EPA-450/2-78-019
May 1978 •*.-*-
(OAQPS No. 1.2-096)
GUIDELINE SERIES
AMBIENT MONITORING
GUIDELINES FOR
PREVENTION OF
SIGNIFICANT
DETERIORATION
(PSD)
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EPA-450/2-78-019
(OAQPS No. 1.2-096)
AMBIENT MONITORING GUIDELINES FOR
PREVENTION OF SIGNIFICANT DETERIORATION (PSD)
Monitoring and Data Analysis Division
Office of Air Quality Planning and Standards
and
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
May 1978
c:
-------
OAQPS GUIDELINE SERIES
The guideline series of reports is being issued by the Office
of Air Quality Planning and Standards (OAQPS) to provide information
to State and local air pollution control agencies for example, to
provide guidance on the acquisition and processing of air quality
data and on the planning and analysis requisite for the maintenance
of air quality. Reports published in this series will be available -
as supplies permit - from the Library Services Office (MD-35), U S
Environmental Protection Agency, Research Triangle Park, North
Carolina 27711, or, for a nominal fee, from the National Technical
Information Service, 5285 Port Royal Road, Springfield, Virginia 22161
Publication No. EPA-450/2-78-019
(OAQPS Guideline No. 1.2-096)
11
-------
FORWARD
Many individuals were involved in the preparation of this document
and should be contacted if any questions arise in the application of
the guideline.
Phone No.
Subject Area Contact (Area Code 919) FTS Number
Ambient Air Quality Alan Hoffman or 541-5351 629-5351
Monitoring David Lutz
Meteorological James Dicke or 541-5381 629-5381
Monitoring Laurence Budney 541-5381 629-5381
Quality Assurance Darryl VonLehmden 541-2415 629-2415
(Ambient Air Quality) or John Clements 541-2196 629-2196
PSD Policy and David Dunbar 541-5497 629-5497
Interpretation of
Regulations
------- An error occurred while trying to OCR this image.
-------
TABLE OF CONTENTS (Continued)
3.4.2 Sulfur Dioxide (SOj 21
3.4.2.1 Horizontal and Vertical Probe Placement 21
3.4.2.2 Spacing from Obstructions 21
3.4.3 Carbon Monoxide (CO) 22
3.4.3.1 Horizontal and Vertical Probe Placement 22
3.4.3.2 Spacing from Obstructions 22
3.4.3.3 Spacing from Roads 22
3.4.4 Photochemical Oxidant 23
3.4.4.1 Vertical and Horizontal Probe Placement 23
3.4.4.2 Spacing from Obstructions 23
3.4.4.3 Spacing from Roads 23
3.4.5 Nitrogen Dioxide (N0?) 23
3.4.5.1 Vertical ana Horizontal Probe Placement 23
3.4.5.2 Spacing from Obstructions 24
3.4.5.3 Spacing from Roads 24
3.4.6 Discussion and Summary 25
4. METEOROLOGICAL MONITORING 27
4.1 Data Required 27
4.2 Exposure of Meteorological Instruments 29
5. INSTRUMENTATION 33
5.1 Air Quality Instrumentation 33
5.2 Meteorological Instrumentation - Specifications 33
5.2.1 Wind Systems (Horizontal Wind) 33
5.2.2 Wind Systems (Vertical Wind) 34
5.2.3 Wind Fluctuations 34
5.2.4 Vertical Temperature Difference 34
5.2.5 Temperature 35
5.2.6 Humidity 35
5.2.7 Radiation - Solar and Terrestrial 35
5.2.8 Mixing Height 35
5.2.9 Precipitation 35
5.2.10 Visibility 36
6. QUALITY ASSURANCE FOR AIR QUALITY DATA 37
6.1 Introduction 37
6.2 Organization Quality Control Requirements 37
6.2.1 Activities to Demonstrate Within Control Conditions 38
for Continuous Methods
6.2.1.1 Reference or Equivalent Method Requirements 38
6.2.1.2 Calibration Requirements 38
6.2.1.3 Span Check Requirements 38
-------
TABLE OF CONTENTS (Continued)
Page
6.2.2 Activities to Demonstrate Within Control Conditions
for Integrated Sampling Methods 39
6.2.2.1 Total Suspended Particulate (TSP) Reference 39
Method (High-Volume Sampler)
t>.2.3 General Description of Activities to Assess 40
Monitoring Data Precision and Accuracy
6.2.4 Activities to Assess Monitoring Data Precision 40
and Accuracy for Continuous Methods
6.2.4.1 Assessment of Data for Precision 40
6.2.4.2 Assessment of Data for Accuracy 41
6.2.5 Activities to Assess Monitoring Data Precision 42
and Accuracy for Integrated Sampling Methods
6.2.5.1 Assessment of Data for Precision 42
6.2.5.2 Assessment of Data for Accuracy 42
6.3 Organization Calculation and Reporting Requirements for 43
Precision and Accuracy
6.3.1 Calculations for Continuous Methods 43
6.3.1.1 Single Instrument Precision 46
6.3.1 .2 Single Instrument Accuracy 43
6.3.2 Calculation for Integrated Sampling Methods 51
6.3.2.1 Single Instrument Precision for TSP 51
6.3.2.2 Single Instrument Accuracy for TSP 53
6.3.3 Organization Reporting Requirements 54
7. QUALITY ASSURANCE FOR METEOROLOGICAL DATA 56
8. DATA REPORTING ,„
bo
8.1 Air Quality Data Reporting 58
8.2 Meteorological Data Format and Reporting 58
9. REFERENCES 5g
VI
-------
1. INTRODUCTION
The Clean Air Act Amendments of 1977, Section 165, regarding
Prevention of Significant Deterioration (PSD), require that the
owner or operator of any potential source of air pollution over
a specified size, which is proposed to be constructed, must conduct
ambient air monitoring for a period of 1 year prior to submission
of the application for a permit to construct the facility. Less
than 1 year of data may be acceptable if it is determined that
complete and adequate analyses can be accomplished with data
collected over a shorter time interval. Continuous air quality
monitoring data must be used in the analyses.
The PSD reviews now include a comparision against the National
Ambient Air Quality Standards (NAAQS), and EPA is focusing the moni-
toring requirements on obtaining the necessary data for this purpose.
Therefore, only preconstruction monitoring will be required, but
existing air quality data may be used provided that it is represen-
tative of the geographic area of concern. The intent is not to
require extensive and costly monitoring programs by sources unless
it is absolutely necessary. If preliminary modeling or other data
indicate that the new source would not pose a threat to a NAAQS,
the source may not need to conduct preconstruction monitoring. The
need for preconstruction monitoring will be decided on a case-by-case
basis by the permit granting authority. The permit granting authority
is defined as the control agency that has the authority to issue new
source review permits.
Hence, it would be prudent (although not required) for the source
to contact the permit granting authority prior to conducting any pre-
construction monitoring. This preliminary review should result in a
determination as to whether it is in fact necessary to monitor. If
monitoring is needed, then the pollutants, locations of the monitors,
and duration of the sampling program can be defined to avoid permit
processing delays.
-------
No post-construction monitoring will be required since the
increment consumption will be determined by modeling. This is
because the accuracy of the measurement methods in relation to
the air quality increments and the year-to-year variability of
air quality data would severely limit the usefulness of the data
collected. The use of air quality monitoring to determine whether
the applicable increments are being maintained assumes that the
baseline air quality from which deterioration is measured is accurately
known. But such baseline data are generally not available, and in
most cases, can only be calculated indirectly through modeling
procedures. Also, the stack height provisions require that a
predicted air quality impact, based upon good engineering practice
stack height, be calculated using dispersion models in any case
where a source uses a stack that exceeds good engineering practice.
The purpose of this guideline is to define the requirements
concerning ambient air quality and meteorological data collection
in support of PSD applications and to ensure consistency among U.S.
Environmental Protection Agency (EPA) Regions and State air pollution
control agencies in terms of the type and amount of monitoring for an
adequate air quality analysis. The first section of this document
covers the general requirements and the exceptions to these require-
ments, and the remaining sections discuss technical considerations
involving the design of an ambient monitoring network, related quality
assurance activities, and collection and submission of data. It must
be emphasized that considerable judgment needs to be exercised in
applying this guidance due to the diversity arid complexity of factors
that affect air quality.
-------
2. GENERAL REQUIREMENTS
2.1 Sources Required to Consider Monitoring in Support of Permit Application
All major emitting facilities must consider monitoring for
a PSD permit application. A major emitting facility is defined
as any one of 28 source categories (Table 1) that have the potential
to emit 100 tons per year or more of any pollutant regulated under
the Clean Air Act. Any other source not in the list of 28 source
categories but having the potential to emit 250 tons per year or
more of any regulated pollutant is included in the definition of
major emitting facilities (this also includes categories 1, 9, 24
and 25 in Table 1 below the sizes indicated). These emission cutoffs
apply to both new sources and modifications to existing sources. Any
new source or modification which is subject to an air quality impact
analysis under the PSD regulations would need to conduct ambient
monitoring except under certain conditions discussed below.
The primary use of preconstruction monitoring is to determine
the status of the particular area (the source site and/or the area
of source impact) with respect to the NAAQS. After it is determined
that a source is subject to an air quality impact analysis, then the
screening procedures outlined in Volume 10 (revised) of the Guidelines
for Air Quality Maintenance Planning and Analysis should be used.
If a source is shown not to pose a threat to a NAAQS and is remote,
then no monitoring (ambient or meteorological) will be required.
Remote means that no significant sources exist within the air quality
impact range of the proposed new source. If a source does not pose
a threat to a NAAQS but is not remote, or if the source does pose a
threat to the NAAQS, then the air quality status of the area must be
determined. This can be done in several ways, i.e., by collecting
air quality and meteorological data prior to construction, by modeling
existing sources, or by using existing air quality and representative
meteorological data.
