Docstoc

10 Fluvial sediment transport Gellis

Document Sample
10 Fluvial sediment transport Gellis Powered By Docstoc
					Sediment




  Allen Gellis, Ph.D
  Research Geomorphologist
  U.S. Geological Survey
  5522 Research Park Drive
  Baltimore, MD 21228
  443 498-5581
  agellis@usgs.gov
BLOOD IS THE RIVER OF LIFE….
“SEDIMENT IS THE WHITE BLOOD CELLS OF THE RIVER”
                                            Gellis
       Sediment Transport

-   Definitions –
-   Sizes and sieving
-   Intro to entrainment/incipient motion
-   Bedforms
-   Suspended/bedload/washload
-   Measurements
-   Hysteresis
-   Watershed perspectives
       WHAT IS SEDIMENT?
• Sediment consists of particles derived
  from rocks or biological materials.
• When transported by, suspended in, or
  deposited by flowing water, referred to
  as fluvial sediment.
           The Sediment Cycle

– Weathering
   • Mechanical (freeze/thaw, abrasion, other)
   • Chemical (carbonic acid, biotic interactions, other)

– Erosion
   • raindrop impact
   • sheet, gully, bank/channel erosion
   • mass wasting, glaciation, volcanic eruption, eolian

– Transport (a number of mechanisms)
– Deposition
   • streambeds, floodplains

– Digenesis
   • compaction, cementation, mineral replacement
• Texture refers to properties of a sediment
  such as particle size, shape,
• roundness, and sorting.
                                                    Particle shape
                                                     is a measure
                                             of the sphericity of a grain.


                                            Sphericity = 3(bc/a2)
                                            Roundness = 2r/a
                     c          a           where r =radius of sharpest corner
                            b




http://duke.usask.ca/~reeves/prog/geoe118/geoe118.017.html
The three mutually perpendicular axes of a non-spherical grain.
Grain size, by convention, is the length of the
intermediate axis (b axis)
                         Density
The bulk density or dry density of a sediment is the dried
mass per unit volume of a sediment bed.
Bulk density of a sandy bed might be about 1.7 g/cm3 =
1700 kg/m3, and of a high-water content mud only
0.5 g/cm3.

Sometimes a non-dimensionalized density is used,
defined as the specific gravity or relative sediment density,
where sediment density is normalized by water density (s
= ps/p).




 Density of the sediment grains.
 Quartz density is s = 2:65 g/cm3.
       Particle Size Classification
                 Systems
• USDA System – Soil description for agricultural, land-
  based wastewater disposal, and most environmental
  applications.
  (i.e., loam)

• AASHTO: American Association of State Highways and
  Transportation Projects – potential use as road base
  (i.e., A-1)

• Unified Soil Classification System (ASTM D2487-92)
  Engineering Applications (i.e., SM)

• Wentworth (phi #)- Geological and Geotechnical
  Studies Using screen or sieve size using the phi number.
  (phi #, sieve no., or mm)
    Each system has unique (“jargon” or terminology)
            The Systems Do Not Group the
                       Same
CLAY/SILT    SILT/SAND   SAND/GRAVEL
SIZE
Wentworth –
preferred by 9 out of
10 geomorphologists
             Sieve Analysis
        Particle-Size Distribution
• Particle size distribution describes the
  abundance (by weight) of the various size
  particles that constitute the mineral portion of
  soil materials.

• The distribution of the size based on size (mm,
  micron, or sieve size)
• Sediment is either dry sieved or wet sieved
  (more accurate)
                  SIEVE ANALYSIS - DRY SIEVED


SITE NAME          Mesa trap1 (lower mesa)-5 surface-10

Collection Date     6/10/2005


Lab Date             7/5/2005

 Initial mass
    sieved            217.35      (g)


           Sieve                   Sieve      Mass of soil
     No.              size        Tare (g) with pan w/o pan % retained Cumulative    % passing
   (US std)          (mm)                     (g)      (g)             % retained

               4           4.75         662.36     674.86     12.5    5.7      5.7          94.3
              10              2         573.58     579.36     5.78    2.7      8.4          91.6
              20           0.85         532.25     553.01    20.76    9.5     17.9          82.1
              40         0.425          448.06     485.38    37.32   17.1     35.1          64.9
             230       0.0625           340.84     465.46   124.62   57.2     92.3           7.7
PAN                    ---              486.21     503.09    16.88    7.7    100.0           0.0



                                                 Totals     217.86   100
             Sieve Analysis
        Particle-Size Distribution
• Particle size distribution describes the
  abundance (by weight) of the various size
      D85
  particles that constitute the mineral portion of
                        D50
  soil materials.

• The distribution of the size based on size (mm,
  micron, or sieve size)             D10


   PARKER GRAIN SIZE CALC.XLS
 Silt and Clay– Pipette Method
•The pipette method utilizes Stoke's Law by the extraction of subsamples of the
soil suspension at a given depth after a predetermined settling time for each size
fraction of interest.

•As time passes, larger particles pass by the sampling depth, and smaller and
smaller size fractions can be sampled.

•After extracting the sample, it is dried weighed, and a calculation can be done to
determine the percentage of the total soil in suspension present in each sample.

