Integrating Supply Uncertainties from Stochastic Modeling into Integrated Water Resource Management: Case Study of the Saskatchewan River Basin

Abstract

A warming climate and land management intensification have altered water supply characteristics in many regions of the world. Incorporation of water supply uncertainties into long-term water resources planning and management is, therefore, significant from both a scientific and societal perspective. This study proposes a set of analyses for integrated water resources management under changing water supply and demand expansion based on a newly developed methodology for vulnerability assessment. The basin of interest for the proposed analysis is the interprovincial Saskatchewan River Basin (SaskRB) in Canada, which supports a wide range of water demands, from municipal and industrial use to irrigated agriculture and hydropower. Proposals for an increase in irrigated area are used as a context for exploring the joint effects of current and future water supply uncertainty and increasing irrigation demand conditions on the water resources system. Changing water supply conditions are represented by perturbing annual volumes and the seasonal timing of the hydrograph peak as input to an integrated water resources model. The analysis enables evaluation of the effects of economic development plans as well as variations in volume and peak timing of flows on water availability and economic productivity, including possibilities for failure to meet demands. Results for the SaskRB show that a large increase in irrigated agriculture raises average net revenues, but these are highly dependent on water supply conditions and loss of revenue may arise under drought conditions. Hydropower production is more sensitive to changes in annual inflow volume than to changes in either annual timing of the peak flow or the magnitude of irrigation expansion. Irrigation expansion can considerably affect the peak flows in the Saskatchewan River Delta, the largest inland delta in North America, during low-flow conditions. For example, a 400% increase in irrigated area under a 25% decrease in inflow volume and 4-week-earlier annual peak timing can reduce the frequency of peak flows in the delta by more than 50%, though potential effects on the riparian and aquatic ecosystems remain uncertain. This case study illustrates the practical utility of stochastic analysis of system vulnerability to feasible futures in such a way that socioeconomic trade-offs can be readily visualized and understood. Such performance assessments are useful for long-term water resources planning and management under water supply uncertainty.