The extent to which modeled air quality estimates and/or existing
air quality and meteorological data can be used to establish the air
-------
4
Table 1. MAJOR EMITTING FACILITIES
================^^ _
D -4.. u *u finfd steam electn'c Plants of more than 250,000,000
British thermal units per hour heat input
2. Coal cleaning plants (with thermal dryers)
3. Kraft pulp mills
4. Portland cement plants
5. Primary zinc smelters
6. Iron and steel mills
7. Primary aluminum ore reduction plants
8. Primary copper smelters
9. Municipal incinerators capable of charging more than 250 tons of
refuse per day
10. Hydrofluoric acid plants
11. Sulfuric acid plants
12. Nitric acid plants
13. Petroleum refineries
14. Lime plants
15. Phosphate rock processing plants
16. Coke oven batteries
17. Sulfur recovery plants
18. Carbon black plants (furnace process)
19. Primary lead smelters
20. Fuel conversion plants
21. Sintering plants
22. Secondary metal production facilities
23. Chemical process plants
24. Fossil-fuel boilers of more than 250,000,000 British thermal units
per hour heat input
25. Petroleum storage and transfer facilities with a capacity exceeding
300,000 barrels
26. Taconite ore processing facilities
27. Glass fiber processing plants
28. Charcoal production facilities
-------
quality status of an area must by necessity be determined on a
case-by-case basis. If air quality data are required, then repre-
sentative meteorological data should be obtained as well.
2.2 Pollutants To Be Monitnrpd
Pollutants that will be increased by the quantity specified
m the PSD regulations and for which an air quality increment or
ceiling (ambient standard) exists must be monitored. This means
the monitoring requirements apply to total suspended particulates
(TSP), sulfur dioxide (so,,), carbon monoxide (CO), photochemical
oxidants (03), and nitrogen dioxide (N02). However, if it can be
established that a source will exceed the emission requirement
for only one pollutant, then only that pollutant need be monitored
Hydrocarbon monitoring is not required since the 0.24 ppm standard
is a guide for developing State Implementation Plarvs to attain the
03 ambient standard and since these emissions are the precursors
in the formation of ozone. Consequently, any source to be located
in an unclassified area for ozone and that exceeds the minimum hydro-
carbon emission limit will be required to monitor ozone and hydrocarbon
monitoring will not be required.
2.3 Minimum Number and Location of Monitoring Sites
The number and location of monitoring sites will be determined
on a case-by-case basis by the owner or operator and reviewed by
the permit granting authority. Consideration should be given to
the effects of existing sources, terrain, meteorological conditions,
existence of fugitive or reentrained dusts, etc. The number of sites
will be directly related to the expected spatial variability of the
pollutant in the area(s) of study. The suggested approach is to
first use appropriate dispersion modeling techniques to estimate
the air quality impact of the proposed source and any existing sources
within the impact range of the proposed source for each pollutant
averaging time. The modeled pollutant contribution of the proposed
-------
source should be analyzed in conjunction with the model results
for existing sources to'determine the general location of maximum
concentration as discussed in Section 3.3. Monitoring should then
be conducted in or as close to these areas as possible. Generally
one to four sites would cover most situations in multisource settings
In remote areas where the expected pollutant variability is small due
to a lack of existing sources, one monitoring site will generally
be sufficient. Further, the location of this site need not necessarily
be in the area of expected maximum pollutant concentration resulting
from the proposed source, i.e., it could be placed at the site of the
proposed facility.
For small sources, i.e., those sources for which the expected
air quality impact is not significant and which will be located in
remote areas, background air quality values can be used in lieu of
monitoring. This is discussed in more detail in Section 3.2.
2'4 Usin9 Existing^Y Quality Datajorjjgteorologlcal Data
If existing data are shown to be representative of the area
under consideration, they can be used to meet the requirements.
Such data might consist of:
1. Data collected by a State or local air pollution control agency.
2. Air quality data collected by a source under these requirements
provided the data are no older than 2 years at the time of
permit application and are considered representative of current
conditions. However, data older than 2 years may be used if
it is updated by the use of models.
3. TSP data being collected by a State agency or local agency
may be used if it is supplemented when necessary by the
source if sufficient sampling is not currently being conducted.
4. Meteorological data collected by the National Weather Service
or by a source under these requirements.
The existing data used for these purposes would not have to meet
the quality assurance requirements as discussed later. The permit
granting authority would have to determine if the data are valid
for determining the air quality status of the particular area.
-------
2.5 Approval of Monitoring Network Prior to Data Collection
Prior approval of the monitoring network is not required. However,
since network size and station locations are determined on a case-by-case
basis, it would be prudent for the owner or operator to seek review
of the network and overall monitoring plan from the permit granting
authority prior to collecting data in order to avoid delays in the
processing of the permit application. Further, the review should also
result in an elimination of unnecessary monitoring. Delays may result
from insufficient, inadequate, poor, or unknown quality data. Table 2
lists information that should be contained in a monitoring plan.
2.6 Quality Assurance
The minimum quality assurance requirements are discussed more
fully in Sections 6 and 7. In summary, the owner or operator will
be responsible for ensuring and documenting that these minimum quality
assurance practices are followed. Further, the audits of the monitors
must be independent. As a minimum, this means the audits must be
performed by someone other than the individuals operating the monitors.
Preferably, an organization other than the one operating the monitor(s)
should conduct these audits. Quality assurance data will be required
so that precision and accuracy of the air quality data can be established.
2.7 Duration of Monitoring
As required by the Clean Air Act, the monitoring must be conducted
for at least 1 year prior to the submission of the application to construct.
However, under some circumstances, less than 1 year of air quality
data may be acceptable. This will vary according to the pollutant
being studied. For TSP, S02, CO, and MA monitoring, less than a full
year will be acceptable if the applicant demonstrates through historical
data or dispersion models that the data are obtained during a time period
when maximum air quality levels can be expected. A minimum of 4
months will be required. Monitoring for ozone will be required for those
months in which the average daily maximum temperatures exceed 20°C
(68°F) in the area under study, or for the 4 months of the year with
the warmest average maximum temperatures for areas where there are not
-------
8
Table 2. MINIMUM CONTENTS OF MONITORING PLAN
- -
I. NETWORK DESCRIPTION
• Topographical description
• Land-use description
• Topographical map of proposed source and environs (including
location of existing sources)
t Climatological description
• Wind rose (if available)
• Location and rationale of air quality/meteorological monitors
II. MONITOR SITE DESCRIPTION
• Universal Transverse Mercator (UTM) coordinates
t Height of probe(s) above ground
§ Distance from obstructions (also indicate height of obstructions)
• Distance from other pollutant sources and roadways
• Photographs of each site (five photos: one in each cardinal
direction looking out from the probe and one closeup of the
probe intake)
III. MONITOR DESCRIPTION
• Manufacturer make and model number, principle of operation
• Age of instrument
0 Description of calibration system
IV. SAMPLING PROGRAM DESCRIPTION
• Time periods for which pollutants will be measured
• Discussion of the use of existing data or model results in
lieu of collecting ambient data
V. QUALITY ASSURANCE PROGRAM
• Calibration frequency
• Independent audit program
• Internal quality control procedures
• Data precision and accuracy calculation procedures
-------
at least 4 months with average maximum temperatures greater than
20°C (68°F). For calculating the daily average maximum temperatures,
a climatic record of at least 10 years should be used except where
no record of that length that can be considered representative of
the area exists.
2-8 Type of Air Quality Monitoring Equipment
All ambient air quality monitoring must be done with continuous
Reference or Equivalent Methods, with the exception of TSP for which
continuous Reference or Equivalent Methods do not exist. For TSP,
samples must be taken in accordance with the Reference Method.
The TSP Reference Method is described in 40 CFR 50. The
identification of continuous Reference or Equivalent Methods for
S02, CO, 03, and N02 which have been designated in accordance with
40 CFR 53, can be obtained by writing Environmental Monitoring and
Support Laboratory, Department E (MD-76), U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711.
2-9 Frequency of Sampling
For S02, CO, N02, 03 and meteorological parameters, continuous
analyzers must be used. Thus, continuous sampling (over the time
period determined necessary) is required. For TSP, daily sampling
(i.e., one sample every 24 hours) is required except in areas where
the applicant can demonstrate that significant pollutant variability
is not expected. In these situations, a sampling schedule less
frequent than every day would be permitted. However, a minimum of
one sample every 6 days will be required for these areas.
2JO Meteorological Parameters and Measurement Methods
At least 1 year of meteorological data should be available for input
to dispersion models used in analyzing the impact of the proposed new
source on ambient air quality and the analyses of effects on soils,
vegetation, and visibility in the vicinity of the proposed source.' In
some cases, representative data are available from sources such as the
National Weather Service. However, in many situations, on-site data
-------
10
collection will be required. Meteorological monitoring, instrumentation,
quality assurance and data reporting requirements are addressed
in Sections 4, 5, 7, and 8.
2.11 Data Reporting
The air quality and meteorological data must be reported
to the permit granting authority at the time of the permit
application, but could be done every 3 months. However the
actual reporting frequency should be based on an agreement
between the applicant and the permit granting authority.
The applicant should submit summaries of the air quality
and meteorological data in a form easily comparable to the
applicable averaging times of the increments and NAAQS. Using
S02 as an example, a frequency distribution of 3-hour and 24-hour
values, monthly arithmetic means, and the arithmetic mean for the
entire sampling period, along with the number of 1-hour observations
recorded, would be considered an appropriate data summary. In addition,
all raw air quality data (i.e., 1-hour values for continuous analyzers/
24-hour values for TSP) should be submitted in hard copy.
Upon request, all air quality data should be submitted in SAROAD
format. A site registration form must be submitted to the permit
granting authority prior to data submission. Data must be submitted
in machine-readable format, preferably on magnetic tape. Also, meteoro-
logical data for which valid SAROAD parameters and method codes exist
should be submitted in SAROAD format. The applicant must also retain
all strip charts, log books, and other records about the site and make
this available upon request.
The quality assurance data, including precision and accuracy
calculations, should be submitted for each site along with the sum-
marized air quality data.
2.12 Summary of Typical Network Design
Table 3 lists typical network sizes, site locations, and moni-
toring frequency and duration for a number of source categories. The
table is not intended to be all inclusive nor binding in all cases. It
is presented strictly for illustrative purposes.
------- An error occurred while trying to OCR this image.