• The pipet method is very accurate, but also time consuming. Pretreatment of
the sample may include the use of dispersing chemicals or oxidizing agents.
 V = (2gr²)(d1-d2)/9µ (Stokes Law)
 Where,
 V = velocity of fall (cm sec-¹), g = acceleration of gravity (cm sec-²),
 r = "equivalent" radius of particle (cm), dl = density of particle (g cm -³),
 d2 = density of medium (g cm-³), and µ = viscosity of medium (dyne sec cm-²).
SILT AND CLAY USING Laser Diffraction Work

In laser diffraction particle size analysis, a representative cloud or
„ensemble‟ of particles passes through a broadened beam of laser light
which scatters the incident light onto a lens. This lens focuses the scattered
light onto a detector array and, using an inversion algorithm, a particle size
distribution is inferred from the collected diffracted light data.

Sizing particles using this technique depends upon accurate, reproducible,
high resolution light scatter measurements to ensure full characterisation of
the sample.
 SEDIMENT
TRANSPORT
                Sediment transport
Sediment transport is the entrainment and
movement of soil and rock particles through
the processes:


  Transport modes
  •   Traction (rolling over the bed surface)
  •   Saltation (jumping over the bed surface)
  •   Suspension (permanent transport within the fluid)
  •   Solution (chemical transport)
                       Entrainment
                                     • Three main forces are
FL, lift
                                       involve in how
                           FD, drag
component
                           component
                                       particles get picked
                                       up:
                                     – Fluid drag force
                                     – Fluid lift force
            FG, gravitational        – Gravitation force
            force
           Drag Force

• The drag force due to water acting on an
  object can be found by:
           FD = ½ ρ CDV2A

  where:   FD = drag force (N)
           CD = drag coefficient (no units)
           V = velocity of object (m/s)
           A = projected area (m2)
           ρ = density of fluid (kg/m3)
              Drag Coefficient: CD
                 FD = ½ ρ CDV2A


• The drag coefficient is a function of the shape of
  the object and the Reynolds Number (Re)
   – Re = (velocity * length-scale) / (kinematic viscosity)
  {The Reynolds number represents the ratio of the importance of
  inertial effects in the flow to viscous effects in the flow}
• For a spherical shape the drag coefficient ranges
  from 0.1 to 300, as shown on the next slide.
Drag Coefficient for Spheres
Lift force is an example of Bernoulli‟s principle, which states
that the sum of the velocity and pressure on an object in a flow
must be constant. Whenever a flow speeds up, it exerts less
pressure than a slower moving part of the flow.
        Entrainment
                        FF, resultant
                        force
                                      • Three main forces are
FL, lift
                                        involve in how
                            FD, drag
component
                            component
                                        particles get picked
                                        up:
                                             –   Fluid drag force
                                             –   Fluid lift force
             FG, gravitational               –   Gravitation force
             force
                                             –   Resultant force
       The different types of fluid forces (FL and FD) combine to produce a
       resultant force which acts downstream (FF). This resultant force can
       move the grain by lifting it or rotating the grain about some pivot.
• Other factors that are important
  include
• Packing of sediment – are particles
  protected or stick out in flow
• Well sorted versus poorly sorted
• Amount of clays (cohesion)
• A given particle will move only when the
  shear stress acting on it is greater than the
  resistance of the particle to movement. The
  magnitude of shear stress required to move
  a given particle is known as the critical
  shear stress (cr). Sometimes a critical
  velocity is also used, as well as a critical
  stream power.
The approach to understand sediment transport and
initiation of motion have used Dimensionless Analysis –
 The properties of most substances can be expressed in
terms of three primary dimensions (1) Mass, M, (2)
Length, L, and (3) Time, T.
For sediment this would include diameter (L), density (),
viscosity (), velocity (U ), flow depth (L)
Shield’s experiments involved determining the critical boundary shear
shield‟s diagram –                       sediment size and density
stress required to move spherical particles of various motion is a
    u*D 
         
      
   function of the the same properties.
over a bed of grains withBoundary Reynolds number
He produced a diagram that allows the determination of the critical shear
stress required for the initiation of motion.
   f   gD( s   )  
                        
                        
          s    gd 
                o0
                




                                   U *d
                            R* 
                                    
       
   




                                                      u*D 
                                                 R*      
                                                        
  Shields using dimensionless analysis and experimental results. Two
  dimensionless groups were used:


                    c            u*D 
   Shields   
                gD( s   )   f   
                              
Shields
                                      
                            
criterion also                                     Boundary Reynolds
called Shields
                   Dimensionless bed
                                             R*    number
Parameter
                   Shear stress

         c                                          
                                           u* = shear velocity; D= grain
                                           diameter;     = kinematic
                 = critical shear stress   viscosity


         g = gravity; ps = sediment
         density; p = density of water;
         D= grain diameter


β* = Flow force/                           R* = Inertial force/
     grain weight                          Viscous
Velocity is an important factor effecting erosion


1. Shape of the velocity profile-
      a) Influenced by the ratio of flow depth (d) to the
         bed material size (D): d/D
      - The larger the roughness element, the steeper the
        velocity gradient toward the bed.
      b) Influenced by channel shape
      - wider channels, velocity gradient is steepest
        towards the bed
      - Narrower channels, velocity gradient is steepest
        towards the banks
Velocity
profiles
Shear Velocity (U*) = measure of shear stress and
and velocity gradient close to bed

                    U* = √τo/ρ
U* = Shear Velocity
τo = Boundary Shear Stress
ρ = Fluid Density


  – U* = √gdS                 d= depth        S= slope
  – Assumes steady, uniform flow
  – Average shear velocity of section of channel
        Subdivisions of turbulent flows
     Turbulent flows can be divided into three layers:


Viscous Sublayer: the region near the boundary that is dominated by
viscous shear and quasilaminar flow (also referred to, inaccurately, as
the “laminar layer”).


