering, which generates a coherent multivariate ensemble from the marginal postprocessed distributions. The ensemble copula coupling (ECC) approach, which is already popular in the field of meteorological postprocessing, is attractive for hydrological forecasts as it preserves the dependence structure of the raw ensemble assumed as spatially and temporally coherent. However, the existing implementations of ECC have strong limitations when applied to hourly streamflow, due to raw ensembles being frequently nondispersive and to streamflow data being strongly autocorrelated. Based on this diagnosis, this paper investigates several variants of ECC, in particular the addition of a perturbation to the raw ensemble to handle the nondispersive cases, and the smoothing of the temporal trajectories to make them more realistic. The evaluation is conducted on a case study of hydrological forecasting over a set of French basins. The results show that the new variants improve upon the existing ECC implementations, while they remain simple and computationally inexpensive.

Evaluation of two hydrometeorological ensemble strategies for flash-flood forecasting over a catchment of the eastern Pyrenees

“…The inclusion of these uncertainties aims at improving the skill and spread of the HEPSs by introducing independent information of all the plausible atmospheric states and processes. Therefore, SREPSs are increasingly used in hydrologic prediction (Cloke and Pappenberger, 2009;Verkade et al, 2013Verkade et al, , 2017Siddique and Mejia, 2017;Benninga et al, 2017;Bellier et al, 2017Bellier et al, , 2018Edouard et al, 2018;Jain et al, 2018). Several studies have stated that probabilistic forecasts could improve decisionmaking if appropriately handled (e.g., Krzysztofowicz, 2001;Todini, 2004;Ramos et al, 2013;Antonetti et al, 2019).…”

Evaluation of a global ensemble flood prediction system in Peru

“…A popular method is to recalibrate the model results to reproduce the climatological distribution (Madadgar et al 2014). Other postprocessing techniques include event bias correction (Smith et al 1992), LOWESS regression (Cleveland 1979), variance inflation (INFL) method (Fundel andZappa 2011, Roulin andVannitsem 2015), ensemble copula coupling (ECC) (Schefzik et al 2013, Bellier et al 2018) and quantile mapping (Wood andLettenmaier 2006, Baigorria et al 2007). Although all of them improve forecast quality significantly (Kang et al 2010), Hashino et al (2007 have shown that the choice of the optimal technique is subject to application requirements.…”

Statistical Postprocessing of Water Level Forecasts Using Bayesian Model Averaging With Doubly Truncated Normal Components

“…Further, following the ideas of Hemri et al (2015) and Bellier et al (2018), one can combine the BMA calibrated forecasts corresponding to different locations and lead times either into temporally or both spatially and temporally coherent multivariate predictions. This can be done with the help of modern techniques such as the ensemble copula coupling (Schefzik et al, 2013) or the Gaussian copula approach (Pinson & Girard, 2012).…”