While the practice of rainwater harvesting (RWH) can be traced back millennia, the degree of its modern implementation varies greatly across the world, often with systems that do not maximize potential benefits. With a global focus, the pertinent practical, theoretical and social aspects of RWH are reviewed in order to ascertain the state of the art. Avenues for future research are also identified. A major finding is that the degree of RWH systems implementation and the technology selection are strongly influenced by economic constraints and local regulations. Moreover, despite design protocols having been set up in many countries, recommendations are still often organized only with the objective of conserving water without considering other potential benefits associated with the multiple-purpose nature of RWH. It is suggested that future work on RWH addresses three priority challenges. Firstly, more empirical data on system operation is needed to allow improved modelling by taking into account multiple objectives of RWH systems. Secondly, maintenance aspects and how they may impact the quality of collected rainwater should be explored in the future as a way to increase confidence on rainwater use. Finally, research should be devoted to the understanding of how institutional and socio-political support can be best targeted to improve system efficacy and community acceptance.

A sampling campaign was conducted in order to characterise the quality and quantity of individual domestic greywater streams. Based on the results, various scenarios of inclusion and / or exclusion of different greywater streams were explored, and their implication for on-site greywater treatment and reuse options are discussed. Domestic greywater was found to contribute as much as 55-70% of the specific daily load of TSS and BOD_t in municipal sewage. The kitchen sink was signalled out as a major contributor of VSS, COD_t, and BOD_t with 58%, 42% and 48%, of their total daily load respectively. The washing machine was established as a significant contributor of sodium, phosphate and COD_t (40%, 37% and 22% of the total load). The dishwasher, although contributing only 5% of the flow, was found to be a significant contributor of phosphate and boron. The wash basin was found to be the least polluting appliance. As "demand" for greywater within the urban environment is lower than its "production", it is logical to recycle only the less polluted greywater streams. In order to explore the consequences of the above concept on discharge volume, pollutants loads and concentrations, 18 scenarios were studied, in each at least one stream was excluded from the combined greywater stream. Exclusion of the joined stream of the kitchen sink plus the highly polluted streams of the washing machine (wash + 1^st rinse) and dishwasher (pre-rinse + wash) significantly improved greywater quality, with the advantage of leaving enough greywater to be reused (65-70 l/c/d).