Sediment transport in wadi

choppedspleenMechanics

Feb 21, 2014 (3 years and 8 months ago)

59 views

Sediment transport in wadi
systems

Part 3
-

Sediment management
structures and canal design

phil.lawrence@sediment.plus.com

Summary sediment management strategy



Limit the diversion of coarser
sediments



Transport fine sediments through
canals to the fields



Make provision for the inevitable rise in
command levels


Limiting the diversion of coarser sediments



Locate intakes at outside of bends




Sediment excluding intakes


Limit diversion when wadi flows high

throttling structures or close gates


Secondary sediment control



Location of intakes at bends



The low flow channel carrying flood
recession flows forms at the outside of a
wadi bend. (Traditional intakes are placed
at the outside of wadi bends for this
reason.)



In floods bed load sweep will move the
largest sediments towards the inside of a
bend and away from an intake.

Bed load sweep at a channel bend

Traditional intake showing location at the
outside of a wadi bend


Limiting the diversion of coarser sediments



Locate intakes at outside of bends



Sediment excluding intakes



Limit diversion when wadi flows high

throttling structures or close gates


Secondary sediment control



Example of a conventional sediment
excluding intake


Example of a spate sediment excluding
intake

Features of the “spate” intake



No divide wall, flows can approach from
any direction including parallel to the weir


Intake aligned to minimise the diversion
angle


Curved channel with floor set lower than
the intake gate sill encourages coarser
sediments to move through the sluiceway


In this case a fuse plug was used

Limitations of sediment excluding intakes in
spate schemes



Spate intakes divert all the wadi flows except for
the short periods, sometimes only minutes,
during flood peaks when wadi flows exceed the
intake capacity. Sediment exclusion only
effective during these periods.


Sluice gates have to be operated in response to
rapidly varying spate flows


mechanised gates
desirable but not often affordable.

Design of sediment excluding intakes
-
use
of physical and numerical models



Physical models of practical scale
overestimate sediment excluding
performance
-

not always made clear in
reports from modelling organisations


Numerical modelling has the potential to
make quantitative predictions of sediment
exclusion without problems in representing
a wide range of grain sizes.


Physical hydraulic model



3 D numerical Model to predict sediment
exclusion


Variation of sediment exclusion with
sediment size

Comparison of predictions





Simple intake model



The 3 d models described require
considerable sediment transport and
numerical modelling expertise to set up,
calibrate, and run.


Simpler models are available that can be
used to provide an indication of the
sediment excluding performance of a
basic intake. ( for example Sharc)

Basic intake

Example of output from simple model
-

impact of sluicing discharge

Impact of Sluicing
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0
2
4
6
8
10
12
14
16
18
Eexcluder discharge m3/s
PR
Limiting the diversion of coarser sediments



Locate intakes at outside of bends



Sediment excluding intakes


Limit diversion when wadi flows high

throttling structures or close gates



Secondary sediment control



Limit diversion from flood peaks



For simple un
-
gated intakes use flow throttling
structures with a rejection spillway to limit the
flows entering a canal.



For gated intakes consider closing canal gates
during short periods of high flow. (There are
problems of responding to rapidly varying flows,
and farmers reluctance to “waste” water) Flow
throttling structures with a rejection spillway are
also used with gated intakes to ensure that
canals are not damaged by excessive flows if
the gates are left open during very large floods .


Limiting the diversion of coarser sediments



Locate intakes at outside of bends



Sediment excluding intakes


Limit diversion when wadi flows high

throttling structures or close gates


Secondary sediment control



Secondary sediment control




Settling basins



Canal sediment extractors

Wadi Mawr settling basins


Desilting a small basin


Models are used design settling
basins/gravel traps


Model predictions include:



Variation in sediment concentrations and grain sizes
passing through a basin it fills with sediment.


Estimates of the frequency of sediment sluicing or de
-
silting operations.


The time period required to flush the basin and the
volume of water needed for flushing.


The dimensions of an escape channel to convey
sediment flushed from a basin to the river or disposal
point.

Minimising trapping fine sediments



A disadvantage of settling basins in spate schemes is their high trap
efficiency for fine sediments at low flows or when basins are empty.


