Rainwater harvesting (RWH) has become a popular water saving and conservation means across the globe. In many cities, a pre-determined rainwater tank size is generally selected without a detail water balance modelling at the location on interest. However, to achieve maximum benefit from a RWH system at a given location, it is desirable to design the system based on the site specific conditions (e.g. local rainfall and loss characteristics) and other relevant design parameters. There have been limited studies on regionalization of RWH system, which can account for the spatial variability in rainfall, loss and water demand characteristics over a given region to provide site-specific design of RWH system. This paper presents a RWH regionalization approach to develop a set of design curves, which assist in the selection of an adequate RWH system at a given location. We use rainfall data from 159 rain gauges across the Greater Sydney region in Australia to develop and test the proposed regionalization approach. It has been found that there is a significant variation in rainfall characteristics across Greater Sydney region and that a common tank size across this region does not present an adequate solution. It has been found that a 5 kL tank can achieve a reliability (i.e. percentage of days the RWH can meet the given demand) of 31–93% for four people depending on the location in the Greater Sydney region and the types of water use. The developed regional design curves could result in increased water savings across Greater Sydney. The proposed regionalization approach can be adapted to other cities across Australia and the world which present a high rainfall gradient.

雨水收集（RWH）已成为全球流行的节水和节约用水手段。在许多城市中，通常会选择一个预定的雨水箱尺寸，而在目标位置没有详细的水平衡模型。但是，为了在给定位置从RWH系统中获得最大收益，希望基于站点的特定条件（例如，本地降雨和损失特征）和其他相关设计参数来设计系统。关于RWH系统区域化的研究很少，它可以解释给定区域内降雨，损耗和需水特征的空间变异性，以提供针对特定地点的RWH系统设计。本文提出了RWH区域化方法，以开发一组设计曲线，有助于在给定位置选择适当的RWH系统。我们使用来自澳大利亚大悉尼地区159个雨量器的降雨数据来开发和测试建议的区域化方法。已经发现，大悉尼地区的降雨特征存在显着差异，并且该地区的共用水箱尺寸不能提供适当的解决方案。已经发现，根据大悉尼地区的位置和用水类型的不同，一个5 kL的水箱可以使四人的可靠性（即RWH可以满足给定需求的天数）达到31–93％。发达的区域设计曲线可能会导致大悉尼地区节水的增加。