One quote from an article about extreme drought in the LA Times attracts the attention on a water resources management method that is underused (exept in India?) but fast gaining ground and attracting increased inerest:
«Griggs cited rainwater harvesting and demand management as the least expansive options for increasing water supplies. Pipelines and dams were among the most expansive options he said.
"Urban storm water is a large untapped source of water generated close to where it is needed. ...In most Australian cities, as much water falls on that city as the city needs", Griggs said.
In Queensland, Australia's fastest-growing state, with 2.7 million residents, about 20% of the population has installed rain-catchement tanks since 2006, when the area received just 7.4% of its average annual inflow to the major dam that supplies it. In 2007, that flow had declined to just 4%.»
In these 3 paragraphs, it's possible to find all the reasons that will make rainwater harvesting a very important component of all water management regimes in the near future: 1) supply side oriented (allowing overall water use increase to some degree); 2) cheapest; 3) locally managed.
In the last few years, rainwater harvesting has already attracted interest in US Western states and Texas. Some German cities are metering residential storm runnoffs to encourage rain harvesting and domestic use (toilet flushes, laundry...). New Building regulations may force rainwater harvesting in large building project in the South of England...
In short, rainwater harvesting will become a big issue in water law and management, especially since it may result in conflicts with prior downstram users and environmental uses.