Quality Assurance Plan for Discharge Measurements Using Broadband Acoustic
Doppler Current Profilers U.S. GEOLOGICAL SURVEY
OPEN-FILE REPORT 95-701
By Stephen W. Lipscomb
CONTENTS
Abstract
Introduction
Training
requirements
Pre-field
office procedures
Software and firmware revisions
Pre-field inspection
Field
procedures
Vessel installation
Instrument check
Configuration file
setup
Discharge measurements
Measurement assessment
Post-field
office procedures
Discharge measurement review
Documentation and archival of discharge
measurements
Annual
instrument check
Periodic
review
Selected
references
The recent introduction of the Acoustic
Doppler Current Profiler (ADCP) as an instrument for measuring velocities and
discharge in the riverine and estuarine environment promises to revolutionize
the way these data are collected by the U.S. Geological Survey. The ADCP and
associated software, however, compose a complex system and should be used only
by qualified per-sonnel. Standard procedures should be rigorously followed to
ensure that the quality of data collected is commensurate with the standards set
by the Water Resources Division for all its varied activities in hydrologic
investigations.
The Acoustic Doppler Current
Profiler (ADCP) is an electronic instrument developed and manufactured by RD
Instruments of San Diego, California, that is used to measure water velocities.
The instrument transmits acoustic signals into the water column. When the
frequency of the transmitted signals is compared with the frequency of
backscatter signals reflected off particles in the water, the velocity of the
particles and, hence, the water, can be calculated.
ADCP's have been widely used by oceanographers to measure currents in the
deep-sea environment. More recently, the technology has been adapted to the
riverine and estuarine environment and promises to revolutionize the way many
streamflow discharge measurements are made by the U.S. Geological Survey (USGS).
As ADCP technology has been refined, it has begun to gain acceptance as a
viable method for obtaining velocity and discharge data. The primary advantages
of making discharge measurements using the ADCP compared with a Price or other
point-current meter are that (1) the time required to complete a measurement is
reduced; (2) data can be collected throughout the water column and cross section
rather than at discrete points; (3) taglines or other stationing devices are
unnecessary because the instrument keeps track of distance traveled,
provided the bed is stable; and (4) the instrument can be boat-mounted, thus
eliminating the installation, maintenance, and liability of costly cableways.
The primary disadvantages of using the ADCP compared with a standard Price
current meter are its high initial cost; inability to function in shallow water;
complexity, requiring an in-depth understanding of the physics, electronics, and
software of the system prior to use; and the frequent revisions to hardware,
firmware, and software due to the newness of the technology. These revisions
generally result in improvements in the accuracy of the data collected; however,
they pose problems in dealing with data collected using earlier systems and
software.
The Office of Surface Water has recognized the utility of the ADCP for many
of the tasks performed by various Water Resources Division offices. However, as
with any technology or methodology adopted by the Division for the collection
and analysis of water resources data, it is critical that strict measures be
implemented to ensure the quality and reproducibility of the data throughout our
agency. The purpose of this report is to provide a quality assurance plan that,
when implemented, will ensure that data collected with ADCP's meet existing
accuracy standards and that the procedures used to collect the data are
consistent throughout the Division. Specific objectives of the plan are to
ensure that:
1. Field and office procedures associated with the use of ADCP's adhere to
applicable Division policies and standards governing the collection, analysis,
and reporting of surface-water data.
2. Personnel who collect and analyze the data are adequately trained in
proper and acceptable ADCP operational procedures.
3. All activities related to the collection, analysis, and reporting of data
collected using ADCP's are reviewed regularly for accuracy, completeness, and
conformance to Division standards.
4. Any deficiencies discovered during review are addressed immediately and
corrective action is taken.
Because of the complex nature
of the ADCP and associated software, at least one person in the field crew is
required to have formal training in the operation and maintenance of ADCP's.
Acceptable training is available through RD Instruments and the USGS National
Training Center in Lakewood, Colorado, or equivalent training can be provided by
personnel within the Water Resources Division. Eventually, qualified personnel
within Districts will be designated as trainers.
Reference materials on the theory, operation, and maintenance of the ADCP are
available from the manufacturer in the form of user's manuals for both the ADCP
and the associated software package "Transect" (RD Instruments, current
versions). These manuals are the primary source of information for the setup and
operation of the instrument and peripheral equipment. The manuals also contain
sections dealing with configuration file setup and the use of Transect for
system configuration, data acquisition, playback, output, and data manipulation.
A report on field procedures for ADCP's (M.R. Simpson, U.S. Geological Survey,
written commun., 1995) is the primary source of information on accepted
Divisionwide procedures for collecting velocity data and making discharge
measurements.
