Content uploaded by John R Gray
Author content
All content in this area was uploaded by John R Gray on Nov 20, 2018
Content may be subject to copyright.
Provided for non-commercial research and educational use.
Not for reproduction, distribution or commercial use.
This article was originally published in Comprehensive Water Quality and Purication published by Elsevier, and the
copy attached is provided by Elsevier for the author’s benet and for the benet of the author’s institution, for non-
commercial research and educational use including without limitation use in instruction at your institution, sending it
to specic colleagues who you know, and providing a copy to your institution’s administrator.
All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing
copies or access, or posting on open internet sites, your personal or institution’s website or repository, are prohibited.
For exceptions, permission may be sought for such use through Elsevier’s permissions site at:
http://www.elsevier.com/locate/permissionusematerial
Gray J.R., and Landers M.N. (2014) Measuring Suspended Sediment. In: Ahuja S. (ed.) Comprehensive Water Quality
and Purication, vol. 1, pp. 157-204. United States of America: Elsevier.
© 2014 Elsevier Inc. All rights reserved.
1.10 Measuring Suspended Sediment
JR Gray and MN Landers, US Geological Survey, Reston, VA, USA
Published by Elsevier Inc.
1.10.1 Introduction 159
1.10.1.1 Terminology 159
1.10.1.2 Criteria for a Sediment Dataset 160
1.10.1.3 Insights on Errors Associated with Measured SSCs and Loads 160
1.10.2 Traditional Suspended Sediment Measurement Techniques 161
1.10.2.1 History of Development of Traditional Sediment Sampling Equipment 161
1.10.2.2 Samplers and Sampling Methods 162
1.10.2.2.1 Manually operated samplers 162
1.10.2.2.1.1 Instantaneous samplers 162
1.10.2.2.1.2 Isokinetic samplers 162
1.10.2.2.1.2.1 Rigid-bottle samplers 165
1.10.2.2.1.2.2 Handheld and handline samplers 166
1.10.2.2.1.2.3 Cable-and-reel samplers 167
1.10.2.2.1.2.4 Bag samplers 167
1.10.2.2.2 Manual sampling methods 168
1.10.2.2.2.1 Single-vertical sampling 169
1.10.2.2.2.1.1 First case 169
1.10.2.2.2.1.2 Second case 169
1.10.2.2.2.1.3 Third case 170
1.10.2.2.2.1.4 Fourth case 170
1.10.2.2.2.2 Multivertical sampling 171
1.10.2.2.2.2.1 EDI method 171
1.10.2.2.2.2.2 EWI method 172
1.10.2.2.2.2.3 Number of cross sections required to quantify a suspended sediment discharge 173
1.10.2.2.2.2.4 Advantages of the EDI and EWI methods 173
1.10.2.2.2.2.5 Transit rates for suspended sediment sampling 174
1.10.2.2.2.2.6 Inspecting samples 174
1.10.2.2.2.3 Point-integrated sampling 175
1.10.2.2.3 The case for depth-integrated sampling with isokinetic samplers 175
1.10.2.2.4 Automatic samplers 177
1.10.2.2.4.1 Automatic pumping samplers 177
1.10.2.2.4.1.1 Installation and use criteria 178
1.10.2.2.4.1.2 Placement and orientation of sampler intake 178
1.10.2.2.4.1.3 Activation 179
1.10.2.2.4.2 Single-stage samplers 179
1.10.2.2.5 Sediment subsampling equipment 181
1.10.3 Surrogate Suspended Sediment Measuring Techniques 182
1.10.3.1 Overview of Selected Suspended Sediment Surrogate Measurement Techniques, Metrics, and
Requirements 182
1.10.3.1.1 Calibration of suspended-surrogate metrics to representative SSC 183
1.10.3.1.2 Acceptance criteria for SSC and PSD data produced by suspended sediment-surrogate technologies 183
1.10.3.2 Technological advances in suspended sediment-surrogate monitoring 184
1.10.3.2.1 Turbidity 184
1.10.3.2.1.1 Background and theory 184
1.10.3.2.1.2 Example field evaluation 185
1.10.3.2.1.3 Advantages and limitations of turbidity 187
1.10.3.2.2 Laser diffraction 187
1.10.3.2.2.1 Background and theory 187
1.10.3.2.2.2 Example field evaluation 188
1.10.3.2.2.3 Advantages and limitations of laser-optic technology 188
1.10.3.2.3 Pressure difference 189
1.10.3.2.3.1 Background and theory 189
1.10.3.2.3.2 Example field evaluation 189
1.10.3.2.3.3 Advantages and limitations of the pressure-difference technology 192
Comprehensive Water Quality and Purification, Volume 1 doi:10.1016/B978-0-12-382182-9.00012-8 157
Author's personal copy
1.10.3.2.4 Acoustic surrogates 192
1.10.3.2.4.1 Acoustic backscatter 192
1.10.3.2.4.2 Acoustic attenuation 193
1.10.3.2.4.3 Acoustic surrogate methods for SSC 193
1.10.3.2.4.4 Multifrequency acoustic surrogates for sediment size 194
1.10.3.2.4.5 Examples of deployments 194
1.10.3.2.4.6 Advantages and limitations of acoustic surrogates 195
1.10.4 Summary and Conclusions 196
1.10.4.1 Traditional Technologies 196
1.10.4.2 Surrogate Technologies 196
References 199
Glossary
Bedload Material moving on or near the stream bed by
rolling, sliding, or skipping.
Bed material The sediment mixture of which the stream
bed is composed.
Composite sample A sample formed by combining two
or more individual samples or representative portions of
the samples.
Concentration of sediment (mass) The ratio of the mass
of dry sediment in a water–sediment mixture to the mass of
the mixture. Also commonly referred to as ‘‘suspended-
sediment concentration.’’
Concentration of sediment (volume) The ratio of the
volume of dry sediment to the volume of the
water–sediment mixture.
Depth-integrated sediment sampler A device that collects
a representative water–sediment mixture at all points along
the sampling vertical.
Diameter, sedimentation The diameter of a hypothetical
sphere of the same specific gravity and the same settling
velocity as the given particle in the same fluid.
Equal-discharge-increment method A procedure for
obtaining the discharge weighted suspended-sediment
concentration of flow at a cross section whereby depth
integration is performed at the centers of three or more
equal flow segments of the cross section.
Equal-width-increment method A procedure of
obtaining the discharge weighted suspended-sediment
concentration of flow at a cross section by performing depth
integration at a series of verticals equally spaced across the
cross section and using the same vertical transit rate at all
sampling verticals.
Fine-material load The part of the total sediment load
that is composed of particles of a finer size than the particles
present in appreciable quantities in the bed; normally, the
fine-material load consists of material finer than 0.062-mm
median diameter.
Fluvial sediment Particles derived from rocks, biological
materials, or chemical precipitants, that are transport by,
suspended in, or deposited by flowing water.
Gaging station A particular site on a stream, canal, lake,
or reservoir at which systematic observations of hydrologic
data are obtained.
Instantaneous sampler A suspended-sediment sampler
that takes a representative specimen of the water–sediment
mixture in a stream at a desired depth and a moment of
time.
Isokinetic sampling To sample in such a way that the
water–sediment mixture moves with no change in velocity
as it leaves the ambient flow and enters the sample intake.
Nephelometer An instrument that measures the amount
of light scattered in a suspension.
Particle size A linear dimension, usually designated as
diameter, used to characterize the size of a particle; the
dimension may be determined by any of several different
techniques, including sedimentation, sieving, micrometric
measurement, or direct measurement.
Point integration A method of sampling at a relatively
fixed point whereby the water–sediment mixture is
withdrawn isokinetically for a specified period of time.
Point-integrating sampler An instrument capable of
collecting a water–sediment mixture isokinetically for a
specified period of time by opening and closing under
water; an instrument suitable for performing point
integration.
Point sample Sample of water–sediment mixture taken at
a single point, either with an instantaneous or a point-
integrating sampler.
Pumping sampler A device that draws the
water–sediment mixture through a pipe or hose, the intake
of which is placed at the desired sampling point in a stream.
Sample vertical An approximately vertical path from the
water surface to the bottom along which one or more
samples are collected to define various properties of the
flow, such as sediment concentration.
Sampled zone That part of a channel transect presumed
to be wholly represented by sediment samples.
Sediment discharge A mass or volume of sediment
passing a steam cross section in a unit time (the term maybe
qualified as suspended-sediment discharge, bedload
discharge, or total-sediment discharge).
Sediment load A general term that refers to material in
suspension or in transport, or both; it is not synonymous
with either discharge or concentration.
Sediment particle A fragment of mineral or organic
material in either a singular or aggregate state.
158 Measuring Suspended Sediment
Author's personal copy
Sediment sample A quantity of water–sediment mixture
deposited sediment that is collected to represent one or
more properties of the sampled medium.
Silt size Particles with median diameters of
0.004–0.062 mm.
Specific gravity Ratio of the mass of any volume of a
substance to the mass of an equal volume of water at 4 1C.
Split sample A single sample separated into two or more
individual parts in a manner such that the content of each
part is representative of the original sample.
Stream discharge The quantity of flow passing through a
cross section in a unit of time.
Suspended sediment Sediment that is carried in
suspension by the turbulent components of the fluid or by
Brownian movement.
Suspended-sediment load A part of the sediment load
which is suspended sediment.
Suspended-sediment sampler A device that collects a
representative portion of the water with its suspended-
sediment load.
Total-sediment discharge The total quantity of sediment
passing a section per unit time.
Total-sediment load All of the sediment in transport; that
part moving as suspended load plus that moving as
bedload.
Turbidity An expression of the optical properties of a
sample which causes light rays to be scattered and absorbed
rather than transmitted in straight lines through a sample.
Wash load The portion of the stream sediment load
composed of particles, usually finer than 0.062 mm in
diameter, which are found only in relatively small
quantities in the bed.
Water discharge See ‘stream discharge’.
1.10.1 Introduction
This chapter presents techniques for measuring suspended
sediment discharges in rivers using traditional samplers and
techniques developed by the Federal Interagency Sedimen-
tation Project (FISP; Skinner, 1989;Glysson and Gray, 1997;
Federal Interagency Sedimentation Project, 2012) and United
States Geological Survey (USGS); and surrogate technologies
being used or tested by the FISP, USGS, and others (Wren
et al., 2000;Gray and Glysson, 2004;Gray and Gartner, 2009).
Sediment and ancillary data are fundamental requirements for
the efficient and proper management of river systems, in-
cluding the design of structures; determining aspects of stream
behavior; ascertaining the probable ramifications of removing
an existing structure; estimating bulk erosion, transport, and
sediment delivery from watersheds and to the oceans; ascer-
taining the long-term usefulness of reservoirs and other public
works; tracking movement of solid-phase contaminants; eco-
logical assessments; restoring degraded or otherwise modified
streams; and as ground-truth data for the calibration and
validation of numerical models. This chapter on ‘‘Measuring
Suspended Sediment’’ has a bearing on several other contri-
butions to this volume, including Chapters 2.13,2.16,4.1,
4.2,4.5,4.9,4.15, and 4.18.
The traditional FISP samplers and techniques described
herein (Davis, 2005;Edwards and Glysson,1999;Nolan et al.,
2005;Gray et al., 2008), which debuted in the mid-twentieth
century, are grounded on sound physical and statistical prin-
ciples and form the basis for production of most fluvial sus-
pended sediment data in the US and in many other countries
on every continent other than Antarctica. More information
on a broader suite of traditional technologies can be found by
perusing the following publications: Manual Sediment
Transport Measurements in Rivers, Estuaries, and Coastal Seas
(van Rijn, 2007) and Methods for Measurement of Suspended
Sediment (ISO, 2002).
Some of the surrogate technologies described herein are
being incorporated into operational programs and show
considerable promise toward providing the temporally dense,
and in some cases spatially robust fluvial sediment data needed
to increase and bring more consistency to sediment-discharge
measurements worldwide. Even as in situ and manually de-
ployed surrogate technologies become operationally ubiqui-
tous, they will continue to require empirical calibrations with
traditional technologies.
1.10.1.1 Terminology
Fluvial sediment can be defined by its origin or operationally
by its method of collection (Figure 1). The total amount of
sediment in transport can be described by its origin as being
composed of bed material load plus wash load. Bed material
load is that part of the total load that is composed of particle
sizes present in appreciable quantities in the shifting parts of
the streambed. Wash load is that part of the total load com-
posed of particles, usually finer than 0.062 mm in diameter,
that originate from the watershed and are found, if at all, only
in relatively small quantities in the streambed (ASTM Inter-
national, 1999).
The operational definition of sediment in transport is in
part a function of the instruments and methods used to obtain
the data. Suspended sediment and bedload discharges are the
quantities of suspended sediment and bedload passing
through a stream cross section per unit time, respectively.
Suspended sediment discharge can include some of the bed
material load component but all of the wash load component.
Bedload discharge includes some of the bed material load
component. The addition of sediment discharges derived from
data collected with physical suspended sediment and bedload
samplers may not equal the sum of bed material load plus
wash load (Figure 1). This is a result of one or more factors
associated with the range in size of sediments in transport, and
the characteristics and deployment methods of the suspended
sediment and bedload samplers.
Sediment-surrogate technologies are defined as instru-
ments coupled with operational and analytical methodologies
that enable acquisition of temporally and (or) spatially dense
fluvial sediment datasets without the need for routine
Measuring Suspended Sediment 159
Author's personal copy
collection and analysis of physical samples other than for
calibration purposes. Although the traditional technologies
described herein, with the exception of automatic samplers,
are manually deployed, most of the surrogate technologies
described can be deployed manually or in situ. The latter are
capable of providing a continuous and dense time series of
data to be empirically converted to the metric of interest.
1.10.1.2 Criteria for a Sediment Dataset
Collection of data to enable reliable sediment transport esti-
mates is often difficult, time consuming, and expensive. It is
frustrating to obtain data for a location and set of conditions
of interest, only to subsequently discover that the collected
data were incomplete (Glysson, 1989a) or were inappropriate
for the requested laboratory analysis. The types of data re-
quired depend on the goals of the assessment and intended
storage medium for the data. For example, suspended sedi-
ment concentration (SSC) and water discharge data are re-
quired to compute continuous records of suspended sediment
discharge (Porterfield, 1972;Koltun et al., 2006). Other rele-
vant data include particle-size distributions (PSDs) of
suspended sediment and bottom material. The integrity of
large-scale, long-term monitoring programs such as the Vigil
Network (Osterkamp and Emmett, 1992), or that proposed
for North America (Osterkamp et al., 1998,2004), the US
(Osterkamp and Parker, 1991;Gray and Shadie, 2010), and
Canada (Day, 1991), is particularly dependent on the re-
liability and comparability of the data collected.
The most reliable databases accept only selected data types
representing sediment and ancillary variables obtained using a
consistent set of protocols. For example, sediment data stored
by the USGS as part of the online National Water Information
System (U.S. Geological Survey, 2012a) and other databases
(Turcios et al., 2000;U.S. Geological Survey, 2012b) are col-
lected by techniques described by the U.S. Geological Survey
(1998a),Edwards and Glysson (1999),Nolan et al. (2005),
and Gray et al. (2008); analyzed in a USGS-approved la-
boratory by techniques described by Guy (1969),Matthes
et al. (1992),Knott et al. (1992,1993), and U.S. Geological
Survey (1999) with laboratory quality assurance verified by
U.S. Geological Survey (1998b).
Glysson (1989a) divided dataset requirements for com-
puting sediment transport using the more common sediment
transport equations for noncohesive sediments into three
categories: sediment, hydraulic, and others. Table 1 summar-
izes and expands on those requirements. Parameter codes are
available for storing most of these metrics in the USGS
National Water Information System (NWIS) database (U.S.
Geological Survey, 2012b).
1.10.1.3 Insights on Errors Associated with Measured
SSCs and Loads
The ‘best’ (considered to be the most reliable and accurate)
data describing SSCs and suspended sediment discharges are
from analyses of physical samples collected using appropriate
samplers and deployment methodologies with adequate
•temporal resolution for the ambient sediment-hydrologic
conditions; and
•spatial resolution throughout the stream cross section.
Errors associated with measured load result from errors in
quantifying stream discharge, sampling and analytical errors
in the sediment samples, and errors in the calibration of fixed
location samples based on flow-weighted cross-section sam-
ples. Water discharge errors at the 95% confidence limit are
likely to be less than 75–8% at many riverine systems using
these techniques (Kennedy, 1984;Sauer and Meyer, 1992).
Laboratory analytical errors from gravimetric analysis of SSC
are larger for very low concentrations and decrease rapidly
with increasing SSC. Sampling errors for SSC measurements
Total sediment load
By origin
Wash load
Bed-material load
Suspended load
Suspended load
Unsampled load1
Bedload
Bedload
By transport By sampling method
1That part of the sediment load that is not collected by the depth-integrating suspended-sediment
and pressure-difference bedload samplers used, depending on the type and size of the sampler(s).
Unsampled-load sediment can occur in one or more of the following categories: (a) sediment that passes
under the nozzle of the suspended-sediment sampler when the sampler is touching the streambed and
no bedload sampler is used; (b) sediment small enough to pass through the bedload sampler's mesh
bag; (c) sediment in transport above, the bedload sampler that is too large to be sampled reliably by the
suspended-sediment sampler; and (d) material too lar
g
e to enter the bedload-sampler nozzle.
Figure 1 Components of total sediment load considered by origin, by transport, and by sampling method. Reproduced from Diplas, P., Kuhle,
R. A., Gray, J. R., Glysson, G. D. and Edwards, T. K. (2008). Sediment transport measurements. In Garcia, M. (ed.) Sedimentation engineering –
Processes, measurements, modeling, and practice. American Society of Civil Engineers Manual 110, ch. 5, pp. 307–353. Reston, Virginia: ASCE.
Available at http://water.usgs.gov/osw/techniques/Diplas_Kuhnle_others.pdf (accessed on 18 May 2012).
160 Measuring Suspended Sediment
Author's personal copy
vary with SSC, PSD, sampler equipment, training and experi-
ence of the field crew, deployment techniques, and sampling
conditions. However, analytical results from replicate field
samples at SSCs above approximately 20 mg l
1
typically have
differences estimated to be approximately 710 % (Edwards
and Glysson, 1999;Horowitz et al., 2001;Horowitz,
2008). Based on published protocols for deployment of
depth-integrated samplers (Edwards and Glysson, 1999;Top-
ping et al., 2011, p. 45) found the maximum total uncertainty
in sand SSCs to be approximately 77%. The accuracy of SSCs
and loads cannot be generalized because of many factors in-
cluding those mentioned here; however, for ‘good’ measure-
ment conditions and data quality, measured load may have an
expected accuracy of 715% (Gray and Gartner, 2009,2010b).
One commonly used analog for SSC data in the US is total
suspended solids (TSS; American Public Health Association,
American Water Works Association, and Water Pollution
Control Federation, 1995). TSS data are not comparable to
SSC data under some circumstances and are fundamentally
unreliable when used to analyze samples collected from
open-channel flows (Gray et al., 2000;U.S. Geological Survey,
2000;Glysson et al., 2000,2001). TSS data tend to:
1. underestimate suspended solid-phase SSCs by a pro-
portionate amount of 25–34% (Gray et al., 2000);
2. be least reliable under the very circumstances under which
most suspended sediment transport typically occurs – at
higher flows when percentages of suspended sand are
comparatively large.
Based on the information above, Gray et al. (2000) con-
cluded that the accuracy and comparability of suspended
solid-phase SSCs determined in the US would be greatly en-
hanced if all such data were produced by the SSC analytical
method (ASTM International, 1999). If TSS data are required
to meet study objectives, then concurrent SSC analyses should
be conducted and any bias associated with the TSS data
quantified.
1.10.2 Traditional Suspended Sediment
Measurement Techniques
The scope of this section includes a brief history of the de-
velopment of traditional suspended sediment samples; de-
scriptions of manual suspended sediment samplers and
methods for their deployment; description, installation, and
operation of automatic samplers; and a summary of equip-
ment used for obtaining water–sediment subsamples.
1.10.2.1 History of Development of Traditional Sediment
Sampling Equipment
The initial attempts to develop suspended sediment sampling
equipment and deployment techniques were made by in-
dependent investigators, with the earliest such available records
dated 1808–09 when Gorsse and Subuors sampled the Rhone
river, Arles, France (Federal Interagency Sedimentation Project,
Table 1 Sediment, hydraulic, and operational parameters for a sediment dataset
a
Sediment and related parameters include: Operational and related parameters include:
Suspended-sediment concentration Sample start and end times
Suspended-sediment discharge Sample method of collection
Bed load discharge Sampler type
Total sediment discharge Sampler nozzle diameter and composi tion
Suspended-sediment particle-size distribution Location of sample collection in cross section from left or right bank
Bedload particle-size distribution Width of sampled vertical
Bed-material particle-size distributions (in channel, overbank) Duration bedload sampler was deployed on bed
Turbidity Bedload sampler bag mesh size
Transparency Analyzing agency and laboratory
Mean specific gravity Duration suspended-sediment sampler collected water
Particle shape Sampler transit rate
Bed-roughness coefficient Number of sampling points in a vertical
Total suspended solids Number of verticals in composite sample
Specific conductance Depth to compute isokinetic transit rate
Dissolved solids concentrations Velocity to compute isokinetic transit rate
Hydraulic and related parameters
b
include: Sample volume to compute isokinetic transit rate
Water discharge Number of bedload samples in a composite
Watercourse stage Tether line used for sampling
Mean depth
Mean velocity
Water temperature
Surface width
Area
Hydraulic radius
Channel slope
a
The National Water Information System Parameter Code Definitions can be accessed at http://nwis.waterdata.usgs.gov/nwis/pmcodes/
b
A site description that may include a morphometric assessment based on one or more channel classification schemes also should be included.
Note: Types of data collected are dependent on their intended use. Parameters marked with an asterisk are considered fundamentally important data types.
Measuring Suspended Sediment 161
Author's personal copy
1940). Baumgarten’s samples collected in the River Garonne at
Marmande, France, from 1839 to 1846 resulted in what were
probably the first sediment discharge computations. Sediment-
discharge measurements in the US began in 1838 when Cap-
tain Talcott sampled the Mississippi river. The fluvial sediment
measurements made in the Rio Grande at Embudo, New
Mexico, beginning 15 January 1889, represent the beginning of
the USGS’s sediment program (Glysson, 1989b). Fluvial sedi-
ment measurements have been made regularly in the Rio
Grande since 1897, lower Colorado river since 1909, and upper
Colorado river basin since 1925. A detailed investigation of
sediment loads starting in 1942 as part of the Missouri River
Project included determination of the feasibility of storage
reservoirs on streams transporting heavy sediment loads. Be-
ginning in about 1930, extensive sediment surveys have been
made in many other streams of the US (Federal Interagency
Sedimentation Project, 1940;Nelson and Benedict, 1950;
Glysson, 1989b;Turcios et al., 2000).
After World War II, the number of sites at which the USGS
collected daily suspended sediment data increased rapidly,
peaking at approximately 364 in 1982 (Glysson, 1989b;
Osterkamp and Parker, 1991). During the period 2006–10,
an average of approximately 65 daily-record sediment sites were
being operated by the USGS (U.S. Geological Survey, 2012d).
Before the mid-1940s, sediment samplers in the US lacked
calibration and were deployed using widely different techni-
ques. Most instruments were designed with limited attention
to, or knowledge of, sediment transport concepts or the in-
fluence of the equipment on the local flow pattern (Glysson,
1989b). As a result, data obtained by different investigators
before the mid-1940s were neither comparable nor could their
accuracy be readily or reliably evaluated. It became apparent
that collection of reliable suspended sediment data requires an
understanding of the physics of sediment transport and of
how the sediment–water mixture is affected by sampler and
deployment characteristics.
In 1939, several US agencies organized an interagency
program to study methods and equipment used in measuring
the sediment discharge of streams and to improve and
standardize equipment and methods where practicable (Fed-
eral Interagency Sedimentation Project, 1941). The FISP was
created under the sponsorship of the Committee on Sedi-
mentation of the Federal Water Resources Council (the func-
tions of this council are now shared by two organizations: the
Technical Committee that oversees the FISP; and the Advisory
Committee on Water Information, Subcommittee on Sedi-
mentation, 2012). The subsequent comprehensive study of
sampling equipment included suspended sediment, bedload,
and bed material samplers, their means of deployment, and
sample analytical methods. As a result of research conducted
by the FISP and others, an integrated system of standardized,
quantifiably reliable sediment samplers, sampling, and ana-
lytical techniques has been developed and is widely used
around the world.
1.10.2.2 Samplers and Sampling Methods
The purpose of a suspended sediment sampler is to obtain a
representative sample of the water–sediment mixture moving
in the stream in the vicinity of the sampler intake. There are
two categories of traditional suspended sediment samplers:
manually operated samplers and automatic samplers. Manu-
ally operated samplers include instantaneous and isokinetic
samplers. Isokinetic samplers – those that collect a filament of
river water without a change in the flow velocity or direction –
include those with rigid sample bottles (bottle samplers) and
with flexible bags (bag samplers). Additional information on
manual sampler specifications, limitations, operation, and
current costs are available in the online FISP catalog available
through Federal Interagency Sedimentation Project (2012).
FISP suspended sediment samplers range in (2012) price from
$95 for a handheld US DH-81 to $7946 for a cable-deployed
US P-63.
1.10.2.2.1 Manually operated samplers
1.10.2.2.1.1 Instantaneous samplers
Instantaneous samplers are applicable for sampling flows that
fail to meet either of the following criteria for deployment
of an isokinetic sampler: sampling depths exceeding ap-
proximately 0.3 m, and (or) mean velocities exceeding ap-
proximately 0.5 m s
1
. At shallow depths, the part of the
stream from the streambed to the nozzle of an isokinetic
sampler deployed at maximum depth, referred to as the
unsampled zone, becomes unacceptably large with respect to
the total depth. At small velocities, only silt- and clay-size
material typically is in suspension, and these finer size frac-
tions tend to be fairly uniformly distributed with depth
(Colby, 1963;Guy, 1970). Under these circumstances, an in-
stantaneous sample from the water column may provide a
reasonably accurate estimate of the SSC at the sampled point
or in the sampled vertical. Instantaneous samplers may also be
deployed at flow velocities too swift to submerge an isokinetic
sampler, or when the presence of debris renders normal
sample collection dangerous or impossible.
Although nonisokinetic samplers may provide acceptable
results under certain sediment-transport conditions, such as
when fine material constitutes most or all of the sediment
load, conditions for which nonisokinetic sampling are ap-
propriate may not be apparent at the time of sample col-
lection. The most reliable suspended sediment samples are
obtained using isokinetic samplers.
The simplest instantaneous sampler is an open bottle used
to obtain a surface, or dip, sample. The weighted-bottle, hand-
held (WBH)-96-weighted bottle sampler (Federal Interagency
Sedimentation Project, 2012) can be deployed with a hand
line in still or slow-moving water. The Van Dorn and Kem-
merer samplers are thief-type samplers that are typically used
for still-water sampling such as in lakes and reservoirs, but
they also may be useful in sluggish streamflows U.S. Geo-
logical Survey, 2006.
1.10.2.2.1.2 Isokinetic samplers
Isokinetic samplers are designed to collect a representative
velocity-weighted sample of the water–sediment mixture.
Water approaching the nozzle of an isokinetic sampler
undergoes essentially no change in speed or direction as it
enters the nozzle orifice (Figure 2). When deployed using
prescribed methods at preselected, strategic locations in a
cross section, an isokinetic sampler integrates a sample
162 Measuring Suspended Sediment
Author's personal copy
proportionally by velocity and area, resulting in a water dis-
charge-weighted sample. A water discharge-weighted sample
contains an SSC and a PSD representative of the material in
transport in the sampled cross section at the time the sample
was collected.
A list of isokinetic samplers available from the FISP is
shown in Table 2. A sampler functions isokinetically when the
relative sampling rate – the dimensionless ratio of the velocity
through the nozzle divided by the approaching stream velocity
– is equal to one. In practice, FISP isokinetic samplers are
designed to ensure that the water velocity in the nozzle is
within 10% of the ambient stream velocity throughout a
sampler’s operating velocity range (Broderick Davis, FISP,
2001, written communication).
SSC errors in samples collected with isokinetic-type
samplers may stem from a combination of the size of the
suspended material and the relative sampling rate (degree of
departure from isokinetic sample collection). The relation
between percent error in SSC and relative sampling rate for
sediments with a density of 2.65 and median diameters of
0.45, 0.15, 0.06, and 0.01 mm in flows of 1.5 m s
1
is shown
in Figure 3. Under these test conditions, relative sampling
rates for 0.45 mm sediments may range from 0.75 to 1.3
without introducing more than about a 10% error in sample
SSC values. Conversely, at relative sampling rates less than
0.25, resultant SSC errors for 0.45 mm sediments may exceed
100%. The range in errors tends to decrease with decreasing
sediment size. For example, 0.01-mm size sediments have less
than a 5% error for relative sampling rates ranging from
approximately 0.2 to almost 5 (Figure 3). In each case,
subunity relative sampling rates result in positive SSC bias;
larger relative sampling rates result in zero or negative
SSC bias.
