OCR error (C:\Conversion\JobRoot\00000AXB\tiff\2000XG1W.tif): Unspecified error
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wideirange of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-010a
January 1979
POWER PLANT STACK PLUMES IN COMPLEX TERRAIN
Description of an Aerometric Field Study
by
Robert C. Koch, W. Gale Biggs,
Douglas Cover, Harry Rector,
Paul F. Stenberg, and Kenneth E. Pickering
GEOMET, Incorporated
15 Firstfield Road
Gaithersburg, Maryland 20760
EPA Contract Number 68-02-2260
Project Officer
George C. Holzworth
Environmental Science Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
ENVIRONMENTAL SCIENCE RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Science Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
A monitoring network consisting of eight fixed sites was set up and
operated in the vicinity of the 712-megawatt, coal-fired, Clinch River,
Virginia Steam Plant during the period of June 1, 1976 to September 30,
1977. At most of the monitoring stations sulfur dioxide, nitric oxide,
total nitrogen oxides, wind direction, wind speed, temperature, and
sulfates were measured. Two of the stations had 30 meter towers from which
meteorological variables were measured. On four or five days per week a
mobile van was operated to supplement the fixed site monitoring, and once
or twice a day balloons were used to measure winds and temperatures aloft
in the vicinity of the plant site. A monitor for continuous measurement
of sulfur dioxide and nitric oxide was operated in one of the two plant
stacks. These data were supplemented by helicopter measurements taken
during ten-day periods in November 1976 and July 1977.
The system used to collect the data, the operational procedures used to
run the system, and its performance record are described in this report.
The data that were collected have been edited and recorded on magnetic
tape and are available from the National Technical Information Service.
An appendix in this report describes the formats required for reading
this tape.
A report on analyses of these data will be published separately.
This report was submitted in fulfillment of Contract Number 68-02-2260
by GEOMET, Incorporated under the sponsorship of the U.S. Environmental
Protection Agency. This report covers a period from June 1, 1976 to
September 30, 1977, and work was completed as of October 24, 1978.
111
-------
CONTENTS
Abstract 111
Figures vi
Tables v111
Acknowledgment x
1. Introduction 1
2. The Clinch River Steam Power Plant 3
Plant selection criteria 3
Plant characteristics 3
Terrain and climatology 5
3. Project Organization 12
Management and staffing 12
Field operations 13
4. Field Measurements Program 16
System characteristics 16
Operating procedures 33
Quality assurance 60
5. Results 88
Overview 88
Data formats 102
6. Project Chronology 118
Calendar of events 118
7. References 124
Appendices
A. Site layout diagrams for eight fixed
monitoring stations 125
B. Major instrument performance characteristics 133
C. Documentation for Clinch River Plume Study
Data Tape 145
-------
FIGURES
Number Page
1 Location of Clinch River Steam Plant in relation
to the Moss No. 3 Coal Preparation Plant 4
2 Views of Clinch River Steam Plant . . . .
3 Views of plant taken from the NW and ESE.
4 Location of the Clinch River Steam Plant 8
5 Surrounding terrain, including monitoring site
locations of the Clinch River Power Plant 8
6 Location of sites 18
7 Standard fixed monitoring station configuration 22
8 Tower monitoring station meteorological signals 24
9 Fixed station network data acquisition station
configuration 26
10 Form to report day's activities 34
11 Form to report computer interventions 37
12 Decision process for operating sulfate samples 39
13 Log for sulfate sampling operations 46
14 Decision process for operating mobile vans 48
15 Log for ground mobile sampling 49
16 Representative helicopter flight path at fixed
distance from the power plant 53
17 Log of helicopter sampling 54
18a Diagram showing location of double theodolite base-
lines relative to the Clinch River Power Plant near
Carbo, VA 57
vi
-------
FIGURES
Number
18b Dimension and angular arrangement of double
theodolite baselines located just south of the
Clinch River Power Plant near Carbo, VA 57
19 Locations of plume photography sites 58
20 Sample plume photos 59
21 Station check list, page one 62
22 Station check list, page two 63
23 Sample operating record - hourly heat rate
loss summary 64
24 Sample operating record - weekly heat rate
loss summary 65
25 Sample operating record - daily generating log 66
26 Double theodolite form 68
27 T-sonde data report 69
28 Sulfate sample collection form 70
29 Mobile sulfate data form 71
30 Plume travel projection form 72
31 Model 8500 Calibrator 73
32 8450 Sulfur Monitor 74
33 Model 8440 Nitrogen Oxide analyzer 75
34 Monitoring station operating periods and measurement
down periods 90
vii
-------
TABLES
Number Page
1 Clinch River Steam Plant Stack Velocities
and Temperatures 5
2 Mean Annual and Seasonal Mixing Heights 11
3 Monitoring Site Characteristics 19
4 Summary of Measurements at Fixed Monitoring sites 28
5 Parameters Measured at Monitoring Stations 89
6 Summary of Sulfur Dioxide Audit Results,
February 15-23, 1977 92
7 Summary of Nitrogen Oxides Audit Results,
February 15-23, 1977 94
8 Audit Results for Oxides of Nitrogen Analyzers,
August 5-12, 1977 96
9 Sulfur Dioxide Analyzer Results from
Audit of August 5-12, 1977 97
10 Comparisons of Embedded Amounts of Sulfate
with Laboratory Analysis of Filter Sulfate 98
11 Comparisons of EPA Method 6 "and EDC Stack Analyzer
Measurements of S02 in Stack Gas 99
12 Comparisons of the EPA Method 7 and EDC Analyzer
Measurements of NO in Stack Gas 100
13 Fixed Station Record Format for Magnetic Tape 104
14 Mobile Ground Station Record Format for
Magnetic Tape 107
15 Helicopter Record Format for Magnetic Tape 107
16 Linear Interpolation Table for Relative
Humidity Measurement 108
17 Sulfate Record Format for Magnetic Tape 109
vlii
-------
TABLES
Number Page
18 Plant Operating Characteristic Record
Format for Magnetic Tape ................. in-
19 Corresponding Measurement of Boiler
Intake and Stack Exhaust Gas Flow Rate
20 Balloon Data Record Format for Magnetic Tape ........ 11 5
21 Chronology of Significant Field Events ........... ]]g
1x
-------
ACKNOWLEDGMENT
The assistance and cooperation of the Appalachian Power Company
and the American Electric Power Company were an important contribution
to this study. The company provided offices and storage and operating
space throughout an 18-month period in which the field program was set
up and performed. In addition, hourly measurements of the plant's
operating characteristics were provided. Field personnel were granted
access to all areas of the plant's operations, and plant personnel
freely offered advice and data in a friendly and timely manner.
-------
SECTION 1
INTRODUCTION
An aerometric survey was conducted in the vicinity of a power plant located
in the complex terrain of the western tip of Virginia. This report describes
the characteristics of the site, the equipment and monitoring operations used
to perform the survey, the kinds of data obtained and available from the program,
and the procedures recommended for accessing the data. The survey is the second
phase of a three-phase study of the behavior of plumes from an elevated contin-
uous source located in complex terrain. The first phase consisted of a litera-
ture review of the current state of knowledge of plume behavior in complex
terrain, including both transformation and removal processes and transport and
dispersion processes. The third phase will consist of an analysis of the data
collected in the second phase in order to identify relationships between terrain
characteristics, meteorological conditions, plant operating conditions and
resulting ground-level concentrations of pollutants.
This study was undertaken because of the generally recognized shortage of
real air quality measurements in mountainous complex terrain. The objective
of the aerometric survey was to obtain a reliable set of measurements over a
relatively long period of time on S02, sulfate, NO and N02 concentrations and
on relationships among power plant emissions, meteorological variables and
ground-level concentrations in areas of complex terrain.
The combustion processes of coal-fired electric power plants are major
sources of sulfur oxides and nitrogen oxides. Permissible ambient concentra-
tions for both pollutants are controlled by national ambient air quality stan-
dards, enforced under State Implementation Plans. S02 has long been considered
a health hazard. Its emission and ambient levels have typically been used as
primary indicators of power plant air pollution. More recently, the transforma-
tion of S02 into sulfates in the atmosphere has been studied as a probable
principal agent of production of the more damaging aspects of sulfur oxide pol-
lution. Nitric oxide and nitrogen dioxide produced in power plant combustion are
of concern because the nitric oxide is rapidly oxidized to N02 in the atmosphere
and N02 has adverse health effects. Nitrogen oxides play an important role in
reactions leading to formation of photochemical oxidants.
The primary emphasis in the measurement program has been on S02» NOx (i.e.,
NO and N02) and meteorological measurements. Sulfates and ozone were also
measured, but much less frequently. There is a general belief that many hours
are required for significant sulfate formation. Therefore, the opportunities
to observe sulfates in the vicinity of the plant which can be attributed to the
plant's emissions should be very limited. The validity of this assumption
can be tested in the third phase of the study. The ozone measurements provide
a limited amount of data on plume reactions, plume definition and background
concentrations.
-------
The program which evolved was constrained by the large expenses associated
with a field program. The measurements made represent an attempt both to get
frequent measurements of the exhaust gas plume from the plant and to observe
special effects caused by the complex topography.
-------
SECTION 2
THE CLINCH RIVER STEAM POWER PLANT
PLANT SELECTION CRITERIA
The locality selected for the field work was required to have the following
characteristics: the emission source was to be a steam-electric plant with a
generating capacity of 700 MW or greater and coal-fired boilers; in a nonurban,
mountainous area, isolated from other sources of S02'» some terrain relief in the
surrounding region was to be more than 1.5 times the height of the stacks at the
steam-electric plant. Ease of movement in the region of the plant and an adequate
network of roads in the plant locality to facilitate mobile sampling and to allow
flexibility in selecting fixed monitoring sites were also important considerations.
The Clinch River Steam Plant, owned by the Appalachian Power Company, is the
steam plant. The installed capacity of electrical generators is 712 MW. The
plant's three boilers are mostly fueled by local low-sulfur (^0.7%) coal; however
during 1977 the plant began using larger quantities of higher sulfur coal from
more distant sources. The Clinch River site has ridges and valleys both near
the plant and at a considerable distance from it which are readily accessible by
conventional on-the-road vehicles. A scattered rural population, served by
power and telephone lines, lives throughout the area. Monitoring sites for key
topographic and climatological conditions were easily available. The Clinch
River plant is isolated; the only other S02 pollution source within a 30 km
range, except for a few private residences, is a coal-drying operation (Moss
No. 3 Coal Preparation Plant), located 3 km northeast of the Clinch River plant,
which burns a small amount of local low-sulfur coal (see Figure 1).
Another important consideration in selecting the Clinch River Steam Plant,
as the site for this study was the power company's willingness to cooperate.
PLANT CHARACTERISTICS
The Clinch River Steam Plant has a generating capacity of 712 MW. The fuel,
local low-sulfur coal, is burned in three boilers. Exhaust gases are emitted
through two stacks, 150 ft apart, each 453 ft high. The stack diameters are
15.6 ft and 12.5 ft. The larger diameter stack serves two boilers. Table 1
shows the exit velocity and temperature of each stack for 50, 75 and 100% load
as reported by the plant. The plant operates essentially as a base load plant,
with a normal diurnal variability in its output load of about 200 MW (i.e., load
varies from a normal daytime peak of about 650 MW down to a nighttime low of
about 450 MW).
The plant is approximately square, 400 ft on a side, and 194 ft high, with
the base being 1513 ft above mean sea level. The boiler water is cooled through
the use of 5 mechanical-draft cooling towers. Four of the towers, serving the
first 2 boiler units, measure 67 ft by 281 ft and are 61 ft high. Each tower
contains 10 22-ft fans and can cool water 19° F at a rate of 55,000 gallons a min.
-------
0 V
NORTH
\\
MOSS NO. 3 COAL
PREPARATION PLANT
1560'
RNER
PLANT
1520' MSL
226
(3360 M)
TOWER SITE
1920' MSL
Figure 1. Location of Clinch River Steam Plant in
relation to the Moss No. 3 Coal Preparation Plant.
-------
The fifth tower, serving the plant's third boiler unit measures 67 ft by 391 ft
and is also 61 ft high. It contains 14 24-ft fans and is capable of cooling
water 20° F at a rate of 110,000 gal a min. The plant uses electrostatic pre-
cipitators with a design efficiency of 99.7% to remove flyash from the exhaust
gases. The flyash that 1s removed is transported by truck to landfill areas near
the plant. Figure 2 shows views of the plant looking north from the southside
of the plant. The cooling towers are seen on the east (right hand) side of the
plant.
TABLE 1. CLINCH RIVER STEAM PLANT STACK VELOCITIES AND TEMPERATURES*
Stack #1 Stack #2
475 mw load capacity 237. 5 mw load capacity
Load Velocity Temperature Velocity Temperature
50%
75%
100%
96 fps
125 fps
131 fps
282 F
294 F
302 F
75 fps
97 fps
102 fps
282 F
294 F
302 F
* Reported by plant engineer based on stack measurements.
The Clinch River Plant burns approximately 6,000 tons of coal a day to power
its three turbine-generators. The coal is ground into a powder before it is fed
into the plant's three 10-story high boilers. These boilers produce steam at
2000 Ib per square in. and 1050° F to drive the turbines. The turbine-generators
produce electricity at 15,000 V which is then stepped-up to 138,000 V for trans-
mission to several substations in Virginia, West Virginia, Kentucky, and
Tennessee.
TERRAIN AND CLIMATOLOGY
The site of the field measurement program was the vicinity of the Clinch
River Power Plant in Carbo, Virginia that could be affected by exhaust gases
from the plant. Two views of the plant site are shown in Figure 2 and the
complexity of the surrounding terrain in Figure 3. The location of this plant
in western Virginia is shown in Figure 4. The details of its surrounding
terrain, including monitoring site locations, are shown in Figure 5.
There are ridges exceeding 1-1/2 times stack height within 3 to 5 km of the
plant in all quadrants of the compass. Elevations exceeding stack height by
factors of 3 or more occur within 8 km to the northwest and 14 km to the southeast,
Clinch Mountain, located 14 km southeast of the plant, runs northeast-southwest
for approximately 25 km and reaches elevations in excess of 4,200 ft. Big A
Mountain, located 20 km northeast of the plant, is a finger-like projection
-------
View of the Clinch River Steam Plant
View of the Clinch River Steam Plant
Figure 2.
6
-------
View of the plant taken approximately
2 km NW looking toward the SE.
View of the plant taken approximate!/
7 km ESE looking toward the WNW.
Figure 3.
-------
^^?4^€fe
£l^^P.YsDk^Hraci
^rfasa4wSgs3s»S:
LCK ZZV /MGWQ
Figure 4. Location of the Clinch River Steam Plant.
8
-------
I
a!
3
•8
a
U
o
I
U
2
"53
fl
•a
i
M)
I
5
U
a*
li,
-------
that reaches elevations In excess of 3,800 ft. Flat Top Ridge, situated 8 km
northwest of the plant, is oriented northeast-southwest and reaches elevations
up to 3,000 ft. The major pattern of ridges and valleys trending northeast-
southwest is transected by many streams running northwest-southeast to form
a truly complex terrain.
The small town of Cleveland (estimated population - 250) lies next to the
Clinch River 4 km east-northeast of the plant. The town consists largely of
private residences and churches, but 1t also has a few small, locally owned
stores, a gas station, an elementary school, and a high school. The village of
Carbo lies just outside the plant grounds to the southeast. Carbo, with a
population of about 50, consists of private residences, a couple of churches,
and a country store.
The land north of the Clinch River Plant is rich in low-sulfur coal. A
large number of strip mines, as well as deep mining operations, dot the country-
side.
Climatological influences for the region in which the plant is located are
represented by the 40-year record of weather observations for the tri-city area
of Bristol, Kingsport and Johnson City, about 35 miles south-southwest of the
Clinch River Steam Plant. The National Weather Service Office located at the
Tri-City Airport is the closest first order station to the Clinch River power
plant. The climatological data recorded there are representative of the larger-
scale features which affect both sites and many of the smaller-scale features as
well. The data collected at the plant site during this study identifies the
nature of the highly local influences.
During the summer season, meteorological conditions are mostly under the
influence of weak high pressure systems. The storm track lies well to the north
of the area and only occasionally do frontal systems pass by. During the winter,
the prevailing storm track is to the south and east, along the Gulf Coast and up
the Atlantic coast. The fall transition period is associated with a high fre-
quency of occurrence of a stable air mass. Both clear days and very foggy days
are common and precipitation is less frequent. During the spring transition
season unstable air masses are common; the frequency of occurrence of dense fog
is very low; cloudiness and precipitation are more frequent. The winter season
is characterized by the highest frequency of cloudy conditions and precipitation.
The area is most likely to be affected by well-defined cyclones and front systems
during the winter and spring seasons.
The most common wind directions observed at Bristol are from the west-
southwest and the northeast. This is typical of other stations in the Appala-
chian region also. The mean resultant surface wind is westerly in winter,
becoming southwesterly in spring and summer. In the fall, the mean resultant
surface wind is northwesterly. The mean wind speed varies from 2 mps in summer
to 3.5 mps in late winter and early spring.
Clear days are most frequent in the fall, and cloudy days are most frequent
in the winter. Late summer and early fall is the period of most frequent heavy
fog (with visibility of one-quarter mile or less), while March and April are the
10
-------
months with least frequent heavy fog. These observations combined with the wind
speed climatology suggest that atmospheric dispersion conditions are least favor-
able during the early fall and most favorable during the late winter.
Precipitation occurs at a minimum monthly mean of 2.2 in. in October and at
a maximum monthly mean of 5.0 in. in July. Snow accumulations in excess of 1 in.
or more occur a mean of 5 d per year. Thunderstorms occur on a mean of 45 d per
year, mostly in the summer7
The normal diurnal temperature variation is 20 to 25° F and is fairly con-
sistent all year round. The monthly mean temperature varies from 36° F in
January to 75° F in July. The monthly mean relative humidity is 50 to 60% at mid-
day all year round. In summer the mean 7 a.m. humidity is 92% and in winter it is
80%. The time of maximum humidity is around 7 a.m., and it lies between the
summer and winter values in spring and fall. The high frequency of dense fog in
the fall compared to spring is associated with the greater stability of the air
in the fall more than with humidity.
The upper air observing station of Huntington, West Virginia (about 150 km
to the north-northwest of the plant) is the closest point for which upper air
data are available. Based on 3 years' data, Holzworth (1972) reported the sea-
sonal and annual mean afternoon mixing heights at Huntington to be as shown in
Table 2. The afternoon mixing height is lowest during the winter and highest
during the spring. Holzworth1s (1972) national scale analysis of mixing height
data suggests that mixing heights near the Clinch River Steam Plant are very
close to the Huntington values. However, the effects that the complex terrain
in the vicinity of the plant will have in increasing the mixing height are not
well established. When the mixing height is below or slightly above the height
of the plant exhaust plumes, this characteristic has an important relationship
to the plume behavior.
TABLE 2. MEAN ANNUAL AND SEASONAL MIXING HEIGHTS AT
HUNTINGTON, WEST VIRGINIA (HOLZWORTH 1972).
Season Afternoon mixing height (meters)
Winter 1079
Spring 1986
Summer 1641
Autumn 1340
Annual 1511
11
-------
SECTION 3
PROJECT ORGANIZATION
MANAGEMENT AND STAFFING
The aerometric survey was conducted by a field team which was resident
in Abingdon, Virginia for a period of about 20 months. This included time
to assemble equipment and supplies on-site, to set up and shake down the
sensing and data collection equipment, to operate and maintain the equipment
over the operating period, and to disassemble and remove the equipment at the
end. The primary management of the field operation and various support func-
tions were performed from the main offices in Gaithersburg, Maryland, where
specialists in instrumentation, procurement, and data processing were avail-
able as needed. The project manager exercised direct management of personnel,
supplies and funds and kept the Government project officer, the power plant
personnel, local authorities and other interacting parties informed of the
project's needs and progress. Nearly daily interaction occurred by telephone
between field personnel and the project manager.
