Sediment Transport Concepts - Sediment Transport in Irrigation Canals: A New Approach - page 127

in flumes for sediment transport experiments the boundary shear is not
constant due to the presence of bed forms. For laboratory data a sidewall
correction is applied, so that the data can be treated as that of a wide
canal and accordingly empirical relationships are developed. Different
techniques are in use for the sidewall correction (Einstein, 1942; Vanoni
and Brooks, 1957), where the main objective is to find an average bed
shear stress after making due allowance for the friction of the sidewall.
Engelund and Hansen (1967) pointed out that even after the theoretical
side wall correction, the experimental flume data should be considered
with caution, since the shear stress not only depends on the relative rough-
ness of bed and walls but also on the B / y ratio of the flume. Moreover
Brownlie (1981a) mentions that field data have slightly higher sediment
concentrations than laboratory data for similar ranges of dimensional
groups. For a theoretical analysis of the discrepancy in laboratory and
field observations, he used a typical river section and showed that the dif-
ference was due to the changing water depth along the wetted perimeter
compared to a constant depth in a laboratory flume.
An analysis of the applicability of the sediment transport predictors
in irrigation canals should consider two aspects; namely the side slope
m and the B / y ratio. The majority of the canals have a trapezoidal cross
section with side slopes ranging from 1:1 to 1:4 depending upon the soil
type and bank stability with the exception of small and lined canals that
may be rectangular. The changing water depth on the sides will influence
the overall shear stress distribution along the perimeter. This effect is more
pronounced if the B / y ratio is small. The majority of the irrigation canals
are non-wide and their B / y ratio is less than 8 (Dahmen, 1994). Hence,
the assumption of a uniform velocity and sediment transport across the
section and expressing them per unit width of canal is not correct.
For the same hydraulic and sediment characteristics different sediment
transport predictors give widely varying results. For one specific condi-
tion one predictor gives better results than the other and it is not possible
to adjust all the predictors to produce the same value for a given condition.
Hence, the purpose of the adjustment is to adapt the equation for a spe-
cific canal condition and with this fine-tuning the predictability should
be improved.
5.3.4 Effects of the canal geometry and flow characteristics
on the sediment transport
For the flow conditions and sediment characteristics prevailing in the
irrigation canals, Méndez (1998), after evaluating the available total
load predictors with field and laboratory data, concludes that the predic-
tors given by Brownlie, Ackers and White, and Engelund and Hansen are
better compared to other predictors. However, a prediction within an
error factor of less than 2 was not possible.