832R799O6
MONITORING REQUIREMENTS,
METHODS, AND COSTS
FOR THE
NATIONWIDE URBAN RUNOFF
October 1979
Water Planning Division
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
MONITORING
REQUIREMENTS, METHODS, AND COSTS
by
Philip E. Shelley, Ph.D.
EG&G Washington Analytical Services Center, Inc.
Rockville, Maryland 20850
Reprinted from the
AREAWIDE ASSESSMENT PROCEDURES MANUAL
EPA 600/9-76-014
for the
NATIONWIDE URBAN RUNOFF PROGRAM
Water Planning Division
U.S. Environmental Protection Agency
Washington, D.C. 20460
October 1979
-------
APPENDIX D
MONITORING REQUIREMENTS, METHODS, AND COSTS
D. 1 Introduction
The nation's waters are as mixed and varied as its population and, just as
there is no single measure of human health, there is no single measure of
water quality. Furthermore, the nation's waters themselves (ground waters,
streams, lakes, estuaries, and coastal waters) vary considerably in size,
geological features, flow characteristics, climate and meteorological in-
fluences, and the type and extent of human impacts on them, and all these
factors have a bearing on water quality.
Most definitions of water quality today are use-related, and each water use
is sensitive to different pollution types and levels. For example, suffi-
cient dissolved oxygen is critical to fish and other aquatic life but of
little significance to drinking water supplies or swimming. On the other
hand, coliform bacteria counts are a classical water pollution measure for
human contact or ingestion but have little significance for most industrial
uses or aquatic life. Even for the same parameter, the critical concentra-
tion for which one use begins to be impaired may be quite different from the
level at which another use is affected. Thus, water quality monitoring -
the collective activity that allows determination of the suitability of a
particular water source for a specific use - is heavily use dependent. It
is one thing to evaluate the lower Mississippi River as a drinking water sup-
ply and quite another to evaluate Lake Erie for swimming, a small stream in
Michigan for trout fishing, or the South Platte River for irrigation. A dif-
ferent monitoring effort would be required for each.
-------
D.1.1 How to Use This Appendix
In view of the foregoing, this Appendix cannot be a cookbook. Its overall
objective is to provide the 208 planner with a range of information, consid-
erations, and techniques that will allow him to design and implement a water
quality monitoring program that is suited to his particular requirements. As
indicated in Chapter 1 of this Manual, special emphasis is placed on equipment
and methods suitable for storm-generated discharges.
The organization of this Appendix is indicated in Table D-l. By referring to
it, the reader can locate information on the topic of immediate interest,
e.g., where to look for available water quality data, how to select test
catchments for stormwater model calibration and verification, how to choose
an automatic sampler, etc. The topical organization is intended to support
the chapters in the main body of this Manual by allowing quick reference to
specific information, but it is recoamended that this entire Appendix be read
and understood thoroughly before implementation.
D.I.2 Purposes and Objectives of 208 Monitoring
The broad objective of a monitoring activity is to provide information upon
which decision-makers can1 act. A more specific statement of objectives is
required, however, for design and implementation of a monitoring effort.
Examples of more specific monitoring objectives of interest to 208 agencies
include:
1. Establishing baseline conditions
2. Determination of assimilative capacities of streams
3. Following the effects of a particular project or activity
4. Pollutant source identification
5. Long-term trend assessment
6. Waste load, allocation
7. Projecting future water characteristics
D-2
-------
TABLE D-l. ORGANIZATION OF APPENDIX D
Page
D.I Introduction D-l
0.1.1 How to Use This Appendix ........... D-2
- D.I.2 Purposes and Objectives of 208 Monitoring . . , D-2
D.I.3 Types of Monitoring Activities D-8
D.I.3.1 Reconnaissance Survey D-9
D.I.3.2 Point Source Characterization D-9
D.I.3.3 Intensive Survey D-10
D.I.3.4 Fixed Station Monitoring Networks D-l2
D.I.3.5 Ground Water Monitoring . . D-12
D.I.3.6 Biological Monitoring ............. D-13
D.I.4 Coordination With Other Monitoring Programs ....... D-14
D.I.5 Available Data Sources D-16
D.I.5.1 Meteorological Data D-l7
D.I.5.2 Geographical Data D-18
D.I.5.3 Water Quality Data D-18
D.2 Measurement Site, Parameter, and Frequency Selection D-22
D.2.1 Site Selection D-22
D.2.1.1 Overall Site Location Guidance D-22
D.2.1.2 Site Selection for Waste Load
Allocation Surveys D-27
D.2.1.3 Catchment Selection for Stormwater Model
Calibration and Verification D-29
D.2.1.4 Specific Site Selection Criteria D-32
0.2.2 Parameter Selection D-34
0.2.2.1 Parameters for Storm-Generated Discharges . . . D-36
0.2.2.2 Parameters for the National Water
Quality Monitoring Program D-41
D.2.2.3 Parameters for Waste Load Allocations D-43
0.2.3 Measurement Frequency Selection. D-43
D.2.3.1 Frequency for Background and Trend Data .... D-43
D.2.3.2 Frequency for Waste Load Allocation
Surveys D-50
D.2.3.3 Frequency for Storm-Generated Discharges .... D-51
__-
-------
TABLE D-l. ORGANIZATION OF APPENDIX D (Cont'd)
Page
D.3 Flow Measurement Considerations, Equipment, and Procedures . . . D-52
D.3.1 General Considerations D-52
D.3.2 Flow Measurement Equipment D-57
D.3.2.1 Desirable Equipment Characteristics D-58
D.3.2.2 Evaluations of Some Promising Devices D-63
D.3.2.3 Review of Commercially Available Equipment
and Costs D-68
D.3.2.4 Review of Recent Field Experience D-73
D.3.3 Flow Measurement Field-Procedures D-76
D.4 Sampling Considerations, Equipment, and Procedures D-79
D.4.1 Sample Types D-80
D.4.2 Automatic Sampling Equipment D-84
D.4.2.1 Elements of an Automatic Sampler System .... D-84
D.4.2.2 Considerations in Automatic
Sampler Selection D-91
D.4.2.3 Survey of Commercially Available Equipment . . . D-92
D.4.2.4 Review of Recent Field Experience D-97
D.4.3 Manual Versus Automatic Sampling D-99
D.4.4 Sampling Field Procedures D-100
D.4.4.1 Manual Sampling Procedures D-100
D.4.4.2 Automatic Sampling Procedures D-l08
D.4.S Sample Quantity, Preservation, and Handling D-112
D.4.6 Sampling Accumulated Roadway Material D-115
D.5 Cost Estimation D-122
D.5.1 Instrumentation Costs D-122
D.5.2 Related Equipment Costs D-124
D.5.3 Manpower Costs D-l25
D.5.4 Field Operations Costs : D-127
D.5.5 Laboratory Analysis Costs D-129
D.5.6 Data Analysis and Reporting D-129
D.5.7 Example USEPA Costs D-130
D.6 Waste Load Allocation Study Procedures D-133
-------
These monitoring objectives include both point and noripoirit source considera-
tions involving variable and intermittent as well as continuous flows (see
Chapter 1 of this Manual). In particular, Section 208 of PL 92^500 has fo-
cused money and attention on stormwater runoff and the need for urban runoff
quality planning. The goals of such planning efforts are to define the run-
off problem, identify potential solutions and costs, and measure the effec-
tiveness of solution alternatives versus costs. Planning of this nature
requires a method of evaluation that can provide comprehensive and areawide
analysis, including the prediction of alternative futures. Mathematical
models represent a developing tool that can be used by planners to meet these
needs (see Appendix A). Such models require field data for their calibration
and verification, and monitoring for this objective, along with problem as-
sessment monitoring, will be emphasized in this Appendix.
A detailed listing of some USEPA uses for monitoring information is given in
Table D-2. From a review of this table, it is readily apparent that a proper
understanding of what is sought is paramount in the design and implementation
of any given monitoring activity. Furthermore, the objectives should be re-
duced to writing, not only to ensure careful consideration of what they ac-
tually should be and help prevent misunderstandings by those.involved, but
also to set the limits, and thus discourage the pursuit of interesting but
nonessential bypaths. These objectives willsalso.provide a basis for measur-
ing the extent to which the results of the effort .meet the needs that justi-
fied the undertaking.
To illustrate the form such objective statements might take, several examples
will be given. These were taken from actual 208 program efforts that are be-
ing designed and implemented now.
The stated objective for an instream sampling survey is to provide water qual-
ity and flow data for calibration and verification of a continuous water qual-
ity simulation model which will be used to simulate existing and future
D-5
-------
TABLE D-2
SOME USEPA USES OF MONITORING INFORMATION
Develop/revise water quality standards
Develop/revise 303 basin plans
Develop/revise 208 areawide plans.
Develop/revise 201 facilities plans
Document progress toward achievement/
maintenance of ambient standards and
legislative goals
Monitor primitive areas for background
levels and significant deterioration
Development of baseline information
Model validation/development
Develop health research/control techniques
Develop/evaluate Environmental Impact
Statements
Develop/revise effluent standards
Formulate/revise discharge permits
Determine permit compliance
Develop/revise drinking water standards
Develop/revise pesticides monitoring plan
Develop/revise toxic standards
Develop/revise pretreatment standards
Investigate single pollution incidents (fish
kills, oil spins)
Develop/assess/revise point source control
strategies
Develop/assess/revise nonpoint source
control strategies
Allocate resources
Report indices, trends, etc., to the public
Support enforcement actions
Develop/revise waste load allocations
-------
conditions in selected streams and rivers in northeastern Illinois. Since
organic pollutants and nutrients are considered the most general and wide-
spread water quality problems in the region, the sampling and analysis pro-
gram is designed to provide information necessary to simulate these
parameters.
The stated objective for a land use runoff study is to determine nonpoint
source pollution loading functions for homogeneous land uses. Transfer-
ability of data is required, since these loading functions will then be ap-
plied to other areas throughout the region.
The stated objective for a lake study is to determine, in terms of quantity
and quality, the pollutional load from nonpoint sources that enters Lake
Michigan during storm events. Note that this objective does not suggest that
a complete survey of Lake Michigan be undertaken (a task of great magnitude)
but, rather, seeks to determine what is going into the lake.
The three foregoing statements of objective were selected to illustrate that
being specific and concise can go together (and should). As a final example,
the following eight objectives are stated for an urban nonpoint source
monitoring network:
Collect basin rainfall and runoff.offl* for 14 Philadelphia area
drainage basins.
Calibrate the USGS Dawdy parametric rainfall runoff model using
3 to 5 years of data.
Using long-term Weather Service rainfall records as input to the
calibrated model, develop flood frequency duration curves for
14 urban drainage basins.
Measure physical basin characteristics of the 14 urban drainage
basins.
D-7
-------
• Relate physical basin characteristics to optimized model
parameters.
• Using developed regression relationships between model and basin
characteristics, develop flood-frequency duration curves for un-
gaged basins.
v Verify results with collected data on selected test basins.
• Collect average stream quality data for development of quality
trends as related to type of development.
Once the objective statement has been clearly formulated, the survey design
can begin, but not before.
D.I.3 Types of Monitoring Activities
There exist a number of types of monitoring activities that can be employed
in meeting overall monitoring requirements. Their suitability and applica-
bility will depend upon the purposes and objectives of the particular effort
involved. Included are (1) reconnaissance surveys, (2) point source charac-
terizations, (3J intensive surveys, (4) fixed station network monitoring
networks, (5) ground water monitoring, and (6) biological monitoring. The
last two types of monitoring activities are broken out separately only be-
cause they require skills, equipment, and techniques that are markedly dif-
ferent from those used in the first four. None of these should be considered
as completely separate activities in actual practice. Comprehensive data in-
terpretation will require that all monitoring data be considered together.
A brief description of these monitoring activities follows, with emphasis
placed on typical objectives of each. By comparing them, the reader can see
how they differ and how they may be combined to meet overall 208 monitoring
objectives.
D-8
-------
D.I.3.1 Reconnaissance Survey
A reconnaissance survey is a general or overall examination of a particular
area. It is a visual or superficial qualitative (and sometimes quantitative)
survey. Typical objectives of a reconnaissance survey include:
1. Getting the "lay of the land" in preparation for an intensive survey.
2. Identification of all waste sources in a particular catchment.
3. Identification of water uses in terms of types, locations, quan-
tities, and frequencies.
4. Determination of general stream characteristics.
5. Obtaining information necessary for establishing the overall design
of a fixed-station network.
6. Investigation of reported pollution incidents'or spills.
D.1.3.2 Point Source Characterization
A point source characterization (or effluent monitoring) study is one con-
ducted to determine the characteristics of an identifiable, discrete dis-
charge (either continuous or intermittent) into a receiving body of water.
Although several point sources are usually involved in a complete survey,
the mechanics of execution are basically similar.:,, .and the same general con-
siderations apply. It is also possible that. morie.-ithan one measurement site
(i.e., sampling and flow determination) might, b.e, involved as, for example,
in a treatment plant efficiency study. Mass loading discharges rather than
simple parameter concentrations are usually sought.: Some objectives are:
1. Determination of frequency, quantity, and strength of combined
sewer overflows.
2. Characterization of storm sewer discharges.
3. Determination of treatment plant efficiency.
4. Verification of a permit application.
5. Infiltration/inflow determination at a given site.
D-9
-------
6. Verification of self-monitoring data with regard to permit
compliance.
7. Determination of pretreatment requirements or verification of
compliance with pretreatment standards.
8. Verification of toxic substances sources.
9. Case preparation (as part of an enforcement action).
D.I.3.3 Intensive Survey
Intensive surveys are major elements in a monitoring program. The intensive
survey: (1) bridges the gap between the data bases generated by effluent
monitoring and fixed-station monitoring; (2) provides a definitive basis for
understanding and describing receiving water quality and the mechanisms and
processes that affect water quality; (3) provides the documentation required
to explain the trends observed at fixed network stations; and (4) is a tool ^-^
for determining the ultimate fate of polluiiints in the water environment. "*
Some generalizations concerning the overall nature of intensive surveys and
their planning and execution follow.
1. Repetitive measurements of water quality are made at each station
(sources and receiving water). The stations will typically comprise
a short, very dense, sampling network throughout the durat'ion of
the field effort.
2. The duration of an intensive survey is dictated by the objectives
of the survey, with 3 to 14 days being typical for freshwater
streams, lakes, and reservoirs. Surveys in tidal bodies are typ-
ically more complex and longer in duration as are nonpoint source
surveys (e.g., for calibrating a stonnwater management model).
3. The measurements taken during an intensive study vary. A study
may be oriented towards one particular type of data (chemical,
biological, sediment, etc.) or it may involve the collection of
many types of data.
D-10
-------
4. Continuous and intermittent point and nonpoint sources within the
survey area are usually monitored during the study.
Some major objectives of intensive surveys are:
1. Determining quantitative cause-and-effect relationships of water
quality for making load allocations, assessing the effectiveness
of pollution control programs, or for developing alternative
solutions to pollution problems.
2. Setting priorities for establishing or improving pollution
controls.
3. Supporting and setting priorities for enforcement actions.
4. Identifying and quantifying nonpoint sources of pollution and as-
sessing their impact on water quality.
5. Assessing the biological, chemical, physical, and trophic status
of publicly-owned lakes and reservoirs. /
6. Providing data for the classification or reclassification of
stream segments as being either effluent limited or water quality
limited.
7. Evaluating the locations and distribution of fixed monitoring
stations.
8. Calibrating and verifying stormwater management models.
Such objectives should be considered mutually compatible. The incremental
cost of expanding a single-purpose survey into a multipurpose survey should
always be evaluated prior to conducting the survey.
D-ll
-------
D.I.3.4 Fixed Station Monitoring Networks
The fixed monitoring network is a system of fixed stations that are sampled
in such a way that well-defined histories of the physical, chemical, and
biological conditions of the water and sediments can be established. In
general, other monitoring data will be needed to explain, in detail, the
trends observed at the fixed stations. Thus, a high level of coordination
between the fixed-station monitoring network and other monitoring activities
is essential for developing a useful data base. The basic objectives of
fixed monitoring networks are to provide data and information that, when
taken in combination with other data, will:
1. Characterize and define trends in the physical, chemical, and
biological condition of surface waters, including significant
publicly-owned lakes and impounded waters.
2. Establish baselines of water quality.
/
3. Provide for a continuing assessment of water pollution control
programs.
ft
4. Identify and quantify new or existing water quality problems
or problem areas.
5. Aid in the .identification of stream segments as either effluent
limited or water quality limited.
6. Act as a triggering mechanism for intensive surveys, enforcement
proceedings, or other actions.
»
0.1.3.5 Ground Water Monitoring
Because of the increasing threat to the quality of ground water posed by
some waste management practices and a general lack of comprehensive informa-
tion on the origins, scope, and nature of existing ground water pollution
D-12
-------
problems, it is important that programs be established and maintained to mon-
itor ground water quality. Some objectives of ground water monitoring are:
1. Obtaining data for the purpose of determining baseline conditions
in ground water quality and quantity.
2. Providing data for the early detection of ground water pollution
or contamination, particularly in areas of ground water use.
3. Identifying existing and potential ground water pollution sources
and maintaining surveillance of these sources, in terms of their
impact on ground water quality.
4. Providing a data base upon which management and policy decisions
can be made concerning the surface and subsurface disposal of
wastes and the management of ground water resources.
Ground water monitoring has been extensively treated in a recent series of
USEPA reports (1-5) and will not be discussed further in this Appendix. It
is only mentioned here to point up its importance to the 208 planning process.
0.1.3.6 Biological Monitoring
Aquatic organisms and communities act as natural pollution monitors. Some
organisms tend to accumulate or magnify toxic substances, pesticides, radio-
nuclides, and a variety of other pollutants. Organisms also can reflect the
synergistic and antagonistic interactions of point and nonpoint source
pollutants within the receiving water system. Some objectives of a biolog-
ical monitoring program are to gather biological data in such a manner as to:
1. Determine suitability of the aquatic environment for supporting abun-
dant, useful, and diverse communities of aquatic organisms.
D-13
-------
2. Provide information adequate to detect, evaluate, and characterize
changes in water quality through the study of biological produc-
tivity, diversity, and stability of aquatic systems.
3. Detect the presence and buildup of toxic and potentially hazardous
substances in aquatic biota.
4. Provide information adequate to periodically update the eutrophic
condition classification of freshwater lakes.
D.I.4 Coordination With Other Monitoring Programs
An attempt to put 208 monitoring somewhat in perspective is presented in Fig-
ure D-l, taken from the National Water Monitoring Panel (6). It is obvious
that if each functional purpose is to be productive, the proper information
must be provided by the monitoring program. It also should be clear that
persons responsible for monitoring must maintain a frequent and substantive
contact with those programs requiring information. Finally, there is abun-
dant need for coordination among all aspects of a monitoring program.
The importance of this last statement regarding coordination among monitoring
activities can be emphasized by considering the following. At the federal
level, legislative authority for monitoring is contained in at least six
Acts:
• The Federal Water Pollution Control Act
• The Safe Drinking Water Act
• The Refuse Act
• The Marine Protection, Research and Sanctuaries Act
• The Federal Insecticide, Fungicide, and Rodenticide Act
• The Solid Waste Disposal Act
When combined with State and local legislation, the legislative mandates form
an almost staggering dimension. The activities responsible for monitoring
implementation form an equally large dimension. At the federal level alone,
D-14
-------
o
>—
in
MANAGEMENT
PROGRAM EVALUATION;
PRIORITIES - POLICIES
REPORTING. MONITORING NEEDS
PROURAM PIIOHITIES
REPOTTING. MnNITORINT. NEEDS
PLANNING
1ASIN PLANS
AREAHIDE PLANS
PROCRAM PRIORITIES
PERMIT ISSUANCE
MUNICIPAL PERMITS
INDUSTRIAL PERMITS
LOAD ALLOCATIONS
/DATA FUR LOAD / /
ALLOCATIONS, FACILITY / /
SITING, ETC. / /
REPORTING. MONIITWIW; NEEOS
PROGRAM PRIURITIhS
COMPLIANCE
LIST OF VIOLATORS
ENFORCEMENT PRIORITIES
PERMIT CONDITIONS
ADDITIONAL DATA
FOd PERMIT
CONDITIONS
/ /
/ /
/ /
ENFORCEMENT
HINT, VIOIATORS INTO
COMPLIANCE
PERMIT VIOLATIONS
DATA FROM
COMPLIANCE
SURVE>S
7 /
IIATA FOR
IVimNrl
// DATA FOR PROGRAM f:VAL-I
/ UATION: FIXED STATIONS. /
/ TRENDS. NtN PROBLEMS, /
/ KATLB QIIAI IT» UIANIJIS /
MONITORING
FIGURE D-l
MONITORING IN PERSPECTIVE
-------
they include the U.S. Environmental Protection Agency; the U.S. Geological
Survey; the U.S. Department of Agriculture; the Bureau of Reclamation; the
Department of Defense, including the Army Corps of Engineers and the Naval
Facilities Engineering Command; the Bureau of Mines; the National Aeronautics
and Space Administration; the Occupational Safety and Health Administration;
the Food and Drug Administration; the Energy Research and Development
Administration; and others, not to mention special purpose monitoring efforts
conducted by federal activities such as the National Science Foundation, the
Council on Environmental Quality, the Office of Manpower and Budget, the
Office of Technology Assessment, and so on. These efforts must be combined
with those of the states, designated agencies, and all pollutant dischargers
operating under effluent permits. Typically, monitoring efforts are far from
centralized. For example, in the USEPA alone, monitoring responsibilities -
encompassing both the collection and use of information - are found in
16 Headquarters offices under 5 assistant administrators. Similarly, USEPA
field responsibilities are dispersed among the 10 regional offices and 13 re-
search laboratories.
D.I.5 Available Data Sources
The prudent use of resources dictates that the maximum use practicable be
made of existing data. For 208 planning, these can be grouped into three
categories: meteorological, geographical, and water quality. Available data
sources for each category will be discussed in turn. One caveat more or less
applicable to each must be mentioned, however. All prior data may not be of
acceptable quality (i.e., truthfulness, suitability, accuracy). Where at all
possible, attempt to determine the original- source and some indication of the
"goodness" of the data. For example, USGS stream gage records are annotated
with a somewhat subjective indication of the quality of the record, e.g.,
poor, fair, good, etc. Unfortunately, this is the exception rather than the
rule. Be especially chary of water quality records; attempt to determine
how the samples were taken, whether or not they were handled properly, and
how the analyses were run.
D-16
-------
D.I.S.I Meteorological Data
The best source of long-term rainfall data in the United States is the
National Weather Service (NWS). Data can be obtained from the NWS either
on tape files or through published daily and hourly summaries. Tapes can
be obtained by contacting:
U.S. Department of Commerce
National Climatic Center
NOAA Environmental Data Service
Federal Building
Ashville, N.C. 28801
Telephone (704) 258-2850
Data are available on two record files: Deck 448-USWB HOURLY PRECIPITATION
and Deck 345-WBAN SUMMARY OF DAY. Most first-order stations are covered.
The period of record is usually from August 1949 to the current data with
some gaps. Long-term 5-minute data are also available from the NWS for over
50 major U.S. cities, and can be generated for most cities having a NWS city
or airport office.
One word of caution; be sure to determine if there is a high aerial vari-
ability of rainfall for the region in question. For example, the total
rainfall measured at the NWS station at Philadelphia International Airport
was 44.47 inches for 1975. The totals for individual gaged catchments within
the city for the same period ranged from 40 to over 60 inches, with a city-
wide average of 51.37 inches.
Other meteorological data available from NOAA include snowfall, temperature,
wind, sunshine and sky cover, evaporation, and humidity. Local data sources
and the possible existence of data from previous studies should also be
investigated.
D-17
-------
D.I.5.2 Geographical Data
In this context, the term geographical is used in its broadest meaning.
Chief among this category are land use data, but other physical, cultural,
and demographic data will also be desired (e.g., catchment slopes and terrain,
soil types, sewer maps, population distributions, etc.)- Sources of such data
are described in detail in Appendix C of th.is Manual, but generally include:
• U.S. Census Bureau
• Metropolitan Sanitary Districts
• State and local planning agencies
• Office of the County Surveyor (or equivalent)
• U.S. Coast and Geodetic Survey
• U.S. Department of Housing and Urban Development
• USDA Soil Conservation Service
• Standard Metropolitan Statistical Area Data
• Previous basin (303e) or facilities (201) plans
D.I.5.3 Water Quality Data
The STORET system of the USEPA is the largest source of water quality data in
the nation. The system is operated as a utility serving states, areawide
agencies, and other organizations. Data are stored in the system by the data
collecting organization for their own purposes as well as for sharing with
others. The STORET system should be queried for existing data during the
initial design phase of the 208 areawide monitoring effort. USEPA headquar-
ters and regional offices may be contacted for assistance in the use of
STORET.
Other existing water quality data are widespread, but the recent establish-
ment of the National Water Data Exchange (NAWDEX) should considerably assist
users in locating and acquiring needed data. Unlike STORET, NAWDEX is not a
large depository of water data. Rather, its objective is to provide the user
with sufficient information to define what data are available, where these
D-18
-------
data may be obtained, in what form the data are available, and some of the
major characteristics of the data.
The U.S. Geological Survey has the lead-role responsibility for NAWDEX. In
this capacity, it has established the NAWDEX Program Office at its National
Center in Reston, Virginia. This office became active in November 1975 and
provides the central management for NAWDEX. It also has the responsibility
for coordinating all operational activities within the program. This
includes serving as liaison between NAWDEX members and users of the system.
The service capabilities of NAWDEX will be supported by a nationwide network
of Local Assistance Centers established in the offices of NAWDEX members to
provide local and convenient access to NAWDEX and its services. This network
will initially be established in late 1976 in the 46 district offices of the
U.S. Geological Survey. These offices are located in 45 states and Puerto
Rico. Most are equipped with computer terminals, thereby providing an
extensive telecommunication network for access to the computerized directory
and indexes being developed for the NAWDEX program. As the NAWDEX membership
increases, additional centers will be added in large population areas and
areas of high user interest to provide improved access to NAWDEX and its
services.
The NAWDEX Program Office is currently developing a Water Data Sources
Directory. This directory will identify organizations that collect water
data, locations within these organizations from which water data may be
obtained, the geographic areas in which water data are collected by these
organizations, the types of water data collected, alternate sources for ac-
quiring the organization's data, and the media in which the data are avail-
able. This directory is scheduled for release in 1977.
A computerized Master Water Data Index is also being prepared which is sched-
uled for nationwide use in November 1976. This index will identify individ-
ual sites for which water data are available, the locations of these sites,
the organizations collecting the data, the hydrologic disciplines represented
by the data, the periods of record, water data parameters, the frequency of
D-19
-------
measurement of the parameters, and the media in which the data are available.
More than 350,000 water data sites are currently being indexed from informa-
tion contributed by 19 federal agencies and more than 300 non-Federal
agencies.
Through its Water Data Sources Directory, Master Water Data Index, and indexes
and other reference sources made available by its participating members,
NAWDEX assists its users in locating data of special interest. These data
include water data in computerized and in both published and unpublished
forms. The user is then referred to the organization(s) having the needed
data. NAWDEX thus serves as a central point of contact for locating water
data that may be held by several different organizations. Data search assist-
ance may be obtained from the NAWDEX Program Office or from any of the Local
Assistance Centers.
.
To expedite locating exdu^i'^t'v- *v.ta, NAWDEX and STORET should be queried at
the same time. In addition "to referring the user to STORET, NAWDEX will pro-
vide information on other data sources for the area under consideration in
many instances.
Requests for services or additional information related to NAWDEX and STORET
may be directed to:
National Water Data Exchange STORET (WH-553)
U.S. Geological Survey U.S. Environmental Protection Agency
421 National Center 401 M Street, S.W.
Reston, VA 22092 Washington, D.C. 20460
Telephone (703) 860-6031 Telephone (202) 426-7792
Local points of contact for the USEPA STORET system and state Water quality
.agencies are given in Chapter 2 of this Manual (Tables 2-4 and 2-7). Se-
lected federal sources for water quality information are given in Table D-3.
A call to the Federal Information Center, (202) 755-8660, with its staff of
trained information specialists, will assist the user in finding the appro-
priate contact within any of these federal agencies.