Field personnel responsible for data collection using the ADCP must be
completely familiar with these reference materials. Procedures for data
collection and ranges of acceptable conditions described in these documents must
be strictly adhered to.
To avoid delays in the
field and to ensure that the data collected are complete and of the highest
quality, some preparations prior to departure are recommended. These
preparations should include a determination as to whether the instrument is
utilizing the most recent software and firmware upgrades and that the complete
system, including the field computer and power supply, is operational.
Software and firmware revisions
Because of the newness of ADCP technology, there have been, and likely will
continue to be, frequent upgrades to both software and firmware associated with
the system. Many of these upgrades will result in only minor improvements to
internal processing speed or data output capabilities and will not substantively
affect the quality of discharge measurements made with the instrument.
Less frequently, major upgrades could affect the quality of measured velocity
or discharge data. Upgrades of this type will require field testing at one of
the validation sites described in the report by Morlock (in press) prior to
Division acceptance, to ensure that the instrument is performing up to required
standards. The Office of Surface Water will determine, prior to implementation
of all software and firmware upgrades, whether field testing is necessary.
Before the ADCP is taken to the field, the most recent Division-accepted
version of Transect should be installed on the primary and backup field
computers. In addition, a backup of this software should be copied to a diskette
and kept with the computer in the event that the copy on the hard drive is
inadvertently erased. In every office having an ADCP, a contact person will be
assigned to obtain and distribute the most recent copies of Transect from the
manufacturer. This person will be responsible for notifying users of software
and firmware revisions and for installing firmware upgrades as they become
available from the manufacturer. These upgrades can be installed either by
computer download from RD Instruments via modem or by manually replacing chips
within the instrument's circuit boards. If the latter approach is required, an
electronic technician, trained in the procedures for opening the ADCP pressure
case, should install the firmware upgrades in a static-free environment.
Pre-field inspection
A pre-field inspection must be made to ensure that the instrument is
functioning correctly. This inspection consists of connecting the ADCP to a
computer, preferably the one that will be used in the field, connecting a power
supply to the deckbox, and powering the system up. The system then can be
checked by entering the "Acquire" mode of the Transect software and attempting
to "wake the system up." If any problems are encountered at this point, the ADCP
and Transect user's manuals (RD Instruments, current versions) should be
consulted.
Inclusion of all required cables and connectors should be ascertained and any
spare parts that might be difficult to obtain in the field, such as fuses for
the deckbox, batteries for the ADCP and the computer, cables, and diskettes for
making backup copies of data files should be assembled as part of the pre-field
inspection.
General field procedures for making
discharge measurements with the ADCP are described in a report by M.R. Simpson
(U.S. Geological Survey, written commun., 1995) and therefore will not be
discussed here. However, certain precautions should be taken that will ensure
the accuracy, precision, and completeness of the data collected.
Vessel installationThe ADCP is typically mounted on either side of a
boat or in a well through the hull. The instrument should be rigidly attached in
a vertical position so that the transducers are submerged at least 3 inches (8
cm) below the water surface. In rough water, the transducers may have to be
lowered further to ensure that their heads stay submerged and that no cavitation
occurs in their vicinity during the entire measurement. The mount used to attach
the ADCP to the boat should be designed to withstand the combined forces of
water velocity resulting from the boat's movement and the ambient stream
velocity.
The mount should be designed so that the instrument can be quickly raised or
rotated out of the water for moving from one site to another or for quickly
traversing the cross section. The ADCP must not be mounted near steel or any
other ferrous material that would affect the functioning of the internal
flux-gate compass. For this reason, a boat with a steel hull should not be used
and the instrument should be mounted as far as possible from any ferrous objects
on the boat, such as an engine with a cast-iron block or heads. A rule of thumb
is to keep the compass, located near the top of the ADCP pressure case, away
from any ferrous object on the boat by at least the longest dimension of the
object. For instance, if the longest dimension of a steel davit mounted on the
boat is 4 feet, the ADCP should be mounted no less than 4 feet from the davit to
avoid interference with the compass.
