Designation: D4411 − 03 (Reapproved 2014)´1

Standard Guide for

Sampling Fluvial Sediment in Motion1
This standard is issued under the fixed designation D4411; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.

ε1 NOTE—The footnotes of Table 1 were editorially corrected in July 2014.

1. Scope 2. Referenced Documents

1.1 This guide covers the equipment and basic procedures 2.1 ASTM Standards:2
for sampling to determine discharge of sediment transported by D1129 Terminology Relating to Water
moving liquids. Equipment and procedures were originally D3977 Test Methods for Determining Sediment Concentra-
developed to sample mineral sediments transported by rivers tion in Water Samples
but they are applicable to sampling a variety of sediments
transported in open channels or closed conduits. Procedures do 3. Terminology
not apply to sediments transported by flotation. 3.1 Definitions—For definitions of other terms used in this
1.2 This guide does not pertain directly to sampling to guide, see Terminology D1129.
determine nondischarge-weighted concentrations, which in 3.1.1 isokinetic—a condition of sampling, whereby liquid
special instances are of interest. However, much of the descrip- moves with no acceleration as it leaves the ambient flow and
tive information on sampler requirements and sediment trans- enters the sampler nozzle.
port phenomena is applicable in sampling for these 3.1.2 sampling vertical—an approximately vertical path
concentrations, and 9.2.8 and 13.1.3 briefly specify suitable from water surface to the streambed. Along this path, samples
equipment. Additional information on this subject will be are taken to define various properties of the flow such as
added in the future. sediment concentration or particle-size distribution.
1.3 The cited references are not compiled as standards; 3.1.3 sediment discharge—mass of sediment transported per
however they do contain information that helps ensure standard unit of time.
design of equipment and procedures. 3.1.4 suspended sediment—sediment that is carried in sus-
1.4 Information given in this guide on sampling to deter- pension in the flow of a stream for appreciable lengths of time,
mine bedload discharge is solely descriptive because no being kept in this state by the upward components of flow
specific sampling equipment or procedures are presently ac- turbulence or by Brownian motion.
cepted as representative of the state-of-the-art. As this situation 3.2 Definitions of Terms Specific to This Standard:
changes, details will be added to this guide. 3.2.1 concentration, sediment—the ratio of the mass of dry
1.5 This standard does not purport to address all of the sediment in a water-sediment mixture to the volume of the
safety concerns, if any, associated with its use. It is the water-sediment mixture. Refer to Practice D3977.
responsibility of the user of this standard to establish appro- 3.2.2 depth-integrating suspended sediment sampler—an
priate safety and health practices and determine the applica- instrument capable of collecting a water-sediment mixture
bility of regulatory limitations prior to use. Specific precau- isokinetically as the instrument is traversed across the flow;
tionary statements are given in Section 12. hence, a sampler suitable for performing depth integration.
3.2.3 depth-integration—a method of sampling at every
point throughout a sampled depth whereby the water-sediment
1
This guide is under the jurisdiction of ASTM Committee D19 on Water and is
the direct responsibility of Subcommittee D19.07 on Sediments, Geomorphology,
2
and Open-Channel Flow. For referenced ASTM standards, visit the ASTM website, www.astm.org, or
Current edition approved Jan. 1, 2014. Published March 2014. Originally contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
approved in 1984. Last previous edition approved in 2003 as D4411 – 03 (2008). Standards volume information, refer to the standard’s Document Summary page on
DOI: 10.1520/D4411-03R14E01. the ASTM website.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States

1
 D4411 − 03 (2014)´1
mixture is collected isokinetically to ensure the contribution stable stage-discharge relationship. In small rivers and man-
from each point is proportional to the stream velocity at the made channels, artificial controls as weirs can be installed.
point. This method yields a sample that is discharge-weighted These will produce exceptionally stable and well defined
over the sampled depth. Ordinarily, depth integration is per- stage-discharge relationships. In large rivers, only natural
formed by traversing either a depth- or point-integrating controls ordinarily exist. Riffles and points where the bottom
sampler vertically at an acceptably slow and constant rate; slope changes abruptly, such as immediately upstream from a
however, depth integration can also be accomplished with natural fall, serve as excellent controls. A straight uniform
vertical slot samplers. reach is satisfactory, but the reach must be removed from
3.2.4 point-integrating suspended-sediment sampler—an in- bridge piers and other obstructions that create backwater
strument capable of collecting water-sediment mixtures isoki- effects.
netically. The sampling action can be turned on and off while 5.3 A sampling site should not be located immediately
the sampler intake is submerged so as to permit sampling for a downstream from a confluence because poor lateral mixing of
specified period of time; hence, an instrument suitable for the sediment will require an excessive number of samples.
performing point or depth integration. Gaging and sampling stations should not be located at sites
3.2.5 point-integration—a method of sampling at a fixed where there is inflow or outflow. In rivers, sampling during
point whereby a water-sediment mixture is withdrawn isoki- floods is essential so access to the site must be considered.
netically for a specified period of time. Periods of high discharge may occur at night and during
3.2.6 stream discharge—the quantity of flow passing a inclement weather when visibility is poor. In many instances,
given cross section in a given time. The flow includes the bridges afford the only practical sampling site.
mixture of liquid (usually water), dissolved solids, and sedi- 5.4 Sampling frequency can be optimized after a review of
ment. the data collected during an initial period of intensive sam-
pling. Continuous records of water discharge and gauge height
4. Significance and Use (stage) should be maintained in an effort to discover parameters
4.1 This guide is general and is intended as a planning that correlate with sediment discharge, and, therefore, can be
guide. To satisfactorily sample a specific site, an investigator used to indirectly estimate sediment discharge. During periods
must sometimes design new sampling equipment or modify of low-water discharge in rivers, the sampling frequency can
existing equipment. Because of the dynamic nature of the usually be decreased without loss of essential data. If the
transport process, the extent to which characteristics such as sediment discharge originates with a periodic activity, such as
mass concentration and particle-size distribution are accurately manufacturing, then periodic sampling may be very efficient.
represented in samples depends upon the method of collection. 5.5 The location and number of sampling verticals required
Sediment discharge is highly variable both in time and space so at a sampling site is dependent primarily upon the degree of
numerous samples properly collected with correctly designed mixing in the cross section. If mixing is nearly complete, that
equipment are necessary to provide data for discharge calcu- is the sediment is evenly and uniformly distributed in the cross
lations. General properties of both temporal and spatial varia- section, a single sample collected at one vertical and the water
tions are discussed. discharge at the time of sampling will provide the necessary
data to compute instantaneous sediment-discharge. Complete
5. Design of the Sampling Program mixing rarely occurs and only if all sediment particles in
5.1 The design of a sampling program requires an evalua- motion have low fall velocities. Initially, poor mixing should
tion of several factors. The objectives of the program and the be assumed and, as with sampling any heterogeneous
tolerable degree of measurement accuracy must be stated in population, the number of sampling verticals should be large.
concise terms. To achieve the objectives with minimum cost, 5.6 If used properly, the equipment and procedures de-
care must be exercised in selecting the site, the sampling scribed in the following sections will ensure samples with a
frequency, the spatial distribution of sampling, the sampling high degree of accuracy. The procedures are laborious but
equipment, and the operating procedures. many samples should be collected initially. If acceptably stable
5.2 A suitable site must meet requirements for both stream coefficients can be demonstrated for all anticipated flow
discharge measurements and sediment sampling (1).3 The conditions, then a simplified sampling method, such as
accuracy of sediment discharge measurements are directly pumping, may be adopted for some or all subsequent sampling.
dependent on the accuracy of stream discharge measurements.
Stream discharge usually is obtained from correlations between 6. Hydraulic Factors
stream discharge, computed from flow velocity measurements, 6.1 Modes of Sediment Movement:
the stream cross-section geometry, and the water-surface el- 6.1.1 Sediment particles are subject to several forces that
evation (stage). The correlation must span the entire range of determine their mode of movement. In most instances where
discharges which, for a river, includes flood and low flows. sediment is transported, flow is turbulent so each sediment
Therefore, it is advantageous to select a site that affords a particle is acted upon by both steady and fluctuating forces.
The steady force of gravity and the downward component of
3
The boldface numbers in parentheses refer to the list of references at the end of turbulent currents accelerate a particle toward the bed. The
this standard. force of buoyancy and the upward components of turbulent

