Time series analysis of hydrometeorological data

Analyses of hydrometeorological time series for practical purposes, for example to aid in the design of reservoirs for hydroelectric power generation, have generally assumed stationarity of the data. Similarly, climatologists have related their analyses to 30-year climate `normals'. The most obvious exception to this general assumption has occured when the effects of man on a watershed are so large as to force attention; for example, the various diversions into and out of the Great Lakes.Recently, attention has concentrated on a much larger scale anthropogenic effect, the `greenhouse warming' and it's anticipated change in climate and, hence, in water resources. Projections have forecast decreases in runoff of up to 50% resulting from an increase of 1 degree C in temperature.