September 2008 Morpeth Flood addition - Performance of ensemble rainfall forecasts in relation to pluvial flooding impacts

在2007年的大洪水之后，联合王国政府委托编写了一份报告（《皮特评论》），其中制定了应对这一问题所需的行动。该报告的一项关键建议是需要“改进所有形式的建模”。到目前为止，大多数注意力都集中在开发河流洪水（河流洪水）的计算机模型上，但很大一部分洪水来自降雨事件，导致局部径流（洪水）。这些洪水事件很难预测，因为它们通常具有局部高强度降雨（通常与对流风暴有关），并通过地下排水沟和陆地在城市地区径流。新版本的Met。Office的业务天气预报模型最近开发了高分辨率（1.5公里），能够更准确地代表这些当地降雨事件，有可能用于支持更好地预测洪水。该提案旨在将新模型的输出用于9月6日的案例研究活动。2008年莫尔佩斯洪水影响了大约1,000处房产。这个案例研究是特别有价值的，因为它是在以前的项目中从公开来源的信息详细测量洪水动态的极少数事件之一，这使得可以详细重建不同的洪水来源，以便与洪水模型进行比较。1.5公里天气预报模型的详细降雨数据的质量为9月。2008年莫尔佩斯洪水事件将通过与雨量计观测和英国目前使用的其他标准方法进行比较进行评估，包括评估降雨数据的不确定性。然后，这些数据将被输入局部集水区径流的水文模型，并用于模拟流入城镇的水量，以及洪水如何到达洪泛区的不同区域，从而与此次事件的观测洪水位进行比较。该案例研究是第一个使用新的天气预报模型降雨预测，包括洪水建模的不确定性，并将有助于支持洪水预测模型的发展，从英国各地的洪积源。

Following the major floods in 2007, the UK Government commissioned a report (the Pitt Review) which formulated the actions needed to respond to this issue. A key recommendation of the report was the need for 'improved modelling of all forms of flooding'. Most attention so far has been paid to developing computer models of flooding from rivers (fluvial flooding), but a significant percentage of flooding is from rainfall events which lead to localised runoff (pluvial flooding). These flood events are difficult to predict as they are often have local high intensity rainfall (typically associated with convective storms) and runoff in urban areas through subsurface drains as well as overland. A new version of the Met. Office's operational weather forecast model has recently been developed with a high (1.5km) resolution that is capable of representing these local rainfall events more accurately, with the potential to be used to support better predictions of pluvial flooding. This proposal aims to use the outputs from the new model for the case study event of the 6th Sept. 2008 Morpeth flood that affected around 1,000 properties. This case study is particularly valuable, as it is one of the very few events for which the dynamics of the flooding have been measured in detail in a previous project from publicly-sourced information, which has allowed detailed reconstruction of the different sources of flooding for comparison against flood models. The quality of the detailed rainfall data from the 1.5km weather forecast model for the Sept. 2008 Morpeth flood event will be assessed through comparison with raingauge observations and other standard approaches currently used in the UK, including assessment of the uncertainty in the rainfall data. These data will then be fed into hydrological models of runoff from the localised catchment areas, and used to simulate the volumes of water flowing into the town, and how the floodwater reaches the different areas of the floodplain, allowing comparison against the observed flood levels for this event. This case study is the first to use the new weather forecast model rainfall predictions including uncertainty for flood modelling, and will help to support development of flood prediction modelling from pluvial sources across the UK.