Water supply planning has traditionally relied on infrastructure systems in which the risk and uncertainty of reliabile supplies is characterized as a “safe yield”. Modern design criteria grounded in variations of benefit-cost analysis also identify the need to manage residual risk in decision making. A systems perspective recognizes the need to manage groundwater and surface water as a single resource. By definition the water balance is “fully allocated” so new demands can only be accommodated by altering existing uses; increased water appropriations reduce system resilience for all users. Sustainable water uses must therefore be viewed not as an appropriation with “zero impact”, but in terms of acceptable impacts – i.e. acceptable degradation of the hydrologic services realized from the climatic water budget. The challenge of managing sustainable water supplies is to match supply and demand in time and space, with acceptable impacts among uses.
Increased pressure on limited supplies combined with uncertain projections of non-stationary hydroclimatic change threatens sustainability through both shortages that decrease reliability, and through increased competition to appropriate growing shares of a limited supply. Supplies and reliability can be enhanced – at cost- through interbasin transfers, technologies for the recycling, reuse, and ultimately desalination of brackish groundwater and estuarine surface water supplies, and aggressive treatment and reuse of wastewater. Given these economic alternatives, simple decision heuristics and hydrologic metrics - such as “safe yield” or ecosurplus- are insufficient to inform the sustainable management of water supplies.
The need to manage residual risk is framed as the inherent challenge of planning for “failure”, and illustrated with contingency planning for sustainable water supplies under uncertainty. Contingency planning provides resilience for decision making under uncertainty, empowering a decomposition of decision making into traditional decisions under risk, and the management of residual risk under uncertainty – including the management of acceptable states of system failure and recovery. Examples from water supply planning and instream flow management illustrate the use – and misuse – of simple decision heuristics, framing the pportunities for science to inform planning and policy decisions for managing sustainble water supplies in the face of uncertainty.