Transition Layer: intermediate between
quasilaminar and fully turbulent flow.


Outer Layer: fully turbulent and
momentum transfer is dominated by
turbulent shear.
     GRAIN ON THE BED
Reynolds Number

Re = Inertial forces/viscous forces = VR/μ

where ρ V = velocity, R= hydraulic radios, and μ =
fluid viscosity.

Laminar <500
Transitional 500-2500
Turbulent >2500
i) Viscous Sublayer (VSL)


The thickness of the VSL (d the lower case Greek letter delta) is known
from experiments to be related to the kinematic viscosity and the shear
velocity of the flow by:

              12        Where U* is the shear velocity of the        o
           d            flow:                                   U* 
              U*                                                      
i) Viscous Sublayer (VSL)



So the Boundary Reynolds number                         u * D  11 .6 D
combines the effect of near-bed velocity      Re *  f        
and grain size. Compares grain size with                       d sub
thickness of laminar sublayer


                           u* = shear velocity; D= grain diameter;         = kinematic
                           viscosity; d   = thickness of laminar sublayer
     Critical Dimensionless Shear Stress


                                     cr
                              c (   )d
                                *
d= mean depth                      s
S= surface water slope (m/m)
D = particle diameter (m)
s = specific weight of sediment (1000 kg/m3)
 = specific weight of water (2650 kg/m3)

 = shear stress (PA) PA= N/m2; N=kg.m/(s2)
                                             = specific weight of water
      Critical Dimensionless Shear Stress



                c  (   )d
                   *
        cr                            s


                  Shields parameter

THIS IS WHAT
WE WANT TO
SOLVE FOR
Shield criteria
Typical range
0.01 to 0.07


0.01 loosely
packed


.07 tightly
packed
(armored)
If  for flow of interest is > cr
      We have motion
   NOTE ON SHIELDS PARAMETER

For channel beds composed of particles coarser than 8 mm, depending on
whether the channel bed is loosely consolidated or tightly packed, Church
(1978) demonstrated that the Shields parameter can vary by a factor of two.

Recently deposited sediments can be poorly packed, making it easier for those
particles to be entrained

Reid et al. (1985) demonstrated that the shear stress needed to entrain
particles could be up to three times higher than the average when the flood
occurred after an extended period of no bed disturbance.
Mixed-size sediments can exhibit differential mobility and
selective transport that, in turn, influence identification of
incipient motion thresholds (Shields 1936c; Neill and Yalin
1969; Wilcock 1988). The most easily mobilized particles in
a mixed-size bed are those with high protrusions (Kirchner
et al. 1990) and low intergranular friction angles (Miller and
Byrne 1966) ---distributions of which are functions of
sediment size, shape, rounding, sorting, and packing (Miller
and Byrne 1966; Li and Komar 1986; Kirchner et al. 1990;
Buffington et al. 1992; Carling et al. 1992).



                                       From Buffington, 1999
                                 Determine ratio di / d50
 CALCULATION OF                 Where: di = bed material D50 of
 SHIELDS                        riffle;
 PARAMETER                      d50 = subpavement Dˆ50 or bar
 (Andrews, 1983)

                                    Equation 1 ARMORED

If ratio ratio Di/Dˆ50
                                                     d             0.872


                            *  0.0834  
                                                           i
= 3.0 – 7.0 then use Eq1

                                        d 
                            c


                                                           50




                                                       d
                                    Equation 2
If ratio Dˆ50/Di                                                        0.887


                            *  0.0384  
                                                                i
= 1.3 – 3.0 then use Eq2


                                        d 
                                c


                                                                50
Condition of streambed                          Shields Parameter
Loosely packed, ‘quicksands’ and gravels with   0.01-0.035
large voids
Normal, uniform grain sizes                     0.035-0.065
Closely packed with smaller particles filling   0.065-0.10
voids
Highly imbricated                               >0.10



                                                        Gordon et al. (1992)
Source: Leopold et al (1992) – Fluvial Processes in Geomorphology
   c    0.73 D   Carlson and
                   Griffiths,
                   1987
Modified critical shear stress
The modified critical shear stress equation is based on the relationship
between the particle size of interest (Di) and D50, (Andrews 1983).
For many stream simulations, the particle size of interest, Di, is usually
D84 and/or D95, because if these particles begin to move, much
of the streambed is in motion and the structure of the channel bed will
change (Komar 1987; Komar and Carling ,1991)




                                           0.4         0.6
              0.045

Where   ci = critical shear stress for sediment
           D50 =
movement, 0.045 = shields parameter for D50; Di =
sediment percentile of interest;
The modified critical shear stress equation is
appropriate for assessing particle stability in
riffles and plane-bed channels (i.e., where flow is
relatively uniform or gradually varied between
cross sections) with channel-bed gradients less
than 0.05 (5 percent) and D84 particles ranging
between 10 and 250 mm (2.5 to 10 inches).
Critical unit discharge
(Bathurst, 1987)