2
 D4411 − 03 (2014)´1
currents accelerate a particle toward the surface. Relative in steady flow, the bedload discharge at a point fluctuates
motion between the liquid and the particle is opposed by a drag significantly with time. Also, the discharge varies substantially
force related to the fluid properties and the shape and size of from one point to another.
the particle. 6.1.6 Within a river or channel, the sizes of the particles in
6.1.2 Electrical charges on the surface of particles create transport span a wide range and the flow condition determines
forces that may cause the particles to either disperse or the mode by which individual particles travel. A change in flow
flocculate. For particles in the submicron range, electrical conditions may cause particles to shift from one mode to the
forces may dominate over the forces of gravity and buoyancy. other.
6.1.3 Transport mode is determined by the character of a 6.1.7 For transport purposes, the size of a particle is best
particle’s movement. Clay and silt-size particles are relatively characterized by its fall diameter because this describes the
unaffected by gravity and buoyant forces; hence, once the particle’s response to the steady forces in the transport process.
particles are entrained, they remain suspended within the body
of the flow for long periods of time and are transported in the 6.2 Dispersion of Suspended Sediment:
suspended mode. 6.2.1 The various forces acting on suspended-sediment
6.1.4 Somewhat larger particles are affected more by grav- particles cause them to disperse vertically in the flow. A
ity. They travel in suspension but their excursions into the flow particle’s upward velocity is essentially equal to the difference
are less protracted and they readily return to the bed where they between the mean velocity of the upward currents and the
become a part of the bed material until they are resuspended. particle’s fall velocity. A particle’s downward velocity is
6.1.5 Still larger particles remain in almost continuous essentially equal to the sum of the mean velocity of the
contact with the bed. These particles, termed bedload, travel in downward currents and the particle’s fall velocity. As a result,
a series of alternating steps interrupted by periods of no motion there is a tendency for the flux of sediment through any
when the particles are part of the streambed. The movement of horizontal plane to be greater in the downward direction.
bedload particles invariably deforms the bed and produces a However, this tendency is naturally counteracted by the estab-
bed form (that is, ripples, dunes, plane bed, antidunes, etc.), lishment of a vertical concentration gradient. Because of the
that in turn affects the flow and the bedload movement. A gradient, the sediment concentration in a parcel of water-
bedload particle moves when lift and drag forces or impact of sediment mixture moving upward through the plane is higher
another moving particle overcomes resisting forces and dis- than the sediment concentration in a parcel moving downward
lodges the particle from its resting place. The magnitudes of the through the plane. This difference in concentration produces a
forces vary according to the fluid properties, the mean motion net upward flux that balances the net downward flux caused by
and the turbulence of the flow, the physical character of the settling. Because of their high fall velocities, large particles
particle, and the degree of exposure of the particle. The degree have a steeper gradient than smaller particles. Fig. 1(2) shows
of exposure depends largely on the size and shape of the (for a particular flow condition) the gradients for several
particle relative to other particles in the bed-material mixture particle-size ranges. Usually, the concentration of particles
and on the position of the particle relative to the bed form and smaller than approximately 60 µm will be uniform throughout
other relief features on the bed. Because of these factors, even the entire depth.

FIG. 1 (2) Vertical Distribution of Sediment in the Missouri River at Kansas City, MO

3
 D4411 − 03 (2014)´1
6.2.2 Turbulent flow disperses particles laterally from one 7. Spatial and Temporal Variations in Bedload Discharge
bank to the other. Within a long straight channel of uniform
7.1 Bedload discharges vary both within a section and along
cross section, lateral concentration gradients will be nearly
symmetrical and vertical concentration gradients will be simi- the channel due to variations in the sediment and mean flow
lar across the section. However, within a channel of irregular properties, turbulence, patterns of secondary circulation and
cross section, lateral gradients will lack symmetry and vertical position relative to the bed relief. (See 13.1, also 7.2.) Also,
gradients may differ significantly. Fig. 2(3) illustrates the because of the intimate relationship between bedload discharge
variability within one cross section of the Rio Grande. and the flow forces, particles that move as bedload at one
6.2.3 Sediment entering from the side of a channel slowly section may be immobile or may move as suspended load at
disperses as it moves downstream and lateral gradients may another cross section. As a result, the proportion of bedload
exist for several hundred channel widths downstream. In or discharge to total sediment transport may vary longitudinally
near a channel bend, secondary flow accentuates both horizon- and bedload discharge observed at one section may not be
tal and vertical gradients. Until data have been collected to representative of the bedload discharge at another section.
prove the contrary, one must assume both gradients exist and
7.2 Although data on the temporal variation in bedload
design sampling procedures accordingly.
6.2.4 At sections where spatial variability exists, samples discharge are far from abundant, observations with bedload
must be collected from many regions within a cross section. samplers have shown that discharges vary dramatically and
Only for special conditions will samples from one or two tend to be cyclic. In one study (4) of a river having bed material
points be adequate. of coarse cobbles, bedload samples collected every 3 min
6.2.5 Despite turbulent currents that disperse particles along during a 3-h test showed a coefficient of variation of 41 % and
the direction of flow, the concentration at a fixed point will an oscillation period of about seven minutes. Another study (5),
vary with time even if flow conditions are steady. Temporal conducted in a laboratory flume with a bed of coarse gravel,
variability depends upon many factors. Within a group of showed that the coefficient of variation of bedload samples
samples collected during a short period of time, the concen- collected every minute during a 1-h test was 100 %. Temporal
tration of any sample generally will not deviate from the mean variations at a fixed sampling point are caused, in large
by more than approximately 20 %; however, every sample measure, by the passage of bed forms. Because a single
must be composed of a stream filament at least 50 ft long. measurement at a point probably will not be representative of

FIG. 2 (3) Cross-Sectional Variability of Suspended Material in Two Different Size Ranges, Rio Grande, near Bernardo, NM (a) Contours
in mg/L for Material Between 0.0625 and 0.125 mm; (b) Contours in mg/L for Material Between 0.25 and 0.5 mm

