The concept of ‘reverse’ flow routing in rivers is proposed, in which upstream flow hydrographs are specified from a downstream flow requirement. An example is given of reverse flow routing by the implicit method of solution of the equations of momentum and continuity. Reverse flow routing is further described in the context of the x‐t plane, and then the point is made that solutions were not found in all possible directions of computation in the x‐t plane. Solutions were found only in the directions of increasing time and decreasing distance (upstream).

Bank seepage occurs along natural streams bounded by an alluvial material when the passage of a flood wave produces a rise and fall in the stream stage. Owing to the difficulties of measuring flow rates in a porous medium, the quantity of bank seepage and its effect on the attenuation of a flood wave have not been determined from experimental observations. As an alternative for the purposes of this study, the equations of motion were used to describe flood flows in an idealized rectangular channel bounded by banks of a homogeneous, isotropic, porous medium in which the Boussinesq equation governs flow. The channel and the banks were assumed to be underlain by an impervious layer. Numerical simulation of a sinusoidal flood wave in the channel showed that bank seepage causes significant attenuation of the wave in narrow channels bounded by a relatively pervious material. This attenuation is characterized by a reduction in the hydrograph peak and sustenance of the flow during recession. In wide channels and in channels bounded by a relatively impervious material, no significant attenuation of the flood wave occurs.

The variability of dependable flows for fixed storage is an important parameter that should be incorporated into the economic analysis of a water resource storage project. Flow values representative of the James River were computed by a hydrologic model that uses data randomly generated from monthly precipitation distributions. Dependable flows from fixed storage were found to have large variance. Confidence limits were placed on the dependable flows illustrating the wide range of possible dependable flows for a specified reservoir size, and suggesting the desirability of expressing benefit-cost ratios for water resource projects in terms of confidence limits.

No abstract

No abstract