The FISP’s suite of depth- and point-integrating samplers
(Davis, 2005;Table 2) are isokinetic samplers. A depth-inte-
grating sampler is designed to isokinetically and continuously
accumulate a representative sample from a stream vertical
while transiting the vertical at a uniform rate (Federal
Direction of flow
(a)
(b)
(c)
Sediment
particles
Cs
Cv
VnIntake nozzle
Isokinetic sampling
When v = Vn
Then C = Cs
Cs
Cv
Vn
Nonisokinetic sampling
When v < Vn
Then C > Cs
Cs
Cv
Vn
Nonisokinetic sampling
When v > Vn
Then C < Cs
Figure 2 Relation between intake velocity and sample SSC for (a) isokinetic and ((b) and (c)) nonisokinetic sample collection of particles larger
than 0.062 mm, where v is the mean stream velocity, V
n
and Cis the flow velocity and SSC incident on the sampler nozzle, respectively, and C
s
is the sample SSC. Reproduced from Edwards, T. E. and Glysson, G. D. (1999). Field methods for measurement of fluvial sediment. U.S.
Geological Survey Techniques of Water-Resources Investigations, Book 3, ch. C2, 89 pp. Available at http://water.usgs.gov/osw/techniques/
sedimentpubs.html (accessed on 12 June 2012).
Measuring Suspended Sediment 163
Author's personal copy
Table 2 Designations and characteristics for Federal Interagency Sedimentation Project (FISP) manually operated isokinetic samplers (Davis, 2005)
Sampler
designation
a
Nozzle inner
diameter, cm (in.)
Container type and
capacity
Mode of
suspension
Maximum depth,
m (ft)
Minimum isokinetic
velocity, m s
1
(ft s
1
)
Maximum recommended
velocity
b
,ms
1
(ft s
1
)
Unsampled zone,
cm (in.)
Mass, kg
(weight, lbs)
US DH-48 0.48 (3/16)
c
, 0.64
(1/4)
Rigid bottle 0.47 l (pint) Wading rod 2.7 (9) 0.5 (1.5) 2.7 (8.9) 8.9 (3.5) 2 (4)
US DH-59 0.48 (3/16) Handline or
cable reel
4.6 (15) 1.5 (5.0) 11 (4.5) 10 (22)
US DH-59 0.64 (1/4) 2.7 (9)
US DH-76
d
0.48 (3/16), 0.64
(1/4)
Rigid bottle 0.95 l (quart) 4.6 (15) 2.0 (6.6) 8.1 (3.2) 11 (25)
US DH-81 0.48 (3/16) Rigid bottle1 l Wading Rod 2.7 (9) 0.6 (2.0) 1.9 (6.2) 10 (4.0) 0.5 (1)
US DH-81 0.64 (1/4) 2.3 (7.6)
US DH-81 0.79 (5/16) 2.1 (7.0)
US DH-95 0.48 (3/16) Handline or
cable reel
4.6 (15) 0.6 (2.1) 1.9 (6.2) 12 (4.8) 13 (29)
US DH-95 0.64 (1/4) 0.5 (1.7) 2.1 (7.0)
US DH-95 0.79 (5/16) 0.6 (2.1) 2.3 (7.4)
US DH-2 0.48 (3/16) Flexible 1-l bag 11 (35) 0.6 (2.0) 1.8 (6.0) 8.9 (3.5) 14 (30)
US DH-2 0.64 (1/4) 6.1 (20)
US DH-2 0.79 (5/16) 4.0 (13)
US D-74 0.48 (3/16) Rigid bottle 0.47 l (pint)
or 0.95 l (quart)
Cable reel
4.6 (15) 0.5 (1.5) 2.0 (6.6) 10 (4.1) 28 (62)
US D-74 0.64 (1/4) 2.7 (9) pint
4.6 (15) quart
US D-74AL 0.48 (3/16) 4.6 (15) 1.8 (5.9) 19 (42)
US D-74AL 0.64 (1/4) 2.7 (9) pint
4.6 (15), quart
US D-95 0.48 (3/16) Rigid bottle 1 l 4.6 (15) 0.5 (1.7) 1.9 (6.2) 12 (4.8) 29 (64)
US D-95 0.64 (1/4) 0.6 (2.0) 2.0 (6.7)
US D-95 0.79 (5/16)
US D-96 0.48 (3/16) Flexible 3-l bag 34 (110) 0.9 (3.0) 3.8 (12.5) 10 (4.0) 60 (132)
US D-96 0.64 (1/4) 18 (60)
US D-96 0.79 (5/16) 12 (39)
US D-96-A1 0.48 (3/16) 34 (110) 1.8 (6.0) 36 (80)
US D-96-A1 0.64 (1/4) 18 (60)
US D-96-A1 0.79 (5/16) 12 (39)
US D-99 0.48 (3/16) Flexible 6-l bag 67 (220) 1.1 (3.5)
US D-99 0.64 (1/4) Flexible 6-
d
or 3 l
(6.3- or 3.2-quart) bag
e
4.6 (15.0) 24 (9.5) 125 (275)
US D-99 0.79 (5/16)
37 (120) 1.1 (3.5)
c
or 0.6 (2.0)
d
24 (78)
US P-61-A1 0.48 (3/16) Rigid bottle 0.47 l (pint)
or 0.95 l (quart)
55 (180), pint 0.5 (1.5) 3.0 (10.0) 11 (4.3) 48 (105)
37 (120), quart
US P-63 0.48 (3/16) 4.6 (15.0) 15 (5.9) 91 (200)
US P-72 0.48 (3/16) 22 (72), pint 1.6 (5.3) 11 (4.3) 19 (41)
16 (51), quart
US P-6 0.48 (3/16) Rigid bottle 0.95 l (quart)
f
49 (160), quart 0.5 (1.5) 4.0 (13) 8.9 (3.5) B45 (B100)
a
Samplers designated in italics may also be used for water quality sampling as described in the US Geological Survey National Field Manual for the Collection of Water Quality Data (variously dated).
b
For rigid-bottle samplers, the maximum recommended velocity for sampler deployment is based either on measured isokinetic limitations, or, for prototypes of samplers tested at Anthony Falls Hydraulic Laboratory flume, on the maximum
velocities used in tests. Bag samplers were determined to retain isokinetic characteristics at the highest velocities tested. Their maximum recommended velocity was selected to correspond with the velocity at which the angle of the suspension
cable was drawn back just shy of ‘excessive’ by testing personnel – from 251to 301– and on safety considerations.
c
The 0.48-mm (3/16-in.) internal diameter nozzle is designated for use in high-velocity flows.
d
A minimum isokinetic velocity of 1.1 m s
1
(3.5 ft s
1
) applies to the D-99 sampler using a 6-liter flexible bag and a 0.48-mm (3/16-in.) internal diameter nozzle.
e
A minimum isokinetic velocity of 0.61 m s
1
(2 ft s
1
) applies to the D-99 sampler using a 3-l flexible bag and 0.64-mm (1/4-in.) or 0.79-mm (5/16-in.) internal diameter nozzle.
f
The sampler is designed to use various sizes of bottles as sample containers. The 1-ll HDPE bottle, Item # 621-0032 sold by Dynalab Corp., has been found to work well with the sampler (FISP, 2011, written communication).
164 Measuring Suspended Sediment
Author's personal copy
Interagency Sedimentation Project, 1952). A depth-integrating
sampler collects and accumulates a velocity- or water dis-
charge-weighted sample as it descends and ascends at a con-
stant rate through the sampling vertical provided that the
appropriate transit rate is not exceeded and the sample con-
tainer is not overfilled.
Point-integrating samplers are more versatile than the
depth-integrating types. Point-integrating samplers are de-
signed for depth integration of streams too deep (or too swift)
to be sampled in a continuous round-trip integration. They are
equipped with an electrically activated valve that enables the
operator to isokinetically sample points in, parts of, or the
entire vertical. For stream cross sections less than 9 m deep,
the full depth can be traversed in one direction at a time by
opening the valve and depth integrating either from surface to
bottom or vice versa. Stream cross sections deeper than 9 m
can be integrated in segments of 9 m or less by collecting
integrated sample pairs consisting of one or more downward
integration(s) and a corresponding upward integration(s)
in separate containers (Federal Interagency Sedimentation
Project, 1963).
1.10.2.2.1.2.1 Rigid-bottle samplers When a rigid-bottle
suspended sediment sampler is submerged with the nozzle
pointing directly into flow within the sampler’s calibrated
velocity range, a part of the streamflow enters the sampler
container via the nozzle and air in the container exhausts
under the combined effect of three forces:
1. A positive dynamic head at the nozzle entrance due to the
flow.
2. A negative head at the outlet of the air exhaust tube due to
flow separation.
3. A positive pressure due to difference in elevation between
the nozzle entrance and the air exhaust tube outlet.
Under these conditions, a calibrated isokinetic sampler will
collect a sample with an SSC and a PSD essentially unchanged
from those at the sampling point in the stream and a represen-
tative sample will result. However, when the sample in the
container reaches the level of the air exhaust, the intake flow rate
drops and circulation of the streamflow into the nozzle and out
oftheairexhausttubeoccurs.Becausethevelocityofthewater
flowing through the bottle is less than the stream velocity, some
sediment – particularly coarser particles – tend to settle in the
sample bottle resulting in a large-biased SSC value. Additionally,
the resulting subefficient sampling rate may favor positive SSC
bias (Figure 3). Substantial errors in an SSC and a PSD can result
from samples collected using an incorrect or uncontrolled sam-
ple rate. The magnitude of errors tends to increase concomitantly
with increases in the percentage and size of suspended sand-size
material (Federal Interagency Sedimentation Project, 1941;
Figure 3). Edwards and Glysson (1999) and the U.S. Geological
140
Standard nozzle
Stream velocity = 1.5 m s−1
0.45 mm Sediment
0.15 mm Sediment
0.06 mm Sediment
0.01 mm Sediment
120
100
80
60
40
20
−20
−40
−50
0.15 0.2 0.2 0.4 0.6 1.0 1.5 2.0 3.0 4.0 5.0
Relative sampling rate = Mean intake velocity
Mean stream velocit
y
0
Error in concentration (%)
Figure 3 Effect of sampling rates on measured SSCs for four sediment-size distributions. Reproduced from Gray, J. R., Glysson, G. D. and
Edwards, T. E. (2008). Suspended-sediment samplers and sampling methods. In Garcia, M. (ed.) Sediment transport measurements.
Sedimentation engineering – Processes, measurements, modeling, and practice. American Society of Civil Engineers Manual 110, ch. 5.3,
pp. 320–339. Available at http://water.usgs.gov/osw/techniques/Diplas_Kuhnle_others.pdf (accessed on 18 May 2012) and adapted from Federal
Interagency Sedimentation Project (1941). Laboratory investigation of suspended-sediment samplers. Interagency Report No. 5, Hydraulics
Laboratory, Iowa University, Iowa City, 99 pp. Available at http://water.usgs.gov/fisp/docs/Report_5.pdf (accessed on 12 June 2012).
Measuring Suspended Sediment 165
Author's personal copy
Survey (1998a) provided more information on ranges in transit
rates required for isokinetic sampling.
1.10.2.2.1.2.2 Handheld and handline samplers US depth-
integrating, hand-held (DH)-81, US DH-48, US DH-59, US
DH-76, and US DH-95: where streams are wadable or access
can be obtained from a culvert, low-bridge span, cableway or
boat, and any of six lightweight samplers (Table 2)canbe
used to obtain suspended sediment samples via a wading rod
or hand line. The US DH-81 sampler (Figure 4(a);Table 2),
which is deployed by a wading rod, consists of a US DH-81A
adapter and US D-77 cap and nozzle (U.S. Geological Survey,
2006; Edwards and Glysson, 1999;Gray et al., 2008).
All parts can be autoclaved, enabling the collection of a
depth-integrated sample for a flow-weighted, unbiased bac-
terial analysis. Any bottle having standard Mason jar threads
can be used with the US DH-81 sampler. The unsampled
zone – the distance from the centerline of the nozzle to the
streambed when the sampler contacts a flat bed – varies
depending on the size of bottle used.
The US DH-81 is advantageous for sampling in subfreezing
temperatures because the plastic sampler head and nozzle
(a) (b)
(c) (d)
(e)
Figure 4 Handheld and handline samplers. (a) The US DH-81 suspended sampler with an attached wading rod (the D-77 cap and nozzle is
located to the left of the bottle and sampler body). (b) The US DH-48 suspended sediment sampler with an unattached wading rod. (c) The US
DH-59 suspended sediment sampler with hanger bar. (d) The US DH-76 suspended sediment sampler with hanger bar. (e) The US DH-95
suspended sediment sampler with hanger bar.
166 Measuring Suspended Sediment
Author's personal copy
attach directly to the plastic or glass bottle. Sampler bodies
constructed of metal are relatively efficient at conducting heat
away from the nozzle, air exhaust, and bottle, resulting in
increased potential for ice blockage of the nozzle and (or) the
exhaust port and likely subsequent negation of isokinetic
sampling properties.
The rod-suspended US DH-48 sampler (Figure 4(b);
Table 2) features a streamlined aluminum casting that par-
tially encloses the sample container (Federal Interagency
Sedimentation Project, 1952;Edwards and Glysson, 1999;
Gray et al., 2008). The container, usually a 0.45-l glass milk
bottle, is sealed against a gasket recessed in the head cavity of
the sampler by a hand-operated, spring-tensioned pull-rod
assembly at the tail of the sampler.
The US DH-59 and US DH-76 samplers (Figures 4(c) and
(d),respectively;Table 2 ) are designed for use in unwadable
streams with maximum depths less than 4.6 m and flow vel-
ocities up to approximately 1.5 m s
1
. The fundamental dif-
ference between the samplers is that the US DH-59
accommodates a 0.45-l sample bottle, whereas the US DH-76
uses a 0.9-l container. The tailfin assembly for each sampler
ensures sampler alignment parallel to the direction of the am-
bient flow with the intake nozzle entrance-oriented upstream.
The US DH-95 sampler (Figure 4(e);Table 2) is designed
to enable collection of unbiased 1 l samples for trace element
analyses in addition to samples for suspended sediment ana-
lyses (McGregor, 2000a) in depths less than 4.6 m and flow
velocities up to approximately 2.4 m s
1
. The sample bottle,
nozzle cap, and nozzles are available in plastic or, for trace-
element sampling, teflon. The bottle cavity is machined from a
low-lead bronze casting and is plastic coated. The tail section
is constructed from plastic.
1.10.2.2.1.2.3 Cable-and-reel samplers US D-74, US D-95,
US P-61A1, US P-63, US P-72: the US D-74 (Figures 5(a) and
(b)), US D-74AL, and US D-95 (Figure 5(c)) depth-inte-
grating samplers can be used to obtain suspended sediment
samples in unwadable streams less than 4.6 m deep
(Table 2).
The bronze US D-74 and aluminum US D-74AL are de-
signed to be suspended from a bridge, cableway, or boat. The
US D-74 sampler completely encloses a 0.9-l sample container
or a standard 0.45-l milk bottle when an adapter is used. The
sampler head is hinged at the bottom and swings downward
to provide access to the sample container chamber. The body
includes tail vanes that serve to align the sampler and intake
nozzle with the flow.
The US D-95 (Figure 5(c)) sampler, like the US DH-95
(Figure 4(e)), is designed to make possible collection of un-
biased samples for trace element analyses in streams not ex-
ceeding 4.6 m in depth (McGregor, 2000b) at stream velocities
ranging from 0.5 to 2.3 m s
1
. The bronze body casting is
coated with plastic, and the tail section is constructed from
plastic to help avoid metal contamination during water
quality sampling.
Point-integrating suspended sediment samplers in wide use
are the US P-61A1 (Figure 5(d)), US P-63, and US P-72
(Table 2). Each is equipped with an operator-controlled
sampler solenoid valve powered by a nonsubmersible battery
pack, which makes possible collection of a sample at a discrete
depth or can start and stop depth-integrated sample col-
lection. Automatic pressure equalization at depth precludes a
sudden inrush of sample due to a static head differential when
the valve is opened. All point-integrating samplers can be used
in depth-integration mode.
The US P-61A1 (Figure 5(d);Table 2) is calibrated for use
in velocities up to 2 m s
1
, but there is evidence to suggest that
it can collect samples isokinetically at velocities of at least
3ms
1
(Wayne O’Neal, FISP, 2000, written communication).
The US P-63 and US P-72 are heavier and lighter versions and
have higher and lower flow-velocity limits, respectively, but
otherwise are functionally similar to the US P-61A1. The
newest of the FISP series of point-integrating samplers – the
US P-6, approved for use in 2010 – has operational range
similar to the US P-61A1 but with a simpler valve assembly.
Because of the comparatively complex nature of point-
integrating samplers, the user may find it useful to seek add-
itional information given in FISP reports (Federal Interagency
Sedimentation Project, 1952,1963;Davis, 2005) or to obtain
information directly from the Federal Interagency Sedimen-
tation Project (2012).
1.10.2.2.1.2.4 Bag samplers Samplers using collapsible
bags as the sample container have been used since the 1970s
(Stevens et al., 1980). Nordin et al. (1983) tested a large-
volume bag sampler in the Rio Orinoco and Rio Amazonas,
South America. Moody and Meade (1994) deployed a bag
sampler of the type devised by Stevens et al. (1980) in the
Mississippi river and selected tributaries.
As with rigid-bottle isokinetic samplers, water enters the
bag sampler through a nozzle. However, a bag sampler has no
exhaust port, and the sample container is a collapsible bag. Air
is manually expelled from the bag before submersion of the
sampler. The empirically determined minimum transit rate for
a bag sampler is constrained by the intake nozzle and bag
volume; and the maximum transit rate is 0.4 times the mean
flow velocity in the vertical. When a Teflon bag is used, bag
samplers are capable of collecting unbiased samples for trace
element analyses in addition to those collected for suspended
sediment analyses.
The US D-96 collapsible bag sampler (Figures 5(e) and (f);
Table 2) was developed in part to address the limitations and
disadvantages associated with bottle and previous bag samplers
(U.S. Geological Survey, 2006; Davis, 2001, 2005). This cable-
suspended sampler can provide up to 3 l of sample for sub-
sequent unbiased trace element analyses in addition to physical
sediment analyses. It is fabricated from bronze and aluminum
castings with a high-density polyethylene tail. All metal parts
are plastic coated with commercially available ‘PlastiDip.’ A
sliding tray (Figure 5(f)) in the sampler holds the nozzle
holder with nozzle in place and supports a perfluoroalkoxy bag.
The US D-96 sampler can collect velocity-weighted samples in
streams with velocities from 0.6 to 3.8 m s
1
,althoughmin-
imum velocities require the larger diameter nozzles. At a max-
imum transit rate of 0.4 times the mean flow velocity in the
vertical, the US D-96 sampler is capable of sampling to a depth
of 34 m with a 4.8-mm diameter nozzle (Davis, 2001;Table 2).
Bag samplers with smaller and larger capacities than the US
D-96 sampler are also available. The 13-kg US DH-2 is a
Measuring Suspended Sediment 167
Author's personal copy
handline sampler capable of collecting a 1 l sample. The
heaviest FISP sampler at 125 kg – the US D-99 – is capable of
collecting a 6 l sample at a maximum depth of 67 m when a
4.8-mm diameter nozzle is used.
1.10.2.2.2 Manual sampling methods
The most common purpose of sediment sampling is to de-
termine the instantaneous mean water discharge-weighted
SSC at a cross section. Derived SSC values are combined with
water discharge to compute the measured suspended sediment
discharge. A water discharge-weighted SSC representative of
the mean value in the cross section is desired for this purpose
and for the development of coefficients to adjust data
collected by observers, automatic samplers, and selected sus-
pended sediment surrogate instruments.
Ideally, the best method for sampling any stream to de-
termine suspended sediment discharge would be to collect
the entire flow of the stream over a given time period, and
extract, dry, and weigh the sediment. This method is rarely
feasible. Instead, the SSC in the flow is determined by col-
lecting depth-integrated suspended sediment samples that
define the mean water discharge-weighted SSC in the sample
vertical, and collecting samples from sufficient verticals to
define the water-mean discharge-weighted SSC in the cross
section (Edwards and Glysson, 1999;Nolan et al., 2005;
Gray et al., 2008).
(a) (b)
(c) (d)
(e) (f)
Figure 5 (a) The US D-74 suspended sediment sampler. (b) The US D-74 suspended sediment sampler open. (c) The US D-95 suspended
sediment sampler. (d) The US P-61A1 point-integrating suspended sediment sampler. (e) The US D-96 suspended sediment sampler. (f) The US
D-96 suspended sediment sampler with tray extended.
168 Measuring Suspended Sediment
Author's personal copy
1.10.2.2.2.1 Single-vertical sampling
The objective of collecting a single vertical sample is to ob-
tain an SSC value representative of the mean water discharge-
weighted SSC in the vertical being sampled at the time the
sample was collected. An isokinetic sampler deployed at a
constant rate in a downward and an upward transit will
collect a sample weighted for the variations in velocity and
SSC in the vertical from the surface to the top of the
unsampled zone. The following equation demonstrates this
concept:
Ci¼ZDi
BiþUZ
cisðÞvisðÞds
ZDi
BiþUZ
visðÞds
½1
where C
i
is the mean SSC in vertical i,B
i
the elevation of the
stream bed in vertical i,UZ the distance from the bed to the
nozzle of a sampler resting on the bed (unsampled zone), D
i
the elevation of the water surface in vertical i,c
i
(s) the SSC at
depth sin vertical i,sthe depth in the vertical, and v
i
the
velocity at depth sin vertical i.
The method used to obtain the mean SSC in a vertical
depends on the flow conditions and PSD of the sediment in
transport. These conditions can be generalized to four types of
situations (Edwards and Glysson, 1999), where v equals the
mean ambient velocity in the sampled vertical:
1. Low velocity (vo0.6 m s
1
) when little or no sand is being
transported in suspension.
2. High velocity (0.6rvr4.6 m s
1
) when depths are less
than 4.6 m.
3. High velocity (0.6rvr4.6 m s
1
) when depths are greater
than 4.6 m.
4. Very high velocity (v44.6 m s
1
).
1.10.2.2.2.1.1 First case When the mean ambient velocity
in the sampled vertical is less than 0.6 m s
1
, barring very
shallow depths, the velocity is low enough so that little if any
sand is in suspension. The distribution of any silt- and clay-
size material (o0.062 mm in diameter; Guy, 1969)in
transport is likely to be relatively uniform from stream sur-
face to bed (Guy, 1970). The sampling error for this case is
10% or less with relative sampling rates in a range from ap-
proximately 0.2 to at least 5 (Figure 3). Consequently,
it is less important to collect the sample isokinetically with
fines in suspension than it is when sand-size particles
(0.062–2 mm in diameter; Guy, 1969) are in suspension. In
shallow streams, a sample may be collected by manually
submerging an open-mouthed bottle into the stream to ob-
tain a ‘grab’ sample. There are at least two ways to collect a
proper ‘grab’ sample; the sample bottle should be:
1. held tilted upward at an approximate 451angle from ver-
tical in the upstream direction. The bottle should be filled
by moving it from the surface vertically toward the
streambed and back at a constant rate, or
2. inverted and thrust vertically through the water column to
a depth no closer than approximately 8 cm of the bed;
tilted upward; and filled as it is raised at a constant rate to
the surface.
In neither case should the bottle be filled to the brim, nor
should the bottle mouth come closer than approximately
8 cm from the bed (the unsampled zone) to obtain samples
that are compatible with those obtained using depth-inte-
grated samples at higher velocities. In exceptionally shallow
water where the 8-cm unsampled zone limitation might pre-
clude sampling in a substantial proportion of the vertical, an
effort should be made to sample a substantial proportion of
the vertical without disturbing the bed material.
1.10.2.2.2.1.2 Second case When 0.6rvr4.6 m s
1
and
the depth is less than 4.6 m, a depth-integrating sampler
described in Table 2 that is suitable for the ambient
streamflow condition should be used. Note that for many
samplers, the maximum velocity for this case is less than
4.6 m s
1
(Table 2). The method of sample collection is
basically same for all of the aforementioned samplers, whe-
ther used while wading or deployed from a bridge, cableway,
boat, or other platform:
1. Insert a clean container into the sampler and ensure that
the nozzle and the air exhaust tube (where applicable) is
unobstructed.
2. Lower the sampler to the water surface, so that the nozzle is
above the water and the lower tail vane or back of the
sampler is in the water with the nozzle pointing directly
into the oncoming flow.
3. Lower the sampler at a constant rate until it touches the
bottom (see following caveat), and immediately retrieve it
at a constant rate until it clears the water surface.
In rare cases, site conditions are such that it is unadvisable
for the descending sampler to contact the bed. For example,
the presence of anchored or mobile bed features capable of
snagging or engulfing the sampler can present unacceptable
safety concerns or at least the prospect of losing the sampler.
Also, the presence of a soft bed into which the sampler sinks –
or a bedform that the sampler nozzle gouges – can result in
contamination of the sample with bed material. Under these
and perhaps other extenuating circumstances, it may be
acceptable or necessary to sound the depth before sampling
and reverse direction of the sampler before contacting the
bottom. Some USGS hydrographers sound bottom at a vertical
with an echo sounder or sounding weight so as to determine
the depth from which to retrieve the sampler before contacting
the bed. Some samplers used in the lower Mississippi river and
Atchafalaya rivers, for example, are deployed no lower than
the 0.9 depth due to safety or sample contamination concerns
(Paul Frederick and Michael Manning, USGS, written com-
munications, 2012). Theoretically, partial-depth samples
would tend to have low-biased SSC values compared to those
obtained in the full-sampled depth due to SSC gradients that
favor larger SSC values with depth, particularly when sand-
size fractions are in suspension (American Society of Civil
Engineers, 2006).
Those who consider using this modified 0.9-depth sampling
method (or another user-determined sampling depth using a
Measuring Suspended Sediment 169
Author's personal copy
technically supportable rationale, such as the 0.95 depth) in
lieu of the standard sampler-deployment procedure (Edwards
and Glysson, 1999) should be cognizant of the objective to
balance the unequivocal need for sampling safely and repre-
sentatively, with the desire to minimize the size of the
unsampled zone over that associated with a sampler contacting
the bed. If a modified procedure of this type is used, it must be
fully documented. Additionally, it is advisable to measure SSC
profiles in verticals with a point-integrating sampler (or to
collect depth-integrated samples from the full depth to compare
to those collected at the 0.9 depth when it can be done safely,
checking carefully for evidence of excess or inordinately coarse
sediment in the full-depth sample) to aid in estimating errors in
the mean SSC from use of the nonstandard technique.
For streams that transport substantial suspended sand loads
or otherwise exhibit marked spatial SSC heterogeneity in the
cross section, at least two complete depth integrations of the
sample vertical should be made as close together in time as
possible, one bottle for each integration. Each bottle then
constitutes a sample and can be analyzed separately or, for the
purposes of computing the sediment record (a time series of
sediment discharges often reported as daily values), SSC values
representing two or more bottles can be averaged as a set and
tagged with a single time of collection. This set is used as a
single SSC value for computing the sediment record. In an
analysis of uncertainties arising from potential inadequate time
averaging in sample collection at a verticals and over a cross
section, Topping et al. (2011) recommend collecting a total of
four water-column transits in as many verticals as practical, with
each transit consisting of lowering and raising the sampler at a
constant rate between the surface and river bed.
1.10.2.2.2.1.3 Third case When 0.6rvr4.6 m s
1
and
the depth is greater than 4.6 m, rigid-bottle depth-integrating
samplers cannot be used because the depth exceeds the
maximum allowable depth for these samplers (some rigid-
bottle depth-integrating samples are limited to deployment
in depths less than 4.6 m; see Table 2). In this case, a point-
integrating or bag-type sampler must be used. The method
for collection of a sample using the bag-type sampler is
similar to that used with the depth-integrating samplers.
A point sampler may be used to collect depth-integrated
samples in verticals where the depth is greater than 4.6 m. For
streams with depths of 4.6–9.1 m, a procedure for sampling
modified from that described by Edwards and Glysson (1999)
is as follows:
1. Insert a clean bottle in the sampler and close the sampler
head.
2. Lower the sampler to the streambed, keeping the solenoid
valve closed; note the depth to the bed. If there is a high
probability of contaminating the sample with bottom
material gouged from the bed, then it is acceptable to
sound the depth before sampling and initiate sample col-
lection as shallow as the 0.90 depth.
3. Simultaneously, open the valve and start raising the sam-
pler to the surface, using a constant transit rate.
4. Keep the valve open until after the sampler has cleared the
water surface. Close the valve.
5. Remove the bottle containing the sample, check the vol-
ume of the sample, and mark the appropriate information
on the bottle. (If the sample volume exceeds allowable
limits, discard the sample and repeat depth integration
using a faster transit rate that does not exceed 0.4 times the
mean velocity in the vertical.)
6. Insert a clean bottle into the sampler and close the
sampler head.
7. Lower the sampler until the nozzle above water but with
the lower tail vane touching the water, allowing the
sampler to align parallel into the flow.
8. Open the valve and lower the sampler at a constant transit
rate until the sampler to the same depth at which the initial
sample was collected (the bed, or to as shallow as the 0.9
depth).
9. Close the valve when the sampler touches the bed or at the
same minimum depth at which the first sample was col-
lected (by noting the depth to the streambed in step 2
above, the operator will know when the sampler is ap-
proaching the bed under all but the most dynamic bed-
load-transport conditions).
