FLOOD03643 - Laserfiche WebLink

Home Browse Search

The values of alpha, which ranged from 1.0 to 2.0, shown in table and the correlation coefficients shown in table 3 indicate a slight tendency for alpha to decrease with discharge and depth (hydraulic radius). There is also a sl ight tendency for alpha to increase with channel roughness, slope, and particle size. Because the alpha values consist of two subsets, one including natural channels and another including channels having manmade obstructions, subsequence analyses were made on each subset. Attempts made to develop relations between alpha and other hydraulic or bed-material properties were unsuccessful because of the complexity of the changes in alpha and these variables. The mean of all the alpha values and the means of the two subsets ranged from 1.33 to 1.34. Inspection or the values of alpha in table 1 indicates they are much greater than the value of 1 assumed in equations 6, 7, and 9. the alpha of all cross sections at high flows. The majority of the However, the solution of equation 9 for a multisection reach involves an evaluation of the difference between the alpha coefficients of upstream and downstream sections. Therefore, although the value or alpha may be greater than 1.0, what is important and consequently what would affect I the accuracy of the computed n value is the relative difrerence between alpha upstream and downstream. It would be nearly impossible to measure reaches used are basically uniform throughout, although slightly con- tracting, as indicated by the basic hydraulic properties. Therefore, the higher values of alpha should not introduce much error in the com- puted n values. However, those studies that evaluate the velocity head to determine mean channel velocity at one cross section, such as techniques using superelevation around bends or dynamic uprise (velocity head buildup), could be in considerable error in assuming alpha equals 1 in a simple trapezoidal-shaped channel cross section. If'!