Instrument checkAfter the ADCP is mounted and the required cables to
the computer and power supply are connected, the instrument must be checked to
ensure that all circuits and sensors are operating properly. Several subroutines
in BBTALK, which is a program provided as part of the ADCP software package,
perform internal diagnostic and calibration tests on the ADCP and display its
internal setup. These subroutines will test the circuits and sensors of the ADCP
and, upon request, will write the results to a user-designated file. Specific
procedures for initializing the self-test subroutines in BBTALK are detailed in
a predeployment test procedures document (RD Instruments, current version)
provided by the manufacturer with each ADCP or upon request and must be
performed prior to each deployment of the ADCP. If the instrument fails to pass
any of the self-test subroutines, the ADCP technical manual and Transect user's
manual (RD Instruments, current versions) should be consulted. If the problem
cannot be corrected, the manufacturer's field service representative should be
contacted and corrective action taken. If the ADCP has failed any of the
self-tests and data are collected, the data should be clearly marked as suspect
and the nature of the test failure documented in the remarks section of the ADCP
fieldnote sheet.
During the initiation of communication between the computer and the ADCP, a
comparison of the configuration file and the ADCP setup is performed. If the
ADCP fails to respond, or if any error messages are displayed, the ADCP and
Transect user's manuals (RD Instruments, current versions) should be consulted.
Many times, the problem can be solved by ensuring that all connections,
including the power supply to the ADCP, are properly made.
Sometimes an error message will warn that the configuration file is not in
agreement with the ADCP setup. Again, these problems usually can be solved by
rechecking the configuration file for proper communication settings, transducer
frequency, or head orientation (upward or downward looking). If the problem
cannot be discerned by consulting the user's manual, a manufacturer's field
service representative should be contacted and all problems related to failure
of self-tests or error messages should be resolved before proceeding.
Configuration file setupThe configuration file must be matched to the
physical conditions of the cross section. Such parameters as bin size, mode,
bottom track and water pings per ensemble, and blanking distance require setting
by a trained user to optimize the quality of data collected for the existing
conditions. Proper setup of the configuration file is beyond the scope of this
report but is addressed in a report describing discharge measurement procedures
using ADCP's (M.R. Simpson, U.S. Geological Survey, written commun., 1995). That
report provides specific details on selecting cross sections and setting up
configuration files for a variety of conditions and should be consulted prior to
collecting data.
Currently, the configuration file is stored on the computer separately from
the files that contain the raw velocity data. Therefore, each raw data file must
be linked in some way to the configuration file used during data collection so
that the parameters used during playback and postprocessing of the raw data
files are the same as those that were used during data acquisition. The raw data
files and configuration file used for a specific measurement must be saved on
the hard drive of the computer and copied to a diskette for a backup as soon as
possible following completion of the discharge measurement. Space on an ADCP
fieldnote sheet adopted by the Office of Surface Water has been designated for
listing raw data filenames and associated configuration filenames.
Discharge measurementsThe ADCP discharge measurement procedures guide
(M.R. Simpson, U.S. Geological Survey, written commun., 1995) describes
procedures for making discharge measurements, including site selection criteria,
configuration file setup, and postprocessing of raw data. That reference
provides details for field use of ADCP's and postprocessing of data as accepted
by the Office of Surface Water and should be adhered to rigorously. Any
variation from the procedures described in that publication must be documented
and reviewed before the data can be released.
Before discharge data are collected using the ADCP, preliminary information
describing the site, date, personnel, equipment, and versions of software and
firmware used should be entered on the ADCP fieldnote sheet. The remainder of
this sheet should be completed as data are collected. Information related to the
distance to riverbanks at the beginning and ending of individual transects and
raw data and configuration filenames must be entered immediately. This
information is critical to the accurate calculation of discharge, and any delays
in entering it on the fieldnote sheet will increase the likelihood of errors.
Individual transects can be viewed as instantaneous discharge at a particular
site. It is therefore necessary to average multiple transects to reduce
variation due to turbulence and velocity surges. Averaging is analogous to the
Division's policy of measuring velocity with a Price current meter over at least
40 seconds to minimize instantaneous fluctuations in stream velocities. In
general, at least four transects must be made at each site to ensure a valid
determination of discharge. More transects will be needed under certain
conditions, such as extremely turbulent water. If any one of the first four
transects differs from the mean discharge by more than 5 percent, it should
first be evaluated to determine if there is any reason to justify discarding it.
A determination to discard a single transect might be made on the basis of a bad
ensemble; a "Made Good" distance that differs significantly from the total
distance; anything that might have affected the velocity profiles during the
measurement, such as boat traffic in the vicinity; or any number of other
factors. If a transect is discarded, another should be made so that the
discharge measurement is calculated from the average of at least four transects.
If there is no justifiable reason for discarding the transect, four additional
transects should be made and all of them, including the outlier, should be
averaged to determine discharge.
During a transect, the ADCP measures the velocity in the water column
relative to the movement of the vessel to which it is attached. The vessel
velocity relative to the channel bed is also measured and is used to calculate
the actual water velocity. This calculation assumes a fixed bed. If, however,
the bed is in motion, the calculation of actual water velocity will be in error.
