Inter-basin transfers as a supply option: the end of an era?

Large hydraulic projects have underlain the success of ancient civilizations, and they are part of humankind's water culture. However, most of them have been confined to a single river-basin. The aqueducts of the Roman Empire are a notable exception, and it remains fascinating that the Romans spent so much effort to build such impressive and long-lasting hydraulic architecture, but then, in the end, to transfer so little water. It would be interesting to link this with the economy of the time, e.g. looking at the slavery factor. In our capitalist economy, the value of labor and the depreciation of capital over time make such projects impossible to finance. And because water is heavy, hydraulic infrastructure usually comprises specific assets with high investment costs and slow technical depreciation. Yet in order to provide clean water to cities, the idea resurfaced of transferring water over long distances, an idea that returned in the classical period and rose during the 19th and 20th centuries. Additionally, the invention of the steam engine, of sealed pipe junctions, and also of tight-closing faucets, made it possible to raise water or build siphons and develop IBTs at an unprecedented scale. However, at the turn of the 20th century, water treatment was invented. This allowed a reduction in the initial investment cost by increasing operation and maintenance costs. In turn, this new engineering based on water treatment and sanitation made it possible to extract water from nearby surface water; it made urban water much more a local affair, and shifted the burden of finance from the state or rich entrepreneurs to the customers. From that point on, control of water volumes became essential, and this did not favor large IBTs, where the bigger the flow, the smaller the marginal cost. Paradoxically, IBTs still fared very well in the colonies, and later in developing nation-states, for intensive agriculture: the motto after WW2 was multipurpose water projects for global economic development, meaning that industrial and urban water users would indirectly subsidise agriculture. But as soon as environmental impacts were translated into dollars, and the global economy became less confident in the future (thus increasing interest rates), many IBTs were weakened. Their future would have been better assured if they had been re-integrated into some regional water resources management scheme, perhaps associated with the conjunctive use of ground water and surface water. Large IBTs have been frequently associated with the power of upper layers of government, especially centralised ones. At a time when multi-level governance is on the global agenda, it becomes difficult to believe that transferring large quantities of water suffices to improve humankind's welfare: it has to be combined with less conventional technologies and other institutional designs.