The system was designed in the main office and coordinated with the
Government project officer. It was recognized that the initially designed
system was less extensive than was desired and it was allowed to evolve
as additional resources became available by increasing the number of mon-
itoring stations from six to eight and by adding additional sensors for
pollutants and meteorological conditions. Basically, the system consists
of fixed monitoring stations, which operate continuously and automatically
and which transmit data to a central site, and mobile monitoring stations
which are moved to be in the vicinity of the plume during each operating
period. The details of the system and how it is operated are described in
Section 4, FIELD MEASUREMENTS PROGRAM.
The number of people who were resident at the monitoring site varied
from three to five, but generally consisted of four people. A field crew
supervisor was responsible for coordinating the activities of people at the
site, for reporting problems and progress to the project manager, and for
interacting with the project manager in planning future activities. Small
procurement actions for supplies and minor maintenance services were per-
formed by the field staff, but all procurements were subject to the approval
of the project manager. The primary functions of the field crew were opera-
ting a mobile van, taking upper air observations, maintaining the sensors
and data processing equipment, record keeping, and filing and shipping of
data and supplies. One member of the field crew was responsible for run-
ning the mobile van, one was responsible for maintenance of equipment, and
two were responsible for upper air observations. The two responsible for
upper air observations also performed record keeping, data handling functions
and support to the maintenance and mobile van operations as needed.
12
-------
Airborne monitoring was performed during two intensive 10-day monitoring
periods. The field staff was increased to eight members for these operations.
This permitted two men to be assigned to the helicopter operations, one man
to coordinate information at the central processing site and one man to help
with the increased load of errands and data records. A helicopter was leased
from the Appalachian Flying Service, who furnished a pilot and provided assis-
tance in installing the instrumentation in the helicopter. . The helicopter
was based at the Tri-City Airport, near Bristol, Tennessee. Each day's
airborne operation began at the airport. However, subsequent refueling was
arranged at a local helicopter pad at Lebanon, Virginia. The convenient
refueling greatly increased the amount of flying time devoted to sampling
operations.
Other organizations who assisted in the execution and support of the
monitoring operations included the following:
• U.S. Environmental Protection Agency—Provided resources
for the study, advice on instrumentation and field
activities, guidance for selection of alternatives, and
critical review of all activities.
t American Electric Power Company and its subsidiary,
Appalachian Power Company—Provided office and storage
space, use of company grounds, security for vehicles
and trailers, data on plant operations and facilities,
advice on plant operating characteristics, and critical
review of monitoring data.
t Appalachian Flying Service—Contracted to GEOMET for
helicopter service.
• Midwest Research Institute and PEDCo Environmental Specialists,
Incorporated—Contracted to GEOMET for sulfate analysis of hi-vol
filters.
•
Research Triangle Institute—Contracted to EPA to field audit
the monitoring equipment.
•
Engineering-Science, Incorporated—Contracted to EPA to measure
stack gas characteristics as a check of other available measurements,
FIELD OPERATIONS
The field monitoring system was designed to include both stationary
and mobile sampling systems. This permitted conventional stationary
ground-level observations to be supplemented by observing stations which
are moved directly into the path of the plume during routine daily sampling
operations. The monitoring network was fully instrumented with meteoro-
logical sensors in order to permit micrometeorological influences induced
13
-------
by the complex terrain to be observed. The emission characteristics of
the plant's two stacks were determined by direct measurements of NO and
S02 concentrations in one stack as a part of the automated monitoring
network and by measurements of air temperature, boiler intake, air flow
rates, generator output loads, fuel consumption and fuel sulfur content,
which are taken by the plant.
The fixed monitoring network provides a continuous record of how
emissions from the plant affect a number of selected locations. Eight
locations were selected because of their relation to terrain character-
istics which are expected to influence the plume of plant emissions.
A part of the field operation was concerned with keeping these observing
stations operating properly. These stations collect a large amount of
data with a relatively small amount of manual support.
The fixed station sampling was supplemented by mobile sampling in
which the sampling station is moved to locations where the plant plume
can best be observed. Ground sampling in a mobile van was usually per-
formed 4 or 5 days a week. Most of the sampling was performed during
the day, but some nighttime sampling was carried out. Two types of
ground mobile sampling strategies were used. In one strategy the samples
were collected in a moving state. This allowed a large number of differ-
ent locations to be sampled in a short time to help define the width and
length of the area in which the plume was present at ground level. The
second strategy was to park the van at an opportune spot to provide a
1-hour sample which supplements the observations available from the fixed
stations. A mobile sampling station requires continuous manual supervi-
sion and produces data directly proportional to the amount of manning time
assigned. However, unlike the fixed stations, which only observe the
plume when the wind is carrying the plume to the station, the mobile sta-
tion can be moved to a good observing location (within limits imposed by
available roads) so that it is usually observing the plume. The fixed
stations are most suited to observing long-term plume effects, while the
mobile stations are most suited to observing short-term plume effects.
Both types of observation were important parts of this program.
While ground-based meteorological measurements were continuously
measured at the fixed monitoring stations, upper air measurements were
obtained from balloons. These measurements require continuous manual
supervision and were made during daylight hours. They provide a valu-
able supplement to the ground-based meteorological measurements from
fixed stations by adding a vertical dimension to the meteorological
data base.
The most comprehensive set of observations during the field opera-
tion was obtained by helicopter sampling. These observations allow
the plume to be measured in three dimensions. These data provide the
14
-------
most detailed picture of plume behavior. It is planned that the plume
behavior observed by helicopter will be used to interpret plume behav-
ior during other periods in which the plume is partially observed by
the fixed and mobile ground sampling stations.
In summary, the field program was organized to include fixed and mobile
ground sampling stations, emissions data from the plant, upper air meteoro-
logical data and airborne sampling. The equipment and supplies necessary
to carry out the program were procured and set up in the vicinity of the
Clinch River Steam Plant in Carbo, Virginia. The program was conducted by
a four-man field crew resident at the site and supported by specialized
services in management procurement and data processing in the home office
and by various services provided by the plant, by a helicopter service,
by an analytical laboratory and by maintenance and supply organizations
as required.
15
-------
SECTION 4
FIELD MEASUREMENTS PROGRAM
SYSTEM CHARACTERISTICS
The field program consisted of five distinct activities: continu-
ous data collection at eight fixed stations, frequent collection using
a mobile monitoring unit for surface observations, T-Sonde/Pibal observa-
tions, scheduled use of a helicopter monitoring unit for airborne observa-
tions and collection of plant operating parameters. Data were collected
over a 16-mo period.
The actual number of fixed stations was determined by the cost and manpower
required to maintain each station. The general design criteria for the fixed
station monitoring sites were:
• Wind speed, wind direction, temperature, S0?,
NO and N02 concentrations were to be continu-
ously measured near ground-level (within 10 m).
• Frequency of occurrence of concentrations above
detection thresholds was to be high; '10 percent
or more was the goal.
• Three monitoring sites were to be oriented to
measure variations ;in the ratio of S0? to sul-
fate and NO to NOp concentrations witn distance
from the power plant.
• Two monitoring sites were to be oriented to mea-
sure valley-ridge concentration differences.
• Three monitoring sites were to be oriented to
measure concentrations for different roughness
fetches; e.g., upvalley, downvalley, cross-
valley.
t One site was to detect presence of valley drain-
age flow.
!
''"] • Two sites were to measure plume impingement on
,* ridges.
16
-------
• Sulfate was to be measured on request in hourly
samples; system had to be capable of collecting
24 hours of sulfate samples, unattended.
t Two sites were to include 100-foot towers for
measuring vertical variation in temperature
and/or wind.
t The sites had to be accessible for setting up
equipment and shelters and for electricity and
telephone service; this meant being near a road
with residents served by telephone service and
electricity.
• Site shelters were to be secure from vandals.
The eight fixed monitoring station locations were selected to meet
design criteria for climatology, local topography, existing roadways and
utility services, and the ability to negotiate for the land on which to
place the instrument shelter. Locations were selected using guidance from
dispersion model calculations and representative climatological data. The
site locations are contained in Appendix A. Initially only six stations were
operational with the Lambert site coming on line on October 27, 1976 and the
Johnson site becoming operational November 15, 1976. The Castlewood station
washed away in the flood of April 5, 1977. The Lambert instrumentation was
installed at Castlewood during July 1977. Site characteristics are sum-
marized in Figure 6 and Table 3, and they are diagrammed in Appendix A.
Site No. 1 was the Tower site which had, in addition to the air qual-
ity equipment, a 30 m tower which measured winds at two levels and tem-
perature at three levels. This site was 3.4 km to the northeast of
the plant at an elevation of 585 m. The site was in open terrain and air
flows in the region were unobstructed by vegetation or nearby terrain
features. The nearest terrain feature was a ridge located directly south
of the Tower site at about 1 kilometer distance. This was the closest site
to the Power Plant and was located 124 m above the base of the plant.
Site No. 2 was the Munsey site which was located a little over 4 kilome-
ters to the southeast of the Power Plant. It was located approximately
one-third of the way up the hillside from the valley floor and was placed
in that location so that, in conjunction with the Hockey site located on
approximately the same line but at 200 m higher elevation, it could assist
in resolving the characteristics of the plume as it traveled in that direc-
tion.
Site No. 3 was the Nash's Ford site which was located approximately
11 kilometers to the east of the Power Plant in the Clinch River Valley.
The site was located on a small plateau between the Clinch River and Thompson
Creek. This site was representative of upvalley flows from the Power Plant
as well as any wind channeled up the valley.
17
-------
\
\
\
\
p
bo
\
18
-------
o
>.
,
>
o
w
V
1
a
I
«
o
fl
•s
O
stance
O
01
o
"o
ro
o
cM
oo
o
^
t--.
o
oo
rJ
o
t^
oo
ffi
N
o
ws
N.
o
vo
ul
oo
— ico
»H CM
0)
i
Z
-------
Site No. 4 was the Hockey site which was located on a ridge (792 m msl)
just above the Munsey site and on a line with Munsey and the Power Plant.
This site was selected to observe the concentration characteristics of the
Power Plant plume in relation to a ridgetop.
Site No. 5 was the Castlewood site which was located approximately 8
kilometers west-southwest of the Power Plant. It was very close to the Clinch
River, however, it had an unobstructed upvalley view so that winds from that
direction would be characteristic of drainage flows if they were occurring.
This site was about 10 m below the plant elevation. The station washed away
in the flood of April 5, 1977.
Site No. 6 was the Kent's Ridge site which was located 30 km east-
northeast of the Power Plant. This site was chosen for both its elevation
and distance and was the most distant site from the Power Plant.
Site No. 7 was the Johnson site which was located south-southwest of
the Power Plant. This was located on a ridge just to the south of the Power
Plant that runs from the west-southwest to the east-northeast. This is the
same ridge line that Hockey was located on.
Site No. 8 was the Lambert site and was located approximately one-third
of the way up the valley side. It was chosen to determine the plume charac-
teristics in that valley. The instrumentation was moved to Castlewood in
July 1977.
A stack monitor was located in the duct work at the base of the Unit 3
stack. The monitor was between the electrostatic precipitators and the base
of the stack. Valid data was collected after December 1, 1976.
The data acquisition station and office were located within the Clinch
River Steam Plant. The station contained all the equipment for communications
with the eight satellite stations, including a small data processing center;
further, it housed a facsimile machine for weather maps. All operations were
conducted from that station.
Standard Fixed Monitoring Stations
Five standard fixed stations (Nos. 2, 3, 4, 5 and 6) measured S02>
N0x» NO, wind speed, wind direction and temperature continuously. Two addi-
tional stations (Nos. 7 and 8) monitored S02, wind speed and wind direction.
The Tower site (No. 1) included a more extensive set of meteorological mea-
surements in addition to the standard fixed station measurements. Supplementary
wind information was obtained at Site No. 4 (Hockey). These additional wind
data were in the form of wind speed and wind direction at the 30 m level.
Equipment for this purpose was supported by a 100-foot guyed tower.
At all sites, the gas analyzer for nitrogen oxides measured total
nitrogen oxides (NOX) and nitric oxide (NO). The difference between
the two measurements was recorded as nitrogen dioxide (N02). Although
20
-------
all three signals were available from the analyzer, only the NO and NO
signals were calibrated and processed. The recorded N0? value is the x
difference between the calibrated NO and NO values. For every fixed
and mobile station measurement of nitrogen oxides, values of NO, NO , and
NOp are available, unless either the NO or the NO channel was not oper-
ating, in which case only one measurement is available.
Hourly samples of sulfates were collected on request at stations 3, 4,
5, 6, and 7, using a 24-port high volume air sampler.
For gaseous pollutant measurements, air was collected from a height of
about 1 m above the roof of the instrument shelter and pumped through a
sampling manifold from which it was available to the gas analyzers. The
wind and temperature sensors were located at the top of a 5 m tower which
was guyed to the roof of the instrument shelter. The resultant sampling
height was about 8.5 m above the ground. In general, this has been referred
to as a 10 m measuring height in this report. The 24-port high volume air
sampler was located on the roof of the instrument shelter.
In addition to the two gas analyzers, the instrument shelter, which was
maintained at 20°C, housed a transmuter to condition the meteorological
signals, an automated calibrator to dilute and regulate gas flow to the
gas analyzers, and a data logger. The data logger digitized analog sig-
nals, recognized commands from a central processor, and responded to
requests to turn on or off switches or to collect and transmit data.
The shelter contained working space for maintenance work and storage
space for filters, a log book and miscellaneous supplies.
Access to the roof was obtained by a ladder stored in the shelter.
The filters on the particulate samplers located on the roof were changed
during the routine twice weekly visits to the shelters by field personnel.
Sulfate sampling could be conducted continuously by collecting hourly
averaged samples for up to 24 hours at any remote fixed station without
requiring filter collection and replacement. When this equipment was
operated continuously, it was necessary for field staff to visit the sta-
tion once a day for servicing. However, the sequential air samplers were
commanded by an operator at the central site to activate collection of an
hour's particulate sampling only if the simultaneous S0? reading was
expected to be high or the station was selected to maintain background
levels. We expected that this procedure would permit the operation of most
fixed station sulfate sampling units unattended until the regularly scheduled
maintenance of the entire fixed station (which was normally twice a week),
since sulfate measurements were usually made only during those hours when
the plume was in the vicinity of the station.
As a back up to the telecommunication system, N0~ and S02 were con-
tinuously recorded on strip charts. The configuration of equipment at each
standard fixed monitoring station is illustrated in Figure 7.
21
-------
Figure 7. Standard fixed monitoring station configuration.
22
-------
Tower Monitoring Station
In addition to single-level continuous measurements of SCL, NO, and
NO , the Tower station made continuous measurements of wind direction and
speed (three dimensions) at two levels (10 and 30 m), temperature at three
levels (0.5, 4 and 30 m), and also measured relative humidity, ultraviolet
radiation, and precipitation. Relative humidity was measured on the tower
at a height of 4 m. Ultraviolet radiation and precipitation were measured
at appropriate points with unobstructed exposures due to vegetation, terrain
or man-made structures. The tower was approximately 30 m tall and was sup-
ported by a concrete base. Except for the meteorological items, the Tower
monitoring site was configured like the other standard fixed monitoring
stations. Figure 8 illustrates the processing of meteorological measure-
ments.
Stack Monitoring Station
Exhaust gases exiting the number 2 stack at the Clinch River Power Plant
were continuously monitored by an in situ gas analyzer beginning November 4,
1976. The sampler was located in the breeching leading from the electro-
static precipitator to the stack. Sampling was performed on gases exhausted
to the stack serving one boiler rather than the stack serving two boilers in
order to simplify the relationship between boiler operating characteristics
and the exhaust gas measurements. The signals were digitized and transmitted
to the central station by the same procedure used at other remote sites.
Central Data Acquisition
The data acquisition system for the fixed station network utilized
leased telephone lines for telemetry from the satellite fixed stations and
a minicomputer at the data acquisiton station for data processing and system
control. Each fixed monitoring station had a Monitor Labs 9400 remote data
logger with scanner, analog to digital converter, and telemetry interface.
The central data acquisition station processing and control system was a
Monitor Labs 700 Ambient Data Acquisiton System controlled by a Data General
Nova Computer 2/10 with 24 K core storage and with automatic memory protec-
tion, program loading, and restart. System time was established and main-
tained through any power outages by a Monitor Labs 3100 digital clock with
automatic battery power backup. The Nova Computer could be controlled,
programmed, and interrogated by a console printer which provided automatic
and requested data printout. Computer-edited and processed data were
stored on a magnetic tape ready for subsequent analysis on a larger com-
puter. The satellite stations reported data on query from the computer
at a scan rate of one complete cycle per 2 min. Any values desired by the
resident meteorologist could be printed out on request via the printer and
the program was protected against operator error. The base station auto-
matically controlled the air quality analyzers for zero readings and single
value calibrations. These routine zero/calibration readings were programmed
to be taken once a day, shortly after midnight, but could be initiated at
any time on operator request.
23
-------
Elevation 30M
Elevation 10M
>
DATA LOGGER
»
* v y ^ > ^
^ in cr>
t Hd H6
£ w 3
8 S 1
O. v. •$
ail
« ro H
H H
y
t
Relative Humidity
i
>s
Wind Direction, 30M
^ /
o
ro
•a"
CO
£
/* /
ro
tf
fi
be
CO
1
y*. ,
v y*. y\ /
Elevation Sigma, 30M
• Wind Direction, 10M
1 — 1
(U
CO
•a
•M
^
v /s.
Azimuth Sigma, 10M
Elevation Sigma, 10M
>\ yp y^
> > \
SIGNAL CONDITIONER
/^ /
i 1
^ 1
<
/ v >
li.
\ y\. /
13
1 1
« S 2
1 ^
is 0 [T
ft > o
QJ *H W
p, « fi
(3 rt «>
s 3 "
H PS
\
S5 ^
ld
S h
s- i
S <"
H <"
CO
fl
I
o
00
-------
The system interrogated the sensors once every 2 minutes. Hourly
averages, 2-minute values and peak values were stored on magnetic tape
for further processing. Meteorological parameters were treated simi-
larly. The configuration of monitoring stations and the equipment at
the central data acquisition station is shown in Figure 9.
Equipment Used at Fixed Stations
Part or all of the following equipment was located in each of the
stations:
• S02 Analyzer - Monitor Labs (ML) Model 8450
t NO/NOV Analyzer - ML Model 8440
/\
• Total Particulate Sampler - Misco Special 24 Head
Sequential High Volume Air Sampler
§ Wind Sensor - Meteorology Research, Inc. (MRI)
Model 1022
• Temperature Sensor - MRI Model 815-1
• Data Logger - ML Model 9400
• Meteorological Signal Conditioner - MRI Model 1002
with circuits for wind direction (540°), wind speed
(0 to 100 mph), azimuth sigma (0 to 45°) and tempera-
ture (-30* to 50eC)
t Strip Chart Recorder (2 pen)
t Calibrator and Controller - ML Model 8500
• Instrument Shelter - Spacemaster, Special 8' x 12'
Trailer
• Tower - MRI Model 1412-18 Tripod
• Miscellaneous Supplies (cable, mounting arms, tubing,
manifolds, blower, gases, filters, etc.).
The performance characteristics of each of the above sensing instruments
and all others used in this study are described in Appendix B.
The Tower monitoring station was identical to the other fixed mon-
itoring stations except that the meteorological sensors were replaced by
a more extensive set of sensors which were located on a 30-m tower rather
25
-------
Figure 9. Fixed station network data acquisition station configuration.
26
-------
that a 5.5 m tripod on top of the instrument shelters. At the Tower site the
following equipment was used in place of the standard station meteorology
equipment:
• Wind Sensors (2) - Meteorology Research, Inc.
(MRI) Model 1053
• Temperature and Relative Humidity Sensor - MRI
Model 815-4
• Temperature Sensors (2) - MRI Model 815-2
t Precipitation Sensor - MRI Model 370 Tipping Bucket
Rain Gage (heated)
• UV Radiometer - International Light, Inc. Model
PT100-CM108 #600 Photodetector
• Meteorological Signal Conditioner - MRI Model 1001
with circuits for wind azimuth (540°), wind elevation
(+60"), wind speed (0 to 100 mph), azimuth sigma
(U to 45°), elevation sigma (0 to 15°), differential
temperature (+5°C), temperature (-30* to 50"C), rela-
tive humidity (0 to 100%), and rain accumulation
(0.0 to 1.00 in)
• Tower (30 m) - Special Installation.