D-20
-------
TABLE D-3
SELECTED FEDERAL SOURCES FOR WATER QUALITY INFORMATION
Department of Agriculture
Forest Service
Soil Conservation Service
Department of Commerce
National Oceanic and Atmospheric Administration
National Bureau of Standards
Department of Defense
Army Corps of Engineers
Army Civil Engineering Research Laboratory
Navy Facilities Engineering Command
Air Force Civil Engineering Research Center
Department of Health, Education, and Welfare
Public Health Service
Department of Interior
Bureau of Reclamation
Bureau of Land Management
Bureau of Indian Affairs
Bureau of Mines
Bureau of Sport Fisheries and Wildlife
Bureau of Outdoor Recreation
Geological Survey
Office of Saline Water
Fish and Wildlife Service
Office of Water Resources Research
Department of Transportation
Coast Guard
Energy Research and Development Administration
Environmental Protection Agency
National Aeronautics and Space Administration
Nuclear Regulatory Commission
Water Resources Council
Council on Environmental Quality
D-21
-------
D.2 Measurement Site, Parameter, and Frequency Selection
D.2.1 Site Selection
The location of measurement sites is critical to obtaining good quality data
and properly interpreting them. The following discussion covers overall site
location guidance, site selection for waste load allocation surveys, catch-
ment selection for stormwater model calibration and verification, and specific
local site selection criteria.
D.2.1.1 Overall Site Location Guidance
For overall background and problem assessment the following locations are
recommended for the chemical and physical sampling of the water column.
Biological and sediment stations should also be established at these loca-
tions, as appropriate.
1. At critical locations in water quality limited areas. Stations
should be located within areas that are known or suspected to
be in violation of water quality standards, ideally at the site
of the most pronounced water quality degradation. The data
from these stations should gage the effectiveness of pollution
control measures being required in these areas.
2. At the major outlets from and at the major or significant in-
puts to lakes, impoundments, estuaries, or coastal areas that
are known to exhibit eutrophic characteristics. These stations
should be located in such a way as to measure the inputs and
outputs of nutrients and other pertinent substances into and
from these water bodies. The information from these stations
will be useful in determining cause/effect relationships and
in indicating appropriate corrective measures.
3. At critical locations within eutrophic or potentially eutrophic
lakes, impoundments, estuaries, or coastal areas. These
D-22
-------
stations should be located in those areas displaying the most
pronounced eutrophication or considered to have the highest
potential for eutrophication. The information from these sta-
tions, when taken in combination with the pollution source data,
can be used to establish cause/effect relationships and to
identify problem areas.
4. At locations upstream and downstream of major population and/or
industrial centers which have significant waste discharges into
flowing surface waters. These stations should be located in
such a way that the impact on water quality and the amounts of
pollutants contributed can be measured. The information col-
lected from these stations should gage the relative effective-
ness of pollution control activities.
*ja
5. Upstream and downstream of representative land use areas and
morphologic zones within the area. These stations should be
located and sampled in such a manner as to compare the relative
effects of different land use areas (e.g., cropland, mining
area) and morphologic zones (e.g., piedmont, mountain) on water
quality. A particular concern for these stations is the
evaluation of nonpoint sources of pollution and the establishment
of baselines of water quality in sparsely populated areas.
6. At the mouths of major or significant tributaries to mainstern
streams, estuaries, or coastal areas. The data from these sta-
tions, taken in concert with permit monitoring data and intensive
survey data, will determine the major sources of pollutants to
the area's mainstem water bodies and coastal areas. By compari-
son with other tributary data, the relative magnitude of pollu-
tion sources can be evaluated and problem areas can be identified.
7. At representative sites in mainstem rivers, estuaries, coastal
areas, lakes, and impoundments. These stations will provide
data for the general characterization of the area's surface
D-23
-------
waters and will provide baselines of water quality against
which progress can be measured. The purpose of these stations
is not to measure the most pronounced areas of pollution, but
rather to determine the overall quality of the water. Bio-
logical monitorins will be a basic tool for assessing the over-
all water quality of an area.
8. In major water use areas, such as public water supply intakes,
commercial fishing areas, and recreational areas. These sta-
tions serve a dual purpose: the first is public health pro-
tection and the second is for the overall characterization of
water quality in the area. Determining the presence and
accumulation of toxic substances and pathogenic bacteria and
their sources are primary objectives of these stations.
Sediment sampling sites should be located Ji."SBix -areas as determined by in-
tensive surveys, reconnaissance surveys, and historical data. A major con-
cern of sediment monitoring will be to assess the accumulation of toxic
substances, and locations for sediment sampling should be chosen with this
in mind. Sediment mechanics and the hydrological characteristics of the
water body aust be considered. Refer also to Chapter 4 of this Manual.
In general, biological monitoring stations should *e established as follows:
1. At key locations in water bodies that are of critical value for
sensitive uses such as domestic water supply, recreation, and
propagation and maintenance of fish and wildlife.
2. In major impoundments near the mouths of major tributaries.
3. Near the mouths of major rivers where they enter an estuary.
4. At locations in major water bodies potentially subject to
inputs of contaminants from areas of concentrated urban, indus-
trial, or agricultural use.
D-24
-------
5. At key locations in water bodies largely unaffected -by man's .
activities.
For purposes of biological monitoring, a station will normally encompass
areas, rather than points, within a reach of river or area of lake, reservoir,
or estuary adequate to represent a variety of habitats typically present in
the body of water being monitored. Unless there is a specific need to evalu-
ate the effects of a physical structure, it is advisable to avoid areas that
have been altered by a bridge, weir, within a discharge plume, etc. Thus,
biological sampling stations may not always exactly coincide with water column
or sediment stations.
To the extent possible, all monitoring stations should be located in such a
manner as to aid cause/effect analyses. Some station requirements may be
such that, with careful station siting, one particular station could meet the
criteria of a number of types of stations. Caution should be exercised,
however, to avoid compromising the worth of a station for the sake of false
economy. In general, the quality of a monitoring program is not judged
solely by the number of stations. A few critically located stations may be
extremely valuable, while a large number of randomly selected stations may
yield meaningless data. Resource constraints will limit the total number of
stations. Figure 0-2, taken from (6), shows some examples of station
locations.
The stations shown on Figure D-2 are described as follows:
1. At a water supply intake; upstream station of a pair bracketing
a municipal and industrial center.
2. At a critical location in a water quality limited segment; down-
stream station of a pair bracketing a municipal and industrial
center; mouth of a significant input to a reservoir known to
exhibit eutrophic characteristics.
D-25
-------
. (h. Bl WATER
SUPPLY INTAKE
MUNICIPAL-
INDUSTRIAL
COMPLEX
IRRIGATED
CROPLAND
STRIP MINING AREA
WILDERNESS AREA.
MOUNTAINOUS 6 FORESTED
X STATION NUMBER
(X, X) STATION TYPE
W WATER COLUMN
B BIOLOGICAL
S SEDIMENT
FIGURE D-2
STATION LOCATIONS
0-26
-------
3. At a critical location in a reservoir known to exhibit eutrophic
characteristics; in an area of recreation.
4. Upstream of a major land use area (strip mining); major outlet
from a eutrophic reservoir.
5. Downstream of a land use area (strip mining); mouth of a signif-
icant tributary to mainstem river.
6. Upstream of a major land use area (irrigated cropland).
7. Downstream of a land use area (irrigated cropland); mouth of a
significant tributary; representative site for other streams
passing through same land use.
8. Upstream of a major land type area (wilderness).
9. Downstream of a major land type area (wilderness); mouth of sig-
nificant tributary to mainstem river.
10. Representative site in mainstem river.
11. Representative site in mainstem river, mouth of major input to a
potentially eutrophic estuary.
12. Representative site in estuary, recreational area, shellfish
harvesting area.
D.2.1.2 Site Selection for Waste Load Allocation Surveys
Intensive surveys for waste load allocation will form an important part of a
208 agency's monitoring program. Since water quality problems don't manifest
themselves on demand and we can't afford to wait around for the 10-year dry
spell, the use of mathematical models for problem assessment will be required.
These models will require monitoring data from intensive surveys for
D-27
-------
calibration and verification. By and large, at least two intensive surveys
will be required for each waste load allocation study. The first, or pre-
liminary, survey should be performed during slightly higher flow conditions
than the second, or primary, survey which should be conducted when flow condi-
tions are as low as possible.
For the case where only one outfall impacts upon the water quality of a
stream, measurement sites should be located as follows:
A - directly upstream of the outfall.
B - effluent from the outfall.
C - mix point (i.e., where effluent is thoroughly mixed with
stream flow).
D^^ - intermediate points between the mix point and the DO sag point
or a tributary, if one enters the stream ahead of the sag point.
Spacing of 0.1 mile or less is usually warranted.
E - directly upstream of any tributaries.
F - tributary.
G. - intermediate points between tributaries (if more than one) or
between tributary and sag point.
H - sag point.
I. - points downstream from the sag point. Measure at least every
0.1 mile until there is a definite recovery in the DO profile.
Where more than one outfall discharges into the stream, these sources must
also be measured, and the above site locations altered accordingly.
D-28
-------
The problem of determining the mix point deserves special mention. The com-
mon practice of locating the mix point either by visual inspection of the
stream or by simply assuming that the stream is well mixed a certain distance
downstream is simply inadequate. A more rigorous method must be used. One
technique that has successfully been employed is to follow the concentration
of chlorides downstream. The steps in this procedure are:
1. Measure chloride concentration upstream of the outfall.
2. Measure chloride concentration in the effluent.
3. Perform a mass balance calculation to determine the mixed chloride
concentration.
4. Measure chloride concentrations at increasing distances downstream
from the outfall.
5. Locate the mix point where the measured chloride concentration is
equal to the calculated value.
D.2.1.3 Catchment Selection for Stonnwater Model Calibration and
Verification
Field data will be required for calibration and verification of stormwater
models, with details dependent upon the actual model selected (see Ap-
pendix A). However, it must be emphasized at the outset that instrumentation
of a large, multiuse drainage basin can only generate data for verification
of urban planning models. Calibration of these models requires data from
small catchments of uniform land use to provide information for adjusting
model parameters for each individual land use. Since it will not be prac-
ticable to instrument all catchments within a planning area, the effective-
ness of the planning models will depend to a large degree on the ability to
D-29
-------
estimate parameters for catchments that have no calibration data. This im-
plies the selection of catchments that have a high potential for data trans-
ferability as "benchmark" stations and instrumenting them accordingly. Each
instrumented catchment must therefore be viewed as a "sample" of the planning
area's catchments. Selection of representative and (to the extent possible)
uniform "samples" is necessary in order to arrive at a set of transferable
model parameters that cover the variations among catchments for the entire
planning area.
Catchment selection begins with an inventory of catchments in the planning
area. Minimum characterization includes size; present and projected land
use; drainage type (non-sewered, degrees of partial sewer service, fully
sewered, and sewer types); physical catchment characteristics; and relation-
ship to major streams, lakes, or estuaries within the area of interest.
Catchments in urban areas are small and numerous, emphasizing the need for
selecting a small, representative subset. The size will affect the relative
importance of runoff flow and water quality constituent routing. Very small
Ce.g., less than 0.1 square mile) catchments should be avoided as their
(typically) extremely rapid response times may make runoff characterization
impossible. It is unlikely that the requirement for uniformity of land use
will allow utilization of large (e.g., over 5 square miles) catchments in
urban areas.
The extent of sewering will have a significant impact on catchment runoff,
affecting routing, length of overland flow, and the relative importance of
infiltration. In urban areas, the ratio of sewer length to drainage area
typically falls between 8 and 18. The corresponding ratio for natural river
and stream channels would be less than 2. The physical catchment charac-
teristics such as percent imperviousness, ground slope, soil characteristics,
and infiltration potential will obviously affect runoff and must be con-
sidered in selecting representative catchments for instrumentation.
The recommended procedure is to prepare a matrix inventory characterizing
each catchment within the area of interest. These should then be categorized
D-30
-------
using land use as the factor, since parameters in currently available models
are largely functions of land use. One should not feel constrained to use
only the conventional single-family residential, multi-family residential,
commercial, industrial, and open-space land use types. Use types reflecting
the local conditions are more meaningful. For example, it may be desirable
to distinguish between single-family residential areas near the center of a
city and those in the suburbs; review of the catchment inventory may indicate
a number of small suburban shopping centers and the desirability of a mixed
residential/commercial category. Other locally important factors for
determining land use types might be traffic volume, population density, age
of development, family income, percent of streets with curb and gutter,
type of industry, and so on. Although local conditions will determine the
exact number of land use categories to be employed, fewer than five will
probably not allow satisfactory data transfer and more than ten will increase
field data collection costs beyond reason. See Appendix C of this Manual for
further guidance.
The problem of site selection from among those catchments in each land use
category now remains. Budgetary constraints will mandate selection of only
one catchment for instrumentation in each land use category for the most
part and, therefore, the "best" must be selected. Although random selection
may be expedient, a more rigorous and comprehensive approach is usually
desirable. On the other hand, a sophisticated multiple regression analysis
with serial and/or factor differentiation of catchment variables is probably
not warranted. The technique of weighted suitability ratings often employed
in land use planning will be adequate in most instances. It has the advan-
tage that the selection criteria can easily be illustrated on a single chart
for relative catchment comparison. Although the procedure is necessarily
subjective in the selection of factors, suitability values, and weights, so
was the basic selection of land use types.
D-31
-------
D.2.1.4 Specific Site Select '•. n~ CrrY.Ti
Given an identified catchment, stream reach, or other general location where
measurements are desired, there are some general criteria that can aid in
selecting the specific measurement site. They include:
1. Maximum accessibility and safety. Manholes on busy streets
should be avoided if possible; shallow depths with manhole steps
in good condition are desirable. Sites with a history of sur-
charging or submergence by surface water, or both, should be
avoided if possible.
2. Be sure that the site provides the information desired. Famili-
arity with the sewer system is necessary. Knowledge of the ex-
istence of inflow or outflow between the measurement point and
point of data use is essential.
3. Make certain the site is far enough downstream from tributary
inflow to ensure mixing of the tributary with the main stream.
4. Locate in a straight length of channel, at least six widths
below bends.
5. Locate at a point of maximum turbulence, as found in sections
of greater roughness and of probable higher velocities.
Locate just downstream from a drop or hydraulic jump, if
possible.
6. In all cases, consider the cost of installation, balancing cost
against effectiveness in providing the data needed.
The success or failure of selected equipment or methods, with respect to
accuracy and completeness of data collected as well as reasonableness of
cost, depends very much on the care and effort exercised in selecting the
D-32
-------
site. A requirement with regard to flow measurement that appears to be
obvious, but which is frequently not sufficiently considered, is that the
site selected be located to give the desired flow measurement. Does flow
at the site provide information actually needed to fulfill given needs?
Sometimes influent flows, diversions, or storage upstream or downstream from
the selected site would bias the data in a manner not understood without a
thorough study of the proposed site. Such study would include reference to
surface maps and to sewer maps and plans. Sometimes groundwater infiltration
or unrecorded connections may exist. For these reasons, a thorough field
investigation should be made before establishing a flow measurement site.
A basic consideration in site selection is the possible availability of
measurements or records collected by others. At times, data being collected
by the USGS, by the state, or by other public agencies can be used. There
are locations where useful data, although not currently being collected, may
have been collected in prior years. Additional data to supplement those
earlier records may be more useful than new data collected at a different
site.
Requirements that apply to all measurement sites are accessibility, per-
sonnel and equipment safety, and freedom from vandalism. If a car or other
vehicle can be driven directly to the site at all times, the cost in time re-
quired for installation, operation, and maintenance of the equipment will be
less, and it is possible that less expensive equipment can be selected.
Consideration should be given to access during periods of adverse weather
conditions and during periods of flood stage. Sites on bridges or at man-
holes where heavy traffic occurs should be avoided unless suitable protec-
tion for men and equipment is provided. If entry to sewers is required, the
more shallow locations should be selected where possible. Manhole steps and
other facilities for sewer access must be carefully inspected, and any
needed repairs made. Possible danger from harmful gases, chemicals, or
explosion should be investigated. With respect to sites at or near streams,
historical flood marks should be determined and used for placement of access
-------
facilities and measurement equipment above flood level where this is possible.
Areas of known frequent vandalism should be avoided.
In this last regard, the problem of vandalism can be serious and costly, both
in terms of equipment damage and data loss. The selection of sites in open,
rather than secluded, areas may help reduce vandalism as may illumination at
night. Attempts to hide or camouflage equipment have been generally unsuc-
cessful. Instrumentation should be sheltered to the extent possible, trading
off the cost of protective facilities, the latitude afforded by the site, and
the need for easy access. Occasionally, solid masonry or steel shelters sur-
rounded by heavy fencing may be required for measurement sites, and these
additional costs must be included in such instances. Finally, warning signs
are generally unsuccessful; they may only encourage vandalism regardless of
the type of threat -- high voltage, radiation hazard, fine, or imprisonment.
D.2.2 Parameter Selection
A review of the Parameter Handbook points out that the list of possible water
quality parameters that might be of interest to the 208 planner is almost end-
less. Parameter selection must be based on the specific objectives of the
study and a knowledge of general pollution source characteristics. For ex-
ample, nonmunicipal effluent limitations guidelines for existing point
sources, standards of performance for new sources, and pretreatment standards
for new and existing sources discharging to publicly-owned waste treatment
facilities have been published for 28 point source categories (40 CFR 405-432).
Effluent limitations establish the mass of specific pollutants that may be
discharged per unit of production or raw material input. Limitations are es-
tablished for a maximum production day and for the 30-day average. Table D-4
summarizes the effluent parameters included in each of the published effluent
guidelines.
For publicly-owned treatment works in existence on July 1, 1977, or approved
for a Federal construction grant prior to June 30, 1974, effluent limitations
D-34
-------
TABLE D-4
EFFLUENT PARAMETERS BY INDUSTRIAL CATEGORIES
INDUSTRY CATEGORY
1. PULP. PATER AND PAPERBOARDS
2. HJILDERS PAPER AND BOARD
1. TIMIER PRODUCTS
4. SOAP AND DETERGENTS
S. DAIRY PRODUCTS
6. ORGANIC CHEMICALS
'. PETROLEUM REFINING
«. LEATHER TANNING AND FISHING
. CANNED AND PRESERVED
FRUITS AND VEGETAILE5
10. NONFEMOUS METALS
11. GRAIN MILLS
12. SUGAR PROCESSING
1}. FERTILIZERS
14. ASRESTOS
IS. MEAT PRODUCTS
16. FERROALLOYS
n. GLASS
11. ELECTROPLATING
19. PHOSPHATE MANUFACTURING
20. FEEDLOTS
21. CEMENT MANUFACTURING
22. RUIIER PROCESSING
24. INORGANIC CHEMICALS
25. IRON AMD STEEL
26. TEXTILES
,, STEAM ELECTRIC GENERATING
"' EQUIPMENT
2». SEAFOOD PROCESSING
TOTALS:
X
X
X
X
X
X
I
I
X
X
X
X
X
X
X
II
1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
«
«
X
X
X
I
27
X
X
X
X
X
X
X
X
I
I
I
X
X
I
X
X
X
X
X
X
I
X
X
X
27
"
X
2
u
X
I
X
I
X
X
I
n
S
X
X
X
X
1
u
2
•
c
c
X
X
X
X
X
X
X
I
X
X
12
rflRFACTANTS |
X
1
?
I
2
X
X
X
X
s
S
5
=
X
X
2
I
u
X
X
X
X
X
X
7
•0
"
X
X
X
I
4
¥
X
X
3
z
z
X
1
ECAL COLIFORM |
X
X
X
X
X
I
6
X
X
2
| N DINVjai
X
I
r. nnspioRiis |
i
X
X
5
LUORIDE 1
X
I
X
X
X
s
5
X
X
X
X
4
OPPER |
X
X
I
5
i
I
X
X
1
: VAN 11)1 |
X
X
X
3
4ANT.ANESF |
X
1
,
X
X
:
(RSEMIC |
X
1
1
X
1
X
X
:
|
i
i
5
X
X
2
t. DISSOLVED SOLIDS |
1
I
2
D-35
-------
are based upon an effluent standard of secondary treatment. Secondary treat-
ment is defined in 40 CFR 133.102 and consists of:
Parameter
BOD5
Suspended Solids
Fecal Coliform Bacteria
(geometric mean)
Removal Efficiency
PH
7-day Average
45 mg/SL
45 mg/i
400/100 mi
30-day Average
30 mg/i
30 mg/i
200/100 mi
85 percent
6.0 - 9.0
The recommended procedure is to examine the sources and processes involved in
the study area and, on the basis of need-to-know and reasonable expectation,
select measurement parameters accordingly. Flow should always be included.
Parameters should not be limited to those that are known to be a problem, but
should also include those that can reasonably be expected to become a problem.
The 208 monitoring program should identify new problems as well as track
existing ones. The results of early analyses should be used to assess param-
eter coverage and assist in determining whether an increase or decrease is
warranted. Resist the temptation to "look at the whole world." Analyses
cost money, and wise resource management dictates that only parameters, the
knowledge of which directly supports specific study objectives, should be in-
cluded. Put in writing a justification for each parameter selected. Use the
Parameter Handbook for guidance.
0.2.2.1 Parameters for Storm-Generated Discharges
Parameter selection will be facilitated by initially considering water qual-
ity characteristics in gross categories rather than as specific compounds or
D-36
-------
elements. As an example, the following treats the quality characteristics
considered important for storm-generated discharges. See Wullschleger et al.
(7) for elaboration.
D.2.2.1.1 Oxygen Demand
One of the most important quality characteristics in a receiving body of wa-
ter is the dissolved oxygen concentration.' The dissolved oxygen concentra-
tion has a direct bearing on the quality and natural balance of much of the
aquatic biota. Dissolved oxygen concentration can also have an effect on
the recreational and aesthetic uses of a body of water. Storm-generated dis-
charges that contain organic and inorganic compounds that exert a demand for
the oxygen dissolved in water can be considered pollutional discharges in the
sane sense as dry-weather municipal wastewaters.
Oxygen demand is exerted by (1) organic compounds that undergo biochemical
oxidation as a result of microbial activity and (2) by the immediate demand
exerted by the chemical oxidation of inorganic reduced compounds. However,
storm-generated discharges have certain characteristics different from
municipal sewage that affect not only the DO level in the receiving waters,
but also the conventional tests used to measure oxygen demand. Since com-
bined sewer overflows have a variety of sources other than just municipal
sewage, the discharges may contain materials that cause special problems.
During dry weather, when flow through a combined sewer system is low, solids
settle out. At the start of a storm, the first flush of water through the
system may have a high concentration of solids that affects the demand char-
acteristics of the waste. It has been found that the fraction of BOD in the
particulate form can range from 69 to 87 percent, which is considerably
higher than the 30 to SO percent present in most municipal wastewaters.
Also, combined sewer overflows from industrial areas and urban runoff may
contain oils, toxic materials and chemicals which are foreign to the natural
environment and interfere with traditional oxygen demand tests. Finally,
D-37
-------
storm-generated discharges contain a large amount of natural materials such
as silt, vegetation, wood, and other materials such as plastic that may not
exert an immediate demand but will eventually use the oxygen required for
decomposition. These characteristics cause these discharges to be different
from that waste normally encountered in sanitary analyses.
There are numerous tests available for use as potential oxygen demand in-
dicators, including BODg, BOD,0, BOD.., ACOD, COD, TOC, and TOD. The desired
test should have a well established, standardized test procedure and provide
a measurement of the total oxygen demand on the environment. No single
analytical test can meet both of these criteria. Therefore, two parameters
are recommended to indicate oxygen demand for storm-generated discharges, TOD
and BODf. TOD reflects the long-term demand, allowing correct determination
of discharge effects, and lacks the serious interference problems of other
tests, notably COD. BOD5 is recommended, despite its numerous disadvantages,
because of its widespread and historical use. Also, because of toxicity
effects on the BOD. test, comparison of BOD. and TOD results can yield in-
o o
formation about the degree of toxicity and its possible effect on the natural
environment.
D.2.2.1.2 Particulate Concentration
The solid matter present in storm-generated discharges can be divided into
two major categories; namely, particulate solids' and dissolved solids.
Particulate solids are important in combined sewer overflows and storm run-
off applications because they usually represent a large fraction of the total
solids. Also, these solids are generally removed from the flow by physical
treatment processes such as sedimentation, screening, flotation, and filtra-
tion—the type of processes most commonly used for storm-generated dis-
charges. It is, in fact, the relatively high concentration of particulate
solids in these flows which makes such processes attractive.
The recommended parameter for indicating particulate concentration in storm-
generated discharges is nonfilterable residue (suspended solids). The
D-58
-------
analysis is routine and not as time consuming and cumbersome as some of the
other particulate tests and, with a few additional steps, both the volatile
and fixed portions can be determined, yielding another useful piece of
information in most instances. Where settleable residue is desired, the
gravimetric method is recommended, not the Imhoff cone. Turbidity measure-
ments provide little comparable data about particulate matter or concentra-
tion and are not recommended for this purpose.
D.2.2.1.3 Pathogenic Microorganism Potential
Any discharge that includes waters which have come into contact with excre-
ment from warm-blooded animals of any type should be considered as having
the potential for conveying pathogenic bacteria, viruses, protozoa, and other
contagions. It is extremely difficult, if not logistically impossible, to
monitor these discharges for the many pathogens themselves. This problem was
recognized in the water supply field many years ago and has led to the almost
universal usage of the coliform group of bacteria as the indicator or measure
of the sanitary quality of water. The coliforms themselves are not neces-
sarily pathogenic, but their presence should infer the possible presence of
pathogens. However, for a number of reasons the coliform group is not
necessarily the most sensitive indicator as far as storm-generated discharges
are concerned.
The recommended indicator parameters are fecal coliform and fecal strep-
tococcus. Furthermore, it is recommended that the membrane filter (MF)
technique be used rather than the multiple tube fermentation procedure where
results are expressed as the most probable number (MPN) statistic.
D.2.2.1.4 Eutrophic Potential
In addition to sunlight and carbon dioxide, aquatic plants require nutrients
and trace salts. The principal nutrients are compounds which contain the
elements phosphorus, nitrogen, and potassium. The proliferation of aquatic
plants in most water bodies is undesirable. The term "eutrophic" refers to
D-39
-------
a condition in a water body where copious plant growth has resulted in an
undesirable or unsightly'situation of scce legated ..i..?!-.-'? *^:terioration. Al-
though eutrophication is a natural process, it can be accelerated by man's
activities.
Nitrogen and phosphorus are measures of the eutrophic potential of storm-
generated discharges. It is recommended that two nitrogen analyses be con-
ducted, nitrate plus nitrite (run by reducing nitrate to nitrite and
measuring the latter) and Kjeldahl. Of the 14 different phosphorus frac-
tions, total phosphorus is the recommended parameter.
D.2.2.1.5 Toxic and Related Substances
A large number of compounds of varying toxicity and concentration are likely
to be found in combined sewer overflows and storm runoff. However, the
toxicants of major concern can be divided into the general categories of
heavy metals, pesticides, and herbicides.
When studying the quality of storm flows, it is recommended that a composite
sample of the flow be analyzed for lead, zinc, copper, chromium, mercury,
cadmium, arsenic, nickel, and tin four times a year (seasonally). Based
upon the results of these tests, a decision can be made as to how often cer-
tain heavy metals will have to be analyzed thereafter. It is expected that
lead, zinc, copper, and chromium may be measured routinely. In certain com-
bined sewer areas serving known industries, or in certain storm sewer dis-
charges from areas of heavy vehicular traffic, it may be necessary to do
more frequent analysis.
Because of the wide variability of pesticides in use, the periodic nature
of their application depending upon season and nature of the drainage area,
and the complexity of the laboratory analyses, no pesticides or associated
compounds axe recommended for routine analysis. However, it is recommended
that, when evaluating the quality of a storm-generated discharge, a study of
the drainage area should be made to determine the likelihood of pesticide
D-40
-------
application (and the type) and if it is probable that the storm flow may
contain pesticides. At least one discharge should be analyzed to see if that
pesticide is present. Depending upon this result, a decision can be made as
to whether more analyses are needed.
D.2.2.1.6 Other Parameters
There are a host of other parameters that can be used to characterize storm-
generated discharges. In the absence of site specific concerns, however,
only pH is recommended for routine measurement.
D.2.2.2 Parameters for a National Water Quality Monitoring
Program
As a further aid in parameter selection, the proposed minimum parameter list
for a national water quality monitoring program will be discussed.
Temperature, pH, and dissolved oxygen are included because they are the
primary constituents in most chemical reactions that occur within the water-
body. They are also the essential factors that govern whether the ecosystem
will maintain aquatic life. A conductivity measurement is included to
determine the degree to which dissolved solids contribute to the water qual-
ity. This is a most reliable measurement and can be done on site. Salinity
is measured in estuaries and bays.
Fecal coliform is included because it is, at present, the most reliable test
for indicating the possible presence of pathogenic microorganisms in the
system. Trace metals were limited to those that are of high priority and
are toxic. Since the concern of the program is to measure the total load,
total metals instead of dissolved forms are measured.
In order to determine the extent of total nutrient contribution, total
phosphorus, total Kjeldahl nitrogen, and nitrite and nitrate are measured.
Since the basic concern of the program is the total nutrient load, total
phosphorus is measured instead of the other various forms of phosphorus.
