a b s t r a c tThe hydraulic modelling of tidal estuarine environments has been largely limited to complex 3D models that are computationally expensive. This makes them unsuitable for applications which make use of live data to make real/near time forecasts, such as the modelling of storm surge propagation and associated flood inundation risks. To address this requirement for a computationally efficient method a reduced complexity, depth-integrated 2D storage cell model (Lisflood-FP) has been applied to the Humber Estuary, UK. The capability of Lisflood-FP to reproduce the tidal heights of the Humber Estuary has been shown by comparing modelled and observed tidal stage heights over a period of a week. The feasibility of using the Lisflood-FP model to forecast flood inundation risk from a storm surge is demonstrated by reproducing the major storm surge that struck the UK East Coast and Humber Estuary on 5 December 2013. Results show that even for this 2013 extreme event the model is capable of reproducing the hydraulics and tidal levels of the estuary. Using present day flood defences and observed flooding extents, the modelled flood inundation areas produced by the model were compared, showing agreement in most areas and illustrating the model's potential as a now-casting early warning system when driven by publically available data, and in near real-time. The Lisflood-FP model used was incorporated into the CAESAR-Lisflood GUI, with the calibration and verification of the estuarine hydraulics reported herein being a key step in creating an estuary evolution model, capable of operating in the decadal to century timescales that are presently underrepresented in estuarine predictive capability, and ultimately developing a model to predict the evolution of flood risk over the longer term.

Decision-making in flood risk management is increasingly dependent on access to data, with the availability of data increasing dramatically in recent years. We are therefore moving towards an era of big data, with the added challenges that, in this area, data sources are highly heterogeneous, at a variety of scales, and include a mix of structured and unstructured data. The key requirement is therefore one of integration and subsequent analyses of this complex web of data. This paper examines the potential of a data-driven approach to support decision-making in flood risk management, with the goal of investigating a suitable software architecture and associated set of techniques to support a more data-centric approach. The key contribution of the paper is a cloud-based data hypercube that achieves the desired level of integration of highly complex data. This hypercube builds on innovations in cloud services for data storage, semantic enrichment and querying, and also features the use of notebook technologies to support open and collaborative scenario analyses in support of decision making. The paper also highlights the success of our agile methodology in weaving together cross-disciplinary perspectives and in engaging a wide range of stakeholders in exploring possible technological futures for flood risk management.

<p><strong>Abstract.</strong> The inclusion of uncertainty in flood forecasts is a recent, important yet challenging endeavour. In the chaotic and far from certain world we live in, probabilistic estimates of potential future floods are vital. By showing the uncertainty surrounding a prediction, probabilistic forecasts can give an earlier indication of potential future floods, increasing the amount of time we have to prepare. In practice, making a binary decision based on probabilistic information is challenging. The Environment Agency (EA), responsible for managing risks of flooding in England, is in the process of a transition to probabilistic fluvial flood forecasts. A series of interviews were carried out with EA decision-makers (i.e. duty officers) to understand how this transition might affect their decision-making activities. The interviews highlight the complex and evolving landscape (made of alternative <q>hard scientific facts</q> and <q>soft values</q>) in which EA duty officers operate, where forecasts play an integral role in decision-making. While EA duty officers already account for uncertainty and communicate their confidence in the system they use, they view the transition to probabilistic flood forecasts as both an opportunity and a challenge in practice. Based on the interview results, recommendations are made to the EA to ensure a successful transition to probabilistic forecasts for flood early warning in England.</p> <p>We believe that this paper is of wide interest for a range of sectors at the intersection between geoscience and society. A glossary of technical terms is highlighted by asterisks in the text and included in Appendix A.</p>

No abstract

Abstract. By showing the uncertainty surrounding a prediction, probabilistic forecasts can give an earlier indication of potential upcoming floods, increasing the amount of time available to prepare. However, making a decision based on probabilistic information is challenging. As part of the UK-wide policy's move towards forecast-based flood risk management, the Environment Agency (EA), responsible for managing risks of flooding in England, is transitioning towards the use of probabilistic fluvial forecasts for flood early warning. While science and decision-making are both individually progressing, there is still a lack of an ideal framework for the incorporation of new and probabilistic science into decision-making practices, and, respectively, the uptake of decision-makers' perspectives in the design of scientific practice. To address this, interviews were carried out with EA decision-makers (i.e. Duty Officers), key players in the EA's flood warning decision-making process, to understand how they perceive this transition might impact on their decision-making. The interviews highlight the complex landscape in which EA Duty Officers operate and the breadth of factors that inform their decisions, in addition to the forecast. Although EA Duty Officers already account for uncertainty and communicate their confidence in the forecast they currently use, the interviews revealed a decision-making process which is still very binary and linear to an extent, which appears at odds with probabilistic forecasting. Based on the interview results, we make recommendations to support a successful transition to probabilistic forecasting for flood early warning in England. These recommendations include the new system's co-design together with Duty Officers, the preparation of clear guidelines on how probabilistic forecast should be used for decision-making in practice, EA communication with all players in the decision-making chain (internal and external) that this transition will become operational practice and the documentation of this transition to help other institutes yet to face a similar challenge. We believe that this paper is of wide interest for a range of sectors at the intersection between geoscience and society. A glossary of technical terms is highlighted by asterisks in the text and included in Appendix A.