The Hockey site was identical to the other sites except that it had
an additional MRI Model 1022 wind set mounted on top of a 30-m tower.
The following equipment was used to measure and record stack gas con-
centrations:
• Stack Gas Analyzer - Environmental Data Corporation
Monitor for S02 and NO
• Data Logger - Monitor Labs Model 9400
• Strip Chart Recorder (2 Pen)
Table 4 gives a listing of the nine fixed ground stations and the
measurements at those sites. Detailed specifications for each major
equipment item are given in Appendix B.
Equipment located at the base station for accumulating data from
all remote sensing stations and for controlling all field monitoring
activities included the following:
27
-------
a
h
VI
TORINO
1
g
H
CO
H
1
i
o
PS
1
to
«
a
PQ
H
|
I
"w
c
o
> «
3 1
s
a S
H £
II
•3 1
« £
il|ll
V) Q ° S
t) BB 5 C O
| "S -1 2.
« Q "S <
•2 g 75 §
•8 « | 1 S
1 1 1 1 1
in O ° W
•H | -g rS -
|||||
e
o —
"SSI
5 s §.
S
113 U I"
£ v o
it o. *n
* W) *-
ill
.U _
1 « ?
1 J-l
D. "•"
E ^d
" HS
£ o
H
11
H S.
N
i
s1
X X X X
xxxxxxxx
xxxxxxxx
xxxxxxxx
X X X X X X
X X X X X X
X X X X X X
X X X X X X
xxxxxxxx
i;
o
H
8.
H
28
-------
• Data Acquisition System - Monitor Labs Model 7000
including computer interface (ML4100), digital clock
(ML3100), leased-line modem interface (ML4400L),
software (ADAM III), high-speed printer interface
(for Decwriter) and paper tape reader interface
(CP49)
• Computer - Data General NOVA 2/10.
• Uninterruptible Power Supply - Elgard Model 102-1
• Magnetic Tape System - Data General Model 6021
• Printers (2) - Decwriter Model LA35 and LA36
t Paper Tape Reader - Remex Model 6300
• Casette Tape Reader - Metrodata Model DL625
with Interface to Decwriter
• Teletypes (2) - Series A and C (discontinued in
November 1976)
• Facsimile - National Weather Service Fax
• Pibal Theodolites (2)
• T-Sonde receiver and T-Sondes
• Miscellaneous supplies (cable, balloons, magnetic
tape, printer paper, etc.)
• Two-way radio.
Surface Mobile Sampling
Mobile ground sampling was conducted in accordance with the following
criteria:
t Concentrations of SOp, NO, N02, and 03 were to
be continuously monitored for 1 hour at stationary
locations within the footprint area (i.e., all ground
locations where the ambient concentration exceeds the
sensing threshold of the analyzers).
• A 1-hour particulate sample (for sulfate analysis) was
to be collected at each stationary location.
29
-------
Wind speed and direction and temperature were to be
continuously recorded at each stationary location.
Temperature and S0? concentration were to be con-
tinuously recorded using fast response sensors through-
out all mobile traverses with locations periodically
annotated on the recorders. The traverses were to
include a valley to ridge pass which emphasized verti-
cal variations, if possible.
Sampling operations were to be scheduled to give
primary emphasis to reasonably steady meteorological
conditions with secondary emphasis on transient con-
ditions.
Stationary sampling locations were to be selected to
maximize the type of information which was appropriate
to the dispersion situation being observed and to com-
plement the fixed station monitors. Some specific
types of information sought were the following:
~ Ratio of sulfate to SCL and NO to NOp concen-
trations with increasing distance from plant
— Valley to ridge concentration differences
~ Plume axis ground concentration profile for varying
distance from the source
~ Influence of valley circulation on ground concentra-
tions from the plume
~ Influence of terrain roughness on plume dispersion
— Interaction of terrain roughness and meteorological
conditions on plume dispersion.
The latter two types of information sometimes resulted
in the selection of a location in order to obtain data
which was comparable with another sampling day.
Mobile sampling was to be conducted for 5-1/2 hours
a day, 4 days a week.
The surface mobile sampling unit provided data for three important
aspects of the monitoring effort. First, the mobile monitor was able to
30
-------
survey the area upwind of the source to provide a background reference
for all concentrations. Second, the mobile samplinq afforded a much
better resolution of the concentration patterns at the surface by filling
in data points between the fixed station locations and in places of spe-
cial interest such as along ridges or in lower areas where the effects of
drainage air flow could be studied. Third, the mobile unit provided the
opportunity to search for and define the peak concentration areas affected
by the plume.
The following equipment was used for mobile ground sampling:
• NO/NO Analyzer - Monitor Labs (ML) Model 8440
A
• S02 Analyzer - Theta Sensor, Inc. (TSI) Model 7111-1051
t Ozone Analyzer - ML Model 8410
• Particulate Sampler - Misco Model 8000
t Meteorological System - Climatronics Corporation Electronic
Weather Station (included wind speed, wind direction, and
temperature sensors, signal conditioner and recorder)
• Data Logger - Metrodata Model DL 640
• Power Supply - Homelite Model 130A22-1D 2250-watt generator
• Strip Chart Recorders (4) - Esterline Angus Miniservo
Model S22243
• Van - Chevrolet Suburban Model CK10906
t Miscellaneous Supplies (cable, mounting arms, tubing,
manifolds, blower, gases, filters, batteries, strip
charts, casette tapes, etc.)
• Two-way radio.
Detailed specifications for each major equipment item are given in Appendix B.
Airborne Mobile Sampling
The following airborne mobile sampling was performed by helicopter:
• SO^, NO, NOp, and 0., concentrations; temperature and
relative humidity wire continuously measured during
selected periods of each flight.
31
-------
• Measurements of S0?, NO , and 0^ were visually
displayed for an oDserveY accompanying the helicopter
pilot to help locate the plume and to keep track of
vertical and horizontal positions of passes through
the plume.
• Plume concentrations were measured at several dis-
tances downwind varying from 0.1 to 15 km.
• Measurements were made in two 7- to 10-day periods,
one period in the late fall and the second in mid-
summer.
The helicopter sampling was used to measure vertical and horizontal profiles
of SOp, NO and N02> temperature, humidity and 03 in the plant plume.
The helicopter sampling program had two major objectives associated
with its use; the first objective related to plume identification and plume
chasing; the second major objective was plume definition.
The plume identification phase of this program consisted of flying the
helicopter in such a mode so as to identify the bounds of the plume at vari-
ous distances downwind. This technique simply consisted of a flight through
the plume taking measurements while the helicopter was traversing the plume.
One purpose of this type of measurement was to identify the distance down-
wind that the plume could be tracked in its composite form and to identify
the path of the plume relative to the complex terrain.
The plume definition phase consisted of detailed cross-sectional pro-
files of the plume at various elevations. Once the plume had been identi-
fied, continual traverses at different heights through the plume were made
so as to define the detailed characteristics of the various parameters in
the plume.
The following equipment was carried in a helicopter for airborne
sampling:
• SOp Analyzer - Theta Sensor
t NO/NO Analyzer - Monitor Labs (ML) Model 8440
A
• 03 Analyzer - ML Model 8410
• Meteorological and Navigational Data System -
Metrodata Model M-8 with sensors for airspeed,
altitude, temperature and humidity
• Data Logger - Metrodata Model DL640
32
-------
t Power supply
• Two-way radio
• 3 strip chart recorders
• Miscellaneous supplies (cable, tubing, gases, filters,
batteries, strip charts, cassette tapes, etc.)
Detailed specifications for each major equipment item are given in Appendix B.
OPERATING PROCEDURES
This section describes the detailed procedures for performing the vari-
ous field activities that were necessary for the successful completion of
the field portion of this project. Included also are forms that were used
in carrying out the various assignments.
Base Station Operations
The base station served as a coordinating center for the operations of
all other aspects of the study. Here, daily operations were discussed and
altered as necessary to accommodate special events or problems. Each day
a two-sheet report (Figure 10) was completed to summarize the day's activi-
ties. The base station housed facilities for the data acquisition system,
FAX machine, Service A and Service C teletypes, reduction of Pibal and
T-Sonde data, and for computing the plume projection.
Data Acquisition System
At the base station a computer performed the continuous and automatic
running of the data acquisition system. The data acquisition system had
a prescribed routine schedule of functions which consisted of interrogating
all fixed station sensors every 2 min. This was the normal mode of opera-
tion. The routine schedule could be interrupted by operator request to
perform special functions. These functions sometimes were to alter the
routine schedule, to interrogate any of the sensors, to modify the automatic
processing of collected data, or to add or delete sensors from the inter-
rogation list.
The control system received inputs at the control station from a con-
sole, a paper tape reader, a magnetic tape unit, or telephone lines. The
control system sent outputs to a printer, a magnetic tape unit, or telephone
lines. The telephone lines were connected to the data loggers at each
monitoring station. Data from all the remote sites were telemetered to the
base station for processing and storage via leased telephone lines. The
base station contained the central processing unit with a printer for real
time display and for system checkout, and a tape recorder for data storage.
33
-------
DAILY PROJECT REPORT
Clinch River Power Plant Plume Study Project
Carbo Field Office
Julian Day Gregorian Date
Report Prepared by:
Field Staff Members Present:
Major Activities Performed:
Mobile Ground Runs
Unit 1 (SO, NO, 0Y), (where)
A A A
Unit 2 (S02 only), (where)
Pibal Taken, (when)
Routine Station Maintenance - Stations Visited
Sulfate Sampling
Date Preparation and Analyses
Other Routine Activities
Project Visitors:
GEOMET Home Office Staff: (who)
Other Visitors: (who)
Special Events and Problems:
Staff or Equipment Shortages
Unusual or Pollution Events
General Weather Condition:
Logistics:
Material Ordered
Material Received
Material Shipped
(continued)
Figure 10. Form to report day's activities
5/77
34
-------
TOWER
MUNSEY
NASHS
HOCKEY
CASTLE
KENTS
OOHNSN
LAMBER
VAN 1
VAN 2
STACK
DAILY PROJECT REPORT
(Continued)
Major Equipment Status
X = equipment for that parameter not at that station
Enter (in decimal form) the portion of 24-hours that
sensor operated normally e.g., .7 = normal operation
for 16 hours.
* Entries for Part. = A/B where A is time sampler is oper-
able; B = time equipment was operated (time is fraction
of 24-hours in decimal form).
.8
3
_a
a
$-
to
a.
x>
X
X
X
re
cc
X
X
X
X
X
X
X
X
X
X
a.
o
OL.
X
X
X
X
X
X
X
X
X
X
_1
LU
X
X
X
X
X
X
X
X
X
X
LU
co
X
X
X
X
X
X
X
X
X
X
t— 1
a
X
X
X
X
X
X
X
X
X
X
co
CO
3
X
X
X
X
X
X
X
X
X
CO
a
X
X
X
X
X
X
X
X
X
ro
—4
UJ
X
X
X
X
X
X
X
X
X
X
CO
LU
CO
X
X
X
X
X
X
X
X
X
X
CO
£
X
X
X
X
X
X
X
X
X
X
CO
CD
CO
X
X
X
X
X
X
X
X
X
Dichot.
X
X
X
X
X
X
X
X
co
o
X
X
X
X
X
X
X
X
X
X
EQUIPMENT MALFUNCTIONS:
* For example I/. 33 means sampler was available 24 hours and operated 8 hours.
Figure 10. (concluded)
35
-------
The condition of each sensor at each monitoring site was continually
observable on the routine data printout at the central station. Each
sensor was routinely checked, verified and logged at least twice a day.
Each such checkout period included reading a sequence of measurement
values for each sensor at each site to demonstrate that the sensors were
performing meaningfully. Meteorologists technically trained in instru-
ment operation evaluated these readings. Any discrepancies observed were
checked by visiting the appropriate site and taking necessary corrective
actions.
Replacement parts for key components of the system were maintained
on site. Maintenance agreements were arranged for all system components.
Equipment was occasionally moved from one site to another to keep sites
which were exposed to the plume in operation as much as possible.
A log of all computer interventions (Figure 11) was kept to have
a complete understanding of all information recorded on magnetic tape.
A copy of this log is included and is self-explanatory. In addition,
the operator asking for the intervention initialed all entries.
In addition to the above data gathering system operations, routine
meteorological observations were noted in a daily log by an observer
during duty hours. This helped with later data interpretation, especi-
ally during the significant mobile ground sampling periods. Vertical pro-
files of wind speed, direction and temperature were observed by double
theodolite pilot balloon observations supplemented by single theodolite
observations and T-Sondes. Data on the power plant operating conditions
were routinely obtained from the plant personnel.
Plume Projection Techniques—
To aid in determining probable locations for stationary sampling for
Mobile I, a plume projection technique was developed. This technique used
the clear plastic-covered topographic maps as mounted on the wall of the main
office. The projection of the plume was then plotted on this clear plastic
with grease pencils.
The projection usually started at around 5:00 or 6:00 in the morn-
ing, depending somewhat on the wind speed. For high wind speed condi-
tions, only about 2 previous hours needed to be plotted. A color scheme
for grease pencils was selected so that a particular hour then remained
the same color for various plots of the plume on the chart. Each hourly
update was then plotted on the topographic maps leaving the previous plume
projections on the plastic so that a trend could be observed in terms of
plume movement through the day.
The technique itself consisted of calculating the distance that the
plume would traverse in 1 h by using the upper level wind at either the
Hockey or Tower site. Generally, the Hockey site seemed to be more indi-
cative of the travel of the plume. However, the meteorologist would
36
-------
COMPUTER INTERVENTION LOG
Date
Intervention
Time
Return to
Normal Time
Reason for Intervention
Figure 11. Form to report computer interventions.
37
-------
consider both the Tower and the Hockey site winds using his local experi-
ence to determine the best value under various meteorological conditions.
Sulfate Sampling—
The sulfate sampling program was also directed from the base station.
The sulfate monitoring decision process is shown schematically in Figure 12.
Each morning, as the plume projection was plotted on the wall map,
decisions were made as to potential samplers that might be turned on during
that day that would be in a downwind position of the power plant plume.
The synoptic situation from the facsimile data was also reviewed in deciding
whether specific samplers would be activated or not. Sulfate samplers both
upwind and downwind of the plant were usually activated on selected days.
The selected sulfate samplers were turned on through the DEC writer
attached to the computer. Once the samplers were activated, the decision
to leave a sampler on or activate additional sulfate samplers was reviewed
on an hourly basis. If the variation of the plume was such that it appeared
that it would not be returning to the direction of one of the samplers that
had already been activated, generally a 3-hour time span was allowed to elapse
before turning the sampler off. The reason for this elapsed time was related
to the uncertainties of the plume projection position and to the possible low-
level drift of residual sulfates that might have been in the area from earlier
positions of the plume.
A criterion to activate a sulfate sampler was that the wind direction
was within i 45° of the stationary sampling location. This decision was
tempered somewhat by the steadiness of the plume and extensive field experi-
ences If the early morning decision was to not activate any sulfate samplers,
a review of that decision was made at noon based upon additional wind observa-
tions. If the samplers were activated, the hourly review procedure was
followed. If the noon decision was not to activate any samplers, then at
4:00 p.m. another review was made to determine whether any samplers should
be activated for an overnight sampling period.
This 4:00 p.m. review was made on all days using the same criteria as
stated above; however, a longer period of forecast was necessary since this
required a decision that would influence the next 14 to 16 hours (5:00 p.m.
through 9:00 a.m. the next morning).
This sulfate monitoring decision process is shown schematically in
Figure 12. The on and off times for each sulfate monitor are recorded
automatically by the computer and printed on the computer output record.
Fixed Monitoring Network Operations
The daily operation of the fixed station monitoring network consisted
mainly of routine instrument maintenance and calibration. If a malfunction
was detected in any instrument in the system, it was given early attention
38
-------
#
5 =
» S.
•; -3
" a
S^
BE
3 '
*t
!
J 3
c J
X
i
>
at
SULt ATF, SAMPLIN
rl
n
if
H
j
i
•* s
i o "^
" sll
•* « i
D ^
A
I
•g
P|3
L
js .3
5^5;
- < S
;
>•
mpUi 4re
eJ at that
lume b
and Holding,
•Jllj
i^i'
Chech Synoptic Situation
from Facsimile
,i
5 5
5 3
1 *
- 1
31 1 !
lii
i
s
>
* § i; * M
z " * 5 -
5 5 -5. 1 •£.
i * M *
a a £ «. -
*f
H« j ^
j-1 ' . -
a JJ
t is
1 II
i "
^
1
S"^
s 5 1
if Jt
fl J*
!« ll
t« 5
si si
J! S Jl 'S
»3 • a
i ,
s
n
8-
O.
a
O
3
O
(U
a
39
-------
by the field personnel in order to minimize any data loss that could occur.
The maintenance and calibration schedule was designed so that each of the
eight fixed stations received both maintenance and calibration every 7 d.
In addition, the maintenance visit preceded the calibration visit by a
minimum of 2 d to insure stable instrument responses during calibration.
Walk-in Form—-
A form was provided to keep a running account of all equipment at a
station. The procedure for completion was as follows:
1. Enter station and obtain station check list.
2. Without changing any information, complete the check list.
3. Make a comparison between the new check list and the pre-
vious form which will be found on top of the Monitor Labs
8450 Sulfur Analyzer.
4. Note any changes that have taken place.
5. If changes have taken place restore the system to the
original status which will be found in the station and
meteorological equipment log book.
6. Change the hi-vol filters, if required, and note status of
filters. Bring replaced filters and form to main office
for mailing to laboratory.
7. Leave the updated check list on top of the Monitor Labs
8450 Sulfur analyzer and return the previous check list
to the main office file.
Station Log—
Log books were kept at all stations for all equipment. They were
titled as follows:
1. 8450 (Sulfur Analyzer)
2. 8440 (Oxides of Nitrogen)
3. Station and meteorological equipment (9400, Met equipment).
Any visit to a station had to be logged in the appropriate book. If
any work was done or changes were made, the procedure had to be carefully
outlined in the log. Every entry included:
1. Time out of service
2. Reason for outage or nature of repair
40
-------
3. Person's initials
4. Time back in service.
Maintenance Procedure--
General maintenance was performed at all stations on a regular basis,
This work was kept up to date to prevent major periods of down time.
The following procedure was used when a general maintenance visit
was made:
I. 8450 - S02
A. General Inspection
1. Does it respond to span gas?
2. Electrical test?
3. Optical test?
4. Lights on?
5. H2'flow ok?
B. Internal Inspection
1. Synchronous motor running quietly?
2. Chopper belt ok?
3. Pump operating?
4. Recorder and DAS outputs ok?
5. Change burner filters?
6. Water in exhaust?
II. 8440 - NO -NO
/\
A. General Inspection
1. Response to span gas?
2. Electrical test?
3. Optical test?
4. Lights on?
5. NO. NO, 03 flows ok?
6. Converter light work?
7. Vacuum ok?
41
-------
B. Internal Inspection
1. Change dryrite?
2. Change converter catalyst?
3. Change intake filter?
4. Synchronous motor ok?
5. Chopper drive ok?
6. Pump ok?
7. Gelman charcoal filter ok?
8. Span-Zero valves work?
III. 8500 - Calibrator
A. General Inspection
1. Flows on each channel ok?
2. Temperature ok?
3. Switch in run position?
4. NO system pressure?
B. Internal Inspection
1. Pump ok?
2. Charcoal ok?
3. Permeation tubes ok?
4. Filters ok?
5. Ozone light ok in Ozatrog?* !
IV. 9400 - Data Logger
1. Scan properly?
2. Carrier light?
3. Channel assignment ok?
4. Check all channels for following output ranges.
Data should fall within these limits:
- o
1
2
3
4
5
6
7
8
9
10
S0?
