Forecasting water use in a constrained environment

Abstract

Traditional water use forecasting methodologies have been based primarily upon extrapolation of historical trends into the future, using either single or multidimensional regression analysis. However mathematically eloquent, these techniques are in many cases today totally inadequate for providing planners and decision makers with meaningful guidance on expected future water use. The primary reason for this lies with the pure and simple fact that in more and more cases within the United States, the ‘limits to growth’ concepts are beginning to be felt in both the heavy water using portions of the economy and the portions not related to water. The direct and indirect effects of this on future water use are impossible to predict with any confidence by using forecasting techniques based on the analysis and extrapolation of historical trends. To correct this, the following basic concepts should guide the development and utilization of future water use forecasting procedures.

Problems in Forecasting Water Requirements1

“…In order to overcome, at least partially, some of the inadequacies of time series extrapolation, forecasting of water requirements is frequently based on changes in the values of other variables over time (Howe and Linaweaver, 1967;Wollman and Bonem, 1971;Grima, 1973;Berry and Bonem, 1974;Weiss, 1974). For example, population, income, industrial growth, cost (including cost of pollution control), and so forth, are frequently used as explanatory variables in a multiple linear regression equation to predict water use.…”

“…It is particularly serious in the case of forecasting water requirements, in view of the limits on growth. Water requirements forecasting in a constrained environment must be integrated with decision-making and planning methodologies for allocating limited supply in the face of changing, generally increasing, competition for that supply (Weiss, 1974).…”