D-41
-------
This is also more economically sound. In determining the contribution of
nitrogen to the system, the concern of the program is also to arrive at some
understanding of the stage of nitrification within the system. Therefore,
total Kjeldahl nitrogen is included as a measurement of organic nitrogen and
ammonia, and nitrate and nitrite are included to determine the extent of
oxidized nitrogen.
A total suspended solids measurement is included to measure the contribution
of solid material to the system and to give some indication of water clarity
and the probability of chemical adsorption.
A chemical oxygen demand (COD) measurement is included to get an indication
of the oxygen demand placed on the system. Chemical oxygen demand was chosen
over biochemical oxygen demand (BOD) and total organic carbon (TOC) because
it is more reliable than BOD, does not involve problems with holding time
and sample transport as do BOD samples, and does not require the sophisti-
cated equipment required of a TOC measurement. COD is not measured in lakes
and impoundments because it is usually found only in such low concentrations
that it renders the measurement meaningless. TOC is measured in estuaries
because the COD measurement does not yield satisfactory results in salt wa-
ter due to chloride interference.
The trace organics included in the program were chosen because they appear
most frequently on several USEPA priority lists relating to toxic substances;
for example, measurements required for the permit program, measurements re-
quired for the drinking water program, the Section 307(a) list, and several
listings proposed by the Office of Toxic Substances.
The effects of contaminants on aquatic organisms are complex. Synergistic
chemical/physical reactions, biomagnification, and other natural events
cannot be easily quantified. For these reasons and for the purposes of
the program, the best approach to determine the presence and potential
health threat of toxic substances in the ecosystem appears to be the chem-
ical analysis of fish and shellfish tissue. This has, therefore, been
included in the monitoring program.
D-42
-------
D.2.2.3 Parameters for Waste Load Allocations
As an example of possible parameter coverage for waste load allocation sur-
veys, Tables D-S and D-6 indicate minimum parameters for the preliminary and
primary intensive surveys discussed in Section D.2.1.2. The measurement lo-
cations indicated in the tables are those described in Section D.2.1.2. In
addition to the indicated parameters, it will usually be desirable to perform
a metals and pesticide scan on at least one sample from the preliminary sur-
vey and, based on the results, consider additional parameters for the primary
low flow survey.
D.2.3 Measurement Frequency Selection
Monitoring frequencies are established by the variations of the system
(sources and receiving water) and the nature of the pollutants (conservative
and nonconservative). Frequencies selected should be adequate to account for
variations in the flows and quality of pollution sources, the variations in
stream flow, and tidal action. This establishes a spectrum ranging from a
periodic grab sample (suitable for the rare steady-state condition) to con-
tinuous collection over a suitable time period.
D.2.3.1 Frequency for Background and Trend Data
Background and trend data must be representative of the variations in water
quality and changes in pollution occurring over the course of a year, and
the measurement frequency must be less than the shortest anticipated fre-
quency of pollutant variation. To aid in such sampling frequency determina-
tion, Tables D-7, D-8, and D-9 present the proposed sampling frequencies for
the national water quality monitoring program for rivers and streams, lakes
and impoundments, and estuaries and bays..
The sampling frequencies given in the foregoing represent the bare minimum
and, depending upon the anticipated variability, considerations should be
given to utilizing more frequent intervals. If at all possible, new stations
should be sampled on a weekly or biweekly basis for the first 6 months to
D-43
-------
TABLE D-5
PARAMETERS FOR PRELIMINARY SURVEY
Location
A
B
C
Di
E
F
Gi
H
I.
DO
X
X
X
X
X
X
X
X
X
Temp
X
X
X
X
X
X
X
X
X
Distance
Downstream
X
X
X
X
X
X
X
Travel
Time
X
X*
X*
X
X*
Flow
Measurement
X
X
X
X
X
X*
BOD20
X
BOD20
Inh
X
X
Nitrogen
Compounds
X
X
* Measurements need be taken at only one of the multiple locations desig-
nated by each of D., G-, or I..
TRIBUTARY
D-44
-------
TABLE 6
PARAMETERS FOR PRIMARY SURVEY
Location
A
B
C
Di
E
F
G.
H
'i
DO
X
X
X
X
X
X
X
X
X
PH
X
X
X
X
X
X
X
X
X
Temp
X
X
X
X
X
X
X
X
X
Distance
Downstream
X
X
X
X
X
X
X
Travel
Time
X
X
X
X
X
X
Flow
Measurement
X
X
X
X
X
X*
Continuous
DO
X
X
Nitrogen
Compounds
X
X
X
X
X
X
X
X
BOD5
k
X
BOD5
Inh
X
X
X
X
X
BOD2{)
Inh
X
X
X
X
X
o
I
t/l
Measurements need be taken at only one of the multiple locations designated by I..
H
TRIBUTARY
-------
TABLE D-7
PARAMETER LIST AND SAMPLING FREQUENCY
FOR THE NATIONAL MONITORING PROGRAM
Rivers and Streams
Parameter (Units) (STORET Parameter Code)
Sampling Frequency
o
Temperature (C°) (00010)
Dissolved oxygen (mg/l) (00300)
pH (Standard Units) (00400)
Conductivity (UMHOS/cn « 25°C) (00095)
Fecal Coliform (No./lOOmJl) (31616)
Total Kjeldahl nitrogen (mg/f) (00625)
Nitrate + nitrite (mg/£) (00630)
Total phosphorus (mg/£) (00665)
Chemical oxygen demand (mg/l) (00335)
Total suspended solids (mg/1) (0(.'30)
Representative fish/shellfish tissue analysis (see Table D-9)
Flow (CFS) (00060)
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Annually
Monthly
-------
TABLE D-7
PARAMETER LIST AND SAMPLING FREQUENCY
FOR THE NATIONAL MONITORING
PROGRAM (Cont'd)
Lakes and Impoundments, Including the Great Lakes
Parameter (Units) (STORET Parameter Code)
Sampling Frequency
o
i
pH (Standard Units) (00400)
Temperature (°C) (00010)
Dissolved oxygen (mg/l) (00300)
Conductivity (UMHOS/cm 8 25°C) (00095)
Fecal Coliform (No./100m£) (31616)
Total phosphorus (mg/£) (00665)
Total Kjeldahl nitrogen (mg/l) (00625)
Nitrate + nitrite (mg/l) (00630)
Total suspended solids (mg/l) (00530)
Representative fish/shellfish tissue analysis (see Table D-9)
Transparency Secchi Disk (Meters) (00078)
Seasonally
Seasonally
Seasonally
Seasonally
Seasonally
Seasonally
Seasonally
Seasonally
Seasonally
Annually
Monthly
-------
TABLE D-7
PARAMETER LIST AND SAMPLING FREQUENCY
FOR THE NATIONAL MONITORING
PROGRAM (Cont'd)
Estuaries and Bays
Parameter (Units) (STORET Parameter Code)
Sampling Frequency
o
i
oo
Temperature (°C) (00010)
Dissolved oxygen (mg/fc) (00300)
Total organic carbon (mg/fc) (00680)
pH (Standard Units) (00400)
Salinity (°/oo) (00480)
Fecal Coliform (No/lOOml) (31616)
Transparency Secchi Disk (Meters) (00078)
Total Kjeldahl nitrogen (mg/l) (00625)
>&
Total phosphorus (mg/£) (60665)
Nitrate + nitrite (mg/£) (00630)
Total suspended solids (mg/l) (00530)
Representative shellfish tissue analysis (see Table D-9)
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Annually
-------
TABLE D-8
TRACE ORGANICS AND METALS ANALYSES FOR WATER COLUMN
(1)
Parameter (STORET), (ug/1)
PCBs (39516)
Aldrin (39330)
Dieldrin (39380)
o.p-DDE (39327)
p,p'-DDE (39320)
o.p-DDD (39315)
p.p'-DDD (39310)
o.p-DDT (39305)
p,p'-DDT (39300)
Chlordane
cis isomer (39062)
trans isomer (39065)
cis nonachlor (39068)
trans nonachlor (39071)
Endrin (39390)
Methoxychlor (39480)
Hexachlorobenzene (39700)
Pentachlorophenol (39032)
Hexachlorocyclohexane
QL-BHC (39334)
Y-BHC (39810)
Arsenic (01002)
Cadmium (01027)
Chromium (01042)
Copper (01034)
Mercury (71900)
Lead (01051)
(1) For water column analysis when applicable (24).
TABLE D-9
TRACE ORGANICS AND METALS ANALYSES FOR FISH/SHELLFISH TISSUE AND SEDIMENT
Parameter (STORET:tissue, STORET:sediment), (vg/g tissue, ug/kg sediment)
PCBs (39520,39519)
Aldrin (39334, 39333)
Dieldrin (39387, 39383)
o.p-DDE (39329, 39328)
p.p'-DDE (39322, 39321)
o,p-DDD (39325, 39316)
p.p'-DDD (39312, 39311)
o.p-DDT (39318, 39306)
p.p'-DDT (39302, 39301)
Chlordane
cis isomer (39063, 39064)
trans isomer (39066, 39067)
cis nonachlor (39069, 39070)
trans nonachlor (39072, 39073)
Endrin (39397, 39393)
Methoxychlor (39482, 39481)
Hexachlorobenzene (39703, 39701)
Pentachlorophenol (39060, 39061)
Hexachlorocyclohexane
a-BHC (39074, 39076)
Y-BHC (39075, 39811)
Arsenic (01004, 01003)
Cadmium (71940, 01028)
Chromium (71939, 01029)
Copper (71937, 01039)
Mercury (71930, 71921)
Lead (71936, 01052)
D-49
-------
1 year of operation or until the data indicate that less frequent sampling
is warranted.
Fish samples should be collected annually in the fall, since contaminant con-
centrations are at their maximum at this time of year. Only fish samples
that will be most representative of the water quality in the area of interest
should be collected for tissue analysis. Migratory species should be dis-
counted. Two replicate whole fish composite samples of a representative bot-
tom feeder and one whole fish composite sample of a predator species should
be collected at each station. Commercially or recreationally important
species should be collected wherever possible. Each composite should include
at least five fish, each of approximately the same size. Because of their
sedentary existence and great water-filtering capabilities, shellfish are ex-
cellent concentrators of contaminants. Therefore, wherever possible, shell-
fish samples should be collected and analyzed, especially in estuarine
environments.
Where incidents of fish kill occur, the appropriate information should be
recorded. This should include the date of occurrence (or period if it per-
sists), the location or affected area, the species affected, estimates of
the magnitude of the kill (i.e., number of fish), and any other information
that would be useful. The frequency and magnitude of such events should
decrease as a result of implementing the areawide plans, and this can be a
dramatic way of indicating progress.
D.2.3.2 Frequency for Waste Load Allocation Surveys
Samples could be taken at any convenient time if stream conditions did not
vary. The necessary number of samples would be only that dictated by the
desired degree of precision of the results, taking into account the preci-
sion of the laboratory analytical methods. In theory, the times of collec-
tion and numbers of samples are dictated by the need to ensure both an
acceptable measure of the variations in stream conditions and an acceptable
precision of laboratory analysis. In practice, these considerations are tem-
pered by inescapable limitations of budget, personnel, and facilities, and
frequently by the amount of time available.
D-50
-------
There is no fixed number of samples that will yield results within selected
limits of precision in all situations. The number of samples needed for any
point on a stream varies with the variability in water quality at that point.
A preliminary estimate of the variability can be calculated after a limited
number of analytical results has been obtained. A preliminary prediction of
the number of samples needed to ensure final results within selected confi-
dence limits can be based on the preliminary estimate of variability. The
prediction can be refined as the number of analytical results is increased
until the point is reached at which a firm prediction of the number of sam-
ples required becomes possible. Data from a previous study under comparable
conditions may be used to determine variability and predict the number of
samples required.
In the absence of better information, daily grab samples should be taken from
each stream measurement site over at least a 14-day period. Furthermore, the
time of sampling at each site should be varied as much as possible to indi-
cate any diurnal variations. Try to collect at least one set of samples at
night to indicate photosynthetic effects. Review the analytical results from
the early samples, and adjust the frequency accordingly.
For continuous point source discharges, the sampling frequency will also be
dependent upon anticipated pollutant variability. If knowledge about the
time-varying characteristics of the discharge is required, collect a sequen-
tial discrete sample series. Hourly time steps will be adequate for most
continuous discharges, but in some instances either shorter time periods or
flowmeter pacing will be required. If only average daily loadings are re-
quired, twenty-four-hour, flow proportional composite samples represent the
best approach. These should be taken over a minimum of five consecutive
days, and longer if variability indicates.
0.2.3.3 Frequency for Storm Generated Discharges
For intermittent storm related discharges, measurement frequencies must be
quite short, especially for model calibration and verification where knowl-
edge of temporal variations is very important. The measurement interval
D-51
-------
required is related to catchment size, shape, slope, and percent impervi-
ousness. During sampling of the first few storms in a catchment, it is
prudent to estimate sampling intervals on the short side. They can be in-
creased later if the data warrant. Model input frequency requirements must
also be considered. Suggested minimum measurement intervals are given in
Table D-10. The first sample should be collected as close to the beginning
of the storm-generated runoff as possible. This can be accomplished by
triggering an automatic sampler at a predetermined indication of stage or
rate of rise. Subsequent samples can be paced by timer settings or a flow-
meter with flow increments selected so that the rising limb is well charac-
terized. It may not be necessary to analyze all samples on the falling limb.
Early data analysis will indicate if some can be eliminated or composited and
still allow adequate discharge characterization.
D.3 Flow Measurement Considerations, Equipment, and Procedures
Although flow can be thought of as simply another parameter, it is so often
neglected that it should properly be considered as an essential component of
a monitoring program. Flow measurements are absolutely necessary for mass
discharge calculations, stream and runoff studies, and model calibration and
verification.
D.3.1 General Considerations
Concentrations of natural constituents, such as alkalinity, hardness, and
minerals, generally vary inversely with stream flows. Total loads, or quan-
tities, of natural constituents carried by a stream, on the other hand, in-
crease as flow increases. The increasing water carried by the stream more
than balances the decreasing concentration to yield a greater load in terms
of a unit of total quantity, such as pounds per day. Other factors come
into play with unstable constituents. Time-of-water travel increases as flow
decreases, and this serves to accomplish natural purification in shorter dis-
tances. Higher densities of bacteria, for example, occur just below the
D-52
-------
TABLE D-10
MAXIMUM MEASUREMENT INTERVALS
Desirable Maximum Measurement
Interval (rain)
Catchment
Size
50 acres
100 acres
600 acres
3000 acres
Variable
Rainfall
Flow
Water Quality
Rainf al 1
Flow
Water Quality
Rainfall
Flow
Water Quality
Rainfall
Flow
Water Quality
Highly Impervious
Catchment
2
2
3
3
3
4
5
5
7
12
12
15
Highly Pervious
Catchment
3
3
4
5
5
7
12
12
20 .
20
20 .
30
D-53
-------
point of discharge at lower flows, but they die off in shorter distances
because of the longer time of travel. Likewise, BODs are higher near the
point of discharge but stabilize in shorter distances at low discharges.
The natural flow of uncontrolled streams usually varies over a wide range.
Stream flows follow precipitation patterns except in the colder areas of the
country, where precipitation falls as snow in winter and much of the surface
water is frozen. There can be wide differences in stream flow throughout
the year and in the annual flow cycle from year to year. Flow in most areas
tends to be high in late winter and to taper off to minimum quantities in
the fall. High flows usually occur in colder areas when relatively warm
spring rains melt the winter accumulation of ice and snow. However, the
natural cycle may be altered to a considerable extent in streams controlled
by impoundments. Thus, stream flows must be considered in selecting
periods for stream study because of the considerable variations in water
quality that accompany changes in flow. The objectives of the study are im-
portant in this selection, as they are in other decisions.
In manmade conduits, the effects of flow variation are probably greatest in
storm sewers. Although storm sewers are basically designed to carry storm
runoff, during periods of no rainfall they often carry a small but signifi-
cant flow (dry weather flow). This may be flow from ground water, or "base
flow," which gains access to the sewer from unpaved stream courses. Much of
the dry weather flow in storm sewers is composed of domestic sewage or in-
dustrial wastes or both. Where ordinances concerning connections to sewers
are lax or are not rigidly enforced, unauthorized connections to storm
sewers will appear. In some cases, the runoff from septic tanks is carried
to them. Connections for the discharge of swimming pools foundation drains,
sump pumps, cooling water, and pretreated industrial process water to storm
sewers are permitted in many municipalities and contribute to flow during
periods of no rainfall. In some areas, sewers classed as storm sewers are,
in fact, sanitary or industrial waste sewers due to the unauthorized or
inappropriate connections made to them. This may become so aggrevated that
D-54
-------
a continuous flow of sanitary or industrial wastes, or both, discharges into
the receiving stream. Furthermore, this "dry-weather" portion of storm
sewer flow may vary significantly with time.
Storm runoff is the excess rainfall which runs off the ground surface after
losses resulting from infiltration to ground water, evaporation, transpira-
tion by vegetation, and ponding occur. In general, storm runoff is inter-
mittent in accordance with the rainfall pattern for the area. It is also
highly variable from storm to storm and during a particular storm. The
time-discharge relationship, or hydrograph, of a typical storm, with its
synchronous time-precipitation relationship, or hyetograph, is illustrated in
Figure D-3. The meanings of various parameters given in the figure are:
R - Rainfall retained on the permeable portion of the drainage
basin, and not available for runoff.
P - Precipitation in excess of that infiltrated into the ground,
plus that retained on the surface. (Equals the volume of
flood runoff.)
F - Average infiltration of the ground during the storm.
T - Period of rise from the beginning of storm runoff to peak of
the hydrograph.
T - Time from center of gravity of rainfall excess to the hydro-
graph peak (lag time).
b.,b2 - Baseline separating groundwater discharge from surface
runoff.
The total volume of runoff for a particular storm is represented by the areas
between the baseline and the hydrograph.
D-55
-------
o
in
IN./MR
0.50
u
DURATION OF PRECIPITATION EXCESS
0.2S
1
av
av
Jft.
CF.NTROID OF PRECIPITATION EXCESS
NOTE: I INCH » 2.54 CM
1 CUBIC FOOT » 28.3 LITERS
3 4
APRIL. 1959
FIGURE D-3
TYPICAL STORM HYETOGRAPH AND HYDROGRAPH
-------
To illustrate some of the problems in measuring storm runoff in small basins,
peak flows exceeding 85 cubic meters per second per 260 hectares (3000 cfs
per square mile) have been observed. Lag times (t ) of 15 minutes to a
hydrograph peak of about 28 cubic meters per second (1000 cfs) from a 600-
hectare (2.3 sq mi) area are not uncommon. With such rapid changes in the
flow, only highly responsive flow measurement methods can be used. The high
rates of flow, with accompanying high velocities, further limit the usable
flow measuring methods.
All flow data must be synchronized with time, at least on a watch time basis,
to have any useful meaning. A particular need for attention to the time
element occurs in the measurement of flows from small urban storm sewers in
order to define the hydrograph and to provide data for the development and
verification of rainfall-runoff-quality models. Peak flows, storm runoff
volumes, daily flows, or other flow parameters are often correlated with
similar flows at other points on a storm sewer or stream, or with flows of
other storm sewers or streams, to provide a means for flow estimation.
Correlations with temperature, soil moisture, or antecedent precipitation
may be made at times. In most cases, it is essential that the correlated
variables be synchronous, so accurate timing of the data is often required.
It is mandatory if time-series analysis is contemplated.
Timing of measured flows and collection of quality samples can be useful
in determining sources of pollution. For example, they can be related to
time of release of pollutants from industrial plants, or to the time of
accidental spills of pollutants. The time of travel of pollutants along a
stream or storm sewer can be estimated from the time of travel of small
rises or other flow changes in the channel.
D.3.2 Flow Measurement Equipment
This brief discussion is intended to provide an overview to aid the 208 plan-
ner in the selection of equipment for the quantitative measurement of flows.
For further reading, see Shelley and Kirkpatrick (8), ASME (9),
D-57
-------
Replogle (10), McMahon (11), USDI Bureau of Reclamation (12), Leupold and
Stevens (13), and any of the many standard texts on hydraulics and fluid
mechanics.
Any flow measurement system consists of two distinct parts, each having a
separate function to perform. The first, or primary element, is that part
of the system which is in contact with the fluid, resulting in some type of
interaction. The secondary element is that part of the system which trans-
lates this interaction into the desired readout or recording. While there
is almost an endless variety of secondary elements, primary elements are
related to a more limited number of physical principles, being dependent upon
some property of the fluid other than, or in addition to, its volume or mass
such as kinetic energy, inertia, specific heat, or the like. These primary
element physical principles form a natural classification system for flow-
measuring devices as presented in Table D-ll.
D.3.2.1 Desirable Equipment Characteristics
Not all types of flow meters are suitable for measuring wastewater flows.
The severe conditions and vagaries of many of these flows place a number of
very stringent design requirements on flow measurement equipment if it is to
function satisfactorily. No single design can be considered ideal for all
flow measurement activities in all flows of interest. Despite this, one can
set forth some equipment "requirements" in the form of primary design consid-
erations and some desirable equipment features in the form of secondary
design considerations.
The following are primary design considerations for equipment that is to be
used to measure more difficult wastewaters such as storm and combined sewer
flows:
1. Range. Since flow velocities may range from 0.03 to 9 m/s (0.1 to
30 fps), it is desirable that the unit have either a very wide
range of operation; be able to automatically shift scales; or
otherwise cover at least a 100 to 1 range.
D-58
-------
TABLE D-ll
FLOW METER CATEGORIZATION
Division
Quantity
Quantity
Quantity
Quantity
Quantity
Quantity
Quantity
Quantity
Quantity
Quantity
Quantity.
Rate y
Rate -;"
Rate
Rate •'
Rate •
Rate '' |
Rate.v iv,
Rate .-,£•
Rate Msl
Rate
Rateyj; V
Rate . py1
.tate •
Rate" |
Raie
Raie -,j
Rate ;••
Rate -
Volumetric . « "^
Volumetric ^ ^
Volumetric ~~
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Differential Pressure
Variable Area „< <•
Variable Area v V
Variable Area'-' '«=
Head-Area
Head- Area
Head- Area
Head-Area
Head- Area
Head- Area
Head- Area
Head- Area
Head-Area
Head-Area
Head- Area
Flow Velocity
Flow Velocity
Flow Velocity
Flow Velocity
Flow Velocity
Flow Velocity
Flow Velocity
Flow Velocity
Force-Displacement
Force- Displacement
Force-Displacement
Fore*- Oi sp 1 acement
Fore*- Displacement
Force-Momentum
Forc*-Mom*ntum
Force- Momentum
Force-Momentum
Thermal
Thermal
Thermal
Other
Other
Other
Other
Other
Other
Other
Type
Weigher
Tilting Trap
Weight Dump
Metering Tank
Reciprocating Piston
Oscillating or Ring Piston
Nutating Disc
Sliding Vane
Rotating Vane
Gear or Lobed Impeller ,
Deth ridge Wheel •&.' t*\ H^»
Venturi
Da 11 Tube
Flow Nozzle
Rounded Edge Orifice
Square Edge Orifice
Square Edge Orifice
Square Edge Orifice
Square Edge Orifice
Centrifugal
Centrifugal
Centrifugal
Impact Tube
Impact Tube
Linear Resistance
Linear Resistance
Linear Resistance
Gate
Cone and Float
Slotted Cylinder and Piston
Weir
Weir
Flume
Flume
Flume
Flume
Flume
Flume
Flume
Flume
Open Flow Nozzle
Float
Float
Tracer
Vortex
Vortex
Turbine
Rotating Element
Rotating Element
Vane ,'• >- t.:-juit l^ -vtv* ^ ^
Hydrometric Pendulum
Target
Jet Deflection
Ball and Tube
Axial Flow Mass
Radial Mass
Gyroscopic
Mangus Effect
Hot Tip
Cold Tip
Boundary Layer
Electromagnetic
Acoustic
Doppler
Optical V
Dilution . ^
Electrostatic VV***" '^
Nuclear Resonance \ V.-"*^
Subtype
; /
»;M*/>
-
Concentric
Eccentric
Segmented
Gate or Variable Area
Elbow or Long Radius Bend
Turbine Scroll Case
Guide Vane Speed Ring
Pilot-Static
Pitol Venturi
Pipe Section
Capillary Tube
Porous Plug
Sharp Crested
Broad Crested
Venturi
Parshall
Palmer-Bowlus
Diskin Device
Cutthroat
San Dioas
Trapezoidal
Type HS, H, and HL
Simple
Integrating
Vortex-Velocity
Eddy-Shedding
Horizontal Axis
Vertical Axis i ±y^\ (y
1/k.i/l} i^ylmO*'**™"' ' ' ^
VA
C' * • • ,'
^ A ^' V
\v'
D-59
-------
2. Accuracy. For most purposes, an accuracy of ±10 percent of the
reading at the readout point is necessary, and there will be
applications where an accuracy of ±5 percent is highly desirable.
Repeatability of better than ±2 percent is desired in almost all
instances.
3. Flow Effects on Accuracy. The unit should be capable of maintaining
its accuracy when exposed to rapid changes in flow; e.g., depth and
velocity changes in an open channel flow situation. There are
instances where the flows of interest may accelerate from minimum
to maximum in as short a time period as 5 minutes.
4. Gravity and Pressurized Flow Operation. Because of the conditions
that exist at many measuring sites, it is sometimes desirable that
the unit have the capability (within a closed conduit) of measuring
over the full range of open channel flow as well as the conduit
flowing full and under pressure.
5. Sensitivity to Submergence or Backwater Effects. Because of the
possibility of changes in flow resistance downstream of the measuring
site due to blockages, rising river stages including possible re-
verse flow, etc., it is highly advantageous that the unit be able to
continue to function under such conditions or, at a minimum, be able
to sense the existence of such conditions which would lead to
erroneous readings.
6. Effects of Solids Movement. The unit should not be seriously af-
fected by the movement of solids such as sand, gravel, debris, etc.,
within the fluid flow.
7. Flow Obstruction. The unit should be as nonintrusive as possible
to avoid obstruction or other interference with the flow, which
could lead to flow blockage or physical damage to some portion of
the device.
D-60
-------
8. Head Loss. To be usable at a maximum number of measurement sites,
the unit should induce as little head loss as possible.
9. Manhole Operation. To allow maximum flexibility in utilization, the
unit should have the capability of being installed in confined and
moisture-laden spaces such as sewer manholes.
10. Power Requirements. The unit should require minimum power at the
measuring site to operate; the ability to operate on batteries is a
definite asset for many installations.
The following secondary design considerations are desirable features for flow
measuring equipment.
Site Requirements. Unit design should be such as to minimize site require-
ments, such as the need for a fresh water supply, a vertical drop, excessive
physical space, etc.
Installation Restrictions or Limitations. The unit should impose a minimum
of restrictions or limitations on its installation and be capable of use on
or within sewers of varying size.
Simplicity and Reliability. To maximize reliability of results and opera-
tion, the design of the unit should be as simple as possible, with a minimum
of moving parts, etc.
Unattended Operation. For the majority of applications, it is highly desir-
able that the equipment be capable of unattended operation.
Maintenance Requirements. The design of the equipment should be such that
routine maintenance is minimal and troubleshooting and repair can be effected
with relative ease, even in the field.
Adverse Ambient Effects. The unit should be unaffected by adverse ambient
conditions such as high humidity, freezing temperatures, hydrogen sulphide
or corrosive gases, etc.
D-61
-------
Submersion Proof. The unit should be capable of withstanding total immersion
without significant damage.
Ruggedness. The unit should be of rugged construction- and as vandal and
theft proof as possible.
Self Contained. The unit should be self contained insofar as possible in
view of the physical principles involved.
Precalibration. In order to maximize the flexibility of using the equipment
in different settings, it is desirable that it be capable of precalibration;
i.e., it should not be necessary to calibrate the system at each location and
for each application.
Ease of Calibration. Calibration of the unit should be a simple, straight-
forward process requiring a minimum amount of time and ancillary equipment.
Maintenance of Calibration. The unit should operate accurately for extended
periods of time without requiring recalibration.
Adaptability. The system should be capable of: indicating and recording
instantaneous flow rates and totalized flows; providing flow signals to as-
sociated equipment (e.g., an automatic sampler); implementation of remote
sensing techniques or incorporation into a computerized urban data system,
including a multisensor single readout capability.
Cost. The unit should be affordable both in terms of acquisition and instal-
lation costs as well as operating costs, including repair and maintenance.
It is not necessary that all of these primary and secondary design consider-
ations be achieved for all applications. For example, flow measurement
devices used to calibrate others need not necessarily be self-contained, nor
would unattended operations be required. Furthermore, meeting all of the
listed design considerations for all installations and settings would be
difficult, if not impossible, to achieve in a single design. Nonetheless,
D-62
-------
the primary and secondary design considerations can be used to formulate a
set of evaluation parameters against which a given design or piece of equip-
ment can be judged. Since application details may make certain parameters
more or less important in one instance or another, no attempt has been made
to apply weighting factors or assign numerical rank. The evaluation factors
should prove useful, as a check list among other things, for the 208 planner
who has a flow measurement requirement and who may require assistance in the
selection of his equipment. The evaluation parameters together with quali-
tative scales, are presented in the form of a flow measurement equipment
checklist in Table D-12.