N(T
NO,
NO
WSI
WDI
SGI
Temp I
SO. 0-9v;
WT
PCP
0-1 Ov
0-1 Ov
0-1 Ov
0-1 Ov
0-5v
0-5v
0-5v
0-5v
UV @ Tower
0-5v
0-5v
11 ELI 05-v-v
12 SEI 0-5v5v
13 DTI 0-5v5v
14 WS III 0-5v
15 WD III 0-5v
16 EL III 0-5v
17 SE III 0-5v
18 DT III 0-5v
0-lv 19 SG III 0-5v
Ozatrog is a process which generates ozone.
42
-------
V. Station
A. Change glass wool at air intake
B. Clean air conditioner filter
C. Check HiVol for proper operation
D. Check tension on tower support wires
E. Check NO + HL tank pressure
F. Clean trailer and empty trash.
Mobile Ground Monitoring
Operating Procedures--
The mobile sampling operations were scheduled to observe the power
plant plume under a variety of meteorological dispersion conditions, and
included early morning, afternoon, and evening periods. Initially, three
people were employed in the operation, two in van (Mobile 1) unit and one
at the central site. Having two people in the mobile unit allowed one
person to continuously monitor the instruments, while the second person
drove the vehicle through areas affected by the plume and searched for
possible stationary sampling sites. However, when the upper air program
was intensified, the number of people in the mobile unit was dropped to
one. This person had the responsibility of driving the mobile unit to
areas downwind of the plant and setting up the unit for stationary sampling
in those areas. Radio communications were used between the mobile unit and
the central site to pass up-to-date information from the monitoring network
and the facsimile receiver to the mobile unit operator.
Prior to each ground sampling operation there was a period for
planning and setting up the operation. During this period the three
(sometimes two) full-time people at the site reviewed data from the net-
work monitoring system and the NWS weather facsimile (45 charts per day),
made a local observation of winds (up to 3,200 ft) using a pilot-balloon
and a double theodolite, predicted plume behavior for the next 6 h,
selected the mobile sampling route, and set up equipment for the mobile
unit. Prediction of plume behavior took into account present and pre-
dicted estimates of the following features:
• Plume height
• Wind direction and speed at plume height
• Vertical temperature profile
• Atmospheric stability
43
-------
• Valley flow regimes
• Footprint area (locations where ground concentra-
tions exceed sensing threshold).
Initially, these estimates were based on routine forecasting procedures,
such as Briggs plume rise equations, interpolation and extrapolation of
synoptic weather patterns, patterns of diurnal temperature profile, and
workbook dispersion graphs.
Two types of mobile sampling were conducted, including (1) mobile
sampling traverses to identify the footprint area, SCL concentration
profiles and temperature profiles and (2) stationary sampling to meet
selected daily objectives for data. The daily objectives for selecting
sampling locations included obtaining the following:
• One-hour concentration versus distance profiles for
different contributions of terrain roughness and
meteorological conditions. (Select stationary sampl-
ing locations which complement each other and fixed
station monitors.)
t Valley to ridge concentration differences. (Select
moving traverses at appropriate locations and select
stationary sampling locations which complement fixed
station monitors or consecutive mobile unit stationary
locations.)
• For different wind directions representing different
roughness fetches under otherwise similar meteorological
conditions (but different sampling days), selection of
sampling locations at equal distances from the power
plant.
<*
• For similar wind directions but different meteorolog-
ical conditions (and different sampling days), selec-
tion of locations at equal distances from the power
plant.
• Selection of traverse routes and stationary locations
designed to detect the presence and magnitude of special
effects, such as valley circulation, fumigation, etc.,
when conditions favoring such effects are forecast or
observed.
During the planning operation the sampling objective was defined and ten-
tative stationary sampling locations selected.
44
-------
Based on the footprint forecast, the direction and distance to the
closest sensing location was identified. A continuous display of concen-
trations from the SCL analyzer was observed in the mobile unit to iden-
tify the presence of a footprint area, to verify the suitability of the
selected stationary sampling locations, and if necessary to search for
more desirable locations. While within the footprint area, locations
and times were logged for all significant observations. Locations were
designated by means of a cylindrical coordinate system centered on the
plant site. During the period from late December, 1976 through mid-March,
1977 a second mobile van (Mobile 2) was employed to search for the plume
footprint area, while the Mobile 1 van was collecting stationary samples.
Each day's mobile ground sampling survey normally consisted of 4
stationary 1-hour periods during which SCL, NO, NO , wind direction,
wind speed, temperature and ozone were continuously measured and a parti-
culate sample was collected for sulfate analysis. The log used to record
sulfate sampling information is shown in Figure 13. $03 concentrations
were measured continuously during mobile sampling periods within the
5 and 1/2-h sampling operation. If possible, one traverse during the
period included a valley to ridge pass. The Mobile 2 van recorded only
S02 during its 3-mo period of utilization. In the course of the mobile
sampling operation, the operators were able to contact the observer in the
central site to obtain new data from the fixed stations and from Pibal
observations which helped locate the footprint area. A typical day's
operation for a mobile ground survey is shown below:
Elapsed Time Activities
0000 - 0100 Check system status
Obtain and process winds aloft data
Review and analyze data from FAX and mon-
itoring network
0100 - 0200 Predict plume behavior
Set up equipment
Select sampling objectives and determine
route and sampling locations
0200 - 0220 Travel to first location
0220 - 0320 On station
0320 - 0335 Travel to second location
0335 - 0435 On station (30-minute lunch break)
45
-------
SULFATE SAMPLING LOG - MOBILE UNITS
Mobil Unit - Suburban , Helicopter
Grab Sampler # 1 (A.C.) # 2
Observer
(D.C.)
Date
Filter Number
Time
On
Time
Off
Indicated
Flow On
Indicated
Flow Off
Notes:
Figure 13. Log for sulfate sampling operations.
46
-------
Elapsed Time Activities
0435 - 0455 Travel to third location (includes valley
to ridge traverse)
0455 - 0555 On station
0555 - 0610 Travel to fourth location
0610 - 0710 On station
0710 - 0730 Return to central site
Figure 14 is a flow diagram showing the typical operations of the mobile unit.
The Surface Mobile Sampling Log is shown in Figure 15.
Mobile Unit Calibrations--
Analyzers in the mobile unit were calibrated at the base station
using the Monitor Labs Model 8500 calibration unit, which uses S0? and
NOp permeation tubes and precision dilution of bottled gas with certi-
fied analysis of NO. Ozone calibration was accomplished using precisely
generated quantities of ozone. A detailed description of the mobile unit
calibration procedures is given in the Quality Assurance Section.
Power Plant Emission Monitoring
The emissions of S02, NO and exhaust gas characteristics of the two
Clinch River Steam Plant stacks were monitored by obtaining the following
measurements:
• Concentrations of NO and SOp in one stack breeching
• Temperature of exhaust gas in each stack breeching
• Mass flow rate of air into each boiler
• Net output load of each generator
• Daily laboratory analyses of the sulfur content of
the coal delivered to the plant.
The concentrations of NO and SOp in ppm were measured by an EDC
stack gas analyzer. These measurements were recorded continuously on
strip charts and every 2 minutes on magnetic tape by the data acquisi-
tion system. Hourly averages of the temperature (measured at stack base),
the air mass flow rate (measured in boiler intake), and the net output
load from each generator were furnished by the plant each week. Daily
47
-------
nt
lin
M
fo
Plume Position
Steadiness of Pl
Anticipated Plu
Probable Locati
o
jsi
h Sampl
s
<
II
•S £
5 9
s Is
1 ^
o o
5 3
« Z
^ "
1
V
a
UC
1
«
Z
'S o
u o R
« i" f ? s-
™ -J Wi -J S
" g '
'
S
III
,0
O
M
•S
* -o
5
u
a
u
"
S!
6
1 5
•£ 1 - ^ 1 |
2 t 1 s ° = 1
1 S = I |3u
~-l ! ^ l< i t
OOtSCoiSh-P
1
9
a
^
1
o
j;
1
o
n b
J ! Sli
h Q t •«
1 ~« 5 « t 1 i
1/5 5- """ ^ ^ S t *
t§ •* •«• •» 3 f
* & S S Z o
§
•R
o
-------
Surface Mobile Sampling Log
Date:
Mobile Unit #:
Observer:
General Weather Conditions:
Sampling Time:
page 1 of
to
Estimates of
Cloud
Pibal
Wind
Height (ft.)
Amount (tenths)
Low
Middle
High
Wind Direction
Wind Speed
219-420m
420-622m
622-81 3m
Observed Plume Direction (near stack):
Do vapor plumes from cooling towers intersect stack plumes? Yes
Weather Phenomena (e.g., fog, rain, snow, smoke, etc.):
, No
(continued)
Figure 15. Log for ground mobile sampling
49
-------
Surface Mobile Sampling Log continued
Page 2 of
Sampling Route (in general):
Map
Event
Mark
Time
Speedometer
Reading
Remarks
Figure 15. (continued)
50
-------
laboratory measures of the sulfur content of the coal shipments received
were also furnished by the plant. The plant also furnished hourly averages
of the makeup water used in the cooling system which was a measure of the
water vapor emitted from the cooling towers.
Exhaust gases from the third boiler were released from Stack No. 2
in which the exhaust gas monitor was located. Calculated emission rates
of S0? and NO were prepared for each of the two stacks using the collected
data and the methods described in Section 5. Both measured and calculated
values were available for Stack No. 2 which services one boiler.
There is one other significant source of S0? or NO emissions in
the vicinity of the plant and that is a coal preparation* plant which is
3 km north-northeast of the power plant (see Figure 1). This plant when
operating at its maximum capacity 24-hours a day could burn 60 tons of
coal per day, which is 1 percent of the normal coal consumption rate at the
power plant. The only monitor on which it appears the coal plant emissions
had a significant effect was the Tower site, 1 km away. When the wind was
west-northwest, it blew directly from the coal preparation plant to the
Tower site, and spikes of 25 to 30 ppb in S02 readings were recorded
on a strip chart. These spikes were easily aistinguished from the
effects from the power plant which occurred with a southwest wind direc-
tion.
Stack temperatures for both stacks 1 and 2 were measured by thermo-
couples placed in the breeching about one-third of the way from the precip-
itator to stack inlet.
Pressure was measured in the control room. A mercury column barom-
eter was used and was uncorrected, so the pressure measured was the atmo-
spheric pressure above the power plant control room.
Airborne Mobile Sampling
A Bell Jet Ranger helicopter was used to fly a set of air quality
analyzers during two intensive study periods. Approximately 75 hours of
flight time were logged during the intensive study periods.
Air quality measurements of S02, NO, N02, NO and 03 were continuously
scanned and recorded on a digital cassette tape unit which also recorded
data and time at the beginning of each scan. During the second study period,
temperature, relative humidity, altimeter and indicated air speed were recorded
on cassette tape also. Three strip chart recorders were used to display the
concentration trace of SO^, NO and 0,. An observer and a technician rode
with the pilot during all flights to operate the instruments, make necessary
notes on time and location and to direct the sampling operation. Communica-
tions between the helicopter crew and other GEOMET personnel conducting the
ground operation were maintained using two-way business band radios.
51
-------
The objectives of the airborne monitoring work were to obtain enough
data to describe the location and/or dimensions of the power plant plume
under a variety of meteorological conditions. The flight procedure to
obtain a cross section of the plume at various distances downwind along
the plume path is illustrated in Figure 16.
During the first sampling period more emphasis was made on deter-
mining plume location and the behavior of the plume with respect to
topographical features. During the second sampling period emphasis was
primarily on plume dimensions at several downwind distances within 10 km
of the plant.
Event marks were made upon entering or exiting the plume as deter-
mined by the technician watching the strip chart recorder. A log of altim-
eter, air speed and heading values was obtained by reading from the panel
in the cockpit whenever an event was recorded. During the first sampling
period aircraft locations were marked on a 1:24000 scale topographic map
whenever an event was recorded. In the second sampling period ground check
points were selected which were outside of both sides of the plume. The
helicopter flew between these check points at different altitudes. On
each traverse the time that the aircraft passed over the checkpoint was
noted and assigned an event number.
The observer rode in the front seat beside the pilot and performed
the following functions:
• Manually record position, time, event, and pertinent
notes - maintain log of the sample period (Figure 17)
t Communicate with base station regarding aircraft activi-
ties, plume location, etc.
• Communicate with pilot regarding sampling strategy,
location, etc.
The technician rode in the rear seat with the instruments and performed the
following functions:
• Turn on and off analyzers and recorders
• Make event marks on strip charts and data logger
• Observe when entering or leaving plume boundaries
and inform observer when each event occurs so that
it may be entered in the written log
• Make necessary adjustments to instruments.
52
-------
Figure 16. Representative helicopter flight path at fixed distance from the power plant.
53
-------
HELICOPTER SAMPLING LOG
Of
Observer:
Date:
Sampling Time:
to
M8 Package Airspeed ( )
M8 Package Altitude ( )
M8 Package Temperature ( )
MS Package R.H. ( )
DL 640 and Tape Recorder ( )
Strip Chart Recorder ( )
Equipment in Operation:
S02 Analyzer (Sign X) ( )
S02 Analyzer (Theta) ( )
NO, N02, NO Analyzer ( )
Ozone Analyzer ( )
Participate Grab Sampler ( )
MB Package Heading ( )
General Weather Condition:
Measured Plume Elevation (m)(MSL):
Wind Direction at Plume Height (from Helicopter Plume Intercepts):
Wind Speed at Plume Height (mps)(from Pibals or State Other Source):
General Sampling Direction and Ranges:
Position
Ident.
Time
Direction and
Distance from
Plant (Km)
Observations & Remarks
(Landmarks, Comments, etc.)
* Participate grab sampler not used in helicopter sampling program.
Figure 17. Log of helicopter sampling.
54
-------
AIRBORNE SAMPLING LOG
DATE OBSERVER
PAGE TECHNICIAN
S^
o
r*
ra
Figure 17. (cont'd)
55
-------
Zero and calibration tests were performed on the Theta Sensor, NO, NO
analyzer and ozone analyzer before taking off each time a sampling run wasxto
be made. During the second intensive period zero and span data were recorded
after each flight. The analyzers were left running each night to reduce
warm-up time and to reduce drift which might be enhanced by turning the
instruments on and off.
Approximately 30 min before take-off, the helicopter observer called the
power plant base station to get an update on the plume projection and its cal-
culated height. A sampling procedure was discussed and decisions made as to
whether or not they would start off by plume chasing or by getting plume
characteristics.
Upper Air Soundings
To completely characterize the meteorological conditions, upper air tem-
perature and wind data must be known in the vicinity of the Clinch River plant.
The National Weather Service network of soundings taken around the United States
does not include a station near the southwestern Virginia area. The closest
station at Huntington, West Virginia, is about 150 km away. Temperature and wind
soundings taken at the Clinch River site give results most representative of the
surrounding region.
On a daily basis Pilot Balloons were launched at specific times to charac-
terize various meteorological situations. Temperature soundings were obtained
during some periods using the Contel Corporation "MiniSonde" in conjunction with
a Pibal. Temperature data was transmitted back to the ground by a small radio
contained in the Mini-Sonde. The signal was translated into continuous trace of
temperature recorded on a strip chart.
Pibals and Pibal-T Sondes were tracked using two theodolites beginning on
June 2, 1976. Although double theodolite procedures predominated, single theo-
dolite measurements were used if a theodolite station failed during a particular
sounding.
A diagram of the sounding location is found in Figure 18. The baseline used
was selected at the time of each sounding to use a baseline that was not parallel
to the direction of flight in order to increase the resolution of balloon posi-
tions determined by the two theodolite observations.
The number of soundings per day changed from one a day, at the beginning of
the project to three or four a day at the end. T-Sondes were not incorporated
until late 1976 with at least 12 per week being flown at the end of the project.
A precision machined 30g balloon filling device was used to fill the pibals.
Helium was added until a 30g weight just lifted off the ground. When a T-sonde
was attached to a pibal being tracked by a single theodolite, additional helium
was added to the pibal to compensate for the extra weight of the T-sonde,
thereby maintaining the proper buoyancy.
56
-------
Clinch River
Power Plant
Clinch River
Double Theodolite Baselines
Figure 18a. Diagram showing location of double theodolite baselines
relative to the Clinch River Power Plant near Carbo, VA.
N
151.5m
Figure 18b. Dimensions and angular arrangement of double theodolite
baselines located just south of the Clinch River Power Plant near Carbo, VA.
57
-------
All azimuth, elevation and temperature data were introduced into a Texas
Instruments SR-52 calculator program for reduction in the field. This method
gave wind speed, direction, temperature versus height, and vertical wind infor-
mation based on trigonometric reduction of two azimuth angles and one elevation.
Subsequent processing of the data with a POP 11-70 computer system determined
position based on two azimuth and two elevation angles and the most probable
location of the balloon based on the angular inputs.
Plume Photography
Associated with the intensive study period of July 18 through July 28, 1977
was a program of intensive plume photography. Three locations were chosen within
a few km of the plant that had unobstructed views of the plant smokestacks. Site
A was located on a bearing of 108" from the plant at a distance of 6 km, Site B
was located on a bearing of 165° from the plant at a distance of 4 km, and Site C
was located on a bearing of 290" from the plant at a distance of 3 km (see
Figure 19).
The sites were chosen so that regardless of the wind direction at least one
site would provide a good angle of view of the plume as it exited the stacks and
traveled downwind. However, due to the synoptic conditions during the intensive
sampling period, only sites A and B were actually utilized.
The photographic program consisted of taking sequential photographs of the
power plant plume while the Unit 1 electrostatic precipitators were reduced.*
Having the precipitators reduced allowed for better visual observations of the
plume by both the photographic and airborne sampling teams. On 2 days of the
program, the Unit 3 precipitators (serving the second stack) were also reduced in
an effort to see at what point downwind the two plumes merged.
The photographic team used a Nikon F 35 mm camera with motor drive and
interchangeable 200 and 400 mm lenses. The camera was mounted on a tripod and
aimed to take in as much of the power plant plume, while still including the
stacks, as the angle of view would allow. The photographs were taken at 2-minute
intervals using Tri-X 400 ASA Black and White film. Very clear plume photographs
were obtained by using a combination of red and polarizing filters to cut through
local haze. Detailed records were kept to accurately log the time of each
photograph as well as any significant meteorological phenomenon that may have
occurred during the day. The photographic team was kept informed of any possible
wind shift trends by the base station personnel via the two-way radio. When
large shifts in wind direction indicated that one of the other photographic sites
would provide a better view of the plume, the photographic team was able to move
their equipment from their present location and set it up at the other site with
only a minimum loss in photographic time.
* Traning down of the precipitators was coordinated with and reported to state air pollution control authorities.
58
-------
u
"S
a
TI
0)
4>
T3
8-
I
cu
I
59
-------
In 5 days of picture taking, approximately 700 plume images were
obtained. Some examples of these are presented in Figure 20. The objec-
tive of the photographic program was to obtain plume height estimates based
on each photographic image and to prepare a data file containing observa-
tion identification (date, time, location), location of "maximum" plume
height, plume width, and associated meteorological and plant operational
data for further analysis. Comparisons could also be made between the visu-
ally measured plume height and the height of maximum pollutant concentrations
as determined by the airborne sampling program.
QUALITY ASSURANCE
Record Keeping
A comprehensive set of records was maintained in order to permit
detailed tracing of the data collection process when future reviews are
made of the Clinch River data. Records of types of data, where collected,
when collected, how collected, by whom, how and when the data was trans-
ported and where it was sent are all items that have been accounted for.
To acquaint the reader with the supplemental information that is
available, a brief explanation of the individual log sheets follows.
Any data that left the field office was properly logged out so that
there was an identifiable record of the particular data sent, the date
it left, where it was headed and the mode of transportation. There was
a complete log of every item of datum that was collected in the field.
Whenever a strip chart was removed from a field station and transported
to the base station, this was logged in the station manual as well as
the office log. Whenever that segment of data was then sent either
to Gaithersburg or to a particular laboratory for analysis, a complete
log was available as to the whereabouts and status of that data.