-------
Table 3, EXAMPLE PSD MONITORING CASES (continued)
Source
Hydrocarbon
Emissions
remote
multi source
Fugitive Particu-
late Matter
Process Emissions
Stack Particulate
Matter
rural
urban
Pollutant
°3
°3
TSP
TSP
TSP
Number
of sites3
1
1
1
1
1 - 4
Length of
monitoring
4 mos - 8 mos
4 mos - 8 mos
4 mos - 1 yr
4 mos - 1 yr
4 mos - 1 yr
Monitor site
location'3
on site
downwind
on site
on site
Max impact/max cone
Frequency of sampling
continuous
continuous
1/6 to 1/1
1/6 to 1/1
1/6 to 1/1
aln remote areas generally one site is sufficient. In multisource situations or modifications to existing
sources, more sites may be necessary.
bln remote areas, on site monitoring is generally satisfactory. In multisource situations or modifications
to existing sources, areas of maximum impact or maximum concentrations should be monitored.
cl/6 means one 24-hour sample every 6 days.
1/1 means one 24-hour sample each day.
-------
13
3. AMBIENT AIR QUALITY MONITORING
3.1 Objectives and Data Uses
The basic objective of PSD monitoring is to determine the
effect emissions from a source are having or may have on the air
quality in any area that may be affected by the emission. The
data from the PSD monitoring may be used:
1. To classify some portion of an Air Quality Control Region
(AQCR) now unclassified by Section 107(d).
2. To determine the background air quality near the proposed
source site.
3. To determine the air quality status where the emissions from
the proposed source are expected to have the maximum impact.
4. To validate and refine models.
3.2 General Considerations
Monitoring for PSD purposes is not necessarily needed or required
for all situations. Potential emissions from a new source or a modi-
fication to an existing source must be determined. If these potential
emissions for any pollutant regulated by the Clean Air Act are less
than 100 tons per year for any of the 28 specified categories listed
in Table 1 or less than 250 tons per year for other sources, then no
further analyses or monitoring will be required. However, if the
potential emissions are increased by greater than 100 tons per year
for any of the specified categories listed in Table 1 or greater than
250 tons per year for other sources, then the applicant must consider
monitoring for 1 year unless specifically exempt by the PSD regulations,
The next step involves following the screening procedures in
Reference 1, especially in Section 4.2-4.5. If the source
is remote (no significant sources exist within the air quality impact
area of the proposed new source) and if no current representative moni-
toring data are available, then the following air quality levels can
be assumed as background and added to the screening procedure concen-
tration estimates.
-------
14
S02 20 yg/m3
CO 1 ppm
N02 .01 ppm
TSP 30 - 40 yg/m3
03 No values can be assumed. Any HC source or modification
covered under the PSD regulations and for which an ambient
review is needed, must conduct monitoring in any area which
is designated as unclassified in accordance with Section 107(d)
of the Clean Air Act.
These values can be used for all averaging times since in
remote areas (without any significant sources) the air quality should
be uniform throughout the year with the exception of minor fluctuations
due to meteorology and naturally occurring emissions. It should be
emphasized that a source is not required to use these values since
preconstruction monitoring is always allowed in order to establish
the background levels. If the sum of the screening procedure con-
centration estimates plus background does not pose a threat to a
NAAQS, then preconstruction monitoring (ambient and meteorological)
is not required.
If a source does not pose a threat to a NAAQS but is not remote,
or does pose a threat to the NAAQS, the air quality status of the
proposed source site and/or impact area must be determined. This
can be achieved by measuring or modeling the ambient air quality,
modeling, or using existing representative air quality data.
There are some exceptions to the 1-year monitoring requirement.
For example, the applicant may determine if available data or
calculated data can be used in lieu of monitoring. In such cases,
the applicant should explain the technical basis for doing this to
the permit granting authority. Also, the monitoring for ozone will
be required only during the seasons when high concentrations occur.
These exceptions and others have been discussed in Section 2, General
Requirements.
If it is determined that monitoring will be required, then network
design, probe siting criteria, meteorological monitoring, quality
assurance, and data reporting must be considered. These subjects are
discussed more fully in subsequent sections.
-------
15
3.3 Network Design
Many factors, such as topography, climatology, population,
and existing emissions sources will affect the design of a PSD
network. Two situations, i.e., urban or remote, must generally
be considered.
For urban or near-urban areas close to existing sources of
pollutants, more monitors will generally be needed because of the
variability in emissions and the resulting variability in air
quality concentrations. The existing sources in these areas and
their impacts on the population have to be considered along with
the averaging times for each pollutant.
For remote areas in which the permit granting authority has
determined there are no significant existing sources, a minimum
number of monitors would be needed, i.e., one or probably two
at most. Also, some concessions will be made on the location of
these monitors. Since the maximum impact from these sources would
be in remote areas, the monitors may be located, based on convenience
or accessibility, near the source rather than near the maximum impact
area. However, the maximum impact area is still the preferred location.
The ultimate network size for PSD purposes must be decided on
a case-by-case basis by the permit granting authority because of the
diverse factors that affect air quality. Some general procedures are
discussed below and should be followed in determining network size and
general station location. Additional guidance on general siting of
the monitor may be found in References 2-5, which discuss highest
concentration stations, isolated point sources, effects of topography,
etc. The guides presented here should be followed to the maximum
extent possible in developing the final PSD ambient monitoring network.
If the applicant has emission data for existing sources and
meteorological data for use in a recommended EPA dispersion model,
the air quality distribution before and after construction of the
source can be estimated as can the portion of the air quality increment
consumed by the operation of the source. This would provide sufficient
information for the applicant to place a monitor at (1) the location
-------
16
of the maximum concentration increase expected from the proposed
source, (2) the location of the maximum air pollutant concentration
from existing sources of emissions, (3) the maximum impact area
i.e., where the maximum pollutant concentration would hypothetically
occur based on the combined effect of existing sources and the
proposed new source. In some cases, two or more of these locations
may coincide and thereby reduce the number of monitoring stations.
In the case of SO,,, the area of maximum impact should be determined
for both the 3-hour and 24-hour averages. If these two areas do
not coincide, then additional monitors should be installed or the
monitor should be located in the area where the greatest impact
is more likely to occur.
If the applicant has no representative meteorological or
emission data for existing sources, then the emissions for the
proposed new source would be modeled with assumed meteorological
parameters typical of the proposed source location. Monitors would
be located at the modeled point(s) of maximum air pollutant concen-
tration increase, which would also be the maximum impact area A
meteorologist should be consulted prior to application of air quality
models when representative meteorological data are lacking and the
rationale for the assumed meteorological parameters should be provided
to the permit granting authority.
When industrial process fugitive particulate emissions are
involved, the applicant should locate a monitor at the proposed
source site. If stack emissions are also involved, a downwind location
should also be selected. For fugitive hydrocarbon emissions, the
applicant should locate a monitor downwind of the source at the point
of expected maximum ozone concentration contribution. This location
will be found downwind during conditions that are most conducive to
ozone formation, such as temperature above 20°C (68°F) and high solar
radiation intensity. For hydrocarbon emissions from a stack, the
applicant should also locate the monitor in the area of expected
maximum ozone concentration. For both fugitive and stack emissions,
-------
17
the selection of areas of highest ozone concentrations will require
wind speed and direction data for periods of photochemical activity.
Monitoring for ozone will only be necessary during the seasons when
high concentrations occur, i.e., those months when the average daily
maximum ambient temperature exceeds 20°C (68°F) (or as a minimum,
the 4 months with the highest daily average maximum temperature).
Since ozone is the result of a complex photochemical process,
the rate of movement across an area of the air mass containing
precursors should be considered. The distance from the proposed
source to the monitor for an urban situation should be about equal
to the distance traveled by the air moving for 6 to 8 hours at wind
speeds occurring during periods of photochemical activity. In an
urban situation, ozone formation over the initial few hours may be
suppressed by nitric oxide (NO) emissions. For a point source, the
NO interactions may be minimal, and the travel time to the expected
maximum ozone concentration may be 3 to 4 hours downwind.
3.4 Probe Siting Criteria
The desire for comparability in monitoring data requires adherence
to some consistent set of guidelines. Therefore, the probe siting
criteria discussed below must be followed to the maximum extent
possible to ensure uniform collection of air quality data that are
comparable and compatible.
It is recognized that there may be situations when the probe
siting criteria cannot be followed. If the site does not meet the
siting criteria, the differences must be thoroughly documented.
This documentation should help to avoid questions about the data
at a later date.
3.4.1 Total suspended particulates (TSP)
3.4.1.1 Vertical placement - The most desirable height for a TSP
monitor is near the breathing height. Practical considerations such
as prevention of vandalism, security, accessibility, availability of
-------
18
eli
~ :====•"
are the resuH of natural gas coition, no
;;;stTssry except for the iwiita-
' " - ir:r:;::;:d ri?;: s " r o11-
that the p,ume could reasonably be ex e ted „ "*
•j' ue expeccea to imoart nn tha
trudes above the sampler. There mint ai.n h
In at i=» t «. lnere must also be unrestricted airflow
n at l east three of the four cardina, (N, E, S, H, wind directjons
d at ,east one of the three must be the predominate direction for'
the season of greatest potential pollutant concentration.
3.4.1.3
... ^aciag-taiL^ . The Tsp mon1ton.ng Sltes
„; :.U?'y 1nf'UenCed ^ r°adW^ b— ^ concentrated
Plume of part,culate matter generated by vehicle traffic would have
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
21
3.4.2 Sulfur dioxide (SO,,)
3.4.2.1 Hnrl7nnt.a1 and vertical probe placement - As with TSP
monitoring, the most desirable height for an S0? monitor is
near the breathing height. Various factors enumerated before
may require that the inlet probe be elevated. Therefore, the
monitor must be located 3 to 15 meters above ground level.
If the inlet probe is located on a rooftop or an intermediate
height on a building, the inlet probe must be located less than 00
percent of the mean height of the buildings in the vicinity but at
least 3 meters above the ground level. Vertical uniform S0? distri-
bution up to at least the mean building height over the area of interest
in the city can be assumed except near the windward edge of a city
(see Reference 3). The choosing of the 30 percent criteria provides
a safety factor of 20 percent, which is intended to insure that the
effects of any nonuniformity in vertical S02 levels are minimized.
Thus, concentrations measured at or below this level should be similar
to those existing concentrations near the breathing zone (5 to 6 feet
above the ground). The inlet probe should be located on the windward
side of the building relative to the prevailing winter wind direction.
The inlet probe must also be located more than 1 meter vertically and
horizontally away from any supporting structure and also away from
dirty, dusty areas.