• For channels with gradients greater than 1
  percent, and where the flow depth is
  shallow with respect to the channel-bed
  particle size (relative submergence, R/D50
  <10), Bathurst (1987) suggested using
  discharge-per-unit width instead of
  average boundary shear stress for
  determining particle mobility.
    Critical unit discharge
    (Bathurst, 1987)
                                      q is the unit discharge (m2/s)
                                      Q is discharge (m3/s)
                                      w is the active channel width for bedload
                                      transport (fm) at a given cross section.


bed material is uniform well sorted




                       where:
                       qc-D50 is the critical unit discharge to entrain the D50
                       particle size (m2/s)
                       D50 is the median or 50th percentile particle size (m)
                       g is gravitational acceleration (9.81 m/s2)
                       S is bed slope (m/m)
    Critical unit discharge
    (Bathurst, 1987)

bed material is non uniform (poorly sorted)




                   where: qci is the critical unit discharge to entrain the
                   particle size of interest (m2/s)

      Exponent b is a measure of the range of particle sizes that make up
      the channel bed.




        If q >qci = motion
Thompson and Croke, 2008
                         Incipient motion

     •      Other approaches for describing hydraulic
            conditions at incipient motion:
               1. Critical bed velocity
               2. Stream power




Garde and Raju (2000) – Mechanics of Sediment Transport and Alluvial Stream Problems.
               1. Critical bed velocity – Hjulstrom Curve




• Hjulstrom curve (relates flow speeds to
  „threshold‟ velocity for a certain size of
  particle… shows speeds that correspond with
  erosion (entrainment), transport & deposition
The Hjulstrom curve has been attacked for using mean
velocity instead of velocity near the bed.



Stability of the bed, in terms of velocity, has been defined
by Gordon et al. (1992)
                                                  (Vc = critical velocity)
                                                   (Vb = velocity at bed)

                                                      d = grain size

                                                    (V = velocity)



      If Vb > Vc, the bed will be mobile
KISSIMMEE RIVER RESTORATION
18     17.5 MG/L   17.5 MG/L   16     17.5 MG/L
MG/L                           MG/L
   Measurements D50 = 0.18 to 0.33 mm




The discharge measurement
made on March 20, 2008 had
velocities near the bed of 33.1         D50
m3/s which were just above this
threshold to initiate bedload
transport.

Bedload measured on March 20,
2008 was 0.11 Mg.

A discharge slightly below 33.1
m3/s, of 28.3 m3/s, was estimated
to be the discharge at which
bedload transport begins.
Transport of sediment can also be considered in terms
 of stream power, a measure of ability of a stream to
expend energy (i.e., do work) per unit area of channel


            QS   v
    = stream power
    Q = discharge
    S= slope
    = unit weight of water

     = shear stress, v= velocity

  Concept of available stream power; rate of potential
  energy expenditure per length of channel. Thus,
  when excess power exists, it will be available to move
  sediment
Cohesiveness

The cohesion of a sediment
particle is associated with soil
type and particle size. The three
most common minerals
which have electrochemical
forces causing individual
particles to stick together are
illite, kaolinite, and
montmorillonite.
                            effects of cohesion




ASCE (1975) Sedimentation Engineering
  Deposition: What forces control
     the settling of particles?
• As soon as a particle is lifted above the
  surface of a bed, it begins to sink back
  again.
• The distance that it travels depend on the
  drag force of the current, and the settling
  velocity of the Particle.
• The velocity at which a clast settles
  through a fluid is calculated using
  STOKES‟ LAW of settling
                    Sediment transport

Stokes’ Law (settling velocity in a static fluid)

                                gD2 (ρg  ρ f )
                         vg 
                                      18 μ

         vg=settling velocity; D=grain diameter; g=grain density;
                   f=fluid density; =dynamic viscosity


• Stokes’ Law only applies to fine (<100 m), quartz-density grains in
  water
• EXERCISE
LINGANORE CREEK, MD
           4.5
                     Linganore Creek, MD
           4.0                                             XS A-A'
                                                           XS B-B'
           3.5                                             XS C-C'

           3.0
Stage, m




           2.5

           2.0                       20.6 cms

           1.5                     13.2 cms

           1.0

           0.5
                 0    5       10        15       20   25        30
                             Station (LB to RB), m
• How well does the modified critical shear
  stress equation apply here?
• l D84/D50 = 5.2, which is much less than
  30.
• l Slope < 5 percent. YES
• l Channel unit is a riffle. YES
• l D84 particle size of 102 millimeter is
  between the range of 10 and 250
  millimeters.
Movin‟ right „long
Once transport is initiated the modes of
transport are suspended, bedload, and
saltation
OVERLAP
Shear stresses above the critical threshold for transport
mould cohesionless sediment into bedforms whose
geometry depends on flow dynamics and grain size. In
turn these bedforms can play an important role in
channel adjustment.
                          Froude = 1


                                 Relationship
                                 between
                                 Flow
                                 Regime and
                                 Bedform
                                 Type Based
                                 on Froude
                                 Number
                                 (Simons and
Froude < 1
             Froude > 1          Richardson,
                                 1966)
             Froude Number
                            U
                       
                    Fr = Froude Number
      U                     gL
 Fr    Frr 
        F 
                    U = mean flow velocity
                    U
      gL      gL = velocity of shallow water wave
                    g = gravitational acceleration
                    L = water depth

• Ratio between inertial and gravity forces
• Dimensionless value (like Re)
Flow Velocity vs Bedform Type
oV = qs
Bedforms impart drag on the flow, called form drag, and the part
caused by the grains themselves is often called skin friction or
grain roughness drag. Form drag often exceeds skin friction in
magnitude.

