Journal of Hydraulic Engineering

In 1961, the spillway chute of the Karnafuli Hydroelectric Project was damaged by flows ranging up to 3,480 cms (123,000 cfs), about 20% of the design discharge. While not involved with the project prior to the damage, the St. Anthony Falls Hydraulic Laboratory was authorized to conduct model studies of the spillway. It is the objective of this paper to illustrate the damage to the Karnafuli spillway, present information on the probable cause of the damage, and make brief recommendations concerning pressure fluctuations in stilling basins. A separate paper (Toso and Bowers 1988) provides information on a recent study of extreme pressures in stilling basins. The model studies indicated that fluctuating pressures in the hydraulic jump, with magnitudes up to 11 m (36 ft) of head, could have entered the 0.3‐m (12‐in.) drain lines in the chute blocks and caused uplift of the chute slab.

The methods commonly used for sampling the coarser surface layers of gravel‐bed streams are reviewed. It is found that while an areal sample is biased toward the coarser grains compared with a volumetric sample of the same material, the conversion formula suggested by a voidless cube model overcompensates for this effect. A modified version of the cube model that accounts for porosity is proposed. The modified cube model and tests conducted with areal sampling using wax indicate that area‐by‐weight analyses can be converted successfully to the equivalent bulk sieve analyses. The average depth of the wax samples ranged from $D_{65}-D_{91}\text{,}$ increasing with median grain size. The equivalence of the grid by number and bulk sieve analyses is reaffirmed.

One‐dimensional fluidization in a bounded domain is a well‐understood phenomenon. However, in unbounded domains, the formation of an interface between the fluidized and unfluidized regions is a complicated phenomenon that is poorly understood at present. This experimental study advances the fundamental understanding of two‐dimensional fluidization and provides a data base for future numerical modeling. The study was performed in a $365.7\mathrm{cm}\times 121.9\mathrm{cm}\times 30.4\mathrm{cm}(12\mathrm{ft}\times 4\mathrm{ft}\times 1\mathrm{ft})$ tank filled with sand overlying a small source pipe. The back panel contains 143 pressure taps used to determine the hydraulic head distribution. For the bed depths tested, the lines of constant hydraulic head are horizontal in the fluidized region and curve downward in the unfluidized region. The one‐dimensional theoretical superficial velocity needed to initiate fluidization provides a reasonable approximation of the two‐dimensional incipient fluidization conditions. The average sand concentration in the fluidized region decreases with increased flow rate. The total leakage occurring across the interface from the fluidized region to the unfluidized region was determined to be less than 5% of the total fluidization flow rate entering the system.

A method that allows better utilization of available data from a flood event is presented for the three‐parameter Muskingum floodrouting procedure. The procedure is based upon the three‐parameter Muskingum model, where the two conventional parameters (namely, the reach storage coefficient and the inflow/outflow weighting factor) are augmented by a third parameter to account for lateral inflow into the reach. Parameter estimation for the three‐parameter model is performed objectively by a matrix‐inversion method. The flood‐routing procedure consists of randomly splitting a group of flood events into two groups so that the three‐parameter model can be calibrated on one group and evaluated on the other, and vice versa. This procedure is improved by adjusting the previous parameter estimation scheme so that full advantage is taken of both the time interval at which the data is recorded and the lag time between the upstream and downstream hydrograph peaks. An application of the full, improved procedure is given using flood events from the Grey River on the West Coast of the South Island, New Zealand.

A computer model for oil slick transformation in rivers is developed and applied to simulate the fate of oil spills in the connecting channels of the upper Great Lakes. In this model, the transport of the surface oil slick is simulated by a Lagrangian discrete‐parcel algorithm. The model simulates either continuous or instantaneous oil spills and can be used for both open water and ice‐covered conditions. Oil slick transformation processes considered in the model include advection, mechanical spreading, horizontal turbulent diffusion, evaporation, dissolution, and shoreline deposition. The model can be used to develop cleanup measures in the case of an actual spill and to assess likely environmental impacts of possible spills.

Environmental limits on chlorine levels in power plant cooling water discharges have prompted the development of various techniques to reduce the use of chlorine without impacting plant performance. One such technique is targeted chlorination. The concept is to chlorinate a fraction of the condenser tubes at a time with a relatively high chlorine concentration that, after passage through the condenser tubes, will mix with the unchlorinated flow to meet the discharge criterion. This paper describes a mathematical model of the impingement of chlorinated jets on a condenser tubesheet to predict the distribution of chlorine concentrations in the tubes as a result of injection through fixed nozzles. This model is shown to compare well with physical model measurements. In the second half of the paper, the optimization of a fixed‐nozzle targeted chlorination system for practical applications is discussed. An example is given to illustrate the procedure.