Therefore, assessment of bed movement is necessary before a discharge
measurement can be made. Bed movement can be assessed by anchoring the vessel to
the bed or to a fixed object such as a bridge, or by holding the boat at a fixed
location within the channel while a series of ensembles is collected. If the bed
is stable, the Shiptrack display will indicate no significant movement of the
vessel after a period of a few minutes. If, however, the bed is moving, it will
be reflected in the Shiptrack display by an apparent gradual movement of the
vessel in an upstream direction. If this movement is significant relative to the
average water-column velocity, an alternate site should be sought. Often,
bedload movement varies considerably from one point to another across a channel.
As a result, it may be necessary to check for bed movement at several locations
across the channel to ensure that the bed is stable throughout.
After the discharge measurement has been made, each raw data file should be
reviewed using the Playback mode in Transect to ensure that the data are
complete and do not include any bad ensembles, that depths and velocities do not
exceed the prescribed limits set for the instrument in use, and that no velocity
spikes are recorded because stream velocity equaled the ambiguity velocity. The
instrument limitations and ambiguity velocity for specific instruments and
configurations are discussed in a report by M.R. Simpson (U.S. Geological
Survey, written commun., 1992). Measurements of conditions that exceed
instrument limitations or that contain velocity spikes should not be included in
the determination of a final discharge value. If a raw data file contains bad
velocity ensembles or velocity spikes, the measurement should be repeated until
at least four complete measurements with no bad data have been obtained.
Once a sufficient number of transects have been made and reviewed for
completeness, the nearshore discharge estimate section on the back of the
fieldnote sheet must be completed to determine total discharge, including the
unmeasured sections near each bank. The Transect software contains a utility for
making the nearshore estimates on the basis of the first and last valid
velocities and depths measured. Procedures for using this function are provided
in the Transect user's manual (RD Instruments, current version). Care should be
taken to determine the direction of boat movement during collection of the
ensemble that is used for determining the nearshore discharge. Boat direction
determines whether a positive or negative sign should be assigned to the
estimated discharge before the total discharge can be calculated.
A gage height from a staff or some other reference should be obtained before
and after each measurement and entered into the space provided on the front of
the fieldnote sheet. Otherwise, the gage heights corresponding to the time of
each measurement should be obtained from the gage recorder and entered. A mean
gage height and discharge then can be calculated and entered into the space
provided on the upper part of the fieldnote sheet.
Measurement assessmentAn overall assessment of the mean discharge
measurement should be made after completion of the transects composing the
measurement. This assessment is based on a qualitative judgment of conditions
encountered in making the measurement and a quantitative evaluation of the
individual transects. Completeness of the measurement, in terms of the
percentage of the total cross-sectional area measured, and overall measurement
conditions must be assessed. Conditions such as turbulence, eddies, reverse
flows, surface chop, and proximity of the instrument to ferrous objects, which,
under certain circumstances, might affect the results of the measurements and
velocity profiles, should be noted under the appropriate sections on the
fieldnote sheet and used in assigning a rating for the measurement.
The mean and standard deviation of the discharge measurement and the
coefficient of variation (CV), which is the ratio of the standard deviation to
the mean discharge (s/Q), should be calculated. The CV then should be entered as
a percentage in the space provided on the fieldnote sheet. The CV is a measure
of the grouping of individual transects around their mean and is a useful
statistic for making a quantitative assessment of the measurement. If the CV is
greater than 5 percent, additional transects should be made. In some cases, one
or more transects will be apparent outliers and will result in a large CV. If
there is an obvious reason for the variation, the transect can be eliminated
from the mean discharge calculation after the circumstances are noted on the
fieldnote sheet. Additional transects may be needed if one or more is eliminated
so that at least four are used for computing the average discharge. If the mean
discharge varies from the applicable rating or shift curve by more than 5
percent, the measurements should be repeated.
The final discharge rating is assigned by circling the appropriate entry on
the front of the fieldnote sheet. This rating is based on the qualitative
evaluation of the measurement and the CV, which provides a quantitative measure
of precision.
After the fieldnote sheet is completed, all raw data and configuration files
resulting from the discharge measurements should be copied onto a disk-ette as a
backup in case of damage to the computer's internal hard drive.
The ADCP should be
inspected after it has been returned from the field to determine the condition
of the transducers, pressure case, connectors, cables, and deckbox. Damage or
undue wear to any of the components should be reported to the person responsible
for maintenance of the instrument.