4
 D4411 − 03 (2014)´1
the mean bedload discharge, numerous repetitive measure- tration. To avoid such changes at a sampling nozzle,
ments must be made at each measurement point during a time suspended-sediment samplers must operate isokinetically (or
interval that is sufficiently long to allow a number of bed-form nearly isokinetically). If the velocity at the entrance of the
wave-lengths to pass. Alternatively, the sampling position must sampler nozzle deviates from ambient velocity by less than
be moved longitudinally so that samples are obtained randomly 615 %, the error in concentration will seldom exceed 65 %.
over parts of several bed-form wave-lengths. The angle between the axis of the nozzle and the approaching
flow should not exceed 20°.
8. Spatial and Temporal Variations in Total-Sediment 9.2.2 Two basic types of isokinetic instruments are com-
Discharge monly used to sample suspended sediment. One type (integrat-
8.1 Temporal and spatial variations in the total sediment ing) accumulates the liquid-sediment mixture by withdrawing
discharge result from the combined effects of variations in the it during a long period of time. The other type (trap) instanta-
suspended-sediment discharge and the bedload discharge. De- neously traps a volume of the mixture by simultaneously
tailed information on the extent of temporal variations in total closing off the ends of a flow-through chamber. The integrating
load are scarce; however, as with variations in suspended type collects a long filament of flow, hence, the sample
sediment discharge, the variations in total load can be expected concentration is only slightly affected by short-term fluctua-
to change according to particle size. Ordinarily, at normal river tions in the concentration within the approaching flow. For this
sections, the total load cannot be measured as a separate entity; reason, integrating types are recommended over trap types.
therefore, it is obtained by combining observations of the 9.2.3 For integrating-type samplers it is recommended that
suspended load and the bedload. When the total-sediment the nozzle entrance be circular in cross section and have an
discharge is determined from measurements of the suspended- inside diameter of 4.8 mm (3⁄16 in.) or larger. At the nozzle
sediment and bedload discharges, sufficient sampling must be entrance, the wall thickness should not exceed 1.6 mm (1⁄16 in.)
performed to account for the temporal and spatial variations in and the outside edge should be gently rounded.
both quantities. 9.2.4 To ensure an undisturbed flow pattern, the nozzle must
8.1.1 At certain kinds of unusual sections, such as outfalls, extend upstream from its support which may be a tethered body
sills and weirs, or in highly turbulent flow, all of the sediment or a fixed support strut. An upstream distance of 25.4 mm (1
particles may be entrained in the water; consequently, total in.) is adequate provided the support is well streamlined and its
load can be measured by sampling through the nappe or largest dimension lateral to the flow is not more than 40 nozzle
through the entire depth. Such sections are often called diameters.
total-load sections. At total-load sections, spatial variations in 9.2.5 After entering the nozzle, the sample must be
the total sediment discharge can be significant and are func- conveyed, without a change in concentration, to a container. If
tions of the lateral variations in flow properties, suspended- the volume of the conduit is more than approximately 5 % of
sediment concentration, and bedload discharge. At total-load the sample volume, the velocity within the conduit must be
sections, sampling must be carried out in accordance with the adequate to ensure transport as a homogeneous suspension. A
principles of suspended-sediment sampling and replicate velocity exceeding 17 W is recommended where W equals
samples must be collected at a sufficient number of lateral settling velocity of the largest particle in suspension.
locations to account for variations in the discharge of entrained 9.2.6 Integrating samplers that meet the above requirements
bedload particles. are fabricated commercially in the United States. The samplers,
which are listed in Table 1(6), belong to the “US Series”
9. Selection and Design of Sampling Apparatus
designed and sold by the Federal Interagency Sedimentation
9.1 Apparatus selection depends upon the object of the Project. The samplers are of two types, depth-integrating and
sampling program and the physical and hydraulic characteris- point-integrating.
tics of the site. To sample for total sediment discharge within a 9.2.7 Depth-Integrating Samplers—US Series depth-
straight section of open channel, use a suspended-sediment integrating samplers have an intake nozzle and exhaust port but
sampler in conjunction with a bedload sampler. If initial they do not have a valve; therefore, they sample the water-
measurements show that nearly all of the total load is trans- sediment mixture continuously when submerged. They are
ported in suspension, routine sampling can be simplified by highly reliable because they do not contain moving parts;
eliminating bedload measurements. At an outfall, total load furthermore, they are suitable for use in a sampling technique
may be measured by sampling through the nappe with a termed “depth integration” (see 13.1.4). Depth-integrating
depth-integrating sampler. Because these samplers are cali- samplers have a maximum operating depth (see Table 1) (6).
brated when fully submerged, the depth of the nappe should be Fig. 3(7) shows the shape of one member of the US Series of
great enough to ensure the flow contacts the region down- depth-integrators. Auxiliary equipment includes a cable-and-
stream of the air exhaust port. For continuous sampling of total reel suspension system, or for the DH-48 (8) and DH-81, a
load, a traveling-slot or a stationary-slot sampler may be used. wading-rod suspension. During the depth-integration process, a
9.2 Suspended Sediment Samplers: sampler must be lowered and raised at a uniform rate so
9.2.1 Whenever the fluid within a streamtube accelerates by cable-speed indicators or timing devices are used whenever
changing either its direction or speed, sediment particles tend possible.
to migrate across the streamtube boundaries. This migration 9.2.8 Point-Integrating Samplers—US Series point-
causes a local enrichment or depletion in the sediment concen- integrating samplers have an intake nozzle and exhaust port

5
 D4411 − 03 (2014)´1
TABLE 1 (6) Physical Characteristics of US-Series Depth-Integrating and Point-Integrating Samplers for Collecting Samples of Water-
Suspended Sediment Mixtures (after Table 3-3, National Handbook of Recommended Methods for Water-Data Acquisition)
NOTE 1—[Type: DI, depth-integrating; PI, point-integrating. Available nozzle size: A, 4.8 mm; B, 6.4 mm; C, 7.9 mm. Body material: AL, aluminum;
BR, bronze; PL, plastic; FL, fluoropolymer].
Name Type of Method of Mass, Overall Available Sample Maximum Maximum Distance Body Remarks
Sampler Suspension kg Length, Nozzle Size Container Allowable Calibrated Between Material
m Size, mL Depth, m Velocity, Nozzle
m/s and Sampler
Bottom, mm
US DH-48 DI rod 2.0 0.33 AA , B 473 B
2.7 90 AL for wading.
B
US DH-59 DI cable 10.2 0.42 A, B 473 1.5 114 BR for hand-line operation.
B
US DH-76 DI cable 10.9 0.47 A, B 946 2.0 80 BR for hand-line operation.
B
US D-74 DI cable 28.2 0.66 A, B 473 or 946 2.0 103 BR
B C
US D-74AL DI cable 11.4 0.66 A, B 473 or 946 1.8 111 light weight US D-74.
US DH-81 DI rod 0.8 0.3 A, B, C 1000 4.6 2.1 102 PL, FL for wading
US DH-95 DI cable 13.1 0.61 A, B, C 1000 4.6 2.3 122 BR for hand-line operation
US D-95 DI cable 29.0 0.66 A, B, C 1000 4.6 2.0 122 BR
C
US D-96 DI cable 59.9 0.89 A, B, C 3000 3.8 102 BR/AL Collapsible-bag sampler
C
US DI cable 36.3 0.89 A, B, C 3000 1.8 102 BR/AL Collapsible-bag sampler
D-96–A1
D
US D-99 DI cable 124.7 1.00 A, B, C 6000 4.6 241 BR Collapsible-bag sampler
US P-50 PI cable 135.6 1.12 A 473 or 946 61.0 E 3.0 140 BR
41.0 F
US P-61-A1 PI cable 47.5 0.71 AAAA 473 or 946 54.9 E 2.0 109 BR
473 or 946 36.6 F
US P-63 PI cable 90.4 0.85 A 54.9 F 2.0 150 BR
36.6 F
US P-72 PI cable 17.7 0.71 A 473 or 946 22.0 E 1.6 109 AL
15.5 F
A
4.8-mm nozzle available by special order.
B
Varies with nozzle and container sizes as follows:
Nozzle Size Container Size
473 mL 946 mL
A 4.9 m 4.9 m
B 2.7 m 4.9 m
C
Varies with nozzle and container sizes as follows:
Nozzle Size Container Size
A 33.5 m
B 18.3 m
C 11.9 m
D
Varies with nozzle and container sizes as follows:
Nozzle Size Container Size
A 23.8 m
B 36.6 m
C 67.1 m
E
With 473-mL container.
F
With 946-mL container.

that can be opened and closed while the samplers are sub- 10. Bedload Samplers
merged. They also contain a pressure-equalization system to
10.1 Both in Europe and the United States many different
ensure that the pressure within the sample container equals the
hydrostatic pressure whenever the intake-exhaust valve is kinds of bedload monitoring apparatus (9) have been devel-
opened. These features allow the samplers to be used for oped to measure the transport of a wide variety of bed-material
sampling by either the depth integration or point integration particles that occur in nature. In general, each kind of apparatus
(see 13.1.3) techniques. Maximum allowable depths listed for was designed to monitor a particular range of bedload sizes and
these samplers in Table 1(6) apply when they are used for point transport rates. Two broad classifications exist, direct-
integration. When the samplers are used for depth integration measuring apparatus and indirect-measuring apparatus. Direct-
starting at the water surface, the depth limitations given in measuring apparatus collect and accumulate bedload particles
footnote B of Table 1(6) specify the length of the allowable for a given period of time. Indirect-measuring apparatus
two-way vertical sampling path for any single-sample con- monitor some property of the bedload or some phenomena that
tainer; segments of an allowable path length can be sampled occurs as a result of bedload movement. In addition, bedload
throughout all or any part of the maximum allowable depth by discharge can be determined from measurements of the rate of
using multiple containers and opening and closing the intake- (1) migration of bedforms, (2) movement of tracer particles, (3)
exhaust valve appropriately. In addition to a suspension and deposition or erosion in a given area, and (4) change with
speed indicating system, the samplers also require a source of distance in the concentration of some nonconservative property
electrical power.

