Schaarup-Jensen K, Rasmussen M R, Thorndahl S
Division of Water and Soil, Department of Civil Engineering, Aalborg University, Sohngaardsholmsvej 57, DK-9000 Aalborg, Denmark.
Water Sci Technol. 2009;60(1):87-95. doi: 10.2166/wst.2009.290.
In urban drainage modelling long-term extreme statistics has become an important basis for decision-making e.g. in connection with renovation projects. Therefore it is of great importance to minimize the uncertainties with regards to long-term prediction of maximum water levels and combined sewer overflow (CSO) in drainage systems. These uncertainties originate from large uncertainties regarding rainfall inputs, parameters, and assessment of return periods. This paper investigates how the choice of rainfall time series influences the extreme events statistics of max water levels in manholes and CSO volumes. Traditionally, long-term rainfall series, from a local rain gauge, are unavailable. In the present case study, however, long and local rain series are available. 2 rainfall gauges have recorded events for approximately 9 years at 2 locations within the catchment. Beside these 2 gauges another 7 gauges are located at a distance of max 20 kilometers from the catchment. All gauges are included in the Danish national rain gauge system which was launched in 1976. The paper describes to what extent the extreme events statistics based on these 9 series diverge from each other and how this diversity can be handled, e.g. by introducing an "averaging procedure" based on the variability within the set of statistics. All simulations are performed by means of the MOUSE LTS model.
在城市排水模型中，长期极端统计数据已成为决策的重要依据，例如与翻新项目相关的决策。因此，尽量减少排水系统中最高水位和合流制污水溢流（CSO）长期预测的不确定性至关重要。这些不确定性源于降雨输入、参数以及重现期评估方面的巨大不确定性。本文研究了降雨时间序列的选择如何影响检查井中最高水位和CSO流量的极端事件统计。传统上，本地雨量计的长期降雨序列不可用。然而，在本案例研究中，有长期且本地的降雨序列。两个雨量计在集水区内的两个地点记录了大约9年的降雨事件。除了这两个雨量计外，另外七个雨量计距离集水区最远20公里。所有雨量计都包含在1976年启动的丹麦国家雨量计系统中。本文描述了基于这9个序列的极端事件统计数据在多大程度上彼此不同，以及如何处理这种多样性，例如通过基于统计数据集中的变异性引入“平均程序”。所有模拟均通过MOUSE LTS模型进行。