Bedform adjustment represents a mean of regulating, through its
effect on resistance its bed configuration.

Depending on their amplitude and wavelength, bed forms
in sand-bed and gravel-bed rivers (i.e., ripples, dunes, and
bars) can dissipate 10–75% of the total channel shear
stress (Parker and Peterson,1980; Prestegaard 1983;
Dietrich et al. 1984; Hey 1988), causing potentially
significant overestimation of the actual bed shear if
neglected from stress calculations.
                     Bed Load
• The bed load generally constitutes between 5 and
  20 percent of the total load of a stream.
• Particles move discontinuously by rolling or sliding at
  a slower velocity than the stream water.
• The bed load may move short distances by saltation
  (series of short intermittent jumps).
Bedload samplers can be
grouped into four
categories:
    (1)Box or basket
    (2)Tray or pan
    (3)Pressure
       difference
    (4)Slot or pit
During the late 1960‟s, the
U.S. Geological Survey‟s Water
Resources Division Developed the
Helley-Smith Bedload Sampler.


  A pressure- differential type samplers; that is,
  they have nozzles that have larger cross-
  section openings at the rear of the nozzle than
  they have at the entrance.
  Helley-Smith samplers have a flare (ratio of exit
  area to entrance area) of 3.22.
On May 15, 1985, The Technical
Committee accepted a modified nozzle
(area ratio of 1.40) as the “tentative”
standard for use by all Federal agencies.
The tentative acceptance was reaffirmed
on April 1988. (BL-84 & BLH-84)
              Some Pressure-Difference Type
                   Bedload Samplers
Nozzle Size     Area        Type        Nozzle    Suspension           Weight
                Ratio                  Thicknes
                                           s
3” by 3”      3.22      Helley-Smith   ¼-inch     Cable              50-200 lb.
3” by 3”      3.22      Helley-Smith   ¼-inch     Wading Rod
3” by 3”      3.22      Helley-Smith   16-gage    Wading Rod
3” by 3”      1.40      FISP BL-84     ¼-inch     Cable              35-50 lb.
3” by 3”      1.40      FISP BLH-84    ¼-inch     Wading Rod
4” by 8”      1.40      Elwha          ¼-inch     Wading Rod/Cable   10-150 lb.
6” by 6”      3.22      Helley-Smith   ¼-inch     Cable              150-200 lb.
6” by 12”     1.40      Hubbell #5     ¼-inch     Cable              150-200 lb.
6” by 12”     1.40      Toutle River   ¼-inch     Cable              100-200 lb.
                        Type 2
     Acceptable Conditions:
• The following physical conditions exist:
  – The bed material is firm enough physically
    to support the sampler without it sinking
    into the streambed;
  – The streambed is smooth enough for the
    nozzle to lay flat on the bottom;
  – The stream velocity is low enough to allow
    the sampler to sit properly on the
    streambed; and
  – Neither organic nor mineral deposits clog
    the bag to the extent that flow through the
    sampler is restricted.
One of the fundamental
features of bedload movement
is the extreme variation of the
transport rate even when the
streamflow is constant.
Examples of possible distribution of mean bedload transport rates
in a cross section. A, Discharge varies uniformly. B, Discharge is
uniformly consistent. C, Discharge is erratic with varying
tendencies. D, Discharge is an unpredictable combination of
varying tendencies.
Courtesy W.W. Emmett
Temporal variation of bedload transport rates for 120
consecutive bedload samples from a stream with
constant water discharge (Carey, 1985).
Although it is important,
selection of the nozzle type,
area ratio, or sampler type is
probably less critical than the
selection of the proper
sampling method for obtaining
an accurate bedload-
discharge measurement.
Depending on the relative
magnitude of the temporal and
spatial variability of the
bedload transport, different
sampling methods are optimal
when making a bedload-
discharge measurement.
Sampling Procedure:
• Cross-sectional procedure:
  – Single Equal Width Increment (SEWI)
    method
  – Multiple Equal Width Increment (MEWI)
    method
  – Unequal Width Increment (UWI) method
As long as you get 40 samples, you are probably in pretty good shape –
B. Emmett re: J. Gray -11/7/2007
„After 35-40 samples the variance was muted by a sufficient number of
samples‟




                                              Courtesy W.W. Emmett
Analysis:
• Computation of bedload discharge:
  Depends on sampling procedure.