6
 D4411 − 03 (2014)´1

FIG. 3 (7) US D-74 Suspended Sediment Sampler

associated with the bedload particles. This nonconservative 10.3 Indirect-Measuring Apparatus—Most indirect-
property, such as radioactivity, must have a known time rate of measuring apparatus are acoustical devices that measure (1) the
decay. magnitude and frequency of particle-sampler or particle-
10.1.1 No portable direct-measuring apparatus nor indirect particle collisions or (2) the attenuation of energy. Apparatus of
technique is generally accepted at this time as being entirely this type ordinarily give only qualitative information and their
suitable for determining bedload discharge. outputs must be correlated with known discharges to provide
10.2 Direct Measuring Apparatus—Direct-measuring appa- quantitative results. Acoustic devices are seldom used in
ratus can be classified into four general categories; box or routine data collection programs.
basket samplers, pan or tray samplers, pressure-difference
samplers, and slot or pit samplers. 11. Total-Sediment Discharge Samplers
10.2.1 Box or Basket Samplers—Enclosures are open at the 11.1 Because the total sediment discharge is composed of
upstream end and possibly at the top, and have either solid suspended-sediment particles, which moves along within the
sides, mesh sides, or a combination of both. Particles are body of the flow essentially at stream velocity, and bedload
retained within the sampler either by being screened from the particles, which moves in an interrupted fashion essentially in
flow or by settling in regions of reduced flow velocities within continuous contact with the bed, no practical sampler has been
the sampler. designed for sampling total-sediment discharge at regular river
10.2.2 Pan or Tray Samplers—These samplers collect par- sections. Normally, the total sediment discharge is determined
ticles that drop into one or more sections or slots after the from separate measurements of the suspended sediment dis-
particles have been transported up an entrance ramp. charge and the bedload discharge. Conventional sampling
10.2.3 Pressure-Difference Samplers—Essentially box or equipment can be used to measure the total sediment discharge
basket samplers that have entrances or other features that create at certain sections termed total-load sections. At an outfall, a
a pressure drop that overcomes the flow resistance within the sill, a weir, or a section where flow turbulence is sufficient to
sampler and thereby keeps flow velocities at the entrance about entrain all the sediment within the flow, suspended-sediment
the same as the stream velocity. sampling equipment and techniques can be used to determine
10.2.4 Slot or Pit Samplers—These samplers consist of the discharge of particles finer than 2 mm. For particles coarser
collection chambers that accumulate particles as they drop over than 2 mm, use equipment that is capable of collecting and
the forward edge of a chamber that is buried in the stream bed. retaining coarse particles, and that is based on the isokinetic

7
 D4411 − 03 (2014)´1
principles of suspended-sediment sampling. Such equipment sampler requires electric power, and is relatively insensitive to
includes slot samplers. Economic considerations usually pre- approach conditions. In situations where only infrequent clog-
clude the construction of artificial total-load sections except on ging is anticipated, satisfactory performance may be obtained
small streams. The ASCE Manual (10) illustrates a large but by using brushes or other equipment to periodically clean the
expensive turbulence flume. slot. Fiber brushes mounted so that the slot is brushed before
11.2 If sampling can be conducted at a free outfall, slot each pass through the flow nappe will usually assure satisfac-
samplers can be installed. By means of a slotted conduit tory performance. Fig. 5(12) illustrates one type which has
positioned in the outfall, the slot sampler diverts a fraction of been tested.
the water-sediment mixture into a suitable container. Slot 11.6 Rotating (Coshocton-Type) Sampler—The rotating
samplers have been used extensively in erosion research and in (Coshocton-type) sampler, (13) which is in the traversing
laboratory flumes but standard designs have not been perfected category, consists of an elevated slot affixed to a revolving
for sampling sediment or industrial waste water in open water wheel that is mounted on the downstream end of a small
channels or streams. Slot samplers are normally used in H-flume. Discharge from the flume falls on the water wheel
conjuction with a flume, weir, or other flow measuring devices. and causes it to rotate. With each revolution the sampling slot
11.3 The slot width must be adequate to permit free passage cuts across the flow jet and extracts a sample. The sample falls
of the largest sediment particle; the conduit must be stream- into a collecting pan beneath the wheel and is routed through
lined to minimize disturbance to the flow. Sides of the slot may a closed conduit to a storage tank. A typical Coshocton-type
be formed from rigid-metal sheets that are supported so that the sampler is pictured in Fig. 6(13) and Fig. 7(13). Sampler size,
slot opening faces the flow. The slot edges should be knife maximum discharge rate, sampling ratio, and other pertinent
sharp and true to line. A tube or flexible pipe connected to the data are given in Table 2(14).
bottom of the slot carries the sample to a suitable storage 11.6.1 The Coshocton sampler requires no external power
container. The sampler is mounted on the downstream end of a source, however it is sensitive to upstream approach condi-
flow measuring device with a free overfall. The height of the tions. Rotation of the wheel may become erratic at stream
sampler slot must exceed the depth of flow. Some slot samplers discharges that exceed 80 % of the flume capacity.
will not function properly if the flow transports debris capable
of clogging the slot. The slot may be located at a fixed point in 12. Hazards
the flow or it may be propelled across the flow. Accordingly, 12.1 Personal Clothing, Equipment, and Training—
slot samplers may be divided into two broad categories, Operators should wear protective footgear and protective
stationary or transversing. headgear, safety glasses, and leather gloves in addition to
11.4 Stationary Slot Samplers—Stationary slot samplers are high-visibility clothing that is warm enough to prevent hypo-
simple to build and operate. They require no external source of thermia and a Coast Guard approved personal flotation device.
power. To enhance self-clearing of debris, incline the slot at a Where drowning is a hazard, perform sampling by teams that
slight downward angle. Fig. 4(11) illustrates several types that are wearing Coast Guard-approved personal-flotation devices
have been tested. Samples are extracted along one fixed line so and that are proficient in both swimming and first aid.
they are less representative than those collected with a travers- 12.2 Electrical Hazard—(Warning—Equipment powered
ing slot. from low-voltage batteries is safer than equipment powered
11.5 Traversing Slots—Traversing slots collect a sample from 120-V, a-c distribution circuits. Regardless of the power
representative of the entire cross section. The vertical slot source, ground the frames of hoists and other equipment to

FIG. 4 (11) Cross Sections and Installation of Slot-Type Sampler

8
 D4411 − 03 (2014)´1

NOTE 1—1 in. = 25.4 mm. 1 ft = 0.3 m.

FIG. 5 (12) Space Required for the Traversing-Slot Sampler Mounted on a 2-ft Parshall Flume

nearby metal objects such as bridge railings, bridge decks, or 13.1.1 Because of spatial variations in suspended sediment
boat hulls. Use ground-fault detectors where applicable. Dur- concentrations and in flow velocity, the discharge of suspended
ing electrical storms, operators should retreat to low ground or sediment through an area at any given instant is defined by Eq
take cover in a building or metal-topped vehicle.) 1(15) as follows:
12.3 Vehicles—Equip vehicles that must be parked on road G ss 5 * C U dA (1)
shoulders with warning lights and flares in compliance with A

local regulations. Isolate the cargo area from the driver- where:
passenger area; lash the cargo to prevent tipping or sliding.
Gss = “instantaneous” suspended sediment discharge
12.4 Sampling Wadable or Ice-Covered Streams—At cross- through a section of area A,
ings that appear marginal from safety aspects, test the surface U = velocity of sediment particles through an elemental
with a rod or ice chisel, and wear a safety line anchored to a area dA,
firm object on the shore. Wear a Coast Guard approved C = suspended sediment concentration in the elemental
personal flotation device. spatial volume U t'dA,
12.5 Sampling from Overhead Cableways and Bridges— for which:
Inspect supports and safety railings regularly for loose, worn, t' = unit of the time used to express U.
or weak components. At sites where trees or other heavy debris In the practical application of Eq 1, U is considered to equal the
may snag a submerged sampler, the operator should be flow velocity and C is considered to be constant during any
prepared to sever the suspension cable if the sample cannot be given sampling period.
retrieved. Wear a Coast Guard approved personal flotation 13.1.2 Three different techniques are commonly used to
device. evaluate Eq 1; point integration, depth integration, and area
integration. In point integration, samplers and sampling proce-
12.6 Reports and MedFical Treatment—Report all accidents
dures are designed to yield spatial concentrations at a series of
and potentially dangerous situations promptly to the local
points throughout an area. These concentrations together with
safety officer. To save valuable time when an accident occurs,
flow velocities from individual points are used to define
procedures for obtaining professional emergency treatment
concentration and velocity gradients, which, in turn are inte-
should be clearly understood by all operators.
grated according to Eq 1 to give the instantaneous suspended-
sediment discharge through the area.
13. Sampling Techniques 13.1.3 To sample by point integration, divide the flow area
13.1 Techniques for Sampling Suspended Sediment: laterally into increments and collect samples at several depths