3.4.2.2 Spacing from obstructions - No furnace, incineration
flues, or other minor sources of S0? should be nearby. The separation
distance is dependent on the height of the flues, type of waste or fuel
burned, and the quality of the fuel. Inlet probes must be at least 1
meter from walls, parapets, penthouses, etc., if the inlet probe is
located on a roof or other structure.
The sampler must be placed more than 20 meters from trees and
must also be located away from obstacles and buildings. The distance
between the obstacles and the sampler must be at least twice the height
that the obstacle protrudes above the sampler. Airflow must also be
-------
22
unrestricted in at least three of the four cardinal wind directions,
and at least one of the three must be the dominant direction for the
season of greatest potential pollutant concentration. Additional
information on S02 probe siting criteria may be found in Reference 3.
3.4.3 Carbon monoxide (CO)
3-4-3-1 Horizontal and vertical probe placement - Because of
the importance of measuring population exposure to CO concentrations,
air should be sampled at average breathing heights. However, practical
factors require that the air sampler be higher. The required height of
the inlet probe for CO monitoring is therefore 3+1/2 meter, which is
a compromise between representative breathing height and prevention
of vandalism. The recommended 1 meter range of heights is also a
compromise to some extent. For consistency and comparability, it
would be desirable to have all inlets at exactly the same height, but
practical considerations often prevent this. Some reasonable range
must be specified and 1 meter provides adequate leeway to meet most
requirements. The inlet probe must be located more than 1 meter in the
vertical and horizontal direction from any supporting structure.
3-4.3.2 Spacing from obstructions - Airflow must also be unrestricted
in at least three of the four cardinal wind directions, and at least
one of the three must be the dominant direction for the season of
greatest potential pollutant concentration.
3.4.3.3 Spacing from roads - The CO monitoring sites for PSD
purposes should not be unduly influenced by any one line source
of CO. Therefore, the minimum setback from any roadway must be
35 meters for a CO monitoring site for PSD purposes. Additional
information on CO probe siting criteria may be found in Reference 4.
-------
23
3.4.4 Photochemical oxidant
3.4.4.1 Vertical and horizontal probe placement - The inlet probe
for photochemical oxidant monitors should be as close as possible
to the breathing zone. However, the complicating factors discussed
previously require that the probe be elevated. The inlet height
of the probe must be located 3 to 15 meters above ground level.
The probe must also be located more than 1 meter vertically and
horizontally away from any supporting structure.
3-4.4.2 Spacing from obstructions - The probe must be located
away from obstacles and buildings, and the distance between the
obstacles and the sampler probe must be at least twice the height
that the obstacle protrudes above the sampler. The probe must also
be located at least 20 meters from trees. Airflow must be unrestricted
in at least three of the four cardinal wind directions, and at least
one of the three must be in the predominate direction for the season
of greatest potential pollutant concentration.
3.4.4.3 Spacing from roads - It is important in the probe siting
process to minimize destructive interferences from sources of
nitric oxide (NO) since NO readily reacts with ozone. Table 4
provides the required minimum separation distances between roadways
and ozone monitoring sites. The minimum separation distance must
also be maintained between an ozone monitor and any other similar
volume of automotive traffic, such as parking lots. Additional
information on photochemical oxidant probe siting criteria may be
found in Reference 5.
3.4.5 Nitrogen dioxide (NOp)
3.4.5.1 Vertical and horizontal probe placement - The inlet height
of the N02 probe must be located 3 to 15 meters above the ground.
This is a compromise between measuring in the breathing zone and avoidance
-------
24
Table 4. MINIMUM SEPARATION DISTANCE BETWEEN
OZONE MONITORS AND ROADWAYS
Roadway average daily traffic,
vehicles per day
<1 ,000
1,000 to 10,000
>10,000
Minimum separation distance
from roadways to monitor, meters
20
20 - 250'
>250
Distance should be interpolated based on traffic flow
of vandalism, finding suitable sites, etc. For NO,,, the height does
not appear to be a critical factor since the N02 should be fairly
well mixed and somewhat uniform in the vertical direction. The
distance of the inlet probe from the supporting structure must be
greater than 1 meter vertically and horizontally.
3'4'5-2 Spacing from obstructions - Buildings, trees and other
obstacles may possibly scavenge NO,,. In order to avoid this kind
of potential interference, the monitor must be located well away
from such obstacles so that the distance between obstacles and the
sampler is at least twice the height that the obstacle protrudes
above the sampler. For similar reasons, a probe inlet along a
vertical wall is undesirable because air moving along that wall
may be subject to possible removal mechanisms. The inlet probe
must also be at least 20 meters from trees. There must be unrestricted
airflow in at least three of the four cardinal wind directions, and
at least one of the three must be the predominate direction for the
season of greatest pollutant concentration.
3'4'5-3 Spacing from roads, - It is important that the monitoring
site be removed from oxides of nitrogen sources to avoid measurements
being totally dominated by any one roadway and to allow time for
-------
25
conversion (reactions) of NO emission to NCL. Further, the
effects of roadway sources must be minimized by using separation
distances found in Table 5. The minimum separation distance must
also be maintained between a N02 monitor and any other similar
volume of automotive traffic such as parking lots. Additional
information on NCL probe siting criteria may be found in Reference 5.
Table 5. MINIMUM SEPARATION DISTANCE BETWEEN N09
MONITORS AND ROADWAYS '
Roadway average daily traffic,
vehicles per day
<1,000
1,000 to 10,000
>10,000
Minimum separation distance from
roadway to monitor, meters
20
20 - 250a
>250
Distance should be interpolated based on traffic flow.
3.4.6 Discussion and summary
Table 6 presents a summary of the requirements for probe
siting criteria with respect to distances and heights. It is
apparent from Table 6 that different elevation distances above the
ground are shown for the various pollutants. The discussion in the
text for each of the pollutants enumerated the factors why the monitor
must be elevated. The differences in the specified range of heights
are based on the vertical concentration gradients. For CO, the gradients
in the vertical direction are very large, so a small range of heights
has been specified. For S02, N02, TSP and 03 (except near roadways),
the vertical gradients are smaller and thus a larger range of heights
can be used. The upper limit of 15 meters was specified for consistency
between pollutants and to allow the use of a single manifold for moni-
toring more than one pollutant at a site.
------- An error occurred while trying to OCR this image.
-------
27
4. METEOROLOGICAL MONITORING
4.1 Data Required
The preconstruction review of proposed major emitting facilities
will require the use of meteorological data. It is essential that
such data be representative of atmospheric dispersion conditions
at the source and at locations where the source may have a signifi-
cant impact on air quality. The representativeness of the data
is dependent upon (1) the proximity of the meteorological monitoring
site to the area under consideration, (2) the complexity of the
topography of the area, (3) the exposure of the meteorological sensors,
and (4) the period of time during which the data are collected. More
guidance for determining representativeness is presented in Reference 16.
A data base representative of the site should consist of at
least the following data:
1. Hourly average wind speed and direction.
2. Hourly average atmospheric stability based on Pasquill
stability category or wind fluctuations (o"e), or vertical
temperature gradient.
3. Hourly surface temperature at standard height for climato-
logical comparisons and plume rise calculations.
4. Hourly precipitation amounts for climatological comparisons.
In addition, hourly average mixing heights may be necessary for
the air quality impact analysis. In most cases, this may be limited
to an extrapolation of twice-daily radiosonde measurements routinely
collected by the National Weather Service (NWS). Sections 4.2 and 5.2
contain specific information on instrument exposure and specifications.
Requirements for additional instrumentation and data will
depend upon the availability of information needed to assess the
effects of pollutant emissions on ambient air quality, soils, vegetation,
and visibility in the vicinity of the proposed source. The type, quantity,
-------
28
and format of the required meteorological data will also be influenced
by the input requirements of the dispersion modeling techniques used
in the air quality analysis. Any application of dispersion modeling
must be consistent with the EPA Guideline on Air Quality Models.17
The guideline makes specific recommendations concerning air quality
models and data bases. It also specifies those situations for which
models, data, and techniques other than those recommended therein
may be applied.
Site-specific data are always preferable to data collected
off-site. The availability of site-specific meteorological data
permits relatively detailed meteorological analyses and subsequent
improvement of dispersion model estimates. Off-site meteorological
data may be used in lieu of site-specific data only if it is agreed
by source owner and permit granting authority that the off-site
data are reasonably representative of atmospheric conditions in the
area under consideration. The off-site meteorological data can
sometimes be derived from routine measurements by NWS stations.
The data are available as individual observations and in summarized
form from the National Climatic Center, Federal Building, Asheville,
N.C. 28801. On the other hand, if the nearest source of off-site
data is considerably removed from the area under consideration,
and especially if there are significant terrain features, urban areas,
or large bodies of water nearby, it may be necessary that the required
meteorological data be site-specific.
In some cases, it will be necessary that data be collected at
more than one site in order to provide a reasonable representation
of atmospheric conditions over the entire area of concern. Atmospheric
conditions may vary considerably over the area. In some cases, (e.g.,
complex terrain) it will not be feasible to adequately monitor the
entire meteorological field of concern. Then the only recourse is to
site the stations in areas where characteristic and significant airflow
patterns are likely to be encountered. In any event, one of the
meteorological stations should be located so that it represents
atmospheric conditions in the immediate vicinity of the source.
-------
29
A minimum of 1 year of meteorological data should be available.
If more than 1 year of data is available, it is recommended that
such data be included in the analysis. Such a multiyear data
base allows for more comprehensive consideration of variations
in meteorological conditions that occur from year to year. A 5-year
period of record will usually yield an adequate meteorological data
base for considering such year-to-year variations.
In all cases, the meteorological data used must be of at least
the quality of data collected by the National Weather Service.
Desired features of instrumentation for collecting meteorological
data are discussed in Section 5.2.
4.2 Exposure of Meteorological Instruments
-Measurements of most meteorological parameters are affected by
the exposure of the sensor. To obtain comparable observations at
different sites, the exposures must be similar. Also, the exposure
should be such that the measured parameters provide a good repre-
sentation of pollutant transport and dispersion within the area that
the monitoring site is supposed to represent. For example, if wind
flow data over a fairly broad area are desired, the wind sensors
should be away from the immediate influence of trees and buildings,
steep slopes, ridges, cliffs, or hollows.
