Journal of Hydraulic Engineering

This paper presents a new algorithm to identify watershed features like channels, ridges, and overland flow planes from a digitized elevation‐location data of discrete points of a catchment. The lateral inflow to the channels are the outflow from the planes bordering the channels. The flow of water is shown to follow distinct paths across these planes and an algorithm is developed to find out the slope, width, average length, and number of the cascade of planes contributing to a particular channel segment. A model catchment is used to demonstrate the steps of the algorithm and a real catchment is also analyzed to demonstrate the application of the algorithms developed. The data generated can be easily stored to be used in numerically based surface runoff models.

Prediction of land subsidence in a heavy‐snowfall area is one of the most important problems in the determination of optimum use of ground water for melting snow. In this study, land subsidence is predicted by using a regression‐equation model that includes past land subsidence, ground‐water level, and snow‐fall. The components of the long‐period variation and the short‐period variation in land subsidence are predicted by the parameters using the least‐square method and Kaiman filtering, respectively. The memory length of the regression‐equation model is justified by comparing the value of Akaike's information criterion. The proposed method has accurately predicted land subsidence one and two months ahead in snow country in Japan. The input data for this prediction model (regression equations) are past land subsidence, ground‐water level, and snowfall. This method to predict land subsidence is applicable to any snow country in the world.

A modification to the well‐known Beam and Warming implicit scheme is proposed and is applied to one‐dimensional unsteady free‐surface flows. The proposed modification is based on the concept of conservative splitting of flux through an approximate Jacobian. The conservative evaluation of flux vector at known‐time level in the Beam and Warming scheme, hitherto evaluated nonconservatively, improves the accuracy of the solution and effectively eliminates mass balance error. Computational results of a number of illustrative examples are presented and compared with solutions of the original Beam and Warming scheme as well as with analytical solutions. The modified model significantly improves accuracy of results at almost no additional complication or cost of computation.

The Hvorslev method constitutes a rapid means of characterizing the hydraulic conductivity of shallow ground‐water systems. Because the method is widely used, it is of practical benefit to know the limits and accuracy of the method. An important aspect of the Hvorslev analysis is that the role of specific storage $\left(S_s\right)$ is completely ignored. Results of this study show that the Hvorslev method provides acceptable estimates of hydraulic conductivity in the case of zero‐penetration boreholes in which water enters only from the borehole bottom. In this case, relatively large errors arise due to uncertainty in the intake geometry (planate or hemispherical intake) than those arising from neglecting $S_s$. Results presented show that in screened boreholes, in which radial and spherical flow symmetries coexist, $S_s$ cannot be reasonably neglected. Due to the influence of $S_s$, the pressure transient data deviates significantly from the semilogarithmic straight line intrinsic to the Hvorslev idealization, with the deviation increasing with increasing $S_s$. A consequence is that the estimation of $K$ either by the application of the Hvorslev formula or by the basic time‐lag method is rendered ambiguous. For these cases, it appears advisable to dispense with the Hvorslev‐interpretation method and analyze the data using the slug‐test techniques of ground‐water hydrology that account for $S_s$. This strategy of interpretation does not only yield a better estimate of hydraulic conductivity, but also yields one additional hydraulic parameter, namely, $S_s$.

A model for an instantaneous‐unit sediment graph (IUSG), based on attenuation and translation functions of mobilized sediment has been developed and applied to a Chaukhutia watershed (452.25 km²) of the Ramganga reservoir catchment in Northern India. An IUSG is a sediment graph resulting from one cm of mobilized sediment generated uniformly over the basin area during the effective rainfall of infinitesimally small duration. The catchment representative IUSG was converted into a unit sediment graph and convolved with mobilized sediment for generation and prediction of sediment graphs. The predicted sediment graphs showed good agreement with their observed counterparts except for lower peak sediment‐flow rates. The changes in hydrologic regime of catchment due to adoption of conservation measures are mainly reflected by the attenuation of crest segments and peak sediment flow rates of IUSGs for the successive years 1976–77 through 1983–84.

The flow features of a jet device mounted at the bottom of a rectangular channel and issuing a plane wall jet are studied both by computational and experimental means. It is found that the countercurrent jet is able to lift the water dynamically, and that the difference of water levels in the up‐ and downstream portions depends mainly on the slot Froude number and the relative height of slot. From the practical point of view, two applications may be considered: (1) The case in which a difference in water level should be created without usual hydraulic structures such as gates or overflow structures; and (2) the case in which excess energy must be dissipated. In case 2 the slot Froude number should be large, combined with a low tailwater depth. The maximum efficiency of dissipation occurs for a transition from subcritical upstream to supercritical downstream flow, beyond which a conventional secondary stilling basin may be used. These features are shown to be analogous to the classical hydraulic jump.