D.3.2.2 Evaluations of Some Promising Devices
A slightly modified form of the flow measurement equipment checklist given in
Table D-12 has been used to evaluate the various flow-measuring devices and
techniques of Table D-ll, and a matrix summary is given as Table D-13. It
must be emphasized that these evaluations are made with a highly variable
wastewater application such as storm or combined sewer flow measurement in
mind and will not necessarily be applicable for other types of flows.
Only a few of the evaluation parameters normally have numbers associated with
them. To assist the reader in interpreting the ratings, the following gen-
eral guidelines were used. If the normal range of a particular device was
considered to be less than about 10 to 1, it was termed poor; if it was con-
sidered to be greater than around 100 to 1, it was termed good. The inter-
mediate ranges were termed fair. The accuracy that might reasonably be
anticipated in measuring storm or combined sewer flows was considered rather
than the best accuracy achievable by a particular device. For example, al-
though a sharp-crested weir may be capable of achieving accuracies of
±1.5 percent or better in clear irrigation water flows, accuracies of much
better than ±4 to 7 percent should not necessarily be anticipated for a
sharp-crested weir measuring stonnwater or combined sewer discharges. If the
accuracy of a particular flow-measuring device or method was considered to be
better than around ±1 to 2 percent, it was termed good; if it was considered
D-63
-------
O Q C G C O CXQOJD O O p Q.J3.D&t>J3 O_O Or>
c c c o • c:oTroro:o o «D 0:0 o:o~oro:o~b o~oro
U O O • • a.Q'Q-CLO'O O Q O O 3 O D 3 3 O O~O 9
c c_u c • i ;:> o op a,a o D o » gio 0:0:0 op
o c o • q^nnpj3io o z> r)
i c o • < Tra^r) o'o o o o oo
G o.o •• ao oo oo o a
C C C G C Gtmx>X3O O O O O
TABLE D-12
FLOW MEASUREMENT EQUIPMENT CHECKLIST
Designation:
Evaluation Parameter
1
2
3
4
S
6
7
8
9
10
11
12
13
14
IS
16
17
Range
Accuracy
Flow Effects on Accuracy
Gravity & Pressurized Flow
Operation
Submergence or Backwater
Effects
Effect of Solids Movement
Flow Obstruction
Head Loss
Manhole Operation
Power Requirements
Site Requireaents
Installation Restrictions
or Limitation.;
Simplicity and Reliability
Unattended Operation
Maintenance Requirements
Adverse Ambient Effects
Scale
Q Poor O Fair O Good
O Poor O Fair D Good
O High a Moderate O Slight
D No Q Yes
Q High O Moderate Q Low
O High O Moderate D Slight
Q High O Moderate D Slight
O High a Medium a Low
Q Poor o Fair a Good
Q High Q Medium O Low
Q High a Moderate O Slight
Q High Q Moderate a Slight
O Poor O Fair O Good
O No Q O Yes
O High D Medium a low
Weight and Score
Q HiBh - •
Submersion »-- __
-------
to be worse than around ±10 percent, it was termed poor. The intermediate
accuracies were termed fair.
The flow-measuring devices and techniques were not rated on two evaluation
parameters, submersion proof and adaptability, because these factors are so
dependent upon the design details of the secondary element selected by the
user.
In comparison with Table D-13, Table D-14 offers a different (and even more
subjective) comparison of the most promising primary devices or techniques.
Each method is numerically evaluated in terms of its percent of achievement
of several desirable characteristics. Dilution techniques as a class appear
to be most promising of all. In view of the current state-of-the-art, how-
ever, their usefulness is probably greatest as a tool for in-place cali-
bration of other primary devices. They have also been extremely useful for
general survey purposes and have found some application as an adjunct to
other primary devices during periods of extreme flow such as pressurized flow
in a conduit that is normally open channel.
Acoustic open channel devices are also quite promising; but, because of their
dependency upon the velocity profile and the frequently resulting requirement
for several sets of transducers, they are presently only justifiable for very
large flows in view of the expense involved. The usefulness of the Parshall
flume is evidenced by its extreme popularity. The requirement for a drop in
the floor is a disadvantage, and submerged operation may present problems at
some sites. Known uncertainties in the head-discharge relations (possibly up
to 5 percent) together with possible geometric deviations make calibration in
place a vital necessity if high accuracy is required. Palmer-Bowlus type
flumes are very popular overall. They can be used as portable as well as
fixed devices in many instances, are relatively inexpensive, and can handle
solids in the flow without great difficulty.
All point velocity measuring devices have been lumped together in the current
meter category. In the hands of a highly experienced operator, good results
can be obtained (the converse is also true, unfortunately), and they are
D-65
-------
TABLE D-13
FLOWMETER EVALUATION SUMMARY
b
W
C
at
GraviMtric-all types
Volu»etric-all types
Verturi Tube
Oall Tube
Flow Nozzle
Orifice Plate
Elbow Meter
Slope Area
Sharp-Crested Weir
Broad-Creited Heir
Subcrittcal Fluae
Parshall Fluw
Pa.awr-Bowlus t-'luae
Dukin Device
Cutthroat Fluve
San Diaa$ Flu**
Trapezoidal RUM
Type HS, H ( HL Fluae
Open Flow Nozzle
Float Velocity
Tracer Velocity
Udr-Shoddifu
Turfiine Meter
Rotating-Rlefjent Meter
Vane Meter
NrdroMtric Pendulia
Target Meter
Forci-MoMiitui
Hot-Tip Meter
toundary Layer Meter
ElectrougiMtic Meter
Acoustic Mettr
Doppler Meter
Optical Meter
Dilution
legend:
F - Fair
r. - Goal
H - Nigh
L • UM
M • Mediua or Mode
C
P
P
P
P
P
P
F
F
F
F
G
G
G
C
C
F
P
P
F
P
P
G
F
C
P
F
G
rite
G
C
G
G
G
F
F
P
F
F
F
F
F
F
F
P
F
F
F
P
G
C
G
G
C
P
G
t
H
H
S
S
s
s
s
H
M
S
S
s
D
S
S
H
S
S
s
s
s
s
s
s
s
s
M
N -
P .
s -
Y •
Y
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
Y
N
N
Submergence or laciwaler liffccts
Effect of Solids Muveacnt
L H
L H
L S
L M
t. H
L S
H S
M H
H H
M M
L S
H M
H M
L S
L H
L H
1 H
I M
L M
VIS
Y L S
YIN
Y L H
N L S
Y 1 S
No
Poor
Slight
Y«
rlow Obstruct ion
H
H
S
S
H
S
S
H
S
H
S
S
S
s
H
H
M
M
M
S
s
s
s
s
s
M
J
*
H
L
I
H
L
L
H
L
L
L
H
H
L
L
M
1.
L
L
L
I
L
I
L
L
Manhole Operation
P
P
P
P
P
P
G
F
C
G
F
G
C
G
P
F
G
P
P
P
F
F
F
G
VI
I
1
11
I
M
L
L
L
L
L
I
L
L
L
L
L
I.
L
L
I
H
M
H
M
M
L
M
r.
e
i
II
H
H
H
H
H
H
M
M
S
S
s
M
s
H
s
s
H
M
M
M
S
S
u
•c
g
e
II
H
H
M
S
S
S
M
S
S
M
S
H
S
S
H
M
M
M
M
S
S
if
n
Sierilicilr and •<-•
P
F
C
G
G
G
r.
G
c
G
C
G
G
F
G
C
P
F
F
F
F
G
F
Unattended llperjl
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
Y
N
Y
N
Y
Y
Y
Y
Y
N
Y
£
S
Maintenance Mcqui
H
M
M
H
L
L
H
L
H
L
M
L
M
H
H
H
M
M
M
L
M
Advene fcebienl 1
Subversion Proof
M -
M •
M .
M -
M -
M •
M •
M -
H -
M -
M -
H '
S -
S -
N -
s •
s -
s -
s -
s -
H -
S -
1
i f
1 1
F Y
G Y
G Y
F Y
G Y
G Y
G Y
ti V
F Y
r, Y
G Y
C N
r v
F Y
C N
P Y
C Y
F V
F Y
F Y
G N
F N
1 ion
lihral ion
Precal ibra
Ease of Cj
Y C
1 C
Y G
Y C
N F
M F
Y G
Y u
Y F
Y <;
Y G
Y F
Y C
Y r.
Y G
Y C
Y G
Y C
Y C
Y C
N G
r of I'al ihrat ion
Miinlenjnc
F
G
F
G
P
G
G
P
F
G
F
C
F
F
G
G
G
G
G
G
G
G
n
3 5
- H
- H
- H
- M
- H
• L
• I.
- H
- L
- L
- L
- L
- L
- L
• L
- H
- H
- II
- L
- H
- H
- H
- H
• H
- H
- H
- L
- H
I'orl abi 1 i ty
N
N
N
N
Y
V
Y
N
N
Y
Y
V
\
N
Y
N
N
S
N
N
N
Y
Y
D-66
-------
TABLE D-14
COMPARISON OF MOST POPULAR PRIMARY DEVICES OR TECHNIQUES
Prlmry Device
or Technique
Dilution
Acoustic (Open Channel)
Parshall Hate
Palner Bowl us Fluae
Current Meter
Electromagnetic
Acoustic (Pressure Flow)
Open Flow Nozzle
Sharp-Crested Heir
Flow Tube
Venturi Tube
Trajectory Coordinate
Slope Area
Desirable Characteristic (% of Achievement)
Range
100
100
90
80
90
SO
100
60
60
SO
20
80
80
Uncallb.
Accuracy
100
100
9S
90
95
100
100
9S
95
100,
100
70
SO
Head
Loss
100
100
80
85
100
100
100
70
70
95
90
SO
ion
Free Fro«
Upstream
Effects
100
60
90
90
100
100
60
80
80
40
70
100
20
Free Fro*
Downstream
Effects
100
90
80
8S
100
100
90
75
80
100
100
70
100
Solids
Bearing
Liquids
100
95
90
90
90
100
95
80
SO
95
90
100
100
Portability
100
80
70
90
100
0
0
80
80
0
0
100
inn
Unattended
Operation
80
100
100
100
0
100
100
95
90
100
100
0
0
Comment s
Especially useful as a calibra-
tion tool.
Good in large flows hut
expensive.
Requires drop in floor.
Good overall.
Results are very operator
dependent .
Generally requires pressure flow.
Netted transducers recommended .
Good if head drop is available.
Hill require frequent cleaning.
Pressurized flow only.
Pressurized flow only.
Requires free discharge.
Use us last resort.
-------
often used to calibrate primary devices in place or for general survey work.
They are generally not suited for unattended operation, however.
Electromagnetic flowmeters show considerable promise where pressurized flow
is ensured, as do closed pipe acoustic devices. Neither can be considered
portable if one requires that the acoustic sensors be wetted, a recommended
practice for most wastewater applications.
Open flow nozzles and sharp-crested weirs are often used where the required
head drop is available. Weirs will require frequent cleaning and are best
used as temporary installations for calibration purposes. Flow tubes and
Venturis are only suitable for pressurized flow sites such as might be en-
countered, for example, at the entrance to a treatment plant.
Trajectory coordinate techniques, such as the California pipe or Purdue meth-
ods, require a pipe discharging freely into the atmosphere with sufficient
drop to allow a reasonably accurate vertical measurement to be made, a situa-
tion not often encountered in sewers. Slope area methods (e.g., Manning,
formula) must generally be considered as producing estimates only, and con-
sequently should be considered as the choice of last resort (despite their
apparent popularity).
D.3.2.3 Review of Commercially Available Equipment and Costs
The number of commercial firms that offer liquid flow-measuring equipment in
the marketplace today is astoundingly large, probably well in excess of 200.
Many manufacturers offer more than one type of primary device (and these
typically in numerous models) and, when combined with secondary device
choices, the number is virtually overwhelming. Thus, no attempt to cover all
available equipment can be made here. We simply note that two or more firms
offer all devices that were described except for sharp-crested weirs, which
are usually fabricated directly by (or for) the user in accordance with
specifications for the particular measuring site.
D-68
-------
The firms offering flow-measuring equipment as at least a part of their prod-
uct line range from very large, well-known manufacturers that have offered a
wide range of flow-measuring equipment for over a century to relatively small
organizations with a limited product line that has only recently been intro-
duced. This latter category should not be excluded from consideration solely
because of their seemingly novitiate status. The principals involved fre-
quently have many years of experience, and their designs often reflect the
most up-to-date expressions of the state-of-the-art.
The revolution in the electronics industry, especially as regards solid-state
designs and integrated circuitry, has not gone unnoticed by most flowmeter
manufacturers; as a result, many new, sophisticated secondary devices have
recently appeared, and older equipment is frequently being upgraded in design
to reflect the more modern technologies. Furthermore, many of these new
secondary devices are of digital (rather than analog) design and are fre-
quently computer compatible as supplied, offering tremendous possibilities
for system structure.
A listing, by no means complete, of some manufacturers who offer flow-
measuring equipment in the categories listed in Table D-14 is presented in
Table D-15. Under the heading "Company," the name, address, and telephone
number have been provided. Under the heading "Products" only those products
bearing on the flow measurement categories of Table D-14 have been listed,
even though the particular company may have a much more extensive flow meas-
urement product line. The product emphasis was placed on primary devices,
with secondary devices (in the form of level gages) indicated only where
they are offered as "flowmeters." It can be generally assumed that each
manufacturer offers a complete line of secondary elements for use with his
primary devices.
Table D-15 can be used to obtain direct, up-to-date information on all of the
types of equipment discussed from at least two suppliers. Reference can be
made to Shelley and Kirkpatrick (8) for descriptions of the offerings of
these and a number of other manufacturers.
D-69
-------
TABLE D-15
SOME FLOW MEASUREMENT EQUIPMENT MANUFACTURERS
o
i
Co*...,
American Chain and Canla Company. Inc.
ALTO Irislol Division
••terbury. Connecticut O6720
Tel. phone (201) 756-4451
•arffer Neler. Inc.
Instrument Division
4S4S nest troun llcer toed
Nll.auko.. stlsconsln S1221
Telephone |4I4) 1SS-MOO
••ilfer Meter. Inc.
Precision Products Division
6116 East 15th Street
Tulsa. Oklahoma '41 IS
Telephone (Hit) 116-4611
• IP • A Ifatl of Omul Sllnal
1600 Division loed
mest •arutck. n.l. 02*91
Telephone (401) IIS-IOOO
•rooks Instrument Division
Emerson Electric Company
407 Best Vine Stieet
llatrielo. Pennsylvania 19440
Telephone (215) 247-2V*
Conlrolotron Corporation
176 Control Avenue
Farmiotdale. II. »~ »ork II71S
Tele|ikone (SI6| 249-44M
Cushing En|lneerin| Inc.
1J64 Commercial Avenue
Itorthbraok. Illinois 6IIO62
telephone |1I2) S64-D50D
C.K. Stevens, Inc.
P. 0. *>• 619
lennelt Square. Pennsylvania 19141
Telephone (215) 444-IH.I6
Dreielbrook fn|lneerln| Company
20S (el Ik Valley load
Horsham. Pennsylvania 19044
Telephone (215) 674-1214
environmental Measurement Systems
A PI vl s ion oC IfesttBeT
90S Hester Avenue Hnrlh
Seattle. Kashln(tun 981(19
l.-lephunc (206) 215-1621
l*.li la«a-
i|ti< int
ISO Na»»au St r«et
Suitu 1410
Hew York. New lor. 1001*
Tel -.-phone (>|2| S4V-2470
Fischer I Purler t'.n.
•eleph^e ,,,5, 675-6,.,U
Product*
Coflb.nat.on depth and
velocity Maturing device
in a single unit
Flow tuh«». open flow
not ties. Parxhall fluaws
Acoustic (open channel)
Flow lubes, open flow
not i lei, Par!»hall fluaes.
"universal" venturl tube*
Mectroautfnet Ic
•
Acoustic (pressure flow)
Elect ro*u| net 1 c
Acouat ic level gage
Electronic level |a|e
Acoustic (open channel)
Ue.liounnrlic, fl.~
'j.iel!'1" "'""' '""
,:.^,.n,
Carl Fisher ami Coayany
Division of f-oraulahs. Inc.
S29 Vest Fourth Avenue
P. 0. to • 10*6
Eicoodido. Californl.1 9202S
Telephone (714) 74S-6421
Mi MI Co.
P. O. tea S7S
VettfietJ. New Jersey
N. V. Office: Telephone (212) 227-6668
The Foiboro Coakpany
Foxtwro, Ki«*i^achu!«etts O'OJS
Telephone (617) S43-I7SO
Illnde l.n|ineerln| Company of California
P. 0. .roi S6
Sa rat ofa, California 9SII7U
Telephone (40B» 57t-41H
Inlerocean Systems. Inc.
JSIO Curti Street
San Diego. Cilifornia 92110
Telephone (714) 299-4SOO
tah) Sc 1 ent 1 f i c Inst riaaent Cnrporat ioa
P. O. fcii 1166
n Cajon. California
Telephone (714) 4.44- 2I&«
F. B. Leopold CoaBpanx
Division of Syhron Corporation
227 5, Division St .
2 e 1 i e nop 1 e . Pen ns y 1 van i a 1 606 5
Telephone (412) 4S2-610U
teujMild 1 Steven*. Inc.
p. o. *» sat
600 N. H. Hv^ctow Drive
•eaverton. Ihrcnn V704IS
Telephone (Sut| 64r.-9l7|
Manning 1. nvironatenl.il Corp.
l?ll Du fto.*. Street
P. n. ton I3S|>
Telcphnnc (4<>0) 427-i)230
Marl if. ioiS-l.- Air
2110 l^lipMNir Drive
Olytapiii. W.i-.hinj>tOii VBSO.*
lrlc " u""' MI-"»"
Mft r 1 1 .i|i»- . Inc.
•e--lls U.T4/
I«-U-|I||.IMC |ft| 7 | .lf>*}- 7NO(l
NH I'll, Jill t V . IlK .
IS H*-iil.ili Hi. i,l
Nl1** tlllt-IIM4 I't'llfl^ylViliMfl |rt'*H)
Irl- |>li<.,i< (.'If. J tIS («'•»
ITuJu,,,
Fluorescent dyes
Painter -Aow lux f lintws
liluct roavugnet ic . level
l<*fcs
Palnver-lkiwliis II UMTS
Current nkclcrs. level
gages
Current »cters.
f luoresrenl dyes
Open flow noiili-t.
I'll Inter *BOW III!, f llMBCN ,
f;ir^hal 1 f luaK-it
II oat Ifvc 1 (teigos
Atoustiv ami "dl|i|>er"
level R^ge*
1 l«tii «" 1 -v '1 ••'• n*
Curl. ili It- V-noiih ki-ii . ,
l.-vrl K..K,->
-------
TABLE D-15
SOME PLOW MEASUREMENT EQUIPMENT MANUFACTURERS (Cont'd)
Products
Product !t
N-Con Systeas Coapany
MS Main Street
New Kochelle. New York 1(1801
Telephone (914) 23S-I020
Nusonics, Inc.
9 Keystone Place
Paraaus. New Jersey 076S2
Telephone (201) 26S-2400
Ocean Research EquIpaent. Inc.
Falaourh. Massachusetts 02S4I
Telephone (617) S48-S800
The Permit It Coapany
Division of Syhron Corporation
E49 Midland Avenue
Paraaus. New Jersey
Telephone (201) 262-8900
Plasti-Fab. Inc.
II6SO S. M. Rldgcvlev Terrace
Beaverton. Orefon 9700S
Telephone (S02) 644-1428
Plocon. Inc.
An Affiliate of Carl f. Bucttner
6 Associates. Inc.
SI06 liaapton Avenue
St. Louis. Missouri 61109
Telephone (314) 15J-S99J
PIOTAC
Min-EII Coapany, Inc.
1689 Slue Jay Lane
Cherry Mill. New Jersey usual
Telephone (6O9) 429-0421
Robertshaw Controls Coapany
P. O. Box 1S2)
Knoxville. Tennessee* 37917
Telephone (6IS) S46-OS24
Saratoga Systeas, Inc.
10601 South Saratoga-Sunnyvale Road
Cupertino. California 9SOI4
Telephone (41)8) 247-7120
Si-arpa l.iihoratories. Inc.
4k Liberty Street. Brainy Boru Stiition
Heluchin. New Jersey 0884(1
Telephone (2111) S4»-«2l>n
Float and "dipper"
level gages
Acoustic (pressure flow)
Acoustic (open channel)
Flow tubes, open flow
noziles. Parshall fluiaes.
venturi tubes
Palaer-Bowlus fluaws.
Parshall fluaes, V-notch
Heir boxes
Open channel flow tube
Current ajelur flow tube
Parshull fluae*.
level gages
Acoustic (pressure flow)
Acoustic (pressure flow)
Slgaaaotor, Inc.
14 Illiaheth Street
Middleport, New York I4IOS
Telephone (716) 73S-36I6
Singcr-Aaerican Meter Division
13500 I'hilannt Avenue
Philadelphia. Pennsylvania 19116
Telephone (2IS) 637-2100
Sirco Controls Coapany
881S Selkirk Street
Vancouver 14, British Coluabia, Canada
Telephone (604) 261-9321
Taylor
Sybron Corporation
Taylor Instrument Process Control Division
Telephone (716) 235-SOOO
Tri-Aid Sciences, Inc.
161 Norris Drive
Rochester, New York 14610
Telephone (716) 461-1660
Universal engineered Systeas, Inc.
7071 Coaaerce Circle
Pleasanton, California 94S66
Telephone (41S) 462-IS43
Vickcry-Siaas, Inc.
P. O. Box 4S9
Arlington, Texas 76010
Telephone (817) 261-4446
Wallace-Murray Corporation
Carolina Fiberglass Plant
P. O. Box SKO
SIO hast .lanes Street
Wilson. North Carolina 2789.1
Telephone (919) J37-S37I
Hesaar Industrial Systeas Division
90S Dexter Avenue North
Seattle, Nashincton 98109
Telephone (-'(K.I .'85-2420
Nestinghouse Flcelric Corporation
Oceanic Division
I'. I). Ho* I4HH. Mail Stop 91130
Annapolis, Maryland Jl KM
telephone I Mil I 7(iS-S(,Mi
Bubbler level
Pnlaer-Bowlus fluaes.
Parshall fluaes. level
gages
Acoustic level gage
Elect roaagnet ic
Acoustic level gage
Palaer-Bowlus fluaes
Parshall fluaes, venturi
Parshall fluaes
Acoust ie level gages
Acour.t ie (open channel*
-------
In these days of inflation, little can be said about equipment costs except
in a very cursory fashion. For example, one manufacturer is anticipating
a 30-percent increase in the cost of basic flow tube forgings, catalog
pricing is giving way to individual quotes for larger systems, and some
manufacturers are quoting tentative estimates subject to adjustment at de-
livery. Desired features such as remote readouts, digital outputs, recorder
types, battery parts, etc., add another diversion to total system costs.
The following discussion is more indicative them precise, and all costs must
be increased if many accessories are desired.
Dilution flow measurement systems can be put together for under $3K. The
chemicals (salt, dye, etc.) are inexpensive. Acoustic open channel devices
start at around $5K, and larger systems are quoted on an installation basis
only, with $15-40K being a typical charge for a four-path system and some
large, complex installations approaching $100K in cost. Parshall flumes run
from $300 to over $2K in portable versions, depending upon size, and from
$500 to $5K for fixed installations, not counting secondary devices.
Palmer-Bowlus flumes without a level gage will cost between $300 and $3K
depending upon size. Construction materials also affect flume prices.
Simple current meters start at around $300 for basic Price or Ott types and
may run as high as $1,500. Electromagnetic current meters cosf from $2K
to $3K. Electromagnetic pipe meters start at around $2K for small (2 in.)
sizes and run to over $30K in the largest practicable sizes. Acoustic pipe
meters run from $2K to $20K depending upon size. These prices are for
complete systems including secondary devices.
Open flow nozzles, flow tubes, and Venturi tubes are comparably priced with
forging costs and machining accounting for the major portion. In small sizes
(3 in.) they run under $1K, and range up to $15-20K in large sizes (48 in.).
These prices do not include secondary devices.
The liquid level gage market is intensely competitive at the present time,
and prices are similar regardless of technique (e.g., electronic, bubble,
acoustic, dipper, etc.). They run from just under $1K for a basic device
D-72
-------
with visual read-out to over $2K with flow converters, recorders, transmit-
ters, etc., as accessories.
As a closing note, construction, installation, and (importantly) projected
maintenance and repair costs must be considered in addition to the equipment
acquisition costs given above to arrive at true cost of ownership, which is
the only real basis for comparison.
D.3.2.4 Review of Recent Field Experience
A brief review of flow measurement experiences, with emphasis on recent proj-
ects in the storm and combined sewer area, will be given to allow a better
appreciation of the application of some of the flow-measuring devices and
techniques in an actual field setting. The various experiences are presented
by primary device or technique as listed in Table D-14. It should be pointed
out that, although the following discussion focuses more on the negative ex-
periences, instances of good results were encountered with all types of flow
measurement.
Dilution methods were successfully used to calibrate primary devices in sev-
eral instances. In one installation, this technique was used to measure
flows in a sewer under surcharged conditions. A Palmer-Bowlus flume was em-
ployed for normal flow conditions. When the secondary device indicated that
the sewer line was nearly filled, a signal was given to begin chemical in-
jection. An automatic sampler was used to obtain samples for concentration
analysis at a site downstream from the injection equipment. Some other at-
tempts to use dilution methods were less successful, and it was abandoned by
several projects. Erroneous effects due to exposed sludge banks, insuffi-
cient turbulence to ensure mixing, and poor equipment operation (especially
samplers) were among difficulties cited.
Open channel acoustic devices had rather little use in the projects examined
because of their recent origin. Although successful installations exist,
their use has been abandoned at other locations. The primary difficulties
have to do with particles, notably air bubbles, in the flow causing improper
D-73
-------
readings, the complex velocity patterns requiring a number of transverse
sensors, and simple shakedown difficulties typical of early designs of many
complex electronic devices. Acoustic level gages were plagued by wind (in
an open application), foam, standing ripples on the water surface, and false
echoes from manhole structures or other confined areas. More recent indica-
tions are that such problems are being overcome, and satisfaction with these
devices appears to be increasing.
Parshall flumes were used in many projects, and they performed well when
dimensions were faithfully followed, standard approach conditions were pres-
ent, and (especially) when calibrated in place. Unfortunately, far too many
Parshall flume installations are nonstandard, reflecting difficulties in
making precise structures from poured concrete, the improper use of a light-
weight plastic flume liner as a form, etc.
Palmer-Bowlus type flumes were successfully used in a number of instances,
including portable versions intended for short-time application at any given
site. Other than their loss of accuracy as the pipe fills and surcharges, no
general negative comments about the devices themselves were encountered.
There were numerous complaints concerning secondary devices used in conjunc-
tion with Palmer-Bowlus flumes, however, especially bubblers. Instances of
their collecting debris and otherwise requiring frequent cleaning and mainte-
nance abound. In one project, their use was abandoned altogether, and they
were replaced with another type of level sensor.
Current meters were almost exclusively used to spot check flows and verify or
rate existing structures. There were flows where they could not be used at
all, however, because they immediately became fouled by rags, plastic sheets,
and other debris.
Electromagnetic devices were not encountered, except where they had already
been installed for other purposes. They appeared to work well, but the need
for periodic inspection and verification of any fixed flow-measuring device
was illustrated at one installation. As a part of a general flowmeter in-
spection in one district an apparently well performing electromagnetic
D-74
-------
flowmeter was found to be in error by over SO percent. The cause was a piece
of utility pole resting in the meter proper.
No projects examined used pressure flow acoustic meters, but their use in in-
dustrial plant applications has apparently been successful in many instances.
Open flow nozzles performed rather well where sites allowed their use. Fre-
quent inspection and cleaning were required at several installations, however,
to ensure proper readings.
«
Sharp-crested weirs were among the most commonly used (and misused) primary
devices encountered. Problems ranged from failure to properly account for
approach velocity, improper sizing, backwater elevations causing surcharging
and flooding, to almost continual cleaning being required in very trashy
flows.
Flow tubes and venturi tubes were seldom encountered, except where they had
existed for other purposes. They generally seemed to produce complete and
accurate records.
Trajectory coordinate estimates were uncommon, owing to the lack of suit-
able sites.