The standard exposure of wind instruments over level open
terrain is 10 meters above the ground. Open terrain is defined
as an area where the distance between the anemometer and any obstruc-
tion to the wind flow is at least five times the height of the
obstruction. Where a standard exposure is unobtainable at this
height, the anemometer should be installed at such a height that
its indications are reasonably unaffected by local obstructions
and represent as closely as possible what the wind at 10 meters
would be in the absence of the obstructions. Detailed guidance on
assessing adverse aerodynamic effects due to local obstructions is
contained in Reference 18.
-------
30
In locating wind sensors in rough terrain or valley situations,
it will be necessary to determine if local effects such as channeling,
slope and valley winds, etc., are important, or whether the flow
outside those zones of influence is to be measured. If the analysis
concerns emissions from a tall stack, it may be desirable to avoid
the local influences. On the other hand, if pollution from low-level
sources is the main concern, the local influences may be important.
If the source emission point is substantially above the standard
10-meter level for wind measurements, additional wind measurements
at the height of the emission point and at plume height are desirable.
Such measurements are used to determine the wind regime in which the
effluent plume is transported away from the source. (The wind speed
and direction 50 to 100 meters or more above the surface are often
considerably different than at the 10-meter level.) An instrumented
tower is the most common means of obtaining meteorological measure-
ments at several elevations in the lower part of the atmospheric
boundary layer. For wind instruments mounted on the side of a tower,
precautions must be taken to ensure that the wind measurements are
not unduly influenced by the tower. Turbulence in the immediate
wake of a tower (even a lattice-type tower) can be severe. Thus,
depending on the supporting structure, wind measuring equipment
should be mounted (e.g., on booms) at least one structure width
away from the structure, and two systems mounted on opposite sides
of the structure will sometimes be necessary. A wind instrument
mounted on top of a tower should be mounted at least one tower width
above the top. If there is no alternative to mounting instruments
19
on a stack, the increased turbulence problem must be explicitly
resolved to the satisfaction of the permit granting authority.
Atmospheric stability is another key factor in pollutant
dispersion downwind of a source. The stability category is a
function of static stability (related to temperature change with
-------
31
height), convective turbulence (caused by heating of the air at ground
level), and mechanical turbulence (a function of wind speed and
surface roughness). A procedure for estimating stability category
is given by Turner,20 which requires information on solar elevation
angle, cloud cover, ceiling height and wind speed. The hourly obser-
vations at NWS stations include cloud cover, ceiling height and wind
speed. Alternative procedures for estimating stability category
may be applied if representative data are available. For example,
stability category estimates may be based upon horizontal wind
direction fluctuations,21 or vertical gradients of temperature and
wind speed.22 To obtain a representative reading of the air tempera-
ture, the temperature sensor should be protected from thermal radiation
from the sun, sky, earth and any surrounding objects, and must be
adequately ventilated. Aspirated radiation shields are designed to
provide such protection. (Note that ambient temperature data are
also commonly required for plume rise estimates used in dispersion
model calculations.)
Mixing height is another parameter that can be important in
some cases. Mixing height is the distance above the ground to which
relatively free vertical mixing occurs in the atmosphere. For esti-
mating long-term average concentrations, it is adequate to use a
representative annual average mixing height.23 However, in many cases,
and especially for estimates of short-term concentrations, twice-daily
or hourly mixing height data are necessary. Such data can sometimes
be derived23 from representative surface temperatures and twice-daily
upper air soundings collected by selected NWS stations.
Precipitation collectors must be located so that obstructions do
not prevent the precipitation from falling into the collector opening
or force precipitation into the opening. Several collectors may
be required for adequate spatial resolution in complex topographic
regimes.
Visibility systems must be located to provide representative
measurements not only prior to construction of the facility, but also
for facility operational periods. Assessment of visibility impact is
-------
32
currently under study by EPA and other Federal agencies. Visibility
definitions, monitoring methods, data requirements for modeling
purposes and permit review requirements are to be the subjects of
a report to Congress due in February 1979. Since final regulations
for implementing Section 169 of the Clean Air Act will not be promul-
gated until August 1979, specific guidance cannot be given at this
time.
Additional information and guidance on siting and exposure of
meteorological instruments is contained in Reference 24.
-------
33
5. INSTRUMENTATION
5.1 Air Quality Instrumentation
Air quality monitoring for PSD (except for TSP) must use a
continuous designated Reference or Equivalent Method. The designated
methods have appeared in the Federal Register. However, a list
reflecting any new designations or any cancellations of a designation
currently in effect may be obtained from the Environmental Monitoring
and Support Laboratory, Department E (MD-76), U.S. Environmental
Protection Agency, Research Triangle Park, N.C. 27711.
5.2 Meteorological Instrumentation - Specifications
Meteorological instrumentation used for PSD monitoring must
yield reasonably accurate and precise data. Accuracies and allowable
errors are expressed in this section as absolute values for digital
systems; errors in analog systems may be 50 percent greater. For
example, an allowable error expressed as 5 percent means the recorded
value should be within +5 percent of the true value for digital systems,
and +7.5 percent for analog systems. Records should be dated, and
should be accurate to within 10 minutes. Wind speed and direction
(or vector components) should be recorded continuously on strip
charts. All variables may be recorded digitally or on multipoint
recorders at intervals not to exceed 60 seconds for a given variable.
These specifications apply to the meteorological instruments used to
gather the site specific data that will accompany a PSD permit appli-
cation. When the use of existing representative meteorological data
is approved by the permit granting authority, the instrumentation
*? R 9£
should meet, as a minimum, NWS standards. '
5.2.1 Wind systems (horizontal wind)
Wind direction and wind speed systems should exhibit a starting
threshold of less than 0.5 meter per second (m/s) wind speed (at 10
degrees deflection for direction vanes). Wind speed systems should be
accurate above the starting threshold to within 0.25 m/s at speeds
equal to or less than 5 m/s. At higher speeds, the error should not
exceed 5 percent of the observed speed (maximum error not to exceed
-------
34
2.5 m/s). The damping ratio of the wind vane should be between 0.4
and 0.65. Wind direction system errors should not exceed 3 degrees
from true 10-min or greater averages, including sensor orientation
errors. Wind vane orientation procedures should be documented.
5.2.2 Wind systems (vertical wind)
In complex terrain, downwash of plumes due to significant
terrain relief may pose a problem. If such a problem potentially
exists, it may be necessary to measure the vertical component of
the wind at the proposed site, and as close as possible to stack
height. The starting threshold for the vertical wind speed component
should be less than 0.25 m/s. Required accuracy for the vertical
speed component is as specified in Section 5.2.1 for horizontal
speeds.
5.2.3 Wind fluctuations
Determination of the on-site measured standard deviation
of wind fluctuations, or derived standard deviations of cross-plume
concentrations, may be necessary if dispersion parameters are being
developed for use at a specific site. Since the analytical framework
within which such wind fluctuation measurements/statistics are to
be incorporated is expected to be unique or applied on a case-by-case
basis, no general requirements regarding specifications are outlined
in this guideline. Considerable care is required in the selection of
wind instruments and data logging systems, especially in the choice
of sampling and averaging times. Thus, response characteristics
of wind sensors are especially critical. ' Owners or operators
designing programs incorporating these capabilities should submit
a statement from a qualified consultant identifying the adequacy
of such wind system(s) within the context of the overall PSD ambient
monitoring program.
5.2.4 Vertical temperature difference
Errors in measured temperature difference should not exceed
0.003 °C/m.
-------
35
5.2.5 Temperature
Errors in temperature should not exceed 0.5°C if fog
formation, icing, etc., due to water spray or water vapor emitted
from the facility may be a problem. Otherwise, errors should not
exceed 1.0°C.
5.2.6 Humidity
Atmospheric humidity can be measured and expressed in several
ways. Error in the selected measurement technique should not exceed
an equivalent dewpoint temperature error of 0.5CC if the potential
exists for fog formation, icing, etc., due to effluents from the
proposed facility. Otherwise, errors in the equivalent dewpoint
temperature should not exceed 1.0°C.
5.2.7 Radiation - solar and terrestrial
The determination of Pasquill stability class may be based
on whether the solar radiation is termed strong, moderate, or slight.
Stability class can be determined from sun elevation and the presence,
20
height, and amount of clouds, or by using a pyranometer and/or
net radiometer during the daytime and a net radiometer at night.
Such radiation-to-stability relationships are expected to be site-
specific, and the responsibility for demonstrating their accuracy
lies with the permit applicant. General accuracy for pyranometers
and net radiometers used in a PSD monitoring network is expected to
be +5 percent.
5.2.8 Mixing height
Mixing height data may be derived from NWS upper air data.
If available data are determined to be inappropriate by the permit
granting authority, such data may be obtained on-site by the permit
29
applicant. The instrument system to be used is not specified in
this guideline, but its precision and resolution should not exceed
OC OC.
the limits associated with NWS radiosonde systems. '
------- An error occurred while trying to OCR this image.
-------
37
6. QUALITY ASSURANCE FOR AIR QUALITY DATA
6.1 Introduction
The activities described here are the minimum quality assurance
requirements for an organization operating a PSD network to produce
monitoring data acceptable to the permit granting authority. If an
organization has an ongoing quality assurance program more extensive
than the one described here, EPA encourages that the program continue
at that level. Likewise, if an organization wishes to develop a quality
assurance program more extensive than the one required, EPA encourages
that,effort.
Although quality control charts are not a requirement, their use
to document the activities to demonstrate within-control conditions as
described in Section 6.2.1 is strongly encouraged. This recommendation
is made because the control limits given for the activities of Section
6.2.1 are liberal and the development of more restrictive control limits
by each organization is encouraged. Quality control charts provide a
convenient and reliable means to determine when more restrictive control
limits can be reasonably imposed. The construction and use of quality
control charts is discussed in detail in Appendix H of Reference 31, and
this source can be consulted for further information.
Frequent use is made of the term "organization." For purposes here,
"organization" is defined as a source owner/operator, a government agency,
or a contractor who operates an ambient air pollution monitoring network
for PSD purposes.
6.2 Organization Quality Control Requirements
Each organization responsible for operating a PSD network must have
a quality control program that includes the following two items: (1)
activities to demonstrate that measurements are made within acceptable
control conditions, and (2) assessment of organization's monitoring data
for precision and accuracy.