In the following subsections, as each item of data is discussed, the
various forms and procedures for transfering the data from one location to
the next will be included where appropriate.
A Station Check List (Figures 21 and 22) was completed during each visit
to a fixed monitoring site and provided an account of the status of all
equipment at a station. The procedure for completing this form was described
under BASE STATION OPERATING PROCEDURES.
Any visit to a station was also logged in the station log book. If any
work was done or changes were made, the procedure was carefully outlined in
the log.
To determine the S02 and NOX emissions and exhaust gas characteristics
of the plant, operating records were obtained weekly from the Clinch River
Plant records. Sample data are presented in Figures 23, 24, and 25. Gross
generator load, boiler intake air flow rate, and stack gas inlet temperature
were obtained from these data sheets. Coal percent sulfur data was obtained
from records kept by the power company's chemistry laboratory.
60
-------
Figure 20. Sample plume photos.
61
-------
Date
Time
CLINCH RIVER PLUME STUDY
STATION CHECK LIST
_^____ Location
EST Observer
OUTSIDE
Trailer Appearance
Wind Systems: Cupj Rotating | |
Vane(i) in Good Shape | |
Temperature Asperaton Working j j
Hi-Vol Appearl to be OK' | |
Tower General Condition: Ground Cable [ |
Sen»ot Cables | |
BoomjOK | |
Trailer Locked? | |
INSIDE
Clock Time Check OK
Fast
Slow
Hr
Min
General Appearance Inside Trailer: Inlet Sink Trap Clean | |
Manifold Clean | |
Inlet Lines Connected | j
Exhaust Lines to Exhaust Fan | |
SO _ 8450 Both Fans Running Q
Mark with X all lights that are on and all switches that are in an In position. Mark with F all flashing
lights.
Write in H, Flow Rate
FCASCOKTM
,
Hj Vatvm
a a
©
HjAdJ
O
.«- o
Q:
5TATVS
Pump Q
aTto. o
a «— o
en MO— o
az» o
a 'P- o
Suite Slaainr
TtiM COMCUC
I 5 ZO M 120
1 ' I 1 I 1 toe
Avm*i«
Oft S 10 10 M
! i — j — [ — i — iMta
"• — -•— RANGE — — I
! 1 AUTO O
•^ 100 pp6 Q
1 1000 ppb Q
oo
OPERATION j
n »- o |
i i ~«« '
COP«^ o |
7— U |
1
"« D ^.Q
Sulfut Vionrtot
Model S4SO
Figure 21. Station check list, page one.
62
-------
Calibrator 8500
Oven Temperature ^_ °C
o
TEMT
O
RUN
D
HUN
D
STW
a
_0tt_
NO, 8440
J
O O 0
n o« o
i Q o« Q
3/o Dnerite gone
Zero/span/monitor switch (on back) in monitor mode?
DATA UOGCER 9400
TUT MTTtR
o rm o
Any Negative Readings on Active Channels? Ye* | j No [_ ]
Channel Data
Day Hour Mm Sec
Data
O
System
Power
CONTROL
First Point
Last Pome
MON
A
SINGLE
CONT
MANUAL
REMOTE
RECORD
B
MDE
J START
I RESET
TIME
DATA
MON
AUTO
CHART RECORDER - Mark Date/Time Sticker
H2 Tanks - Line Pressure psi Tank Pressure
How Many Tanks
NO Tanks - Line Pressure
>si Tank Pressure
J?S1
psi
!•»
Dm
Figure 22. Station check list, page two.
63
-------
01
1-4
tC
duI3i
13TQI n°^S
OHu
ssajjj -jBg
OHu
doaajBM *«0
°JID "I
iBjox MXW/OLS
8801 31BH JB3H
S^OUJ
3jn AD iLcLKI
^ u £ JlV~3tt'O
SEQ JIH JIV
§AV
MW
pBO"l SSOJ3
ISd
ssaJJ HHJ.
IBIO.L
*H/# "XXV JLHiH
SAW 4
WEns x°(a
SAV Jo
»«*U
0
CM
*
CM
O
00
CM
CO
CM
CM
Oi
s
in
00
00
CM
CM
VO
in
8
p
O
O
ot
in
8
o
§
^_,
CM
VO
oo
CM
O
CM
CM
tv
CM
in
CM
VO
^
oc
o
c
o
s
c
s
s
tn
CM
in
CM
VO
00
CM
00
o
CM
oo
in
00
CM
CO
00
CM
vc
in
CM
O
o
CM
O
s
If
o
8
o
CM
in
CM
01
00
CM
y-t
CM
O
oo
in
in
CO
o
C1
CM
VC
in
a\
o
Cv
O
o
If
s
0
CM
^
CM
O
00
CM
CM
CM
00
S
in
CO
00
CM
CN
in
O1
p
c
o
c
If
if
o
CM
O1
CM
O
00
CM
CM
CO
CM
tn
oo
rt
8
in
CM
«-
in
CM
o
s
o
O1
s
8
o
T-4
o
CM
o
f-
oo
CM
N
CM
oo
in
O
oo
O
oo
CM
in
o
CM
O
C1
O
O
in
8
o
iH
in
CM
S
OO
CM
VO
CM
CM
in
00
£
01
in
CM
in
O
CM
0
2
o
o
s
o
8
0
•rt
in
CM
-
00
CM
8
CM
OO
CTi
CM
O
o
CM
in
O
CM
CO
O
«
O
oo
m
S
ft
0
CM
O
CM
0
OO
CM
in
oo
vd
CM
CO
Cl
oo
in
0
OS
CM
to
o
tn
o
o
CM
O
8
o
cS
8
o
VO
K
CM
CTi
VO
00
CM
in
CM
CM
CM
O
s
CM
CM
in
ro
CO
00
vo
00
CM
o
cs
0
CO
in
cc
8
o
00
CM
VD
00
CM
00
vd
CM
S
T-i
S
-
S
CO
00
VO
00
o
O
CM
0
in
O
O
S
Cv
g
VO
CM
CM
VO
oo
CM
oo
CM
vd
CM
CM
O
00
K
in
TH
CO
00
in
vo
CM
t?
C\
0
3
§
8
CO
O
in
CM
CM
VO
00
CM
oo
CM
vd
CM
o
1-1
CO
00
K
in
CM
oo
oo
«
?
c\
o
3
in
3
8
0
in
^
CM
VO
oo
CM
in
CO
vd
CM
oo
01
oo
^
tn
K
CM
O
00
01
tn
o
0
VO
to
o
s
o
§
0
in
M-
CM
VO
00
CM
VO
ro
CM
oo
01
00
tn
ro
§
CM
O
oo
o
fv
o
CM
o
o
CM
o
3
o
if
VO
0
m
vo
CM
in
VO
00
CM
00
CM
vd
CM
ffl
CT>
CM
00
in
CO
g
CM
OC
01
vo
CM
O
o
CM
O
s
If
o
8
f-
0
in
00
CM
CO
VO
00
CM
in
CM
vd
CM
o
o
00
in
CO
§
CM
O
oo
Ol
s
CM
o
o
CM
O
i
§
00
o
CO
TC
CM
m
VO
00
CM
vd
CM
Jl
in
oo
-
in
00
CM
in
O
in
00
?
c
o
c
o
3
1
o
CO
m
CM
VD
VO
00
CM
f-.
>d
CM
fe
O
00
5
o
00
CM
in
fs.
^
o
c
o
c
I/)
s
§
CO
tn
CM
f-
VO
oo
CM
t-~
vd
CM
CT)
CO
oo
in
CO
o
oo
CM
o
oo
01
CO
00
o
c\
o
o
1
8
o
CM
CO
CM
t^
VO
00
CM
ro
vd
CM
CO
01
1-t
oo
CO
in
oo
CM
O
tv
CM
0
O
C
o
o
tn
3
8
o
in
CM
VO
oo
CM
8
CM
CM
O1
O
OO
vo
t-.
co
oo
CM
CM
VO
CO
o
c
O
8
O
u
o
in
0
CM
O
s
^
s
01
VO
00
VO
CM
rl
VO
CO
O
2!
o
8
o
m
o
8
ff
•c
S
•a
0)
+j
rt
o
o
O
.3
o.
S
CO
CM
64
-------
WEEKLY h£Ai m a LC3J
CLINCH HIVER PLANT
WEEK ENDING
OCT201877
KEY TOUCHED
1 Load (Net MW)
2 Abs. Press. (In Us)
3 Am Power (%}
4 Main Steam (F)
5 RH Steam (F)
fa Throttle (PSIG)
7 Excess Air (%)
8 Make Up (%)
V KH Attemp. (M#/Hr)
UJ Hf Turbine Eff. (%)
11 HH Turbine Eff. (%}
12 Htr. 8 TTD (F)
13 Stack Temp. Diff . (F)
U Air Htr. Lkg. (%}
15 Coal Moisture (%\
lt> Total Ash Comb. (%)
J.V i'rtra load (Hrs;
18 Start Ups
19 Unknown
20 Heat Rate
21 T\i Losses
i2 Net Load (MWH)
23 Coal BTU/LB
±4 Coal Burned (Tons)
Uni
Design
Value
223
1.50-
3.50
1050
1050
2(560
15
.84
0
83.03
tJO.tXI
0.0
/7j*
7.3
4.50
2-^~
XXX
XXX
AAA
J^JX
u
AAA
XXX
XXX
t 1
Aot.Avg
Value
/ft
i-o}]/,^f
jtyj"
/oy?
/OJ3
2032
*./
O. 7O
9, •//
7
/ ' &tf
/&
^f".£^
4.S-S
0
a
AXX
..&?/-?
J/273
// ef&
//.9j"l^
BTU
Loss
XXX
-^?
///
/
/<*
y
- ^
/^
7/3
- 7
<)
- £
i-f
9
-34
6
0
SO
XXX
XXX
XXX
XXX
XXX
Un
)esign
/alue
223
1.50^
3.50
1050
1050
2000
15
.84
0
83.03
Sb.bU
0.0 ,
/ 7 7*
7.3
4.50
j? j'
XXX
XXX
XXX
r/^/
0
XXX
XXX •
XXX
Lt 2
Aot.Avg
Value
/./ 7
.*&;/. rf
¥.74"
/fit
/a/6
U3£
23
i.7
*4*J
XXX
XXX
XXX
,
-------
CR 60
3/9/6Y
20CX1
DAILY ijiiNEKATfNG iOC
unit UNH- 1
•Date
SEP231977
KEYPUNCHED
AU6241977
EXTRA LOAD: Time ON .
Time OFF
Figure 2B. Sample operating record - daily generating log.
66
-------
Each day of the week a Daily Project Report was completed to record all
events of that day. This form was completed at a time when all staff mem-
bers were together to discuss the events of the day and is illustrated in
Figure 10.
Mobile Sampling Logs provided records of mobile data which were very
important in putting the data into the proper context. Vehicle location,
type of sampling (mobile cross-plume or stationary sample), local terrain
features and significant events were recorded for the analysis of mobile
data. Range settings and output voltages were recorded for each instru-
ment. An identification label was affixed to each strip chart or tape to
properly identify the data.
During the airborne sampling an extremely detailed series of notes
was kept to properly identify the helicopter location. Types of equip-
ment, ranges and outputs were also recorded to reduce the data to a
usable form. Channel input to the data logger was recorded at the
beginning of each sampling period to properly identify which instruments
were writing on each channel. This information was recorded for each
helicopter flight on the Helicopter Sampling Log and Airborne Sampling
Log shown in Figure 17.
Regular soundings of wind speed and direction were recorded on the
Double Theodolite Form (Figure 26). Temperature data obtained by suspending
a T-Sonde from the Pibal were recorded on the T-Sonde Data Report (Figure 27}
The Pibal data were initially recorded by voice on a cassette tape recorder
while the temperature sounding was plotted on a strip chart. Reduction of
this data was performed in the field and recorded on the appropriate Pibal
or T-Sonde forms.
Numerous sulfate samples were collected during the Clinch River study.
To properly identify the conditions under which each sample was collected,
the form shown in Figure 28 was completed. Information as to when and
where a sample was collected was important for proper analysis of sulfate
samples. Records of mobile sulfate data were recorded using the form
shown in Figure 29.
A log of all computer interventions was recorded on the form shown
in Figure 11. A new log sheet was kept for each new tape recorded.
The form used to record data on which the plume travel was projected
during a specific period is shown in Figure 30.
Calibration
Fixed Station Calibration--
The folowing procedure was followed when performing analyzer calibra-
tion at all the fixed stations. No adjustments were made to any analyzer
until a complete investigation was made to determine the reason for the
drift of an instrument.
67
-------
DOUBLE THEODOLITE FORM
CLINCH RIVER POWER PLANT PROJECT
(TO BE USED WITH SR52 PROGRAM)
Values to be tiled in SRS2 Program
Punch in this order Elev , Atm^, Azm,
Minute
0
1
2
3
4
3
6
7
3
9
10
11
12
13
14
IS
16
17
18
19
20
Release Station
Elev1
X
Azm
X
Satellite Station
Elev2
X
Amij
X
ws
m/s
X
WD
Corrected
To North
X
a
z
X
c
AZ
X
Cbiemn
(SatelU:i)
Baseline Used (i.e., AC, Afl, BC, TW, or single)
Weight of Balloon ______________ gins T-Sonde Attached
D«e _________^____________ Starting Time __
Attracted and Calculated by ^^^^^^^^^^^^^^^^^^^^
Yes
No
4/77
Figure 26. Double theodolite form.
68
-------
T-SONDE DATA REPORT
ABSTRACTED BY
DAY:
TIME
( )
HEIGHT
( )
SCALE
(1-8)
SONDE READING
TEMPERATURE
Figure 27- T-Sonde data report.
69
-------
SITE:
FILTER NUMBERS:
DATE/TIME ON:
DATE/TIME OFF:
NOTES:
_through
COLLECTORS NAME:
DATE COLLECTED:
For those filters which are discolored, Indicate which shade of grey
most closely represents the discoloration. CNo entry for no discoloration
4 - anything darker than 3.)
Figure 28. Sulfate sample collection form.
70
-------
SULFATE SAMPLING LOG - MOBILE UNITS
Mobil Unit - Suburban , Helicopter
Grab Sampler # 1 (A.C.) # 2
Observer
(D.C.)
Date
Filter Number
Time
On
Time
Off
Indicated
Flow On
Indicated
Flow Off
Notes:
Figure 29. Mobile sulfate data form.
71
-------
Plume Projection Form
Clinch River Power Plant Project
Date
Observer
Time
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
HOCKEY
Wind Speed
Wind
Direction
TOWER
Wind Speed
Wind
Direction
Distance*
(km)
T (Tower)
CO
Plume
Height
* If Tower wind speed is used to calculate distance, please note by putting
"T" in the column next to the distance. Multiply by 3.6 to convert wind
speed in (rn7sec) to kilometers travelled in one hour.
Figure 30. Plume travel projection form.
72
-------
The multipoint check was performed as follows:
Calibrator Steps:
1. Disconnect calibrator from computer command by
removing eight-pin ampherol connector at rear of
calibrator.
Note: Calibrator should be in STBY state when
making this disconnection (see Figure 31).
NO
RUN
D
Oa
D
fl
D
Off
Figure 31. Model 8500 Calibrator
2. Install eight pin ampherol connector (with jumper
between pin A & B) into the same socket that the
computer link was removed from.
3. Turn calibrator to run position (see Figure 31).
S02 Instrument Steps:
1. Turn Monitor Labs 8450 Sulfur Analyzer to zero
mode by depressing the zero button on the front
panel (see Figure 32).
2. Allow analyzer 1/2 hour to stabilize on zero air.
After this period has been completed, read the con-
centration on the digital display of 9400 Data Logger
and record the valiM omtKe calibration graph.
3. Turn Monitor Labs 8450 Sulfur Analyzer to span mode
by depressing the front panel span button (see Figure 32),
73
-------
Air H2
H Valve
cu o
HgPresm
o
0
is-*, o
rr
1 ISP..
v_y
I STATUS 1
Ignition £\
Temp«ratura ()
Pump ^S
| | Tlm« O
I | Remot* Q
a z«° o
1 | Span n
r.s -o
Sulfur Monitor
Time Conftant
1 5 20 60 120
[_ i 1 ls«
Average
Off 5 10 30 60
r | i IMI-
1 1 AUTO O
I | 100 ppb (J
1 1 1000 ppb O
Zero Span
r^^\
( ) ( )
v_y \^
j~| Moato. O
O
Te«t v^
D?^11 o
OB U PweIQ
o«Q w
SuUur Monitor
Model 3450
Figure 32. 8450 Sulfur Monitor
4. Set the S02 flowmeter on the Monitor Labs 8500
calibrator for an actual flow rate of 2500 cc/min.
5. Allow 1/2 hour for the analyzer to stabilize on
span gas and then record the concentration on the
calibration graph.
6. Continue recording span concentrations for 2000,
1500, and 1000 cc/min. calibrator..flow rates.
Note: Wait 15 minutes between recordings to allow
for stabilization at a given concentration.
7. Return Monitor Labs 8450 Sulfur Analyzer to monitor
mode by depressing front panel remote button (see
Figure 32).
8. In addition, the following information should be
recorded on the calibration graph.
a) Time out of service
b) Time in service
74
-------
c) H2 flow
d) Range
e) Time constant
f) Calibrator temperature
g) Calibrator serial number
h) Analyzer serial number.
Oxides of Nitrogen Instrument Steps:
1. Place toggle switch, on back of Monitor Labs 8440
Oxides of Nitrogen Analyzer, in zero position (up),
/^\
0
NO, NO OZONE
Ova* Cow Pwr
0 O O
0. n On D
o« n o« n
NO
S 20
CooBMl
__
GO
O 0
TEST METER SELECT
Ear
Monitor ElM Optic T«t NOX NO NOj D.U
o rm o rm o
Spaa
Figure 33. Model 8440 nitrogen oxide analyzer.
2. Allow unit to stabilize in the zero mode for 1/2 hour.
3. Record results on calibration curve.
Note: Do not use front panel meter for NOX concentra-
tions. The results should be read on the display
of the 9400 Data Logger.
4. Place the toggle switch at the rear of the analyzer in
the span position by pressing the switch all the way
down.
-75-
-------
5. Adjust the NO rotometer of the 8500 calibrator to
give an actual flow at 2500 cc/min.
6. Allow 1/2 hour for the analyzer to stabilize.
7. Record the NO, NOX concentration on the calibration
curve.
8. Record concentrations for calibrator flow rates of
2000, 1500 and 1000 cc/min. allowing 15 minutes for
stabilization at each concentration.
9. Disconnect the supply hose from the intake filter.
10. Connect N02 supply line from 8500 calibrator.
11. Allow 20 minutes for the analyzer to stabilize.
12. Adjust N02 rotometer to flows of 2500, 2000, 1500,
1000 cc/min and record corresponding concentrations
as read from 9400 display.
13. Plot graph of NC>2 input versus output and calculate
converter efficiency, i.e., output/input.
14. Record the following on the calibration graph:
a) NO, NOX, 03 flow rates
b) Range
c) Time constant
d) Time taken out of service
e) Time in service
f) Calibrator temperature
g) Analyzer serial number
h) Calibrator serial number.
15. Return the Monitor Lab Oxides of Nitrogen Analyzer to the
monitor mode by placing the zero-span toggle switch, at the
rear of the analyzer, to the center position which is marked
remote.
Calibrator Steps:
1. Turn the Monitor Labs 8500 calibrator to STBY position.
2. Remove the eight-pin ampherol jumper plug from the cali-
brator and replace the computer command plug at the same
location.
76
-------
3. Return the Monitor Labs 8500 calibrator to the run position.
4. If SOg or oxides of nitrogen concentrations are more than 20
percent from the theoretical values, an investigation should
take place to determine the problem before any adjustments
are made.