Slope-area methods (Manning in particular) were far and away the most fre-
quently encountered. They ranged from proper applications yielding reason-
able discharge estimates to totally unsuitable applications, as in one case
where the combined sewer discharge was found to considerably exceed the
measured precipitation event. Difficulties ranged from accurately measuring
slopes to estimating the proper friction coefficient (n) to use, in the best
instances, to unknowledgeable attempts and improper applications in the
worst. Apparently, far too many persons think that all that has to be done
is to measure stage and plug into a handy formula to obtain flow. It is long
past tine that that situation be corrected.
D-75
-------
D.3.3 Flow Measurement Field Procedures
For flow measurement in natural streams and channels, it is recommended that
USGS assistance be obtained. They will establish gaging stations (temporary
or permanent) at reasonable cost upon request and provide ratings to convert
stage to discharge. Often a culvert or some other control structure for which
a theoretical rating can be developed will be used. In some instances, weirs
or flumes will have to be used. It is prudent to spot check the ratings of
new gaging stations periodically. Be alert to changes in channel character-
istics that would affect the established rating, e.g., sedimentation, ero-
sion, deposition of large stones or boulders, etc.
Follow the manufacturer's recommendations for the installation, calibration,
and operation of the liquid level gages used to record stage. Where stilling
wells are employed, the connecting pipe should be checked for obstruction on
each visit, as should the float and cable operation. Note any instances that
could affect readings in the field log and the corrective action taken. It
is also prudent to verify chart time at each visit if record length exceeds
visit frequency (e.g., weekly flow charts but daily sampling). If a manual
sample is taken, a mark made on the flow chart can assist in subsequent data
analysis.
For manually gaging natural streams at the time of sampling, follow the
guidance given by the Bureau of Reclamation (12). Do not take a stream gag-
ing until all required samples for the site have been collected. Try to min-
imize or avoid walking in the stream until sampling is completed. Stirring
up the bottom may result in nonrepresentative samples. A complete flow
record is more desirable, however, and flow determinations made manually at
the time of sampling should be considered as a last choice.
For flow measurement in man-made channels and conduits, the use of an ap-
propriate primary device (refer to discussion in section D.3.2) is recom-
mended. These should be properly installed, following manufacturer's
recommendations in the case of commercial devices. The Bureau of Reclamation
D-76
-------
Manual (12) provides much helpful information. An independent verification
of the installation (i.e., by someone not on the installation team) will be
prudent in most instances. This is especially true where existing flow-
metering stations are to be used. Checklists for each type of primary device
should be prepared to facilitate field inspection. As an example, a check-
list for a contracted rectangular weir is presented in Table D-16.
Comments made above for secondary devices apply here as well. In closed con-
duits that are subject to occasional surcharging, try to install the level
gage so that it will indicate when this condition occurs. Although the
degree of surcharging cannot be indicated by most designs, knowledge of the
period of time over which the surcharge condition exists may be helpful in
subsequent data analysis. Such sites are best avoided wherever possible,
however.
On each visit, the flow-measuring equipment should be inspected to ensure
proper functioning. Visual verification of stage readings with a staff gage
is recommended at each visit, and results should be noted in the field log,
along with any anomalies discovered (e.g., a rag caught in the notch of a
weir, a stuck float, a clogged stilling well connection tube, etc.) and any
corrective actions taken. The possible buildup of sediment behind a weir
should be checked (the staff gage can be used) and any accumulation removed.
An occasional in-place calibration check is recommended to ensure that subtle
changes that could affect the record have not occurred.
One word of caution as regards the use of sewer maps is in order. Typically,
such maps (elevations especially) reflect intentions rather than installa-
tions. Even so-called as-built drawings may only indicate average invert
slopes from manhole to manhole and tell little about variations in true
slope. It is generally a prudent practice to verify pipe slopes entering
and leaving manholes where flow measurements are to be made.
Flow measurement at outfall sites can present some unique difficulties.
Where there is a. drop from the discharge pipe invert to the upper level of
the receiving stream, the site will probably be acceptable, and a temporary
f»_77
-------
TABLE D-16
CHECKLIST FOR CONTRACTED RECTANGULAR WEIR
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
What is the maximum measurable head?
Is upstream face of bulkhead smooth?
Is upstream face of bulkhead vertical?
(check for plumb with level)
Is upstream face of weir plate smooth, straight.
and flush with upstream face of bulkhead?
Is weir axis perpendicular to channel axis?
(check with line and carpenter's square)
Is entire crest level?
What is thickness of crest in flow direction?
(should be between 0.03 and 0.08 inch)
Is upstream corner of crest sharp and at
right angles to upstream face?
Are both side edges truly vertical and of same
thickness as crest?
Are downstream edges of notch chamfered?
(angle should be 45° or more to crest surface)
What is distance of crest from bottom of
approach channel?
(should be at least twice the depth above
the crest and never under one foot)
What is distance from sides of weir to sides
of approach channel?
(should be at least twice the depth above
the crest and never under one foot)
Does nappe touch only the upstream edges of
the crest and sides? Is nappe free?
Is there free fall?
Does zero head reading match with crest
elevation?
Is head reading taken upstream a distance of
at least 3 times the ma*"*™™ head on the crest?
Is the cross-sectional area of the approach
channel at least 8 times that of the nappe?
Does this condition extend upstream at least
15 times the depth above the crest?
If weir pool is smaller than defined above,
measure velocity of approach with current
meter.
If appreciable velocity of approach is
measured are head readings being corrected?
D
D
D
D
D
D
n
LJ
D
D
D
D
D
D-78
-------
weir box can be installed and used satisfactorily. Where the receiving
stream level is above the invert but below the crown, a pipe extension and
Paliner-Bowlus flume (or a Parshall flume in some instances) can possibly be
used. The real problem occurs where the outfall is completely submerged, and
the expense of a permanent device such as an electromagnetic flow meter
(otherwise, an excellent choice for such a site since it can measure flow in
either direction) cannot be tolerated. The best advice is to find another
site. If that cannot be done, the only recourse is to use a current meter
to obtain a velocity, adjust this to an average value, and multiply by the
pipe area to obtain flow. Where there is insufficient debris in the flow to
cause problems in operation, an oceanographic type recording current meter or
some other recording point velocity sensor can be used. For very trashy
flows, the only solution may be to measure velocities manually, cleaning up
the current meter between observations. This approach may be acceptable for
some intermittent discharges if a man can get to the site on time, but con-
tinuous records are impracticable.
D.4 Sampling Considerations, Equipment, and Procedures
The objective of any sampling effort is to remove, from a defined universe,
a small portion that is in some way representative of the whole. Ideally, a
representative sample will accurately reflect the physical and chemical char-
acteristics of the bulk source in every respect as they were during the sam-
pling period. In water quality, such representativeness is seldom if ever
achieved and, fortunately, seldom required. As used herein, a representative
sample is one that, when examined for a particular parameter, will yield a
value from which that bulk source characteristic can be determined. The
proper sampling methodology, i.e., that which will produce a representative
sample, is dependent upon the type of bulk source to be sampled, e.g., sur-
face water in natural channels (rivers, streams, lakes), municipal waste-
water, ground water, urban runoff, industrial wastewater, treatment lagoon,
and so on. Nonetheless, there are some more or less universal sampling con-
siderations, and they will now be addressed.
D-79
-------
D.4.1 Sample Types
The selection of the type of sample to be collected depends on a number of
factors, such as the rates of change of flow and the character of the water
or wastewater, the accuracy required, and the availability of funds for con-
ducting the sampling program. All samples collected, either manually or
with automatic equipment, are included in the following types, which terminol-
ogy has been recommended for standard usage by Shelley and Kirkpatrick (14).
Discrete Sample
A discrete sample (sometimes called a grab sample) is one that is collected
at a selected point in time and retained separately for analysis. A sequen-
tial discrete sample is a series of such samples, usually taken at constant
time intervals (e.g., one each hour over a 24-hour period), but sometimes at
constant discharge increments (e.g., one for each 100,000 gallons of flow)
when paced by a flow totalizer.
Simple Composite Sample
A simple composite sample is one that is made up of a series of aliquots
(smaller samples) of constant volume (Vc) collected at regular time intervals
(Tc) and combined in a single container. Such a sample could be denoted by
TcVc, meaning time interval between successive aliquots constant and volume
of each aliquot constant.
Flow Proportional Composite Sample
A flow proportional composite sample is one collected in relation to the flow
volume during the period of compositing, thus indicating the "average" con-
dition during the period. One of the two ways of accomplishing this is to
collect aliquots of equal volume (Vc), but at variable time intervals (Tv),
that are inversely proportional to the volume of the flow. That is, the time
interval between aliquots is reduced as the volume of flow increases. Alter-
natively, flow proportioning can be achieved by increasing the volume of each
D-80
-------
aliquot in proportion to the flow (Vv), but keeping the time interval between
aliquots constant (Tc).
Sequential Composite Sample
A sequential composite sample is composed of a series of short-period compos-
ites, each of which is held in an individual container. For example, each of
several samples collected during a 1-hour period may be composited for the
hour. The 24-hour sequential composite is made up from the individual 1-hour
composites.
Continuous Sample
A continuous sample is one collected by extracting a small, continuously
flowing stream from the bulk source and directing it into the sample con-
tainer. The sample flow rate may be constant (Qc), in which case the sample
is analogous to the simple composite, or it may be varied in proportion to
the bulk source flow rate (Qv), in which case the sample is analogous to the
flow proportional composite.
For initial characterization of wastewater flows, sequential discrete sam-
pling is generally desired. It is mandatory for accurate stormwater charac-
terization, since it allows characterization of the wastewater over a time
history and provides information about its variations with time. If the sam-
ples are sufficiently large, manual compositing can also be performed, based
on flow records or some other suitable weighting scheme, and a preferred com-
posite type determined. Some form of automatic compositing will usually be
desired for continued wastewater discharge characterization.
A brief look at the different types of composite samples is in order. Any
scheme for collecting a composite sample is, in effect, a method for mechani-
cally integrating to obtain average flow characteristics. The simple compos-
ite is the crudest attempt at such averaging and will be representative of
the waste flow during the period only if the flow properties are relatively
constant.
D-81
-------
For variable flows, some type of proportioning must be used. This is equiv-
alent to saying that the simple composite is a. very poor scheme for numerical
integration, and a higher Border method is desirable. There are two funda-
mental approaches to obtaining better numerical integration, given a fixed
number of steps. One is to increase the order of the integration scheme to
be used, as in going from the trapezoidal rule to Simpson's rule. The other
is to vary the step size in such a way as to lengthen the steps when slopes
are changing very slowly and shorten them when slopes change rapidly. Typ-
ical of the first approach are the constant time interval, variable volume
(TcVv) proportional composites. There are two straightforward ways of
accomplishing this. One is to let the aliquot volume be proportional to the
instantaneous flow rate, and the other is to make the aliquot volume pro-
portional to the quantity of flow that has passed since extraction of the
last aliquot. Typical of the second approach is the variable time interval,
constant volume (TvVc) proportional composite. Here a fixed volume aliquot
is taken each time an arbitrary quantity of flow has passed.
It is instructive to compare these four composite sample schemes. For the
purposes of this example, four flow functions and five concentration func-
tions are examined. The selections are completely arbitrary (except for
simplicity in exact integration) and, in practice, site specific data should
be used. For each flow/concentration combination, the exact average concen-
tration of the flow was computed (as though the entire flow stream were di-
verted into a large tank for the duration of the event and then its
concentration measured). The ratio of the composite sample concentration to
the actual concentration so computed is presented in matrix form in Fig-
ure D-4 (taken from Shelly and Kirkpatrick, 15). The four rows in each cell
represent the four types of composite samples discussed as indicated in the
legend. The best overall composite for the cases examined is the TcVv, with
the volume proportional to the instantaneous flow rate q. The TcVv where the
volume is proportional to the flow since the last sample, and the TvVc gave
very similar results with a slight edge to the former. However, the dif-
ferences are not large for any case. This brief look at compositing merely
scratches the surface. Flow records and a knowledge of the temporal
D-82
-------
>v CONC
N. k
,.
FLOW >.
b '
b •
k
l\ i t
1 N 1-t
If A simrt
b
1-t
0.90
0.90
0.90
0.90
1.35
0.90
0.86
0.87
0.68
0.95
0.92
0.92
0.90
1.01
0.90
0.90
b
IT
0.97
0.97
0.97
0.97
1.09
0.97
0.96
0.96
0.87
0.98
0.97
0.97
0.97
1.00
0.97
0.97
[^
TTt
COS— r
0.92
0.92
0.92
0.92
1.26
0.90
0.87
0.89
0.72
0.98
0.95
0.93 .
0.88
1.00
0.92
0.92
t-
-t
0.95
0.95
0.95
0.95
1.14
0.97
0.95
0.95
0.82
0.96
0.95
0.95
0.97
1.00
0.95
0.95
b
simrt
0.99
0.99
0.99
0.99
0.99
0.90
0.89
0.97
0.99
1.12
1.09
0.97
0.80
1.01
0.98
0.97
The rows within each flow/concentration cell refer to the following sample
types:
Row 1. TcVc - Simple composite
Row 2. TcVv - Volume proportional to flow rate (q)
Row 3. TcVv - Volume proportional to flow (Q) since last sample
Row 4. TvVc - Time varied to give constant AQ
FIGURE D-4
RATIO OF COMPOSITE SAMPLE CONCENTRATION TO
ACTUAL CONCENTRATION
D-83
-------
fluctuation of pollutants, as can be obtained from discrete samples, are
required in order to choose a "best" compositing scheme for a given
installation.
Continuous samples are also composite in nature but do not fit in the fore-
going discussion since the discrete step integration analogy is not appli-
cable. Had we included the Qv continuous sample in the foregoing example,
its ratio would have been unity for all combinations in Figure D-4. Other
considerations severely limit the instances where a continuous sample is
the composite of choice. For wastewater sampling, it is generally agreed
that the minimum line inside diameter is 0.6 cm (1/4 in.) and that the sample
flow velocity should be at least 0.76 m/s (2.5 fps). A simple calculation
shows that the minimum volume of a 24-hour continuous sample would be
2085 liters (551 gal), hardly a practicable size. For this reason, contin-
uous samples are useful only for very pristine flows (e.g., drinking water),
where the very low flow rates necessary to keep sample volumes reasonable may
still allow a representative sample to be obtained.
D.4.2 Automatic Sampling Equipment
In the following, a systems breakdown of automatic sampling equipment is
given in generic terms to allow the reader to better appreciate their func-
tional purposes and requirements. A survey of commercially available auto-
matic sampling equipment and costs is given, and a review of field experience
with these devices is provided.
0.4.2.1 Elements of an Automatic Sampler System
In a system breakdown by functional attributes, an automatic liquid sampler
may be divided into five basic elements or subsytems. Each of these will
be discussed in turn.
D-84
-------
D.4.2.1.1 Sample Intake Subs/stem
The operational function of the sample intake is to reliably allow the
gathering of a representative sample from the flow stream in question. Its
reliability is measured in terms of freedom from plugging or clogging, to
the degree that sampler operation is affected, and invulnerability to phys-
ical damage due to large objects in the flow. It is also desirable, from
the viewpoint of sewer operation, that the sample intake offer a minimum
obstruction to the flow in order to reduce the possibility of blockage of
the entire pipe by lodged debris, etc.
The sample intake of many commercially available automatic liquid samplers
is often only the end of a plastic suction tube, and the user is left to his
own ingenuity and devices if he desires to do anything other than simply
dangle the tube in the stream to be sampled. Some manufacturers provide a
weighted, perforated plastic cylinder that screens the hose inlet from the
unwanted material that might cause choking or blockage elsewhere within the
sampler. Typical hole sizes are around 1/3 cm (1/8 in.) in diameter and, if
there are sufficient holes to ensure free flow, results have been satisfac-
tory in some applications. Samplers that employ pneumatic ejection have
their own intake chambers that must be used in order for the equipment to
function properly.
0.4.2.1.2 Sample-Gathering Subsystem
Three basic sample-gathering methods or categories can be identified: mechan-
ical, forced flow, and suction lift. The sample lift requirements of the
particular site often play a determining role in the gathering method to be
employed.
Mechanical Methods. There are many examples of mechanical gathering methods
used in both commercially available and one-of-a-kind samplers. One of the
more common designs is the cup on a chain driven by a sprocket drive arrange-
ment. In another design, a cup is lowered within a guide pipe, via a small
automatic winch and cable. Other examples include a self-closing pipe-like
D-85
-------
device that extracts a vertical "core" from the flow stream, a specially con-
toured box assembly with end closures that extracts a short length (plug) of
the entire flow cross section, and a revolving or oscillating scoop that
traverses the entire flow depth.
Some of the latter units employ scoops that are characterized for use with a
particular primary flow measurement device, such as a weir or Parshall flume,
and extract an aliquot volume that is proportional to the flow rate. Another
design for mechanically gathering flow-proportional samples involves the use
of a sort of Dethridge wheel with a sample cup mounted on its periphery.
Since the wheel rotation is proportional to flow, the effect is that a fixed
volume aliquot is taken each time a certain discharge quantity has passed,
and total discharge can be estimated from the size of the resultant composite
sample.
The foregoing designs have primarily arisen from one of two basic considera-
tions: (1) site conditions that require very high lifts, or (2) the desire
to gather samples that are integrated across the flow depth. One of the
penalties that must be traded off in selecting a mechanical gathering unit is
the necessity for some obstruction to the flow, at least while the sample is
being taken. The tendency for exposed mechanisms to foul, together with the
added vulnerability of many moving parts, means that successful operation
will require periodic inspection, cleaning, and maintenance.
Forced Flow Methods. All forced flow gathering methods require some obstruc-
tion to the flow, but usually it is less than with mechanical gathering meth-
ods. It may be only a small inlet chamber with a check valve assembly of
some sort, or it may be an entire submersible pump. The main advantage of
submersible pumps is that their high discharge pressures allow sampling at
greater depths, thereby increasing the flexibility of the unit somewhat, in-
sofar as site depth is concerned. Pump malfunction and clogging, especially
in the pump sizes often used for samplers, is always a distinct possibility;
because of the pump's location in the flow stream itself, maintenance is much
more difficult and costly to perform than on above-ground or more easily
D-86
-------
accessible units. Submersible purnps also necessarily present an obstruction
to the flow and are thus in a vulnerable position as regards damage by
debris.
Pneumatic ejection is a forced flow gathering method used by a number of com-
mercial samplers. The gas source required by these units varies from bottled
refrigerant to motor-driven air compressors. The units that use bottled re-
frigerant must be of a fairly small scale to avoid an enormous appetite for
the gas and, hence, a relatively short operating life before the gas supply
is exhausted. Furthermore, concern has recently been expressed about the
quantities of freon that are being discharged into the atmosphere. The abil-
ity of such units to backflush or purge themselves is also limited. The ad-
vantages of few moving parts, inherent explosion-proof construction, and
high lift capabilities must be weighed against low or variable line veloc-
ities, low or variable sample intake velocities, and relatively small sample
capacities in some designs. Another disadvantage of many pneumatic ejection
units is that the sample chamber fills immediately upon discharge of the
previous sample. Thus, it may not be representative of flow conditions at
•che time of the next triggering and, if paced by a flow meter, correlation
of results may be quite difficult.
Suction Lift Methods. Suction lift units must be designed to operate in the
environment near the flow to be sampled or else their use is limited to a
little over 9m (30 ft) due to atmospheric pressure. Several samplers that
take their suction lift directly from an evacuated sample bottle are
available today. Vacuum leaks, the variability of sample size with lift, the
requirement for heavy glass sample bottles to withstand the vacuum, the dif-
ficulty of cleaning due to the requirement for a separate line for each sam-
ple bottle, the necessity of placing the sample bottles near the flow stream
(and hence in a vulnerable position), and the varying velocities as the sam-
ple is being withdrawn, are among the many disadvantages of this technique.
Other units are available that use a vacuum pump and some sort of metering
chamber to measure the quantity of sample being extracted. These units, in
some designs, offer the advantages of fairly high sample intake and transport
D-87
-------
velocities. The fluid itself never comes in contact with the pump, and the
pump output can easily be reversed to purge the sampling line and intake to
help prevent cross-contamination and clogging.
A variety of positive displacement pumps have been used in the design of suc-
tion lift samplers, including flexible impeller, progressive cavity rotary
screw, roller or vane, and peristaltic types. Generally these pumps are
self-priming (as opposed to many centrifugal pumps), but some designs should
not be operated dry because of internal wearing of rubbing parts. The desir-
ability of a low-cost pump that is relatively free from clogging has led many
designers to use peristaltic pumps. A number of types have been employed
including finger, nutating, and two- and three-roller designs using either
molded inserts or regular tubing. Most of these operate at such low flow
rates, however, that the representativeness of suspended solids is question-
able. Newer high-capacity peristaltic pumps are now available and are find-
ing application in larger automatic samplers. The ability of some of these
pumps to operate equally well in either direction affords the capability to
blow down lines and help remove blockages. Also, they offer no obstruction
to the flow since the transport tubing need not be interrupted by the pump,
and strings, rags, cigarette filters, and the like are passed with ease.
All in all, the suction-lift gathering method appears to offer more advan-
tages and flexibility than either of the others for many applications. The
limitation on sample lift can be overcome by designing the pumping portion of
the unit so that it can be separated from the rest of the sampler and thus
positioned within 6m (20 ft) or so of the flow to be sampled. For many
sites, however, even this will not be necessary.
D.4.2.1.3 Sample Transport Subsystem
The majority of the commercially available automatic samplers have fairly
small line sizes in the sample train. Such tubes, especially at 1/3 cm
(1/8 in.) inside diameter and smaller, are very vulnerable to plugging, clog-
ging due to the buildup of fats, etc. For many applications, a better mini-
mum line size would be 1 to 1.3 cm (3/8 to 1/2 in.) inside diameter.
D-88
-------
For flows that are high in suspended solids, it is imperative that adequate
sample flow rate be maintained throughout the sampling train in order to ef-
fectively transport them. In horizontal runs, the velocity must exceed the
scour velocity while, in vertical runs, the settling or fall velocity must be
exceeded several times to ensure adequate transport of solids in the flow.
Sharp bends and twists or kinks in the sampling lines should be avoided if
there is a possibility of trash or debris in the lines that could become
lodged and restrict or choke the flow. The same is true of some valve de-
signs. In summary, the sampling train must be sized so that the smallest
opening is large enough to give assurance that plugging or clogging is un-
likely in view of the material being sampled. However, it is not sufficient
to simply make all lines large, which also reduces friction losses, without
paying careful attention to the velocity of flow. For many applications,
minimum velocities of 0.6 to 1 m/s (2 to 3 fps) would appear warranted, and
even higher velocities are required for some applications.
D.4.2.1.4 Sample Storage Subsystem
The sample container itself should either be easy to clean or disposable.
Although some of today's better plastics are much lighter than glass and can
be autoclaved, they are not so easy to clean or inspect for cleanliness.
Also, the plastics will tend to scratch more easily than glass and, conse-
quently, cleaning a well-used container can become quite a chore.
The requirements for sample preservation are discussed elsewhere, but it
should be noted here that refrigeration is stated as the best single preser-
vation method and will, in all likelihood, be required unless the sampling
cycle is brief and samples are retrieved shortly after being taken. Light
can also affect samples, and either a dark storage area or opaque containers
would seem desirable. If opaque containers are used, however, they should
be disposable, since it would be difficult to inspect an opaque container
for cleanliness.
D-89
-------
D. 4.2.1.5 Controls and Power Subs/stem
The control aspects of some commercial automatic samplers have come under
particular criticism. It is no simple matter, to provide great flexibility
in operation of a unit while at the same time avoiding all complexities in
its control system. The problem is not only one of component selection but
of packaging as well. For instance, even though the possibility of immersion
may be extremely remote in a particular installation, the corrosive, highly
humid atmosphere, which will, in all likelihood, be present, makes sealing
of control elements and electronics desirable in most instances.
The controls determine the flexibility of operation of the sampler, e.g., its
ability to be paced by various types of flow-measuring devices. Built-in
timers should be repeatable, and time periods should not be affected by volt-
age variations. The ability to repeatedly gather the required aliquot volume
independent of flow depth or lift is very important if composite samples are
to be collected. Provisions for manual operation and testing are desirable,
as is a clearly laid out control panel. Some means of determining the time
when discrete samples were taken is necessary if synchronization with flow
records is contemplated. An event marker is desirable for a sampler that is
to be paced by an external flow recorder. Reliability of the control system
can dominate the total system reliability. At the same time, this element
will, in all likelihood, be the most difficult to repair and calibrate.
Furthermore, environmental effects will be the most pronounced in the control
system.
The required tasks can be best executed, in the light of the current elec-
tronics state-of-the-art, by a solid-state controller element. Such designs
offer higher inherent reliability and are becoming more and more common in
commercially available samplers. In addition, the unit should be of modular ,
construction for ease of modification, performance monitoring, fault loca-
tion, and replacement/repair. Such an approach also lends itself to encap-
sulation, which will minimize environmental effects. Solid-state switching
eliminates the possibility of burned or welded contacts, either of which will
cause complete sampler breakdown.
D-90
-------
Some automatic samplers available today require a 110V AC power supply, but
many tattery-operated units are also available. The latter are, of neces-
sity, smaller in size and sample transport velocity but still have a wide
range of application. Other portable units utilize compressed gas or spring
motors as the only required power source.
D.4,2.2 Considerations in Automatic Sampler Selection
Presently available automatic liquid samplers have a great variety of charac-
teristics with respect to size of sample collected, lift capability, type of
sample collected (discrete or composite), materials of construction, and
numerous other both good and poor features. A number of considerations in
selection of a sampler are:
• Rate of change of wastewater conditions
• Frequency of change of wastewater conditions
• Range of wastewater conditions
• Periodicity or randomness of change
• Availability of recorded flow data
• Need for determining instantaneous conditions, average conditions,
or both
• Volume of sample required
• Need for preservation of sample
• Estimated size of suspended matter
• Need for automatic controls for starting and stopping
• Need for nobility or for a permanent installation
• Operating head requirements
In addition to the foregoing attributes of automatic sampling equipment,
there are also certain desirable features that will enhance the utility and
value of the equipment. For example, the design should be such that mainte-
nance and troubleshooting are relatively simple tasks. Spare parts should
be readily available and reasonably priced. The equipment design should be
such that the unit has maximum inherent reliability. As a general rule, com-
plexity in design should be avoided even at the sacrifice of a certain degree
D-91
-------
of flexibility of operation. A reliable unit that gathers a reasonably rep-
resentative sample most of the time is much more desirable than an extremely
sophisticated, complex unit that gathers a very representative sample 10 «er-
cent of the time, the other 90 percent of the time being spent undergoing
some form of repair due to a malfunction associated with its complexity.
It is also desirable that the cost of the equipment be as low as practical
both in terms of acquisition as well as operational and maintenance costs.
For example, a piece of equipment that requires 100 man-hours to clean after
every 24 hours of operation is very undesirable. It is also desirable that
the unit be capable of unattended operation and remaining in a standby con-
dition for extended periods of time.
The sampler should be of sturdy construction with a minimum of parts exposed
to the sewage or to the highly humid, corrosive atmosphere associated di-
rectly with the sewer. It should not be subject to corrosion or the possi-
bility of sample contamination due to its materials of construction. The
sample containers should be capable of being easily remove.: ~d cleaned;
.-*•-• -•' * '" •
preferably they should be
For portable automatic wastewater samplers, the list of desirable features
is even longer. Harris and Keffer (16) give a number of features of an
"ideal" portable sampler, which are based upon sampler comparison studies
and over 90,000 hours of field experience.
D.4.2.3 Survey of Commercially Available Equipment
Some types of automatic liquid sampling equipment have been available commer-
cially for quite a while. In the last few years, however, there has been a
proliferation of commercial sampling equipment designed for various applica-
tions. New companies are being formed and existing companies are adding au-
tomatic sampling equipment to their product lines. In addition to their
standard product lines, most manufacturers of automatic sampling equipment
provide special adaptations of their equipment or custom designs to meet
unique requirements of certain customers. Some designs that began in this
way have become standard products, and this can be expected to continue.
D-92
-------
The products themselves are also rapidly changing. Not only are improvements
being made as field experience is gathered with new designs, but attention is
also being paid to certain areas that have heretofore been largely ignored.
For example, one company is introducing sampling probes that allow the gath-
ering of oil or various other liquids from the flow surface; solid-state
electronics are being used more and more in sampler control subsystems; new
types of batteries are offering extended life between charges and less weight;
and so on. Table D-17 lists the names and addresses of some 38 manufacturers
who are known to offer standard lines of automatic wastewater sampling
equipment.