If the stream depth is greater than 9.1 m, the process is
similar, except that the descending and ascending integrations
are broken into segments no larger than 9.1 m. Samples col-
lected by this technique may be composited for each vertical if
the same transit rate is used. Otherwise, samples should be
analyzed separately. A single mean SSC is computed for the
vertical (as described in the second case (above), the hydrog-
rapher might opt to retrieve the sampler before touching
bottom due to safety considerations, or if there is a high
probability of contaminating the sample with bottom
material).
1.10.2.2.2.1.4 Fourth case When v44.6 m s
1
, the vel-
ocities are too large to deploy a US D-99 or a US P-63, which
share the highest velocity rating of the FISP suite of isokinetic
samplers. In this case, when the presence of entrained debris
or ice, or other factors, makes normal sample collection
dangerous or impossible, a surface or dip samples, or, better
yet, a partial-depth isokinetic sample may be collected.
A surface sample is one taken on or near the surface of the
water, with or without an isokinetic sampler. At some locations
and at certain times, such as during floods, stream velocities
and depths can be so large that even the heaviest, most
streamlined samplers will not reach the streambed in one or
more sampled verticals. Under such conditions, it can be ex-
pected that all but perhaps the largest sediment particles in
suspension will be well mixed within the flow; and, therefore, a
sample from near the surface, nondepth integrated, may con-
tain an SSC and size distribution representative of the entire
vertical. However, results from these samples should be cor-
related with those from depth-integrated samples collected
under more normal flow conditions as soon as possible after
the large velocities diminish. Along with the depth-integrated
sample, a sample should be collected in a manner duplicating
the sampling procedure used to collect the surface or dip
sample. Analytical results from these samples will be used to
adjust those from the surface or dip sample collected during the
170 Measuring Suspended Sediment
Author's personal copy
higher flow, if necessary, to facilitate the use of these data in
sediment discharge computations and data analyses.
A partial-depth isokinetic sample also requires correlation
to proper depth-integrated samples collected at more normal
flows. If done rigorously, it can be used as a relatively reliable
analog to a full-depth sample. For example, one can only
sample to 4 m or 50% of the 8 m depth at a given vertical
during a flood peak. At the next site visit when the flood has
receded to a maximum depth of 4 m, isokinetic samples are
collected at the same vertical throughout the full depth of 4 m,
and also at 2 m, or 50% of the full depth. The SSC from the
sample collected in the full 4-m depth during the recession is
divided by that at 2 m collected during the same site visit. The
resulting ratio is multiplied by the flood peak SSC determined
from the sample collected in the top half of the vertical. The
product is an empirically derived estimate of the actual full-
depth, flow-weighted SSC during the flood peak. This ap-
proach would have to be repeated using appropriate ratios for
each vertical sampled during the flood peak.
1.10.2.2.2.2 Multivertical sampling
A depth-integrated sampler collected using the manual-
sampling methods described in Section 1.10.2.2.2 will accur-
ately represent the water discharge-weighted SSC in a vertical
at the time of the sample collection. Samples collected at ap-
propriately spaced verticals can be used to calculate the in-
stantaneous SSC at a cross section. The International
Standards Organization (ISO, 2002) lists three methods for
suspended sediment data collection in a cross section: the
equal-discharge-increment (EDI), equal-width-increment
(EWI), and equal-area-increment methods. The equal-area-
increment method is rarely used in the US. The first two
methods are described in the following sections (Edwards and
Glysson, 1999;Nolan et al., 2005;Gray et al., 2008).
1.10.2.2.2.2.1 EDI method With the EDI method, samples
are obtained from the locations representing equal incre-
ments of water discharge. The EDI method is predicated on
three criteria:
1. Samples are collected isokinetically (river conditions
permitting).
2. The vertical represents the mean SSC and PSD for the
subsection sampled.
3. The water discharges on either side of the sampling vertical
are predetermined proportions of the total water discharge,
which requires information on the lateral distribution of
water discharge in the cross section. The lateral distribution
of water discharge is usually obtained by a prior discharge
measurement or a relation developed between river stage
and the lateral distribution of discharge if the channel
geometry is stable.
The mean water discharge-weighted SSC in a cross sec-
tion using the EDI method is calculated from the mean SSC
values from individual verticals (see Section 1.10.2.2.2.1) as
follows:
Cxs ¼1
nX
n
i¼1
Ci½2
where C
xs
is the mean water discharge-weighted SSC in the
cross section, nthe number of verticals used in the EDI
measurement, and C
i
the mean SSC in the vertical i.
The lateral distribution of water discharge can be derived
from a discharge measurement made immediately before se-
lecting sampling verticals (Rantz, 1982), or, if the channel is
relatively stable on an analysis of the lateral distribution of
water discharges measured over a range of historical flows. If
such knowledge can be obtained, the EDI method can save
time and labor (compared to the EWI method, discussed in
the next section), because fewer verticals are required (Hubbell
et al., 1956, p. 35). This conclusion would be particularly
applicable to larger streams.
The inverse of the number of verticals, n, to be sampled by
the EDI method is multiplied by 100% to derive qpercent, the
percentage of water discharge to be represented in samples
collected in each vertical. The location of a vertical nearest the
left bank is selected at-a-point at which the cumulative water
discharge to the left of the vertical is one-half of the total water
discharge times qpercent. Similarly, the location of a vertical
nearest the right bank is selected at-a-point at which the cu-
mulative water discharge to the right of the vertical is one-half
of the total water discharge times qpercent. All other verticals
are selected at points where the cumulative water discharge
between adjacent verticals is equal to the total water discharge
times qpercent.
For example, samples are to be collected from eight in-
crements of equal water discharge from a 100-m wide cross
section of a river flowing at 500 m
3
s
1
. The percentage of the
total water discharge to be represented in samples collected
from each vertical is one-eighth times 100% or 12.5%. The
location of the vertical nearest the left bank is selected at the
point at which the cumulative water discharge to the left of
that vertical is one-half the product of 500 m
3
s
1
times
12.5%, or at the point in the cross section where approxi-
mately 31 m
3
s
1
flows between the vertical and left bank.
Likewise, the vertical nearest the right bank is selected at the
point at which approximately 31 m
3
s
1
flows between the
vertical and right bank. The other six verticals are located at
points separating adjacent verticals by water discharges of
62.5 m
3
s
1
, the product of the total river discharge of
500 m
3
s
1
, times qpercent, 12.5%. The location of each
vertical represents the centroid of the water discharge in its
respective subarea, with each subarea containing equal incre-
ments of water discharge. The lateral locations of the sampling
verticals in natural settings are at irregular intervals due to a
typically uneven channel morphology and variable velocity
distribution.
Samples are collected from each EDI method vertical as
described previously in the Section 1.10.2.2.2.1. Although a
given one-way transit rate must be constant, neither the des-
cending and ascending transit rates in any one vertical need
to be equal nor do the transit rates need to be equal
among verticals. The number of transits in each vertical can
vary as long as no sample bottle overfills. Although different
diameter nozzles for the isokinetic sampler can be used from
vertical to vertical, it may complicate the data collection pro-
cedure, hence, the practice is discouraged.
The EDI method requires 4–9 verticals. At least two sep-
arate samples are recommended to be collected at each
Measuring Suspended Sediment 171
Author's personal copy
vertical. The greater the potential heterogeneity in the distri-
bution of SSC and PSDs in the cross section, the more verticals
should be selected. If an equal amount of sample is collected
at each vertical, the samples can be composited and analyzed
as a single sample. However, to realize the full benefits of the
EDI method, samples should be analyzed separately and the
resulting SSC values summed and then divided by the number
of subsections to derive a mean water discharge-weighted SSC.
One advantage of this method is that data describing the
cross-sectional variation in SSC are produced. Additionally, a
bottle containing an abnormal amount of sediment – par-
ticularly sand – compared to others in the set (because of
recirculation or to gouging the nozzle into the bed) can be
identified and excluded from the calculated mean cross-sec-
tional SSC to minimize the potential for bias.
The bed of a sand channel can shift substantially at single
points and across segments of the width and at timescales as
short as fractions of an hour. This not only makes it difficult at
best to establish a relation between stage and the distribution
of the cross-section water discharge distribution from one visit
to the next but also makes it impossible to be certain that the
water discharge distribution does not change between the time
of the water discharge measurement and sample collection
(see Guy, 1970). Under conditions where the lateral distri-
bution of flow changes rapidly or on rapidly rising or falling
hydrographs, the EDI method may yield unreliable results.
1.10.2.2.2.2.2 EWI method A cross-sectional suspended
sediment sample obtained by the EWI method requires a
sample volume proportional to the amount of flow at each of
10 or more equally spaced verticals in the cross section
(Edwards and Glysson, 1999;Nolan et al., 2005;Gray et al.,
2008). Equal spacing between EWI verticals across the stream
and sampling at an equal transit rate at all verticals yields a
cumulative sample volume proportional to the total water
discharge. This method first was used by Colby in 1946
(Federal Interagency Sedimentation Project, 1963)andis
used most often in relatively shallow, wadable streams, and
(or) alluvial streams where the distribution of water dis-
charge in the cross section is temporally unstable or un-
known. It also is useful where SSCs in the cross section are
markedly heterogeneous, such as at sites where inflow from
an upstream tributary has yet to fully mix with the main
channel flow.
The mean water discharge-weighted SSC in a cross section
using the EWI method is calculated from the mean SSC from
individual verticals (see Section 1.10.2.2.2.1) as follows:
Cxs ¼P
J
j¼1
VoljCj
P
J
j¼1
Volj
½3
where C
xs
is the mean water discharge-weighted SSC in the
cross section, Jthe number of sample bottles used in the EWI
measurement, C
j
the SSC in the sample bottles j, and Vol
j
the
total volume of water collected in sample bottle j.
The number of verticals required for an EWI sediment-
discharge measurement depends on the distribution of SSC
and flow in the cross section at the time of sampling, as well as
on the relative assessment of the desired accuracy of the result.
For many streams, statistical approaches and experience are
needed to determine the desirable number of verticals. Until
such experience is gained, the number of verticals used should
be somewhat larger than that deemed to be minimally ne-
cessary. In all cases, a minimum of 10 verticals should be used
for streams exceeding 1.5 m wide. For streams less than 1.5-m
wide, adjacent verticals should be separated by at least 8 cm to
allow discrete sampling of each vertical while avoiding over-
laps. Through general experience with similar streams, field
personnel can estimate the required minimum number of
verticals to yield a desired relative level of accuracy. For all
but the widest and shallowest streams, or when flood flows
result in consequential floodplain flows, 20 verticals usually
are ample.
The width of the increments to be sampled, or the distance
between verticals, is determined by dividing the stream width
by the number of verticals, n, considered adequate to collect a
water discharge-weighted suspended sediment sample repre-
sentative of the SSC of the flow in the cross section. The lo-
cations of the two verticals nearest to the banks are at a
distance of one-half of the total width divided by n. The lo-
cations of the other verticals are separated from adjacent ver-
ticals by a distance equal to the total width divided by n. The
locations of these verticals represent the centroid of subareas
with boundaries one-half the distance to adjacent verticals.
Hence, only the widths of the subareas necessarily are equal.
The EWI sampling method requires use of the same size
nozzle for a given measurement, and all verticals must be
traversed using a transit rate that will not result in overfilling
the sample bottle at any vertical. The descending and as-
cending transit rates must be equal for all verticals and during
the sampling traverse of each vertical. By using this equal
transit rate technique with a standard depth- or point-inte-
grating sampler at each vertical, a volume of water pro-
portional to the flow in the vertical will be collected.
For example, from the previous paragraphs, samples from
12 verticals are to be collected from a 120-m wide stream. The
location of the leftmost vertical is at a distance of one-half of
the 120-m width divided by 12 (the number of verticals) or at
5 m from the left bank. Each of the 12 verticals are located
10 m apart with the rightmost vertical at 115 m from the left
bank (5 m from the right bank).
The maximum transit rate must not exceed 0.1, 0.2, or 0.4
(a 0.4, transit rate applies to all bag samplers) depending on
the sampler nozzle and bottle size (Edwards and Glysson,
1999). The minimum transit rate must be sufficiently fast to
keep from overfilling the sample container. Therefore, the
minimum transit rate to be used at all verticals is limited by
conditions at the vertical that represents the largest water
discharge per unit width, or, in operational terms, that vertical
with the largest product of depth times mean velocity. A dis-
charge measurement can be made to identify the location of
this vertical. In practice, this location often is estimated from
experience or by sounding for depth and acquiring a feel for
the relative velocity with a deployed sampler, sounding
weight, or wading rod. The transit rate required at the max-
imum water discharge vertical then must be used at all other
verticals in the cross section and often is set to provide the
maximum sample volume in a round-trip transit. It is
172 Measuring Suspended Sediment
Author's personal copy
permissible to sample at multiple verticals using the same
bottle as long as the bottle is not overfilled. If a bottle is
overfilled, the contents must be discarded and all verticals
sampled using that bottle must be resampled, using at least
two bottles to avoid overfilling.
1.10.2.2.2.2.3 Number of cross sections required to quantify a
suspended sediment discharge The requisite number of EDI
and (or) EWI cross sections to compute a reliable suspended
sediment discharge is dependent on at least two factors:
1. Whether or not the purpose of the measurements is to
define a coefficient to estimate cross-section SSC using
samples collected at discrete locations such as at a single
vertical by an observer, by pumping sampler, or by an
in situ suspended sediment surrogate technology.
2. The risk associated with producing an incorrect result due
to errors in the sampling process.
In the first case, Guy and Norman (1970, p. 38) stated that,
‘‘it is desirable to obtainytwo or more mean SSCs [i.e., EDI
and (or) EWI cross sections] yfor a cross section.’’ Edwards
and Glysson (1999, p. 60) implicitly concur, stating that,
‘‘Samples should be collected at the observer’s single vertical
using the observer’s equipment, both before and after each
cross-section samples is taken.’’
In the second case, it is an unfortunate truism (based on
the authors’ cumulative 65 years of experience in hydrology
and sedimentology in 2012) that imprecise or erroneous
sediment data are occasional if inadvertently produced as a
result of mistakes in the sampler selection/sample collection/
processing/transportation/analysis continuum. Production of
spurious data can arise from a number of sources, including
inadequate definition of the SSC in a vertical due to an in-
sufficient number of transits (Topping et al., 2011), incomplete
transits, or use of inappropriate transit rates; gouging the
sampler nozzle or sinking the sampler into the streambed;
alterations arising from subsampling and transport to the
sediment laboratory; and, however infrequent, analytical or
computational errors. Regardless of the source of the error, an
incorrect value from a sample is arguably worse than no data.
Incorrect data nullify the benefits of the resources expended in
their collection and may result in added time and resources in
an attempt to identify and purge the erroneous data.
Based on the facts and rationale above – and on the
findings of Topping et al. (2011) – when the objective is to
define a cross-section SSC coefficient, at least two sets of
samples collected by the EDI and (or) EWI methods are re-
quired in the measurement of suspended sediment discharge.
Additionally, multiple samples need to be collected from the
instrument to be calibrated, usually double the number of
cross sections sampled. For other purposes, multiple EDI and
(or) EWI sample sets are strongly recommended to minimize
the potential for spurious data to remain undetected.
In the case of the EDI method, one may collect multiple
samples at each vertical in a single pass across the stream. The
first sample at each vertical may be designated ‘cross section A,’
the second,‘‘cross section B,’’ etc. Separate mean SSC values are
produced for each cross section, and, if the streamflow con-
ditions have remained more or less the same during both cross
sections and the SSC values are similar among the cross sec-
tions, the mean SSC values from each cross section are aver-
aged and multiplied by the water discharge and a constant
units conversion factor to produce a value of suspended
sediment discharge.
This process to derive a single suspended sediment dis-
charge value from EWI samples is more straightforward. The
mean SSC values produced from each cross section are aver-
aged to produce a single mean SSC value for subsequent
suspended sediment discharge computation. There are no re-
strictions with respect to the starting location for EWI
suspended-sediment sampling in a cross section. The most
common approach is to begin sampling from near one side of
the stream and progress to the opposite bank, and to collect
subsequent EWI sample set(s) starting at or near the location
of the last sample from the previous sample set.
EDI and EWI methods can be used in tandem. For ex-
ample, an EWI sample set may be used to produce a PSD and a
mean SSC, and an EDI sample set used to describe SSC. The
benefits of this approach are from the detailed cross section
distribution of SSC, which can readily identify outliers; and a
PSD that is representative of that in the cross section.
In rare cases, a combination of depth-integrated sampling
and dip sampling to quantify suspended sediment transport
may be appropriate. For example, in the mid-2000s, the right
third of the Rio Grande at Albuquerque, New Mexico (USGS
streamgage 08330000) averaged approximately 0.6-m deep
with mean velocities exceeding 0.6 m s
1
– conditions
amenable for depth-integrated sampling. The left two-thirds of
the channel had a more-or-less uniform 0.2 m depth – con-
ditions amenable to dip sampling. A 10-section EDI meas-
urement – those samples in the deeper section collected with a
depth-integrating sampler and those in the shallow section
dipped – was used to quantify suspended sediment discharge.
Note that an EWI measurement would be inappropriate under
these circumstances, because of the unmet requirement for
uniform transit rates with an isokinetic sampler over the entire
cross section.
1.10.2.2.2.2.4 Advantages of the EDI and EWI methods
Some advantages and disadvantages of both the EDI and EWI
methods have been noted in the previous discussion. It must
be remembered, however, that both methods, if properly
used, will yield similar cross-sectionally averaged results.
The advantages of the EDI method are as follows:
1. The fewer requisite verticals typically result in a reduced
collection time, which is particularly advantageous during
periods of rapidly changing water discharge.
2. Bottles composing a sample set may be composited for
single laboratory analysis when equal volumes of sample
are collected from each vertical.
3. The cross-sectional variation in SSC can be determined if
samples are analyzed individually, and errors often can be
readily identified by a comparison of SSC values in the
cross section.
4. Multiple cross-sectional samples can be collected at each
vertical during a single traverse of the stream.
5. A variable transit rate can be used among verticals although
the rate of a single-direction transit must remain constant.
Measuring Suspended Sediment 173
Author's personal copy
The advantages of the EWI method are as follows:
1. No antecedent knowledge of flow distribution in the cross
section is required.
2. Variations in the distribution of SSC in the cross section
may be better integrated in the composite cross section
sample due to the larger number of verticals sampled.
3. Analytical time and costs are minimized as sample bottles
are composited for single laboratory analysis.
4. This method is easily learned and used due to the
straightforward spacing of sample verticals based on stream
width, rather than on the cross-sectional distribution of
water discharge.
5. Generally, less total time is required on site if no discharge
measurement is deemed necessary and the cross-section
geometry is relatively stable during the measurement.
The advantages of one method are, in many cases, the dis-
advantages of the other. The U.S. Geological Survey (1998a)
considered the EDI method the most universally applicable and
useful water discharge-weighted sampling method.
1.10.2.2.2.2.5 Transit rates for suspended sediment sam-
pling A sample obtained with an isokinetic sampler using
depth integration is quantitatively weighted according to the
velocities through which it passes. Therefore, if the sampling
vertical represents a specific width of flow, the sample is
considered to be water discharge-weighted because, with a
uniform transit rate, the suspended sediment conveyed at
varying velocities throughout the sampled vertical is given
equal time to enter the sampler.
The transit rate used with any depth-integrating sampler
must be regulated to make possible the collection of repre-
sentative samples (i.e., isokinetically collected). Although the
descending and ascending transit rates need not be equal, in
practice it is simpler, ergo less prone to error, for the transit
rates to be equal in both directions. Regardless, the rate of any
one-direction transit must be constant to obtain a valid vel-
ocity- or water discharge-weighted sample. The goal in se-
lecting transit rates is to fill the sample bottle to its optimum
level (Johnson, 1997;Edwards and Glysson, 1999;Nolan
et al., 2005;Gray et al., 2008).
An insufficient transit rate can result in an unacceptable
sample due to overfilling of the sample container. An excessive
transit rate can result in intake velocities less than the stream
velocity due to a large entrance angle between the nozzle and
streamflow lines caused by the vertical movement of the
sampler in the flow (Federal Interagency Sedimentation Pro-
ject, 1952). Transit rates should never exceed the product of
0.4 and the mean velocity (0.4 v) in a vertical with any iso-
kinetic sampler.
Additional limitations may be imposed on maximum
transit rates for rigid-bottle depth-integrating samplers due to
changes in hydrostatic pressure during deployment. The
maximum allowable transit rate is attained when the rate of
change in the internal pressure due to filling equals the rate of
change of hydrostatic pressure. If the sampler is lowered too
fast in the vertical, inflow through the nozzle is insufficient to
increase the pressure in the container at the same rate; con-
sequently, hydrostatic pressure increases at a greater rate than
pressure in the container. The resulting pressure imbalance
causes the sample to enter the nozzle at a velocity greater than
the ambient stream velocity. Stream water can also enter the
exhaust port under these circumstances. Both potential out-
comes result in violation of isokinetic sampling principles
(Stevens et al., 1980). Likewise, if the sampler is raised too
rapidly, the hydrostatic pressure will decrease at a greater rate
than the pressure inside the container. This pressure im-
balance will result in reduced flow of sample into the con-
tainer with respect to the ambient stream velocity. Either
outcome – larger or smaller intake velocities with respect to
the ambient stream velocity – can result in collection of a
sample that contains neither a representative SSC nor PSD of
suspended sediment.
The maximum allowable transit rate for rigid-bottle sam-
plers can be determined with knowledge about the:
1. depth of the sample vertical,
2. mean velocity of the vertical,
3. sampler nozzle size, and
4. bottle size used in the sampler.
Different combinations of nozzle diameters and bottle
volumes result in maximum transit rates ranging from ap-
proximately 0.1 v to 0.4 v. Tables providing isokinetic transit
rates as a function of nozzle diameters and bottle volumes are
provided by the U.S. Geological Survey (1998a). Graphs de-
lineating permissible and optimal transit rates for a combin-
ation of sample container and nozzle sizes as a function of
stream depth and mean velocity are provided by Edwards and
Glysson (1999).
When using a cable-deployed sampler, there are at least
two options toward maintaining constant transit rates: use of a
power-regulated reel (Jelinski, 2008) or an electronic metro-
nome such as the US vertical transit pacer (VTP)-99 Vertical
Transit Rate Pacer (Federal Interagency Sedimentation Project,
2012).
1.10.2.2.2.2.6 Inspecting samples Inspecting individual
samples is fundamentally important toward evaluating the
magnitude and potential sources of variations in sediment
content of sample bottles during a site visit. The primary
objective is to identify samples collected consecutively from
the same vertical or from adjacent verticals in which SSCs
differ markedly. Such checks should be done routinely as part
of the data collection process.
Immediately after collection, each sample should be in-
spected by briefly swirling or agitating the container and then
observing the quantity of sand particles that collect in the
bottom of the tilted container. If it is perceived that the con-
centration of sand inexplicably differs among the bottles when
visually correcting for differences in sample volumes, or the
sand in a sample bottle is inordinately coarse, it is possible
that the sampled concentration is an artifact of the sampler
nozzle gouging or sinking into the bed. In this case, the ver-
tical should be immediately resampled at least once and
additional times until the hydrographer can ascertain if any
samples are contaminated with bed material. Any sample
container perceived to contain anomalously large quantities,
174 Measuring Suspended Sediment
Author's personal copy
or abnormally coarse size fractions of sand should be marked
as such, or, if the hydrographer lacks confidence in a sample’s
validity, the sample should be summarily discarded. If a
container is overfilled or if water ejects from the nozzle during
sampler retrieval, the sample should also be discarded. A clean
container must be used to resample the vertical.
1.10.2.2.2.3 Point-integrated sampling
A point-integrated sample is a sample of the water–sediment
mixture collected isokinetically from a single point in the cross
section. Point-integrated samples are collected using one of
the point-integrating samplers previously presented
(Figure 5(d)). Multiple point samples may be used to define
the distribution of sediment in a vertical, the vertical and
horizontal distributions of sediment in a cross section, and the
mean cross-sectional SSC.
The purpose for which point samples are to be collected
determines the collection method to be used. If samples are
collected for the purpose of defining the horizontal and ver-
tical distribution of SSC and (or) PSDs, samples collected at
numerous points in the cross section with any of the ‘P’ type
samplers will be sufficient. Vertical distributions can be ad-
equately defined by obtaining samples from a number of
points in each sample vertical. Specifically, samples should be
taken with the sampler lightly touching (or, if the nozzle
might gouge the bed, suspended a short distance above) the
bed, 0.3 m off the bed, from 5 to 10 additional points in the
vertical above that point, and from near the surface.
The sampling time for each sample (the elapsed time that
the nozzle is open) must be equal. This result will ensure that
sample volumes collected are proportional to the flow at the
point of collection. These samples may be composited for a
single laboratory analysis. If the EDI method is used to define
the stationing of the verticals, the sampling time may be
varied among verticals. If the EWI method is used, a constant
time for collecting samples from all verticals must be used. In
most cases, it is necessary to analyze each point sample
separately.
The mean water discharge-weighted SSC in a cross section
using the point-integration method is calculated from the
mean SSCs from individual sampling points as follows:
Cxs ¼1
nX
n
i¼1P
Di
d¼1
VolidCid
P
Di
d¼1
Volid
8
>
>
>
<
>
>
>
:
9
>
>
>
=
>
>
>
;
½4
where C
xs
is the mean water discharge-weighted SSC in the
cross section, D
i
the total number of points sampled in vertical
i,nthe number of verticals in which point samples are col-
lected, C
id
the SSC in a sample from point dof vertical i, and
Vol
id
the volume of sample collected from point dof vertical i.
If multiple points are sampled with a single bottle, com-
putation of the mean sample SSC is accomplished by treating
the contents of the bottle as if collected at a single point.
1.10.2.2.3 The case for depth-integrated sampling with
isokinetic samplers
The natural environment presents a variety of challenges to the
hydrographer responsible for measuring riverine suspended
sediment transport rates. The temporal and spatial variability
of sediment transport can be and typically is substantial.
Quantification of sediment transport over time requires a
sufficient number of measurements (Porterfield, 1972), each
of which adequately accounts for the spatial variability of
sediment transport in the cross section. Because of the non-
uniformity in velocity and SSC profiles in a cross section, the
spatial distribution of sediment discharge per unit area creates
a measurement conundrum: How to measure and properly
weight velocity and SSC variations in the cross section to de-
rive the correct value for suspended sediment discharge in the
cross section at the time of the site visit.
Spatial variations in sediment concentrations can result
from a number of factors including the dynamic site charac-
teristics of proximity and relative sedimentary content of tri-
butary inflows, channel roughness characteristics, hydraulic
radius, flow energy, turbulence, sloughing banks; and the
sedimentary characteristics of specific gravities, shapes, and
size distributions. Even at a cross section with relatively well-
mixed flow and sediments of consistent specific gravities and
shapes, spatial SSC variations can be expected due to the
propensity for larger particles to fall at a faster rate than
smaller particles as described by Stokes Law (American Society
of Civil Engineers, 2006). In the absence of coarser (sand size)
particles and with even moderate turbulence, the effect of
Stokes Law can be negligible, yielding a relatively uniform
distribution of silt- and clay-size particles in the cross section.
However, sand-size material in suspension coupled with
minimal to moderate turbulence usually results in a vertical
SSC gradient, with SSCs increasing with depth.
The effects of Stokes Law and other factors that result in
unequal mixing of suspended sediments – ergo the need for
depth-integrated sampling with FISP isokinetic samplers when
their deployment criteria are met – can be demonstrated using
suspended sediment SSC and PSD data collected at the Rio
Grande Floodway near Bernardo, New Mexico, USA (USGS
streamgage 08332010) on 4 May 1966 (Culbertson et al.,
1972). These data were used to generate lines of equal SSC in
the measured cross section (Figures 6(a) and (b)), as distri-
butions of silt-size material (0.004–0.062 mm) and sand-size
material (0.062–2 mm), respectively. The data were obtained
using a DH-48 suspended sediment sampler modified for
point-integration sampling, with SSCs reported per ranges of
phi diameters. With the exception of one measurement of
2 mm sand at an SSC of 31 mg l
1
, all of the remaining 63
point-integrated samples contained material finer than1 mm.
The silt–clay-size range in SSC data from the six verticals
sampled (Figure 6(a)) based on individual samples
(Culbertson et al., 1972; USGS, written communication, un-
dated) is 820–1080 mg l
1
, with a mean silt–clay SSC of
930 mg l
1
. The maximum silt–clay SSC deviation from the
mean value is approximately 14%. If one were to obtain an
open-bottle grab sample from near the water surface at the
rightmost sampled vertical (Figure 6(a)) and use the derived
SSC silt–clay SSC value – 860 mg l
1
– with river discharge to
compute suspended sediment discharge, the result would be a
suspended silt–clay discharge equal to approximately 92% of
the actual value. Alternately, if the silt–clay SSC value from an
unbiased sample pumped from near the right bank at a depth
of approximately 1 m, the resulting silt–clay SSC of
Measuring Suspended Sediment 175
Author's personal copy
1.52
1.83
(a)
(b)
0 3.0 6.1 9.1 12.2 15.2 18.3 21.3 24.4
0
Concentration at
surface (mg l−1)880
941
935 910 915
920
910 860 Discharge = 36.2 m3 s−1
Mean velocity = 1.34 m s−1
Median silt−clay
concentration = 930 mg l−1
907932942930
Mean concentration
in vertical (mg l−1)
0.30
0.61
0.91
Depth (m)
1.22
Distance (m)
1.52
1.83
03.0 6.1 9.1 12.2
Distance (m)
15.2 18.3 21.3 24.4
0
Concentration at
surface (mg l−1)
640
1220
770 690 880
1600
660 350 Discharge = 36.2 m3 s−1
Mean velocity = 1.34 m s−1
Mean sand concentration
(essentially all material
between 0.062−1.0 mm) =
1360 mg l−1
1190142014001350
Mean concentration
in vertical (mg l−1)
0.30
0.61
0.91
Depth (m)
1.22
Figure 6 SSCs as (a) silt–clay and (b) sand-size material in the Rio Grande Floodway near Bernardo, New Mexico, USA (USGS streamgage
08332010) interpolated from data collected on 4 May 1966 (Culbertson et al., 1972). Base graph developed by an anonymous hydrographer
before 1996. Photographs of the ISCO 3700 sampler with jumbo base are reproduced by courtesy of Teledyne Isco, r2012 (use of any trade
or firm names in this report is for identification purposes only and does not constitute endorsement by the US Government).