9
 D4411 − 03 (2014)´1

FIG. 6 (13) The N-1 Coshocton-Type Runoff Sampler

along a vertical in each increment. Select increment widths and the ith element. The suspended-sediment discharge, Gss,
sampling depths so that between adjacent sampling points the through the sampled area then is defined by Eq 2 as follows:
difference in concentration and difference in velocity are small N

enough to conform with desired accuracy. Use a P-61-A1 or G ss 5 ( CQ

i50
i i (2)
any other sampler that meets requirements of Section 9.
13.1.4 In depth integration and area integration, the sam- Since by definition, Ci = wi/vi, w = ∑ Ni = 0 wi, and Cm = w/v,
pling equipment and procedures are designed to mechanically and by virtue of the sampling technique, vi/v = Qi/Q.
and hydraulically perform the integration over the flow area.
With both depth integration and area integration, an isokinetic where:
sampler is traversed through the flow so that each incremental wi and vi = mass of sediment and volume of mixture,
volume of mixture collected from the corresponding element of respectively, collected from the ith element,
traversed area is in the same proportion to the sample volume
as the stream discharge in each corresponding element is to the w, v, and Cm = mass of sediment, volume of mixture, and
stream discharge in the sampled area. This procedure yields the “discharge-weighted” concentration,
samples having “discharge-weighted” concentrations that can respectively, in and of the sample collected
be multiplied directly with the stream discharge through the from the sampled area, and
sampled area to yield the instantaneous suspended-sediment Q = stream discharge through the sampled area.
discharge through the area. The following derivation, which Substituting the defining equations and the “sampling tech-
uses Eq 1 in discrete form, mathematically explains the nique” equation in Eq 2 we obtain:
concept. Consider a sampled area divided into N elemental
(Sv DS D
N
N
areas of size ∆ x∆y. Let Qi be the water discharge and Ci be the wi v iQ Q Qw
suspended-sediment concentration of mixture flowing through
G ss 5
i50 i v
5
v (w 5i50
i
v
5 C mQ (3)

10
 D4411 − 03 (2014)´1

NOTE 1—1 in. = 25.4 mm. 1 ft = 0.3 m.

FIG. 7 (13) Details of the N-1 Coshocton-Type Runoff Sampler

TABLE 2 (14) Size Schedule for Coshocton Samplers concentration and the stream discharge for the entire cross
Sampler No.
Wheel
Capacity
Headroom
Aliquot
Approximate section. In depth integration, the sampled area is only that part
Diameter Requirement Weight of the stream cross section transversed by the intake nozzle at
3
Ft Ft /s Ft Pct Lb a single vertical. To determine the suspended-sediment dis-
N-1 1 1⁄ 3 1 1⁄2 1 26
N-2 2 2 2 1⁄2 1 ⁄2 85 charged through an entire cross section, a series of verticals
N-3 3 5 1⁄2 3 3⁄4 1 ⁄3 270 must be sampled by depth integration. By assuming that the
discharge-weighted sediment concentration from each vertical
represents a certain proportion of the total stream discharge, a
13.1.5 In area integration, the entire flow cross section is discharge-weighted sediment concentration for the entire cross
sampled, hence, Cm and Q are the discharge-weighted sediment section can be obtained and combined with the total stream

11
 D4411 − 03 (2014)´1
discharge to give the suspended-sediment discharge through more verticals are required. Make selection of intervals on the
the cross section. The accuracy of the samples improves with basis of detailed information on the lateral distribution of
an increase in the number of sampling verticals. stream discharge. If such information is unavailable make the
13.1.6 In principle, the fixed slot performs depth integration selection only after a visual survey.
at a single vertical. Instead of sampling along the vertical with 13.4.3 Establish the distance, S, of the first sampling vertical
a moving nozzle, the slot instantaneously samples the whole from the edge of the water, by first dividing the selected
depth. sampling interval, I, into the surface width, W, to ascertain the
number of times, N, the interval will divide into the width
13.2 Depth Integration:
evenly; then, compute the distance E from E = 1⁄2 (W − NI );
13.2.1 To sample by depth integration, collect water- and finally, determine S from (a) S = E + I ⁄2 when E < I/4, (b)
sediment mixture along a vertical line throughout the entire S = I ⁄2 when E = 0, or (c) S = E when E ≥ I/4. Subsequent
depth by using a US D- or P-series sampler. While a sample is sampling verticals are spaced according to the selected sam-
being collected, the sampler must be moved vertically at a pling interval. The last vertical is located at a distance of S units
uniform velocity termed the transit rate (16). With the D series, from the other edge of the water.
sampling is continuous through the entire depth. P series
13.4.4 By experimentation, determine a transit rate at the
samplers are equipped with a valve so sampling may be
vertical with the greatest water discharge per foot of width. The
continuous or interrupted in a series of segments through the
rate should be the slowest rate possible that falls within the
depth. The only basic requirements are that at each sampling
optimum range designated in the appropriate graph in Fig.
vertical, (1) the entire depth is sampled isokinetically, (2) the
8(17) and Fig. 9(18); this rate is also used at all other sampling
vertical transit rate in any given direction is the same over all
verticals.
parts of the depth and the rate never exceeds the product of K,
13.4.5 Lower the sampler to the water surface then, while
(Fig. 8(17) and Fig. 9(18)) multiplied by the mean stream
keeping the nozzle out of the water, allow the sampler to
velocity at the vertical, and (3) sampling does not extend over
become oriented with the flow. To sample, lower the sampler to
half the distance KeVi/An, in two-way integration (K e must be
the streambed and then raise the sampler to the surface at the
≥ 0.4
transit rate determined from 13.4.4. Once established, maintain
and is the largest K value possible for a given nozzle area, An,
the transit rate while lowering and raising; furthermore, main-
and sample volume, Vi). Fig. 8(17) and Fig. 9(18) show graphs
tain the same transit rate at all verticals.
of K versus stream depth for standard nozzle sizes and for 1-
and 2-pint containers. K is a dimensionless number. 13.4.6 Move to the next vertical and repeat the sampling
process. Water-sediment mixture from more than one vertical
13.3 Depth Integration Throughout a Flow Cross Section: can be accumulated in a single bottle; however, care must be
13.3.1 Sampling throughout a cross section requires depth taken not to sample a vertical if the additional mixture will
integration at several verticals spaced across the section. Two cause the bottle to overfill.
methods are commonly used: the equal-width-increment (EWI) 13.4.7 Mark each bottle with appropriate identification in-
method and the equal-discharge-increment (EDI) method. formation. Prior to analysis, samples from all verticals may be
Either method is applicable, but depending on the situation, composited.
one method may be more convenient to apply than the other. 13.4.8 The number of verticals initially selected may be
13.4 EWI (Equal-Width Increment) Method: reduced if it can be established that fewer verticals will provide
13.4.1 The EWI method produces for each increment a the required accuracy.
sample volume proportional to the stream discharge through 13.5 EDI (Equal-Discharge Increment) Method:
the increment. All increments are of equal width and sampling 13.5.1 The EDI method produces a group of samples each
verticals are located at the center of each increment. of which represents the same proportion of the total water
Furthermore, the transit rate is the same at all verticals. This discharge. If the samples are of equal volume (within 10 % of
method produces a gross sample in which the concentration of each other), they can be composited to form a single gross
the sediment is weighted both vertically and laterally according sample representative of the cross section. Otherwise, each
to stream discharge. The method is often used in shallow, sample must be analyzed separately then concentration values
sandbed streams where the discharge rating is unstable. The must be averaged numerically to obtain the mean discharge-
number of verticals depends upon the streamflow, the sediment weighted concentration for the cross section. The method is
characteristics of the stream, and the sampling accuracy commonly used to sample rivers that have stable channels and
desired. In general, not more than 25 verticals will be neces- stable discharge rating.
sary. 13.5.2 Use stream-gaging equipment or other appropriate
13.4.2 Initially, select a sampling interval such that at least apparatus to accurately measure the lateral distribution of water
several sampling verticals are located within each part of the discharge in the cross section. Arrange the results to express
cross section where the suspended-sediment discharge is sub- accumulated discharge as a function of distance from one edge
stantially different from the other parts. In sections that have a of the water. Select a sufficient number of verticals to be
uniform shape and a relatively uniform lateral distribution of sampled so as to assure that at least one vertical is located in
suspended-sediment discharge, 10 or more sampling verticals each part of the cross section where the stream discharge per
are ordinarily sufficient. In sections that have a non-uniform foot of width is distinctly different from the other parts. For
lateral distribution of suspended-sediment discharge, 20 or uniformly shaped sections where the lateral distribution of

12
 D4411 − 03 (2014)´1

NOTE 1—1 in. = 25.4 mm. 1 ft = 0.3 m.

