Search

International audience; We outline a brief timeline of the history of ensemble prediction, which resulted in the setup of HEPEX, starting from the early research on atmospheric predictability and uncertainty in the 1960-70s, through the development of ensemble prediction methods for monthly to seasonal predictions in hydrology and meteorology in the US and UK in the 1980s, the development of the first medium-range (up to 10 days) operational weather ensemble prediction systems at ECMWF and NCEP in the 1990s, the first runs of EFAS with numerical weather ensemble predictions in 2005, up to the most recent developments on seamless prediction systems, ensemble short-range and nowcasting systems, coupled Earth systems and Global multi-model forecasting. The recognition of tradeoffs among different paradigms in hydrological forecasting is emphasized, and the transition from the first model-based, traditional "human-in-the-loop" flood forecasting approaches to more automated, "over-the-loop" probabilistic-based approaches and coupled NWP-hydrological modeling systems. Despite the initial focus on medium-range forecasting, HEPEX has been able to embrace a wide range of spatial and temporal scales. Today, model-based ensemble predictions are investigated for flash flood nowcasting and up to the long-range (several months) forecasting of water resources, and over spatial scales from small to large river systems controlled by different processes. Finally, it is important to note that HEPEX has always been an unfunded/volunteer effort - perhaps the longest in the history of hydrometeorological forecasting

No abstract available.

Hydrological ensemble prediction systems (HEPS) have in recent years been increasingly used for the operational forecasting of floods by European hydrometeorological agencies. The most obvious advantage of HEPS is that more of the uncertainty in the modelling system can be assessed. In addition, ensemble prediction systems generally have better skill than deterministic systems both in the terms of the mean forecast performance and the potential forecasting of extreme events. Research efforts have so far mostly been devoted to the improvement of the physical and technical aspects of the model systems, such as increased resolution in time and space and better description of physical processes. Developments like these are certainly needed; however, in this paper we argue that there are other areas of HEPS that need urgent attention. This was also the result from a group exercise and a survey conducted to operational forecasters within the European Flood Awareness System (EFAS) to identify the top priorities of improvement regarding their own system. They turned out to span a range of areas, the most popular being to include verification of an assessment of past forecast performance, a multi-model approach for hydrological modelling, to increase the forecast skill on the medium range (>3 days) and more focus on education and training on the interpretation of forecasts. In light of limited resources, we suggest a simple model to classify the identified priorities in terms of their cost and complexity to decide in which order to tackle them. This model is then used to create an action plan of short-, medium- and long-term research priorities with the ultimate goal of an optimal improvement of EFAS in particular and to spur the development of operational HEPS in general.

International audience; Ensemble prediction system (EPS) meteorological forecasts are increasingly being used as input to hydrological models in order to extend flood-warning lead-times and to improve forecasts and the knowledge of their uncertainty. Probabilistic skill scores classically used in meteorology are usually also utilized for assessing the quality of hydrological ensemble forecasts. However, the different river discharge magnitudes can make difficult the interpretation and comparisons of these scores, as it is for the continuous rank probability score (CRPS). In this letter, a novel 'Reduction' CRPS (RCRPS), which takes into account the different river discharge magnitudes, is proposed, and its usefulness is exhibited.

Ensemble Prediction System (EPS) meteorological forecasts are increasingly being used as input to hydrological models in order to extend floodwarning leadtimes and to improve forecasts and the knowledge of their uncertainty. Probabilistic skill scores classically used in meteorology are usually also utilized for assessing the quality of hydrological ensemble forecasts. However, the different river discharge magnitudes can make difficult the interpretation and comparisons of these scores, as it is for the Continuous Rank Probability Score (CRPS). In this letter, a novel ‘Reduction’ CRPS (RCRPS), which takes into account the different river discharge magnitudes, is proposed, and its usefulness is exhibited., JRC.H.7-Climate Risk Management

Flooding is a natural global phenomenon but in many cases is exacerbated by human activity. Although flooding generally affects humans in a negative way, bringing death, suffering, and economic impacts, it also has potentially beneficial effects. Early flood warning and forecasting systems, as well as the development of real-time monitoring systems, are recognised measures to reduce the number of flood victims and to support flood disaster responses. Importantly, they may buy time to take appropriate mitigation measures to reduce flood peaks, thus reducing the associated negative impacts. One of the main constraints for global hydrological modelling is the limited availability of observational data for calibration and model verification. This is an even larger issue for real-time flood monitoring and forecasting. This lack of data could potentially be overcome if satellite-retrieved surface water changes signal or streamflow estimates were sufficiently accurate to serve as a surrogate for ground-based measurements. In the first part of the study, the potentials and constraints of river streamflow estimates based on the remote sensing signal of the Global Flood Detection System are evaluated. We used the merged product from the Global Flood Detection System (GFDS) that uses both AMSR-E (Advance Microwave Scanning Radiometer – Earth Observing System) and TRMM (Tropical Rainfall Measuring Mission) to derive surface water extent during the study period. The influence of the local physiographic factors which might influence the retrieval of the satellite signal was also studied. Validation is done based on ground-based streamflow observations. In the second part of the study, it was tested if the GFDS derived streamflow proxy improved the model calibration of the distributed rainfall-runoff routing model LISFLOOD, used by the Global Flood Awareness System (GloFAS). Finally, the GFDS surface water extent data were assimilated into the large-scale hydrological model LISFLOOD using an Ensemble Kalman filter (EnKF). It was evaluated if flood forecasting skill would improve, as well as the timing of the flood peak, as compared to baseline initial conditions (without data assimilation). Furthermore, two additional studies looked at the use of globally near real-time available products for flood forecasting, monitoring, and assessment to support decision makers and humanitarian organisations such as Red Cross Red Crescent and the World Food Programme.