To minimise the trap efficiency for fine sediments:



Basins should be relatively narrow, with sediment storage
obtained by increasing the length, rather than the width or depth
of the basin.



If it is considered necessary substantial reductions in the trap
efficiency for fine sediments can be made if the tail water level in
the basin is lowered for very low basin discharges. One
possibility is to provide a notched weir at the basin exit, so that
tail water levels are substantially lowered when the basin
discharge is very low.

Operating problems with flushed basins



Water level record - Mishrafah March 1981
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
18:00
6:00
18:00
6:00
18:00
6:00
18:00
6:00
Levels (m)
13 March 1981
14 March 1981
15 March 1981
Sediment extractors



vortex tube

Sediment extractors


vortex tube


Secondary sediment control for spate
schemes



Settling basins


Mechanically excavated or flushed basins can
provide high sediment trap efficiencies with a low, or in the case of
mechanically excavated basin, zero, “wastage” of water for sediment
flushing. But sediment trap efficiency varies as a basin fills, and also
with the basin discharge which varies from zero to full supply
discharge in spate schemes.



Canal sediment extractors


Trap coarse sediment with a relatively
constant trap efficiency but require continuous flushing flows of
between 10% and 15 % of the canal discharge. Conventional
extractors not suitable for use in spate schemes.



These disadvantages are minimised in the hybrid system shown on
the next slide.


Hybrid extractor/flushed basin for large
schemes

Hybrid extractor/flushed basin for large
schemes



This system proved to
be extremely
successful in scheme
in Philippines with
massive
sedimentation
problems
-

halting,
and then reversing, a
long term decline in
the irrigation service
area, and providing
very large economic
returns
.


0
2000
4000
6000
8000
10000
1982
1984
1986
1988
1990
1992
1994
1996
Wet Season Irrigated Area at Agno
System (ha)
Spate canal design methods



no scouring


no silting” criteria



not for spate


“Regime” design methods mostly for canals carrying low
sediment loads but Simons and Albertson method
include equations for canals with sand beds and
cohesive banks, carrying “heavy” sediment loads


have
been used in spate systems


Rational methods provide the most logical method of
designing canals to achieve a specified sediment
transporting capacity. Chang, 1985 method provides
predictions of slopes and bed widths that are similar to
that observed in many spate systems.

Comparison of predictions from Chang
method with slopes measured slope of a
wadi Zabid canal


Discharge


(m
3
/s)

Sediment
concentration (sand
load, ppm)

Slope (m/km)

50 (from survey)

na

4.0
(from survey)

50

7000

3.7

37.5 (75%)

7000

3.8

25 (50%)

7000

3.9


Use canal surveys to aid design in
modernised schemes



Canal designs in modernised schemes are best
based on the slopes and cross sections of
(stable) existing canals. Design of enlarged,
extended or new canals can then be derived
using the Chang equation, with a judicious
choice of input parameters to provide a good
match with the slopes and cross sections
observed in existing canals.

Make provision for the inevitable rise in
command levels


Rise rates 5 mm to more than 50 mm year observed in spate
schemes


For existing schemes estimate historical rates of rise of fields from
coring or trial pits, and the history of upstream movement of
traditional diversion structures.


For new schemes base on command increase in near by systems



If no local information available base estimates on regional
catchment sediment yield data, the proportion of the annual
sediment load that will be diverted to a scheme, the scheme
command area, a bulk density for settled silts, and the likely
variation in sedimentation rates between upstream and downstream
fields. (In wadi Laba in Eritrea mean sedimentation rates in
upstream fields were about twice the mean rate for all fields.)


Field rise rates in spate irrigated areas


Scheme

Annual rise rate, cm/year

Upstream fields 0.8

3.2

Middle fields 0.6

1.8

Wadi Laba Eritrea

(Measured 1998/99)

Downstream fields 0.5

0.9

Wadi Laba Eritrea

(
Long term estimate)

3.0

Eastern Sudan

1
3.9

Baluchistan mountain systems

> 5.0

Wadi Zabid

Upstream fields 2

5