-------
38
6-2-1 Activities to demonstrate within control conditions for continuous
methods
6'2-1-1 Reference or equivalent method requirements - All continuous
analyzers must be EPA-designated Reference or Equivalent Methods. A
list of Reference and Equivalent Methods can be obtained by writing to:
Environmental Monitoring and Support Laboratory, Department E (MD-76),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711.
6.2.1.2 CaTjbratjon requirements - Continuous analyzers must be calibrated
during installation and recalibrated whenever any one of the following
conditions occurs: (1) control limit is exceeded for the span check de-
scribed in Section 6.2.1.3, (2) after repair of a malfunctioning analyzer,
and (3) after replacement of major component(s) of an analyzer. Zero plus a
minimum of five calibration points equally spaced over the analyzer range
must be used to generate a calibration curve.
All gaseous calibration standards must be traceable to National
Bureau of Standards,Standard Reference Materials (SRM). The following
SRM must be used to establish traceability of gaseous calibration standards:
(1) S02 calibration standards require traceability to SRM S02 permeation
tubes, (2) NO calibration standards require traceability to SRM NO in
nitrogen, (3) N02 calibration standards require traceability to SRM N02
permeation tubes, and (4) CO calibration standards require traceability
to SRM CO in air. One acceptable protocol to demonstrate traceability
of calibration standards to SRM can be obtained by writing to: Environmental
Monitoring and Support Laboratory, Quality Assurance Branch (MD-77), U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711.
6-2-1-3 Span check requirements - For continuous analyzers, a periodic span
check is used to demonstrate within-control conditions and also to assess
the data for precision. A one-point span check.must be carried out at
-------
39
least once per week on each analyzer at the following analyzer con-
centrations: (1) for S02, NC>2, and 03 analyzers, between 0.08 and
0.10 ppm, and (2) for CO analyzers, between 8 and 10 ppm. In each
case, determine the difference between the known span concentration
and the measured concentration from the existing calibration curve.
The acceptable limits for these differences are as follows: (1) for
S02, N02, and 03 analyzers the limit is +0.025 ppm, and (2) for CO
analyzers +2.0 ppm.
Any time the control limit is exceeded, the analyzer must be removed
from ambient monitoring, checks made to determine assignable cause(s),
and corrective action taken before return of the analyzer to ambient
mointoring. After return of the analyzer to ambient monitoring, the re-
calibration of the analyzer with zero plus a minimum of five calibration
points equally spaced over the analyzer range is required.
The weekly span check data also generate information to assess the
precision of the monitoring data; see Section 6.3.1.1 for procedures to
be followed for calculating and reporting precision.
6.2.2 Activities to demonstrate within control conditions for integrated
sampling methods
6.2.2.1 Total suspended particulate (TSP) reference method (high-volume
sampler) - (a) Sampling flow rate check - Initial flow rate
readings must be observed for each sampler and used to determine when
corrective action is needed. The initial flow rate must be within +15
percent of the established average initial flow rate. If this limit is
exceeded, recalibration or other corrective action is required. See
Section 2.2.4, pages 10 and 11, Reference 32, for details.
(b) Exposed filter reweighing - A sampler of randomly selected ex-
posed filters (filters after sampling) is reweighed. Agreement between
original and repeat weights must be +5.0 mg or less if the original weights
of the entire lot.of exposed filters are to be accepted. If any reweighed
exposed filter differs more than 5.0 mg from its original weight, the entire
-------
40
lot of filters must be reweighed. See Part 8.1.3, Section 2.2.8,
Reference 32,for details.
(c) Recalculation of sample concentration - A sample of randomly
selected calculated TSP values (in yg/m3) is recalculated. The original
and repeat calculations must agree. If this criterion is not met, all
calculations for the lot are checked and corrected as necessary. See
Part 8.1.4, Section 2.2.8, Reference 32, for details.
6'2'3 General description_gf activities to assess monitoring data precision
and accuracy -Descriptions of specific activities to assess
monitoring data for precision and accuracy of individual instruments follow
in Sections 6.2.4 and 6.2.5. Calculation and reporting procedures for
precision and accuracy are described in Section 6.3.
Precision is calculated from periodic checks made by the routine
operator/analyst during the normal operation of the method. Accuracy is
calculated from audits. Audits, as used here, mean independent assessments
of the accuracy of the data obtained in an organization's monitoring
program. Independence is achieved by using personnel, audit standards,
and equipment different from those routinely used during method calibration.
As a minimum, the audit personnel must be different than the operator/analyst
conducting the routine operation of the measurement method. It is preferred,
but not required, that a different organization than the one operating the
measurement method conduct the audit.
6-2-4 Actvities to assess monitoring data precision and accuracy for
continuous methods
6<2-4-1 Assessment of data for precision - Periodic span checks are used
to assess continuous monitoring data for precision. A one-point span check
must be carried out each week on continuous analyzers used to measure S02,
N02, 03, and CO. The concentration for this span check is described in
Section 6.2.1.3. The concentration of the span gas used is the generated
-------
41
concentration (X), and the analyzer response is the measured response (Y)
in the calculations shown in Section 6.3.1.1. All continuous analyzers
must have this span check performed at least weekly.
6.2.4.2 Assessment of data, for accuracy - Periodic independent audits
are used to assess continuous monitoring data for accuracy. Continuous
analyzers that measure S02, N02, Og, and CO are all audited in the same
way. The audit described here applies equally to all designated contin-
uous Reference and Equivalent Methods. The audit required is the periodic
challenge of the continuous analyzers with known concentrations of
pollutant gas at five levels. The results of the audit are used in
assessing the accuracy of monitoring data. The procedure for calcula-
tion of accuracy is described in Section 6.3.1.2.
For each pollutant monitored, it is necessary to develop a means to
provide the audit pollutant gas at five levels. The audit pollutant gases
can be prepared in a manner similar to the preparation of the usual cali-
bration gases. It is required, however, that the concentrations of pollutant
gases in the audit gases be determined independently and that the standards
used to calibrate the analyzers being audited not be used in the analysis
of the audit gases. All audit gas standards must be traceable to National
Bureau of Standards, Standard Reference Materials (SRM) of the same type
described in Section 6.2.1.2 for calibration standards. One acceptable
protocol to demonstrate tracebility of audit gas standards to SRM is re-
ferenced in Section 6.2.1.2 .
The specific requirements for auditing continuous analyzers follow:
once per sampling quarter, each Reference or Equivalent analyzer must be
audited. A five-point audit must be carried out on each analyzer. The
five-point audit must be carried out at the following analyzer concentrations:
(1) for S02, N02,and 03 analyzers, 0.05 +_ 0.01 ppm, 0.10 + 0.01 ppm, 0.20 +_
0.01 ppm, 0.30 Hh 0.01 ppm, and 0.45 ±0.01 ppm, and (2) for CO analyzers
5 + 1 ppm, 10 + 1 ppm, 20 + 1 ppm, 30 + 1 ppm, and 45 +_ 1 ppm. The
difference in concentrations (ppm) between the audit values and the
measured analyzer values are used to calculate accuracy as described in
Section 6.3.1.2.
-------
42
6'2'5 fotivUies to assess monitoring data precision and accuracy far
integrated
6-2>5J Assessment of data_for^precision - Collocated samplers are used
to assess data from integrated sampling methods described in Section 6.2 2
For a given organization's monitoring network, one sampling site must have'
collocated samplers. A site with the highest expected 24-hour pollutant
concentration must be selected. The two samplers must be located at the
same elevation and should be approximately 3 meters apart to preclude
airflow interference. High-volume samplers should have the same roof
orientation. For the pair of collocated samplers, one sampler must be
randomly designated prior to the collection of samples as the official
sampler for normal routine monitoring purposes and the other will be de-
signated as the duplicate sampler. Once designated, the identity of each
sampler must be maintained. Sampling, calibration, operation, and analysis
must be the same for both collocated samplers. The collocated sampler
must be operated as a minimum every third day when continous sampling
is used. When a less frequent sampling schedule is used, as discussed
in Section 2.9, the collocated sampler must be operated at least once
each week. The difference in concentration (yg/m3) between the official
and duplicate samplers is used to calculate precision as described in
Section 6.3.2.1.
6.2.5.2 Assessment of data for accuracy - The assessment of accuracy for
integrated sampling methods is made by auditing a portion of the measurement
process.
(a) TSP method- the portion of the TSP method audited is the flow rate
during sample collection. The specific requirements for auditing the flow
rate of high-volume samplers follow: once per sampling quarter, audit the
flow rate of each high-volume sampler. The flow rate calibration audit is
-------
43
conducted using a reference flow device described in Part 8.1.2, Section
2.2.8, Reference 31. The reference flow device is an orifice having five
different calibrated resistance plates that are mounted on the faceplate
of the high-volume sampler. In conducting the audit, one of the resistance
plates is selected, and the audit flow rate is measured. This procedure
is repeated, using different resistance plates, until the flow rate at five
different levels is obtained. The differences in flow rates (m /min) between
the calibrated resistance plate values and the measured sampler values are
used to calculate accuracy as described in Section 6.3.2.2.
6.3 Organization calculation and reporting requirements for precision
and accuracy
Calculating and reporting estimates of precision and accuracy of
ambient air quality data gathered by organizations operating networks are
described in this section. Precision and accuracy estimates obtained
will be conservative. For example, span drift alone is used to estimate
precision of continuous monitoring data, yet it is known that there are
other sources of imprecision. Also, the accuracy of monitoring data from
integrated sampling is estimated from audits of only a portion of the total
measurement process, namely, flow measurements for TSP.
This section describes the calculation procedures for precision and
accuracy for continuous methods and the TSP integrated sampling method.
Precision and accuracy must be calculated each sampling quarter for each
measurement instrument using information obtained in Section 6.2.3. The
computed precision and accuracy must be recorded on the Data Assessment
Report (Table 7) and adjoined with the quarterly submission of monitoring
data.
6.3.1 Calculations for continuous methods
Calculations for precision and accuracy of individual continuous
instruments are described below.
------- An error occurred while trying to OCR this image.
-------
Table 7. DATA ASSESSMENT REPORT FOR PSD AIR QUALITY DATA - PART II
Organization name:
Organization address:
Reporting period: From
to
month/day/year month/day/year
Audit Results for Continuous Analyzers (CO, NO,,, 03, S02)
Pollutant
Analyzer
identification
code
Concentration, ppm
Point 1
Known
Diff.*
Point 2
Known
Diff.*
Point 3
Known
Diff.*
Point 4
Known
Diff.*
Point 5
Known
Dirt.*
en
Calculate audit difference using the equation below. Show + or - sign when recording difference.