• Basically = constant x width x
  mass / time = tons/day
                  1000
                                                      8/23/08
                         KISSIMMEE        8/28/08
                  100
                                                    8/28/08
Bedload, Mg/day




                   10




                    1        7/19/08


                   0.1   4/20/08             Qb = 1.89E-9 * Q5.196


                  0.01
               10
        Holnbeck,                            100                     1000
        2005                                          3
                                       Discharge, m /s
                THE THRESHOLD OF SIGNIFICANT SUSPENSION
Flow in rivers is invariably turbulent. The eddies of turbulence have the capacity to
waft bed sediment high up into the water column if a) the turbulence is strong
enough and b) the sediment grains are not too heavy. Although the onset of
suspension is not a sharp phenomenon, a standard rule of thumb (Bagnold, 1966)
is
                                     u
                                        1
                                     vs
where vs denotes grain fall velocity and u* denotes the shear velocity, defined as
                                 u  b / 
                         types of suspended load
   • suspended load –
      – wash load – “sediment load of a stream which is
        composed of particles sizes smaller than those found
        in appreciable quantities in the shifting portions of the
        stream bed” – TOO SMALL TO DEPOSIT
      – suspended load – “particles which are moved by and
        suspended in the water column, but can settle in
        locations where the travel velocity is low or settling
        depth is small.” CAN DEPOSIT UNDER SOME
        CONDITIONS.



Garde and Raju (2000) – Mechanics of Sediment Transport and Alluvial Stream Problems.
• MEASUREMENT OF SUSPENDED
  SEDIMENT
Sampled & Unsampled Zones
 with an Isokinetic Sampler
2 mm   Sands   0.062 mm   Silts   0.002 mm



                                    Clays
             Box Coefficient (BC) = Cmean/Cpoint

                                                     Cmean = ~930 mg/l


   BC=~1.1
                                                   BC=~1.1




                                                              BC=1.03




BC=~1
          Box Coefficient (BC) = Cmean/Cpoint

                                                Cmean = ~1,360 mg/l




  BC=~4                                         BC=~5




                      Mean Values


                                                           BC=~1.7




BC=~1.5
Why is Isokinetic Sampling Important?




                           FISP  TM
    SAMPLING TECHNIQUES
• Sample collection methods
  – Depth-integrated sampling
  – Point-integrated sampling
  – Point sampling
  – Grab or dip sampling
  – Pumped samples *
  – Single-stage samples *
  *will be covered in separate lecture
US DH-59




           FISP
              TM
                                                                                     Unsample
  Sampler        Nozzle        Container      Max.         Min. Vel.     Max. Vel.               Weight
                                                                                        d
 Designation     ID (in)         Size       Depth (ft)     (ft/sec)       (ft/sec)                (lbs)
                                                                                     Zone (in)

US DH-48           1/4            pint          9             1.5           8.9         3.5        4

US DH-59         ¼, 3/16          pint         9, 15          1.5           5.0         4.5       22
What sampler
US DH-76         3/16, 1/4       quart          15            1.5           6.6         3.2       25

to use?
US DH-81       5/16, ¼, 3/16      liter         9         1.5 2.0 ,2.0
                                                                         7.0, 7.6,
                                                                           6.2
                                                                                        4.0        1

                                                                         7.4, 7.0,
US DH-95       5/16, ¼, 3/16      liter         15        2.1,1.7, 2.1                  4.8       29
OFR 2005-
US DH-2        5/16, ¼, 3/16      liter     13, 20, 35        2.0
                                                                           6.2

                                                                            6.0         3.5       30

1087
US D-74          ¼, 3/16       pint/quart      9, 15          1.5           6.6         4.1       62

US D-74AL        ¼, 3/16       pint/quart      9, 15          1.5           5.9         4.1       42

                                                                         6.2, 6.7,
US D-95        5/16, ¼, 3/16      liter         15        1.7,1.7, 2.0                  4.8       64
                                                                           6.7

US D-96        5/16, ¼, 3/16    3 liters    39, 60, 110       2.0          12.5         4.0       132

US D-96A1      5/16, ¼, 3/16    3 liters    39, 60, 110       2.0           6.0         4.0       80

                                                           3.0, 3.0,
US D-99        5/16, ¼, 3/16    6 liters    78,120, 220                    15.0         9.5       275
                                                             3.5

US P-61A1          3/16        pint/quart    180, 120         1.5          10.0         4.3       105

US P-63            3/16        pint/quart    180, 120         1.5          15.0         5.9       200

US P-72            3/16        pint/quart     72, 51          1.5           5.3         4.3       41
        AUTOMATIC SAMPLER
• An automatic sampler is a device that, on it
  own, collects a volume of water/sediment
  mixture from a stream, lake, reservoir, well,
  or storm drain and places it in a container
  for further physical, chemical, or biological
  analyses.
             ADVANTAGES
• Can be pre-set to collect samples on a time,
  stage, discharge, volume, rainfall, and (or) in-
  stream surrogate basis.
• Can collect samples over the runoff event.
• Safer to collect samples during high flow events
• May have low labor costs
• Has the capability to be reprogrammed remotely
• Samplers can be packaged in portable units that
  can be easily moved from site to site
           DISADVANTAGES

• Collects a point sample, thus requiring the
  samples be correlated with the “true” cross
  sectional mean
• Has a limited number of samples that can be
  collected before it must be serviced
• Requires power to operate
• Most pumps have a limit of about 28 feet of lift
• Sometimes has high installation costs, especially
  if multiple sites are required
       DISADVANTAGES (cont.)
• Cannot collect samples for analysis of certain
  parameters
• Requires maintenance of site and equipment
• Possible cross contamination if not set up
  properly
• Generally does not collect a sample
  isokinetically
• Has limits on the concentration that can be
  accurately sampled, especially when sand
  concentrations are high
                 Distribution of sediment

Pumping intake
930 mg/L




                          Mean = 929 mg/L
                 Distribution of sediment


Pumping intake
800 mg/L




                           Cross section mean = 1360 mg/L
            What is a “Box
            Coefficient?”
• A mean flow-weighted x-sectional
  constituent concentration divided by
  a mean concentration at a point or
  vertical.