176 Measuring Suspended Sediment
Author's personal copy
900 mg l
1
would yield a suspended silt–clay discharge value
equal to approximately 103% of the actual value. Based on
standard protocols for deployment of depth-integrated
samplers (Edwards and Glysson, 1999), Topping et al. (2011,
p. 43), showed that the maximum total uncertainty in
silt–clay SSCs (four verticals and two transits/vertical) to be
approximately 3%. Hence, collection of a nonisokinetic
sample near the stream’s edge or via an unbiased pumping
sampler, with only silt–clay-size material in suspension may
provide a reasonable estimate (within several percent) of the
actual mean cross-section SSC value. This statement is ne-
cessarily predicated on the assumption that only silt–clay-
size material is in suspension in a river cross section at a
given time. However, it is important to note that only the
more experienced hydrographer can ascertain the relative
contribution of suspended sand to the total SSC with even a
modicum of certitude.
Figure 6(b) depicts a markedly different SSC cross-
sectional distribution for sand-size material (in this case, es-
sentially all material in the range of 0.062–1 mm). The sand-
size range in SSC data from the six verticals sampled
(Figure 6(b)) based on individual samples (Culbertson et al.,
1972; USGS, written communication, undated) is
255–8380 mg l
1
, with a mean sand SSC of 1360 mg l
1
. The
maximum sand SSC deviation from the mean value is ap-
proximately 616%. If one were to obtain an open-bottle grab
sample from near the water surface at the rightmost sampled
vertical and used the derived SSC value – approximately
350 mg l
1
– with water discharge to compute the suspended
sand discharge, the result would equal approximately 26% of
the actual value. Alternately, if the sand SSC value from an
unbiased sample pumped from the right bank at a depth of
approximately 1 m equaled approximately 700 mg l
1
, the
resulting suspended sand discharge value would equal ap-
proximately 194% of the actual value. In comparison, Topping
et al. (2011, p. 45) showed that the maximum total uncertainty
in sand SSCs using standard USGS protocols (Edwards and
Glysson, 1999) is approximately 7%.
The previous paragraphs provide the underpinning for the
answer to the question, ‘How does one measure and properly
weight velocity and SSC variations in the cross section to de-
rive the correct value for suspended sediment discharge in the
cross section at the time of the site visit?’ When substantial
variations in the distribution of flow and SSC in a cross section
exist, proper deployment of a suitable depth-integrating
sampler – the EDI or EWI methods – is required to obtain
samples representative of the flow-weighted sediment content
in the cross section. Such unmixed conditions are typical for
sand- and gravel-bed rivers at most flows; and for rivers with
finer grained or cohesive beds at medium-to-higher flows.
When it is unknown if substantial spatial SSC variations exist,
the same sampling approach should be used to preclude in-
accuracies resulting from nonflow weighting a sample from a
potentially heterogeneous mixture of suspended sediment.
Only when the experienced hydrographer determines un-
equivocally that the amount of sand in suspension is in-
significant with respect to total SSC – or when the limitations
of an isokinetic sampler are exceeded – samples should be
collected with other than isokinetic samplers and their ap-
propriate deployment techniques.
1.10.2.2.4 Automatic samplers
Some sediment-monitoring programs and studies include
sites where collection of sediment samples is required at a
frequency, at a time, and (or) under a set of conditions that
cannot be accommodated through manual sampling. Safety
considerations, remoteness or inaccessibility of site location,
flow conditions, operational costs, and other factors may
render manual collection of sediment and flow data at a site
impractical or impossible. In lieu of manual sampling,
automatic samplers collect a sample for subsequent analysis,
and possibly one or more surrogate technologies may be
deployed to accommodate sediment data collection needs at
some sites. As noted previously and described later, emerging
surrogate technologies for monitoring suspended sediment
are being tested and, in the case of turbidity, incorporated
into operational programs of the USGS (Rasmussen et al.,
2009).
Automatic samplers are useful for collecting suspended
sediment samples during periods of rapid water discharge
changes from storm runoff and in reducing the need for
manual measurements associated with intensive sediment-
collection programs (Federal Interagency Sedimentation Pro-
ject, 1981). However, under some circumstances, use of
automatic samplers to collect data can actually result in costs
greater than those for an observer (Johnson, 1997) at the same
site. Automatic samplers, and particularly pumping samplers,
often require more frequent site visits by the field personnel
than would be required at the conventional observer station
owing to their mechanical complexity, power requirements,
and limited sample capacity. Use of automatic samplers does
not preclude the need for collecting medium- and high-flow
cross-sectional samples. Additionally, use of automatic sam-
plers typically results in reduced data quality because the
samples are rarely, if ever, withdrawn isokinetically. This
conclusion is particularly true for automatic sample collection
from streams conveying high percentages of suspended sand-
size material.
The most commonly used automatic samplers are auto-
matic pumping samplers, which require power to obtain water
samples. Single-stage samplers, which rely on changes in
stream stage and (or) velocity to collect water–sediment
samples passively on the rising phase of a hydrograph, also are
available.
1.10.2.2.4.1 Automatic pumping samplers
Automatic pumping samplers generally consist of a pump,
bottle container unit, sample distribution, activation, and in-
take systems. Ideally, this combination of components should
be designed to meet the following criteria (Bent et al., 2003;
Edwards and Glysson, 1999):
1. Stream velocity and sampler intake velocity should be
equal to allow for isokinetic sample collection if the
intake is aligned into the approaching flow.
2. A suspended sediment sample should be delivered from
stream to sample container without a significant change
in SSC or PSD.
3. Cross contamination of samples caused by residual sedi-
ment in the sampler plumbing between sample collection
periods should be prevented.
Measuring Suspended Sediment 177
Author's personal copy
4. The sampler should be capable of sampling over the full
range of SSCs and particle sizes.
5. Samples should meet minimum sample analysis volume
requirements.
6. The inside diameter of the intake should be at least three
times the diameter of the largest particles sampled, al-
though small enough to maintain a mean sample velocity
that will substantially exceed the fall velocity of largest
particles.
7. The sampler should be capable of vertical pumping lift
heights of approximately 10 m from intake to sample
container for clear water (in most cases, the lift elevation
for river water is a function of sediment concentration,
water temperature, and atmospheric pressure).
8. The sampler should be capable of collecting a reasonable
number of samples – usually at least 24 – dependent on
the purpose of sample collection and the flow conditions.
9. Some provision should be made for protection against
freezing, evaporation, and dust contamination of col-
lected samples.
10. The sample container unit should be constructed to fa-
cilitate removal and transport as a unit.
11. The sampling cycle should be initiated in response to a
timing device, flow or stage change, or external signal
based on a set of criteria that maximizes the potential for
collecting samples at desired points over one or more
hydrographs.
12. The capability of recording the sample collection date and
time should be present.
13. The provision for operation using alternating current (AC)
power or direct current (DC) battery power should be
present.
Nearly all of the automatic pumping samplers in use at
present are commercially produced. The ISCO 6700 and
American Sigma 900 automatic pumping samplers, for ex-
ample, share various features for collecting water samples.
Both are computer controlled and capable of collecting up to
24 1 l samples based on time, flow, and (or) other user-
selected criteria. They use built-in peristaltic pumps and op-
erate on AC power or DC battery power. Both samplers feature
a back-flush cycle to preclude or minimize cross contamin-
ation between consecutively collected samples.
Neither sampler is capable of sampling clear, cold fresh-
water if the peristaltic pump is at a height greater than ap-
proximately 10 m above the water surface at normal sea level
barometric pressures. Cavitation can occur at smaller heads
with larger specific gravities associated with increasing SSCs
and (or) lower barometric pressures and (or) larger water
temperatures. Where lift requirements potentially exceed the
capacity of a sampler, an auxiliary pump may be used to pump
water to the sampler under a positive pressure.
Gray and Fisk (1992) described an automatic pumping
sampling system used to collect samples of highly concen-
trated and hyperconcentrated streamflows in Arizona and New
Mexico, USA. An auxiliary pump in a diving bell affixed at a
height of 1–2 m above that of the water surface at low-flow
pumped stream water to a gaging station. In the gaging sta-
tion, a commercial sampler modified to collect 9 l samples
periodically drew an aliquot of the pumpage from the
auxiliary pump via a Y connector in the intake line. A data
collection platform controlled collection of up to 24 samples
based on time, stage, and rate-of-stage-change criteria. The
data collection platform recorded hydrologic information and
data related to the number and times of samples collected
and periodically updated a USGS database via satellite. Such
capabilities are relatively common in 2012.
1.10.2.2.4.1.1 Installation and use criteria Physical and
fiscal criteria usually drive the decision to use a pumping
sampler for collection of sediment samples. Installation of
an automatic pumping sampler requires careful planning
including selection of the sampler site location and an
evaluation of available or newly collected data to maximize
the potential to obtain useful data from the pumped
samples.
Before installation of an automatic pumping sampler,
many of the problems associated with installing streamgaging
equipment must be addressed. In addition, specific data
concerning the sediment-transport characteristics at the pro-
posed sampling site must be obtained and evaluated before
emplacement of the sampler and location of the intake within
the streamflow. Logistically, the sample site must be evaluated
as to accessibility, availability of electrical power, access to a
bridge, cableway, or other platform from which to safely ob-
tain manual measurements and range in air temperatures
characteristic for the site. The availability of a local observer to
collect periodic reference samples also should be consi-
dered. The sediment-transport characteristics should include
detailed information on the distribution of SSCs and PSDs
throughout the sampled cross section over a range of
water discharges. Additional information is contained in
Section 1.10.1.2 (Table 1).
1.10.2.2.4.1.2 Placement and orientation of sampler intake
The primary concept to consider when placing a sampler
intake in the streamflow at a sample cross section is that
only one point in the flow is being sampled. Therefore, to
yield the most reliable and representative data, the intake
should be placed at the point where the SSC and PSD are
most representative of the mean SSC for the cross section
over the full range of flows. This idealistic concept has great
merit, but the mean cross-section SSC rarely, if ever, exists at
the same point under varying streamflow conditions. It is
even less likely that specific guidelines for locating an intake
under given stream conditions at one stage would produce
the same intake location relative to the flow conditions at a
markedly different stage. These guidelines would have even
less transfer value from cross section to cross section and
from stream to stream. For these reasons, some generalized
guidelines are outlined here and should be considered on a
case-by-case basis in placing a sampler intake in the
streamflow at any given cross section (Edwards and Glysson,
1999):
1. Select a stable cross section in a reach with reasonably
uniform depths and widths to maximize the stability of the
relation between the at-a-point SSC and the mean SSC in
the cross section. This guideline is of primary importance
178 Measuring Suspended Sediment
Author's personal copy
in the decision to use a pumping sampler in a given situ-
ation; if a reasonably stable relation between the at-a-point
SSC and mean SSC in the cross section cannot be attained
by the following outlined steps, an alternate location for
the installation should be considered.
2. Consider only the part of the vertical that could be sampled
using a standard US series depth- or point-integrating iso-
kinetic suspended sediment sampler, excluding the
unsampled zone, because data collected with a depth- or
point-integrating sampler will be used to calibrate the
pumping sampler.
3. Determine, if possible, the depth of the point of mean SSC
in each vertical for each size class of particles finer than
0.25 mm from a series of carefully collected point-inte-
grated samples.
4. Determine, if possible, the mean depth of occurrence of the
mean SSC in each vertical for all particles finer than 0.25 mm.
5. Use the mean depth of occurrence of the mean SSC in the
cross section as a reference depth for placement of the intake.
6. Affix the intake at the desired location in flow. It should
remain stable and free of debris throughout the range of
flows. Its depth should be sufficiently high to avoid inter-
ference by dune migration or contamination by bed ma-
terial but low enough to collect samples at the lowest flows
of interest.
7. Locate the intake in the flow at a distance far enough from
the bank to eliminate any possible bank effects. Never
place the intake in an eddy.
8. Place the intake in a zone of high velocity and turbulence
where the sediment will be relatively well mixed, the po-
tential for deposition on or near the intake is minimal, and
to provide for rapid elimination of any particles during a
purge cycle.
9. Situate the intake line to maintain a downward slope from
the sampler to the orifice without dips, which may retain
water and sediment after a pumping cycle.
Because of the generalized nature of these guidelines and
because selected guidelines may prove to be mutually ex-
clusive, it will rarely be possible to satisfy them all when
positioning a pumping sampler intake into stream flows. The
investigator is encouraged to identify those guidelines that are
most crucial to the success of the installation and endeavor to
satisfy them first.
The orientation of the pumping sampler intake nozzle
can drastically affect sampling efficiency. There are five ways
in which an intake could be oriented to the flow (see
Figure 7): (a) pointing directly upstream (Figure 7(a)), (b)
perpendicular to the flow and horizontal (Figure 7(b)), (c)
vertical with the orifice up (Figure 7(c)), (d) vertical with the
orifice pointing down (Figure 7(d)), and (e) pointing dir-
ectly downstream (Figure 7(e)). Of these five orientations,
(a), (c), and (d) should be avoided because of high sampling
errors and trash collection problems. Orientation (b), with
the nozzle positioned perpendicular to the flow and hori-
zontal, is the most common alternative used. The major
problem with orientation (b) is that sand-size particles may
not be adequately sampled. Orientation (e), pointing directly
downstream, may be advantageous over orientation (b)
(Winterstein and Stefan, 1986).Whentheintakeispointing
downstream, hydraulic theory supports the presence of a
small eddy at the intake, which captures sand particles and
thus enables withdrawal of a more representative sample of
thecoarseload(thefineloadtendstobemorerepresen-
tatively sampled regardless of intake orientation). Regardless
of the intake orientation selected, the ratios of SSCs repre-
sentative of the mean cross-sectional SSC and those from
pumped samples are required to define the sampling effi-
ciency over a range of flows with an emphasis on medium-to-
high flows.
1.10.2.2.4.1.3 Activation The availability of a micro-
processor as an integral part of the sampler, or as an external
controller, provides many options for controlling pumping
samplers that can be tailored to data collection requirements on
hand. As noted previously, Gray and Fisk (1992) described a
method for controlling an automatic water sampler based on
time, stage, and rate-of-stage-change criteria. Their technique is
designed to provide adequate definition of the flood hydro-
graph to make possible reliable computations of daily sediment
and associated solid-phase radionuclide discharges.
Lewis (1996) described a means for controlling an auto-
matic sediment sampler based on real-time turbidity meas-
urements. A technique for controlling an automatic water
sampler that provides unbiased estimates of suspended
sediment discharges, based on time-stratified sampling and
selection at list time, is described by Thomas (1985,1991) and
Thomas and Lewis (1993).
1.10.2.2.4.2 Single-stage samplers
Single-stage samplers were developed to meet the needs for
instruments useful in obtaining sediment data on streams
where remoteness of site location and (or) rapid changes in
stage make it impractical to use a conventional depth-inte-
grating sampler. They are generally less reliable, both in op-
eration and in data accuracy, than depth-integrating samplers.
However, even approximate information on the SSCs between
visits to the stream can be useful if nothing better is available
(Federal Interagency Sedimentation Project, 1961;Edwards
and Glysson 1999).
The US U-59 series single-stage samplers designed and
tested by the FISP consist of a 0.45-l milk bottle or other
sample container, a 4.7-mm inside diameter air exhaust, and a
4.7- or 6.4-mm inside diameter intake constructed of copper
tubing. Each tube is bent to an appropriate shape and inserted
through a stopper sized to fit and seal the mouth of the
sample container.
There are four models of US U-59 samplers (Federal
Interagency Sedimentation Project, 1961). That designated US
U-59A is designed for collection of silt- and clay-size sedi-
ments in low, less than approximately 0.7 m s
1
, stream vel-
ocities. Those designated US U-59B, US U-59C, and US U-59D
are for collection of sand-size and finer material in stream
velocities less than 1.0, 1.6, and 2.1 m s
1
, respectively. A US
U-59D single-stage suspended sediment sampler is shown in
Figure 8(a).
The US U-59 series of samplers obtain a sample on the
rising phase of the hydrograph from a point near the water
surface when the water level inside the intake tube reaches the
Measuring Suspended Sediment 179
Author's personal copy
weir height. As the sample siphons from the intake orifice into
the sample bottle, air from the sample bottle vents out of the
exhaust tube. The sampler is designed to cease filling when
the sample height reaches the inner exhaust tube orifice. The
sample velocity in the intake tube is a function of various
factors including stream velocity on the orifices, intake orifice
orientation, turbulence, and the presence of obstructions in
either tube.
The sampling operation just described is somewhat ideal-
istic because, in reality, the operation is affected by various
factors including flow velocity and turbulence. These factors
alter the effective pressure at the entrance and exit nozzles,
which in turn alters the sampler’s intake velocity.
The US U-59 sampler has many limitations with respect to
desirable sampling objectives. It is a type of point sampler
because it samples a single point in the stream at whatever
stage the intake nozzle is positioned when immersed in flow.
Its primary purpose is to collect a sample automatically, and
the US U-59 usually is used on flashy streams or other lo-
cations that are difficult to visit in time to manually collect
samples. Besides being automatic, the US U-59 is simple and
inexpensive compared to automatic pumping samplers; a
bank of them can be used to obtain a sample at various depths
during the rising hydrograph. However, despite these seem-
ingly important advantages, the US U-59 sampler has many
limitations. Following are the more important of these
limitations:
1. Samples are collected at or near the stream surface, so
that, in the analysis of the data, theoretical adjustments
for vertical distribution of SSCs or size are necessary.
2. Samples usually are obtained near the edge of the stream
or near a pier or abutment; therefore, theoretical adjust-
ments for lateral variations in sediment distribution are
required.
3. Even though combinations of size, shape, and orientation
of intake and air exhaust tubes are available, the installed
system may not result in intake ratios sufficiently close to
unity to sample sands accurately at parts of the runoff
hydrograph.
4. Covers or other protection from trash, drift, and vandalism
often create unnatural flow lines at the point of sampling.
5. Water from condensation may accumulate in the sample
container before sampling.
6. The sediment content of a sample may change during
subsequent submergence.
Sampler
Sampler
Sampler
Intake Intake
Flow
(a) (b)
(d)(c)
(e)
Flow
Flow
Sampler
Intake
Flow
Sampler Flow
Intake
Figure 7 Examples of pumping-sampler intake orientations. (a) Pointing directly upstream, (b) Perpendicular to flow and horizontal, (c) Vertical
with the orifice up, (d) Vertical with the orifice down, and (e) Pointing directly downstream.
180 Measuring Suspended Sediment
Author's personal copy
7. The device is not adapted to sampling on falling stages or
on secondary rises.
8. No specific sampler design is best for all stream
conditions.
9. The time and stage at which a sample was taken is often
unknown.
10. At high velocities, flow can circulate into the intake nozzle
and out the air exhaust. This can result in an increase in
the SSC of coarse material in the sample by at least an
order of magnitude.
Gray and Fisk (1992) developed a modified single-stage
sampler that provides a measure of protection against van-
dalism and flood damage while minimizing the potential for
water circulation (Figure 8(b)). Various single-stage samplers
are arranged vertically inside a protective polyvinyl chloride
(PVC) pipe capped at both ends. Screw cap 0.9 l bottles are
used to provide a larger sample volume and a more positive
seal. External air exhaust orifices extend through the top cap to
the highest elevation feasible for the site, reducing the po-
tential for its inundation. External intake orifices are set flush
with the exterior PVC pipe so that debris cannot snag on them.
A hinged lockable door provides access to the 0.9-l sample
bottles.
The investigator using single-stage samplers may find pro-
tective measures necessary to avoid blockage of intakes or air
exhausts due to nesting insects. In freezing temperatures,
precautions against sample container breakage due to expan-
sion of a freezing sample are advised. The percent sand-size
material should be analyzed for all samples collected by sin-
gle-stage samplers. This analysis will help identify instances of
bias in SSCs resulting from sample recirculation.
1.10.2.2.5 Sediment subsampling equipment
Samples of water–sediment mixtures are sometimes sub-
sampled or split into multiple aliquots to make possible dif-
ferent analytical determinations on the subsamples. The validity
of data obtained from subsamples depends on the compar-
ability of to the SSC of the original sample. Subsamples tend to
have larger constituent variances than the original and also may
be biased. Subsampling should be avoided unless it is necessary
to achieve the ends of the sampling program.
Before 1976, USGS guidelines on manual sample splitting
required compositing the water sample into a large, clean jug
or bottle, shaking it for uniform mixing and then withdrawing
the required number of samples (U.S. Geological Survey,
1976). In 1976, the 14-l Scienceware
s
Churn Sample Splitter
(churn splitter) (Figure 9(a)) was introduced to facilitate the
withdrawal of a representative subsample of a water–sediment
mixture (Capel and Larsen 1996; Lane et al. 2003). A fluor-
opolymer version of the churn splitter for trace element sub-
sampling is also available (Federal Interagency Sedimentation
Project, 2012). The cone splitter (Figures 9(b) and (c)), a
device developed to split water samples for suspended sedi-
ment and other water quality constituents into approximately
10 equal and representative aliquots, was introduced for wide-
scale use in 1980 (Capel and Nacionales, 1995;Capel and
Larsen, 1996).
Based on test results on the sediment-splitting efficiency of
the churn and cone splitters (U.S. Geological Survey, 1997), the
USGS has approved the use of the churn splitter for providing
subsampling when the original sample’s SSC is less than
1000 mg l
1
at mean particle sizes less than 0.25 mm. The cone
splitter is approved for providing subsamples at SSCs up to
10 000 m g l
1
at mean particle sizes less than 0.25 mm. The
Air exhaust
tubes
Bank of
modified
single-stage
samplers
102 mm
Inner diameter
polyvinylchloride
(PVC) pipe
(a) (b)
Internal air
exhaust tube
orifice Internal air
intake tube
orifice
External air
exhaust tube
orifice
Weir elevation
of intake tube
External intake tube
orifice flush
to outside of PVC pipe
Bottle cap affixed
to flange
1−l
glass bottle
Figure 8 (a) US U-59D single-stage suspended sediment sampler and (b) Modified single-stage sampler. Reproduced from Gray, J. R. and Fisk,
G. G. (1992). Monitoring radionuclide and suspendedsediment transport in the little Colorado river basin, Arizona and New Mexico, USA. IAHS
Publication No. 210, pp. 505–516. Wallingford, OX: IAHS Press, Institute of Hydrology.
Measuring Suspended Sediment 181
Author's personal copy
test data suggest that the cone splitter’s acceptable SSC range
exceeds 10 000 mg l
1
and may be as large as 100 000 mg l
1
.
1.10.3 Surrogate Suspended Sediment Measuring
Techniques
The scope of this section includes an introduction to advances
in suspended sediment surrogate technologies, calibration
issues, data acceptance criteria, and descriptions of four surro-
gate technologies that are being or may be used in large-scale
operational monitoring programs. Descriptions of the theory,
applications, some advantages, limitations, and costs of each
surrogate technology are presented and compared. This is fol-
lowed by an evaluation of the efficacy of each technology for
riverine suspended sediment transport monitoring.
All of the surrogate methods described herein must be
calibrated using representative physical samples to obtain
mass SSC and in some cases PSD, with the possible exception
of the manually deployed version of the described laser
technology. The reader is thus encouraged to become familiar
with the information under Section 1.10.2. Those interested in
surrogate technologies for measuring bedload may opt to ac-
cess Gray et al. (2010) and (or) Gray and Gartner (2010a); and
for bed material and other surrogate technologies, Gray
(2005).
The prospect of continuous and quantifiably accurate
monitoring of riverine suspended sediment transport is a
revolutionary concept when considered from an operational
perspective. The benefits of such applied capability could be
enormous, providing for safer, more frequent and consistent,
arguably more accurate, and ultimately less expensive fluvial
sediment data collection for use in managing the world’s
sedimentary and related resources.
1.10.3.1 Overview of Selected Suspended Sediment
Surrogate Measurement Techniques, Metrics,
and Requirements
Advances in suspended sediment surrogate techniques for
transport monitoring programs in rivers show varying degrees
of promise toward supplementing and (or) supplanting tra-
ditional data collection methods based on routine collection
of physical samples and subsequent laboratory analyses.
Research-based and commercially available technologies op-
erating on turbidity (transmissometry and nephelometry)-,
laser-, and digital optics; pressure-difference; and acoustic
principles have been or remain the foci of field or laboratory
tests by the USGS and other organizations. Advantages and
limitations associated with each suspended sediment-surro-
gate technology should be factored into monitoring program
network design to select the most appropriate technology.
General considerations should include deployment site
physical, flow, sedimentological characteristics, and moni-
toring objectives. Examples of specific factors that can limit or
enhance the efficacy of a surrogate technology include cost
(purchase, installation, operation, and data analysis), re-
liability, robustness, accuracy, measurement volume, sus-
ceptibility to biological fouling, volumetric versus mass-SSC
determinations, and suitability to the range of in-stream mass
SSCs and PSDs.
Selected sediment-surrogate technologies have been shown
to provide the types, quality, and density of fluvial sediment
Flange
Flange
Funnel
Cylindrical
reservoir
Stand pipe
Cone splitte
r
housing
Drawing not to scale
Outlet ports
(a) (b) (c)
Figure 9 The14-l churn splitter (a), and the cone splitter ((b) and (c)). Permission to reproduce the photograph of the Scienceware
s
Churn
Sample Splitter is by courtesy of Bel-Art Products (2012). Churn sample splitter. Available at http://www.belart.com/shop/advanced_search_
result.phpkeywords=churn þsplitter&x=6&y=3 (accessed on 18 July 2012) and the photograph and diagram of the cone splitter is reproduced by
courtesy of Geotech Environmental Equipment, Inc. (2012). Dekaport sample splitter. Available at http://www.geotechenv.com/pdf/ground_
water_sampling_equipment/dekaport_sample_splitter.pdf (accessed on 18 July 2012).
182 Measuring Suspended Sediment
Author's personal copy
data needed to improve suspended sediment discharge com-
putations in research settings and in limited operational de-
ployments. Potentially useful instruments and methods for
inferring the physical characteristics of fluvial sediments
(Wren et al., 2000;Gartner et al., 2003;Bogen et al., 2003;
Gray et al., 2003;Gray, 2003b;Gray and Glysson, 2004;Gray,
2005;Topping et al., 2007;Gray and Gartner, 2009,2010a,
2010b, 2010c; Gray et al., 2010) are being developed and tested
worldwide. For example, through the informal USGS Sedi-
ment Monitoring Instrument and Analysis Research Program
(Gray, 2003a,2003b;Gray and Simo
˜es, 2008), the USGS and
collaborators in other government agencies, academia, and
the private sector are testing several instruments for measuring
SSCs and, in some cases, PSDs. To make the transition from
research to operational monitoring applications, these new
technologies must be rigorously tested with respect to accur-
acy, robustness, and reliability in different physiographic and
(or) laboratory settings as appropriate, and their performances
must be compared to data obtained by the aforementioned
traditional methods and to available quality control data. In
most cases, performance comparisons should include con-
current collection of data by traditional and new techniques
for a sufficient period – probably years – and in a variety of
river types and flow conditions to optimize calibration and
identify potential bias between the old and new technologies.
Calibrations must cover the range of SSC and PSD ob-
served in the system and, once established, must be verified
periodically to test for temporal changes. Also, the in situ
technologies presented herein measure a fixed point in a
stream and require periodic site-specific calibrations to infer
the sedimentary characteristics representative of the entire
channel cross section or reach segment. This requirement is
anticipated to be substantial for new river-monitoring appli-
cations but may diminish as calibration data accumulate.
None of the technologies represents a panacea for sedi-
ment monitoring in all rivers under all flow and sediment-
transport conditions. However, with careful matching of
surrogate monitoring technologies to selected river reaches
and objectives, it is becoming possible to remotely, continu-
ously, and accurately monitor SSCs and suspended sediment
discharges (and in some cases, PSDs) in a variety of river types,
flow conditions, and sedimentological regimes. In some cases,
the computed SSC values and perhaps other data types may
be qualified with estimates of uncertainty (U.S. Geological
Survey, 2005).
1.10.3.1.1 Calibration of suspended-surrogate metrics to
representative SSC
All of the surrogate methods described use metrics that must
be calibrated using physical samples representative of the cross
section to obtain mass SSC or PSD. The reliability and efficacy
of data produced by a sediment-surrogate technology are
predicated on the adequacy of its calibrations. Three general
types of calibrations may be used, depending on the metric of
interest. These are calibration of:
1. the surrogate instrument metric to some standard such as
blank water, a turbidity standard, or a single- or double-
blind standard reference sediment sample (U.S. Geological
Survey, 1998b);
2. the surrogate metric to representative physical environ-
mental samples of SSC and/or PSD; and
3. the monitored volume to cross-section average volume.