Calibration Curves--
Field calibration curves for NOX, NO, and S02 were generated and
identified using the following information:
1. A theoretical curve based on the flow rate of the calibrator
for theoretical span concentrations
2. An actual curve of indicated flow rate of the calibrator versus
concentrations measured by the analyzer at the same indicated
flows that are used in the theoretical curve
3. Calibrator temperature
4. Calibrator serial number
5. Analyzer serial number
6. Time out of service for calibration
7. Time in service after completion of calibration
8. Zero concentration
9. For $02: H2, sample (air) flow
10. For Oxides of Nitrogen: NOX, NO, and 03 flows.
Hi-Volume Sampler--
The Misco 24-port sequential sampler was calibrated miJttemn orifice
calibrator that has been referenced to a positive displacement meter. The
procedure for individual port calibration was as follows:
1. Turn remote switch to local and remove filters from retaining
rings
2. Connect orifice to sampling head per manufacturer's instructions
3. Connect U-tube manometer to port on orifice can
4. Place new circular chart on recorder
77
-------
5. Set sequences to port #1 and turn hi-vol to on position;
allow about 3 to 5 minutes for flow to stabilize
6. Note orifice number, change in height of water in
manometer and chart deflection.
7. Repeat procedure for 23 additional ports.
8. Change orifice restriction and repeat procedure for
all 24 ports (should be done for at least four restric-
tions, #5, #9, #10, #15).
9. Allow a single port to run for one sampling increment.
Record the exact sampling time run and the reset time
from the end of the sampling run to the start of
sampling for the next port.
10. Return sampler to remote operation after replacing
filters.
Flow Calibration--
8500 Calibrator—All flows in each calibrator were checked every 6
months to verify the dilution effects. The following procedure was fol-
lowed:
1. Observe the sample outputs on rear of 8500 calibrator
(should be 2 to 4 outputs depending on calibrator).
2. Note that each output consists of a supply line and
a vent.
Supply to analyzer
Vent~24 - 28"
Calibrator
78
-------
3. Remove supply to analyzer and cap opening.
4. Connect bubble flow meter to vent line. (Set up meter
per manufacturer's instructions.)
5. Adjust rotometer to desired position (i.e., 2500, 2000,
1500, 1000 or 500 cc/min.j.
6. Generate bubbles in flow meter.
7. Record length of time that bubble takes to traverse
limits of volume. (Note volume may have to be changed
to get good reading of time required.)
8. Repeat process for remaining flows on each parameter.
9. Repeat for all parameters.
10. Plot graph of indicated flow versus actual flow and
record indicated flows for 2500, 2000, 1500, 1000, and
500 cc/min.
11. Unplug supply line and reconnect supply line.
NO Supply Flow--
1. Open cover of 8500 calibrator.
2. Remove NO capillary supply line from tee connection.
3. Connect capillary to bubble flow meter.
4. Generate bubbles and record time to traverse the desig-
nated volume.
5. Repeat procedure for different NO flows which are
regulated by the front panel NO pressure regulator.
6. Reconnect NO capillary to dilution Tee.
Mobile Calibrations (Ground and Airborne)
In calibrating instruments in the van and the helicopter readings
were taken with instruments connected to the field generator power supply.
The mobile calibrator was placed inside the vehicle and put in the RUN
mode. Calibration graphs were prepared with the same information that
appeared in the fixed station procedure.
79
-------
Calibrations—The procedure for calibrating the SCL analyzers
was as follows:
1. Connect zero air supply to Theta Sensor intake.
2. Allow to stabilize 15-20 minutes.
3. Make necessary adjustments for zero reading.
4. Record reading of what appears on assigned channel of
data logger.
5. This reading will be volts and must be converted to ppm.
6. Disconnect zero supply and connect SO,, supply to intake.
7. Adjust 8500 calibrator S0? channel to a rotometer
reading of 1500 cc/min.
8. Allow to stabilize for 15-20 minutes.
9. Make necessary adjustments to Theta Sensor for proper span.
10. Repeat procedure for 2500, 2000 and 1000 cc/min dilution
flows and plot on calibration graph.
11. Return Theta Sensor to sample mode by reconnecting sample
line to intake.
NO-NO Calibrations—Because an NO dilution system did not exist in
the mobilexcalibrator, NOp was used as the calibration gas for the NO-NO
instrument. This was done by passing N0? through the Molybdenum converter
in the 8440 NO analyzer and utilizing tne efficiency of the converter.
The converter efficiency was determined at a fixed station that had an NO
dilution supply. This measuremement was made every 3 to 4 months. The cali-
bration procedure followed in the mobile vehicles was:
1. With NO analyzer in proper operating mode, connect zero
supply to intake.
2. Allow system to settle for 15-20 min.
3. Make necessary zero adjustments and record final reading on
calibration graph.
4. Introduce N02 supply to 8440 NOX analyzer.
5. Adjust N0~ rotometer in 8500 calibrator to 1500 cc/min.
80
-------
6. Allow system to stabilize 15-20 minutes and then make neces-
sary span adjustment to NO channel only.
A ' '
7. Record reading of NO calibration graph.
/\
8. Repeat for dilution flows of 2500, 2000, and 1000 cc/min.
9. Interchange the NO and NO interconnect lines on back
of 8440 NO analyzer. This is done to allow N02 supply
to pass through converter so that the NO channel may be
calibrated.
10. Set N02 dilution flow to 1500 cc/min.
11. Allow system to stabilize for 15 to 20 minutes and record
final output.
12. Repeat for dilution flows of 2500, 2000 and 1000 cc/min.
13. NO concentrations must be adjusted using N0? - NO con-
verter efficiency by the following:
(N02 x Converter Efficiency) = (NO)
14. The NO and NO interconnect lines must be returned to the
proper connection mode for proper instrument operation.
15. Return 8440 NO analyzer to the sample mode by connecting
supply line to intake.
Ozone—The procedure for calibrating the ozone analyzers was as
follows:
1. Connect zero supply to intake of 8410 ozone analyzer.
2. Allow to stabilize 15-20 minutes.
3. Make necessary zero adjustments and record data on
calibration graph.
4. Turn 8500 calibrator ozone generator to setting #1,
set dilution to rotometer 1500 cc/min. and allow sys-
tem to stabilize for 30 minutes.
81
-------
5. Make necessary span adjustments and record on calibra-
tion graph.
6. Repeat procedure for dilution flows of 2500, 2000 and
1000 cc/min.
7. Turn ozone generator off and return 8410 ozone monitor
to sample mode by connecting supply line to intake.
Sulfate Sampler—The Misco 800 grab sampler is calibrated using the
orifice calibrator and a water manometer:
1. Connect orifice to front of grab sampler.
2. Connect manometer to port on orifice can.
3. Turn 8000 grab sampler on and record manometer reading
along with sampler gauge reading.
4. Repeat for orifice restrictions of #5, #7, #10 and #15.
5. Mount filter on grab sampler.
6. Plot graph of grab sampler gauge reading versus manom-
eter change in height.
Data Editing
The editing and correction of data collected in the field program
started in the field as the field observers scanned the computer printer
output and observed the performance of the instruments. The observers
had the capability of editing data directly in the field or even taking
instruments off line so that data were deleted if it appeared the instru-
ment was malfunctioning. Some data were actually reduced and analyzed
in the field and used in the decision-making process relating to field
operations.
After the data were received in Gaithersburg, additional screening
was accomplished and those data were presented in quarterly reports to the
project officer. After the aerometric measurements were completed, the
data were thoroughly screened and edited into a final comprehensive data
tape that is available from the National Technical Information Service.
Field Data Editing--
Upon arrival at the power plant in the morning, the first item
of business was to scan the computer printout for continuity of data
from the field sensors. All zero and span values were checked to see
that each instrument performed that sequence at the proper time during
the night as well as checking for the reasonableness of the values.
82
-------
A scan of the meteorological parameters was made to check the reason-
ableness of the temperature sensors at the various stations. If one sta-
tion was recording a temperature in large variance to the majority of
other stations, and no reasonable explanation was readily available, then
that sensor was marked for checking and added to the maintenance list of
items to be covered. The same procedure was followed for the wind speed and
wind direction sensors. The wind direction sensors sometimes showed large
variations during light wind conditions due to their being affected by
local terrain-induced flows on certain occasions. Local effects could be
a dominate feature in wind values that do not correspond with other stations.
These considerations were evaluated before effort was expended to visit the
site to check the sensor.
So that the results of the double theodolite and T-Sonde data could
be readily available for use in directing the mobile sampling vehicle,
a programmable hand calculator (Texas Instrument SR-52) was available for
use by the field personnel. The field procedure was to record the azimuth
and elevation angles on a voice recorder; transcribe these angles onto a
double theodolite recording form prepared especially for this project;
enter the angles into a SR-52 programmable calculator; and record the
calculated wind speed, wind direction, height and change in height of
the balloon at each observation point.
After ascents with a temperature sensing device (T-Sonde) attached
to the balloon, the strip chart record was reduced by field personnel
who entered temperatures at appropriate heights on a form which presents
a detailed temperature profile with height.
Office Editing--
All data were logged in and recorded as received in the Gaithersburg
office, and a form documenting the data received was transmitted back to
the field personnel to corroborate their records as to what was sent. All
digital tapes were transmitted to the computing center (in the Gaithersburg
office) and logged into their magnetic tape storage files. All other data
were stored in their appropriate file or log book depending upon the type
of data.
The zero and span values received from the field were recorded on
computer forms for keypunching and merging into the appropriate zero and
span files as maintained on the computer disk. Zero and span values for the
fixed stations as well as the mobile stations were evaluated and keypunched.
The field data tapes were reblocked into convenient data records for
handling on the POP 11/70 computer available in the Gaithersburg office.
As these data were reblocked, the zero and span corrections were applied
to the air quality data and values for N02 were calculated. All data
(instantaneous as well as hourly) were recorded on a system tape at high
density, thus reducing the tape handling procedures for further analysis.
Between seven to nine field tapes could be condensed onto one system tape
by this process.
83
-------
Simultaneous with the writinq of the new system tape for the
corrected field data, the hourly averages were abstracted out of the
data set and recorded on the disk. This became the active working file
from which the reports (both quarterly and final) were generated.
The disk files were then sorted by station and a print of these
results was made available in hard copy form to the project personnel
for reviewing and editing of any erroneous data.
The daily log records were reviewed so that any data that the field
personnel indicated might be either questionable or erroneous were then
scrutinized using the hard copy printout. Any data that could not be cor-
rected was edited out as invalid data. Forms were available to the per-
sonnel screening the data to code corrections which were then entered
into an edit program that replaced all incorrect or missing values with
either a correct value or a missing indicator.
Plant operating data was obtained from the Clinch River Power Plant
personnel and transmitted to the Gaithersburg office. These forms were
given directly to keypunch personnel who had been trained to pick off
and punch the appropriate numbers directly onto cards. These data were
then placed on the disk file and merged with the air quality data for
the printout of the fixed station data used in the quarterly reports.
It was also used in the appendices which presented these data as plant
operating data.
The sulfate analysis data were screened when the analysis was received
from PEDCo in addition to the quality control that PEDCo performed. Any
high values were usually brought to our attention by PEDCo with the indica-
tion that they had rechecked their calculations to confirm the correctness
of these numbers. These sulfate measurements were corrected for background
filter sulfate content. The adjusted measurements were then combined with
the flow recorded for each high-volume sample to determine the air concentra-
tion mass per volume as presented in the report. Calibrations of the high
volume samplers were used to correct the recorded flow rate in these calcula-
tions.
The first port of each high volume sampler was never turned on
(the logical circuit was set up to initiate running on port 2) so that
a check of any sulfate deposition on the filters while in the high
volume sampler could be ascertained. All other ports (i.e., 2 through
24) were active ports which collected sulfates by drawing air through
the filters.
Field Audits
Obtaining a complete and reliable air quality and meteorological data
set is the primary objective of the Phase II Clinch River field project.
To ensure quality, data were collected and on-site audits were conducted.
84
-------
A series of four independent audits were carried out by members of Research
Triangle Institute. Additionally, GEOMET instituted periodic checks of
meteorological and air quality equipment.
The four audits performed by RTI consisted of the following:
1. System Audit (January 20 through 21, 1977)
2. Performance Audit #1 (February 15 through 23, 1977)
3. Technical Assistance Audit (July 12 through 14, 1977)
4. Performance Audit #2 (August 5 through 13, 1977).
The system audit was an evaluation of the GEOMET field project as a com-
plete system. Of primary concern was the record-keeping system, data
validation procedure, maintenance procedure, equipment installation and
field personnel.
The first and second performance audits were dynamic calibration
checks of all air quality equipment. This consisted of introducing a
known pollutant concentration to an analyzer and obtaining the resultant
output from the computer system. All samples introduced were NBS trace-
able and verified by laboratory techniques before and after the audit.
SOp samples were generated by permeation tubes and dilution of SOp tank
gas. NO concentrations resulted from the dilution of a concentrated NO
tank supply, and gas phase titration of NO produced known N0« samples.
Temperature measurements were checked during the first audit by
mounting an aspirated sensor on the roof of each sampling station and
comparing to the output received at the base station.
Prior to the July intensive study, a technical assistance audit was
conducted at the Munsey station. The purpose was to verify the data
quality at the Munsey station, which was representative of the entire
system, review any weaknesses that had been pointed out during the two
prior audits and finally to make any suggestions that would assist GEOMET
in obtaining a high percentage of quality data during the intensive study.
During the preliminary field audit, some basic weaknesses were
pointed out. The particular items are listed below:
1. Traceability of calibration standards and permeation
tubes to NBS-SRM's via supplier. Primary dependence
upon the manufacturer's quality control.
2. Lack of flow devices to check volumetric flow rate
within _+ 2 percent accuracy. Flow devices not trace-
able to NBS.
85
-------
3. Independent check (standard of higher accuracy) of NO
standard not available.
4. Capability of verification of ozone generator output
by KI method or apparatus to perform gas phase titra-
tion not apparent.
5. No quality assurance coordinator designated to insure
maximum quality and quantity of data.
6. Possible altitude effects on airborne data, i.e.,
effect of altitude on analyzer calibration.
7. Zero air not checked against an independent source to
define its quality.
8. Calibration of SCL analyzer not through sample inlet
line, i.e., not tnrough H?S scrubber.
All traceability of calibration data pertaining to permeation devices and
tank gas have been retained at the Gaithersburg office of GEOMET. This was
a security procedure. An NBS traceable flow device was purchased in early
April 1977. A check was then run against the flow device that had been in
use. Because of problems experienced with the Monitor Labs 8440 NO-NO
analyzer, an NBS traceable tank of NO gas was purchased and delivery t8ok
place in early November 1976. Initially, the tank was used to eliminate
the NO-NO analyzer problems. Periodic checks (four to six months) were
then made on all NO tanks in the field by comparing them to this NBS tank.
Ozone quality checks were made during all performance audits. A
reliable gas phase titration device was not available during the Clinch
River project.
The plumbing of the Monitor Labs 8450 sulfur dioxide analyzer was
modified to allow calibration gas as well as sample air to pass through
the HpS scrubber. Tests were conducted to determine the influence of
the HpS scrubber on the calibration gases and no appreciable impact
was ooserved. For the purpose of consistency, the new plumbing system
was utilized.
Cross checks of calibration devices at all stations have been
carried out on a periodic basis. This is accomplished with a portable
calibrator that was checked against all other calibration devices. If
a majority of stations yielded consistent results with one or two show-
ing discrepancies then the calibrator was accepted and the stations
showing variations were investigated. On the other hand, if all sta-
tions showed variation from the actual concentration, then the calibra-
tor was suspected.
86
-------
The two individual performance audits pointed out particular malfunc-
tions with specific instruments. This information was the basis for inves-
tigations on those problems and subsequent repairs. Furthermore, data
corrections could be made based on GEOMET calibration data for the periods
when the instrument was in question.
GEOMET conducted periodic audits on an informal basis. Air quality
equipment, consisting of calibrators, SOp and NO-NO analyzers, was
checked in addition to wind speed and wind direction equipment.
SOp analyzers were checked against a known S0? concentration gen-
erated Dy a permeation device. Multiple points were checked to ensure
reliability through the entire instrument range. Oxides of nitrogen anal-
yzers were referenced to an NBS traceable NO supply. After a multipoint
NO calibration was performed, N02 samples generated from a permeation
tube were introduced and comparea to the NO calibration. This served two
purposes: first a check of the NO calibration technique; second, and most
important, a measure of the Molybdenum converter's (N0? to NO) effi-
ciency could be obtained.
Wind direction sensors were oriented to a known location that could
be identified on a topographic map and verified by a compass. Wind speed
sensors were checked electronically by generation of a known signal cor-
responding to a particular wind speed. The resultant wind speed was then
compared to the input data.
High-volume samplers were tested for flow versus chart reading by
utilizing an orifice calibration device. The orifice calibrator was
supplied with a calibration curve which was then checked against a posi-
tive displacement meter.
87
-------
SECTION 5
RESULTS
OVERVIEW
The field data which were collected over a 16-mo period between
June 1, 1976 and September 30, 1977 are recorded on magnetic tape. They
are also tabulated in preliminary form in five quarterly reports. The
data were collected at the 13 monitoring stations identified in Table 5.
The characteristics of the 13 stations, which are described in detail in
Section 4, are summarized in Table 5 by 11 types of measurements. Four
of the measurements are aerometric measurements of pollutants in the air,
including sulfur dioxide, nitrogen oxides (i.e., NO and NOX), ozone, and,
sulfate aerosols. The remaining seven measurements are meteorological
measurements, including wind speed and direction, temperature, relative
humidity, precipitation and ultraviolet radiation intensity. At some
stations vertical profiles of temperature and/or wind can be constructed
from the observations made at two or more heights.
Figure 34 shows when during the operating period each station became
operative and when there were major periods of down time of five days or
more. The heavy lines with arrowheads at each end indicate periods when
the station was collecting data. The light lines without arrowheads show
periods of down time of five days or more when various measurements were
not being taken. The numbers below the lines refer to the measurements
listed in Table 5 to identify the types of measurements which were not
operational.
Figure 34 provides a convenient index of the types of data which are
available for various parts of the field monitoring operations. The six
primary monitoring stations which measured S02, NO, NOx. wind speed and
direction, and temperature were operational from July 1 to the end of the
period, except for the Castlewood station. A flood on April 5, 1977,
washed the Castlewood station away. In July, the Lambert station was
moved to the Castlewood site which was provided with full utility service
in August. Two supplementary stations which measured S02 and wind speed
and direction became consistently operational in November. The stack
monitor was also installed in November to begin measuring S02 emissions.
The NO emission measurements became operational in December. The stack
monitor was put out of commission briefly when the plant shut down suddenly
due to the April flood. Another significant weather phenomenon which
seriously disrupted the monitoring network was summer thunderstorms.
Lightning strikes on the tower knocked out the meteorological instruments
once each summer.
88
-------
O
H
O
§
g
(-
I
s
00
u->
3
<
a
o
5
to
i — t
£
. S
s *•>
*T-I O
A O
HI
2
O W
a
s
rt
1
0)
H-l
n
1
£
CO Q)
%» &£
§ 2
W ^
1
Q) T3
to O
rt fe
U f
(U
•s
o
EC
VI
% T3
rt 0
Z
Sr
c
8
o
E-
Measurement
X
X
X
X
X
X
X
X
X
X
X
X
C\J
s
iH
X
X
X
X
X
X
X
X
X
Nitrogen Oxides
N
X
X
X
X
X
X
Sulfates
to
X
X
(U
8
•**
X
X
X
X
X
X
X
X
X
1
s
o
T-t
•0
*
X
X
X
.2
g
o
CO
VO
X
X
X
X
X
X
X
X
X
Temperature
^
X
X
X
£
Temperature Prof:
*
oo
X
X
^
Relative Humidit
•
X
Precipitation
o
X
UV Radiation
T-l
t
•a
89
-------
O
i
£
Jfl
1C
^
I
I
i
5
z
tl.
c
£
1
J
1
f
*
f
I ~
i
*
-
L
»
«
,
1 "1
1'
r-
V
Sf
»
Tow er
j-'
1.
j
u
t _.
K
-
i
N
l"
i-
~
!-
X
z
u
|4.