An overall matrix, which summarizes the equipment characteristics to facil-
itate comparisons, is presented in Table D-18. There are several column
headings for each sampler model (or class of models). "Gathering Method"
identifies the actual method used (mechanical, forced flow, suction lift)
and type (peristaltic, vacuum, centrifugal pump, etc.). Depending upon the
gathering method employed, the sample flow rate may vary while a sample is
being taken, vary with parameters such »s lift, etc. Therefore, the "Flow
Rate" column typically lists the upper end of the range for a particular
piece of equipment, and values significantly lower may be encountered in a
field application. "Lift" indicates the maximum vertical distance that is
allowed between the sampler intake and the remainder of the unit (or at least
its pump, in the case of suction lift devices).
"Line Size" indicates the minimum line diameter of the sampling train.
"Sample Type" indicates which type or types of sample the unit (or series)
is capable of gathering. Not all types can necessarily be taken by all units
in a given model class; e.g., an optional controller may be required to
enable taking a TvVc type sample, etc. The "Installation" column is used to
indicate if the manufacturer considers the unit to be portable or if it is
primarily intended for a fixed installation. "Cost Range" indicates either
the approximate cost for a typical unit or the lowest price for a basic model
and a higher price reflecting the addition of options (solid-state control-
ler, battery, refrigerator, etc.) that might enhance the utility of the
D-93
-------
TABLE D-17
AUTOMATIC WASTEWATER SAMPLER MANUFACTURERS
o
«o
A I II Enterprises
1711 South III Avenue
Omaha, Nebraska 68144
Advanced Instrumentation, Inc.
Box 2216
Santa Cruz. California 9S063
T. A. Baldwin Company. Inc.
16760 Schoenborn Street
Sepulveda. California 91343
Bestel-Dean Limited
92 Morsley Road North.
Norsley
Manchester. England M28 SQW
BIF Sanitrol
P.O. Box 4
1-argo, Florida
33S46
Brallsford and Company. Inc.
Milton Road
Rye. New York I0580
Brand/wine Valley Sales Co.
20 tost Main Street
lloney Brook. Pennsylvania 19S44
Chandler Development Company
1031 East Duane Avenue
Sunnyville. California 94086
Chicago Pmp Division
I:MC Corporation
1800 PMC Drive
Itasco. Illinois 60143
Collins Products Co.
P.O. Box 382
Livingston. Texas 773S1
Environmental Marketing
Associates
3331 Northwest lilmwood Dr.
Corvallls. Oregon 97330
ETS Products
12161 Lackland Road
St. Louis. Missouri
63141
Fluid Kinetics. Inc.
3120 Production Drive
Fairfield. Ohio 4SOI4
llorlton Ecology Company
7435 North Oak Park Drive
Chicago, Illinois 60648
Ilydro-Numatic Sales Co.
65 Hudson Street
llackensack. New Jersey 07602
llydraguard Automatic Samplers
8SO Kecs Street
Lebanon, Oregon 973SS
Instrumental Inn Specialties Co.
Environmental Division
P.O. Box 5347
Lincoln, Nebraska 68SOS
.Kent Cambridge Instrument Co.
73 Spring Street
Ossining, New York I0562
Lakeside Equipment Corp.
1022 East Devon Avenue
Bartlelt. Illinois 60103
Manning Environmental Corp.
120 DiiBois Street
P.O. Box 1356
Santa Cruz. California 98061
Mark I and Specialty F.n|>. Ltd.
Box 145
Etobicoke. Ontario (Canada)
Nalco Chemical Company
180 N. Michigan Avenue
Chicago, Illinois 60601
Nappe Corporation
Crnton Falls Industrial Complex
Route 22
Croton Fulls. New York 10519
N-Con Systems Company
308 Main Street
New Roche lie. New York 1080)
Paul Nouscono Company
805 Illinois Avenue
.:ullin*ville, Illinois 62234
HI- Industries. Inc.
P.O. Box 746
Niagara Falls. New York 14302
Peri Pump Company, Ltd.
180 Clark Drive
•Connor.-, New York 14223
I'hipps ami Bird, Inc.
303 Sou til (>lh Street
Kiclmontl. Virginia 232(15
Protech, Inc.
Roberts Lone
Ma Iyer n, Pennsylvania I'JJSS
Quality Control Equipment Co.
P.O. Box 2706
DCS Moines. Iowa S03I5
Sigmamotor, Inc.
14 Elizabeth Street
Middleport. New York 14105
Sirco Controls Company
8815 Selkirk Street
Vancouver. B.C.
Sonford Products Corporation
400 East Broadway, Box B
St. Paul Park, Minnesota 55071
Testing Machines. Inc.
400 Bayview Avenue
Amityvillc. New York 11701
Tctradyne Corporation
1681 South Broadway
CarrolIton, Texas 75006
Trl-Aid Sciences, Inc.
161 Norris Drive
Rochester. New York 14610
Hilliams Instrument Co., Inc.
P.O. Box 4365. North Annex
San Fernando, California 0134.?
Universal F.nginecred Systems, Inc
7071 Commerce Circle
I'lcasanton, California 9-1566
-------
TABLE D-18
SAMPLER CHARACTERISTIC SUMMARY MATRIX
«o
Cn
Saajpler
•eslel-lieeei MI |l
•estel-Uean Crude
• IF 41
•rellsford OL-f ^ IF
• rellsfbril rVS
•rails ford OV-2
•vs rp-ioo
•vs PF: 400
1VS SK-IOO
•vs pre-400
CklcafO Plaap
Collins 42
Coll Hi 40
IMA 200
ETS FS-4
llorlion S7S70
llorliao S7S76
Itorliaej S7S7I
Hvdrefuard NT
Hvdrafuerd A
llvdra NUMIIC
ISCO 1192
ISfO I4U
ISCO ISM)
«eiu S.SA
lent ss»
(cm SSt
Ukeslde I i
Hu»in| S-4IHM
Mark lend I1OI
Nerkland 101 t 102
Markland I04T
Hid Kb ML 1000 « 2000
Nldlab Ml 3000
Nalco S-IUU
Nappe Porte-Posilcr
Nappr Series 46
Noasconn Snift
N-Coii Scout 1 1
N-M|*
S-fleiiblr i>|>clliT
S-flt.il.l. inpelU'i
S-prriMjItic
S-prri^l lit l»-
S-pcrisldltic
Hu.
•at>
(•l/ejin)
09"
talk
NA
III
S
10
•
7.600
J.bOO
•
111.000
>1.J8S
~&,OOO
laik
-20
100
100
100
•
•
5.700
Lion
HA
1.400
ISO
*OO
31. mm
NA
3,nno
•
•
•
1.1.90
1 . l>ao
29.400
II.4CIII
1 .1 . 200
B
ISO
ISO
l.ifl
<•)
6.1
6.1
4.!)
..'
J.7
<2
• S
«.t
•i.t
as
NA
NA
NA
•1
« S
9.1
9.1
».l
>»
•»
4.6
7.»
7.»
7.9
4.9
4.0
S.O
0
6.7
la.]
II. 1
la.. I
9.1
•1.1
'.6
l.«
4.6
9. 1
S.S
S S
l.lnr
Slie
(•")
6.4
19.1
2S.4
t.t
4.»
4.1
J.2
12.7
12.7
1.2
2S.4
2.4
2.4
9.S
b.4
o.a
08
o.a
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
.'5.4
I.'. 7
».S
6. 1
6.1
6.4
6.4
6.4
I-1. 7
6.4
9.S
4.*
...4
(..4
Saeple
t/pe
0. IcVc. I«Vc
0. TcVc. r>Vc
IcVc. TvVc
Tont i nwms
T.Vc. TvVc
TcVc. T.Vc
IcVc. TvVc
TrVc, IvVc
l>. TcVc. TvVc
IcVc. TvVc
UVc. TvVc
IcVc. TvVc
IcVc. TWc
TcVc. TvVc
Continuous
Crab
TcVc
Continuous. TcVc
TcVc
TcVc
TcVc. TvVc
0. TcVc. TvVc, S
TcVc. TvVc
IcVc. TvVc
Discrete
0. TcVc. TWi . S
It. TcVc. IvVc. S
IcVv
II. S
TcVc. TvVc
II. TcVc
II. TcVc. IvVc
TcVc. IvVc
IfVc. Ii-Vv
K-Vc. IvVt
IcVc
TcVi. IvVc
tonl iniiniis
KVc. IvVt
^et|iiriil i jl
InMal lat iun
Pnrlvltle
Portable
riicd
Portible
Portable
Piirlable
Portable
Portable
riicd
P or F
»i»ed
P or F
P or F
Portable
Portable
Portable
Portable
Portable
Portable
Portable
Portable
Portable
Portable
Portable
Portable
1 lied
Fiied
M.ed
Portable
Portable
Filed
fiied
Portable
Portable
Portable
Portable
1 ned
PoMahli-
Con jl>le
IMM.il.le
Purl.il.li-
Cos 1 Ranee
(S)
ml
link
-l.ooo
«6-»7J
S.'0.672
371
IS3-I.S2S
I.SOO-2.SIII
S.6SO
I.4SO-3.3SO
2.6OO-J.200
98S-2.47I
«15-2.32«
199- 4S6
l.09S-up
-410
-220
S9S
246- S4I
MS-661
1.100
I.09S- 1.491
645- 1.020
7 SO- I.I 10
1.240
2.JS4
2.3S4
700 op
1.290
I.09S-I.3SO
S94- 2.H9
1 ,O9t-2.644
I.SOO-2.SOII
.1.1100 -3. Still
Itok
22S-2H5
i . ion- 1 .mm
llnh
S7S-91S
1 .I.'S- 1 .-'OS
Po><:r
AC/la:
AC
AC
IK:
AC/O
i>:
AC/IK
AC/ID:
At:
«:/!•:
AI:
AI:
AC
Ai:/nr
AC
AC/OC
AC
DC
Air
Air ( AT
AC
AC/laJ
AC/nc
AI/IK:
Atr/n:
AI:
AT
AC
IK:
Air « l«
Air ( 1C
Air ( AC
AI
AI:
AC
AC/IN'
AI
AI
«vi«
AlVIM'
-------
TABLE D-18
SAMPLER CHARACTERISTIC SUMMARY MATRIX (Continued)
Saipler
N-CiHl Treblrr
It-Con Sentinel
Perl 704
Phlppi and lird
ProTech CO- II II
ProTeeh Ct-l2S
Pro Tech CU-I2SFP
ProTech (XC-200
ProTech CU-1OH
Pralech nil 4005
lf.fr. CVt
qCKC IVE II
qci;c *
• Ice tartan
SF.RCO HH- ]
SliUCO TL-I
Sigauwutor VA-I
Slfuaulor HAP-;
SlfMaotor IM-1-24
Sifauanlor sJA-S
Sifoaanlor mi'-S
SlfMKlar •» 5-24
Si no C/ST-VS
Slrco I/IC-VS
Slrco I/MP-VS
Sirca M-VS
Son ford IIC-4
Streiagard DA- .MSI
ml Fluid SlreM
nil Hk 11 (Hunt)
Tri-AIJ
Nilliun Itaclllwilit
rmlherlnl Method
H- scoop
user supplied
S- peristaltic
N-cup on chain
r-pneujiat ic
F-pneiaial tc
r'-pneuaat Ic
F-pneuwt ic
r'-siibaersible ptakp
F-suhaersible puap
S-vacuuai ptaip
S-vacuuat puap
N-cup on chain
S-vocuuB) puap
user supplied
S-pcristvll ic
S-perlstall ic
S-peristallic
S-perislallic
S-perlstaltlc
S- peristaltic
S-vacuua) ptvp
N-cup on cable
user supplied
S-vacuuat puap
M dipper
user supplied
F-pneuMt Ic
S-eviicuated jars
S-perislaltic
S-dlaphru|ai type
Ho.
Rate
(•l/«in)
NA
bl.(MM>
160
NA
1.000
I.OOfl
1.0(10
1.000
6.000
6.000
1.000
l.OOO
HA
Unk
42,000
60
60
60
10
SO
M
12. 000
NA
6.0OO
NA
NA
•
Varies
SIM)
60
Lift
(•)
0
MA
7.6
It.l
9.1
«.l
9.1
16.1
91
9.1
6.1
6.1
HI
1.7
NA
6.7
6.7
0.7
5.S
S.S
S.S
o.r
61
NA
6,7
II. S
NA
7.6
^J
7.S
l.ft
l.lm-
Si
S - Suction lift
• - Ik-pends on pressure and 11 ft
NA • Mil Applicable
link - llnkiitiwn MI liaM' 4if MrilinK
-------
device. Finally, the "Power" column is used to indicate whether line current
(AC), battery (DC), or other forms of power (e.g., air pressure) are required
for the unit to operate.
D.4.2.4 Review of Recent Field Experience
In order to assess the efficacy of both commercially available samplers and
custom engineered units in actual field usage, a survey of recent USEPA
projects, many of which were in the storm and combined sewer pollution con-
trol area, was conducted. None of these projects was undertaken solely to
compare or evaluate samplers, but all required determination of water qual-
ity. In the following paragraphs, difficulties encountered with various
elements of the liquid samplers are described.
The small diameter, low intake velocity probes found in several commercial
units were felt to be unable to gather as representative a sample of the
flow as could be obtained manually. There were many instances of inlet tube
openings being blocked by rags, paper, disposable diapers, and other such
debris. Although less a fault of the equipment than an installation prac-
tice, there were several instances of intake tubes being flushed over
emergency overflow weirs, up onto manhole steps, etc., during periods of
high flow and left high and dry and unable to gather any samples when the
flow subsided.
There were numerous instances of pre-evacuated bottle samplers losing their
vacuum in 24 to 48 hours, resulting in little or no data. Furthermore, per-
sonnel find these units with their 24 individual intake tubes virtually im-
possible to clean in the field. The low suction lifts on many commercial
units render some sites inacessible. In one project, three sites required
manual sampling because none of the samplers on hand could meet the 5- to
6-meter lifts required at these sites. There were several instances of sam-
ple quantity varying with sewage level as well as with the lift required at
the particular site. On at least two occasions, submersible pumps were dam-
aged or completely swept away by heavy debris in the flow.
D-97
-------
Within the sampling train itself, line freezing during winter operation was
a problem in several projects with instances of up to 60-percent data loss
reported. In one project, the intake line was too large, which allowed
solids to settle out in it until it ultimately became clogged. There were
numerous instances of smaller suction tubes becoming plugged with stringy
and large-sized material. A very frequent complaint, applied especially to
discrete samplers, was that they gathered inadequate sample volumes for the
laboratory analyses required.
On one project, although not directly the fault of the sampling equipment
itself, data were lost for 14 storms due to improper sterilization of non-
disposable sample bottles.
The control subsystems of commercial units probably came in for more criti-
cism than any other. Comments on automatic starters ranged from poor to
unreliable to absolutely inadequate. There were instances where dampness
deteriorated electrical contacts and solenoids causing failure of apparently
well-insulated parts. The complexity of some electrical systems made them
difficult to maintain and repair by field personnel. Inadequate fuses and
failures of microswitches, relays, and reed switches were commonly encoun-
tered. The minimum time between collection of samples for some commercial
units was too long to adequately characterize some rapidly changing flows.
Collected USEPA experience in one region reported by Harris and Keffer (16)
involved over 90,000 hours use of some 50 commercial automatic liquid sam-
plers of 15 makes and models. They found that the mean sampler failure
rate was approximately 16 percent with a range of 4 to 40 percent among
types. They also found that the ability of an experienced team to gather a
complete 24-hour composite sample is approximately 80 percent. When one
factors in the possibility of mistakes in installation, variations in per-
sonnel expertise, excessive changes in lift, surcharging, and winter opera-
tion, it is small wonder that projects on which more than 50 to 60 percent
of the desired data were successfully gathered using automatic samplers were,
until recently, in the minority.
0-98
-------
In fairness to present day equipment, it must be pointed out that some of the
above cited complaints stem from equipment designs of up to 10 years ago, and
many commercial manufacturers, properly benefitting from field experience,
have modified or otherwise improved their products' performance. The would-
be purchaser of commercial automatic samplers today, however, should keep in
mind the design deficiencies that led to the foregoing complaints when select-
ing a particular unit for his application.
D.4.3 Manual Versus Automatic Sampling
The decision whether to sample manually or use automatic samplers is far from
straightforward, and involves many considerations in addition to equipment
costs. Experience has indicated that operator training is necessary if manual
sampling is to produce reproducible results. Instances have been noted
wherein two different operators were asked to obtain a sample at a particular
site with no other guidance given. Analyses of samples taken at almost the
same instants in time have shown differences exceeding SO percent. Other
work conducted solely to compare manual sampling methods has indicated such
discrepancies in results that suspicion must be cast upon manual methods that
involve dipping of samples out of raw waste sources and has raised questions
regarding the suitability of such manual grab sampling as a yardstick against
which to measure other techniques.
The decision to use automatic sampling equipment does not represent the uni-
versal answer to water and wastewater characterization, however. For initial
characterization studies, proper manual sampling may represent the most eco-
nomical method of gathering the desired data. It is also prudent from time
to time to verify the results of an automatic sampler with manual samples.
In gene.al, manual sampling is indicated when infrequent samples are required
from a site, when biological or sediment samples or both are also required
from a site, when investigating special incidents (e.g., fish kills, hazard-
ous material spills), where sites simply will not allow the use of automatic
devices, for most bacteriological sampling, etc. Manual sampling will often
be the method of choice in conducting stream surveys, especially those of
D-99
-------
relatively short duration where only a single daily grab sample is required
from each site. For large rivers, lakes, and estuaries, manual sampling will
almost always be required.
Automatic samplers are indicated where frequent sampling is required at a
given site, where long-term compositing is desired, where simultaneous sam-
pling at many sites is necessary, etc. Automatic sampling will often be the
method of choice for storm-generated discharge studies, for longer period
outfall monitoring, for treatment plant efficiency studies, where 24-hour
comp.osite samples are required, and so on.
Typically, the wide spectrum of 208 agency monitoring activities will require
a capability for both manual and automatic sampling, and so the question is
not which capability to obtain but when to use each. The answei should be
determined in the design of each survey, using the above information as
guidance.
D.4.4 Sampling Field Procedures
D.4.4.1 Manual Sampling Procedures
The preferred method of gathering manual samples from a raw waste stream is
to use a pump to actually extract the fluid and tubing of appropriate size to
transport it to the sample container. Pump and tubing sizes should be such
that effective collection and transport of all suspended solids of interest
is ensured. Both small, flexible impeller centrifugal pumps and progressive
cavity screw pumps have been successfully used with good repeatability of
results. It should be noted, however, that the collection of flow propor-
tional or sequential composite samples can become quite tedious if performed
manually at the sampling site. Locate the intake at approximately the three-
quarters depth point (i.e., one-fourth of the way up from the bottom) and
point it upstream into the flow. Adjust the pump speed until intake velocity
approximately equals the mean flow velocity (obtained from a flow-measuring
device or current meter) and, after about 60 seconds, direct the stream into
the sample container. Avoid using an intake screen unless absolutely
necessary.
D-100
-------
When manually sampling natural streams, use a depth-integrating sampler at
the center of the stream if the flow is laterally homogeneous. Check the
site for this by occasionally taking samples from the quarter points and
comparing results. If significant differences are found, either choose
another site or take a number (5 to 20 depending upon stream width) of depth
integrated samples along a transect perpendicular to the flow. Based on the
results, choose the minimum number of transverse stations that will yield
acceptable results.
Depth integrating samplers for use in more swiftly running streams are rela-
tively heavy, and so some type of hoist or winch is normally used to facil-
itate handling. These can be mounted on boats for river and estuary cruises,
on trucks or trollies for bridge sampling, etc. Contact the nearest USGS
field office for more information on availability and use of different depth
integrating samplers.
Samples may be manually gathered at a given depth in the water column by
using a Juday bottle or one of its modifications (e.g., Kemmerer, Van Dorn).
This type is essentially a cylinder with stoppers that leave the ends open
while the sampler is being lowered to allow free passage of water through
the cylinder. When the desired depth is reached (as determined by markings
on the line, for instance) a messenger is sent down the line and causes the
stoppers to close the cylinder, which is then raised and the sample trans-
ferred to its container. These devices can be used to approximate depth in-
tegration through the water column, to investigate stratification in lakes,
or wherever a sample from a particular depth is desired. When using such
devices from bridges, take precautions so that the messenger, when dropped
from the height of the bridge, does not batter and ruin the triggers that
release the stoppers. One simple way to avoid this is to support the mes-
senger a few feet above the sampler with a string and release it when the
desired depth is reached.
If vertical concentration gradients are not severe, a single grab sample
will suffice. It is recommended that a container smaller in volume than the
desired total sample volume be used, and that the required sample volume be
D-101
-------
obtained by repeated dippings at one minute intervals. Rinse the container
two or three times in the water to be sampled prior to taking the first ali-
quot. Comparison of the results between depth integrated and simple grab
samples will indicate when the latter technique will suffice.
For reproducibility of manual sampling results, operator training is abso-
lutely essential; 208 agencies can ill afford to entrust this task to well-
intentioned but untrained staff or volunteers. Also, it is time that we
forget about using a beer can nailed to a stick as a sample gathering device.
All in all, the manual pumping sampler described earlier in this section will
produce the most reproducible results, and its use is recommended whenever
feasible. One subject that should also be touched on briefly is manual com-
positing according to flow records. Given a series of discrete samples of
equal volume taken at regular time intervals and a flow record, the question
is what size aliquot should be taken from each discrete sample container to
form the flow proportional composite sample? Recall from Section D.4.1 that
this can be done in one of two ways: either extract an aliquot volume that
is proportional to the instantaneous flow rate at the time the discrete sam-
ple was taken, or extract an aliquot volume that is proportional to the total
discharge that has occurred since the last discrete sample was taken. The
formula used for this can be written as:
a. = f, Vc,lt.
where: a. = aliquot volume to be extracted from the i-th discrete
sample, i.e., the one taken at time t.
i ° index indicating the order in which the discrete samples
were taken,
f. * flow variable; either the flow rate when the i-th discrete
sample was taken (q.) or the total discharge that has
occurred since the (i-l)-th sample was taken (AQi'Qi-Qi.j)
V = composite sample volume desired
n = number of discrete samples taken
The desired composite sample volume is determined based on the requirements
for the analyses to be conducted. The subtle problem is that one does not
D-102
-------
have complete freedom in selecting V because of the fixed discrete sample
volume (V,), and the entire sequential discrete series may be wasted if this
is not recognized, because there might not be enough sample in one bottle to
fulfill its aliquot requirements. This is best illustrated by an example
(see Table D-19). Note that if steps 3 and 4 had not been carried out, when
the operator came to bottle number 5 he would not have been able to continue,
since he would be 250 mi. short. This has happened. Also, it is incorrect
to use leftover liquid from the adjacent discrete samples to make up the
deficit (which has also occurred).
In actuality, one can compute the maximum composite sample volume that can
be formed from a series of discrete samples. The formula is
(V ) = V, Ef./(f.)
k c'max d i v i/max
If this quantity is greater than the amount desired, the formula given earlier
for determining aliquot volume can be used. If not, the aliquot size should
be computed from
ai = fi V»x
This will be illustrated by a second example, shown in Table D-20. Since the
available composite sample is nearly half a liter less than was desired, a
new decision on how to allocate the available volume must be made.
Example III (Table D-21) is included to indicate how to manually prepare a
time-constant, volume-proportional-to-discharge-since-last-sample-was-taken
composite when a record of flow rate rather than discharge is available. The
results of Examples II and III agree because the same flow function (q=5,000
sin TTt/8) was used in each case and the trapezoidal integration scheme worked
well.
The-details for manually preparing a time-constant, volume-proportional-to-
instantaneous-flow-rate composite sample using the flow rate record given
in Example III will not be presented (a.^191, 354, 462, 500, 462, 354, 191,
0; la.=2,514 mi), but it is of interest to contrast the measured concentration
of a constituent of interest obtained by this method as opposed to the method
of Example II. For this purpose, assume that the constituent behavior is a
D-105
-------
TABLE D-19
MANUAL COMPOSITE SAMPLE EXAMPLE I
Example:
Given:
Steps :
1.
2.
3.
4.
Manually preparing a time-constant, volume-proportional-to-
instantaneous- flow-rate composite sample.
A 500 ml discrete sample was taken at the end of each hour over
an 8-hour shift. A 2-liter composite is desired. A recording
of flow rate is available.
Sample No. (i) q. a.
1 300 47
2 600 94
3 1,200 188
4 2,400 375
5 4,800 750
6 2,000 312
7 1,000 156
8 500 78_
Eq.^ » 12,800 2,000
Enter q. from record and sum.
Calculate ai«q.jVc/Zqi«2000qi/12,800.
Check to see if maximum a. exceeds discrete
Compute new aliquot volume *> a.xSOO/750.
a.xSOO/750
31
63
125
250
500
208
104
52_
1,333
sample volume.
D-104
-------
TABLE D-20
MANUAL COMPOSITE SAMPLE EXAMPLE II
Example: Manually preparing
a time-constant, volume-proportional-to-
discharge-since- last-sample-was-taken composite.
Given: A 500-mil discrete
an 8-hour shift.
Steps :
of totalized flow
samp le was
A 3-liter
taken at the
composite is
end of each hour over
desired. A recording
is available.
Sample No. (i) Q. AQ. - a.
0
1
2
3
4
5
6
7
8
0
969
3,729
7,860
12,732
17,605
21,736
24,496
25,465
SAQ,
_
969
2,760
4,130
4,873
4,873
4,130
2,760
969
= 25,464
1. Enter 0. from record and calculate AQ. » Q
i i
2. Calculate (Vc)max - (500) (25, 464) /4, 873 =
_
99
284
424
500
500
424
284
9£
2,614
!i " Qi-r
2,614 mi.
3. Since (V ) is less than desired, calculate aliquot size from
a.. » 500 AO^/4,873.
D-105
-------
TABLE D-21
MANUAL COMPOSITE SAMPLE EXAMPLE III
Example: Manually preparing a time-constant, volume-proportional-to-
discharge- since- last- sample-was- taken
Given: A 500-mJl discrete
an 8-hour shift.
sample was taken at
A 3- liter composite
composite.
the end of each hour over
is desired. A recording
of flow rate is available.
Steps :
Sample No. (i)
0
1
2
3
4
5
6
7
8
-------
simple linear decay (i.e., conc.=9-t). The true concentration in the flow
rate proportional sample would be 5.0 (assuming the discrete samples from
which the composite was formed were 100 percent representative). The corre-
sponding true concentration of the discharge proportional composite
(Example II) would be 4.5, a difference of around 10 percent due solely to
the method of compositing.
The possible importance of sediment oxygen demand (SOD) measurements to
208 agency plans is well illustrated by Butts (17) who noted, as a result of
an extensive SOD study, that "... it is doubtful that the aquatic ecology of
the (Illinois) waterway can be measurably enhanced solely by achieving cur-
rent water quality standards." The subject of SOD measurement remains some-
what controversial, but it is recommended that determinations be made in situ
rather than in the laboratory. Ascertaining the relationship between SOD
rates and DO content of the overlying waters is better accomplished by perform-
ing in situ measurements. This can be done, for example, by setting a bell-
shaped shallow cover over the spot on the bottom where the measurement is to
be made, circulating the water within this "sampler" with a small pump, and
measuring the change in DO with time.
The design of an in-situ SOD measuring device developed by the Illinois State
Water Survey is described by Butts (17), who also reports favorably on its
use. The cover was made from a 14-inch-diameter by 24-inch-long steel pipe
split longitudinally in half. End plates were welded on, and angle iron was
welded around the lower edge to act as cutting edges and seating flanges.
Fittings for raising and lowering the device, two hose attachments to allow
connection of a pump for water circulation, and a split collar to hold the
DO/temperature probe were also welded in place. The "sampler" covered a flat
bottom area of about 0.2 square meter (336 sq in.), and the total volume of
water within the system was around 31 liters. The device is handled with a
USGS bridge winch adapted for use on a boat.
D-107
-------
D.4.4.2 Automatic Sampling Procedures
When using automatic samplers, the greatest problem comes in mounting the
intake. Screened intakes should be used in waters containing large solids,
trash, or debris to prevent clogging. Screen openings should be slightly
smaller than the smallest opening in the sampling train. More and more com-
mercial devices are now provided with intake screens by their manufacturers.
When using these, the end of the intake hose should be approximately at the
center of the screen. If intake screens are not provided with the samples,
they can be fabricated quite simply by drilling a large number of appropri-
ately sized holes in a piece of plastic pipe, cementing on end covers, and
drilling out one end to accept the sample tube and fastening it with a hose
clamp and fitting. Clear plastic is recommended to facilitate inspection.
A typical size for an intake screen to accommodate a 3/8 inch ID tube is ap-
proximately 1.5 to 2 inches in diameter by 6 to 10 inches long. Hole diam-
eters could be 1/4 inch if the rest of the sampling train is larger.
The flexible plastic intake tubing commonly used in most commercial automatic
samplers will require some protection in many installations, or wear from
particles in the flow and damage from debris will necessitate frequent re-
placement. Flexible electrical conduit and reinforced garden hose have been
successfully used in this regard. Even with such protection, it is recom-
mended that sample intake lines be trenched in where they run over earthen
surfaces.
