Reliable flood damage assessment is important for decision-making in flood risk management.Flood damage assessment is often done with damage curves based only on water depth. These depthdamage curves are usually developed based on data from a specific location and specific flood conditions. Such depth-damage curves tend to be applied outside the scope of their validity. Validation studies show that in such cases depth-damage curve are not very reliable, probably due to excluded influencing variables. The expectation is that the inclusion of more variables in a damage function will improve its transferability. We compare multi-variable models based on Bayesian Networks and Random Forests developed on the basis of flood damage data sets from Germany and The Netherlands. The performance of the models is tested on a validation sub-set of both countries' data. The models are also updated with data from the other country and then tested again. The results show that the German models (BN/RF-FLEMOps) perform better in the Netherlands than the Dutch models (BN/RF-Meuse) perform in Germany. This is probably because the FLEMOps models are based on more heterogeneous data than the Meuse models. The FLEMOps models, therefore, are better able to capture damages processes from other events and in other locations. Model performance improves via updating the models with data from the location to which the model is transferred to. The results show that there is high potential to develop improved damage models, by training multi-variable models with heterogeneous data, for example from multiple flood events and locations.

Abstract. Flood damage assessment is usually done with damage curves only dependent on the water depth. Several recent studies have shown that supervised learning techniques applied to a multi-variable data set can produce significantly better flood damage estimates. However, creating and applying a multi-variable flood damage model requires an extensive data set, which is rarely available, and this is currently holding back the widespread application of these techniques. In this paper we enrich a data set of residential building and contents damage from the Meuse flood of 1993 in the Netherlands, to make it suitable for multi-variable flood damage assessment. Results from 2-D flood simulations are used to add information on flow velocity, flood duration and the return period to the data set, and cadastre data are used to add information on building characteristics. Next, several statistical approaches are used to create multi-variable flood damage models, including regression trees, bagging regression trees, random forest, and a Bayesian network. Validation on data points from a test set shows that the enriched data set in combination with the supervised learning techniques delivers a 20 % reduction in the mean absolute error, compared to a simple model only based on the water depth, despite several limitations of the enriched data set. We find that with our data set, the tree-based methods perform better than the Bayesian network.

Abstract. This paper addresses the large differences that are found between damage estimates of different flood damage models. It explains how implicit assumptions in flood damage functions and maximum damages can have large effects on flood damage estimates. This explanation is then used to quantify the uncertainty in the damage estimates with a Monte Carlo analysis. The Monte Carlo analysis uses a damage function library with 272 functions from seven different flood damage models. The paper shows that the resulting uncertainties in estimated damages are in the order of magnitude of a factor of 2 to 5. The uncertainty is typically larger for flood events with small water depths and for smaller flood events. The implications of the uncertainty in damage estimates for flood risk management are illustrated by a case study in which the economic optimal investment strategy for a dike segment in the Netherlands is determined. The case study shows that the uncertainty in flood damage estimates can lead to significant over-or under-investments.

Abstract:The Netherlands has just finished implementing the Room for the Rivers program along the Rhine and Meuse Rivers in response to increasing river discharges. Recently, making more room for the river is, however, being challenged for future application because the flood defenses are assessed to be too weak and will need reinforcement anyway. To be able to decide on the most desirable policy for the remainder of the century, we require knowledge of all benefits and costs of individual interventions and strategic alternatives for flood mitigation. In this paper, we quantify some benefits of making more room for the rivers. We recognize and quantify two risk-reducing effects and provide results of analyses for the Rhine and Meuse Rivers in The Netherlands. Making room for rivers was originally advocated because it (1) reduces the consequences of flooding, as well as (2) reduces the probability of failure of the embankments. We have now quantified these effects allowing translation into risk reduction proper. Moreover, larger floodplain surface area may influence the relationship between discharge and flood level, which implies that rivers with widened floodplains are less sensitive to uncertainties about future river discharges. This does not reduce risk proper, but makes the river system more robust, as we shall argue in the discussion where we present risk reduction and robustness as complementary perspectives for assessing strategic alternatives for flood risk management.

Abstract. This paper addresses the large differences that are found between damage estimates of different flood damage models. It explains how implicit assumptions in flood damage models can lead to large uncertainties in flood damage estimates. This explanation is used to quantify this uncertainty with a Monte Carlo Analysis. As input the Monte Carlo analysis uses a damage function library with 272 functions from 7 different flood damage models. This results in uncertainties in the order of magnitude of a factor 2 to 5. The resulting uncertainty is typically larger for small water depths and for smaller flood events. The implications of the uncertainty in damage estimates for flood risk management are illustrated by a case study in which the economic optimal investment strategy for a dike segment in the Netherlands is determined. The case study shows that the uncertainty in flood damage estimates can lead to significant over- or under-investments.

Abstract. Increasing amounts of data, together with more computing power and better machine learning algorithms to analyse the data, are causing changes in almost every aspect of our lives. This trend is expected to continue as more data keep becoming available, computing power keeps improving and machine learning algorithms keep improving as well. Flood risk and impact assessments are also being influenced by this trend, particularly in areas such as the development of mitigation measures, emergency response preparation and flood recovery planning. Machine learning methods have the potential to improve accuracy as well as reduce calculating time and model development cost. It is expected that in the future more applications will become feasible and many process models and traditional observation methods will be replaced by machine learning. Examples of this include the use of machine learning on remote sensing data to estimate exposure and on social media data to improve flood response. Some improvements may require new data collection efforts, such as for the modelling of flood damages or defence failures. In other components, machine learning may not always be suitable or should be applied complementary to process models, for example in hydrodynamic applications. Overall, machine learning is likely to drastically improve future flood risk and impact assessments, but issues such as applicability, bias and ethics must be considered carefully to avoid misuse. This paper presents some of the current developments on the application of machine learning in this field and highlights some key needs and challenges.

Residential assets, comprising buildings and household contents, are a major source of direct flood losses. Existing damage models are mostly deterministic and limited to particular countries or flood types. Here, we compile building-level losses from Germany, Italy and the Netherlands covering a wide range of fluvial and pluvial flood events. Utilizing a Bayesian network (BN) for continuous variables, we find that relative losses (i.e. loss relative to exposure) to building structure and its contents could be estimated with five variables: water depth, flow velocity, event return period, building usable floor space area and regional disposable income per capita. The model’s ability to predict flood losses is validated for the 11 flood events contained in the sample. Predictions for the German and Italian fluvial floods were better than for pluvial floods or the 1993 Meuse river flood. Further, a case study of a 2010 coastal flood in France is used to test the BN model’s performance for a type of flood not included in the survey dataset. Overall, the BN model achieved better results than any of 10 alternative damage models for reproducing average losses for the 2010 flood. An additional case study of a 2013 fluvial flood has also shown good performance of the model. The study shows that data from many flood events can be combined to derive most important factors driving flood losses across regions and time, and that resulting damage models could be applied in an open data framework.

No abstract