PREDICTABILITY OF SEDIMENT TRANSPORT USING SELECTED TRANSPORT FORMULA
Keywords:Grain roughness, Kramer’s uniformity coefficient, non uniform sediment, samaga et al flume data, sediment transport, suspended load and statistical parameters of analysis.
In alluvial rivers the sediment load is transported as bed load, suspended load and wash load. The bed material of an alluvial
channel moves as bed load (contact load or saltation load) at low value of average shear stress and they constitute the suspended load at
the higher value of shear stress. Prediction of the suspended sediment transport rate is necessary as it is affected by hydrological as w ell
as hydraulic characteristics of the alluvial stream. Methods are available for predicting the suspended load transport rate f or uniform
sediments. How ever, limited studies exist for non-uniform sediment transport. Microscopic approach of Swamee and Ojha (1991) which
considers the non-uniformity of sediment is used in the present study to compute the suspended load transport rate. Suspended load
transport formula is tested against Samaga et al flume data set (1984b) for four sediment mixtures (M-1, M-2, M-3 & M-4). The data set is
peculiar as it has varying size of sediments in mixture w ith different mean diameter of sediments and geometric deviation for each mixture.
Suspended load function is also tested using Sacramento River data measured at two USGS gauging station namely Butte city and
Colusa, California. Performance of the Swamee and Ojha’s suspended load function has been evaluated using statistical parameters such
as Root Mean Square Error (RMSE), Discrepancy Ratio (r) and Inequality Coefficient (U). Predicted results for the suspended load
transport rate are w ithin the wide error range of -9% to +4200% for all the four mixtures and hence failed to provide satisfactory results.
Predicted results for the Sacramento River data also scatters in w ide range of errors (1063%) providing better results as compared to
mixture M-1 of Samaga et al flume data set.