Operational forecast uncertainty assessment for better information to stakeholders and crisis managers

Abstract

Abstract. Flood forecasting uncertainty is crucial information for decision makers. However, deterministic only forecasts have been communicated in France until now, like in many other countries. The French Flood Forecast Centres (FFCs) recently set up a new service which aims at publishing quantitative forecasts along with their associated uncertainty. Two surveys (one of the greater audience and identified end-users, another of FFCs worldwide) were conducted to design it. The forecasters' toolbox was then supplemented with two new tools. The first one provides automatic forecasting uncertainty estimations calibrated on past forecasting error series. The second one allows the forecasters to incorporate their own expertise to adjust the automatically calculated uncertainty estimation. The evaluation of the forecast uncertainty estimations issued in real time in 2014 suggests that even if these assessments are perfectible, they are already informative and useful for end-users. The first feedbacks from forecasters and crisis managers also show that if the assessment of probabilistic forecasts remains a technical challenge, their use is foremost a human challenge. This move is a paradigm change for both forecasters and decision makers. Therefore, they have to be accompanied in order to achieve this deep shift in their professional practice.

Advancing lake and reservoir water quality management with near-term, iterative ecological forecasting

“…Identifying the factors contributing to total forecast uncertainty can be used to prioritize improvements to the forecast workflow (Dietze 2017). In addition, knowing the uncertainty associated with a forecast may allow managers to make more informed decisions about the likelihood of a forecasted event (Morss et al 2008, Berthet et al 2016.…”

“…Specifically, knowing the likelihood of different forecasted water quality impairment events (e.g., hypolimnetic anoxia) may improve managers' ability to interpret their risk. For example, a study of managers receiving flood forecasts revealed that the majority thought that "adding uncertainties to forecasts bring[s] useful operational information" (Berthet et al 2016). When no uncertainties are provided with forecasts, many stakeholders will infer their own estimate of uncertainty for decision-making, which may be biased and less accurate than quantified uncertainties from forecast models (Morss et al 2008, Berthet et al 2016).…”

“…For example, a study of managers receiving flood forecasts revealed that the majority thought that "adding uncertainties to forecasts bring[s] useful operational information" (Berthet et al 2016). When no uncertainties are provided with forecasts, many stakeholders will infer their own estimate of uncertainty for decision-making, which may be biased and less accurate than quantified uncertainties from forecast models (Morss et al 2008, Berthet et al 2016). Because little is currently known about manager interpretation of water quality forecast uncertainty, studying how managers respond to different levels of uncertainty will help improve forecast visualization and decision support tools.…”

A Near‐Term Iterative Forecasting System Successfully Predicts Reservoir Hydrodynamics and Partitions Uncertainty in Real Time

“…Multiple statistical approaches, such as Bayesian state‐space modeling, particle filters, and ensemble filters, are used to estimate and propagate the different sources of uncertainty that contribute to the total uncertainty in a forecast (e.g., driver data, initial conditions, model process, and parameters; Clark et al, 2008; Dietze, 2017a; Ouellet‐Proulx, Chimi Chiadjeu, et al, 2017). Fully specifying all of these uncertainty sources provides both an assessment of confidence in a forecast for managers as they interpret the forecasts for decision‐making, and valuable information for researchers about how to improve forecasts (Berthet et al, 2016; Dietze, 2017a; Morss et al, 2008).…”

Keeping Everyone on the Same Page During Flash Flood Emergencies

“…While uncertainty in hydrologic forecasts is unavoidable, incorporating uncertainty as a part of the forecast information has met with limited success. Berthet et al [27] describe an assessment of incorporating uncertainty measures communicated as part of the forecast information but found only limited acceptance among governmental flood forecasting centers and hydroelectricity suppliers because of potential confusion associated with the uncertainty bounds communicated. Vieux [28] describe case studies involving forecast uncertainty in real-time hydrologic forecasting of reservoir inflows to hydroelectric facility with gated dam operations.…”