This study presents a novel approach for urban flood forecasting in drainage systems using a dynamic ensemble-based data mining model which has yet to be utilised properly in this context. The proposed method incorporates an event identification technique and rainfall feature extraction to develop weak learner data mining models. These models are then stacked to create a time-series ensemble model using a decision tree algorithm and confusion matrix-based blending method. The proposed model was compared to other commonly used ensemble models in a real-world urban drainage system in the UK. The results show that the proposed model achieves a higher hit rate compared to other benchmark models, with a hit rate of around 85% vs 70 % for the next 3 h of forecasting. Additionally, the proposed smart model can accurately classify various timesteps of flood or non-flood events without significant lag times, resulting in fewer false alarms, reduced unnecessary risk management actions, and lower costs in real-time early warning applications. The findings also demonstrate that two features, “antecedent precipitation history” and “seasonal time occurrence of rainfall,” significantly enhance the accuracy of flood forecasting with a hit rate accuracy ranging from 60 % to 10 % for a lead time of 15 min to 3 h.

@en

With the increase of the frequency and severity of flooding events, coupled with population growth, the risks posed to people from flooding is ever more apparent. This paper proposes a methodology to examine the risks posed to vehicles' occupants and pedestrians simultaneously in an urban context. Through considering stability functions of a range of vehicle types and pedestrian, a risk assessment profile for a vehicle occupant was derived. Using a historical 1-in-20-year rainfall flood event that took place in the city of Exeter (UK) in 2014, and a synthetic 1 in 100-year rainfall flood event, the potential risks posed to vehicle occupants were analysed. The results showed that for these events the potential risks posed to people travelling by car and caught in flood waters were likely to be more severe if they were to remain within their vehicles than if they were to exit said vehicles. Analysis of the changes in risk over time further revealed that if a vehicle was to become immobilised in flood water, they would only have a short timeframe (~10 min) before the level of risk increases. This is a critical finding, highlighting that remaining inside an immobilised vehicle during flood event and waiting for assistance may increase the level the risk the individual is exposed to, with the results showing the significance of such studies in reducing the risk of flooding to people.

@en

Urban flooding is a major problem for cities around the world, with significant socio-economic consequences. Conventional real-time flood forecasting models rely on continuous time-series data and often have limited accuracy, especially for longer lead times than 2 hrs. This study proposes a novel event-based decision support algorithm for real-time flood forecasting using event-based data identification, event-based dataset generation, and a real-time decision tree flowchart using machine learning models. The results of applying the framework to a real-world case study demonstrate higher accuracy in forecasting water level rise, especially for longer lead times (e.g., 2–3 hrs), compared to traditional models. The proposed framework reduces root mean square error by 50%, increases accuracy of flood forecasting by 50%, and improves normalised Nash–Sutcliffe error by 20%. The proposed event-based dataset framework can significantly enhance the accuracy of flood forecasting, reducing the occurrences of both false alarms and flood missing and improving emergency response systems.

@en

This paper proposes a novel methodology for modelling the impacts of floods on traffic. Often, flooding is a complex combination of various causes (coastal, fluvial and pluvial). Further, transportation systems are very sensitive to external disturbances. The interactions between these two complex and dynamic systems have not been studied in detail so far. To address this issue, this paper proposes a methodology for a dynamic integration of a flood model (MIKE FLOOD) and a microscopic traffic simulation model (SUMO). The flood modelling results indicate which roads are inundated for a period of time. The traffic on these links will be halted or delayed according to the flood characteristics—extent, propagation and depth. As a consequence, some of the trips need to be cancelled; some need to be rerouted to unfavourable routes; and some are indirectly affected. A comparison between the baseline and a flood scenario yields the impacts of that flood on traffic, estimated in terms of lost business hours, additional fuel consumption and additional CO2 emissions. The proposed methodology will be further developed as a workable tool to evaluate the flooding impact on transportation network at city scale automatically.@en