Discharge measurement review
All discharge measurements must be reviewed by a second qualified person in
the office. Special attention should be given to the calculation of total
discharge, including nearshore estimates. All calculations used to derive the
mean discharge and gage height should be checked and the fieldnote sheet should
be reviewed for accuracy and completeness. The reviewer then must initial the
front of the fieldnote sheet.
Documentation and archival of discharge measurementsEventually,
Transect will include an option to generate a standard output file that will
contain the information in the configuration file and a summary of the discharge
measurement. This file will include information similar to that on the standard
9 - 275 discharge measurement form -- width, depth, area, mean velocity, and
discharge at incremental locations across the section. Until this option is
available, the raw data should be processed and an ASCII output file generated
that contains a summary of stationing and velocity data at approximately 30
equally spaced locations across the section.
Copies of the configuration, subsystem self-test, and the discharge
measurement summary files should be printed and attached to the fieldnote sheet
for filing as soon as practical after returning from the field. Information from
these documents should be used to enter the measurement on the 9 - 207 discharge
measurement summary form and into the computer database of discharge measurement
summaries (HT - 1). In both the paper and computer summaries of the measurement,
it should be noted in the remarks section that the measurement was made using an
ADCP.
Generation of a paper copy of the raw data files is not practical because of
their size and binary format. However, contained in these files is a significant
amount of information, such as horizontal and vertical velocity vectors
calculated for each ensemble, that should be retained. At this time, there is no
accepted method for permanent archival of digital data files within the Water
Resources Division. Until an acceptable archival method is approved, the raw
data and configuration files should be copied to as permanent a storage medium
as is available. Preferably, the files should be copied to some type of
semipermanent storage medium such as CD-ROM. Eventually, these files could be
transferred to a more permanent storage medium when the technology becomes
available.
Each ADCP must be checked
annually by making a discharge measurement at a site where the ADCP-measured
discharge can be compared with a known discharge derived from some other source.
An example of such a site would be one where a stable stage-discharge relation
with no significant shifting has been established over a period of several
years. The site ideally would be chosen to minimize the amount of unmeasured
sections near the banks or in shallows and should not be near any large steel
structures, such as bridges, that might affect the ADCP's compass. The discharge
obtained using the ADCP must be within 5 percent of the known discharge. If
these measurements fail to agree with the known discharge, the ADCP must be
returned to the manufacturer for further evaluation and calibration if
necessary. These check measurements must be fully documented and a summary log
of the results kept on file in the District or Field Office and noted in the
applicable station analysis.
Every 3 years, each District's
procedures for ADCP measurements, documentation, and data archival will be
reviewed by Office of Surface Water personnel or their designees during the
District Surface Water Review. This review will include an evaluation of site
selection, suitability of configuration files, measurement completeness and
accuracy, documentation of discharge and check measurements, and archival of
data files. In addition, the review will evaluate the maintenance procedures
followed by each office to ensure that the most current firmware and software
upgrades have been implemented and that the instrument and peripheral equipment
are being properly maintained. Recommendations by the Surface Water Review team
will be immediately addressed by the appropriate District personnel.
In summary, it is the primary objective of this plan to ensure that all data
collection and analyses performed in conjunction with the use of ADCP's be
conducted in a professional manner commensurate with the standards set by the
Water Resources Division for all its varied activities in hydrologic
investigations. Because of the complexity of the ADCP and associated software
and the dynamic nature of its application to the riverine and estuarine
environment, the ADCP should be used only by qualified personnel and standard
procedures should be followed
rigorously.
Morlock, S.E., in press, Evaluation of Acoustic Doppler Current Profiler
measurements of river discharge at 12 sites within the continental United
States: U.S. Geological Survey Water-Resources Investigations Report 95 - 4218.
RD Instruments, 1989, Acoustic Doppler Current Profilers -- principles of
operation: a practical primer: San Diego, Calif., RD Instruments,
36 p.
RD Instruments, current version, Broadband direct reading ADCP pre-deployment
test procedures: San Diego, Calif., RD Instruments, 10 p.
RD Instruments, current version, Direct-reading broadband Acoustic Doppler
Current Profiler technical manual: San Diego, Calif., RD Instruments, 49 p.
RD Instruments, current version, User's manual for RD Instruments Transect
program (for use with broadband ADCP): San Diego, Calif., RD Instruments, 88 p.
Simpson, M.R., and Oltmann, R.N., 1992, Discharge-measurement system using an
Acoustic Doppler Current Profiler with applications to large rivers and
estuaries: U.S. Geological Survey Water-Supply Paper 93 - 2395, 32 p.
|