Diff. = measured cone. - known cone.
-------
46
6.3.1.1 Single Instrument precision - Estimates of precision for ambient
air quality measurements from continuous methods are calculated from the
results of weekly span checks as described in Section 6.2.1.3. Each
organization, at the end of each sampling quarter, calculates and reports
a precision probability interval for each reference or equivalent analyzer.
Directions for calculations are given below, and for reporting are given
in Section 6.3.3. If monitoring data are invalidated because span drifts
exceeded the limits specified in Section 6.2.1.3, DO NOT include these
span drift results in the calculations of estimates of precision given
below for single-instrument precision.
Let Xj represent the known concentration of the span gas and Y-
represent the measured concentration from the weekly span check. Cal-
culate the percentage difference (dj) for each span check using Equation
1.
rf Yi " xi
di = ~ Lx 100
Xi (1)
For each instrument, calculate the quarterly average (dj), Equation
2, and the standard deviation (S,-), Equation 3
J
d. = z d./n (2)
S, =
n "1/2
9 9
V - (S=1 d.JVn
n-1
(3)
where n is the number of span check of the instrument made during the
sampling quarter. For example, n is 13 if span checks are made once
during each of the 13 weeks in a quarter.
------- An error occurred while trying to OCR this image.
-------
48
Applying Equations 2 and 3:
3. = +3/13 = +0.2%
J
S. =
_|99 - (3)2/13
12
1/2
Applying Equations 4 and 5:
Upper limit = 0.2 + 1.96(2.9) = 5.9 or +6%
Lower limit = 0.2 - 1.96(2.9) = -5.5 or -6%
On the Data Assessment Report (Table 8), -6 is reported for the lower
95 percent limit and +6 is reported for the upper 95 percent limit
for the CO analyzer precision.
6.3.1.2 Single instrument accuracy - Estimates of accuracy for ambient
air quality measurements from continuous methods are calculated from
the results of independent audits as described in Section 6.2.4.2.
Each organization, at the end of each sampling quarter, calculates
and reports an accuracy probability interval for each reference or
equivalent analyzer. Directions for calculation are given below
and directions for reporting are given in Section 6.3.3.
Let X. represent the known concentration of pollutant in
"hh
the audit gas for the i audit point and Y. represent the corresponding
j_U I
measured value of the i audit point. Calculate the percentage dif-
ference (d.) for each audit point using Equation 1. For each quarterly
audit, calculate the average percentage difference (d.) and the standard
deviation (S.) using Equations 2 and 3, respectively. Calculate the
\J
95 percent probability limits for accuracy using Equations 4 and 5.
As an example, consider the following data from the audit of
an N02 analyzer:
Audit cone., Y-, cone., X., Difference, d.,
point ppm n ppm ~* percent
Measured
cone., Y -,
ppm
0.053
0.107
Known
cone., X-.
ppm ^
0.051
0.102
1 0.053 0.051 3.9
2 0.107 0.102 4.9
3 0.212 0.204 3.9
4 0.321 0.307 4.6
5 0.480 0.460 4.3
z = +21.6
------- An error occurred while trying to OCR this image.
-------
Tables. DATA ASSESSMENT REPORT FOR PSD AIR QUALITY DATA - PART II (Sample Entries)
Organization name: CXCxyyig—-- - 1 OruX^. i ~-k&xX-X _
Organization address:
Reporting period: From
5'LI. O
ZJK.
pZ
to
1
27 5Q 3,
month/day/year month/day /year
Audit Results for Continuous Analyzers (CO, NO-, 0,, S0?)
Pollutant
Analyzer
identification
code
l\i CA, -S .
c*. -*~s I
l^ ^
Concentration, pjDm
Point 1
Known
,05/
Diff.*
4-. 00^
Point 2
Known
. /C<2
Diff.*
f- ODjT
'
Point 3
Known
.3ci
Diff.*
+-.cet
Point 4
Known
• '3^7
Diff.*
t-.O/^
Point 5
Known
Diff.*
^cwc>
i
en
o
*Calculate audit difference using the equation below. Show + or - sign when recording difference.
Diff. = measured cone. - known cone.
-------
51
Applying Equations 2 and 3:
3. = +21.6/5 = 4.3%
J
, _ [94.08 - (21.6)2/5
'J L 4
Applying Equations 4 and 5:
- 0.44%
Upper limit = 4.3 + 1.96(0.44) = 5.2 or +5%
Lower limit = 4.3 - 1.96(0.44) = 3.4 or +3%
On the Data Assessment Report (Table 8), +3 is reported for the
lower 95 percent limit and +5 percent is reported for the upper 95 percent
limit for the N02 analyzer accuracy.
6.3.2 Calculation for integrated sampling methods
Calculation methods for precision and accuracy of individual
integrated samplers are described below:
6.3.2.1 Single instrument precision for TSP - Estimates of precision
for ambient air quality measurements from the TSP method are calculated
from results obtained from the collocation of two samplers at one
sampling site as described in Section 6.2.5.1. Each organization,
at the end of each sampling quarter, calculates and reports a precision
probability interval using weekly collocation sampler results. The
calculated precision at the one sampling site is considered indicative
of the precision at all sampling sites for the TSP method. Directions
for calculation are given below, and directions for reporting are given
in Section 6.3.3.
Using the paired measurements described in Section 6.2.5.1,
let Y. represent the concentration of pollutant measured by the
duplicate integrated sampler and X.. represent the calculation of
pollutant measured by the designated official sampler during the
ith sampling period. Calculate the percentage difference (d^) using
------- An error occurred while trying to OCR this image.
-------
53
Applying Equations. 2 and 3:
f31/12
217 -
3. = +31/12 - 2.6%
J o
__ 3
11
Applying Equations 6 and 7:
Upper limit = 2.6 + 1.96(3.5)//F = 7.4 or 7%
Lower limit - 2.6 - 1.96(3.5)//T - -2.3 or -2%
On the Data Assessment Report (Table 3), -2 is reported
for the lower 95 percent limit and +7 is reported for the upper
95 percent limit for the TSP sampler precision.
6.3.2.2 Single instrument accuracy for TSP - Estimates of accuracy
for ambient air quality measurements from the TSP method are calculated
from the results of independent audits as described in Section 6.2.5.2.
Once each sampling quarter, the flow rate of each high-volume sampler
is audited. Each organization, at the end of each sampling quarter,
calculates and reports and accuracy probability interval for each TSP
sampler. Directions for calculation are given below and directions
for reporting are given in Section 6.3.3.
Let X. represent the known flow rate for each resistance plate,
and Yi represent the measured flow rate. Calculate the percentage
difference (d..) for each flow rate using Equation 1. For each quarterly
audit, calculate the average percent difference (3.) and the standard
deviation (S.) using Equations 2 and 3, respectively. Calculate the
J
95 percent probability limits for accuracy using Equations 4 and 5.
As an example, consider the following data from the audit
of a TSP sampler:
-------
54
Measured
flow rate, Y
rn3 /min (cfm)
Resistance
plate no.
13
10
7
18
5
Apply Equations 2 and 3:
3.= +26/5 = 5.2%
J
158 - (26)
Known
flow rate, X.,
m /min (cfm)
Difference, d->
• percent 1
1.53 (53.9)
1.39 (49.1)
1.21 (42.8)
1.73 (61.0)
0.99 (34.9)
1.47 (52.0)
1.34 (47.2)
1.11 (39.1)
1.68 (59.3)
0.93 (32.8)
4
4
9
3
6
Z = +26
= 2.4%
Apply Equations 4 and 5:
Upper limit = 5.2 + 1.96(2.4) = 9.9 or 10%
Lower limit = 5.2 - 1.96(2.4) = 0.5 or 1%
On the Data Assessment Report (Table 8), +1 is reported for the
lower 95 percent limit and +10 is reported for the upper 95 percent
limit for TSP sampler accuracy.
6.3.3 Organization reporting requirements - At the end of each sampling
quarter, the organization must calculate precision and accuracy for
each continuous analyzer and each TSP sampler, and record the calculated
results on Part I of the Data Assessment Report (Table 7). In addition,
for each audit of a continuous analyzer (CO, NO , 0,, and SO,,), record
C. 3 C-
the five audit concentrations and the difference between the measured
concentration and the known (audit) concentration on Part II of the
Data Assessment Report (Table 7). The quarterly Data Assessment Report
-------
55
must be submitted with the air monitoring data. All data used to
calculate reported estimates of precision and accuracy, including
span checks, collocated sampler results, and audit results, must
be made available to the permit granting authority upon request.
The results from the example calculations in Sections 6.3.1
and 6.3.2 have been recorded on Table 8 to illustrate how to complete
the Data Assessment Report. A blank copy of the Data Assessment Report
is included as Table 7. This blank report may be copied and used by
the organization submitting estimates of precision and accuracy for
air monitoring data.
-------
56
7. QUALITY ASSURANCE FOR METEOROLOGICAL DATA
New equipment requires only the field checkout and calibration
procedures recommended by the manufacturer. Used equipment should
receive an appropriate examination (overhaul if necessary) and
calibration prior to initial installation to assure the acquisition
of the maximum amount of usable data within the error limits specified
herein. Inspection, servicing and calibration of equipment must be
scheduled throughout the measurement program at appropriate intervals
to assure at least 90 percent data retrieval for each variable measured.
In addition, the joint frequency for the recovery of wind and stability
data should not fall below 90 percent on an annual basis; missing
data periods must not show marked correlation with the various
meteorological cycles.
Calibration of systems should be accomplished no less frequently
than once every 6 months. In corrosive or dusty areas, the interval
should be reduced to assure adequate and valid data acquisition.
If satisfactory calibration of a measuring system can be
provided only by the manufacturer or in special laboratories, such
as wind-tunnel facilities, arrangements should be made for such
calibrations prior to acquisition of the equipment. A parts inventory
should be maintained at a readily accessible location to minimize
delays in restoring operations after system failures.