   Mean Concentration x-section mg/L
   -----------------------------------------
   Mean Concentration pumped mg/L
                                  SSC vs TSS
                                 14,466 paired samples

            150000




            100000
SSC, mg/L




            50000




                 0
                     0   40000                    80000   120000
                                      TSS, mg/L
                               SSC vs TSS
                              Samples < 5000 mg/L




            4000




            3000
SSC, mg/L




            2000




            1000




               0
                   0   1000    2000         3000    4000
                                   TSS, mg/L
            SUMMARY

• TSS Values Tended to be Smaller than
  SSC
• Degree of Agreement Between % Fines
  and TSS Varied Between Stations
• SSC VS TSS Relation is Better at a Site
  than for the Total Data Set
COMPUTATION OF SUSPENDED-SEDIMENT LOADS
Computed Sediment
   Discharge
   X tons/day
Computed Sediment
    Discharge
  ~103X tons/day
                              SEDIMENT SUBDIVISION

                         13    14   15   16   17   18         19      20      21       22
                   250

                                                        Sampled suspended-sediment
                                                        concentration
                   200                                  Estimated suspended-sediment
                                                        concentration
Discharge, ft3/s




                   150                                  Hydrograph



                   100



                   50



                    0




                         13    14   15   16   17   18        19      20      21        22

                                              HOUR
The computation of sediment discharge is

Qs = Qw x C * 0.0027 (ENGLISH)
(Qs = sediment discharge tons/day (Qw – water discharge; C = sediment
concentration‟ 0.0027 conversion, assuming sediment water mixture = 62.4
Ib/ft3)

Qs = Qw x C * 0.0864 (Metric; Mg/day)
 REGRESSION METHOD


1. Generate a Q vs
    tons/day plot
2. Convert to log-log
3. Generate Regression line
4. Convert Mean daily Q to
Mean Daily Load                                                                 LINGANORE CREEK
use Bias correction                                                10000




                               Suspended-sediment load, tons/day
                                                                   1000



                                                                     100



                                                                      10                                  SL =0.00014Q ^2.45

MAY HAVE ERRORS OVER
100%; most applicable to                                              1
                                                                                                               All data R2=0.63
large rivers, should not use
                                                                     0.1
in small streams
                                                                    0.01
                                                                           10               100                                   1000

                                                                                 Mean daily discharge, ft3/s
FLOW DURATION METHOD

1. Determine % of time for each Q class
2. Use Q determine SL from transport
   curve
3. Weight for % of time

(Piest, 1962; Walling, 1977
    https://www.wou.edu/las/physci/taylor/
    andrews_forest/refs/walling_1977.pdf
                  • Comparison of
                    methods for
                    annual loads of
                    Sediment
                    – A) f(Turb)
                    – B) f(Q)
                    – C) Discrete
                      samples
                  • Error analysis
                    shows that
                    turbidity in
                    realtime is a
                    more accurate
                    predictor than
Kansas District     Q
Chesapeake Bay
                           Rio Grande de Loiza at Caguas
                           1989-1995
                 300
Number of Days



                                                on average 75% of
                                                annual load transported in eight days
                 200

                           on average 31% of
                 100       annual load transported in one day



                  0
                       0         20        40       60       80       100
                  Percent of Suspended Sediment Load Transported
          Transport curves
• A) Instantaneous discharge vs suspended-
  sediment concentration
• B) Mean daily discharge vs daily
  suspended-sediment load
SEDIMENT
VARIABILITY                                               Rio Piedras at El Senorial
                                           105
  Suspended-sediment concentration, mg/L



                                           104


                                           103


                                           102


                                           101


                                           100
                                             0.01   0.1      1        10     100   1000


                                                          Discharge (m3/s)
Walling, 1974
1. Flow
2. Time relation of sample to
Hydrograph peak
3. Baseflow at start
4. Index of flood intensity


Guy, 1964
1. Flow
2. Baseflow at start
3. Peakedness Index
Hysteresis
Figure 1. Hysteresis types (after Williams, 1989; Kurashige,
1994). Arrows indicate direction through time. Type 1 -
clockwise, Type - 2 clockwise early in storm reversing to
counterclockwise, Type 3 - counterclockwise, Type 4 -
counterclockwise early in storm reversing to clockwise, Type 5
- no exhaustion
Hysteresis type             Cause of hysteresis shape                                Reference
Type 1 (clockwise)          Depletion of sediment source                             Walling, 1974; Wood, 1977; Costa, 1977
                            Formation of armored layer                               Williams, 1989
                            before peak discharge

