Sediment transport in wadi systems

1
Sediment transport in wadi systems

• Part 3 - Sediment management structures and canal
  design

phil.lawrence_at_sediment.plus.com
2
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

3
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

4
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.

5
Bed load sweep at a channel bend
6
Traditional intake showing location at the
outside of a wadi bend
7
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

8
Example of a conventional sediment excluding
intake
9
Example of a spate sediment excluding intake
10
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

11
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.

12
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.

13
Physical hydraulic model
14
3 D numerical Model to predict sediment exclusion
15
Variation of sediment exclusion with sediment size
16
Comparison of predictions

17
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)

18
Basic intake
19
Example of output from simple model - impact of
sluicing discharge
20
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

21
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 .

22
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

23
Secondary sediment control

• Settling basins
• Canal sediment extractors

24
Wadi Mawr settling basins
25
Desilting a small basin
26
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.

27
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.

28
Operating problems with flushed basins
29
Sediment extractors vortex tube
30
Sediment extractors vortex tube
31
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.

32
Hybrid extractor/flushed basin for large schemes
33
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.

34
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.

35
Comparison of predictions from Chang method with
slopes measured slope of a wadi Zabid canal
36
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.

37
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.)

38
Field rise rates in spate irrigated areas
39
(No Transcript)