Calibration of the surrogate instrument metric to a stand-
ard or blank water provides a test for instrument drift, fouling,
or malfunction and typically follows manufacturer recom-
mendations. Calibration to measured SSC and/or PSD is a
primary analytical component of using any surrogate par-
ameter and requires concurrent physical samples and surro-
gate measurements over the range of sediment conditions for
the period of monitoring. The calibration will typically use
statistical methods and inclusive of model uncertainty, which
should be carefully documented. The surrogate measurements
may be calibrated to representative physical samples taken
from the stream cross section or may be calibrated to samples
collected at the point or within the volume being measured by
the surrogate. In the latter case, the point-estimated SSC must
be calibrated to the cross-sectionally averaged SSC.
Calibration of surrogate metrics to PSDs can be more dif-
ficult than to SSCs because the analytical results of PSD ana-
lyses are somewhat dependent on the methods used to
measure and analyze the PSD and on handling of water
samples between the stream and location of analysis. Gravi-
metric PSDs determined by sieving and/or settling velocity
methods would not be expected to be the same as volumetric
PSD obtained by laser diffraction methods. PSD determined
by multifrequency acoustics methods will depend on the fre-
quencies used in the analysis. Furthermore, in situ PSD would
not be expected to be the same as laboratory PSD where
samples are affected by handling and potential disaggregation
of flocs due to sonic or chemical dispersant. Nonetheless, for
data continuity in sediment studies, the results need to be
quantitatively comparable. Preliminary results from recent
studies indicate that PSD can be quantified or qualified using
surrogate technologies, although the results may require
important metadata on assumptions and limitations.
1.10.3.1.2 Acceptance criteria for SSC and PSD data
produced by suspended sediment-surrogate
technologies
Validation of a suspended sediment-surrogate technology
requires evaluation criteria and a well-conceived and well-
administered testing program (Gray et al., 2002;Gray and
Glysson, 2005;Gray et al., 2010). Following are some quali-
tative criteria for selecting and deploying a surrogate
technology.
•Capital and operating costs should be affordable with re-
spect to the objectives of the monitoring program in which
the surrogate instrument is deployed.
•The technology should be able to measure SSCs, and, if
sought, PSDs, throughout the range of interest (but not
necessarily throughout the entire potential environmental
range).
•The equipment should be robust and reliable, that is,
prone to neither failure nor signal drift.
•The method should be sufficiently simple to deploy and
operate by a field technician with a reasonable amount of
appropriate training.
Measuring Suspended Sediment 183
Author's personal copy
•The derived data should be relatively simple and straight-
forward, or rendered so with advanced computational
tools, to use in subsequent computations and (or) ac-
companied by standard analytical procedures as compu-
tational routines for processing the data.
Quantitative criteria for acceptable accuracies of the de-
rived data are difficult to develop for all potential applications,
in part, because of substantial differences in river sedimentary
and flow regimes. For example, accuracy criteria for rivers
transporting mostly silt and clay can and should be set more
stringently (intolerant of larger-magnitude uncertainties) than
those for rivers that transport comparatively large fractions of
sand. However, there is a clear need for consistency in SSC
criteria on the part of instrument developers, marketers,
and users.
To this end, quantitative acceptance criteria developed for
PSD and SSC data produced by a laser diffraction instrument
(Gray et al., 2002) have been generalized for evaluating data
from other suspended sediment surrogate instruments. At least
90% of PSD values between 0.004 and 0.5 mm median
diameter are required to be 725% of true median diameters.
Absent a more rigorous evaluation, this criterion has been
applied to all particle sizes in suspension (Gray et al., 2010).
SSC acceptance criteria range from 750% uncertainty at
lowest SSCs to 715% uncertainty for SSCs exceeding
1000 mg l
1
. The criteria presented in Table 3 are adapted
from Gray et al. (2002,2008), and Gray et al. (2010).
These criteria pertain solely to the performance of a sur-
rogate technology within its physical realm of measurement.
Routine calibrations to correlate instrument signals to mean
cross-sectional SSC values are required for all of the in situ
instruments presented herein.
1.10.3.2 Technological advances in suspended sediment-
surrogate monitoring
The following sections in this sub-chapter describe theoretical
principles (Gray and Gartner, 2004,2010b), capital costs, ex-
amples selected field applications, and advantages and limi-
tations of four suspended sediment-surrogate technologies
that cover a range of transport conditions that are being used
or considered promising by the USGS. Capital cost infor-
mation provided herein is in 2012 US dollars relative to the
approximate purchase price of an in situ turbidimeter with
sonde (sensor), wiper, and controller, totaling approximately
$5500 US in 2012.
1.10.3.2.1 Turbidity
1.10.3.2.1.1 Background and theory
Turbidity is an expression of the optical properties of a sample
that causes light rays to be scattered and absorbed rather than
transmitted in straight lines through the sample (Ziegler,
2003;Anderson, 2005). Turbidity was the most common
surrogate used for determining water clarity and computing
SSCs in US rivers in 2003 (Pruitt, 2003) and remained so
in 2012.
A number of commercially available optical instruments
operate on one of two physical principles: transmissometry
and nephelometry. Transmissometers employ a light source
beamed directly at a light detector. The instrument measures
the fraction of visible light from a collimated light source
(typically at approximately 660 nm) that reaches the detector.
The fraction of light reaching the detector is converted to a
beam attenuation coefficient, which is related to SSC.
Nephelometry is the measurement of light scattering usu-
ally with a light detector at 901from the incident light
(adapted from U.S. Environmental Protection Agency, 1999)
in visible or infrared (IR) spectra. Most laboratory turbidi-
meters measure 901scattering. According to D&A Instrument
Company (1991), optical backscatterance (OBS) instruments
collectively are a type of nephelometer designed to measure
backscattered (1401–1651) IR in a small (concentration-
dependent) volume on the order of a few cubic centimeters.
Transmittance and scatterance are functions of the number,
size, color, index of refraction, and shape of suspended par-
ticles (Conner and De Visser, 1992;Sutherland et al., 2000).
Figure 10 shows examples of five types of nephelometry
sensors.
A wide variety of turbidimeters are available, most of which
operate on the nephelometric principle. For example, Landers
(2003) described bench tests as part of a workshop at which
variances in measurements from nine different types of tur-
bidimeters using blind reference samples were evaluated. One
instrument that was first described in the early 1980s
(Downing et al., 1981;Downing, 1983) and is now widely
used for in situ applications is the OBS-3 (originally manu-
factured by the D&A Instrument Company, now by Campbell
Scientific, Inc.).
Turbidity instruments lack moving parts (unless outfitted
with optical wipers) can be deployed in situ and provide rapid
sampling capability. Site-specific empirical calibrations are
required to convert measurements to reliable cross-sectional
SSC estimates. The technology is relatively mature, and has
been shown to provide reliable data at a number of USGS
streamgages (Uhrich, 2002;Schoellhamer and Wright, 2003;
Melis et al., 2003;Uhrich and Bragg, 2003;Wright and
Schoellhamer, 2005;Rasmussen et al., 2005;2009;Schoell-
hamer et al., 2007;Buchanan and Morgan, 2011) and other
sites (Pratt and Parchure, 2003;Lewis, 2002). As noted pre-
viously, the cost of an in situ turbidimeter with sonde (sensor),
wiper, and controller in 2012 is approximately $5500. The
purchase price of an OBS without a wiper but with cable is
about equal to the fully equipped in situ nephelometric
turbidimeter cost.
Maximum SSC limits for these instruments depend in part
on PSDs. The OBS has a generally linear response at SSC less
than approximately 2000 mg l
1
for clay and silt and
Table 3 Acceptance criteria for suspended sediment concentration
(SSC) data. The data are considered acceptable when they meet these
criteria 95% of the time with any deviant values randomly distributed
over the range in SSCs
SSC Acceptable uncertainty
Minimum, mg l
1
Maximum, mg l
1
Percent, 7
0o10 50
10 o100 50–25 computed linearly
100 o1000 25–15 computed linearly
1000 – 15
184 Measuring Suspended Sediment
Author's personal copy
10 000 mg l
1
for sand (Ludwig and Hanes, 1990), although
Kineke and Sternberg (1992) described the capability to
measure SSC up to approximately 320 000 mg l
1
(in the
nonlinear region of the OBS response curve). The specification
sheet for the OBS-5 þmanufactured by Campbell Scientific
(2008) lists an applicable range of up to approximately
500 000 mg l
1
(specific gravity 1.3). The upper SSC limit for
transmissometers depends on optical path length but may be
as low as approximately 50 000 mgl
1
(D&A Instrument
Company, 1991). Thus, transmissometers are more sensitive at
low SSC, whereas optical backscatter sensors have superior
linearity in turbid water (Downing, 1996). In general, the
wider a turbidimeter’s turbidity-measurement range, the less
precise the within range derived turbidity data and vice versa.
Biological fouling of sensor optical windows can be
problematic, particularly in warmer, nutrient-rich waters.
Biological fouling results in a tendency for the output to shift
from the calibration curve to spuriously larger values over
timescales of days or more, particularly in warmer, micro-
biologically active waters. Commercially available mechanical
wiper systems available with some sensors may alleviate this
problem.
Because of the relation between OBS signal response and
PSD, OBS (like all single-frequency optical instruments) is
best suited for application at sites with relatively stable PSDs.
For a given mass SSC, OBS response increases with decreasing
particle size (Conner and De Visser, 1992;Downing, 1996;
Sutherland et al., 2000). OBS signal response is minimally
affected by changes in PSD in the range of 200–400 mm and
greatly affected by changes when particles are smaller than
approximately 44 mm(
Conner and De Visser, 1992). Conner
and De Visser (1992) caution against using OBS in environ-
ments where changes in PSDs occur and particle sizes are less
than 100 mm. Additionally, the OBS signal can vary as a
function of particle color. Sutherland et al. (2000) found a
strong correlation between observed and predicted OBS
measurements of varying SSC and ratios of black and white
suspended sediment. They found the smallest OBS signal-gain
response for black sediment and the largest for white sedi-
ment, with responses from other colors falling between. They
suggest that the level of blackness of particles acts to absorb
the near-IR signal of the OBS, thus modifying its output.
Hence, caution should be exercised in deployments under
varying particle-size and particle-color conditions, unless the
instrument is recalibrated for ambient conditions.
1.10.3.2.1.2 Example field evaluation
Continuous turbidity measurements have been shown to
provide reliable continuous SSC values with a quantifiable
uncertainty at the USGS streamgage on the Kansas river at
DeSoto, KS, USA, since the 1990s. Simple linear regression
analysis explained in Christensen et al. (2000) was used to
develop a site-specific univariate model using turbidity to
compute SSC (Figure 11). The model explains approximately
93% of the variance in SSC. Continuous suspended sediment
discharge values computed from the model and subdaily time
series water discharge data are available online (U.S. Geo-
logical Survey, 2005). The advantages of regression-based
estimates using continuous turbidity measurements over dis-
crete sample collection are that regardless of flow conditions,
SSC and sediment-discharge values are obtained essentially
continuously at the interval in which water discharges are
recorded.
Some researchers are using turbidity in conjunction with
other variables to compute time series of SSC. Jastram et al.
(2009) had monitored turbidity at a USGS streamgage on the
James river at Cartersville, VA, USA, since 2003. Figure 12
shows a time series of computed SSC, sampled SSC, and
streamflow data for this station from 22 October 2006 to 30
April 2007. The continuous SSC data were computed by using
(a)
(b) (c)
(e)
(d)
Figure 10 Photographs showing nephelometry sensors: (a) YSI model 6136, (b) Hydrolab turbidity sensor with wiper, (c) Forrest Technology
Systems model DTS-12, (d) D&A Instrument Company model OBS 3 þ, and (e) Hach OptiQuant with wiper.
Measuring Suspended Sediment 185
Author's personal copy
a multiple regression technique from square root-transformed
time series data describing turbidity, streamflow, and water
temperature. The model explains approximately 97% of the
variance in SSC.
Schoellhamer et al. (2007) described a multistation, mul-
tiyear field investigation to continuously monitor SSC in
California’s San Francisco Bay and Delta system that began in
1991. As of 2010, the program consisted of 13 monitoring
10
10 100
Turbidity (NTU)
SSC = 1.797 NTU0.905
R2 = 0.93
1000 10 000
100
1000
10 000
SSC (mg l−1)
Figure 11 Linear regression comparing field turbidity in nephelometric turbidity units and instantaneous SSCs in milligrams per liter for the
Kansas river at DeSoto, KS, USA, 1999 through 2002. Reproduced from Gray, J. R., Gooding, D. J., Mellis, T. S., Topping, D. J. and Rasmussen,
P. P. (2003). U.S. Geological Survey suspended-sediment surrogate research, Part II: Optic technologies. In: Jane W. and Judy P. (eds.)
Proceedings of the Virginia Water Research Symposium 2003, Water Resource Management for the Commonwealth. Virginia Polytechnic Institute
and State University, Blacksburg, 8–10 October, pp. 58–64. Available at http://vwrrc.vt.edu/pdfs/proceedings/
2003WaterResearchSymposium_proceedings.pdf (accessed on 13 June 2013).
0
100
200
SSC (mg l−1)
300
400
500
600
700
Estimated SSC Observed SSC Streamflow
10/22 11/10 11/29 12/18 1/6 1/25 2/13
Date (2006−07)
3/4 3/23 4/11 4/30
10
100
Streamflow (m3 s−1)
1000
10 000
Figure 12 Time series plot of continuous SSCs (computed by multiple linear regression from square root-transformed time series of turbidity,
streamflow, and water temperature data), sampled SSCs in milligrams per liter, and streamflow in cubic meters per second for the James river at
Cartersville, VA, USA, 22 October 2006–30 April 2007. Reproduced from Jastram, J. D., Moyer, D. L. and Hyer, K. E. (2009). A comparison of
streamflow-based and turbidity-based estimates of suspended sediment concentrations in three Chesapeake bay tributaries. U.S. Geological
Survey Scientific Investigations Report 2009-5165. Available at http://pubs.usgs.gov/sir/2009/5165/ (accessed on 14 June 2012).
186 Measuring Suspended Sediment
Author's personal copy
stations (with turbidity sensors at multiple depths) at which a
cumulative 280 years of sensor data have been collected.
Turbidity sensors are calibrated with water samples collected
by a Van Dorn sampler after sensor cleaning at each site. As an
example of data quality, results from the 2008 water year
(1 October 2007–30 September 2008) from 9 records at 5
stations in San Francisco bay had an average of approximately
82% data considered acceptable (after deletion of records
compromised by biological fouling and other factors)
(Buchanan and Morgan, 2011). Improved self-cleaning sensors
have increased the fraction of acceptable data from approxi-
mately 50% in the 1990s. Calibration curves indicated gen-
erally good correlations between SSC samples and turbidity
readings. Before October 1997, calibrations were performed
using ordinary least-squares regressions; starting with water
year 1998, a robust, nonparametric, repeated median method
was used (see Buchanan and Morgan, 2011, for a description
of the method). San Francisco bay sensors are calibrated to
point measurements and Delta sensors are calibrated to dis-
charge-weighted, cross-sectionally averaged SSC values. Sus-
pended sediment discharge is determined by multiplying the
discharge-weighted, cross-sectionally averaged SSC by water
discharge, accounting for tide-driven bidirectional flow
(Wright and Schoellhamer, 2005).
1.10.3.2.1.3 Advantages and limitations of turbidity
Advantages:
•As the most ubiquitous of the field-deployed surrogate
technologies, results from a large number of field settings
are available for evaluation.
•The technology is relatively mature and reliable.
•Calibration techniques are documented and largely
straightforward.
•At a cost for a fully equipped turbidimeter of approxi-
mately $5000, this is one of the more affordable sediment-
surrogate technologies.
Limitations:
•The at-a-point turbidity time series data may not be rep-
resentative of the sedimentary conditions in the river cross
section.
•Saturation of the turbidimeter signal can occur resulting in
erroneous (constant) values for all SSC values that exceed a
maximum value.
•Biological fouling or damage to optical windows may
require frequent site visits to service the instrument.
•Instrument response to grain size, composition, color,
shape, and coating can be variable and, hence, can reduce
the accuracy of derived SSC values without additional
calibration.
•A lack of consistency in measurement characteristics
among commercially available instruments impinges on
the comparability of turbidity measurements made with
different types of instruments.
1.10.3.2.2 Laser diffraction
1.10.3.2.2.1 Background and theory
Laser diffraction instruments exploit the principles of small-
angle forward scattering to infer volumetric PSDs. At small
forward scattering angles, laser diffraction by spherical par-
ticles is essentially identical to diffraction by an aperture of
equal size (Agrawal and Pottsmith, 1994). Thus, this method
of determining PSDs (and, by inference, volumetric SSC val-
ues) is mostly insensitive to changes in particle color or
composition, although departure from sphericity produces a
bias in the computed PSD compared to that for sieving. For
example, Agrawal et al. (2008) had shown that natural par-
ticles measured by laser diffraction are inferred to be ap-
proximately 20–40% larger than sieved spheres of known
sizes. They reported a new method that permits computation
of PSDs of equivalent sieved sizes.
At present, an in situ version of this type of instrument is
commercially available from only one manufacturer (Sequoia
Scientific, Inc., 2012). First used in the early 1990s (Agrawal
and Pottsmith, 1994), the present version of a laser diffraction
instrument that can be deployed unattended to provide a time
series of PSD and volume SSC values is the laser in situ scat-
tering and transmissometry (LISST)-100X, shown in
Figure 13(a) (Sequoia Scientific, Inc., 2012). LISST instru-
ments use a 32-ring detector to sense light scattered by par-
ticles illuminated by a laser beam into small forward angles.
These data are inverted to determine PSDs in 32-size classes
between 1.25–250 and 2.5–500 mm (limits that are not fun-
damental and which may be expanded in the future). The
standard sample path of this device is a cylindrical volume
with a diameter of approximately 6 mm and a length of
50 mm (essentially a point measurement). An isokinetic,
cable-suspended, streamlined version of the LISST-100X, the
(a) (b)
Figure 13 LISST: (a) an LISST-100X in situ instrument and (b) an LISST-SL (streamlined) manually deployable instrument. Photographs
courtesy of Sequoia Scientific, Inc.
Measuring Suspended Sediment 187
Author's personal copy
LISST-SL shown in Figure 13(b), features the capability of
real-time velocity measurement. This measurement is, in turn,
used to control a pump to withdraw a filament of water and
route it through the laser beam at the ambient current velocity
(Gray et al., 2002;Gray and Gartner, 2004;Gray et al., 2004;
Agrawal and Pottsmith, 2006;Gray and Gartner, 2009;Gray
et al., 2010). The performance of the LISST-SL has been
evaluated by the Federal Interagency Sedimentation Project
(2012) in a laboratory, a flume, and by USGS researchers in
rivers in Washington and Illinois, USA. The purchase price of
one of the LISST instruments (in situ or manually deployed)
described in this section ranges from approximately 2–6 times
that for a fully equipped turbidimeter, depending on the
instrument of interest.
Because an LISST determines PSD for all measurements, it
is not subject to potential inaccuracies in the calculation of
SSC associated with single frequency (optical and acoustic)
instruments that occur due to changes in PSDs as long as
particle sizes fall within the instrument measurement range
(Agrawal and Pottsmith, 2000).
Field and laboratory tests have shown the LISST-100X to be
capable of determining PSDs of natural materials and the size
of monosized particle suspensions with an accuracy of ap-
proximately 10% (Traykovski et al., 1999;Gartner et al.,
2001). The LISST-100X can also be used to determine mass
SSC from volume SSC if mean particle density is known from
field calibrations or some other means (Gartner et al., 2001).
Size-dependent mass density (as can happen due to floccu-
lation) can be used to convert volume to mass SSC.
As is the case with all types of in situ optical instruments,
biological fouling can degrade measurements. Antifouling
shutters for some LISST instruments are available from the
manufacturer. In addition, the technology has an SSC range
limitation associated with multiple scattering in the presence
of high SSCs. The LISST-100X requires approximately 30% or
more laser optical transmission. The range limitation is a
function of the laser path length, PSDs, and SSCs. For SSC, the
usable limits range from tenths of a g l
1
for small particle
sizes to several thousand mg l
1
for larger particle sizes. Op-
tical blocks that reduce the path length of LISST-100X by 50%,
80%, or 90% are available, reducing the optical path from the
standard 5 cm to 5 mm can extend measurement limits ap-
proximately from 500 to 5000 mg l
1
for 25 mm particles
(Yogesh Agrawal, Sequoia Scientific, Inc., written communi-
cation, 2008; Agrawal et al., 2008). A prototype LISST-Infinite
has been tested by the USGS (Konrad et al., 2006) for appli-
cation in very high SSCs. The system pumps a water–sediment
mixture sample to the instrument and then uses automated
multistage dilution (if necessary) before measuring PSDs and
SSCs with built-in LISST-100X optics. However, as is the case
with any nonisokinetic point sampler, pumping a water
sample from a fixed point in the channel may result in a PSD
and an SSC that is nonrepresentative of that in the cross sec-
tion. The same is true for the LISST-Streamside, which was
developed specifically for monitoring applications such as for
real-time continuous sediment monitoring in shallow streams
and rivers, including periods of storm runoff.
A somewhat simpler and less expensive version of the
LISST-100X instrument – the LISST-25X – measures mean SSC
and a mean particle size (Sauter mean size) in two size classes
(0.0025–0.062 and 0.062–0.5 mm) (Sequoia Scientific, Inc.,
2012). This device is also based on the same small-angle
scattering principles as the LISST-100X, but it obtains the SSC
through a weighted summation of the output of ring detectors,
bypassing the inversion step from rings output to PSD. The
cost of the LISST-25X is about double that of a fully equipped
in situ turbidimeter.
1.10.3.2.2.2 Example field evaluation
Laser sensors are being investigated as an alternative moni-
toring protocol for tracking reach scale suspended sediment
supply at a USGS streamgage on the Colorado river near
Grand Canyon, AZ, USA, located 164 km downstream from
Glen Canyon dam (Topping et al., 2004). A canyon wall-
mounted LISST-100B provides continuous suspended sedi-
ment transport data (SSCs and PSDs in the range of
0.0012–0.25 mm) that may reduce uncertainty in estimates of
the transport of sand and finer material.
An example of data collected by a LISST-100B at a fixed-
depth, canyon wall mounted installation on the Colorado
river is shown in Figure 14. Data were obtained averaging 16
measurements at 2-min intervals during a 24-h deployment in
July 2001. The time series of 720 LISST at-a-point measure-
ments are compared with cross-sectional data obtained by US
D-77 isokinetic bag sampler (the US D-77 bag sampler has
been phased out of use by the U.S. Geological Survey (2002))
concurrent with some of the LISST measurements using
techniques described by Edwards and Glysson (1999) and
Nolan et al. (2005) (more recent calibrations are performed
using the US D-96 bag sampler). In addition to accurately
tracking sand-size SSCs, the LISST-100B also recorded the in-
crease of variance in the SSCs of sand-size particles expected
with increasing flows; peak sand-size SSC values ranged up to
150 mg l
1
(Figure 14).
The FISP has performed laboratory bench tests of the
sedimentological characteristics of a LISST-SL. The range in
SSC used in tests was 10–3000 mg l
1
. Material used for test-
ing was primarily less than 0.15 mm, although some tests
included coarser material that was difficult to keep suspended
in the test system. Sedimentological results from these LISST-
SL tests fall within the acceptable uncertainty values for the
corresponding SSC levels shown in Table 2 (Broderick Davis,
FISP, written communication, 2008).
1.10.3.2.2.3 Advantages and limitations of laser-optic
technology
Advantages:
•The instrument provides in situ or real-time PSD meas-
urements in 32 size classes.
•Calculated volumetric SSC values are not affected by
changes in PSD.
•A manually deployed isokinetic version of the LISST tech-
nology is available.
Limitations:
•In situ laser measurements may not be representative of
sedimentary conditions in the river cross section.
188 Measuring Suspended Sediment
Author's personal copy
•Saturation of the laser-optic signal can occur at a SSC level
of about half of that at which a standard in situ turbidi-
meter saturates.
•As with turbidimeters, frequent field visits may be required
to clean the optics if antifouling shutters are not used.
•The cost of a LISST instrument is 2–6 times that for a fully
equipped in situ turbidimeter depending on the instrument
of interest.
1.10.3.2.3 Pressure difference
1.10.3.2.3.1 Background and theory
The pressure-difference technique for monitoring SSC relies
on simultaneous measurements from exceptionally sensitive
pressure transducer sensors arrayed at different fixed depths in
a water column. The difference in pressure readings is con-
verted to a water density value from which SSC is inferred after
correcting for water temperature (dissolved-solids concen-
trations in freshwater systems are rarely of consequence in the
density computation).
The following equation describes this conversion for a
2-orifice instrument (refer to Figure 15(d)):
g¼ðr1r2Þ=ðz1z2Þ½5
where gis the specific weight of the fluid, which is an analog for
SSC in many freshwater systems, r
1
,r
2
the simultaneous pres-
sure measurements at orifices 1 and 2, respectively, and z
1
,z
2
the depths below water surface of orifices 1 and 2, respectively.
Implicit assumptions are that the same water surface lo-
cation is measured by all sensors, and that the density of the
water–sediment mixture above the lowest sensor is more or
less constant at a given instant. The technology has both la-
boratory and field applications (Lewis and Rasmussen, 1999).
One of the first uses of the pressure-difference technique was
for monitoring the density of crude oil in pipes (William
Fletcher, D&A Associates, oral communication, 1999).
The instrument produced by Design Analysis Associates
(2012) and evaluated by the USGS in Puerto Rico and Ari-
zona, USA – the Double Bubbler Differential Instrument
(Double Bubbler) – was based on a dual sensor configuration
(Figure 15). Although the Double Bubbler is no longer
manufactured, an Accubar
s
Constant Flow Dual Orifice
Bubble Gauge/Recorder produced by Sutron Corporation
(2012) is available in two- and three-sensor configurations.
The accuracy of the three-sensor version is advertised as thrice
that of the dual sensor unit.
The technique has been applied in the laboratory with
promising results of better than 3% accuracy (543714 mg
l
1
) for determining mass concentrations of suspensions of
glass microspheres (Lewis and Rasmussen, 1999). However,
application of this technique in the field can be complicated
by small signal-to-noise ratios associated with small
SSCs, turbulence, substantially large dissolved solids concen-
trations, and (or) water temperature variations. Additionally,
analyses may be complicated by density variations in the
suspended material. These complications coupled with the
sensitivity limitations of the pressure transducer sensors may
render this technology unreliable at concentrations below
10 000–20 000 mg l
1
.
1.10.3.2.3.2 Example field evaluation
Information on the field performance of the pressure-differ-
ence technology is available from USGS streamgages on the
lower Rı
´o Caguitas in Puerto Rico (Larsen et al., 2001;Gray
and Gartner, 2010b) and the Paria river in AZ, USA (Gray and
Gartner, 2010b). Continuous pressure-difference data were
collected during October–December 1999 at the Rı
´o Caguitas
streamgage using a Double Bubbler, composed of a digital
recorder, bubbler system, and two precision pressure sensors
with orifices anchored at fixed depths in a vertical (Design
Analysis Associates, Inc., 2008) (Figure 15). Most of the an-
nual suspended sediment discharge in the lower Rio Caguitas
occurs as runoff from a few storms during which SSCs exceed
24:00
0
20
Suspended-sand
concentration (mg l−1)
40
60
80
100
120
140
12:00
7/19
24:00 12:00
7/19
24:00
Explanation
Calibrated LISST 100B point measurement
Cross-sectionally integrated with D-77 sampler
Dischar
g
e of water
0
100
Discharge (m3 s−1)
200
300
400
500
Date (2001)
24:00
0.09
0.10
0.11
0.12
0.13
0.14
0.15
0.16
0.17
0.18
0.19
0.20
Suspended-sand D50 (mm)
12:00
7/19
24:00 12:00
7/19
24:00
0
100
Dischar
g
e (m3 s−1)
200
300
400
500
Date (2001)
Figure 14 Comparison of sand concentrations in milligrams per liter and median grain sizes in millimeters measured at the USGS streamgage
at the Colorado river near Grand Canyon, AZ, USA, using a LISST-100B and a US D-77 bag sampler. Reproduced from Melis, T. S., Topping,
D. J. and Rubin, D. M. (2003). Testing laser-based sensors for continuous in situ monitoring of suspended sediment in the Colorado river,
Arizona. In Bogen, J., Fergus, T. and Walling, D. E. (eds.) Erosion and sediment transport measurement in rivers, technological and
methodological advances. International Association of Hydrological Sciences Publication 283, pp. 21–27.
Measuring Suspended Sediment 189
Author's personal copy
approximately 500 mg l
1
. The maximum SSC measured at
the streamgage during the Puerto Rico Double Bubbler tests
based on water samples collected by an autosampler (Edwards
and Glysson, 1999) was 17 700 mg l
1
.
Data analyses involved data smoothing and removal of
outliers. To calculate the weight density of suspended sedi-
ment and dissolved solids, the weight density of pure water at
27 1C was subtracted from the smoothed data values. Even
with these manipulations, the tests of the Double Bubbler
instrument at the Puerto Rico site during October–December
1999 showed relatively poor agreement with water discharge,
SSC, and water density (Figure 16). The Double Bubbler data
contained a large amount of signal noise, making interpret-
ation difficult. Lacking a thermistor for temperature compen-
sation, 12 of 15 base-flow instrument measurements inferred
negative SSC values (an impossibility) concurrent with in-
stream measured SSC values of 10 000–100 000 mg l
1
.