«
~
fe-
ft
_
-
•
-
»
r
L'
>
C
J
D
1,
«
<
'
L
c
1
s
I
t
;
i —
"*"•*
*"
i-
J~
h
Cr
J.
J
-
tn
C
V
t
1
E
'•
Jt
Si
\
I
\
•Sv
P»
Lambert
c
10*,
>U
a
J
A
\
1
•
-
a
j
]
\
:
[
-l C
c
rt
> :
t
L
1
u
8
c S
H V
> I
f
I
in
u
1
•o
f
B)
J!
t
&
•3
•s
^
f-
0
1
8
etnent wai
i
a
Ame that
ndlcate t
f
I
*
*!*
!"i
SSI
III
5 5 o
CO M O
8 1 ^
3 a n «
S "t
T!
O
bo
.S
a
o
bO
fl
S
.Sf
fc,
90
-------
The pibal station and mobile van were manned fairly continuously;
however, there were several periods scattered throughout the study when
only one or two operations a week were conducted with pibals or the van.
These periods are shown in Figure 34. There was a period from January to
March when a second van was used to supplement the mobile sampling
operation.
The two most intensive monitoring periods were conducted using a
helicopter to make repeated passes through the plume. One 10 d period of
helicopter operations was performed during November, and the second 10 d
period was performed in July 1977.
The procedures followed to control the quality of data were tested by
independent measurements made by organizations other than GEOMET or its
subcontractors or suppliers. Field performance audits were conducted by
an independent contractor hired by EPA. Fifteen strips containing unknown
quantities of integrated sulfates were furnished by EPA and submitted to
the laboratory which performed the sulfate analysis. The stack gases were
measured by an independent contractor using the EPA reference method to com-
pare with simultaneous measurements by the automated continuous stack
monitor.
Two independent audits of the monitoring instruments were performed by
Research Triangle Institute. In the first audit, conducted during the
period of February 15 to 23, 1977, measurements of S02, NO, NOX, N02 and,
temperature were tested at the eight fixed monitoring trailers and in the
primary van (Smith et al., 1977). The effects of variations in C02 concen-
trations on the S02 analyzers were also examined.
The S02 analyzers at the eight fixed monitoring stations were tested
with five audit concentrations, including a zero value, concentrations near
50, 100 and, 150 ppb, and a higher value which varied from 350 to 850 ppb at
different stations. A summary of the measured audit concentrations at each
station and the station analyzer responses are listed in Table 6. One instru-
ment (Lambert Station) was not performing properly on the day of the audit.
The cause of the poor performance was later identified and corrected. The
instruments are calibrated with span gases generated using a permeation tube
in a system with a maximum output of about 150 ppb. The instrument responses
compared to the audit values show that the absolute differences are smallest
in the range of 0 to 150 ppb. Responses near the high audit value are more
variable in absolute accuracy, although not in percentage differences. At
the fixed stations only 1% or less of the recorded S02 concentrations are
above 200 ppb. If one omits the highest measurements in the audit and the
values for Lambert station, there are 28 remaining responses to audit compari-
sons. The root-mean-square difference of these 28 values is 10 ppb which is
about the expected accuracy of the instrument. Because the instrument was
91
-------
TABLE 6. SUMMARY OF SULFUR DIOXIDE AUDIT RESULTS, FEBRUARY 15-23, 1977.
Station Name
(Purchase date
of analyzer)
Castlewood
(Feb. -Mar. , 1976)
Johnson
(Feb. -Mar. , 1976)
Hockey
(Feb. -Mar. , 1976)
Munsey
(Feb. -Mar. , 1976)
Lambert
(Feb. -Mar., 1976
Tower
(Feb. -Mar., 1976)
Nash
(Feb. -Mar. , 1976)
Kent
(Feb. -Mar. , 1976)
Analyzer Audit
Audit Range SO2
Date (ppb) (ppb)
2/15/77 0-1000 0
50
109
138
544
2/16/77 0-1000 0
65
127
188
349
2/17/77 0-1000 0
50
96
146
721
2/18/77 0-1000 0
51
97
150
725
2/19/77 0-1000 0
51
96
148
468
2/20/77 0-1000 0
51
96
149
460
2/21/77 0-1000 0
51
99
145
841
2/22/77 0-500 0
46
87
138
340
Response
so2
(Ppb)
10
37
92
135
484
0
58
116
175
325
4
57
110
174
746
5
46
88
139
536*
0
80
160
251
787
1
46
101
157
678
' -1
40
95
149
732
-2
43
94
161
395
Difference (d)
(response- audit)
ppb
+10
-13
-17
-3
-60
0
-7
-11
-13
-24
+4
-7
-14
+28
+25
+5
-5
-9
-11
-189
0
+29
+64
+103
+319
+ 1
-5
+5
-8
+218
-1
-11
-4
+4
-109
-2
-3
+7
+23
'55
J]
_
-26
-16
-2
-11
_
-11
-9
-7
-7
_
-14
+ 15
J-19
^
_
-10
-9
-7
-26
.
+57
+67
+70
+68
_
-10 -
+5
+5
-47
_
-22
-4
+3
-12
_
-6
+8
+17
-16
Regression of audit
on response
A - mR i- b
m b
1.13 -2.73
1.07 1.59
0.977 -10.5
1.12 -2.99
1.60 -1.05
0.938 2.45
1.15 -6. 59
0.847 5.21
Data point does not fit linear pattern and was not included in calculation of regression coefficients.
92
-------
not routinely calibrated for higher concentrations, values above 200 ppb
are associated with a higher degree of uncertainty. The audit results
found the instruments varied from 26% below to 16% above audit values
(excluding Lambert) at high concentrations (between 340 and 841 ppb). The
root-mean-square percentage difference was 13% for these higher values.
The Lambert S02 measurements were a special situation during the first
audit. The results in Table 6 show that these measurements were consis-
tently higher than audit values by 57 to 70%. After the audit it was dis-
covered that there was a marked increase in the instrument response when
the calibration gas was introduced through the instrument "sample port"
(as was done in the audit evaluation) rather than through the instrument
"span port" as was routinely done in the field calibrations). When the
instrument solenoid valve was replaced there was no difference between
readings for span gas introduced through the two ports. Since the span
readings are quite consistent prior to changing the solenoid valve, it is
assumed that the solenoid was defective from its initial installation. The
dilution effect of the faulty solenoid valve on the span gas used to cali-
brate the instrument prior to replacement is relatively constant. Thus it
has been possible to correctly interpret the instrument response to the
span gas by taking the dilution effect into account and assuming the S02
content of the diluting ambient air is zero.
During the audit it was demonstrated that high C02 concentrations can
interfere with S02 analyzer readings. The effect, expressed as a percent
of the S02 reading, is independent of S02 concentration, but it varies from
an average depression of 5% at a C02 concentration of 360 ppm to an average
depression of 15% at 750 ppm. The difference between C02 concentrations in
ambient air and in the dilution air used to calibrate the instruments was
tested once each day during the audit at whichever station was being audited
that day. The average difference was 126 ppm with a range of +23 to -324 ppm.
These results suggest that the suppression effect is generally less than 5%.
The greatest uncertainty is likely to occur with very high S02 measurements.
The magnitude of the maximum effect is expected to be about a 15% depression
in the reading from the actual concentration.
The NO-NOx analyzers at five fixed stations and in the mobile van were
tested for response to NO, N02 and, total oxides of nigrogen. The fixed
stations were tested for each gas with concentrations near 400 ppb, near
800 ppb, and, a higher value (for NO and NOX only). The van was tested
with concentrations near 200 and 400 ppb. The audit concentrations are
close to the concentrations of the span gases used to calibrate the instu-
ments; however, these are much higher than are normally measured. The
results of the audit for concentrations near 400 ppb are shown in Table 7.
The results for Munsey show that this instrument was not functioning properly.
The source of the problem was corrected shortly after the audit. The mean
percentage differences between response and audit aalues for NO, NOX, and,
N02 were -2%, +3%, and, +18%, respectively, when the Munsey station results
are omitted. The root-mean-square percentage differences were 9, 11, and 21%,
respectively. Since N02 is determined as the difference between NOX and NO
measurements, it is not surprising that a larger error is associated with N02
than with either of the other two measurements. The audit results for the
higher concentrations were similar to those presented in Table 7.
93
-------
TABLE 7. SUMMARY OF NITROGEN OXIDES AUDIT RESULTS, FEBRUARY 15-23, 1977.
Station
C astlewood
Munsey
Tower
Nash
Kent
Van
Concentrations
Gas
Zero
NO
NOX
NO2
Zero
NO
NOX
NO 2
Zero
NO
NOX
N02
Zero
NO
NOX
NOa
Zero
NO
NOX
NO2
Zero
NO
NOX
N02
Audit
Level
fP?B)
0
374
374
398
0
383
363
312
0
428
428
398
0
401
401
398
0
371
371
398
0
404
404
312
Analyzer
Response
(PPB)
- 1
428
420
519
- 2
699
693
649
- 3
427
487
483
- 3
360
354
416
6
334
348
513
4
394
435
333
Difference
(R esponse -Audit)
(PPB)
1
+ 54
+ 46
+121
- 2
+336
+330
+337
- 3
- 1
+ 59
+ 85
- 3
- 41
- 47
+ 18
+ 6
- 37
- 23
+115
+ 4
- 10
+ 31
+ 21
%
_
+ 14
+ 12
+ 30
+ 93
+ 91
+108
_
0
+ 14
+ 21
_
- 10
- 12
+ 5
-
- 10
- 6
+ 29
_
- 2
+ 8
+ 7
94
-------
The temperature sensors at six fixed monitoring stations and the van
were audited. The mean difference between audit and response temperature
was 0.6°C. The root-mean-square difference was 0.9CC. The small differ-
ences which did occur could be attributed to differences in location of
the audit and network sensors which were not always at the same heights
above ground. Differences of as much as 20 feet were present during some
of the comparisons.
A second field audit was performed during the period of August 5 to 12,
1977, by Research Triangle Institute (Frazier and Williams, 1977). One
ozone, six oxides of nitrogen and three sulfur dioxide analyzers were audited
following procedures used in the first audit. The ozone analyzer response
in the van differed from the audit test concentrations by no more than 10 ppb,
which is within the instrument accuracy, when challenged by concentrations
of 0, 30, 70, 110 and, 160 ppb of ozone.
The oxides of nitrogen analyzers at six stations were tested, including
the van. As in the previous audit the audit concentrations were much higher
than the normally measured ambient air concentrations. It is also observed
in the audit results that the percentage difference between audit and
response concentrations increases with increasing concentration. In order
to get an error estimate that is representative of the values normally
measured, we have determined the root-mean-square percentage difference
between audit and response concentrations for audit concentrations below
600 ppb. For all NO and NOX comparisons this includes responses to a single
audit value of 560 ppb at each site. For N02 comparisons this includes
responses to two or three audit values since varying concentrations of N02
were used at each station. The results obtained for the van show that it
is clearly not operating properly. The source of this problem was later
identified and corrected. The audit results for the van have not been used
in estimating the uncertainty of the network oxides of nitrogen observations.
The results presented in Table 8 show that the mean errors of NO, NOx and,
N02 are 4, 3 and, 8%, respectively, averaged over the five good measuring
stations; the root-mean-square percentage errors are 10, 8 and, 15%,
respectively.
Sulfur dioxide analyzers at three stations were audited. In contrast
to the first audit, higher concentrations were tested in the second audit
with the results listed in Table 9. Audit results for measurements below
400 ppb are most relevant to the ambient air measurements compiled in the
field program. The mean percentage difference between audit and response
concentrations below 400 ppb (excluding 0 values) was 1%; the root-mean-
square percentage difference was 8%.
In order to estimate the accuracy of the laboratory assessment of sul-
fate content on the filters submitted for analysis, a set of filters with
known sulfate amounts embedded in the filters was obtained from EPA. The
PEDCo laboratory analysis of these filters is presented in Table 10. The
sulfate content varied from 0 to 6550 ug. The results show that the differ-
ence between the embedded amounts and the measured values is best expressed
as an absolute difference rather than a percentage of the embedded amount.
The mean difference was 109 pg, and the root-mean-square difference was
223 jjg. The higher measurement results may be due to the tendency of the
95
-------
TABLE 8. AUDIT RESULTS FOR OXIDES OF NITROGEN ANALYZERS,
AUGUST 5-12, 1977
Station
Tower
Munsey
Kent's Ridge
Nash's Ford
Hockey
Concentrations
Gas
Zero
NO
NOX
N02
Zero
NO
NOX
N02
Zero
NO
NOX
NO 2
Zero
NO
NOX
NO 2
Zero
NO
NOX
N02
Audit
(PPB)
0
560
560
240
330
590
0
560
560
340
490
0
560
560
260
360
580
0
560
560
260
360
0
560
560
220
300
Analyzer
Response
(PPB)
10
640
600
220
330
630
0
600
580
380
590
0
620
590
250
380
630
0
540
630
, 340
490
0
510
500
200
320
Difference
(Response -Audit)
(PPB) %
10
80
40
- 20
0
40
0
40
20
40
100
0
60
30
- 10
20
50
0
- 20
70
80
110
0
- 50
- 60
- 20
20
_
14
7
- 8
0
7
_
7
4
12
20
_
11
5
- 4
6
9
_
- 4
12
31
29
_
- 9
-11
- 9
7
96
-------
TABLE 9. SULFUR DIOXIDE ANALYZER RESULTS
FROM AUDIT OF AUGUST 5-12, 1977
Station
Name
Munsey
Kent's
Ridge
Hockey
Analyzer
» * _ j i
Model
and S/N
Monitor Labs
Model 8450
S/N 85
Monitor Labs
Model 8450
S/N 128
Monitor Labs
Model 8450
Concentration
(ppm)
0.00
0.20
0.38
0.48
0.66
0.87
0.00
0.20
0.38
0.48
0.00
0.20
0.38
0.48
0.66
0.87
Analyzer
Response
(ppm)
0.02
0.18
0.38
0.49
0.67
0.87
0.01
0.22
0.43
0,55
0.01
0.19
0.37
0.46
0.61
0.77
Difference
PPM
0.02
-0.02
0.00
0.01
0.01
0.00
0.01
0.02
0.05
0.07
0.01
-0.01
-0.01
-0.02
-0.05
-0.10
Percent
__
-10.0
0.0
2.1
1.5
0.0
--.
10.0
13.2
14.6
__
- 5.0
- 2.0
-4.2
-7.6
-11.5
Regression on Analyzer
Response on Audit
Concentration
Y = mX + 3
m
0.998
1.126
-
0.882
b
0.004
0.004
0.021
r
0. 9990
0. 9996
0. 9987
m = slope
b = intercept
r = correlation coefficient
filter material to absorb S02, which is converted to sulfate during the
process of transporting and handling the filters. Otherwise, the measure-
ments are within the accuracy of the assessment technique.
In order to test the accuracy of the EDC stack gas analyzer, which
measures NO and S02 concentrations in the breeching of one stack, the
Engineering Science Company was hired under contract to EPA to measure the
stack gases by the standard reference methods. Comparative tests were made
during the 4-d period of November 15 to 18, 1977,(Engineering Science 1978).
The comparisons for S02 measurements are presented in Table 11. The com-
parisons for NO measurements are presented in Table 12. The plant was
operating its number three generator at a very constant and high-load
level throughout the measurement period. The output averaged 240 megawatts
and did not vary from the average by more than 1 or 2 megawatts. The EDC
analyzer produced very consistent results which were continuously recorded
on a strip chart. The S02 measurements showed a standard deviation of 12 ppm
compared to a mean of 459 ppm or a variation of less than 3% of the mean.
The stack samples drawn by probe and later analyzed in a laboratory showed
a slightly higher mean of 479 ppm and over twice as much variation in the
samples, i.e., a standard deviation of 27 ppm or about 6% of the mean.
97
-------
TABLE 10. COMPARISONS OF EMBEDDED AMOUNTS OF SULFATE
WITH LABORATORY ANALYSIS OF FILTER SULFATE.
Filter
No.
1204
1205
1211
1207
2159
2202
3338
3390
3463
4302
4306
4326
5141
5226
5249
6080
6125
6345
7156
7222
7343
8104
8155
8307
9220
9221
9222
Mean
Standard
Deviation
Sulfate
Embedded
(yg)
200
200
200
6550
6550
6550
0
0
0
500
500
500
2670
2670
2670
5695
5695
5695
4240
4240
4240
1735
1735
1735
745
745
745
2482
2375
Measured
(yg)
425
360
300
6750
7000
6750
<150
<150
<150
475
581
484
2531
2844
2500
6000
5821
5786
4450
4200
3700
2052
1958
2021
833
933
858
2587
2405
Difference
( Measured-Embedded)
(yg)
225
160
100
200
450
200
-
_
-
- 25
81
- 16
-139
174
-170
305
126
91
210
- 40
-540
317
223
286
88
188
133
109
223
Difference/ Embedded
%
113
80
50
3
7
3
-
_
-
- 5
16
- 3
- 5
7
- 6
5
2
2
5
- 1
- 13
18
13
16
12
25
15
IS
32
98
-------
TABLE 11. COMPARISONS OF EPA METHOD 6 AND EDC STACK ANALYZER
MEASUREMENTS OF SO2 IN STACK GAS
Run No. *
6-1
6-3
6-5
6-7
6-9
6-11
6-12
6-13
6-14
6-15
6-16
Date
11-15-77
11-16-77
11-16-77
11-16-77
11-17-77
11-17-77
11-17-77
11-17-77
11-18-77
11-18-77
11-18-77
Time
1527-1612
1141-1211
1322-1352
1515-1545
1042-1112
1234-1304
1328-1358
1413-1443
0844-0914
0921-0951
1023-1053
Mean
Standard Deviation
Plant
Load
(mw)
241
240
242
240
239
240
239
240
242
241
239
240
1
Stack Concentration
EPA
Method 6
(ppm)
495
472
488
483
452
474
466
420
516
506
496
479
27
EDC
Analyzer
(ppm)
450
450
470
460
440
450
460
450
470
470
480
459
12
Difference
(EDC -Method 6)
(ppm)
-45
-22
-18
-23
-12
-24
- 6
30
-46
-36
-16
-20
21
Difference
Over
Method 6
(°/o)
-9
-5
-4
-5
-3
-5
-1
7
-9
-7
-3
-4
4
* Results for EPA Method 6 during runs 6-2, 6-4, 6-8, 6-10 produced obvious outline results believed
due to a leak in the right angle probe-to-impinger connector.
99
-------
TABLE 12. COMPARISONS OF THE EPA METHOD 7 AND EDC ANALYZER
MEASUREMENTS OF NO IN STACK GAS
Run No.
7-1
7-2
7-3
7-4
7-5
7-6
7-7
7-8
7-9
7-10
7-11
7-12
7-13
7-14
7-15
7-16
7-17
7-18
7-19
7-20
7-21
7-22
7-23
7-24
7-25
7-26
7-27
7-28
7-29
7-30
7-31
7-32
7-33
7-34
7-35
7-36
Date
11-15-77
11-15-77
11-15-77
11-15-77
11-15-77
11-15-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-16-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-17-77
11-18-77
11-18-77
11-18-77
11-18-77
11-18-77
11-18-77
Time
1535
1550
1608
1623
1638
1655
1015
1045
1115
1145
1215
1245
1345
1415
1445
1515
1545
1615
1100
1130
1200
1230
1300
1330
1400
1430
1500
1530
1600
1630
0915
0930
0945
1000
1015
1030
Mean
Standard Deviation
Plant
Load
(mw)
241
241
241
241
241
241
239
239
239
240
240
240
242
242
241
241
240
240
239
240
240
240
240
239
239
240
240
240
240
239
242
241
241
241
241
239
240
1
Stack Concentrations
EPA
Method 7
(ppm)
956
982
893
965
939
1005
1249
927
842
828
951
976
890
918
922
900
949
952
877
920
820
858
978
910
925
1041
950
1031
1009
1051
977
975
967
1064
1052
1056
958
81
EDC
Analyzer
(ppm)
900
890
900
900
900
900
900
880
880
890
880
890
880
880
870
860
900
900
870
880
950
840
930
930
940
940
950
960
950
940
960
970
960
960
960
960
912
36
Difference
(EDC -Method 6)
(ppm)
- 56
- 92
- 7
- 65
- 39
-105
-349
- 47
38
- 62
- 71
- 86
- 10
- 38
- 52
- 40
- 49
- 52
- 7
- 40
130
- 18
- 48
20
15
-101
0
- 71
- 59
-111
- 17
- 5
- 7
-104
- 92
- 96
- 46
73
D if fere nee
Over
Method 6
(°/°)
- 6
- 9
1
- 7
- 4
-10
-28
- 5
5
7
- 7
- 9
- 1
- 4
- 6
- 4
- 5
- 5
- 1
- 4
16
- 2
- 5
2
2
-10
0
- 7
- 6
-11
- 2
- 1
- 1
-10
- 9
- 9
- 4
7
100
-------
The EDC analyzer consistently underestimates the EPA Method 6 estimate by
about 4% or 20 ppm. In view of the uncertainties and differences in the
techniques of the two measurements, this small difference strongly supports
the validity of the stack gas analyzer measurements.
