Florian Pappenberger

Ignorance is bliss: Or seven reasons not to use uncertainty analysis

Uncertainty analysis of models has received increasing attention over the last two decades in water resources research. However, a significant part of the community is still reluctant to embrace the estimation of uncertainty in hydrological and hydraulic modeling. In this paper, we summarize and explore seven common arguments: uncertainty analysis is not necessary given physically realistic models; uncertainty analysis cannot be used in hydrological and hydraulic hypothesis testing; uncertainty (probability) distributions cannot be understood by policy makers and the public; uncertainty analysis cannot be incorporated into the decision-making process; uncertainty analysis is too subjective; uncertainty analysis is too difficult to perform; uncertainty does not really matter in making the final decision. We will argue that none of the arguments against uncertainty analysis rehearsed are, in the end, tenable. Moreover, we suggest that one reason why the application of uncertainty analysis is not normal and expected part of modeling practice is that mature guidance on methods and applications does not exist. The paper concludes with suggesting that a Code of Practice is needed as a way of formalizing such guidance.

Development of a European flood forecasting system

Abstract Recent advances in meteorological forecast skill now enable significantly improved estimates of precipitation quantity, timing and spatial distribution to be made up to 10 days ahead for model scales of 40 km in forecast mode. Here we outline a prototype methodology to downscale these precipitation estimates using regional Numerical Weather Prediction models to spatial scales appropriate to hydrological forecasting and then use these to drive high‐resolution scale (1 or 5 km grid scale) water balance and rainfall‐runoff models. The aim is to develop a European Flood Forecasting System (EFFS) and determine what flood forecast skill can be achieved for given basins, meteorological events and prediction products. The output from the system is a probabilistic assessment of n‐day ahead discharge exceedence risk (where n < 10) for the whole of Europe at 5 km resolution which may then be updated as the forecast lead time reduces. At each stage the discharge estimates can be used to drive detailed (25–100 m resolution) hydraulic models to estimate the flood inundation which may potentially occur. Initial results are presented from a prototype version of the system used to perform a hindcast of the January 1995 flooding events in NW‐Europe (Rhine, Meuse).

Ensemble predictions and perceptions of risk, uncertainty, and error in flood forecasting

Abstract Under the auspices of the World Meteorological Organization, there are a number of international initiatives to promote the development and use of so-called ensemble prediction systems (EPS) for flood forecasting. The campaign to apply these meteorological techniques to flood forecasting raises important questions about how the probabilistic information these systems provide can be used for what in operational terms is typically a binary decision of whether or not to issue a flood warning. To explore these issues, we report on the results of a series of focus group discussions conducted with operational flood forecasters from across Europe on behalf of the European Flood Alert System. Working in small groups to simulate operational conditions, forecasters engaged in a series of carefully designed forecasting exercises using various different combinations of actual data from real events. Focus group data was supplemented by a follow-up questionnaire survey exploring how flood forecasters understand risk, uncertainty, and error. Results suggest that flood forecasters may not instinctively use ensemble predictions in the way that promoters of EPS perhaps think they should. The paper concludes by exploring the implications of these divergent ‘epistemic cultures’ for efforts to apply ensemble prediction techniques developed in the context of weather forecasting to the rather different one of flood forecasting.

High-Resolution 3-D Flood Information From Radar Imagery for Flood Hazard Management

This paper presents a remote-sensing-based steady-state flood inundation model to improve preventive flood-management strategies and flood disaster management. The Regression and Elevation-based Flood Information eXtraction (REFIX) model is based on regression analysis and uses a remotely sensed flood extent and a high-resolution floodplain digital elevation model to compute flood depths for a given flood event. The root mean squared error of the REFIX, compared to ground-surveyed high water marks, is 18 cm for the January 2003 flood event on the River Alzette floodplain (G.D. of Luxembourg), on which the model is developed. Applying the same methodology on a reach of the River Mosel, France, shows that for some more complex river configurations (in this case, a meandering river reach that contains a number of hydraulic structures), piecewise regression is required to yield more accurate flood water-line estimations. A comparison with a simulation from the Hydrologic Engineering Centers River Analysis System hydraulic flood model, calibrated on the same events, shows that, for both events, the REFIX model approximates the water line reliably

Toward global drought early warning capability: Expanding international cooperation for the development of a framework for monitoring and forecasting

Drought is a global problem that has far-reaching impacts, especially on vulnerable populations in developing regions. This paper highlights the need for a Global Drought Early Warning System (GDEWS), the elements that constitute its underlying framework (GDEWF), and the recent progress made toward its development. Many countries lack drought monitoring systems, as well as the capacity to respond via appropriate political, institutional, and technological frameworks, and these have inhibited the development of integrated drought management plans or early warning systems. The GDEWS will provide a source of drought tools and products via the GDEWF for countries and regions to develop tailored drought early warning systems for their own users. A key goal of a GDEWS is to maximize the lead time for early warning, allowing drought managers and disaster coordinators more time to put mitigation measures in place to reduce the vulnerability to drought. To address this, the GDEWF will take both a top-down approach...