However, all but two of the samples collected during seven
higher flow periods showed concomitant increases in inferred
positive SSC values.
A complicating factor in the pressure-difference method is
in-stream turbulence, which introduces noise about equal to
the magnitude of the signal of interest, particularly during
high flows that occur more or less concomitant with the
largest SSCs. Additionally, diel and storm-related fluctuations
in water temperatures resulted in a daily range as much as
10 1C. The high relative humidity characteristic of this humid
tropical site may also complicate the use of the Double Bub-
bler because of the sensitivity of the narrow diameter bubbler
gas lines to moisture, unless the gas lines are equipped with
dryer tubes. This test of the Double Bubbler instrument
showed the need for temperature compensation, and possibly
the need to deploy the instrument at a site where the range in
the density of the water–sediment mixture is substantially
larger than the 1–1.02 range occurring in the Double Bubbler
evaluation at the Rı
´o Caguitas streamgage.
In 2004, the Puerto Rico Double Bubbler system was re-
located to the USGS streamgage on the Paria river at Lees Ferry,
AZ, USA, where SSCs as high as 10
6
mg l
1
have been meas-
ured during storm runoff (Beverage and Culbertson, 1964).
Deployment of the Double Bubbler in the Paria river was
predicated on the hypothesis that Paria river SSCs, commonly
exceeding peak measured Rio Caguitas SSCs by a factor of at
least three and in some cases by 1–2 orders of magnitude,
would subject the instrument to a substantially larger density
range than that inferred for higher flows at the Rio Caguitas
streamgage in Puerto Rico. However, even with the addition of
a thermistor for monitoring water temperatures, results were
inconclusive (Gray and Gartner, 2010b).
Double Bubbler data were collected, at 5-min intervals, dur-
ing periods of elevated flow at the Paria river streamgage from
July 2004 through September 2006. Data collected from more
than 14 storm-runoff hydrographs were examined and com-
pared to SSCs from samples collected during storm
runoff. The elevated flows had peaks ranging from approximately
7–90 m
3
s
1
; the maximum SSC measured was 382 000 mg l
1
from a sample collected using an autosampler. Of the 261 sus-
pended sediment samples collected during the 14 storm-runoff
periods, 86% had SSC values larger than 50 000 mg l
1
(Nancy
Hornewer, USGS, written communication, 2008).
Dry air or nitrogen
Orifice gas supply system
Precision differential
pressure measurement
system
Low
pressure
orifice
line
High
pressure
orifice
line
Sciencid
valve Sciencid
valve
Atmospheric
reference
Water
level
Upper
orifice
d
d is distance between orifice
tube elevations = 304.8 mm
±1.76 × 10−4 mm
Orifice line
mounting
block
z1
z2
(a)
(b)
(c)
(d)
2 = z31 = z1
Water
temperature
probe
Lower
orifice
Figure 15 Double Bubbler pressure differential instrument (a) in-stream components before installation, (b) controller and orifice bar, (c) air
compressor and tank assembly, and (d) schematic. Photographs (b) and (c) courtesy of Design Analysis Associates (2012). Home page.
Available at http://www.waterlog.com/ (accessed on 14 June 2012).
190 Measuring Suspended Sediment
Author's personal copy
Similar to data collected at the Rio Caguitas in Puerto Rico
and contrary to the aforementioned hypothesis, the Double
Bubbler data collected at the Paria river at Lees Ferry streamgage
seemed to have a large amount of signal noise, also making
interpretation difficult (Gray and Gartner, 2010b). The Double
Bubbler data were collected only during periods of elevated
stages (water discharges) because the instrument was not fully
submerged during normal shallow flows. Data were filtered in a
manner similar to that for the Rı
´o Caguitas data but not
smoothed. Relations between measured SSCs and those calcu-
lated from bubbler data tended to be inconsistent. It is likely
that bed movement caused the lower orifice to become partially
or fully blocked at times, contributing to erroneous data. Also,
the paired stage readings necessary for the density calculation
could not always be obtained because both orifices were only
submerged during infrequent periods of high flow. Neither
condition constitutes a fair test of the technology.
The performance of the Double Bubbler has been proven
neither inadequate nor adequate for USGS data collection
purposes in streams with large SSCs (Gray and Gartner,
2010b). Regardless, due to its strong theoretical underpin-
nings, continuous monitoring capability, and – not unim-
portantly – a lack of any other proven technology for
monitoring SSCs in high- and hyperconcentrated streamflow
conditions the technology is considered worthy of additional
testing. To this end, a 3-orifice pressure-differential system
(Sutron Corporation, 2012) is being tested at the USGS
streamgage Rio Puerco near Bernardo, New Mexico, USA.
120
100
80
60
40
Weight density of suspended
sediment,
mg l−1 × 1000 from bubbler
20
0
−20
−40 0.1
10 000
1000
100
10
1
0.1
10/1 11/1
Note: Only symbols denote measure values; dashed interpolation lines are included for viewin
g
purposes only
Date, 1999−2000
Explanation
12/1 1/1
−40
−20
0
Weight density of suspended sediment,
mg l−1 × 1000 from bubbler
Suspended sediment concentration, (mg l−1)
stream discharge (mg3 s−1)
20
40
60
80
100
120
1
Stream discharge (m3 s−1)
10 100
120
100
80
60
40
20
0
−20
−40
10 100
Suspended-sediment concentration (mg l−1)
1000 10 000
Weight density of suspended sediment,
mg l−1 × 1000, from bubbler
Weight density of suspended sediment,
mg l−1 × 1000, from bubbler
Stream discharge,
m3 s−1
Suspended-sediment concentration,
mg l−1
Figure 16 Scatter plots and time series of stream discharges, SSCs, and weight density of suspended sediments and dissolved solids measured
with a Double Bubbler (Design Analysis Associates (2012)), 1 October 1999–1 January 2000, lower Rio Caguitas, Puerto Rico. Water discharge
and sediment data are instantaneous values in cubic meters per second and milligrams per liter, respectively; Double Bubbler weight density
values are expressed in milligrams per liter as 30-min mean values of measurements made at 5-min intervals. Reproduced from Larsen,
M. C., Figueroa-Alamo, C., Gray, J. R. and Fletcher, W. (2001). Continuous automated sensing of streamflow density as a surrogate for
suspended-sediment concentration sampling. Proceedings of the 7th Federal Interagency Sedimentation Conference, Reno, NV, 25–29 March,
vol. I, pp. III-102–III-109. Available at http://pubs.usgs.gov/misc_reports/FISC_1947-2006/pdf/1st-7thFISCs-CD/7thFISC/7Fisc-V1/7FISC1-3.pdf
(accessed on 14 June 2012).
Measuring Suspended Sediment 191
Author's personal copy
Beverage and Culbertson (1964) indicated that hyperconcen-
trated streamflows of at least 560 000 mg l
1
have occurred at
this site.
1.10.3.2.3.3 Advantages and limitations of the pressure-
difference technology
Advantages:
•The pressure-difference technology’s inference of SSC in a
single vertical is an improvement over at-a-point meas-
urements but still may not provide SSC data representative
of mean cross-sectional values.
•The technology is relatively robust, being prone to neither
signal drift nor biological fouling.
•The technology doubles as a redundant stage sensor for
the site.
•The technology may be unique in that the accuracy of its
measurements theoretically improves with concentrations
increasing above approximately 10 000–20 000 mg l
1
.
•The theoretical underpinnings of the technology are rela-
tively simple and straightforward.
Limitations:
•The required computational scheme presupposes that the
SSC in the vertical profile above the lower pressure sensor
is more or less constant to the surface. This assumption,
which is difficult to verify, may not be valid.
•The technology may be incapable of measuring SSCs below
approximately 10 000–20 000 mg l
1
in turbulent flows
and where the bedforms cover one or both orifices. The
field performance of the technology has yet to be
adequately resolved at any SSC.
•The technology is incapable of measuring SSC when the
top orifice is not submerged or the bottom orifice is buried
in sediment.
•Spurious data are numerous and are believed to be asso-
ciated with flow turbulence.
•The Double Bubbler (Design Analysis Associates, Inc.,
2008) is no longer marketed; however, three versions of the
Accubar
s
Bubble Gauge Recorder are marketed by Sutron
(2012). The third version, employing three precision
pressure sensors, is reputed to be thrice as accurate as those
composed of dual density sensors.
1.10.3.2.4 Acoustic surrogates
Hydroacoustics is a compelling technology that can provide
surrogate measurements of SSC with several advantages in-
cluding large sample volumes, potential for simultaneous vel-
ocity measurements, and environmental robustness (Landers
et al., 2012). Characterization of suspended sediment using
backscatter and attenuation of acoustic signals in water has
been described and developed for several decades (Urick, 1948,
1975;Flammer, 1962;Hay, 1983;Sheng and Hay, 1988;Flagg
and Smith, 1989;Thorne et al., 1991;Hay and Sheng, 1992;
Lynch et al., 1994;Holdaway et al., 1999;Gartner, 2004;Wal l
et al., 2006;Gray and Gartner, 2009;Simmons et al., 2010;Gray
and G ar tne r, 2 010b;Guerrero et al., 2011;Thorne et al., 2011).
The basic principles are that acoustic waves passing through a
water–sediment mixture will scatter and attenuate as a function
of sediment, fluid, and instrument characteristics. The acoustic
metrics of backscatter and attenuation relate functionally to
sediment characteristics (SSCs, PSDs, and shapes) within an
ensonified volume after adjusting for the influence of fluid and
instrument characteristics.
Early investigations of acoustic surrogates relied on in-
struments with a separate sound source and receiver, rather
than a combined source and receiver such as modern trans-
ceivers (referred to here and typically as transducers). The
transducer emits an acoustic pulse and then, after an interval
just long enough to stop ‘ringing,’ it receives the echoes
backscattered from particles suspended in the acoustic path as
illustrated in the simplified cartoon of Figure 17. Acoustic
Doppler velocity meters measure the Doppler shift in the
frequency of the backscattered signal to determine the velocity
of the particles scattering the signal (the assumed water vel-
ocity) relative to the transducer. Two or three transducers at
fixed beam angles may be used to resolve a 2- or 3-dimen-
sional flow velocity vector. As acoustic Doppler velocity meters
are now ubiquitous in streamflow monitoring, the acoustic
backscatter provides tremendous potential for sediment
monitoring using existing instrumentation.
1.10.3.2.4.1 Acoustic backscatter
The amplitude of the backscattered acoustic signal depends on
several factors including SSC, PSD, acoustic frequency, range
from the transducer to the ensonified particles, the volume of
sediment-water ensonified, water temperature, sediment
density, and sediment shape. In order to use acoustic surro-
gates, the effects of all of these factors must be measured,
modeled, or assumed. Physically, the acoustic signal to sedi-
ment interaction involves both acoustic scattering and at-
tenuation. Because there are unique physical mechanisms
driving backscatter and attenuation, these two metrics can
Transmitting
Transmitted acoustic ping
Receiving
Reflected acoustic energy
Scatterers
Transducer
Transducer
Figure 17 Acoustic backscatter from suspended particles.
192 Measuring Suspended Sediment
Author's personal copy
provide independent information relating to SSC and PSD. In
many applications, the acoustic backscatter is calibrated to
measure SSC, as described in the following paragraphs, and
the effect of changes in the sediment size, shape, and density
are assumed to be negligible.
1.10.3.2.4.2 Acoustic attenuation
Acoustic absorption due to sediment (a
s
) in decibel per meter
varies linearly with SSC for a given frequency and sediment
size, shape, and density. The physical process driving acoustic
attenuation for a given SSC include shear (or viscous) energy
losses at the sediment particle-to-water boundary and energy
transference due to scattering. Viscous losses are primarily due
to the SSC of finer particles (viscous loss size range), whereas
scattering losses are primarily due to coarser particles (scat-
tering loss size range). The acoustic attenuation due to viscous
loss is caused by shear at the fluid–particle boundaries because
of a lag between the sound wave-induced vibration of the
particle and that of the fluid. The magnitude of the viscous
loss is a function of the particle surface area, sound frequency,
fluid viscosity, and the ratio of particle to fluid density. The
scattering loss is due to reradiation of the acoustic energy in-
cident on a particle. Scattering loss is a function of the ratio of
acoustic wavelength, l, to particle circumference 2pa
s
.The
scattering attenuation reaches a maximum at particle diam-
eters of approximately 1050, 840, and 420 mm for frequencies
of 1.2, 1.5, and 3 MHz, respectively.
The minimum acoustic attenuation occurs at the transition
between viscous and scattering losses. The particle size asso-
ciated with this minimum attenuation increases with wave-
length (decreasing frequency). For example, the minimum
acoustic attenuation occurs at particle diameters of 90, 74, and
42 mm for frequencies of 1.2, 1.5, and 3 MHz, respectively,
using 1484 m s
1
as the speed of sound in water. These par-
ticle diameters are in the silt to very fine sand range. Know-
ledge of this transition particle size is relevant to using
multifrequency acoustics to assess characteristics of the sedi-
ment PSD. Acoustic attenuation may be computed using
forms of the equations by Urick (1948) or Sheng and Hay
(1988) as described in Thorne and Hanes (2002) or Landers
(2012). Acoustic attenuation may be measured with much
greater potential accuracy than available by computations
using multicell acoustic Doppler current profilers or by using
the method developed by Topping et al. (2007).
1.10.3.2.4.3 Acoustic surrogate methods for SSC
A semiempirical backscattering theory and acoustic surrogate
methodology was progressively developed by several re-
searchers working in fluvial (Urick, 1948,1975;Flammer,
1962;Gartner, 2004;Topping et al., 2007;Wall et al., 2006;
Gray and Gartner, 2009;Simmons et al., 2010) and marine
environments (Sheng and Hay, 1988;Thorne et al., 1991;
Hay and Sheng, 1992;Downing et al., 1995;Crawford and
Hay, 1993;Thorne and Hanes, 2002;Thorne et al., 2011). The
methods developed in fluvial and marine environments are
distinct, yet similar and can be equated mathematically. For
the purposes of this summary, the form of equations pre-
sented is that developed and used primarily by researchers in
fluvial systems.
The acoustic sediment surrogate methods used primarily in
fluvial systems begins with Urick’s (1975) sonar equation and
has been used by several authors (Thevenot and Kraus, 1993;
Reichel and Nachtnebel, 1994;Gartner, 2004;Wall et al.,
2006;Topping et al., 2007). The sonar equation is written in
logarithmic units of decibels as:
RL ¼SL 2TL þTS ½6
where
2TL ¼20log10ðrÞþ2rðasþawÞ½7
In this method, RL is the reverberation level (measured
backscatter intensity at the transducer face recorded by the
acoustic velocity meter) of the received signal, SL is the source
level of the emitted signal, 2TL is the two-way transmission
loss equal to the sum of the spherical energy spreading and
attenuation, and TS is the intensity of the signal echoed by the
particles in the ensonified volume equal to 10 log
10
(SSC).
In the equation for TL, ris the range from the transducer to
the ensonified volume, acoustic attenuation is expressed in the
coefficients for acoustic absorption, a
s
, and water absorption,
a
w
. The first term to the right of the equality in the TL equation
accounts for physical spreading of the acoustic energy as the
signal propagates from its source and spreading is linear in
logarithmic space. Attenuation in pure water at depths less
than 100 m is a function of temperature and acoustic fre-
quency only and may be computed using standard equations
such as that by Fisher and Simmons (1977). Water tempera-
tures are measured and stored by most acoustic meters at each
measurement interval, so continuous time series of a
w
can be
computed. The relative backscatter (RB) is computed as
RB¼RL þ2TL, which is equivalent to the total scattering by
suspended particles.Then, log
10
(SSC) is a function of RB and:
SSC ¼10ðAþBRBÞ½8
The coefficients Aand Bare evaluated using regression for
paired physical SSC and acoustic measurements. Source level
is evaluated in the regression coefficient for this method
(Thevenot and Krause, 1993;Gartner, 2004), and sediment
acoustic attenuation is sometimes assumed to be negligible for
low-concentration systems (Gartner et al., 2001;Wall et al.,
2006).
Measurements of the acoustic return signal can be digitally
sliced into specific range-gated ‘cells’ to provide data on vel-
ocity and acoustic metrics at integral points along the acoustic
beam. Most commercially available acoustic velocity meters
have this profiling capability. These multicell data offer an
effective means to directly measure the sediment acoustic at-
tenuation at a high temporal resolution. Figure 18 shows the
acoustic backscatter in decibels measured by a 1.5 MHz unit
with 10 cells of 20 cm axial distance each measured between
0.2 and 2.2 m from the transducer face for a river with an
average SSC in the channel cross section of approximately
694 mg l
1
at the time of this acoustic measurement (Landers,
2012). The RL line is the measured backscatter intensity and
its slope is the combined two-way signal strength loss due
to spherical spreading plus fluid and sediment acoustic at-
tenuation. The (RL þ20log
10
(r)) line is the measured back-
scatter intensity corrected for spherical spreading, and the
Measuring Suspended Sediment 193
Author's personal copy
(RL þ20log
10
(r)þ2ra
w
) line is further corrected for fluid at-
tenuation. Solving for the slope of this line provides the two-
way acoustic attenuation, 2a
s
.Topping et al. (2007) first used
this method to solve for a
s
. This multicell method has the
powerful advantages of measuring a
s
and of normalizing for
the effects of sediment scattering properties and transducer-
specific characteristics.
Methods to obtain SSC using acoustic surrogates include/
involve elements common to all sediment monitoring studies
(identifying information goals and accuracy requirements,
budgets, site selection, training) and elements specific to
acoustic surrogates (selection of monitoring section, instru-
ment set-up, data retrieval, data processing, concurrent phys-
ical sampling, and calibration). The acoustic meters may be
mounted in situ for time series data or used from a mobile
platform for discrete measurements. The ensonified volume
for in situ monitoring should be selected to maximize the
measured volume for all flow conditions and generally limited
by low-flow conditions. The system should be set up to collect
multicell data with initial set up using at least 10 cells to op-
timize calibration with measured SSC. Once the instrument
set up is complete and deployed, concurrent acoustic and
physical sample measurements are required over the range of
sedimentological conditions that occur for the monitoring
period. The TL data are computed for each cell using the range
to the cell, computed a
w
, and measured or computed a
s
. The
RB is computed for each time step (for in situ) or location (for
mobile) using the measured backscatter (RL) and the TL as
RB¼RL þ2TL.This analysis is performed for each time step to
obtain a time series of the acoustic surrogates of acoustic at-
tenuation by sediments and normalized acoustic backscatter.
The acoustic surrogates RB and a
s
are then calibrated using
concurrent measurements of cross-section representative SSC.
Sensitivity analyses may be performed to determine whether
to use the entire ensonified profile or selected cells to best
represent the SSC. Sensitivity analyses should also be done to
determine whether using both RB and a
s
provide statistically
significant and independent data explanatory variables or
whether RB should be used alone as the acoustic surrogate.
Computed estimates of SSC should be reported with the SSC-
acoustic surrogate model error, as well as assessments of the
calibration dataset uncertainty.
1.10.3.2.4.4 Multifrequency acoustic surrogates for sediment
size
A major limitation of single-frequency systems is that the
metrics of acoustic attenuation and RB change in response to
changes in both SSC and PSD, creating a SSC–PSD ambiguity.
Relative acoustic backscatter from sediment may increase with
increased concentration at a fixed size distribution or with
increased sediment size at a fixed SSC, and acoustic attenu-
ation also varies with size. Multifrequency acoustic systems,
however, have been successfully used to estimate both sedi-
ment concentration and general size characteristics (Crawford
and Hay, 1993;Gartner, 2004;Hay and Sheng, 1992;Thorne
et al., 1991;Smith et al., 2006;Topping et al., 2007). Two
approaches have been developed that use multifrequency
acoustics to evaluate sediment size, one by Hay and Sheng
(1992), which is effectively summarized in Thorne and Hanes
(2002) and one by Topping et al. (2007). Three frequencies
have been used in most multifrequency sediment surrogate
studies.
Topping et al. (2007) observed the transition from viscous
to scattering losses in sediment acoustic attenuation, as noted
in Section 1.10.3.2.4.3 for Urick’s equation and from this
describe two acoustic size classes of sediment – a finer acoustic
size class in which viscous attenuation is dominant and a
coarser acoustic size class in which scattering attenuation is
dominant and backscatter is more significant. Applying this
method, Topping et al. (2007) report computed concen-
trations within 5% of the values computed using conventional
data and median sand grain size typically within 10% of the
values obtained by conventional measurement.
1.10.3.2.4.5 Examples of deployments
Landers (2012) used horizontally mounted, in situ 1.2, 1.5,
and 3.0 MHz systems at a site on the Yellow river near Atlanta,
GA, USA. Acoustic, turbidity, and laser-diffraction surrogates
were compared with traditional SSC measurements and
computed sediment flux. Figure 19 shows a time series plot of
measured SSC (from physical samples), RB, and attenuation
for a 1.5-MHz system for a small storm in April 2010 at the
Yellow river site.
More than 180 representative, concurrent SSC measure-
ments were obtained during 2009–10 for calibration with the
acoustic surrogates. Both RB and a
s
proved statistically sig-
nificant and independent as explanatory variables. The re-
gression fit between these variables and measured SSC had R
2
values from 0.79 to 0.80 for the three frequencies tested, with
model standard errors from 34% to 40%, as indicated in
Figure 20 for the 1.5 MHz acoustic Doppler current profiler.
By comparison, a traditional model using streamflow dis-
charge as the explanatory variable for these data had a model
standard errors of 73%. The measured SSC calibration dataset
was estimated to have an uncertainty of 710%. However, as
noted previously, the surrogate measurements can measure at
very high resolution compared with physical sampling over
time, thus providing improved accuracy and information
over time.
Sediment flux measured by detailed samples was estimated
using all three frequency acoustic surrogates with an error of
prediction of less than 10%, which is within the estimated
uncertainty of the sediment flux calibration dataset. Landers
50
01 2 3
Ran
g
e from transducer face (m)
4 5 6 7
60
70 RL: unadjusted
RL: adjusted for spreading
RL: adjusted for spreading and water attenuation
RB: fully adjusted backscatter
80
1.5 MHz; SSC = 185 mg l−1
Acoustic backscatter (dB)
90
100
110
120
Figure 18 Acoustic backscatter amplitude profiles along beam axis
for 1.5 MHz.
194 Measuring Suspended Sediment
Author's personal copy
(2012) also found that ratios of measured acoustic attenuation
for 1.2, 1.5, and 3 MHz systems are significantly related to
PSD, with increasing significance for smaller size fractions.
Wall et al. (2006) used a bottom-mounted, broadband, 4
beam, 0.614 MHz acoustic Doppler current profiler on the
Hudson river below Poughkeepsie, NY, USA (USGS stream-
gage 01372058). They collected concurrent SSC samples
from the ensonified volume to calibrate the sonar equation.
They also measured depth- and width-integrated SSC for
the stream cross section and developed the relation between
sediment flux in the acoustic sampling area and in the river
cross section. Their work resulted in the computation of
daily sediment flux for this tidally affected location of the
Hudson river and is included in near real time as part of the
USGS National Water Information System (U.S. Geological
Survey, 2012a).
1.10.3.2.4.6 Advantages and limitations of acoustic surrogates
Advantages:
•Acoustic velocity meters are used broadly in streamflow
monitoring to collect acoustic backscatter data, which can
generally be used with calibration and analysis to infer
sediment characteristics.
•Acoustic meters are highly robust in the stream environ-
ment requiring little cleaning and maintenance.
•Acoustic meters measure a substantially large volume
of the flow compared to the other described surrogate
technologies.
•Most commercially available acoustic velocity meters
measure profiles of backscatter, which can be used to in-
dicate acoustic attenuation as well as backscatter.
•As with other surrogate technologies, acoustic meters can
operate at very high temporal resolutions that provide
improved accuracy and information for dynamic sediment
characteristics.
•Multifrequency acoustic instruments can provide infor-
mation on PSD as well as SSC.
Limitations:
•As with turbidimeters, acoustic metrics from single-
frequency systems may change due to changes in sediment
concentration or sediment size, creating a size–
concentration ambiguity.
•Acoustic metrics are sensitive to the ratio of sediment size
to acoustic wavelength and a somewhat narrow frequency
is optimal for evaluation of PSD using multifrequency
acoustics.
•The sensitivity of acoustic surrogates of suspended sedi-
ment is limited for low concentrations and generally may
not be applicable for concentrations less than approxi-
mately 10 mg l
1
for frequencies in the 0.5–5 MHz
frequency range.
•All methods that use measurements of a subsection or
single point of the channel cross section require concurrent
point and cross-section SSC measurements for calibration
112
Concentration, cross section SSC
Relative backscatter, 1.5 MHz
Attenuation, 1.5 MHz
110
108
106
104
102
100
98
96
94
04/27/10
04/27/1004/26/1004/26/1004/25/1004/25/10
04/24/10
0
40
80
120
160
200
Concentration SSC (mg l
−1
)
3.5
3.0
2.5
2.0
Attenuation (dB m
−1
)
1.5
1.0
0.5
0.0
Figure 19 Measured SSC, acoustic attenuation, and relative acoustic backscatter during runoff on the Yellow river near Atlanta, GA, USA.
10 102
1
10
10
1
2
2
Predicted cross section SSC by
1.5 MHz acoustic metrics (mg l−1)
3
4
5
6
7
8
2
2
3
4
5
6
SSC
95% Cl
Observed cross section SSC (m
g
l−1)
345678 2 3456
Figure 20 Observed and predicted SSC using acoustic backscatter
and attenuation from 1.5 MHz acoustic Doppler current profiler.
Reproduced from Landers, M. N. (2012). Fluvial suspended sediment
characteristics by high-resolution, surrogate metrics of turbidity,
laser-diffraction, acoustic backscatter, and acoustic attenuation. Ph.D.
Dissertation, Georgia Institute of Technology. Available at http://
hdl.handle.net/1853/43747 (accessed on 14 June 2012).
Measuring Suspended Sediment 195
Author's personal copy
to the entire cross section and assumed stability of those
calibrations. This limitation is more restrictive for surro-
gates based on small at-a-point volume measurements than
for acoustic surrogates, which typically are based on a
much larger volume than point measurements.
1.10.4 Summary and Conclusions
The efficient and proper management of river systems is
predicated on the availability of sufficient high-quality fluvial
sediment and ancillary data. Two general means for obtaining
sediment data have been described: those based on traditional
instruments and techniques and those using technologically
advanced instruments. Traditionally, collected data tend to be
spatially robust but temporally deficient. The opposite is true
for in situ surrogate technologies. Some of the surrogate
technologies are being incorporated into operational pro-
grams and show considerable promise toward providing the
temporally dense and, in some cases, spatially robust fluvial
sediment data needed to increase and bring more consistency
to sediment-discharge measurements worldwide.
1.10.4.1 Traditional Technologies
The previously described in Section 1.10.2.2 traditional FISP
samplers and techniques, which debuted in the mid-twentieth
century, are grounded on sound physical and statistical prin-
ciples and form the basis for production of the bulk of fluvial
suspended sediment data in the US and in selected countries
of every continent other than Antarctica (Federal Interagency
Sedimentation Project, 1941;Edwards and Glysson, 1999;
Nolan et al., 2005). The key for collection of representative,
unbiased physical samples is the use of any one of a suite of
FISP isokinetic samplers – devices capable of collecting a
filament of flow without change in the ambient velocity or
direction – using the EDI or EWI methods deployment tech-
niques. The derivative data are used to directly quantify in-
stantaneous suspended sediment discharges and calibrate
physical and surrogate intermittent or time series data col-
lected by in situ devices. Failure to use flow integrating in-
struments and techniques often results in production of data
of unknown veracity. Additionally, even if in situ surrogate
technologies become operationally ubiquitous, they will
continue to require empirical calibrations with traditional
technologies. Ergo, traditional instruments and technologies
will remain necessary and relevant for the foreseeable future.
1.10.4.2 Surrogate Technologies
Four advanced in situ technologies for monitoring fluvial
suspended sediment transport are among instruments and
techniques being tested by the USGS: turbidity (nephelometry
and transmissivity), laser optics, pressure difference, and
acoustic backscatter. Although none is a panacea for sediment
monitoring needs in all rivers, the capability for consistent,
large-scale monitoring of suspended sediment transport in
many of the world’s rivers may be possible.
Table 4 summarizes selected attributes of the four tech-
nologies that are germane to their potential use as a sedi-
ment-surrogate technology. Each technology, with the
possible exceptions of manually deployed laser-optic instru-
ments, requires periodic calibration with data produced from
traditionally collected water samples to calculate the mean
value in the cross section. When properly configured and
deployed, each is capable of providing a dense and con-
tinuous time series of SSC for use in computation of con-
tinuous suspended sediment transport. Laser optics and
possibly multifrequency acoustic backscatter may provide the
added capability of sediment discharge computations by
particle-size class. The ability to determine continuous, high
frequency, time series of SSC is a major advantage over tra-
ditional data collection techniques, obviating the need for
routine, potentially subjective interpolations between sample
values, and providing the capability to determine high-fre-
quency SSC and PSD fluctuations not revealed by traditional
measurements. Calibrations with somewhat larger un-
certainty bounds might be considered more acceptable in
that the vastly increased derived data density preclude the
routine need for interpolations between infrequent discrete
measurements.