One of the most challenging sample intake mounting problems is in a natural,
wet weather stream. If the intake is allowed to rest on the bottom where it
could obtain samples at very low flows and, hence, more readily determine
first flush effects, there is a possibility that flow fields around the in-
take may induce scour and cause artificially high solids readings. Mounting
the intake well above the bottom obviates this problem but prevents acquiring
samples of very low flow. The best compromise seems to be to mount the in-
take horizontally, at right angles to the flow, in the middle of the stream
and with its lowest surface around 2 inches above the bottom (higher if sig-
nificant bedload depths are anticipated). The stream bottom at this point
D-108
-------
should be reasonably flat and free of stones or other flow-altering obstruc-
tions upstream of the intake. For cobble-strewn bottoms, follow the above
procedure but measure from a sheet of plywood resting on the stones.
To anchor the sample intake to the bottom, use screw augers or metal rods
driven well into the soil. Simple hose clamps can be used to affix the intake
screen to these supports.
For continuously flowing natural streams, similar considerations pertain.
The main difference will be in the vertical location of the intake. In the
absence of other factors, mount the intake near the low flow mid-depth. If
stream depth allows, the intake should be mounted with its center line ver-
tical, and suction taken from the bottom. In this -configuration, a single
mounting rod can be used. It should be located to one side of the intake
(never in front of it).
The foregoing has been written with smaller streams, typical of those that
would be encountered in an urban runoff study, in mind. As indicated earlier
in this section, it is not expected that automatic samplers will find wide
use in river monitoring.
In man-made channels and conduits, there is no longer a concern for bottom
scour. For those carrying intermittent flows, the intake screen can be
allowed to rest on the bottom unless significant bedload depths are antici-
pated. Where large debris is likely to be encountered, a spring-loaded
intake screen mounting should be considered to help prevent destruction. It
is a fairly common practice to simply let the intake screen trail downstream
by its tubing. In very low or no-flow periods it will rest on the invert
and, during higher flows, hydrodynamic forces will tend to lift it up. The
chief objection to this practice is that probes facing downstream do not
gather representative solids due to momentum effects. Data on the degree of
under-representation caused by this practice are virtually nonexistent, how-
ever. Use this practice as a last resort.
D-109
-------
Where the flow is continuous (but variable), position the intake screen near
the low flow mid-depth. As opposed to natural streams, however, in many man-
made conduits it will be more convenient to dangle the intake from above with
the suction tube pointing down. Although the vertically up orientation is
preferable, this practice is also acceptable. The chief disadvantage of
"dangling" approaches to intake mounting is that you never really know where
the intake is. Be certain that there is no possibility of full flow posi-
tioning the intake where it could be left "high and dry" as the flow recedes.
Manhole benches, steps, weirs, and the like have taken their toll in careless
intake installations.
For the (rare) case where relatively steady flow is anticipated in either
natural or man-made channels, position the intake at about the three-
quarter depth point. If two automatic sampling devices are used for redun-
dancy at a critical site, position one intake at the eight-tenths depth point
and one at the four-tenths depth point. Shelley (18) discusses the rationale
for sample intake location in some detail and presents designs for maintain-
ing intakes at a constant percentage of depth in variable flows, noninvasive
intakes, etc.
All of the foregoing has been written primarily with suction lift intakes
in mind, but similar considerations apply if forced-flow devices are used.
For samplers employing mechanical gathering methods, follow the manu-
facturer's directions.
Mounting the main body of the automatic sampler is rather straightforward;
follow the manufacturer's directions. Keep the lift as short as possible
commensurate with the likelihood of submergence. If excess sample tubing
exists, cut it off. Do not simply coil it out of the way, thinking that the
extra length might be useful at the next installation.
D-110
-------
After setting up the controls and power subsystem according to the operator's
manual for the particular sampler being used, manually cycle it a few times
and measure the quantity of sample actually being taken. This is especially
important where fixed aliquot volume composite samples are to be collected.
Verify sample volume gathered on each site visit. Partial plugging, intake
blockage, or other occurrences that might not be immediately obvious can
affect the sample quantity in most designs. Also, use a stopwatch to record
the time that it takes to gather the sample and verify this on subsequent
visits. For battery-operated units, frequent voltage checks are in order
until service life can be established for the installation. Manufacturers
are not noted to be conservative in estimating battery life, and it will be
affected by a number of factors such as sample lift, temperature, etc.
Always inspect the sample intake at each visit.
For operation in very cold weather, a heated enclosure for the sampler body
will be required. Sample lines should be wrapped with heater tape and
insulated — large plastic trash bags work well for this. Check for possible
ice buildup at each visit. Should frozen (or partially frozen) samples be
encountered, do not discard them, but immediately enter the facts in the
field log and also report the condition to the analytical laboratory when
the samples are delivered.
Maintenance and troubleshooting of automatic samplers are so design-dependent
that little general guidance can be given other than to follow manufacturer's
instructions and recognize the importance of these activities in contributing
to project success. However, one word of caution pertaining to suction lift
samplers using peristaltic pumps must be made. Some of these pump designs
require that the tubing be lubricated. This must be done or tube life will
be considerably shortened; failures after less than 2 hours of operation
have been reported for some designs when inadequate lubrication was applied.
With care and consideration, most automatic samplers can be made to work
reasonably well; with carelessness and disregard, almost none will.
D-lll
-------
D.4.5 Sample Quantity, Preservation, and Handling
Since the required sample volume is dependent upon the type and number of
parameters to be analyzed for and the instrumentation and methods to be
employed in the analysis at the laboratory, the laboratory analyst is the
best person to specify the quantity needed. A preliminary estimate of
sample volume can be obtained as follows. Determine the parameters to be
analyzed for and, from the Parameter Handbook, obtain the sample volume
required.for each analysis. Sum these to obtain the minimum volume, and
increase this amount as necessary to allow for spillage, mistakes, sample
splitting, and for analytical laboratory quality control purposes. In the
absence of better information, doubling the minimum volume should be
adequate.
Having collected a representative sample of the fluid mixture in question,
there remains the problem of sample preservation and analysis. It is a
practical impossibility either to perform instant analyses of the sample on
the spot or to completely and unequivocally preserve it for subsequent ex-
amination. Preservation methods are intended to retard biological action,
retard hydrolysis of chemical compounds and complexes, and reduce volatility
of constituents. They are generally limited to pH control, chemical addi-
tion, refrigeration, and freezing. The USEPA (19) has compiled a list of
recommendations for preservation of samples according to the measurement
analysis to be performed. In order to provide an overview for some common
parameters, this list has been reproduced here as Table D-22. For other
parameters and program design, reference should be made to the Parameter
Handbook.
Proper sample handling is also essential to obtaining successful results from
any monitoring program. A few general guidelines are given below.
1. Each sample container must have a designation, normally a number,
that uniquely distinguishes it from all other samples in the survey.
D-112
-------
TABLE D-22
RECOMMENDATIONS FOR PRESERVATION OF SAMPLES ACCORDING TO MEASUREMENT
(1)
o
I
N*ai«raM»l
Acldllp
Altallaitp
Artoolc
•u*
•r~l4*
am
Chlorite
Color too too,
Color
CpO.140*
OUaolvod pfoo
ProW
•|p4l*r
.,...»
«•!*.»
1.414.
MAS
MBt.lt
•I...I.M
lmp,»Ji4
Unl
H.r,.r,
Mllol»4
.,!„,..
— ...
•Jol4o*l
Mitral a
Nltralt
tol
(ol»
IM
IM
IM
IM*
IM
St
to
M
U
SM
MO
MO
Mt
IM
IM
JS*
no
IM
IM
4M
SM
IM
SO
Cooioloor
P.C(1*
P.G
r.c
r.c
r.c
P.G
r.c
r.c
P.C
r.c
cool.
C Miy
r.c
r.c
fjii
P.C
r.c
P.I;
r.c
r.c
P.C
r.c
Pratorva.il va
Cool. 4*C
Cool. 4'C
HMO. to pH <2
*
Cool. 4 C
Cool, 4 C
H SO to pN «;
' *
MOM loo.
fool. 4'C
Cool. 4*C
Cool. 4*C
NotM to pH 12
0.1 o. ,U.
Fli M *!!•
Cool. 4V
Cool. 4*r
Cool. 4*C
tool. 4'C
Fllt«f a* *Ur
HMO, to ptl «2
rlltcr oo »lt*
«, ,. „, .,
tlltcr
IMO, to prl «2
tool. 4*t:
tool. 4*C
II^SO^ to pit <2
Cool. 4*C
11^ lo otl < 2
tool. 4*C
««,«.««
24 Hr»
24 Mra
ft •*>»
ft Mn(J*
24 m>
7 O.r.
7 Dar,
24 Mr*
24 llr •
24 Mr*
NO IfeldlMg
Mo HuUI«|
7 Oap*
7 Dan
24 Mr*
24 Hri
ft Nu«
ft No*
ft Ho*
U Oop* (Cl»i«|
11 Dor* (Ibrd
24 Ifcr*'*'
i. «,.'"
14 Hr»l4t
» II,,'4'
»..,„,«.,
NT*
Oil aad Craoso
iHfajilc Itarbo*
•"
PWnollc*
Phoiphurow*
^ ^_
Hrdrolptafala
Total
Ol**ol*od
*••**"*
f.U.r^).
*,.m,.,^i.
Total
VolallU
SottUatilr Matter
Silica
Spc^fU
ToodiK t aoco
suir.,.
Sull.J.
Sti 1 1 1 1 *
•-»"
Thr«»lM>ld lldor
i..rt»4lir
Vol
Sfl
IOM
2S
2S
SOD
SO
SO
IM
IM
IM
100
IOOU
SO
IM
SO
SO
SO
IOUI
,-00
ion
24 NT,
.*4 Hrt
No Holding
21 Mt>
7 Oar.
I. If *MpU* caoMl »o rotwrood to HM laooratorp IN la** than
ft hour* ao4 lw»Mla| UM aita*4* ikl* Halt, th« ftn*l rapontd
tiiw of 40 Of/l, •ftpoclallp If • lo*gar hol4ln| i10*
S. If lk« iaool« I* *ta»lllie
to 2S*C for raadiNf, or lo
r«*ult, report *J at 2S't:.
It ka, b«*o ihow* th«t tooola* proparlp praiarvW aMp fc*
kfld for ••tCiwIoJ parludi be rood tk« i
llM.
-------
2. When frequent sampling over a long time period is involved, con-
sideration should be given to incorporating a temporal indication
as a part of the sample identification number; e.g., the number
of the week in a year, the last two digits of the year, etc. The
temptation to code too much information about the sample into its
identification number must be resisted, however, or else the risk
of mixups due to unauthorized abbreviations becomes too great.
3. Consideration should be given to the use of preprinted, sticky-back
labels in many instances. Be certain, however, that they are
waterproof. Rubberband and tie-on tags have also been used success-
fully.
4. The use of color-coded labels has been successful where sample
splitting or different preservation techniques are employed. In
the latter case, for example, a green label could indicate that
nitric acid had been added and that, therefore, an analyst could
obtain aliquots from this sample for metal analyses, etc.
5. Where possible, the type of sample, date, and any preservatives added
should be written on the sample label prior to collecting the sample
in the field. The time of day should be added when the sample is
collected. Additional information should be noted in the field
notebook and on supplemental forms where used.
6. The foregoing should be observed in addition to any chain-of-custody
procedures that are involved. See (20) for USEPA recommendations
for a chain-of-custody program.
The proper cleaning of all equipment used in the sampling of wastewater is
essential to ensuring valid results from laboratory analyses. Cleaning
protocols should be developed for all sampling equipment early in the design
of the monitoring program. Here, also, the laboratory analyst should be
consulted, both to ensure that the procedures and techniques are adequate as
D-114
-------
well as to avoid including practices that are not warranted in view of the
analyses to be performed. The possibility of the container affecting the
sample analyses should be checked periodically. Distilled or demineralized
water should be placed in a typical container for a period of time similar
to that of a normal sample. Then the particular constituent of interest
should be measured in the water from this blank. Also, checks for sample
adsorption on the container should be made by placing a known amount of a
particular constituent in a typical container. After a specified holding
time, analyses should be made to determine if any of the material was ad-
sorbed into the container or changed in any other manner. These checks
should be done after sample bottles have been used for a series of samples.
In this way the cleaning techniques used can be tested for thoroughness.
The use of blanks and spikes just mentioned brings up the subject of qual-
ity control in general. Although outside the scope of this Appendix, each
208 agency must have a viable quality assurance program. The USEPA (21, 22)
has published minimal requirements for a water quality assurance program and
a handbook for analytical quality control in the laboratory. The recommenda-
tions in these two references should be followed by all 208 agencies.
D.4.6 Sampling Accumulated Roadway Material
Accumulated roadway material may represent a significant source of pollution
during storm-generated discharges in urban areas. In order to quantify this
source, provide inputs for models, determine if better urban housekeeping
practices would produce commensurate water quality improvements, etc., sam-
pling of accumulated roadway material will be required. The following dis-
cussion is abstracted from Wullschleger et al. (7).
Samples of materials deposited on roadways are collected using a combination
of sweeping, vacuuming, and water flushing techniques. Each sample will con-
sist of three fractions: litter, dust and dirt, and water flush. The par-
ticulate materials collected by sweeping and vacuuming are separated on the
basis of particle size into a litter fraction and dust and dirt fraction.
The litter fraction consists of that portion of the particulates retained by
D-115
-------
a U.S.A. No. 6 sieve (i.e., greater than 3.35 mm in diameter). This fraction
is usually composed of stones, gravels, wood fragments, and other larger
sized materials in addition to bottles, cans, paper production, etc., which
are normally thought of as litter. The dust and dirt fraction will contain
particulates smaller than 3.35 mm in diameter. The water flush fraction con-
tains those components of the dust and dirt fraction which were not picked up
at high efficiencies by the sweeping and vacuuming techniques. The flush
plus the dust and dirt constitute a total dust and dirt fraction which is the
major source of water pollutants found in runoff from urban roadways.
If a physical and chemical description of the street surface contaminants is
needed, the sample should be collected by hand sweeping, followed by flush-
ing. All of the dry solid material collected from the test area should be
placed in clean containers and shipped back to the laboratory. There it
should be air dried thoroughly and sealed for storage until analyzed. All of
the flushed material should be measured for volume, but only a portion of it
need be retained for analysis. The liquid sample should be stored in clean
containers (glass, if pesticide analyses are to be made) and cooled to <4°C
if possible. The analyses of the liquid fraction should be made as soon as
possible after collection. To reduce the number of chemical analyses re-
quired, the dry and liquid samples can be combined on an equal sample area
basis before the analyses are performed.
If only physical loading information (such as kg (Ib) of solids per curb km
(mile)) is needed, hand sweeping is probably sufficient. In most cases, the
additional quantity of material that can be obtained by subsequent vacuuming
and/or flushing is insignificant. If information regarding particle size
distribution is required, then the sample should be collected using a combi-
nation of hand sweeping and dry vacuuming. The vacuum is more efficient in
removing the fine particles which are needed for size distribution analyses.
If size distribution of the solids in the wet phase is needed, then flushing
will also be required.
D-116
-------
The basic procedures for the collection of samples are:
Hand sweeping - Hand sweeping for dry solids collection should utilize a
standard stiff-bristled push broom. The sweeping pattern should be from the
center of street or from one edge of the test area towards the gutter or op-
posite side of the test area. After concentrating the material along this
edge, the sample should be collected, using a whisk broom and dustpan.
Vacuuming - Vacuuming the test area usually removes more smaller-sized par-
ticles than is possible by only using sweeping techniques. The vacuuming
pattern should approximate the pattern described for hand sweeping. An in-
dustrial wet/dry "shop" vacuum cleaner with a 5-7.6 cm (2 in. to 3 in.) di-
ameter hose is recommended. Other types of units, ranging from small
household vacuums to large motorized vacuum sweepers, may also be satisfac-
tory, depending on the size of the test area.
Flushing - The test area can be flushed with water after hand sweeping to re-
move soluble films and other nonswe'epable material. The materials removed
with this method more closely resemble those which are removed by a runoff
event. The test area is first slightly wetted to soften and facilitate re-
moval of soluble materials. It is then flushed with a stream of water from a
garden hose and spray nozzle connected to a fire hydrant or other water sup-
ply. Begin at the road crown and flush toward the edge. The downs lope gut-
ter is dammed with sandbags to create a collection area. A small vacuum
collector is used before an industrial wet/dry vacuum cleaner to remove the
sample water from the collection area. All water and contaminants are col-
lected using this vacuum-operated collector trap. This is an air-tight box
or drum with a capacity of several gallons to several hundreds of gallons
(depending upon specific test procedures), outfitted to function as a "trap"
in a vacuum line. The inlet hose of the collector trap has a pickup nozzle
on the open end. The outlet hose of the collector trap is connected to an
industrial shop vacuum.
The vacuum cleaner used for collection of roadway particulates consists of a
pick-up head attached to a 38* (10 gal) canister on the top of which is
D-117
-------
mounted an exhaust motor. Exhaust ports from the canister leading to the
motor are covered by a filter bag to retain solids picked up during the vacu-
uming operations. Since the finer particles found on roadways are relatively
more heavily laden with water pollutants, experiments have been performed to
determine the retention of smaller-sized particles by the filter bag. Re-
coveries of 99, 93, and 94 percent were obtained using a new filter bag with
each sampling run. These tests indicate satisfactory retention of fine par-
ticulates by the filter bags as well as quantitative removal and recovery of
vacuumed particles from the canister walls and bags.
The water flush procedure has also been tested in the field. It was found
that a roadway area of 92 sq m (1000 sq ft) could be thoroughly flushed with
about 951 (25 gal) of water. In most cases, over SO percent of the applied
flush was recovered by vacuuming of the impounded water along the curb.
A specific stepwise sampling procedure for the collection of street surface
contaminants is given below.
1. Select a roadway sampling site 30.48 continuous curb meters
(100 ft) or more. The street surface and curbing should be
in relatively good condition. Mark the limits of the sampling
length selected.
2. Rake and/or brush along the curb for 3.0 or 4.6m (10 or
15 ft) from the limit markings away from the section to be
sampled.
3. Knock the brush clean. Rake and/or brush from the higher ele-
vation limit. Shovel bulk litter plus swept dust and dirt
into a clean galvanized garbage can.
4. Vacuum along the entire curb length of the roadway sampling
site out to a distance of four to five feet from the curb.
Three vacuumings of the site should be carried out to collect
the dust and dirt sample fractions. Two vacuum cleaners are
D-11S
-------
used simultaneously to speed up the operation with particular
attention at the litter pickup point.
5. Position several sand bags at the curb of the lower limit of
the sampling area to impound the flush water.
6. Place the nozzle of a dual motor shop vacuum at a low point in
front of the sand bags so as to suck water into a 2081 (55-gal)
drum.
7. Place the intake hose from a rotary screw pump into a 208£.
(55-gal) drum filled with water and begin flushing the roadway
using the garden hose.
8. Flush the entire roadway surface area toward the curb and finish
by flushing the gutter toward the sand bags.
9. Approximately 57 to 95£ (15 to 25 gal) of water are required to
flush 56-93 sq m (600-1000 sq ft) of roadway. Generally greater
than 50 percent of the flush water applied is recovered by the
vacuum.
10. Take out the filter bags and shake well into garbage can with
bulk material. Save the bags.
11. Empty vacuum canisters into garbage can. Brush canisters well.
12. Take combined litter and dust and dirt in garbage can and the
flush fraction to the laboratory. Other equipment may proceed
to next sampling site.
Sampling sites should be chosen to represent the range of conditions that oc-
cur in the area. Important variables may include land use, average daily
traffic, type of adjacent landscaping, and street surface material. It is
recommended that at least a single complete analysis be made for each land
D-119
-------
use area, with total solids analyses being made on samples representing other
identified variables. If several sampling sites are established in each land
use area, a portion of each sample could be combined for complete composite
chemical analysis representing that land use.
For one 12-month field study, seven area roadways were chosen based primarily
upon the range of average daily traffic levels and road use categories encom-
passed. Other factors considered in the roadway selections were speed limit
and roadway surface material. Satisfactory condition of the street surface
and a sufficient length of curb against which the sample could be deposited
and collected were important factors in selection of the specific sampling
sites on the area roadways chosen.
In general, the following information should be collected for a sampling
site: sampling location; date; local land use; parking restrictions; traffic
characteristics; composition, type and condition of the street, gutter, and
curb; the size of the test area; and a description of the adjoining area.
Photographs of the area are often valuable. Data concerning the cleaning
frequency, the date of the last recorded cleaning, and the recent rainfall
history should also be obtained for each test area.
If the selected study area is subject to vehicular traffic, it will be neces-
sary to establish some type of traffic control for the protection of the
field workers. Flagmen and traffic cones are probably a minimum precaution
which should be used in all areas.
The type of study area (street surface, parking lots, or other large sur-
faces) and sampling objectives will determine the size of sampling area. A
typical secondary street can usually be sampled using a single test area of
about 93 sq m, 7.6m x 12.2m (1000 sq ft, 25 x 40 ft). Large paved surfaces
nay be better sampled using several smaller test areas (0.9 sq m (10 sq ft))
and averaging the results. Experimental design procedures should be incorpo-
rated to determine the necessary types of study areas to sample to satisfy
specific study objectives.
D-120
-------
As with the selection of the study area, the frequency of sampling will de-
pend on the objectives of the sampling program. For one 12-month field
study, a schedule was set up early in the program such that the roadways
were sampled during several seasons of the year in order that seasonal ef-
fects on pollutant deposition rates might be studied. However, during the
winter season, freezing conditions prevented the collection of some of the
flush fractions.
Sampling periods were scheduled to begin on a Monday and end one week later
on the following Monday. Sample collections were planned to be carried out
in the following manner:
1. An initial sample was obtained by cleaning the roadway surface
and quantitative collection of materials initially found on
the site. No measurements of traffic were taken to correspond
with the initial sample; however, records of precipitation and
dates of the most recent antecedent cleaning of the roadway
surfaces were maintained throughout the 12-month field study.
2. The site was sampled a second time after an accumulation period
of approximately 24 hours during which time a measured volume
of traffic passed the roadway site. As many as four samples
having a one-day accumulation period were taken during the re-
mainder of the week. Traffic counts were taken with each one-
day sample.
3. The final sample of the period was gathered following the week-
end. Ideally then, a sampling period consisted of an initial
sample, four one-day samples, and a weekend sample with traffic
data for all samples except the initial one.
4. Precipitation frequently interrupted the planned pattern of the
sampling periods. Samples were gathered after rainstorms in
a few cases; however, it was felt that such samples would be
atypical; and, therefore, collections after runoff events were
D-121
-------
abandoned early in the program. The roadway site was cleaned
as soon as convenient after precipitation had ceased and a new
sample accumulation period begun. Sampling periods were ex-
tended in some instances in order to make up for loss of sam-
ples due to precipitation.
Experimental design procedures should be incorporated to determine the re-
quired sampling frequency and sample numbers to satisfy specific study ob-
jectives. The published results of previous sampling programs may be useful
in this design process.
D.5 Cost Estimation
It is difficult to provide precise program cost information, since costs are
dependent upon so many program and locality related factors, e.g., insti-
tutional setting and accounting procedures, area complexities and program
size, opportunity free labor, etc. This section presents a methodology for
cost estimation for any given program, some "ball park" rules of thumb for
preliminary rough cut costing, and some specific examples. The costing
methodology is divided into six steps:
1. Estimate Instrumentation Costs
2. Estimate Related Equipment Costs
3. Estimate Manpower Costs
4. Estimate Field Operations Costs
5. Estimate Laboratory Analysis Costs
6. Estimate Data Analysis and Reporting Costs
These will be discussed in turn. Note that modeling costs are not included.
D.5.1 Instrumentation Costs
Instrumentation costs were discussed in Section 0.3 and will only be sum-
marized here. These costs represent capital acquisition costs for the most
part, and amortization schedules will be a matter of local accounting
D-122
-------
procedures and discretion. Resist the temptation to lower apparent program
costs by using long amortization periods, especially for equipment used in
storm-generated discharge studies. Such hostile flows take a great instru-
mentation toll, and one or two years life is much more typical than ten or
twenty.
For flow measurement, the types and numbers of primary and secondary devices
must be determined. These are multiplied by the cost of each to arrive at
the total dollars required. Make an allowance for spare parts for secondary
devices (say 10% in the absence of more specific information). Consider the
purchase of at least one complete extra unit to allow for quick field fixes.
Instrument breakdown is most likely to occur during important data collection
periods, and record interruptions should be as brief as possible.
Langbein and Harbeck (23) reported that a sample of four USGS districts
yielded the following costs (in 1972 dollars) for flow-gaging stations: $SK
to S10K for installation of an indefinite-term full-record station; $2.5K
to $4K for installation of a short-term full-record station; station
operating costs of $0.8K to J1.3K per year; office costs for processing the
record of $0.5K to $1.3K per year. For partial-record stations, costs as a
percentage of full-record stations were stated as: 5 percent for low flow
only; 15 to 20 percent for crest stage record; and up to 50 percent for a
flood hydrogram. They also noted that there could be extremely large
variations outside of these nominal ranges.
In the absence of better information, it is suggested that $10K be budgeted
for each urban flow measurement site for the acquisition and installation of
a primary flow measurement device. Allow $2K each for the secondary device.
The former number takes into account that some sites will allow relatively
inexpensive portable devices to be used, while others may require consider-
able modification, e.g., installation of a below ground metering vault. The
desirability of using existing USGS gaging sites where feasible is obvious.
Where possible, try to use existing meteorological instrumentation operated
by others. One notable exception will be raingages. They typically cost
D-123
-------
less than $400 each; recorders may add another $500-$1,000 each. Give
serious consideration to the use of leased telephone lines to provide rain-
fall indication back to a central location. The cost is nominal, and the
information is invaluable in crew dispatching and operations. At least one
raingage per catchment will be required.
For automatic samplers, the number and types must be determined. For storm-
generated discharge sampling, the bulk of the commercially available devices
will not be suitable without some modification. Most manufacturers will do
this, but it adds cost. Between $2K and $4K should be allowed for each unit.
Allow 10 percent for spares and purchase at least one complete extra unit.
Also give consideration to installing two separate automatic samplers at
critical sites for redundancy. If both function flawlessly, the extra sample
quantity won't hurt, and the likelihood of missing a critical storm event is
considerably diminished. Manual sampling equipment must be provided to each
field crew. Allow two sets for each crew and plan on $100 for each set.
D.5.2 Related Equipment Costs
Shelters will be required for monitoring instrumentation at most sites.
Costs can range from under $300 for a metal garden shed to well over $2K if
concrete slabs and heavy fencing are required. Consideration should be given
to ease of moving instrumentation from site to site; transportable (i.e.,
trailer) shelters have been used successfully in this regard. For some
installations, e.g., one with a large mechanically refrigerated sampler, AC
power will be required, and the expense of running electrical lines should
be included as part of the overall station cost. Such costs may not be in-
significant; $6.4K was spent just to get power to one 208 stormwater monitor-
ing site in Illinois. Site preparation costs should not be capitalized.
Other related equipment that will be required includes small tools, personnel
safety and protective gear (e.g., waders, hard hats, respirators, harnesses,
etc.), and miscellaneous field hardware. These may or may not be already
D-124
-------
available. As a very rough estimate, allow 2 percent to S percent of the
total instrumentation acquisition cost for this purpose. Do not capitalize
them.
Other related equipment includes vehicles (automobiles, vans, trucks, etc.),
boats, motors, generators, pumps, and the like. These are capital equipment
items but, because of their multiplicity of uses, they should probably not
be fully charged to a 208 program alone. In the event that the local 208
agency does not have such equipment at its disposal, consideration should
be given to leasing rather than purchase. Lease rates vary with locality,
but for longer term rentals, i.e., months, not days, rates can be quite
reasonable. For mobility of field crews, consideration should be given to
leasing extra vehicles during periods of intense activity. Of course all
leasing costs should be directly charged to the monitoring program.
D.5.3 Manpower Costs
The manpower costs associated with a 208 monitoring program will vary tre-
mendously from agency to agency, depending upon the size of the program
(and hence the number and skill mix of personnel required) as well as local
wage scales (including fringe benefits) and accounting practices (applica-
tion rates for overhead and general and administrative expenses). Therefore,
skill levels here will be indicated by estimating equivalent federal govern-
ment service ratings. Salaries can be adjusted up or down to suit local
conditions, and burdens can be applied according to local accounting
practices. Table D-23 indicates the types of talent that a 208 program may
require.