NOTE 2—Sampler’s transit rate equals K multiplied by mean flow velocity at the sampling vertical.
FIG. 8 (17) Transit Rates for Depth-Integrating Sampler Type US D-74 with Pint Sample Container

suspended sediment discharge is relatively uniform, seven sample at one vertical that is located so half of the increment’s
sampling verticals will usually be satisfactory. For irregular discharge is on one side of the vertical. The transit rate at a
sections more verticals are required; 14 verticals will usually vertical must be uniform; but transit rates may differ among the
be sufficient. Divide the number of verticals, N, into the total verticals. Care must be taken not to overfill the sample
water discharge to determine the discharge to be represented by container nor to exceed the maximum transit rate designated in
each sampling vertical, and then partition the width into N the appropriate graph in Fig. 8(18) and Fig. 9(18).
increments of equal discharge (19). Within each increment,

13
 D4411 − 03 (2014)´1

NOTE 1—1 in. = 25.4 mm. 1 ft = 0.3 m.

NOTE 2—Sampler’s transit rate equals K multiplied by mean flow velocity at the sampling vertical.
FIG. 9 (18) Transit Rates for Depth-Integrating Sampler Type US D-74 with Quart Sample Container

13.6 Area Integration: sample element to the volume of the entire sample is equal to
13.6.1 To sample by area integration, a traversing slot must the ratio of the discharge through the area element to the
be installed in accordance with 11.3. The slot must sample discharge through the entire cross section; hence, the concen-
through the entire depth of flow and must sample isokinetically tration of the sample is “discharge weighted.”
at all depths. At a uniform velocity, move the slot laterally
across the flow. The sample consists of all liquid and sediment 14. Techniques for Sampling Bedload
that enters the slot during one traverse. By this procedure an 14.1 With direct-measuring bedload samplers, information
element of sampled volume is collected from every area on the bedload discharge is obtained by accumulating particles
element in the cross section. The ratio of the volume of a for a given period of time. To average spatial and temporal

14
 D4411 − 03 (2014)´1
variations in the transport rate, a number of individual samples data necessary to locate the position of sampling verticals for
must be collected at each of several different lateral positions the EDI sampling method.
across the width of the flow section. Whenever it is possible, 15.1.2 If temporal variations in both stream discharge and
many samples should be collected and analyzed statistically to suspended-sediment concentrations are significant, the quantity
determine the optimum number necessary for a given accuracy. of suspended sediment, S, discharged through an area during a
Because the rate of bedload transport may vary over the length specified duration, T, must be determined as follows:
of the individual bed forms, samples must be collected from a
number of random positions along the lengths of several bed S5 * * C U dA dT 5 * Q dT
T A T
s (4)
forms.
Whenever sampling is in accordance with depth- or area-
14.2 The number of lateral sampling positions required to integration procedures and concentrations are discharge
adequately represent the discharge depends on the character of weighted, Eq 4 becomes the following:
the lateral variations in transport. If both the transport rate and
the bed form configuration are fairly uniform laterally, sam- S5 *C
T
m QdT (5)
pling sequentially at numerous points across a section provides
samples from a number of random positions along the length of If, during time period T, Cm, or Q or both, are nearly
the bed forms. Hence, if traverses are made repetitively over a constant, integration of Eq 5 can be simplified as follows:
long enough time for several bed-form wave lengths to pass by
the measuring section, the necessity to sample both randomly if C m 5 constant 5 C̄, then S 5 C̄ * Q dT 5 C̄Q̄T
t
(6)
along the length of the bed forms and at different lateral
positions is satisfied. if Q 5 constant 5 Q̄, then S 5 Q̄ *C
T
m dT 5 C̄ m Q̄T (7)

14.3 If the sampler is supported on a suspension line and if

if C m 5 constant 5 C̄ m and Q 5 constant 5 Q̄, (8)
the velocities or bed-form relief is high, extreme care must be
taken to avoid inadvertently collecting stationary bed material. then S 5 C̄ m Q̄ * dT 5 C̄ m Q̄T
As the sampler is lowered into layers of progressively lower T

velocity, the drag force on the sampler continually decreases so 15.1.3 If the sample was collected by a slot sampler, the
the unit achieves an upstream motion which must be arrested sample is a time-integrated composite. Measure both the mass
before it strikes the bed and scoops bed material. Also, when of sediment and volume of the mixture in the sample container.
the sampler is retrieved, there is a tendency for it to slide Divide each quantity by R, the sampling ratio, to obtain the
upstream prior to lifting off the bed; during this process mass of sediment and volume of mixture discharged by the
scooping is possible. Scooping can sometimes be reduced by channel.
using a front stay-line to limit the downstream motion. Even 15.1.4 For a fixed slot:
with a stay-line, extreme turbulent pulsations combined with
R 5 W s /W j (9)
line elasticity may cause the sampler to oscillate on the bed and
thereby scoop some material. Scooping can be minimized by 15.1.5 For the vertical traversing slot:
installing devices on the sampler that cause its entrance to lift
R 5 W s T i /W j T 2 (10)
off the bed immediately upon retrieval.
where:
15. Calculation Ws = slot width,
15.1 Calculation of Suspended Sediment Discharge: Wj = width of the flow jet,
Ti = time the sampler is in the flow jet per traverse, and
15.1.1 Instantaneous suspended-sediment discharge can be T2 = time for one complete traverse.
computed from either Eq 1 (see 13.1.1) or Eq 3 (see 13.1.4). In
either case, the accuracy of the result depends upon the 15.1.6 For the Coshocton-type sampler (20):
representiveness of the samples and the accuracy of the R 5 W s /2πr (11)
stream-discharge measurements. When sediment discharge is
where:
determined from point-integrated samples in accordance with
Eq 1, accurate measurements of vertical and lateral velocity r = radial distance from the center of rotation.
gradients are required. These are best obtained from point 15.1.7 Eq 1-6 are written without specification of units. In
velocities measured at the time of sampling. Suspended- addition to the preferred units for concentration, mass of
sediment discharge computed according to Eq 3 requires sediment per unit volume of mixture, concentration may be
accurate data on stream discharge at the time of sampling. At expressed in terms of volume of sediment per unit volume of
many locations, an accurate discharge can be obtained from the water-sediment mixture or in terms of mass of sediment per
gauge-height record and a stage-discharge rating curve. If an unit mass of water sediment mixture. Parts per million by mass
accurate rated discharge is not available, stream discharge and milligrams per litre are commonly used. Stream discharge
should be measured immediately before sampling or immedi- is ordinarily expressed as volume per unit time and sediment
ately after sampling. When the stream discharge is measured discharge is usually expressed as mass per unit time. Because
prior to sampling and is measured according to conventional of the disparity in units, each of the equations requires a units
stream-gaging techniques, the measurement also provides the conversion constant to make it dimensionally correct. The