An independent meteorological audit (by other than one who
might be retained by the owner to install and operate the network)
should be performed to provide an onsite evaluation of (1) the network
installation, (2) inspection, maintenance and calibration procedures
and logging thereof, (3) data reduction procedures, including spot
checking of data, and (4) data logging and tabulation procedures.
The onsite visit (requiring as little as 1 day in many cases) should
be made within 60 days after the network is first in full operation,
and a written audit/evaluation should be provided to the owner. This
report should be retained by the owner. Any problems should be corrected
-------
57
and duly noted as to action taken in an addendum to the audit
report. A reproducible copy of the audit report and the addendum
should be furnished with the source construction permit application,
Such audit-evaluation by an independent meteorological
consultant should be performed about each 6 months. The last
such inspection should be made no more than 30 days prior to the
termination of the measurement program, and while the measurement
operation is in progress.
-------
58
8. DATA REPORTING
8.1 Air Quality Data Reporting
Data generated from the PSD monitoring, as well as the quality
assurance data, should be submitted to the permit granting authority.
This could be done on a quarterly basis. The applicant and the
permit granting authority should have a prior agreement as to the
format, procedure, and schedule for the data submission. The permit
application would include the fourth quarter data as well as a
summary of data collected during the previous 12 months. The periodic
submission of data is intended to identify any problems in the data
as they occur and avoid delays in processing the application due to
inadequate or poor quality data. For continuous analyzers, at least
80 percent of individual hourly values are expected to be reported
for the time period for which the sampling is performed. For manual
methods (TSP), 80 percent of individual 24-hour values should be
reported. See Section 2.11 for other considerations.
8•2 Meteorological Data Format and Reporting
Because of the different data requirements for different types
of analyses that might be used to evaluate various facilities, there
is no fixed format that applies to all data sets. However, a generali-
zation can be made:-all meteorological parameters must be collated in
chronological order and tabulated according to observation time, and
be furnished to the permit granting authority upon request. All units
should be in the SI system (International System of Units). All input
data (in the format required by the analytical procedures selected)
used in, and all results of, the air quality analyses must be furnished
to the permit granting authority upon request.
-------
59
9. REFERENCES
1. Budney, L.J., Guidelines for Air Quality Maintenance Planning and
Analysis Volume 10 (Revised): Procedures for Evaluating Air Quality
Impact of New Stationary Sources. U.S. Environmental Protection
Agency, Research Triangle Park, NC. OAQPS No. 1.2-029, EPA
Publication No. EPA-450/4-77-001. October 1977.
2. Ludwig, F.L., J.H. Kealoha, and E. Shelar. Selecting Sites for
Monitoring Total Suspended Particulates. Stanford Research Institute,
Menlo Park, CA. Prepared for U,S. Environmental Protection Agency,
Research Triangel Park, NC. EPA Publication No. EPA-450/3-/7-018.
June 1977, revised December 1977.
3. Ball, R.J. and G.E. Anderson. Optimum Site Exposure Criteria for
S0~ Monitoring. The Center for the Environment and Man, Inc.,
Hartford, CT. Prepared for U.S. Environmental Protection Agency,
Research Triangle Park, NC. EPA Publication No. EPA-450/3-77-013.
April 1977.
4. Ludwig, F.L. and J.H.S. Kealoha. Selecting Sites for Carbon Monoxide
Monitoring. Stanford Research Institute, Menlo Park, CA.
Prepared for U.S. Environmental Protection Agency, Research Triangle
Park, NC. EPA Publication No. EPA-450/3-75-077. September 1975.
5 Ludwig, F.L. and E. Shelar. Site Selection for the Monitoring of
Photochemical Air Pollutants. Stanford Research Institute, Menlo
Park CA Prepared for U.S. Environmental Protection Agency
Research'Triangle Park, NC. EPA Publication No. EPA-450/3-78-013.
April 1978.
6. Bryan, R.J., R.J. Gordon, and H. Menck. Comparison of High Volume
Air Filter Samples at Varying Distances from Los Angeles Freeway.
University of Southern California, School of Medicine, Los Angeles,
CA (Presented at 66th Annual Meeting of Air Pollution Control
Association, Chicago, IL., June 24-28, 1973. APCA 73-158.)
7. Teer, E.H. Atmospheric Lead Concentration Above an Urban Street.
Master of Science Thesis, Washington University, St. Louis, MO.
January 1971.
8. Bradway, R.M., F.A. Record, and W.E. Belanger. Monitoring and
Modeling of Resuspended Roadway Dust Near Urban Arterials. GCA
Technology Division, Bedford, MA. (Presented at 1978 Annual
Meeting of Transportation Research Board, Washincton, DC.
January 1978.)
9 Pace, T.G., W.P. Freas, and E.M. Afify. Quantification of Relationship
Between Monitor Height and Measured Particulate Levels in Seven U.S.
Urban Areas. U.S. Environmental Protection Agency, Research Triangle
Park NC (Presented at 70th Annual Meeting of Air Pollution Control
Association, Toronto, Canada, June 20-24, 1977. APCA 77-13.4.)
-------
60
10
Harrison, P. R. Considerations for Siting Air Quality Monitors
in Urban Areas. City of Chicago, Department of Environmental
Control, Chicago, IL. (Presented at 66th Annual Meeting
?973 U;0ntr01 Associatl'on' Chicago, IL., June 24-28,
11. Study of Suspended Particulate Measurements at Varying Heights
12. Rodes, C.E. and G.F. Evans. Summary of LACS Integrated Pollutant
Data. In: Los Angeles Catalyst Study Symposium. U.S. Environmental
park- Nc- EPA
13' P™M °'Au r-' 4' MNa^'onal Assessment of the Urban Particulate
Problem: Volume 1, National Assessment. GCA Technology Division,
Bedford, MA. U.S. Environmental Protection Agency, Research
Traingle Park, NC. EPA Publication No. EPA-450/3- 75-024
June 1976.
14. Pace, T.G Impact of Vehicle-Related Particulates on TSP Concentra-
nlnol a.n.dcRaJlonale for Siting Hi-Vols in the Vicinity of Roadways.
OAQPS, U.S. Environmental Protection Agency, Research Triangle Park,
iiL . Apri i i y/o .
15. Wedding, J.B., A.R. McFarland, and J.E. Cermak. Large Particle
Collection Characteristics of Ambient Aerosol Samplers. Environ
Sci. Techno!. _n_: 387-390, 1977.
16. Cole, H.S. Guidance for NAQTS: Review of Meteorological Data
Sources. OAQPS, U.S. Environmental Protection Agency, Research
Triangle Park, NC. December 1977 (draft).
17. Guideline on Air Quality Models. OAQPS, U.S. Environmental
Protection Agency, Research Triangle Park, NC. OAQPS No. 1.2-080.
Apri i i y /o .
18. Technical Support Document for Determination of Good Engineering
Practice Stack Height. OAQPS, U.S. Environmental Protection Agency,
Research Triangle Park, NC. February 1978 (draft).
19. Gill, G.C., L.E. Olsson, J. Sela, and M. Suda. Accuracy of Wind
Measurements on Towers or Stacks. Bull. Amer. Meteorol Soc 48-
665-674, September 1967. ' — '
20. Turner, D.B. Workbook of Atmospheric Dispersion Estimates, Revised
Office of Air Programs, U.S. Department of Health, Education and
Welfare, Research Triangle Park, NC. Publication No. AP-26. 1970
-------
61
21. Onsite Meteorological Programs. Nuclear Regulatory Commission,
Washington, DC. NRC Guide 1.23. February 1972.
22. Weber, A.H. Atmospheric Dispersion Parameters in Gaussian Plume
Modeling - Part 1: Review of Current Systems and Possible Future
Developments. U.S. Environmental Protection Agency, Research
Triangle Park, NC. EPA Publication No. EPA-600/4-76-030a.
July 1976.
23. Holzworth, G.C. Mixing Heights, Wind Speeds, and Potential for
Urban Air Pollution Throughout the Contiguous United States.
Office of Air Programs, U.S. Department of Health, Education and
Welfare, Research Triangle Park, NC. Publication No. AP-101.
1972.
24. Guidelines for Siting and Exposure of Meteorological Instruments
for Environmental Purposes. Meteorology and Assessment Division,
U.S. Environmental Protection Agency, Research Triangle Park, NC.
January 1976 (draft).
25. Hoehne, W.E. Progress and Results of Functional Testing. National
Oceanic and Atmospheric Administration, Sterling, VA.
NOAA Technical Memorandum NWS T&EL-15. April 1977.
26. Stone, R.J. National Weather Service Automated Observational
Networks and the Test and Evalaution Division Functional Testing
Program. In: Preprint Volume for Fourth Symposium on Meteorological
Observations and Instrumentation, Denver, CO. April 10-14, 1978.
27. Mazzarella, D.M. Meteorological Sensors in Air Pollution Problems.
In: Proceedings of the Second Joint Conference on Sensing of Environ-
mental Pollutants. Instrument Society of America, Pittsburgh,
PA. 19/3.
28. Mazzarella, D.M. Meteorological Instruments for Use Near the
Ground - Their Selection and Use in Air Pollution Studies.
Science Associates, Inc., Princeton, NJ. (Presented at
Conference on Air Quality Meteorology and Atmospheric Ozone,
Boulder, CO., 1977.)
29. Johnson, W.B. and R.E. Ruff. Observational Systems and Techniques
in Air Pollution Meteorology. In: Lectures on Air Pollution and
Environmental Impact Analyses. American Meteorological Society,
Boston, MA. 1975.
30. George, D.H. and K.F. Zeller. Visibility Sensors in Your Air Quality
Program. In: Proceedings of the Second Joint Conference on Sensing
of Environmetal Pollutants. Instrument Society of America, Pittsburgh,
PA. 1973.
-------
62
31. Quality Assurance Handbook for Air Pollution Measurement Systems;
Volume I - Principles. U.S. Environmental Protection Agency (MD-77),
Research Triangle Park, NC. EPA Publication No. EPA-600/9-76-005.
January 1976.
32. Quality Assurance Handbook for Air Pollution Measurement Systems;
Volume II - Ambient Air Specific Methods. U.S. Environmental
Protection Agency (MD-77), Research Triangle Park, NC. EPA Publication
No. EPA-600/4-77-027a. May 1977.
US GOVERNMENT POINTING OFFICE 1978-740-261 /4153
Region No 4
------- An error occurred while trying to OCR this image.
------- |