As an example of the use of Table D-23 to estimate manpower requirements,
consider the following. It is desired to conduct one intensive survey per
month for a one-year period. The objective of these intensive surveys is to
provide information for waste load allocation studies. The basic unit man-
power for the estimates made here consist of a field party chief, three
qualified technicians, a chemist, a microbiologist, and a biologist. It is
assumed that the minimum sampling period would be 5 consecutive days. The
D-125
-------
TABLE D-23
TALENT REQUIREMENTS
Skill Area
Environmental Engineer (1)
Sanitary Engineer
Hydrologist
Chemical Engineer
Chemist
Oceanographer (2)
Biologist
Limno legist (3)
Field Technicians
Lab Technicians
Clerical
Federal
GS Rating
GS 13-14
GS 11-12
GS 9-11
GS 9-11
GS 11-13
GS 11-12
GS 9-12
GS 9-11
GS 3-6
GS 5-7
GS 2-4
Annual
Pay Rate
$23K to $35K
$16K to $25K
$13K to $21K
$13K to $21K
$16K to $30K
$16K to $25K
$13K to $25K
$13K to $21K
$7K to $13K
$9K to $14K
$6K to $10K
NOTES:
1. Assumed to be responsible for overall monitoring program.
2. Required for estuarine or near coastal studies.
3. Required for lake surveys.
D-126
-------
basic intensive survey unit manpower estimates are shown in Table D-24.
Using the salary figures from Table 0-23 as a guide, the direct labor costs
for one intensive survey are as follows: Field Party Chief, $26K x 3.75 MW =
$1.9K; Chemist, $18K x 16 MW = $S.5K; Biologist, $16K x 5 MW = $1.5K; Field
Technicians, $9K x 5 MW * $0.9K; Lab Technicians, $11K x 3.25 MW * $0.7K;
Typist $7K x 1 MW = $0.1K. Thus, the total salary requirements for one
intensive survey would be approximately $10.6K.
Of course it must be kept in mind that not all personnel time can be utilized
at 100 percent efficiency due to a number of reasons (e.g., in runoff stud-
ies the field crews have to be paid whether it rains or not), and so a better
procedure for budget estimation is to calculate annual salary costs for all
required personnel rather than on a work unit basis. One last comment on
manpower costs deals with the actual hours worked while in the field.
Twelve hour (or longer) days are the rule rather than the exception, and
extra compensation for this overtime must be allowed for in arriving at total
manpower costs. Non-professional (i.e., non-exempt) personnel must be paid
in accordance with applicable wage/hour laws (e.g., time and one-half for
over 8 hours per day or sixth straight day in a week, etc.). Professional
(exempt) personnel will be paid straight time for hours worked, given com-
pensatory time off, or some such consideration depending upon local policy,
but this also represents cost to the program and must be accounted for.
D.5.4 Field Operations Costs
This is a miscellaneous category that covers costs incurred incidental to
field operations and that do not logically fit in any of the foregoing dis-
cussions. Included are such items as personnel travel costs and per diem
as appropriate, miscellaneous supplies (as opposed to equipment, e.g., ice
for samples if required, chemical preservatives, sample containers, gasoline,
etc.)» performance bonds for site restoration if required, charges for util-
ities (electricity, telephone lines, etc.), and so on.
D-127
-------
TABLE D-24
ESTIMATED MANPOWER REQUIREMENTS FOR INTENSIVE SURVEYS
Activity
Personnel
Time
(man-weeks)
Remarks
do
Initial planning
Reconnaissance
(if needed)
Mobilize field equip-
ment and crew
Field sampling
Fixed lab analyses
chemistry' and biology
Data analyses and
report preparation
Field party chief*
and lab personnel
Field party chief*
and biologist
Field party chief*
technicians and
lab crew
Field party chief*
2 laboratory crew
3 technicians
1 biologist
Chemist
Biologist
Field party chief*
chemist and
microbiologist, typist
2 MW
1 MW
1 MM
1 MM
3 MW
4 MW
1 MW
15 MW
3 MW
3 MW
Assemble maps and post data
Select sampling sites and
synoptic biological screening
Get all equipment together
and ensure it is in working
order
Field sample collection and
field lab analyses
Assume 20 samples per day
for 15 parameters, chemistry
and plankton, and invertebrate
identification and enumera-
tion
Analyze data, write and type
report
In the case of estuarine or near coastal studies this would be an oceanographer.
-------
Taken individually, these items do not represent major sums of money but,
collectively, they form a sum that may not be insignificant for many 208
monitoring programs and, therefore, must be considered in total cost estima-
tion. They are so project and locality specific that no specific guidance
can be given for cost estimation. Lacking anything else, add 2 to 5 percent
of the total survey cost to cover this category and adjust as appropriate
during detailed survey design.
D.5.5 Laboratory Analysis Costs
Use costs for analyses of the selected parameters quoted by the chosen labo-
ratory where possible. Use costs given in the Parameter Handbook for pre-
liminary estimating if local cost data are not available. Add 10 percent to
20 percent to the total for quality control costs.
As an example, Table D-25 contains average analysis costs for the minimum
recommended parameter list for characterizing urban runoff. Summing the
costs for the individual analyses results in a total estimated analysis cost
of $163 per sample. If a sequential discrete sample series of 24 bottles
was collected to characterize a storm event, the total lab fee would be
$163 x 24 = $3,912. Adding 15 percent for quality control, the final esti-
mated laboratory analysis cost would be approximately $4.5K per storm event.
Thus, laboratory analysis costs are one of the major operating costs and may
amount to 30 percent to 50 percent of the total cost for this portion of the
program budget.
D.5.6 Data Analysis and Reporting
Costs in this category will depend upon the complexity of the survey, the
degree of data interpretation required, computer charges for statistical
analyses where necessary, and the type of report being generated, e.g., event
summary, annual project, etc. Ball park estimates of 20 percent to 50 per-
cent of the estimated professional manpower costs for field work will be
adequate in most instances. Use more refined, project specific cost informa-
tion as it becomes available.
D-129
-------
TABLE D-25
AVERAGE ANALYSIS COSTS FOR URBAN RUNOFF PARAMETERS
Parameter
Cost
BOD5
TOD
Suspended Solids (NFS)
Volatile Suspended Solids
Fecal Colifonn (MF)
Fecal Streptococcus (MF)
Nitrate-Nitrite Nitrogen
Kjeldahl Nitrogen
Total Phosphorous
Lead
Zinc
Copper
Chromium
pH
$ 10
30
8
7
10
10
15
15
15
10
10
10
10
3
D.5.7 Example USEPA Costs
Harris and Keffer (16) have provided some information on the costs of a USEPA
Surveillance and Analysis Field Investigations Section engaged in effluent
monitoring for compliance verification purposes and technical assistance to
208 agencies, e.g., stream monitoring. Major field equipment with approxi-
mate initial costs is listed in Table D-26. The Field Investigations Section
professional staff includes two sanitary engineers (GS-13 and 11), one
chemical engineer (GS-11), and one hydrologist (GS-9). The subprofessional
U-130
-------
staff consists of four engineering technicians in grades ranging from GS-3
to 6. The regional laboratory, with a staff of eight professional chemists
(GS-7 to 13) and three microbiologists (GS-7, 9, and 12), is responsible
for operating the mobile laboratories of the section during field surveys.
TABLE D-26
MAJOR FIELD EQUIPMENT AVAILABLE TO USEPA REGION VII
SAD FIELD INVESTIGATIONS SECTION
Quantity
Equipment
Approximate Initial Cost
1
1
7
5
SO
Mobile Laboratory
Mobile Laboratory (on loan)
GSA Vehicles
Boats and Motors
Automatic Samplers
Flow-Measuring Devices
Field Analysis Devices
Portable Detector
Metal Detector
$15,000
5,000
28,000
6,600
6,100
1,200
300
In areas outside the range in which analytical support can be provided by the
regional laboratory, field sampling teams normally operate within a 161-km
(100-mile) radius of a mobile laboratory, which is generally set up at a
wastewater treatment facility in a community within the area of interest.
Because of logistics problems in some of the more sparsely populated areas of
the region, it is frequently necessary to work field teams outside of this
161-km (100-mile) radius. Ten to twenty-five percent of the total field ac-
tivity may be conducted at distances up to 322 km (200 miles) from the labo-
ratory base. Operating at these greater distances reduces capability by an
estimated 50 percent and greatly increases the unit cost of sample collection.
Prior to mounting a survey, every effort is made to ascertain and consoli-
date the various data needs of the Agency and of the State in order to avoid
D-131
-------
duplication of effort and to minimize the number of laboratory setups. It
requires a minimum of 1 week to 10 days to prepare and stock a mobile
'laboratory; get it on site; have electricity, water, and phone installed; and
then'torn down and returned to the base station following completion of a
survey. If possible, field activities in areas requiring mobile laboratory
support are restricted to surveys of 30 days duration or longer.
Under favorable conditions, a mobile laboratory field operation works best
with a crew of seven people including: two engineers, two engineering tech-
nicians, one chemist, one microbiologist, and one laboratory technician.
Working[entirely within a 161-km (100-mile) radius of the mobile laboratory,
this^staff (which is rotated at 2-week intervals) would be able to install
samplers and collect approximately 100 samples per week for field and labo-
ratory analyses. Total time and costs for a 30-day field survey are esti-
mated as follows:
Engineers
1 man-month office preparation
•.:.-,• 2 man-months field work
2 man-months data analyses and report writing
Engineering Technicians
2 man-months mobile laboratory and equipment repair and preparation
4 man-months field work
Laboratory Personnel
6 man-months mobile laboratory work
6 man-months regional laboratory analytical work
Clerical
2 man-months planning and report preparation
Costs
Salaries $23,500
Per Diem 7,300
Travel of Personnel 400
Government Bill of Ladings 400
' •'•" Vehicles 1,000
••::•'•'• Miscellaneous Equipment 1,500
-ev~,.. (Ice, batteries, containers, utilities, chemicals, etc.)
$34,100
0-132
-------
When reviewing the foregoing costs, it should be kept in mind that theyc.dp
not reflect any burdens (e.g., office space, heating, employee fringe bene-
fits), that equipment costs have increased considerably, and that this team
is proficient, well trained, and one of the most efficient in the country;
1 : ' ' L
D.6 Waste Load Allocation Study Procedures *•••: -;,u...
This section contains procedures for conducting a waste load allocation
study and some recommendations and guidelines for implementation of the'
field surveys that will be necessary to carry it out. Most waste load al-
location studies will require, as a minimum, one reconnaissance"survey "arid
two intensive surveys, one to gather model calibration data and the other
to gather verification data. The procedural steps to be followed in con-'
ducting a waste load allocation study are outlined in Table 0-27.
The first step is to obtain all relevant existing data. This will include
all available water quality and flow data for surface water and known
sources of wastes. Locations of water use (and a list of legitimate uses)
should be determined. Obtain maps and either mark or prepare overlays
showing land uses, outfall locations; existing stream gaging and monitoring
site locations; stream slopes, cross sections, and flows; locations of
rapids, dams, pools, etc.
Analyze the existing data; estimate stream velocities, relative loads from
each point source and nonpoint source area, water quality coming into and
leaving the planning area, travel times downstream from discharge locations,
etc. Use the results to determine first approximations of station locations
and parameters to be covered. Plan the reconnaissance survey accordingly.
It is recommended that the reconnaissance survey include a toxics scan of
all major discharges and others of concern to the 208 agency in order to
identify toxic parameters that should be included in the intensive survey.
The next step is the conduct of the reconnaissance survey. The information
gathered at this time will form the basis for the intensive survey imple-
mentation plan. In conducting the reconnaissance survey, the field survey
D-133
-------
TABLE D-27
PROCEDURAL" STEPS "FOR CONDUCTING A WASTE LOAD ALLOCATION STUDY
(1) Obtain all relevant existing data.
' (2) Analyze existing data and perform preliminary calculations.
C3) Based on (1) and (2), plan reconnaissance survey.
""(4) 'Conduct reconnaissance survey.
(S) Analyze results of reconnaissance survey and make preliminary
model runs.
'(6) Based on (3) and (S), plan calibration survey.
(7) Conduct intensive survey for model calibration.
(8)~ Reduce data from calibration survey and analyze results.
(9) Fit model using results from (8) and run.
(10) Review results of model runs.
(11) Based on (8) and (10), plan verification survey.
(12) Conduct intensive survey for model verification.
(13) Compare results of verification survey with model predictions.
(14) If results of (13) are favorable, use the model for planning.
If not, make adjustments to the model as warranted in view of
(13), using the verification data gathered in (12) as additional
calibration data. New verification data will now be required,
so repeat steps (11) through (13).
manager should be accompanied by persons who supplement his own skills, e.g.,
if he is a sanitary engineer, he may have with him a biologist and a chemist.
The biologist is an especially important member of the reconnaissance team.
An experienced aquatic biologist in a very short time can collect and ex-
amine bottom organisms that will reveal both the severity of pollution in a
general way, and the length of stream affected. His findings can, for ex-
ample, reveal whether the effects of the wastes have extended farther down-
stream in the past then they do at the time of the reconnaissance and will
have an important influence on the final planning of the study, especially
with regard to the number and locations of biological sampling sites.
D-134
-------
A quick tour of the area and the streams at readily accessible ,points may be
taken to get the general "lay of the land" and the relationships among water
uses, waste sources, and the stream. After this, the individuals of the :tearn
may go about their separate duties. The field survey manager needs to cover
much of the ground that each of the others does,, though in less detail. 'He
must have the entire situation in mind to develop the final s.tudy plan,
supervise the subsequent field operation, and prepare the repor£. -, '
The field survey manager should become thoroughly familiar with,.characteris-
tics of the streams. A trip throughout each reach by boat, if the,stream is
deep enough, provides the best opportunity for observation.. Access^to the
stream may be limited to bridges and roads that parallel the stream if a j
boat cannot be used. An overall view of the stream may be obtained from a
plane or helicopter, but observation of detail from the height involved is
limited. Walking often is difficult because of undergrowth or rough terrain,
and can be extremely time consuming unless the stream reach is very short.
Detailed notes of observations should be made promptly; don't depend on ;
memory. Notes should include general impressions of depths, currents, veloc-
ities, bends, widths, types of bottom, water uses, waste discharges and
mixing of wastes, availability of access, and sensory evidences of pollution,
such as excessive plankton or attached growth, floating materials, oil, color,
suspended matter, sludge deposits, gas bubbles, and odor. Special attention
should be paid to tentative sampling stations selected in the preliminary
planning. Accessibility of stations, as well as suitability for sampling,
must be considered. Stations should be marked or otherwise identified to en-
sure sample collection at the proper points. For example, the stream miles
may be painted on bridges, with arrows indicating the sampling points.
A dry run of the sampling route or routes should be made and timed. This in-
formation will be needed in estimating the final number of sample collectors
that will be necessary and the maximum time samples will be held. The routes
should be marked on a map, and notes made of any check points that will
assist in following the routes. Stream samples for preliminary analysis
D-13S
-------
Ijl be, collected,at this time to assist in parameter selection for the
intensive survey and to familiarize the laboratory personnel with what to
anticipate when the study starts, e.g., determination of colifo-rm and BOD
will..assist in selection, of proper dilutions, possible interferences can
be identified,,,.etc.. simple, field determinations, such as those .of tempera-
ture ,,.D02,;, and pH, may be made at the same time.
Potential locations for. a mobile laboratory, if one is to be used, should be
...investigated., . frequently the site is a local water or sewage treatment plant.
Accessibility and, suitability of an area where the unit may be parked must
be considered. Availability of necessary water and electrical connections
must be .checked. Arrangements for metering water or electricity should be
.made,.if necessary. An area, sewer, or drain to which wastes can be dis-
charged from the laboratory without nuisance is needed. Arrangements for
access at any time, day or night, must be made if the area is fenced or other-
,^ise protected. A nearby storage room or space for supplies and materials
that are not in immediate use in the laboratory is useful. Convenient tele-
phone^, service is a must, especially if the laboratory is to serve as head-
. quarters for the field crew.
Facilities may be established in a local laboratory of a water or sewage
treatment plant, high school, university, or industrial plant as a substitute
for a mobile laboratory. The chemist in the reconnaissance survey crew
should review such local facilities to determine their adequacy and what ad-
ditional equipment and supplies will be needed.
If stormwater runoff is a survey concern, try to include a rainy-day visit as
part of the reconnaissance effort. Much valuable information can be obtained,
as well as a better appreciation of the conditions the field crew will be
working under. For urban areas, sewer maps should be verified as to discharge
locations, and the possibility of unrecorded outfalls investigated. Obtain
information on traffic density by driving during rush hours and use these
travel times for urban field crew logistics planning.
D-136
-------
After completion of the reconnaissance survey, 'analyze the results ain'd ma;ke
preliminary'model runs using the reconnaissance data to get the model'"s'eg^-
mented properly and to better locate stations. With the results ovf the re-
connaissaiiice survey and model runs as a -guide, "a' workable intensive "survey
plan may be" generated. Start by a careful revfevTof the bbje'ctivesr'can-
they all be accomplished? Now is the time for any additions "aif deletions,
not halfway through the field activity. Put the detailed plan down in writ-
ing and have it reviewed by all involved prior to finalizing/ Don't dis-1
count helpful comments and suggestions from the field technician's^ •' Include
samples of all field data sheets, equipment checklists', etc'. '"' ' -' -s'-:-
• : •••/ . :-r. .33., -.TO;.- -a^
Pay especial attention to all logistics aspects during preparation of tire*
intensive survey plan. For example, if ice is used to cool samples and°the
survey calls for round-the-clock activity, locate sources where ice can be
obtained at odd hours, e.g., automatic machines at service stations, all-"
night convenience stores, etc., and write them down so all will know. The
25-cent do-it-yourself car wash facilities at some service stations represent
sources of high-pressure hot water (and soap, if detergent analyses a'fe^n'ot
performed) that can be used for cleaning automatic sampling and other* field
equipment, and their locations should be indicated.
The'parameters to be measured must be listed, taking into account the prob-
lem assessment of parameters determined from the toxics scan, along with
any special handling precautions to be observed, preservatives to be used,
sample volumes required, etc. Lists of special supplies and equipment and
personnel requirements should be prepared at this time. The funds allo-
cated for the survey and the anticipated cost of the field operations should
be reviewed here also.
For around-the-clock survey efforts (with two crews working 12-hour shifts,
for example), allow for communication of significant information at shift
change by having each shift leader report at least 30 minutes early. If such
surveys are for extended periods of time, plan on changing crews every 2:weeks
to avoid excessive fatigue. In any event, a system of communication that
D-137
-------
allows any crew member to be contacted within a reasonable period of time
•fsay 3 hours) i's highly recommended. Radios on vehicles are also useful
communication aids-. '•" J • ~ :
;:, [;•• :-\ • :•-", .:> '•; W: :'. i i:.T7v~ v: • "'J
Prepare' cbmple'te equipment invelftrbries that show locations, status, and main-
••tenanceantf calibration'sdiedulVs'. Be certain that maintenance responsibil-
, • i jtc~
ities are clearly defined. In addition to the more obvious equipment,
instrumentation, and spares discussed earlier in this Appendix, there are a
-number of other miscellaneous items that will prove useful in the field, and
'a few examples will be mentioned. A fairly strong magnet on a line is use-
ful for retrieving metal objects dropped in water. A metal detector is also
a handy-device"at times. A set of basic surveying gear (transit, distance
tape or chain, stadia poles, optical rangefinder, etc.) will be useful in
some instances. A pick, shovel, ax, and saw will find several applications,
such as improving rural stream access for sampling. Some of the basic car-
t ji. x .•
penter's tools on hand will be needed at times. A Danaides (orifice) bucket
rj ?•
is useful for estimating moderate pipe discharges into open air. A number
of uses will arise for rope, string, wire, and reinforced sticky tape. A
walkie-talkie set greatly facilitates communications in the field.
• .'i • • ' '
Obviously, not every contingency can be allowed for, and experience in con-
*>
ducting field surveys will facilitate future planning. However, the field
survey manager should feel very uncomfortable in reviewing the intensive
survey implementation plan if he:
•)
o Does not clearly understand the survey objectives,
• Has strong preconceived notions about the results,
• Has not personally visited each measurement site,
•"•'• Has not consulted with laboratory personnel,
• Has not clearly assigned responsibilities,
• Has trusted to luck or favorable opportunities.
'Conduct the intensive survey in accordance with the implementation plan
insofar as practicable. Be certain that there are compelling reasons for
aliy 'changes or deviations and document them. The importance of recording
, "^ •' - . b.;'
D-138
-------
all information that might aid in the interpiLetajtipn and analysis -p£
data that are being collected has been stressed throughput this Appendix:;
Even seemingly insignificant bits of information may be very useful in
fitting the entire puzzle together. Howeyexv. if ,thf-y are not; recorded-tin-'
. ?„ .lire -,._j.!..- - •'.•••" .-•.'
a clear and intelligible way, they are likely £ta be los.t.cpmplete^y
trust to memory. • r.;Si;i ^.i.
Field logbooks should be provided to each leader,, of a. sury.ey teamiand^Others
as appropriate. Standard procedures for field,data taking,should?be observed,
e.g., logbooks should be bound with pages numbered^serially,.-entries should
be made with ballpoint pen, erasures should never be permitted Ou§eH3_£riJce.-
outs), all entries should be signed and dated, etc. Field?logbooks -.should: be
clearly titled on the cover (to prevent the aquatic biologist from acciden-
tally picking up the equipment crew's log, for example) and should have* an
assigned location when not in use. The field survey managers should frer ,
quently review all field logbooks and initial and date them when this is ,
done. . .. ^-^ -;C
j" '•"•'-':••- . " -. .'••"'«
The field logbooks are the main source of data annotation information. Be
sure to record information so that it will be useful to future surveys as
well as the present one. For example, entries may range from snake sight-
ings to traffic jams that delayed getting samples to the laboratory.
Knowledge of the former can reduce possible future danger to.personnel, while
information about the latter may suggest the desirability of route alteration.
It would be improper to assume the snake would never pose a problem (even if
it were killed, others may be around) or that anyone could tell that there
had been a delay in getting the samples to the laboratory by comparing the
time they were removed from the sampler with the time they were logged in at
the laboratory.
The importance of time synchronization of data has been stressed earlier.
This is equally important with data annotations. Write down the time pfrday
(use watch time) whenever an entry is made in a field logbook. This will,.as-
sist in subsequent interpretation. Finally, make full use of field.logs.and
other annotation records in report preparation. The perfect survey has yet
D-139
-------
to be performed, and mistakes and errors will happen. Sweeping these under
the table is much more censorable than admitting that they occurred, espe-
. ci.a&'y;' where; a si^ificantfrlMp%^7o,h%)tta"quality or interpretation is
likely to result. The worst sin is data fudging (or outright falsification)
in an attempt,to cover up mistakes._ This abhorrent practice should be subject
to the most severe reprimand possible. Do not confuse this with-data adjust-
ment based on the best available information and professional judgment, e.g.,
adjustment ,to a.Cfldw record "time base to account for a uniformly slow-running
,cJLock drive, accounting for a zero offset, etc. Data adjustment is an accept-
able practice if it is clearly annotated and explained.
After completing the intensive survey for model calibration, reduce and
analyze the data and prepare it for model input. Fit the model using the
calibration data and make several runs. After reviewing the results of
the calibration survey and the model runs, plan the intensive survey to
collect verification data. Unless confidence in the model results is very
high, it will be prudent to plan and conduct the verification sufvey with
the same degree of'coverage as the calibration survey. Conduct the inten-
sive survey for verification data following the above guidance for the
calibration survey and taking advantage of lessons learned from it.
3 '*
Finally, compare the results of model runs with the data gathered during
the verification, survey. If the comparison is favorable, the model may be
considered verified and can be used for waste load allocation planning.
If. differences are significant, make adjustments to the model as warranted
from a review of the discrepancies between the model predictions and the
results of the verification survey, using the verification data now as
additional calibration data. Once this latter step is done, there is no
longer any information about the verification of the model. The importance
of this point- must not be overlooked. To verify the model it will be neces-
sary to plan and conduct another intensive survey solely for this purpose.
In many instances,^however, it will not need to be as extensive as the
earlier intensive surveys, since the level of confidence in model results
should be considerably higher than before.
D-140
-------
REFERENCES
1. Tod^.J^.K. , ztaJL., "Monitojqji.g ..Groundwater Quality.:. >kmitoriTvg.
Methodology." USEPA Environmental Monitoring '"Series , EPA'-~60d/4-7lb-026
I'Oi'i'
2. Everett, L. G. , it aJL. , "Monitoring GroiaidWater Quality: ":Meth"bds" and
Cost st? !£> USEPA Environmental Monitoring, Series, EPA-600/34-d7r6>025r'(1976) .
3. Hampton, "N. F. , "Monitoring Groundwaf'ef "QuJalityT""ba'ta Maha'gement. '*"'''
USEPA Environmental Monitoring SeriesV"±-EPA-.600/4^£6rfQl:9
4. Crouch, R. L. , it aJL. , "Monitoring Groundwater ''Quality : Economic Frame-
work and Principles." USEPA (in press), '.(?1;976D .
5. Tinlin, R. M. , "Monitoring Groundwater Quality: Illustrative Examples."
USEPA Environmental Monitoring Series, EPA-600/4-76-0^6 (1976).
6. "Model State Water Monitoring Program." USEPA National Water Monitoring
Panel, EPA-440/9-74-002 (1975).
7. Wullschleger, R. E., Zanoni, A. E. and Hansen, C. A., "Methodology 'for
•!••••• thes.Study of Urban Storm Generated Pollution and Control." USEPA. 1-
., . Environmental Protection Technology Series, EPA-,600/2-76-145 (1976).
8.. Shelley, P. E. , and Kirkpatrick, G. A., "Sewer- Fl-ow! Measurement:,.;- A<
State-of-the-Art Assessment." USEPA Environmental Protection Technology
Series, EPA-600/2-75-027 (1976). ' -:/
9. FtuAd Me^etA - Tfeex/i Thtoiy and kppLLc&tion. American Society of
Mechanical Engineers , Report of the ASME Research Committee on Fluid
Meters, Sixth Edition, New York, NY (1971)
TO. Replogle, J. A. ,; "Flow Meters for Water Resource Management." WoteA.
I, ReiouAcaA Botfce/Cin. pp 345-374 (1970). . .
fl :' McMahon, J. P.', "Flow of Fluids." Section 5-1 of Handbook 0(J App&ced
-'.' IrUt/UunervCcttcon, D. M. Considine, ed., McGraw-Hill Book ..Co., New York,
NY (1964).
12. WateA MeoAUA&ment Manual. Second Edition, U.S. Dept. o£ Interior,,]
Bureau of Reclamation, GPO, Washington, D.C. (1967).
13. Leupold and Stevens, Inc., WoJtVi RuouA£.eA Uota Book. : First Edifion,
„,. P.O. Box 688, Beaverton, OR 97005 (1974).
14. 'Shelley, P. E. , and Kirkpatrick, G. A., "An Assessment , of Automatic
Sewer Flow Samplers." In tfcttet VoUMtLon A44e44mejtt: Auiomotitc
Samp&ung and MeaAuiuntnt, ASTM Special Tech. Pub. 582,
(1975).
D-141
-------
15. Shelley, P. E., and Kirkpaierick, G. A., "An Assessment of Automatic Sewer
Flow Samplers - 1975." USEPA Environmental Protection Technology Series,
EPA-600/2-75-065 (1975).
16. Harris, D. J., and Keffer, W. J., "Wastewater Sampling Methodologies
and Flow Measurement Techniques." USEPA, EPA 907/9-74-00$ (1974).
17. Butts, T. A., "Measurements of Sediment Oxygen Demand Characteristics
of the Upper I1linois. Waterway." Report of Investigation'76, ISWS-74-
R176, .Illinois State Water S;urvey, Urbana, IL (1974).
18. Shelley, P. E., "Design and Testing of a Prototype Automatic Sewer
Sampling System." USEPA Environmental Protection Technology-Series,
EPA-600/2-76-006 (1976).
19. "Methods for Chemical Analysis of Water and Wastes." USEPA'Environ-
.mental" Monitoring and Support Laboratory (formerly Methods development
and Quality Assurance Research Laboratory), EPA-625-/6-74-003 (1974).
20. "Model1 "State Water Monitoring Program." USEPA National Water Monitoring
Panel, EPA-440/9-74-002 (1975).
21. "Minimal Requirements for a Water Quality Assurance Program." USEPA,v
EPA-440/9-75-010 (1976).
22. "Handbook for Analytical Quality Control in Water and Wastewater
Laboratories." USEPA Environmental Monitoring and Support Laboratory
(formerly Analytical Quality Control Laboratory) (1972).
23. Langbein, W. B., and Harbeck, G. E., Jr., "A Note on Costs of Collecting
Hydrometric Flow Data in the U.S." HydAjotoQ^caJL Science* 8u£., 19,
pp 227-229 (1974).
24. "A Basic Water Monitoring Program," Standing Work Group on Water
Monitoring, EPA-440/9-76-025 (January 28, 1977).
D-142
------- |