15
 D4411 − 03 (2014)´1
constant may involve the density of either the solid sediment or 16.1.3 Particles transported in suspension can be sampled
the water-sediment mixture in addition to mass, volume and more accurately than particles transported near or on the bed;
time conversions. Reported values for concentration, stream consequently, for a given flow condition, sampling errors
discharge, and sediment discharge should be specified in metric increase with an increase in the size of the particles in
units. transport.
15.1.8 A sampler in the US Series will not sample from a 16.2 Suspended Sediment Samplers:
zone close to the streambed. This zone extends upward from 16.2.1 With the US Series of samplers the largest errors
the bed to a height equal to the distance from the sampler occur while sampling close to the bed. Each sampler is
nozzle to the sampler bottom (see Table 1). As a result, constructed with its nozzle several centimeters above the
depth-integrated samples collected with these samplers do not bottom of the sampler; therefore, in a channel with a firm bed,
contain a contribution from this unsampled zone. When the the flow between the bed and the nozzle entrance is unsampled.
suspended-sediment discharge is computed from Eq 3 by using Samplers should be chosen or designed to minimize this
the discharge-weighted concentration from such samples, there unsampled depth. In a channel having a bed composed of soft
is an unaccounted portion of the total suspended-sediment deposits or dunes, oversampling may result if the sampler is
discharge that is called the “unmeasured suspended-sediment allowed to settle into the bed or into a trough. In deep streams
discharge.” This discharge equals Q"(Cm " − Cm) in which Q" the long length of submerged cable and the heavy weight of the
is the water discharge in the unsampled zone and Cm"is the sampler act to increase the time required to reverse the
discharge-weighted concentration in the unsampled zone. If the direction of sampler motion. This reversal time combined with
suspended sediment is distributed uniformly with depth so that the high concentration near the bed may result in oversam-
Cm" = Cmor if the unsampled depth is very small relative to the pling.
total stream depth, then the unmeasured suspended-sediment 16.2.2 A point-integrating or a depth integrating sampler
discharge is usually considered to be inconsequential. should be checked to ensure it samples isokinetically. The first
However, if the suspended sediment concentration gradient step is to adjust the balance of the sampler. The nozzle must be
increases significantly near the streambed or if the unsampled level to keep the entrance facing directly into the approaching
zone is a large fraction of the sampled depth, then the flow. Adjustments can be made by adding or removing weights
unmeasured discharge can be significant. from the lower tail vane (see Fig. 3).
16.2.3 The hydraulic tests can be performed either in a
16. Precision and Bias flume or in a tow tank (21). If a flume is used, the breadth and
16.1 General Comments: depth must be sufficient to eliminate blockage effects between
16.1.1 The selection of a sampling procedure and the the walls of the flume and the body of the sampler. Tests
evaluation of probable measurement error must be based on indicate a flume width of 3.0 ft and a water depth of 6.0 ft are
statistical parameters that are site dependent. These parameters sufficient for testing P-61’s, which are the largest of the
can be estimated by analyzing results from an initial sampling commonly used suspended-sediment samplers. The test section
program and by studying the nature of the sediment source. of the flume must be longer than about 18 ft.
This initial sampling should be intensive, so that spatial 16.2.4 The first step in performing the hydraulic tests is to
variability and temporal variability can be determined. The set the desired flow speed at the test point, which is usually
basic data can also be used to estimate errors associated with taken midway between the flume walls and about two feet
more abbreviated programs or with more simplified sample- below the water surface. Measure the speed Vf with a rotating-
collecting procedures. bucket current meter. Remove the meter before proceeding to
16.1.2 In a large river, sampling errors that occur because of the next step.
a change in equipment or procedure can be estimated by 16.2.5 If the sampler is a point integrator, close the sampling
comparing the data sets; however, this comparison will not valve before positioning the nozzle at the test point. When all
reveal sampling errors that are common to all data sets. In is ready, switch the valve open and at the same instant, start a
some small streams and flumes, more accurate error analysis timer. After a few seconds, close the valve and stop the timer.
can be performed by comparing sediment discharge computed 16.2.6 Hoist the sampler above the flow and measure the
from samples with sediment discharge obtained from an volume collected in the sample container. The optimum vol-
independent measurement. This comparison will reveal all ume is about 80 % of the bottle’s rated capacity. If the volume
sampling errors regardless of their source. One experiment is too small, discard the sample and collect another during a
technique involves the use of a vibrating-type of material longer interval; if the volume is too large, shorten the interval.
feeder. With a feeder, inject dry sediment into the flow at a 16.2.7 The flow speed inside the nozzle, Vn (in feet per
suitable location upstream of the sampling site. Sediment second), is given by the equation Vn = KQ/T where Q is the
discharge determined by sampling can be compared with the sample volume in millilitres, T is the sampling interval in
sediment injection rate. A correction must be made for sedi- seconds, and K is a constant that depends on the diameter of the
ment stored within the channel between the sampling site and hole at the nozzle’s entrance. K is 0.1841 for a 3⁄16 in. hole,
the feeder. 0.1036 for a 1⁄4 in. hole, and 0.0663 for a 5⁄16 in. hole.

16
 D4411 − 03 (2014)´1
16.2.8 For isokinetic sampling, Vn must match Vf; however, sampler. A calibrated tank will greatly facilitate the measure-
in practice a 10 % deviation from this ideal condition is ment of sample volume.
permissible. In other words, Vn should be between 0.9 Vf and 16.3.3 Determining the sediment sampling ratio is more
1.1 Vf. At flow speeds slower that about 1.5 ft/s, a larger difficult. Both the liquid and sediment discharge must be
deviation is acceptable. known. Thorough mixing of a known quantity of sediment in
16.2.9 Vn can be increased by enlarging the hole where the the flow upstream from the sampler and accurate measure-
sample water emerges from the nozzle. Use a reamer with a ments of sample volume and weight are required. When the
taper of 1⁄4 in./ft to cut away a small amount of material. sediment concentration in the sample is the same as the
Proceed cautiously: alternative between reaming and hydraulic sediment concentration in the flow, the sediment sampling ratio
testing. Once removed, the material cannot be replaced. is assumed to equal the liquid sampling ratio.
Normally, this reaming operation is performed only once when 16.3.4 If either sampling ratio varies with stream discharge,
a sampler is new. a composite sample collected during a period of variable flow
16.2.10 If the sampler under test is a depth integrator, set Vf may not be representative of the total flow. Samples compos-
as explained in 16.2.3. Then lower the sampler and start the ited under such conditions must be discharge weighted to be
timer at the instant the nozzle penetrates the water surface. truly representative.
Hold the sampler at the test point for a few seconds, then hoist 16.3.5 Normally the assumption is made that the sediment
the sampler above the flow. Stop the timer when the nozzle sampling ratio is the same as the liquid sampling ratio. This
breaks through the water surface. may not be true if the sediment is composed of heavy particles.
16.2.11 Refer to 16.2.6 through 16.2.8 for computations and Calibration to determine the sediment sampling ratio is re-
reaming. quired.
16.2.12 If a tow tank is used for the hydraulic tests, the 16.4 Vertical Slot Samplers:
breadth and depth must be sufficient to eliminate blockage 16.4.1 The liquid sampling ratio of the vertical slot, Fig. 5,
effects between the walls and sampler. Satisfactory results have (12) increases with increasing flow depth (discharge) and
been obtained in a tank 12 ft wide and 12 ft deep. The increases as Ws/Wj increases.
minimum length of the tank must be about 450 ft if the tests are 16.4.2 The accuracy of vertical slot samplers is dependent
to be run at speeds ranging up to 10 ft/s. on the precision used in construction. The slot must be straight
16.2.13 If the sampler is a point integrator, close the and must have knife-sharp edges. Flow within the sampler and
sampling valve before positioning the sampler at the test point. from the sampler to the storage tank must be unrestricted to
When the carriage has stabilized at the desired speed, open the prevent sediment deposition.
sampling valve and at the same instant start the timer. 16.5 Inclined Slot:
16.2.14 After a few seconds, close the valve and, at the 16.5.1 The accuracy of inclined slot samplers is dependent
same instant, stop the timer. Refer to 16.2.5 for the optimum on the precision used in construction. The slot must be straight
sample volume. It may be necessary to collect samples smaller and have knife-sharp edges. Flow within the sampler and from
than the optimum in order to stop the carriage safely at the end the sampler to the storage tank must be unrestricted to prevent
of the run. sediment deposition.
16.2.15 Refer to 16.2.6 through 16.2.8 for computations and 16.5.2 Limited test data (11) for the inclined slot (Fig. 4)
adjustments. In the towing tests, Vf corresponds to the towing show that the sampling ratio (S): (a) does not change with
speed that is usually read from a meter aboard the towing moderate changes in Reynolds number (from 1600 to 3600);
carriage. (b) is relatively constant for angles of inclination (β) between
16.2.16 If the sampler is a depth integrator, the tow tank 5° and 20°; (c) increases as the ratio of slot width (Ws) to jet
carriage must be moving at a steady speed before the sampler flow width (Wj) increases; (d) decreases as the angle (α) formed
is lowered into the water. Start the timer at the instant the by the sides of the sampler increases; and (e) decreases as flow
nozzle penetrates the water surface. Wait a few seconds, then depth (discharge) increases. A relatively constant sampling
hoist the sampler. Stop the timer at the instant the nozzle clears ratio may be obtained for a range of flow depths (discharge) by
the water. Refer to 16.2.6 through 16.2.9 for computations and using a slightly diverging slot that increases in width toward
adjustments. the downstream end. The width of the slot at any distance from
16.3 Slot Samplers: the end of the flume will depend upon the shape (width) of the
16.3.1 To assure acceptable performance, slot samplers nappe at that point (22).
must be calibrated under conditions for which they were 16.6 Coshocton Samplers:
designed and are to be used. Their liquid sampling ratio and 16.6.1 Coshocton samplers will not function precisely as
sediment sampling ratio should be calibrated for a wide range indicated by Eq 11. The liquid sampling ratio is influenced by
in discharge, sediment concentration, and sediment particle the following: (a) the speed and uniformity of rotation
size. of the sampling wheel and slot; (b) size and geometry of
16.3.2 The volumetric or weight method of calibration is the slot; (c) position of the slot relative to the jet; (d) flow
recommended. If the stream discharge is known from the flow approach conditions upstream from the sampler; (e) splashing;
measuring device used in conjunction with the sampler, the and (f) velocity and direction of the jet.
liquid sampling ratio is readily determined from timed volu- 16.6.2 The width of the diverging slot at any radius (r) is
metric or weight measurements of discharge from the slot approximately the following:

