Abstract: Stormwater run-off generally refers to pluvial, i.e. rainfall related, water that does not soak into the ground at the point at which it falls. The volume and timing of stormwater run-off, specifically from roof tops is highly important to urban flood control and its capture has the potential for non-potable uses within (e.g. for WC flushing and for washing machines) and outside the home (e.g. car washing and garden watering). The former runs a risk of flash floods where local and downstream stormwater (or combined sewer) systems become overburdened in times of extreme rainfall events. The later will influence potential future urban water supplies, which is particularly important at time(s) where mains water availability is scarce (e.g. times of drought or when the national demand for water in the UK increases beyond supply capabilities) population. Rainwater harvesting (RWH) systems can benefit flood risk and water supply however their ability to do either / both is dependent on the subtleties of filling and emptying (i.e. stored water volume or spare storage capacity) which are not fully understood, particularly in peak flow events. Through the use of five years worth of daily rainfall data for Birmingham (2007 - a record breaking year for UK flooding, to 2011) these subtleties are investigated through a sensitivity type analysis of tank size, occupancy rates and technology efficiency. The results show that RWH tanks sized according to BS8515 would not have been capable of capturing rainfall that fell in peak flow events. Moreover not all yearly non-potable demands would have been met. If tanks were over-sized by a factor of 3.0 (i.e. use the larger of 15% yearly non-potable demands or rainfall) this would have been sufficient to meet all demands and eliminate roof-top run-off.