The applicability of each technology is dictated in part
based on the physical and hydrological characteristics of the
monitoring site, monitoring objectives, and on the instru-
ment’s advantages and limitations. Each deployed surrogate
instrument provides time series data representative of the
sedimentological characteristics in but a fraction of the cross
section. Both optical technologies provide at-a-point SSC data
during periods when in-stream SSC values remain below the
instrument’s saturation limit. The SSC data provided by laser
optics are computed from PSDs associated with each
measurement.
The location and volume of flow measured by an instru-
ment is an important factor in correlating the measurements
to mean cross-sectional SSC values. Assuming production of
reliable data collected from the instrument’s measurement
realm, SSCs computed utilizing acoustic-backscatter technol-
ogy (employing a profile of vertical or horizontal measure-
ments) may correlate better with the mean SSC value for
the river cross section than those computed with the
pressure-difference method, which in turn may be better
than those computed from at-a-point turbidity or laser-optic
measurements.
Nevertheless, the most ubiquitous in situ surrogate tech-
nology utilizes turbidimeters, which have been shown to
provide useful data for computing SSCs in a number of field
settings. However, issues associated with instrument sensor
saturation can result in failure to record reliable data at the
higher values of SSCs that tend to be the most influential in
sediment transport. SSCs computed from at-a-point turbidity
data may not be representative of the mean cross-sectional
SSC, particularly when sand-size material composes an ap-
preciable fraction of total suspended sediment transport.
Biological fouling can reduce signal integrity in the absence
of a mechanical wiper or manual cleaning to keep the optical
window clean. Turbidimeter costs are a small fraction of
the annual cost of monitoring suspended sediment transport
using traditional techniques, but the potential for increased
196 Measuring Suspended Sediment
Author's personal copy
Table 4 Summary of selected attributes of four suspended sediment surrogate technologies
Technology Turbidity (bulk optics) Laser Pressure difference Hydroacoustics
Instrument or type In situ turbidimeter In situ OBS In situ LISST-100 or
LISST-streamside
(pumping)
Manually deployed
LISST-SL
In situ Double
Bubbler
In situ single-
frequency acoustic
Doppler profiler
In Situ Multiple-
frequency acoustic
Doppler profiler
Price relative to
in situ turbidimeter
$5000 and up
(summer 2011)
Approximately 1X Approximately 5X Approximately 6X Approximately 1X Approximately
2X–3X
Approximately
3X–6X
Approximate
concentration
measurement
range
Standard 0–2 g l
1
Standard 0–5 g l
1
Depending on
versions:
0–2 g l
1
(particle
size dependent)
Approximately
0–2 g l
1
(particle
size dependent)
Larger than
approximately
10 g l
1
, but
needs more
research; no upper
limit
0.02–5 g l
1
0.02–5 g l
1
Available at larger
ranges. Sensor
saturation
(censoring) occurs
above maximum
Available at larger
ranges
Variable as function
of PSD and
frequency
Variable as function
of PSD and
frequency
Approximate
measurement
range, PSD in mm
Does not measure
PSD
Does not measure
PSD
0.0025–0.5 or
0.0025–0.38
0.002–0.38 or
0.00125–0.25
Does not measure
PSD
Does not measure
PSD
May measure or
qualify PSD
Measurement
metrics
Formazin
nephelometric or
nephelometric
turbidity units
Optical Backscatter
in millivolts
Volumetric particle
concentration and
PSD within
instrument
sediment size
limits
Volumetric particle
concentration and
PSD within
instrument
sediment size
limits
Calibrated to
concentrations
from physical
samples in mass
units
Relative acoustic
backscatter and
attenuation
Relative acoustic
backscatter and
attenuation for
multiple
frequencies
Ancillary
measurements
None None Depth and water
temperature
Depth, ambient
velocity, and water
temperature
Stage Velocity,
temperature, stage
or depth, if
oriented down
Velocity,
temperature, stage
or depth, if
oriented down
Reliability,
robustness, and
frequency of
servicing
Reliable technology.
For wiped sensor
models, cleaning
every 2–6 weeks
depending on
stream and
temperature
Reliable technology.
For wiped sensor
models, cleaning
every 2–6 weeks
depending on
stream and
temperature
Robustness and
reliability is
variable to low,
depending on
stream. Cleaning
every 1–3 weeks
Robustness and
reliability is
variable to low,
depending on
stream. Cleaning
every use.
Low concentration
data unreliable;
veracity of higher
concentrations
unresolved
Very reliable and
robust. More or
less unaffected by
fouling
Very reliable and
robust. More or
less unaffected by
fouling
(Continued )
Measuring Suspended Sediment 197
Author's personal copy
Table 4 Continued
Technology Turbidity (bulk optics) Laser Pressure difference Hydroacoustics
Region of
measurement
Fixed point Fixed point Fixed point (100X) or
pumping
(streamside)
Point, vertical, or
multiple verticals
Single fixed vertical,
mean
concentration
value
Conic beam with
data available at
selected distances
from the sensor
Conic beam with
data available at
selected distances
from the sensor
Accuracy for
derivation of
suspended
sediment data
When within
measurement
range has been
used to develop
reliable SSC-
turbidity
regression
relations
When within
measurement
range has been
used to develop
reliable SSC-
turbidity
regression
relations
Deemed reliable in
some field
applications
Deemed reliable in
some field
applications
Unresolved based on
two field tests;
additional work
planned
Shown useful in field
applications where
size distribution
does not change
dramatically
Shown to provide
accurate silt–clay
versus sand-size
fractions
Potential for
application in
large-scale
monitoring
programs
Very high (given
appropriate in-
stream
sedimentological
conditions,
calibration, and
ability to maintain
instruments)
Very high (given
appropriate
in-stream
sedimentological
conditions,
calibration, and
ability to maintain
instruments)
High (given
appropriate
in-stream
sedimentological
conditions, known
density, and ability
to maintain
instruments)
High (used for
calibrating in situ
instruments)
Unknown pending
additional testing
using
modifications of
the physical
system and
algorithms
High for SSC for
systems with
relatively stable
PSD
High for SSC;
Moderate for
silt–clay versus
sand-size fractions
Abbreviations: LISST, laser in situ scattering and transmissometry; OBS, optical backscatterance; PSD, particle-size distribution; SSC, suspended sediment concentration.
198 Measuring Suspended Sediment
Author's personal copy
site visits for maintenance may result in increased operating
costs.
In situ laser-optic instruments also suffer from the draw-
backs associated with sensor saturation, biological fouling,
and at-a-point measurement limitation characteristics of in situ
turbidimeters. Additionally, laser data are in the form of vol-
ume SSC and mass SSC may be calculated only if particle
density is known or can be reliably inferred. The purchase
price of an in situ laser-optical instrument (LISST-100) is ap-
proximately 5 times the cost of a fully equipped in situ tur-
bidimeter. However, these instruments have the major
advantage in providing continuous PSDs from which the
volumetric SSCs are inferred.
The pressure-difference technology is designed for moni-
toring SSCs exceeding approximately 10 000 mg l
1
in a single
vertical, which is near or above the maximum range of the
other technologies examined herein. The purchase price of
this relatively uncomplicated technology is similar to that for a
turbidimeter. It is relatively robust in that it integrates the
density of a water column as opposed to a single vertical, and
it is not subject to biological fouling. The theoretical under-
pinnings of this technology are straightforward. However,
performance of the pressure-difference technology has been
marginal at best in field tests in Puerto Rico (maximum SSCs
of approximately 17 700 mg l
1
) and Arizona (maximum
concentrations of approximately 380 000 mg l
1
). Because
this technology addresses a unique monitoring niche for
measurements in highly or hyperconcentrated flows and be-
cause of large benefits associated with the production of a
dense time series of surrogate measurements, it remains under
consideration for future testing and use.
Acoustic backscatter technology shows the most promise for
meeting the needs of large-scale fluvial sediment monitoring
programs. The technology measures several orders of magni-
tude more flow than those technologies associated with point
measurements. SSC data computed from backscatter data ob-
tained using a three-frequency instrument array and appropri-
ate postprocessing techniques range from 10 to 20 000 mg l
1
(silt- and clay-size material) and 10–3000 mg l
1
(sand-size
material). These data are deemed by the principal investigators
to be at least as accurate – within 5% – as measurements made
by traditional techniques. At present, the cost of using a three-
frequency Doppler array (three separate instruments) is ap-
proximately sixfold that of a fully equipped in situ turbidimeter.
Although at least one multifrequency acoustic backscatter meter
is commercially available, it lacks Doppler velocity capability.
Multifrequency Doppler velocity profiler is becoming com-
mercially available; however, the signal processing does not as
yet support analysis for sediment properties. Fortunately, there
are indications that development of applicable, self-
contained, multifrequency Doppler velocity units are planned,
making more economic monitoring of sediment transport
possible in the future, at least under some hydrological and
sedimentological conditions.
Most suspended sediment data obtained by Federal agen-
cies at present have their underpinnings in instruments and
techniques conceived before the mid-1940s. Hence, the pro-
spect of broad application of one or more suspended sedi-
ment surrogate technologies presented herein – and perhaps
others in development – is a revolutionary concept in fluvial
sedimentology. The benefits of such applied capability could
be enormous, providing for safer, more frequent and con-
sistent, arguably more accurate, and ultimately less expensive
fluvial data collection for use in managing the world’s sedi-
mentary resources.
References
Advisory Committee on Water Information, Subcommittee on Sedimentation (2012).
Home page. Available at http://acwi.gov/sos/ (accessed on 12 June 2012).
Agrawal, Y. C. and Pottsmith, H. C. (1994). Laser diffraction particle sizing in
STRESS. Continental Shelf Research 14, 1101–1121.
Agrawal, Y. C. and Pottsmith, H. C. (2000). Instruments for particle size and
settling velocity observations in sediment transport. Marine Geology 168/1–4,
89–114.
Agrawal, Y. C. and Pottsmith, H. C. (2006). The isokinetic streamlined suspended-
sediment profiling LISST-SL – Status and field results. Proceedings of the 8th
Federal Interagency Sedimentation Conference, pp. 288–295, Reno, NV, 2–6
April. Available at http:/.pubs.usgs.gov/misc_reports/FISC_1947-2006/pdf/1st-
7thFISCs-CD/8thFISC/Session%202C-1_Agrawal.pdf (accessed on 14 June
2012).
Agrawal, Y. C., Whitemore, A., Mikkelsen, O. A. and Pottsmith, H. C. (2008). Light
scattering by random shaped particles and consequences on measuring
suspended sediments by laser diffraction. Journal of Geophysical Research 113,
11, C04023.
American Public Health Association, American Water Works Association, and Water
Pollution Control Federation (1995). Standard methods for the examination of
water and wastewater (19th edn.). Washington, DC: American Public Health
Association.
American Society of Civil Engineers (2006). Sedimentation engineering. In Vanoni,
V. (ed.) ASCE Manuals and Reports on Engineering Practice No. 54, 418 pp.
New York, NY, ISBN: 0-7844-0823-8.
Anderson, C. A. (2005). Turbidity. U.S. Geological Survey National Field Manual for
the Collection of Water-Quality Data. U.S. Geological Survey Techniques of
Water-Resources Investigations, Book 9, ch. 6.7. Available at http://
water.usgs.gov/owq/FieldManual/Chapter6/6.7_contents.html (accessed on 14
June 2012).
ASTM International (1999). Terminology for fluvial sediment: Active standard
D4410-98. West Conshohocken, PA: ASTM International.
Bent, G. C., Gray, J. R., Smith, K. P. and Glysson, G. D. (2003). A synopsis of
technical issues for monitoring sediments in highway and urban runoff. In
Granato, G. E., Zenone, C. and Cazenas, P. E. (eds.) The national highway runoff
data and methodology synthesis, vol. 1, pp. 113–163. U.S. Department of
Transportation Publication No. FHWA-EP-03-054; also available as U.S.
Geological Survey Open-File Report 00-497. Washington, DC: Office of Natural
and Human Environment. Available at http://ma.water.usgs.gov/fhwa/products/
ofr00497.pdf (accessed on 14 June 2012).
Beverage, J. P. and Culbertson, J. K. (1964). Hyperconcentrations of suspended
sediment. Journal of the Hydraulics Division, Proceedings of the American
Society of Civil Engineers 90, 117–128.
Bogen, J., Fergus, T. and Walling, D. E. (2003). Erosion and sediment transport in
rivers, technological and methodological advances. International Association of
Hydrological Sciences Publication 283, 238 pp. Wallingford, OX: IAHS Press,
Centre for Ecology and Hydrology.
Buchanan, P. A. and Morgan, T. L. (2011). Summary of suspended-sediment
concentration data, San Francisco bay, California, water year 2008. U.S.
Geological Survey Data Series Report 634. Available at http://pubs.usgs.gov/ds/
634/ (accessed on 29 June 2012).
Capel, P. D. and Larsen, S. J. (1996). Evaluation of selected information on
splitting devices for water samples. U.S. Geological Survey Water-Resources
Investigations Report 95-4141, 103 pp. Available at http://pubs.usgs.gov/wri/
1995/4141/report.pdf (accessed on 13 June 2012).
Capel, P. D. and Nacionales, F. (1995). Precision of a splitting device for water
samples. U.S. Geological Survey Open-File Report 95-293, 23 pp. Available at
http://pubs.usgs.gov/of/1995/0293/report.pdf (accessed on 13 June 2012).
Christensen, V. G., Xiaodong, J. and Ziegler, A. C. (2000). Regression analysis and
real-time water-quality monitoring to estimate constituent concentrations, loads,
and yields in the little Arkansas river, south-central Kansas, 1995–99. U.S.
Geological Survey Water-Resources Investigations Report 00-4126, 36 pp.
Measuring Suspended Sediment 199
Author's personal copy
Available at http://ks.water.usgs.gov/Kansas/pubs/reports/wrir.00-4126.html
(accessed on 13 June 2012).
Colby, B. R. (1963). Fluvial sediments – A summary of source, transportation,
deposition, and measurement of sediment transport. U.S. Geological Survey
Bulletin 1181-A, 47 pp. Available at http://pubs.usgs.gov/bul/1181a/report.pdf
(accessed on 12 June 2012).
Conner, C. S. and De Visser, A. M. (1992). A laboratory investigation of particle
size effects on optical backscatterance sensor. Marine Geology 108, 151–159.
Crawford, A. M. and Hay, A. E. (1993). Determining suspended sand size and
concentration from multi-frequency acoustic backscatter. Journal of the
Acoustical Society of America 94, 3312–3324.
Culbertson, J. K., Scott, C. H. and Bennett, J. P. (1972). Summary of alluvial-
channel data from Rio Grande conveyance channel, New Mexico, 1965–69. U.S.
Geological Survey Professional Paper 562-J, 48 pp. Available at http://
pubs.usgs.gov/pp/0562j/report.pdf (accessed on 24 May 2012).
D&A Instrument Company (1991). OBS-1
&
3 Instruction Manual. Port Townsend,
WA: D & A Instrument Company. 41 pp.
Davis, B. E. (2001). The US D-96: An isokinetic suspended-sediment/water-quality
collapsible bag sampler. Federal Interagency Sedimentation Project Report PP,
Vicksburg, MS, USACE Engineer, and Research Development Center, 37 pp.
Available at http://water.usgs.gov/fisp/docs/Report_PP,_US_D-96_011114_.pdf
(accessed on 12 June 2012).
Davis, B. E. (2005). A guide to the proper selection and use of federally approved
sediment and water-quality samplers. U.S. Geological Survey Open-File Report
2005-1087, 20 pp. Available at http://pubs.usgs.gov/of/2005/1087/ (accessed
on 14 June 2012).
Day, T. J. (1991). Sediment monitoring in Canada. Proceedings of the Fifth Federal
Interagency Sedimentation Conference, Subcommittee on Sedimentation of the
Interagency Advisory Committee on Water Data. Las Vegas, NV. vol. I, pp. 1-
14–1-29. Available at http://pubs.usgs.gov/misc/FISC_1947-2006/pdf/1st-
7thFISCs-CD/5thFISC/5Fisc-V1/5Fisc1-1.PDF (accessed on 13 June 2013).
Design Analysis Associates (2012). Home page. Available at http://
www.waterlog.com/ (accessed on 14 June 2012).
Diplas, P., Kuhle, R. A., Gray, J. R., Glysson, G. D. and Edwards, T. K. (2008).
Sediment transport measurements. In Garcia, M. (ed.) Sedimentation engineering
– Processes, measurements, modeling, and practice. American Society of Civil
Engineers Manual 110, ch. 5, pp. 307–353. Reston, Virginia: ASCE. Available at
http://water.usgs.gov/osw/techniques/Diplas_Kuhnle_others.pdf (accessed on 18
May 2012).
Downing, J. P. (1983). An optical instrument for monitoring suspended particles in
ocean and laboratory. Proceedings of Oceans 1983. San Francisco, CA, 29
August–1 September, pp. 199–202. DOI: 10.1109/OCEANS.1983.1152088.
Downing, J. P. (1996). Suspended sediment and turbidity measurements in streams:
What they do and do not mean. Paper Presented at Automatic Water Quality
Monitoring Workshop, Richmond, BC, 12–13 February.
Downing, J. P., Sternberg, R. W. and Lister, C. R. B. (1981). New instrumentation
for the investigation of sediment suspension processes in the shallow marine
environment. Marine Geology 42, 19–34.
Downing, A., Thorne, P. D. and Vincent, C. E. (1995). Backscattering from a
suspension in the near field of a piston transducer. Journal of the Acoustical
Society of America 97(3), 1614–1620.
Edwards, T. E. and Glysson, G. D. (1999). Field methods for measurement of fluvial
sediment. U.S. Geological Survey Techniques of Water-Resources Investigations,
Book 3, ch. C2, 89 pp. Available at http://water.usgs.gov/osw/techniques/
sedimentpubs.html (accessed on 12 June 2012).
Federal Interagency Sedimentation Project (1940). Field practice and equipment
used in sampling suspended sediment. Interagency Report No. 1, Hydraulics
Laboratory, Iowa University, Iowa City, 175 pp. Available at http://water.usgs.gov/
fisp/docs/Report_1.pdf (accessed on 12 June 2012).
Federal Interagency Sedimentation Project (1941). Laboratory investigation of
suspended-sediment samplers. Interagency Report No. 5, Hydraulics Laboratory,
Iowa University, Iowa City, 99 pp. Available at http://water.usgs.gov/fisp/docs/
Report_5.pdf (accessed on 12 June 2012).
Federal Interagency Sedimentation Project (1952). The design of improved types of
suspended-sediment samplers. Interagency Report No. 6, St. Anthony Falls
Hydraulics Laboratory, Minneapolis, MN, 103 pp. Available at http://
water.usgs.gov/fisp/docs/Report_6.pdf (accessed on 12 June 2012).
Federal Interagency Sedimentation Project (1961). The single-stage sampler for
suspended sediment. Interagency Report 13, St. Anthony Falls Hydraulics
Laboratory, Minneapolis, MN, 105 pp. Available at http://water.usgs.gov/fisp/
docs/Report_13.pdf (accessed on 13 June 2012).
Federal Interagency Sedimentation Project (1963). Determination of fluvial sediment
discharge. Interagency Report 14, St. Anthony Falls Hydraulics Laboratory,
Minneapolis, MN, 151 pp. Available at http://water.usgs.gov/fisp/docs/
Report_14.pdf (accessed on 12 June 2012).
Federal Interagency Sedimentation Project (1981). Test and design of automatic
fluvial suspended-sediment samplers. Interagency Report W, St. Anthony Falls
Hydraulics Laboratory, Minneapolis, MN, 53 pp. Available at http://
water.usgs.gov/fisp/docs/Report_W.pdf (accessed on 14 June 2012).
Federal Interagency Sedimentation Project (2012). World Wide Web home page:
U.S. Geological Survey. Available at http://water.usgs.gov/fisp/ (accessed on 18
May 2012).
Fisher, F. H. and Simmons, V. P. (1977). Sound absorption in sea water. Journal of
the Acoustical Society of America 62(3), 558–564.
Flagg, C. N. and Smith, S. L. (1989). On the use of the acoustic Doppler current
profiler to measure zooplankton abundance. Deep–Sea Research 36(3), 455–474.
Flammer, G. H. (1962). Ultrasonic measurement of suspended sediment. U.S.
Geological Survey Bulletin 1141-A, 48 pp. Washington, DC: Government
Printing Office. Available at http://pubs.usgs.gov/bul/1141a/report.pdf (accessed
on 14 June 2012).
Gartner, J. W. (2004). Estimating suspended solids concentrations from backscatter
intensity measured by acoustic Doppler current profiler in San Francisco bay,
California. Marine Geology 211, 169–187.
Gartner, J. W., Cheng, R. T., Wang, P. F. and Richter, K. (2001). Laboratory and
field evaluations of LISST-100 instrument for suspended particle size
determinations. Marine Geology 175/1–4, 199–219.
Gartner, J. W., Mueller, D. S., Wall, G. R. and Gray, J. R. (2003). Breakout session
4: Other fluvial-sediment surrogates. Proceedings of the Federal Interagency
Workshop on Turbidity and Other Sediment Surrogates. Reno, NV, 30 April–2
May 2002, pp. 21–27. Available at http://pubs.usgs.gov/circ/2003/circ1250/
(accessed on 13 June 2012).
Glysson, G. D. (1989a). Criteria for a sediment data set. Proceedings of the
International Symposium on Sediment Transport Modeling, Hydraulics Division,
the American Society of Civil Engineers, New Orleans, Louisiana, 14–18 August,
7 pp.
Glysson, G. D. (1989b). 100 years of sedimentation study by the USGS. In: Wang,
S. S. Y. (eds.) Proceedings of the International Symposium, Sediment Transport
Modeling, 14–18 August, pp. 260–265. New Orleans: American Society of Civil
Engineers. Available at http://kleene.er.usgs.gov/sdct/images/2/21/
Glysson_100Years.pdf (accessed on 12 June 2012).
Glysson, G. D. and Gray, J. R. (1997). Coordination and standardization of federal
sedimentation activities. In Gray, J. R. (ed.) Expanding Sediment Research
Capabilities in Today’s U.S. Geological Survey. Technical Committee of the
Subcommittee on Sedimentation, St. Petersburg, FL, 17–19 February,
3 pp. Available at http://water.usgs.gov/osw/techniques/workshop/glysson.html
(accessed on 28 January 2013).
Glysson, G. D., Gray, J. R. and Conge, L. M. (2000). Adjustment of total
suspended solids data for use in sediment studies. Proceedings of the ASCE’s
2000 Joint Conference on Water Resources Engineering and Water Resources
Planning and Management. Minneapolis, MN, 30 July–2 August 2000, 10 pp.
Available at http://water.usgs.gov/osw/pubs/ASCEGlysson.pdf (accessed on 12
June 2012).
Glysson, G. D., Gray, J. R., and Schwarz, G. E. (2001). A comparison of load
estimates using total suspended solids and suspended–sediment concentration
data. Proceedings of the ASCE World Water
&
Environmental Resources
Congress. Orlando, FL, 20–24 May, 9 pp. Available at http://water.usgs.gov/osw/
pubs/TSS_Orlando.pdf (accessed on 12 June 2012).
Gray, J. R. (2003a). The need for sediment surrogate technologies to monitor
fluvial-sediment transport. In: Gray, J. R. and Glysson, G. D. (eds.) Proceedings
of the Federal Interagency Sedimentation Workshop on Turbidity and Other
Sediment Surrogates, 30 April–2 May 2002, U.S. Geological Survey Circular
1250, 4 pp. Reno, NV: U.S. Geological Survey. Available at http://
water.usgs.gov/osw/techniques/TSS/gray.pdf (accessed on 28 January 2013).
Gray, J. R. (2003b). U.S. Geological Survey suspended-sediment surrogate
research, Part I: Call for a sediment monitoring instrument and analysis research
program. In: Jane W. and Judy P. (eds.) Proceedings of the Virginia Water
Research Symposium 2003, Water Resource Management for the
Commonwealth, 8–10 October, pp. 53–57. Blacksburg, VA: Virginia Polytechnic
Institute and State University. Available at http://vwrrc.vt.edu/pdfs/proceedings/
2003WaterResearchSymposium_proceedings.pdf (accessed on 13 June 2013).
Gray, J. R. (ed.) (2005). Proceedings of the Federal Interagency Sediment
Monitoring Instrument and Analysis Research Workshop. Flagstaff, AZ, 9–11
September, U.S. Geological Survey Circular 1276, 46 pp. Available at http://
pubs.usgs.gov/circ/2005/1276/ (accessed on 13 June 2012).
Gray, J. R., Agrawal, Y. C. and Pottsmith, H. C. (2004). The LISST-SL streamlined
isokinetic suspended-sediment profiler. In: Liu, C. (ed.) Proceedings of the 9th
200 Measuring Suspended Sediment
Author's personal copy
International Symposium on River Sedimentation. 18–21 October, vol. IV,
pp. 2549–2555. Yichang, China: Tsinghua University Press. Available at http://
water.usgs.gov/osw/techniques/Lisst_Gray_Agrawal_Pottsmith.pdf (accessed on
14 June 2012).
Gray, J. R. and Fisk, G. G. (1992). Monitoring radionuclide and suspended-
sediment transport in the little Colorado river basin, Arizona and New Mexico,
USA. IAHS Publication No. 210, pp. 505–516. Wallingford, OX: IAHS Press,
Institute of Hydrology.
Gray, J. R. and Gartner, J. W. (2004). Surrogate technologies for continuous
suspended-sediment monitoring in the United States. In Cheng, L. (ed.)
Proceedings of the 9th International Symposium on River Sedimentation, 18–21
October, vol. IV, pp. 2515–2522. Yichang, China: Tsinghua University Press.
Gray, J. R. and Glysson, G. D. (2004). Traditional and new techniques for deriving
sediment-discharge information in the United States (invited lecture). In
Tan, Y. (ed.) Proceedings of the 9th International Symposium on River
Sedimentation, 18–21 October, vol. 1, pp. 493–504. Yichang, China:
Tsinghua University Press.
Gray, J. R., Glysson, G. D. and Turcios, L. M. (2000). Comparability of total
suspended solids and suspended-sediment concentration data. U.S. Geological
Survey Water-Resources Investigations Report 00-4191, 14 pp. Available at
http://water.usgs.gov/osw/pubs/WRIR00-4191.pdf (accessed on 12 June 2012).
Gray, J. R., Glysson, G. D. and Edwards, T. E. (2008). Suspended-sediment
samplers and sampling methods. In Garcia, M. (ed.) Sediment transport
measurements. Sedimentation engineering – Processes, measurements,
modeling, and practice. American Society of Civil Engineers Manual 110, ch.
5.3, pp. 320–339. Available at http://water.usgs.gov/osw/techniques/
Diplas_Kuhnle_others.pdf (accessed on 18 May 2012).
Gray, J. R. and Shadie, C. (2010). Determination of daily sediment, nutrient, and
sediment-associated chemical concentrations and loads for the conterminous
United States – A proposal to establish a long-term, base-funded, network-
design national monitoring network to generate sediment, nutrient, and
sediment-associated chemical concentrations, loads, budgets and temporal
trends. Piloted in the Mississippi river basin: http://acwi.gov/sos/minutes/
index.html (accessed on 18 May 2012, under the 25 June 2010, subcommittee
on sedimentation meeting minutes).
Gray, J. R. and Glysson, G. D. (2005). Attributes of a sediment monitoring
instrument and analysis research program. Proceedings of the Federal
Interagency Sediment Monitoring Instrument and Analysis Research Workshop.
Flagstaff, AZ, 9–11 September, 6 pp. Available at http://water.usgs.gov/osw/
techniques/sediment/sedsurrogate2003workshop/gray_glysson.pdf (accessed on
13 June 2012).
Gray, J. R., Glysson, G. D. and Mueller, D. S. (2002). Comparability and accuracy
of fluvial-sediment data – A view from the U.S. Geological Survey. Proceedings
of the ASCE Specialty Conference, Hydraulic Measurements
&
Experimental
Methods. Estes Park, CO, 28 July–1 August, 6 pp. Available at http://
water.usgs.gov/osw/techniques/asce.pdf (accessed on 28 January 2013).
Gray, J. R. and Gartner, J. W. (2009). Technological advances in suspended-
sediment surrogate monitoring. Water Resources Research 45, W00D29.
Available at http://water.usgs.gov/osw/techniques/2008WR007063.pdf (accessed
on 13 June 2012).
Gray, J. R., Gooding, D. J., Mellis, T. S., Topping, D. J. and Rasmussen, P. P.
(2003). U.S. Geological Survey suspended-sediment surrogate research, Part II:
Optic technologies. In Jane W. and Judy P. (ed.) Proceedings of the Virginia
Water Research Symposium 2003, Water Resource Management for the
Commonwealth. Virginia Polytechnic Institute and State University,
Blacksburg, 8–10 October, pp. 58–64. Available at http://vwrrc.vt.edu/pdfs/
proceedings/2003WaterResearchSymposium_proceedings.pdf (accessed on 13
June 2013).
Gray, J. R. and Simo
˜es, F. J. M. (2008). Estimating sediment discharge. In Garcia,
M. (ed.) Sedimentation engineering – Processes, measurements, modeling, and
practice. American Society of Civil Engineers Manual 110, Appendix D,
pp. 1067–1088. Reston, Virginia: ASCE. Available at http://water.usgs.gov/osw/
techniques/Gray_Simoes.pdf (accessed on 13 June 2012).
