Access to daily stream flow data, at the river reach scale is a central component of many aspects of water resource and water quality management. However, the majority of river reaches within the UK are ungauged. Hence, there is an operational requirement for a quick, consistent and reliable method for simulating historical stream flow records within ungauged catchments. The overall objective of this research has been to develop a rainfall runoff model for predicting natural daily stream flows within a catchment without recourse to the calibration of model parameters against observed stream flow data. Implicit within this objective is a requirement that the model parameters can be estimated from readily available data describing the physical characteristics of the catchment. The fundamental approach to the research has been to develop and calibrate suitable models within a large, representative sample of UK catchments and to subsequently develop predictive, statistical relationships for estimating model parameters from the climatic and physiographic characteristics of the catchments. The predictive capacity of the regionalised model forms has been extensively evaluated through comparisons with gauged flow data, calibrated models and existing industry-standard methods for estimating historical flow time series within ungauged catchments. The regionalised model forms developed represent a significant advance over existing, low-cost methods for estimating historical flow regimes within ungauged river catchments. The errors in the simulated stream flows are sufficiently small for the techniques to be a useful aid in the management of water resources within the UK.