Probabilistic Urban Flood Forecasting System for Chinese Mega Cities Using Advanced Active Phased Array Radar

This study aims to bring together academics, industrial partners and government agency from UK and China to conduct a pilot study on the use of the active phased array radar to provide early urban flood warnings to Chinese mega cities. This is the first such project in China of using the cutting-edge active phase array radar (APRA) to provide rainfall monitoring and storm warning information for cities at large scale. The X-band, dual-polarized active phase array radars (APAR) are one of the latest developments of radar rainfall measurements technology, aiming to provide very high spatial-temporal resolution (30metres/30secs), easy and cost-effective maintenance and suitable for covering small to medium, built-up areas. The technological advances of APARs, lies in not only their ability of much refined precipitation distributions, but their potential to be used as a real-time driver for model-based probabilistic urban flood forecasting/warning systems.

In this paper, we present a framework of APAR based urban flood modelling approach that can utilize and maximize the advantage of APRA to support the research on urban flooding impact triggered by localized extreme storms. The framework is intended to achieve 1) Appraisal of APAR based QPE and short-range forecasts for urban flood modelling warnings; 2) Assessment of the system in the context of operational use; 3) providing best-practice recommendations to both local government and the industry in the use of APAR; 3) Supporting probabilistic ensembles to represent the uncertainties associated with the nowcasts; 4) Based upon the research outcome of the project, advocating government agencies at various levels and influence policy making on the uptake of new technology in urban flood risk management. An example of such system is discussed with the implementation using Short Term Ensemble Prediction System (STEPS) to obtain radar QPFs and ensemble nowcasting, coupled with ICM Live Model, the urban flood inundation model provided by the HR Wallingford, to generate real-time pluvial simulation on local scale. Further, by embedding into the existing national radar network and further coupled with numerical weather prediction, the APARs can offer extra lead time up to days, this is also important in the context of urban flood warning.