Gray, J. R., Laronne, J. B., Marr, J. D. G. (2010). Bedload-surrogate monitoring
technologies. U.S. Geological Survey Scientific Investigations Report 2010-5091,
37 pp. Available at http://pubs.usgs.gov/sir/2010/5091; the papers listed in
Table 2 are available only online from http://pubs.usgs.gov/sir/2010/5091/
papers/listofpapers.html (accessed on 28 January 2013).
Gray, J. R. and Gartner, J. W. (2010a). Surrogate technologies for monitoring bed-
load transport in rivers. In Poleto, C. and Charlesworth, S. (eds.),
Sedimentology of aqueous systems, ch. 2, pp. 45–79. London: Wiley-Blackwell.
Available at http://water.usgs.gov/osw/techniques/sed_aq_sys_chap_2_pdf_
from_wb_3_16_2010.pdf (accessed on 13 June 2012).
Gray, J. R. and Gartner, J. W. (2010b). Surrogate technologies for monitoring
suspended-sediment transport in rivers. In Poleto, C. and Charlesworth, S.
(eds.) Sedimentology of aqueous systems. ch. 1, pp. 3–45. London: Wiley-
Blackwell. Available at http://water.usgs.gov/osw/techniques/sed_aq_sys_chap_
1_pdf_from_wb_3_16_2010.pdf (accessed on 13 June 2012).
Gray, J. R. and Gartner, J. W. (2010c). Overview of selected surrogate technologies
for high-temporal resolution suspended-sediment monitoring. Proceedings of the
9th Federal Interagency Sedimentation Conference, Las Vegas, NV, 27 June–1
July, 15 pp.
Guerrero, M., Szupiany, R. N. and Amsler, M. (2011). Comparison of acoustic
backscattering techniques for suspended sediments investigation. Flow
Measurement and Instrumentation 22(5), 392–401.
Guy, H. P. (1969). Laboratory theory and methods for sediment analysis. U.S.
Geological Survey Techniques of Water-Resources Investigations, Book 5, ch.
C1, 58 pp. Available at http://pubs.usgs.gov/twri/twri5c1/ (accessed on 12 June
2012).
Guy, H. P. (1970). Fluvial sediment concepts. U.S. Geological Survey Techniques of
Water Resources Investigations, Book 3, ch. C1, 55 pp. Available at http://
pubs.usgs.gov/twri/twri3-c1/ (accessed on 12 June 2012).
Guy, H. P. and Norman, V. W. (1970). Field methods for measurement of fluvial
sediment, U.S. Geological Survey Techniques of Water-Resources Investigations
Report, Book 3, ch. C2, 59 pp.
Hay, A. E. (1983). On the remote acoustic detection of suspended sediment at long
wavelengths. Journal of Geophysical Research 88(C12), 7525–7542.
Hay, A. E. and Sheng, J. (1992). Vertical profiles of suspended sand concentration
and size from multi-frequency acoustic backscatter. Journal of Geophysical
Research 97(C10), 15661–15677.
Holdaway, G. P., Thorne, P. D., Flatt, David, Jones, S. E. and Prandle, D. (1999).
Comparison between ADCP and transmissometer measurements of
suspended sediment concentration. Continental Shelf Research 19,
421–441.
Horowitz, A. J. (2008). Determining annual suspended sediment and sediment-
associated trace element and nutrient fluxes. Science of the Total Environment
400(1–3), 315–343, Available at http://www.sciencedirect.com/science/article/pii/
S0048969708004543 (accessed on 12 June 2012).
Horowitz, A. J., Smith, J. J. and Elrick, K. A. (2001). Selected laboratory
evaluations of the whole-water sample-splitting capabilities of a prototype
fourteen-liter Teflon
s
churn splitter. U.S. Geological Survey Open-File Report
01-386, 12 pp. Available at http://pubs.usgs.gov/of/2001/ofr01-386/ (accessed
on 12 June 2012).
Hubbell, D. W., Jordan, P. R., Culbertson, D. M. and Hembree, C. H. (1956).
Investigations of some sedimentation characteristics of a sand-bed stream.
Progress Report No. 1, U.S. Geological Survey Open-file Report, Lincoln, NE,
78 pp (the authorship is inferred from page 2 of the text, which is officially
listed as ‘D. M. Hubbell and others’). Available at http://pubs.usgs.gov/of/1956/
hubbell/Hubbell_et_al_1956.pdf (accessed on 25 June 2012).
International Standards Organization (ISO) (2002). Methods for measurement
of suspended sediment, 27 pp. Available at http://www.iso.org/iso/iso_
catalogue/catalogue_tc/catalogue_detail.htmcsnumber=31797 (accessed on
21 May 2012).
Jastram, J. D., Moyer, D. L. and Hyer, K. E. (2009). A comparison of streamflow-
based and turbidity-based estimates of suspended sediment concentrations in
three Chesapeake bay tributaries. U.S. Geological Survey Scientific Investigations
Report 2009-5165. Available at http://pubs.usgs.gov/sir/2009/5165/ (accessed
on 14 June 2012).
Jelinski, J. C. (2008). Electric reel drive. U.S. Geological Survey, Hydrological
Instrumentation Facility Preliminary Information Sheet. Available at http://
1stop.usgs.gov/uo/publications/hifpubs/techsheets/ElectricReelDriveTechSheet
Revision-04012008.pdf (accessed on 13 June 2012).
Johnson, G. P. (1997). Instruction manual for U.S. Geological Survey sediment
observers. U.S. Geological Survey Open-File Report 96-431, 33 pp. Available at
http://pubs.usgs.gov/of/1996/0431/report.pdf (accessed on 12 June 2012).
Kennedy, E. J. (1984). Discharge ratings at gaging stations. U.S. Geological Survey,
Techniques of Water-Resources Investigations, Book 3, ch. A10, 59 pp. Available
at http://pubs.usgs.gov/twri/twri3-a10/ (accessed on 12 June 2012).
Kineke, G. C. and Sternberg, R. W. (1992). Measurements of high concentration
suspended sediments using the optical backscatterance sensor. Marine Geology
108, 253–258.
Knott, J. M., Glysson, G. D., Malo, B. A., and Schroder, L. J. (1993). Quality
assurance plan for the collection and processing of sediment data by the U.S.
Geological Survey, Water Resources Division. U.S. Geological Survey Open-File
Report 92-499, 18 pp. Available at http://pubs.usgs.gov/of/1992/0499/report.pdf
(accessed on 12 June 2012).
Measuring Suspended Sediment 201
Author's personal copy
Knott, J. M., Sholar, C. J. and Matthes, W. J. (1992). Quality assurance
guidelines for the analysis of sediment concentration by U.S. Geological
Survey sediment laboratories. U.S. Geological Survey Open-File Report
92-33, 30 pp. Available at http://pubs.usgs.gov/of/1992/0033/report.pdf
(accessed on 12 June 2012).
Koltun, G. F., Eberle, M., Gray, J. R. and Glysson, G. D. (2006). User’s
manual for the graphical constituent loading analysis system (GCLAS). U.S.
Geological Survey Techniques and Methods, Book 4, ch. C1, 50 pp.
Available at http://pubs.er.usgs.gov/usgspubs/tm/tm4C1 (accessed on 14
June 2012).
Konrad, C., Pottsmith, H. C. Mellis, T. S. and Rubin, D. M. (2006). Real-time
analysis of concentrated fluvial suspended sediments. Proceedings of the 8th
Federal Interagency Sedimentation Conference, Reno, NV, 2–6 April, pp.
585–591. Available at http://pubs.usgs.gov/misc_reports/FISC_1947-2006/pdf/
1st-7thFISCs-CD/8thFISC/Session%208C-2_Konrad.pdf (accessed on 14 June
2012).
Landers, M. N. (2003). Summary of blind sediment reference sample measurement
session. In: Gray, J. R. and Douglas Glysson, G. (eds.) Proceedings of the
Federal Interagency Workshop on Turbidity and Other Sediment Surrogates,
Reno, NV, 30 April–2 May, U.S. Geological Survey Circular 1250, pp. 29–30.
Available at http://water.usgs.gov/osw/techniques/turbidity.html (accessed on 13
June 2012).
Landers, M. N. (2012). Fluvial suspended sediment characteristics by high-
resolution, surrogate metrics of turbidity, laser-diffraction, acoustic backscatter,
and acoustic attenuation. Ph.D. Dissertation, Georgia Institute of Technology.
Available at http://hdl.handle.net/1853/43747 (accessed on 14 June 2012).
Landers, M. N., Arrigo, J., and Gray, J. R. (2012). Advancing hydroacoustics
technologies for sedimentary research and monitoring. Joint CUAHSI-USGS
Workshop on Sediment Hydroacoustic Techniques in Rivers and Streams,
Shepherdstown, West Virginia, 20–22 March 2012. EOS, Transactions, American
Geophysical Union, vol. 83, No. 26, p. 244.
Lane, S. L., Flanagan, S. and Wilde, F. D. (2003). Selection of equipment for
sampling (ver. 20). US Geological Survey Techniques of Water-Resources
Investigations, vol. 9, chap. A2, Reston, VA. Available at http://water.usgs.gov/
owq/FieldManual/Chapter2/Ch2_contents.html (accessed on 28 January 2013).
Larsen, M. C., Figueroa-Alamo, C., Gray, J. R. and Fletcher, W. (2001). Continuous
automated sensing of streamflow density as a surrogate for suspended-sediment
concentration sampling. Proceedings of the 7th Federal Interagency
Sedimentation Conference, Reno, NV, 25–29 March, vol. I, pp. III-102–III-109.
Available at http://pubs.usgs.gov/misc_reports/FISC_1947-2006/pdf/1st-
7thFISCs-CD/7thFISC/7Fisc-V1/7FISC1-3.pdf (accessed on 14 June 2012).
Lewis, A. J. and Rasmussen, T. C. (1999). Determination of suspended sediment
concentrations and particle size distributions using pressure measurements.
Journal of Environmental Quality 28, 1490–1496.
Lewis, J. (1996). Turbidity-controlled suspended sediment sampling for runoff-event
load estimation. Water Resources Research 32(7), 2299–2310.
Lewis, J. (2002). Estimation of suspended sediment flux in streams using
continuous turbidity and flow data coupled with laboratory concentrations. In:
Gray, J. R. and Douglas Glysson, G. (eds.) Proceedings of the Federal
Interagency Workshop on Turbidity and Other Sediment Surrogates, Reno, NV,
30 April–2 May, U.S. Geological Survey Circular 1250, 3 pp. Available at http://
water.usgs.gov/osw/techniques/TSS/LewisTSS.pdf (accessed on 13 June 2012).
Ludwig, K. A. and Hanes, D. M. (1990). A laboratory evaluation of optical
backscatterance suspended solids sensors exposed to sand-mud mixtures.
Marine Geology 94, 173–179.
Lynch, J. F., Irish, J. D., Sherwood, C. R. and Agrawal, Y. C. (1994). Determining
suspended sediment particle size information from acoustical and optical
backscatter measurements. Continental Shelf Research 14(10/11), 1139–1165.
Matthes, W. J., Sholar, C. J. and George, J. R. (1992). Quality assurance plan for
analysis of fluvial sediment by laboratories of the U.S. Geological Survey. U.S.
Geological Survey Open-File Report 91-467, 31 pp. Available at http://
pubs.usgs.gov/of/1991/0467/report.pdf (accessed on 12 June 2012).
McGregor, J. (2000a). Development of the US DH-95 suspended-sediment sampler.
Federal Interagency Sedimentation Project, Report MM, 27 pp. Available at
http://water.usgs.gov/fisp/docs/Report_MM_US_DH-95_000619.pdf (accessed
on 12 June 2012).
McGregor, J. (2000b). Development of the US D-95 suspended-sediment sampler.
Federal Interagency Sedimentation Project, Report LL, 24 pp. Available at http://
water.usgs.gov/fisp/docs/Report_LL_US_D-95_000619.pdf (accessed on 12
June 2012).
Melis, T. S., Topping, D. J. and Rubin, D. M. (2003). Testing laser-based sensors
for continuous in situ monitoring of suspended sediment in the Colorado river,
Arizona. In Bogen, J., Fergus, T. and Walling, D. E. (eds.) Erosion and sediment
transport measurement in rivers, technological and methodological advances.
International Association of Hydrological Sciences Publication 283, pp. 21–27.
Moody, J. A. and Meade, R. H. (1994). Evaluation of the method of collecting
suspended sediment from large rivers by discharge-weighted pumping and
separation by continuous-flow centrifugation. Hydrological Processes 8,
513–530.
Nelson, M. E. and Benedict, P. C. (1950). Measurement and analysis of suspended
sediment loads in streams. American Society of Civil Engineers, Transactions,
Paper No. 2450, pp. 891–918.
Nolan, K. M., Gray, J. R. and Glysson, G. D. (2005). Introduction to suspended-
sediment sampling. U.S. Geological Survey Scientific Investigations Report
2005-5077. Available on CD-ROM at http://pubs.er.usgs.gov/pubs/sir/
sir20055077 (accessed on 28 January 2013).
Nordin, C. F., Meade, R. H. and Curtis, C. C. (1983). Data from sediment studies
of the Rio Orinoco, Venezula, 15–25 August, 1982. U.S. Geological Survey
Open-File Report 83-679, 25 pp. Available at http://pubs.usgs.gov/of/1983/
0679/report.pdf (accessed on 12 June 2012).
Osterkamp, W. R. and Emmett, W. W. (1992). The vigil network: Erosion and
sediment transport monitoring programmes in river basins. Proceedings of the
Oslo Symposium. International Association of Hydrological Sciences Publication
210, Oslo, Norway, pp. 397–404. Wallingford, OX: IAHS Press, Institute of
Hydrology.
Osterkamp, W. R., Heilman, P. and Lane, L. J. (1998). Economic considerations of a
continental sediment-monitoring program. International Journal of Sediment
Research 13(4), 12–24.
Osterkamp, W. R., Heilman, P., Gray, J. R., et al. (2004). An invitation to participate
in a North American sediment-monitoring network. Eos, Transactions, American
Geophysical Union 84(40), 386–389.
Osterkamp, W. R. and Parker, R. S. (1991). Sediment monitoring in the United
States. Proceedings of the 5th Federal Interagency Sedimentation Conference.
Available at http://pubs.usgs.gov/misc/FISC_1947-2006/pdf/1st-7thFISCs-CD/
5thFISC/5Fisc-V1/5Fisc1-1.PDF (accessed on 18 May 2012).
Porterfield, G. (1972). Computation of fluvial-sediment discharge. U.S.
Geological Survey, Techniques of Water-Resource Investigations, Book 3,
ch. C3, 66 pp. Available at http://water.usgs.gov/pubs/twri/twri3-c3/ (accessed
on 14 June 2012).
Pratt, T. and Parchure, T. (2003). OBS calibration and field measurements. In: Gray,
J. R. and Douglas Glysson, G. (eds.) Proceedings of the Federal Interagency
Workshop on Turbidity and Other Sediment Surrogates, Reno, NV, 30 April–2
May, U.S. Geological Survey Circular 1250, 3 pp. Available at http://
water.usgs.gov/pubs/circ/2003/circ1250/ (accessed on 13 June 2012).
Pruitt, B. A. (2003). Uses of turbidity by states and tribes. In: Gray, J. R. and
Douglas Glysson, G. (eds.) Proceedings of the Federal Interagency Workshop on
Turbidity and Other Sediment Surrogates, Reno, NV, 30 April–2 May, U.S.
Geological Survey Circular 1250, pp. 31–46. Available at http://pubs.usgs.gov/
circ/2003/circ1250/ (accessed on 13 June 2012).
Rantz, S. E. (1982). Measurement and computation of streamflow. U.S. Geological
Survey Water-Supply Paper 2175, vols 1 and 2, 631 pp. Available at http://
pubs.usgs.gov/wsp/wsp2175/ (accessed on 14 June 2012).
Rasmussen, P. P., Gray, J. R., Glysson, G. D. and Ziegler, A. C. (2009). Guidelines
and procedures for computing time-series suspended-sediment concentrations
and loads from in-stream turbidity-sensor and streamflow data. U.S. Geological
Survey Techniques and Methods, Book 3, ch. C4, 53 pp. Available at http://
pubs.usgs.gov/tm/tm3c4/ (accessed on 14 June 2012).
Rasmussen, T. J., Ziegler, A. C. and Rasmussen, P. P. (2005). Estimation of
constituent concentrations, densities, loads, and yields on lower Kansas river,
Northeast Kansas, using regression models and continuous water-quality
modeling, January 2000 through December 2003. U.S. Geological Survey
Scientific Investigations Report 2005-5165, 117 pp. Available at http://
pubs.usgs.gov/sir/2005/5165/ (accessed on 28 January 2013).
Reichel, G. and Nachtnebel, H. P. (1994). Suspended sediment monitoring in a
fluvial environment: Advantages and limitations applying an acoustic Doppler
current profiler. Water Research 28(4), 751–761.
van Rijn, L. C. (2007). Manual sediment transport measurements in rivers, estuaries
and coastal seas, 500 pp. Available at http://www.theseusproject.eu/wiki/Manual_
Sediment_Transport_Measurements_in_Rivers,_Estuaries_and_Coastal_Seas
(accessed on 19 May 2012).
Sauer, V. B. and Meyer, R. W. (1992). Determination of error in individual discharge
measurements. U.S. Geological Open-File Report 92-144, 21 pp. Available at
http://pubs.usgs.gov/of/1992/ofr92-144/ (accessed on 12 June 2012).
Schoellhamer, D. H., Mumley, T. E. and Leatherbarrow, J. E. (2007). Suspended
sediment and sediment-associated contaminants in San Francisco bay.
Environmental Research 105, 119–131.
202 Measuring Suspended Sediment
Author's personal copy
Schoellhamer, D. H. and Wright, S. A. (2003). Continuous measurement of
suspended-sediment discharge in rivers by use of optical backscatterance
sensors. In Bogen, J., et al. (eds.) Erosion and sediment transport measurement
in rivers, technological and methodological advances. International Association
of Hydrological Sciences Publication 283, pp. 28–36.
Sequoia Scientific, Inc. (2012). Home page. Available at http://www.sequoiasci.com/
default.aspxSectionName=home (accessed on 15 July 2012).
Sheng, J. and Hay, A. E. (1988). An examination of the spherical scatterer
approximation in aqueous suspensions of sand. Journal of the Acoustical
Society of America 83(2), 598–610.
Simmons, S. M., Parsons, D. R., Best, J. L., et al. (2010). Monitoring suspended
sediment dynamics using MBES. Journal of Hydraulic Engineering 136,
45–49.
Skinner, J. V. (1989). History of the federal interagency sedimentation project. In:
Wang, S. S. Y. (ed.) Proceedings of the International Symposium, Sediment
Transport Modeling, 14–18 August, pp. 266–271. New Orleans, LA: American
Society of Civil Engineers American Society of Civil Engineers.
Smith, C. K., Wren, D. and Chambers, J. (2006). Estimation of particle sizes for a
range of narrow size distributions of natural sands suspended in water using
multifrequency acoustic backscatter. Proceedings of the 8th Federal Interagency
Sedimentation Conference, Reno, NV, 2–6 April, pp. 531–538. Available at
http://pubs.usgs.gov/misc/FISC_1947-2006/pdf/1st-7thFISCs-CD/8thFISC/
Session%206C-2_Smith.pdf (accessed on 13 June 2013).
Stevens, H. H., Jr., Lutz, G. A. and Hubbell, D. W. (1980). Collapsible-bag
suspended-sediment sampler. American Society of Civil Engineers, Proceedings,
106(HY4), pp. 11–16.
Sutherland, T. F., Lane, P. M., Amos, C. L. and Downing, J. (2000). The calibration
of optical backscatter sensors for suspended sediment of varying darkness
levels. Marine Geology 162, 587–597.
Sutron Corporation (2012). Dual orifice constant flow Accubar bubble gauge and
recorder 56-0134-25-2C. Available at http://www.sutron.com/products/
DualOrificeCF_Bubbler.html (accessed on 22 May 2012).
Thomas, R. B. (1985). Estimating total suspended sediment yield with probability
sampling. Water Resources Research 21(9), 1381–1388.
Thomas, R. B. (1991). Systematic sampling for suspended sediment. In
Proceedings of the Fifth Federal Interagency Sedimentation Conference. Las
Vegas, NV, 18–21 March, vol. 1, pp. 2-17–2-24. Available at http://
pubs.usgs.gov/misc/FISC_1947-2006/pdf/1st-7thFISCs-CD/5thFISC/5Fisc-V1/
5Fisc1-2.PDF (accessed on 13 June 2013).
Thomas, R. B. and Lewis, J. (1993). A comparison of selection at list time and
time-stratified sampling for estimating suspended sediment loads. Water
Resources Research 29(4), 1247–1256.
Turcios, L. M., Gray, J. R. and Ledford, A. L. (2000). Summary of U.S. Geological
Survey on-line instantaneous fluvial sediment and ancillary data. Available at
http://water.usgs.gov/osw/sediment (accessed on 12 June 2012).
Thevenot, M. M. and Kraus, N. C. (1993). Comparison of acoustical and optical
measurements of suspended material in the Chesapeake estuary. Journal of
Marine Environmental Engineering 1, 65–79.
Thorne, P. D. and Hanes, D. M. (2002). A review of acoustic measurement of
small-scale sediment processes. Continental Shelf Research 22, 603–632.
Thorne, P. D., Hurther, D. and Moate, B. D. (2011). Acoustic inversions for
measuring boundary layer suspended sediment processes. Journal of the
Acoustical Society of America 130(3), 1188–1200.
Thorne, P. D., Vincent, C. E., Hardcastle, P. J., Rehman, S. and Pearson, N. (1991).
Measuring suspended sediment concentrations using acoustic backscatter
devices. Marine Geology 98, 7–16.
Topping, D. J., Melis, T. S., Rubin, D. M. and Wright, S. A. (2004). High-resolution
monitoring of suspended-sediment concentration and grain size in the Colorado
river in Grand Canyon using a laser acoustic system. Proceedings of the 9th
International Symposium on River Sedimentation, 18–21 October,
pp. 2507–2514. Yichang, China: Tsinghua University Press.
Topping, D. J., Rubin, D. M., Wright, S. A. and Melis, T. S. (2011). Field
evaluation of the error arising from inadequate time averaging in the standard
use of depth-integrating suspended-sediment samplers. U.S. Geological Survey
Professional Paper 1774, 95 pp. Available at http://pubs.usgs.gov/pp/1774/
(accessed on 12 June 2012).
Topping, D. J., Wright, S. A., Melis, T. S. and Rubin, D. M. (2007). High-resolution
measurement of suspended-sediment concentrations and grain size in the
Colorado river in Grand Canyon using a multi-frequency acoustic system.
Proceedings of the 10th International Symposium on River Sedimentation.
Moscow, Russia, 1–4 August, vol. III, pp. 330–339.
Traykovski, Peter, Latter, R. J. and Irish, J. D. (1999). A laboratory evaluation of the
LISST instrument using natural sediments. Marine Geology 159, 355–367.
Uhrich, M. A. (2002). Determination of total and clay suspended-sediment loads
from instream turbidity data in the North Santiam river basin, Oregon,
1998–2002. In: Gray, J. R. and Douglas Glysson, G. (eds.) Proceedings of the
Federal Interagency Workshop on Turbidity and Other Sediment Surrogates,
Reno, NV, 30 April–2 May, U.S. Geological Survey Circular 1250, 3 pp.
Available at http://water.usgs.gov/pubs/circ/2003/circ1250/ (accessed on 13 June
2012).
Uhrich, M. A. and Bragg, H. M. (2003). Monitoring instream turbidity to estimate
continuous suspended-sediment loads and yields and clay-water volumes in the
upper north Santiam river basin, Oregon, 1998–2000. U.S. Geological Survey
Water-Resources Investigations Report 03-4098, 44 pp. Available at http://
pubs.usgs.gov/wri/WRI03-4098/ (accessed on 13 June 2012).
Urick, R. J. (1948). The absorption of sound in suspensions of irregular particles.
Journal of the Acoustical Society of America 20, 283–289.
Urick, R. J. (1975). Principles of Underwater Sound, 2nd edn., 384 pp. NY:
McGraw Hill.
U.S. Environmental Protection Agency (1999). Guidance manual for compliance with
the Interim Enhanced Surface Water Treatment Rule – Turbidity provisions.
Washington, DC: USEPA. Office of Water, EPA 815-R-99-010.
US Geological Survey (1976). Sampling mixtures of water and suspended sediment
in streams. Office of Water Quality Technical memorandum 76.17. Available at
http://water.usgs.gov/admin/memo/QW/qw76.17.html (accessed on 28 January
2013).
U.S. Geological Survey (1997). Comparison of the suspended-sediment splitting
capabilities of the churn and cone splitters. U.S. Geological Survey Office of
Water Quality Technical Memorandum 97.06. Available at http://water.usgs.gov/
admin/memo/QW/qw97.06.html (accessed on 13 June 2012).
U.S. Geological Survey (1998a). Guidance for collecting discharge-weighted
samples in surface water using an isokinetic sampler. U.S. Geological Survey
Office of Surface Water Technical Memorandum 99.01. Available at http://
water.usgs.gov/admin/memo/SW/sw99.01.pdf (accessed on 18 May 2012).
U.S. Geological Survey (1998b). A national quality assurance program for sediment
laboratories operated or used by the Water Resources Division. U.S. Geological
Survey, Office of Surface Water Technical Memorandum No. 99.01. Available at
http://water.usgs.gov/admin/memo/SW/sw99.01.html (accessed on 12 June
2012).
U.S. Geological Survey (1999). Guidelines from the1998 Sediment Laboratory
Chiefs’ Workshop. U.S. Geological Survey Office of Surface Water Technical
Memorandum 99.04. Available at http://water.usgs.gov/admin/memo/SW/
sw99.04.html (accessed on 18 May 2012).
U.S. Geological Survey (2000). Collection and use of total suspended solids
data. Office of Surface Water Technical Memorandum 2000.03. Available at
http://water.usgs.gov/admin/memo/SW/sw01.03.html (accessed on 12 June
2012).
U.S. Geological Survey (2002). Water-quality field methods phase-out of the US D-
77 and frame-type samplers. Office of Water Quality Technical Memorandum
2002.09. Available at http://water.usgs.gov/admin/memo/QW/qw02.09.html
(accessed on 18 May 2012).
U.S. Geological Survey (2005). Real-time water-quality concentrations and loads
estimated using regression analysis. Available at http://ks.water.usgs.gov/Kansas/
rtqw/sites/06892350/htmls/2005/p63680_2005_all_uv.shtml (accessed on 28
January 2013).
U.S. Geological Survey (2012a). USGS water data for the nation. National Water
Information System. Home Page. Available at http://waterdata.usgs.gov/nwis
(accessed on 28 January 2013).
U.S. Geological Survey (2012b). U.S. Geological Survey historical daily-value
suspended-sediment database. Available at http://co.water.usgs.gov/sediment/
(accessed on 12 June 2012).
U.S. Geological Survey (2012c). USGS parameter code definition. Available at
http://nwis.waterdata.usgs.gov/nwis/pmcodes/ (accessed on 18 July 2012).
U.S. Geological Survey (2012d). Statistics on hydrologic data-collection stations
operated by the U.S. Geological Survey. Available at http://1stop.usgs.gov/
StationCounts/sc2011/index.cfm (accessed on 18 May 2012).
Wall, G. R., Nystrom, E. A. and Litten, S. (2006). Use of an ADCP to
compute suspended- sediment discharge in the tidal Hudson river,
New York. U.S. Geological Survey Scientific Investigations Report 2006-5055,
16 pp. Available at http://pubs.usgs.gov/sir/2006/5055/ (accessed on 14 June
2012).
Winterstein, T. A. and Stefan, H. E. (1986). Effects of nozzle orientation on sediment
sampling. Proceedings of the Fourth Federal Interagency Sedimentation
Conference, Las Vegas, Nevada, vol 1, pp. 1-20–1-28. Available at http://
pubs.usgs.gov/misc/FISC_1947-2006/pdf/1st-7thFISCs-CD/4thFISC/4Fisc-V1/
4Fisc1-1.PDF (accessed on 28 January 2013).
Measuring Suspended Sediment 203
Author's personal copy
Wren, D. G., Barkdoll, B. D., Kuhnle, R. A. and Derrow, R. W. (2000). Field
techniques for suspended-sediment measurement: American Society of Civil
Engineers. Journal of Hydraulic Engineering 126(2), 97–104.
Wright, S. A. and Schoellhamer, D. H. (2005). Estimating sediment budgets at the
interface between rivers and estuaries with application to the Sacramento–San
Joaquin river delta. Water Resources Research 41, W09428, doi:10.1029/
2004WR003753.
Ziegler, A. C. (2003). Breakout session 1 – Definition of optical methods for
turbidity and data reporting. In: Gray, J. R. and Douglas Glysson, G. (eds.)
Proceedings of the Federal Interagency Workshop on Turbidity and Other
Sediment Surrogates, 30 April–2 May, Reno, NV, U.S. Geological Survey
Circular 1250, pp. 9–13. Available at http://water.usgs.gov/pubs/circ/2003/
circ1250/ (accessed on 13 June 2012).
204 Measuring Suspended Sediment
Author's personal copy