Sensitivity analysis based on regional splits and regression trees (SARS-RT)

A global sensitivity analysis with regional properties is introduced. This method is demonstrated on two synthetic and one hydraulic example. It can be shown that an uncertainty analysis based on one-dimensional scatter plots and correlation analyses such as the Spearman Rank Correlation coefficient can lead to misinterpretations of any model results. The method which has been proposed in this paper is based on multiple regression trees (so called Random Forests). The splits at each node of the regression tree are sampled from a probability distribution. Several criteria are enforced at each level of splitting to ensure positive information gain and also to distinguish between behavioural and non-behavioural model representations. The latter distinction is applied in the generalized likelihood uncertainty estimation (GLUE) and regional sensitivity analysis (RSA) framework to analyse model results and is used here to derive regression tree (model) structures. Two methods of sensitivity analysis are used: in the first method the total information gain achieved by each parameter is evaluated. In the second method parameters and parameter sets are permuted and an error rate computed. This error rate is compared to values without permutation. This latter method allows the evaluation of the sensitivity of parameter combinations and thus gives an insight into the structure of the response surface. The examples demonstrate the capability of this methodology and stress the importance of the application of sensitivity analysis.

Visualizing probabilistic flood forecast information: expert preferences and perceptions of best practice in uncertainty communication

The aim of this article is to improve the communication of the probabilistic flood forecasts generated by hydrological ensemble prediction systems (HEPS) by understanding perceptions of different methods of visualizing probabilistic forecast information. This study focuses on interexpert communication and accounts for differences in visualization requirements based on the information content necessary for individual users. The perceptions of the expert group addressed in this study are important because they are the designers and primary users of existing HEPS. Nevertheless, they have sometimes resisted the release of uncertainty information to the general public because of doubts about whether it can be successfully communicated in ways that would be readily understood to nonexperts. In this article, we explore the strengths and weaknesses of existing HEPS visualization methods and thereby formulate some wider recommendations about the best practice for HEPS visualization and communication. We suggest that specific training on probabilistic forecasting would foster use of probabilistic forecasts with a wider range of applications. The result of a case study exercise showed that there is no overarching agreement between experts on how to display probabilistic forecasts and what they consider the essential information that should accompany plots and diagrams. In this article, we propose a list of minimum properties that, if consistently displayed with probabilistic forecasts, would make the products more easily understandable. Copyright

Challenges of operational river forecasting

Skillful and timely streamflow forecasts are critically important to water managers and emergency protection services. To provide these forecasts, hydrologists must predict the behavior of complex coupled human–natural systems using incomplete and uncertain information and imperfect models. Moreover, operational predictions often integrate anecdotal information and unmodeled factors. Forecasting agencies face four key challenges: 1)making themost of available data, 2)making accurate predictions usingmodels, 3) turning hydrometeorological forecasts into effective warnings, and 4) administering an operational service. Each challenge presents a variety of research opportunities, including the development of automated quality-control algorithms for the myriad of data used in operational streamflow forecasts, data assimilation, and ensemble forecasting techniques that allow for forecaster input, methods for using humangenerated weather forecasts quantitatively, and quantification of human interference in the hydrologic cycle. Furthermore, much can be done to improve the communication of probabilistic forecasts and to design a forecasting paradigm that effectively combines increasingly sophisticated forecasting technology with subjective forecaster expertise. These areas are described in detail to share a real-world perspective and focus for ongoing research endeavors. Open Access Content

Hyperresolution information and hyperresolution ignorance in modelling the hydrology of the land surface

There is a strong drive towards hyperresolution earth system models in order to resolve finer scales of motion in the atmosphere. The problem of obtaining more realistic representation of terrestrial fluxes of heat and water, however, is not just a problem of moving to hyperresolution grid scales. It is much more a question of a lack of knowledge about the parameterisation of processes at whatever grid scale is being used for a wider modelling problem. Hyperresolution grid scales cannot alone solve the problem of this hyperresolution ignorance. This paper discusses these issues in more detail with specific reference to land surface parameterisations and flood inundation models. The importance of making local hyperresolution model predictions available for evaluation by local stakeholders is stressed. It is expected that this will be a major driving force for improving model performance in the future.

The TIGGE Project and Its Achievements

AbstractThe International Grand Global Ensemble (TIGGE) was a major component of The Observing System Research and Predictability Experiment (THORPEX) research program, whose aim is to accelerate improvements in forecasting high-impact weather. By providing ensemble prediction data from leading operational forecast centers, TIGGE has enhanced collaboration between the research and operational meteorological communities and enabled research studies on a wide range of topics.The paper covers the objective evaluation of the TIGGE data. For a range of forecast parameters, it is shown to be beneficial to combine ensembles from several data providers in a multimodel grand ensemble. Alternative methods to correct systematic errors, including the use of reforecast data, are also discussed.TIGGE data have been used for a range of research studies on predictability and dynamical processes. Tropical cyclones are the most destructive weather systems in the world and are a focus of multimodel ensemble research. Their ...