Some Findings on Effective Discharge Determination for Gravel Bed Rivers

The variability in the effective discharge (Qeff) calculations was assessed and attributed to the flow frequency distribution (FFD) determination, the type of sediment data used in the analysis and the methodology used in the calculation. Improvements to the existing methods for Qeff determination included the use of a Log-Pearson Type III model to represent the FFD in an analytical approach and the use of density frequency distributions to avoid dependence on the class size of histograms, and to replace the concept of frequency of discrete events by frequencies of a continuous variable. A sensitivity analysis by using the analytical approach demonstrate that the use of the suspended sediment load data to infer the bedload transport rates is inappropriate and that FFD with less skew in the data tend to have low values of effective discharge. Qeff for three gravel rivers (i.e., SF Salmon, SF Payette and Boise Rivers) were calculated. Methods used yield to similar results. Effective discharge is reached or exceeded about 6, 17 and 7 days per year for the SF Salmon, SF Payette and Boise River, respectively. The values of effective discharge and bankfull discharge are nearly identical for Boise and Payette River and the effective discharge is 1.2 times the bankfull discharge for Salmon River. This paper will emphasize understanding the basis of the formula or methodology for correct application. Numerical results are presented in comparative tables as well as conclusions.