                            River bed                                                Kurashige, 1994


                            Water depth and water slope increases before peak        Kurashige, 1994
                            discharge **

                            Increased baseflow after peak discharge                  Costa, 1977


                            Seasonality                                              Sidle and Campbell, 1985


Type 2 (clockwise then      Ice breakup
counterclockwise)

Type 3 (counterclockwise)   Floodwave traveling faster than mean flow velocity *     Marcus, 1989;Williams, 1989

                            High soil erodibility                                    Williams, 1989


                            Distant sediment source                                  Williams, 1989


                            upstream tributaries                                     Asselman, 1999


                            Thin, exposed soil surfaces                              Kurashige, 1994


Type 4 (counterclockwise
then clockwise)

Type 5 (stationary)         Uninterrupted supply of sediment ranging from ample to   Wood, 1977; Williams, 1989
Hysteresis - types Puerto Rico


                                   FOREST

                                   PASTURE

                                   CROPLAND

                                 URBAN
Conclusion (hypothesis)




                                                             sediment
                                                             flushing
Hysteresis type

                  Counter




                                                                         Previous Event
                  clockwise



                  Clockwise




                                                              Sediment
                                                              increase
                              Undisturbed   Disturbed
                              (forest)      (construction)
        SURROGATE
     TECHNOLOGIES FOR
    SUSPENDED SEDIMENT
Those based on:
 Bulk Optics (Attenuation and
  Scatterance)
 Multi-Frequency Optics
 Laser Diffraction
 Digital Photo-Optic Imaging
 Pressure Difference
 Gamma Attenuation
 Acoustic Backscatter
Optical Backscatterance, ~
Turbidity




  Paul Buchanan (USGS), San Francisco/Delta Bay, April 1999
   BULK OPTIC NOTABLES
      Conc at a point, contact
 Positives:
  - Cheap ($1-$3K)
  - Relatively mature technology
  - Most common surrogate tech.
 Negatives
  - Small op range, pegs
  - Prone to fouling
  - High maintenance
Linganore Creek, MD
An acoustic Doppler current profiler (aDcp)
measures three-dimensional velocity profiles within
the water column using the Doppler shift principle,
whilst the bottom tracking function and acoustic
backscatter can be used to measure bed load
velocity and estimate suspended sediment
concentration.
              BEDLOAD SURROGATE



In cases where the bed is mobile, the bottom
tracking Doppler shift is a function of both the
velocity of the boat and the mobile bed.

The aDcp bottom-track provides a direct estimate
of bed load velocity but not of bed load flux. Bed
load velocity ub can be used to estimate bed load
transport qb from:
              SUSPENDED SEDIMENT SURROGATE



The strength of an aDcp acoustic return is a function of
the instrument and the flow properties, and has been
used to estimate the concentration of suspended
sediment in a wide range of flows. Instrument response
is a function of frequency, transmit power, receiver
sensitivity and distance to the measurement volume,
whilst the size, type and concentration of suspended
sediment are the most important flow properties.
Suspended-Sediment Concentration
   by Acoustic Doppler Current
             Profiler




 From printed literature, Dredging Research, LTD, UK
     A General Comparison of In-Situ ~Cost
         Min –
     Factor    Measures Robust / Inspection
          Max     PSD =
Suspended-Sediment Surrogate Technologies*
 Tech
               particle-size
                             Reliable Frequency $ US
               Conc., %         dist.
    Bulk      0-0.2 fine    Point conc.    Fouls         Often (days       $1K-3K
   Optics     0-1.0 sand                   easily,       to weeks)
                                           Miss ^
                                           conc?
  Acoustics
   1-F…..     0-?           Beam           Yes, very     Infrequent       $5K-10K
   2-F…..     0-3K sand     conc.          Yes, very     Infrequent        $More
              0-20K fines   Beam
                            conc. (and
                            PSD)
    Laser
 LISST-100.. 0-~0.5         conc. &        ? Can foul ? ?                    **
 LISST-SL.. 0-~0.5          PSD            Should not; Manual
                            conc. &        isokinetic   deploy
                            PSD
   Digital    0-? Only      conc. &        ? Can foul?   To be            $3K-$6K
   Optics     tested in     PSD                          determined
              lab.
  Pressure                                                  Infrequent Gray, $3K-5K
           *All require calibration until reliability is proven
               Min ~ 1-      Conc. in a      Yes                   (J.R.     9/2005)
          ** 2%, no top vertical
 Differential The LISST series of instruments range from $5K to $35K in the USA.
     Sediment Transport Models
                                        • Difficult problem – most
                                          models are empirical.
                                        • Usually make simplifying
                                          assumptions about flow.
                                        • Many different formulas
                                          exist.

Most have the form                gs=a(q-qc)
Where gs = bedload transport rate
a=coefficient
 = water density
q = discharge
qc = critical value for initiation of bed material movement
                                 Sediment
                             Transport Models



Table from Chanson, p. 198
             flow-transport interaction
                models - size matters
=f(bed stress, transport rate, grain size (of
  transport and bed surface)
http://www.wdfw.wa.gov/hab/ahg/shrg/24-shrg_sediment_transport.pdf
http://water.usgs.gov/software/


         SEDDISCH

				
DOCUMENT INFO