17
 D4411 − 03 (2014)´1
2 πr q/Q sampler had been absent. Many samplers have a sampling
Ws 5
111.6 T efficiency that varies from one particle size to another. Except
for slot or pit samplers, which ordinarily have efficiencies close
where:
to 100 % for all conditions, the overall sampling efficiency of
T = thickness of metal at the slot edge (in.), and a specific sampler is not constant but varies with the size
q/Q = desired proportion of stream discharge to be ex- distribution of the bedload particles, the flow conditions, the
tracted; typically 0.01, 0.005, and 0.0033 for the 1, 2, rate of bedload transport, and the degree of filling of the
and 3 ft samplers, respectively. sampler. The sampling efficiency is influenced, but is not solely
16.6.3 At high stream discharges the liquid sampling ratio determined by the sampler’s hydraulic efficiency, which is the
tends to decrease rapidly with an increase in the stream ratio of the volume of water that passed through the sampler
discharge; therefore, the sampler should not be used if the entrance in a unit of time to the volume of water that would
discharge exceeds 80 % of flume capacity (23). have passed through this entrance area if the sampler had been
absent. Hydraulic efficiency and sampling efficiency can only
16.7 Bedload Samplers: be determined accurately through comprehensive and detailed
16.7.1 A bedload sampler must be calibrated to determine procedures.
its sampling efficiency, which is defined as the ratio of the mass
of bedload collected during any single sampling time to the 17. Keywords
mass of bedload that would have passed through the width 17.1 fluvial sediment; sampler calibration; samplers; sedi-
occupied by the sampler entrance during the same time if the ment discharge; sediment transport

REFERENCES

(1) Guy, H. P., and Norman, W. W., “Field Methods for Measurement of (10) American Society of Civil Engineers, Sedimentation Engineering,
Fluvial Sediments,” Techniques of Water-Resources Investigations of by Task Committee, V. A. Vanoni, ed., ASCE, New York, NY, 1975,
the United States Geological Survey, Book 3, Chapter C2, 1970, pp. p. 342.
23–24. (11) Barnes, L. L., and Frevert, R. K., “A Runoff Sampler for Large
(2) Inter-Agency Committee on Water Resources, “Determination of Watersheds, Part 1, Laboratory Investigations,” Agricultural
Fluvial Sediment Discharge,” Rept. 14, A Study of Methods Used in Engineering, Vol 35, No. 2, 1954, pp. 84–90.
Measurement and Analysis of Sediment Loads in Streams; Subcom- (12) Dendy, F. E., “Traversing Slot Runoff Sampler for Small
mittee on Sedimentation, Minneapolis, MN, 1963, p. 28. Watershed,” ARS-S-15, U.S. Dept. of Agriculture, Southern Region,
(3) Bennet, J. P., and Nordin, C. F., Jr., “Suspended Sediment Sampling 1973.
Variability,” Chap. 17 in Environmental Impact on Rivers (River
(13) Parsons, D. A., “Coshocton-Type Runoff Samplers—Laboratory
Mechanics III), ed. and pub. by H. W. Shen, Fort Collins, CO, 1973.
Investigations,” TP-124 , U.S. Dept. of Agriculture, Soil Conserva-
(4) Ehrenberger, R.,“ Direkte Geschiebemessungen an der Donau bei
tion Service, 1954, Sheet 1.
Wein und deren bisherige Ergebnisse” (Direct Bedload Measurements
on the Danube at Vienna and Their Results to Date):Vienna, Die (14) Dendy, F. E., Allen, P. B., and Piest, R. F., Chapter 4, Sedimentation,
Wasserwirtschaft, Issue 34, 1931, pp. 1–9. Translation No. 39–20, in “Field Manual for Research in Agricultural Hydrology,” Agricul-
U.S. Waterways Experiment Station, Vicksburg, MS. tural Handbook No. 224; Science and Education Administration,
(5) Einstein, A. H., 1937, “Die Eichung des im Rhein verwendeten U.S. Dept. of Agriculture, Beltsville, IN 20705.
Geschiebefangers” (Calibrating the Bedload Trap as Used in the (15) ASCE, ibid, p. 338.
(Rhine): Schweizer. Bauzeitung, Vol 110, No. 12. Translation, Soil (16) Inter-Agency Committee on Water Resources, Report No. 6, ibid, p.
Conservation Service, California Institute of Technology, Pasadena, 28.
CA. (17) Skinner, J. V., and Beverage, J. P., 1976, ibid, Fig. 2.
(6) Federal Interagency Work Group III 1978, Chapter 3—Sediment, in (18) Skinner, J. V., and Beverage, J. P., 1976, ibid, Fig. 3.
National Handbook of Recommended Methods for Water Data (19) American Society of Civil Engineers, Sedimentation Engineering, by
Acquisition, Office of Water Data Coordination, Geological Survey, Task Committee, V. A. Vanoni, ed., ASCE, New York, NY, 1975, p.
U.S. Dept. of Interior, Reston, VA, pp. 3–20. 339.
(7) Skinner, J. V., and Beverage, J. P., 1976, Instructions for Sampling (20) Carter, C. E., and Parsons, D. A., “Field Test on the Coshocton-Type
with Depth-Integrating, Suspended-Sediment Samplers D-74, Wheel Runoff Sampler,” ASAE Transactions, 1967, Vol 10, No. 1,
D-74AL. D-74-TM, and D-74AL-TM, Federal Inter-Agency Sedi- pp. 133–135.
mentation Project, St. Anthony Falls Hydraulic Laboratory, (21) Beverage, Joseph P., and Futrell II, James C., “Comparison of Flume
Minneapolis, MN, Fig. 1. and Towing Methods for Verifying the Calibration of a Suspended-
(8) Inter-Agency Committee on Water Resources, “The Design of Im-
Sediment Sampler,” Water Resources Investigations Report 86-4193
proved Types of Suspended Sediment Samplers,” Rept 6, A Study of
of the U.S. Geological Survey, 1986.
Methods Used in Measurement and Analysis of Sediment Loads in
(22) Barnes, K. K., and Johnson, H. P. “A Runoff Sampler for Large
Streams; Subcommittee on Sedimentation, Minneapolis, MN, 1952,
pp. 60–63. Watersheds, Part II. Design of Field Installation,” Agricultural
(9) Hubbell, D. W., “Apparatus and Techniques for Measuring Bedload,” Engineering, Vol 37, No. 12, pp. 813–815, 824.
U.S. Geological Survey Water-Supply Paper 1748, Washington, DC, (23) Dendy, F. E., Allen, P. B., and Piest, R. F., ibid, pp. 248–50, 378–87.
1964.

18
 D4411 − 03 (2014)´1
ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; http://www.copyright.com/

19