Thirty-six comparisons were made for NO measurements. The EDC anal-
yzer measurements showed a standard deviation of 36 ppm compared to a mean
of 912 ppm or a 4% ratio of variability. The EPA Method 7 measurements
showed a standard deviation of 81 ppm compared to a mean of 958 ppm or an
8% ratio of variability. The EDC measurements were consistently lower than
the Method 7 measurements by an average of 46 ppm, which is about 5% of the
Method 7 measurements. As with the S02» the NO measurements show very good
agreement considering the differences and uncertainties in the two measure-
ment techniques.
Overall, these results suggest that the NO and S02 measurements by the
EDC gas analyzer are accurate to within 5% when the analyzer is working
properly. Because of inherent differences in the two measurement techni-
ques, it is not meaningful to estimate the accuracy of the EDC analyzer
with any greater precision from the results of these tests.
Other significant measurements made during the stack gas analysis
included the temperature, volume flow rate, oxygen content, C02 content
and the NOX/NO ratio. Exhaust gas temperature and velocity were determined
in two exhaust gas profiles, each of which consisted of 48 traverse point
samples from the breeching cross-section. The cross-section is 7.3 m high
by 3.7 m wide. The average traverse temperatures were 392°K and 395°K com-
pared to temperatures of 391°K and 394°K reported by the plant for the cor-
responding times. Based on these results, it is likely that the plant tem-
perature records are within 1°K of what a traverse would determine the
stack temperature to be.
The average traverse velocities were 14.8 and 14.4 m/sec. indicating
volume flow rates of 397 and 385 m3/sec respectively. These numbers com-
pared to boiler air flow rates of 249.5 and 239.7m3/sec reported by the
plant for this same time period. The average ratio of air flow in the
breeching to air flow in the boiler indicated by these measurements is 1.60.
Accordingly, all boiler air flow measurements reported by the plant have
been scaled by this ratio to estimate the stack gas flow rate.
Because of the large difference between the stack breeching measure-
ments of air flow rate and the plant measurements of boiler air flow rate,
a correlation was run between the boiler flow rate and generator load. The
air flow, if it is meaningful, should be well correlated with the plant
load. The correlation was 0.90. As a result, it was decided that the plant
measurements multiplied by a factor of 1.60 are probably a good estimate of
the stack gas exhaust flow rate. The scaled plant measurements are included
in the hourly file of stack gas characteristics for each stack.
101
-------
DATA FORMATS
The data collected in the field program has been organized on magnetic
tape into the following six categories:
1. Fixed station data
2. Mobile ground station data
3. Helicopter data
4. Sulfate data
5. Plant operating data
6. Balloon data.
Each category of data is organized using a different format. A descrip-
tion of the format and the data items is presented for each of these types
of data. All of the data files are available on magnetic tape in unlabeTed,
1600 bpi, ASCII character format.
Fixed Station Data
The fixed station data is organized into a separate data file for
each station. A separate record is entered in the file for each hour of
recorded observations. The hour of the day indicated for each record
refers to the 60 minute period ending at that whole hour. The record
format is listed in Table 13. The table entries indicate the variable in
each record, their order of appearance, a description of each variable
including units, and the number and format of the characters (or numbers)
for each variable. There are eight fixed station data files, one for
each site. The following six-character codes designate each station:
Site 6-Character Code*
Tower TOWERb
Munsey MUNSEY
Nash's Ford NASHSb
Hockey HOCKEY
Castlewood CASTLE
Johnson JOHNSN
Lambert LAMBER
Kent's Ridge KENTSb
The following codes are used as flags:
W - Measurement is based on less than 22 scan values within the hour.
M - N02 calibration gas used to adjust instrument was not properly
controlled. Measurement shows changes relative to other measure-
ments within a few days, but may not be quantitatively correct.
S - Derived from strip chart.
* b designates a blank character.
102
-------
A - Arithmetic mean rather than vector mean of 2-minute scan values
because wind speed or direction is missing.
L - Wind direction and speed are not reliable because wind speed is
near or below the sensing threshold of the sensor.
E - Estimated zero applied to NOX, NO data. Data is only qualita-
tively useful.
C - Incorrect zero/span values have been applied to data. Data is
only qualitatively useful.
U - Data is uncalibrated (no zero/span applied). Data is only quali-
tatively useful.
N - N02 not computed since NOX and NO are uncalibrated or calibrated
incorrectly.
K - Kent's Ridge wind direction data unreliable.
Codes used to designate a missing or unacceptable measurement are listed
in Table 13. Values are unacceptable if the instrument was malfunctioning
and producing unuseful information. Incorrect data periods were deter-
mined by reviewing operating logs, records of calibration data and listings
of the measurements.
Only one flag is assigned to each measurement. The following priorities
are used to select a flag if more than one might be assigned:
Priority Wind Speed Wind Direction Pollutant
1 L K N
2 A AM
3 W WE
4 U
5 C
6 S
7 W
Mobile Ground Station Data
Mobile Ground Monitoring data consists of data collected at different
locations by the two mobile ground monitoring vans, designated Mobile 1
and Mobile 2. The data was originally recorded every 30 seconds as a
voltage output on cassettes in the mobile units and then transcribed to
9 track ASCII reels. Each location sampled was identified by an event
marker and time, which were also recorded on the cassette. The azimuth,
distance and elevation of each location sampled were then abstracted from
the operating log and merged with the data on tape using the time and event
flag to match records.
Channels recorded are S02, for Mobile 2 and S02, NOx, N02, NO, 03,
wind speed, wind direction and, temperature for Mobile 1. The channel
data were converted to engineering units using the appropriate scale for
each channel. Zero and span calibrations were applied using values taken
before each run. The calibrated data was recorded on a file in the format
103
-------
TABLE 13. FIXED STATION RECORD FORMAT ON MAGNETIC TAPE
Channel #
1
1A
2
2A
3
3A
4
4A
S
6
7
8
9
10
11
12
13
14
IS
Variable Description
Station Code
Year
Julian Day
Hour of Day
SO2 Validity Flag
Mean SOj
Peak SO2
NOx Validity Flag
Mean NOX
Peak NOX
N02 Validity Hag
Mean NO
Peak NO2
NO Validity Flag
Mean NO
Peak NO
10 M Wind Speed Validity Flag
10 M Wind Speed
10 M Wind Direction Flag
10 M Wind Direction
10 M Sigma Azimuth Hag
10 M Sigma Azimuth***
10M Temperature Validity Flag
10M Temperature''"
UV Radiation Validity Flag
UV Radiation
Relative Humidity Flag
Relative Humidity-
Precipitation Flag
Precipitation
10 M Wind Elevation Angle Flag
10 M Wind Elevation Angle
10 M Sigma Elevation Flag
10 M Sigma Elevation***
Lower Temperature Difference Flag
0. 5 to 4M Temperature Difference**** <
30 M Wind Speed Flag
30 M Wind Speed
Column
1-6
7-8
9-11
12-13
14-15
16-20
21-25
26-27
28-32
33-37
38-39
40-44
45-49
SO-S1
52-56
57-61
62-63
64-68
69-70
71-75
76-77
78-82
83-84
85-89
90-91
92-96
97-98
99-103
104-105
106-110
111-112
113-117
118-119
120-124
125-126
127-131
132-133
134-138
Format*
A6
12
13
12
A2
F5.0
F5.0
A2
F5.0
F5.0
A2
F5.0
F5.0
A2
F5.0
F5.0
A2
F5.2
A2
F5.1
A2
F5. 1
A2
FS. 1
A2
FS.2
A2
F5.0
A2
F5.2
A2
FS.O
A2
F5. 1
A2
F5. 1
A2
F5.2
Units
Code
Last Two Digits
1-366
1-24
Code
PPB by Volume
PPB by Volume
Code
PPB by Volume
PPB by Volume
Code
PPB by Volume
PPB by Volume
Code
PPB by Volume
PPB by Volume
Code
Meters per Second
Code
Degrees
Code
Degrees
Code
Degrees C
Code
Milliwatts per CM2
Code
Percent
Code
Inches
Code
Degrees
Code
Degrees
Code
Degrees C
Code
Meters per Second
Missing**
-999.
-999.
-999.
-999.
-999.
-999.
-999.
-999.
999.
999.
999.
999.
999.
999.
999.
999.
999.
999.
999.
* All decimals implied
** A blank field indicates null channel
*** Sigma means the standard deviation of the two-minute scan values.
****Difference is upper level minus lower level
'Tower temperature at 4M
104
-------
TABLE 13. (cont'd)
Channel 4
16
17
18
19
20
Variable Description
30 M Wind Direction Flag
30 M Wind Direction
30 M Elevation Angle Flag
30 M Elevation Angle
30 M Sigma Elevation Flag
30 M Sigma Elevation***
Upper Temperature Difference Flag
0.5 to 30 M Temperature Difference****
30 M Sigma Azimuth Flag
30 M Sigma Azimuth***
Filler Character
Column
139-140
141-145
146-147
148-152
153-154
153-159
160-161
162-166
167-168
169-173
174-176
Format*
A2
FS.l
A2
F5.0
A2
FS.l
A2
FS.l
A2
FS.l
UniB
Cod*
Degrees
Code
Degrea
Code
Degrees
Code
Degrees C
Code
Degrees
Blank
Missing**
999.
999.
999.
999.
999.
* All decimals implied.
** A blank field Indicates null channel.
*** Sigma means the standard deviation of die two-minute scan values.
**** Difference is upper level minus lower level.
105
-------
listed in Table 14. Records for both Mobile 1 and Mobile 2 data are 76 bytes
long. Missing value indicators of -999 are used for all pollutants and
meteorological data.
Each data record corresponds to an observation period. An event number
and a corresponding location and time designate the beginning of the observa-
tion period. The measurements are mean or peak values observed between the
beginning time and location and the ending time and location found in the
following record (which designates the beginning time and location for the
next observation period). The last record entered in the file for each
mobile run contains all missing indicators and gives the ending location
and time a corresponding event number.
The same editing procedures were used to identify missing and unaccept-
able values as were applied to the fixed station data.
Helicopter Data
Helicopter data were collected during two 10-d periods. The measure-
ments were digitized and recorded on magnetic cassette tape while in flight.
The data logger, which digitized and recorded data, scanned the sensing
instruments in continuous cycles. The cycle frequency was generally between
4 and 5 times a second. The measurements of S02, NO and 03 were also con-
tinuously recorded on strip charts. The strip charts and notes made by the
observers during flight have been used to select those parts of the cassette
tape records which have information relevant to the plume.
The sensing instruments produce an output voltage which is digitized
and recorded on magnetic tape. The digital records were scaled to engi-
neering units and corrected for nonlinearities in the output. All of the
outputs are sufficiently close to being linear that deviations can be
ignored, except for the relative humidity measurement. The response of
that instrument will vary with the resistance sensor, which was changed
each day. A mean response curve was derived by testing four elements. It
was found that linear interpolation of the value listed in Table 16 repro-
duces the response of the tested elements with only a few percent error
in the estimated relative humidity reading. As a result, this table was
used to scale all the relative humidity readings. In addition, the follow-
ing temperature correction was applied:
R = RO + 1 (T - 25)(RO - 33)
R = relative humidity, percent.
RQ = relative humidity measurement scaled by Table 16,
percent.
T = ambient air temperature, °C.
Differences among relative humidity values above 95% are not significant,
nor are differences below 20% significant, for the helicopter sensor.
106
-------
TABLE 14. MOBILE GROUND STATION RECORD FORMAT FOR MAGNETIC TAPE
Variable
UNIT
YEAR
JDAY
TIME
SEC
FLAG
EVENT
EASTING
NORTHING
ELEV
S02
NOX
NO2
NO
03
ws
WD
TEMP
Position
2-2
4-5
6-8
10-13
14-15
16-16
17-19
20-27
28-35
36-40
44-47
48-51
52-55
56-59
60-63
64-67
68-71
72-75
76
Format
11
12
13
14
12
Al
13
F8. 2
F8. 2
15
14
14
14
14
14
14
14
14
Description
Mobile unit # (1 or 2)
Last 2 digits of year (76 or 77)
Julian day
Time of scan (Eastern Standard Time)
Second of scan
A or P (indicating average or peak for the event
Event marker
Easting distance from plant (decameters)
Northing distance from plant (decameters)
Elevation (ft above sea level) at start of event
SO2 concentration (ppb)
NOx concentration (ppb)
NO2 concentration (ppb)
NO concentration (ppb)
03 concentration (ppb)
Wind speed (mph)
Wind direction (degrees)
Temperature (°C)
Blank
TABLE
15. HELICOPTER RECORD
FORMAT FOR MAGNETIC TAPE
Variable
UNIT
YEAR
JDAY
TIME
SEC
EVENT
EASTING
NORTHING
ELEV
S02
NOX
N02
NO
03
RH
TEMP
Position
2-2
4-5
6-8
10-13
14-15
17-19
20-27
28-35
36-40
44-47
48-51
52-55
56-59
60-63
64-67
68-83
84-87
88
Format
11
12
13
14
12
13
F8.2
F8.2
15
14
14
14
14
14
14
14
Description
Mobile Unit* (3)
Last 2 digits of year (76 or 77)
Julian day
Time of scan (Eastern Standard Time)
Second of scan
Event marker
Easting distance from plant (decameters)
Northing distance from plant (decameters)
Elevation (ft above sea level) of helicopter
SO2 concentration (ppb)
NOx concentration (ppb)
NO2 concentration (ppb)
NO concentration (ppb)
03 concentration (ppb)
Relative humidity (96)
Blank
Temperature (°C)
Blank
107
-------
TABLE 16. LINEAR INTERPOLATION TABLE FOR
RELATIVE HUMIDITY MEASUREMENTS
Recorded Voltage Relative Humidity, Percent*
< 0.5
0.5
0.8
1.0
1.5
2.0
3.0
5.0
>5.0
Reject
21
47
53
65
72
82
100
Reject
* Unconnected for temperature influence (see text).
The record format of the helicopter is listed in Table 15. There are
16 files of records covering the two 10-d observation periods. Each record
corresponds to a set of readings made during a single cycle of the data
logger. The data were also edited using the same procedures to identify
unaccpetable data as were applied to fixed station data. Record lengths
of 88 bytes are used. Missing value indicators of -999 are used for the
pollutants and relative humidity data. Any negative number can be con-
sidered missing for the temperature data.
Sulfate Data
Sulfate measurements consist of particulate matter analyzed for sul-
fate content collected on filters at any of six fixed stations (there were
only five samplers, but one was moved) or in the Mobile 1 van. The net
amount of sulfate collected as airborne aerosol during a 1-hour sampling
period must be estimated from the sulfate content measured by laboratory
analysis by corrections for several interferences, including:
t Sulfate content of filter material
• Sulfate accumulated during nonsampling exposure to the air
and during transport and handling
t S02 absorbed during sampling which is converted to sulfate
prior to analysis.
The sulfate content of the filter material is determined by submitting
blanks which are not exposed in the field to analysis and determining the
average value of the filter sulfate content. The accumulated sulfate is
determined by submitting one filter out of every 24 to the same exposure
treatment as sample filters, including mounting on a hi-vol sampling head,
except that no sample air is drawn through the filter. The sulfate content
108
-------
of the exposed blank is a measure of the accumulated sulfate. The amount
of absorbed S02 transformed to sulfate has not been identified. However,
the S02 concentration when the sample was taken is known, and the field
conditions in terms of temperature and relative humidity can be recon-
structed. When more is known about this mechanism, it may be possible to
correct for this effect.
A strip chart record of the average air flow rate during a sulfate
sampling hour is manually interpreted and recorded as is the duration of
the sample (which may be a minute or so more or less than an hour). The
net sulfate on a filter, corrected for filter sulfate, and accumulated
nonsample sulfate, divided by the air flow rate times the sample time, is
the measured sulfate concentration. Table 17 lists the record and charac-
ter formats used to store the sulfate data on magnetic tape. The time
designates when the sample was started. All of the data records are
entered In one data file with all of the records for a single station
grouped together in chronological order. The stations are identified
by the following two character codes and appear in the following order
in the data file:
Station
Castlewood
Hockey
Johnson
Kent's Ridge
Munsey
Nash's Ford
Mobile 1 Van
Code
CA
HO
JO
KE
MU
NA
MO
TABLE 17. SULFATE RECORD FORMAT FOR MAGNETIC TAPE
Variable
Station Name
Year
Month
Day of Month
Hour of Day
Minute of Hour
Filter Number
Net Sulfate Corrected
Sampled Air Volume
Sulfate Concentration
Elevation
Distance from Plant
Direction from Plant
Record
Position
1-2
3-4
5-6
7-8
9-10
11-12
13-16
17-20
21-25
26-30
31-34
35-38
39-42
43-44
Character
Format
A2
12
12
12
12
12
14
14
F5.1
F5.1
14
F4. 1
14
Units
Code
76 or 77
1 to 12
1 to 31
0 to 24 (Eastern Standard)
0 to 59
1 to 9999
Micro grams
Cubic meters
Micrograms per cubic meter
Meters above mean sea level
(mobile unit only)
Kilometers (mobile unit only)
Degrees (mobile unit only)
Blank
109
-------
Plant Operating Data
The plant operating data includes hourly measurements of the plant's
exhaust gases and operating characteristics plus the plant's weekly deter-
minations of heat conversion efficiency, fuel heat content and fuel sulfur
content. The basic information is used to generate daily and hourly
records of plant data which are stored in the format shown in Table 18. A
single data file containing, for each day during the monitoring period, a
daily header and 24 hourly records was formed and is available on magnetic
tape.
Each of the information items in Table 18 was generated in a consis-
tent manner. The daily sulfur content is a 7-d running average of measure-
ments made by the plant on samples drawn from coal shipments as they are
received. Each shipment is dumped into a storage pile, where it is mixed
with successive shipments on a daily basis. The 7-d averaging time was
found to be the best measure of sulfur content of fuel burned when compared
with measurements of S02 concentrations in the flue gas.
The heat content of the fuel and the heat conversion rate of each
boiler-generator combination is determined on a weekly basis for the fuel
fed to each boiler. These numbers are very stable on a week-to-week basis
as well as between boilers. No adjustments are made on a daily basis to
the plant's weekly estimates.
The temperature of the exhaust gas from each boiler is measured in
the breeching leading to the stack. The temperature, reported by the
plant in °F, is converted to °K.
Although the exhaust temperature is measured, the exhaust flow rate
is not measured. The air flow into each boiler and the oxygen content in
the air as it leaves each boiler are measured. The mass of the exhaust
gas is considerably greater than the sum of the fuel plus the intake air
due primarily to the leakage of additional air into the system. The ratio
of exhaust flow to intake flow has been measured in a few instances giving
the results listed in Table 19. The intake air flow rates reported in the
data records has been scaled to a reasonable estimate of exhaust flow rates
using the average ratio of exhaust to intake flow rates for each unit.
The hourly generator loads are included in the data set because they
are the best available measure of hourly variations in fuel consumption
and pollutant emission rates. These values are entered as received from
the plant's records.
The remaining numbers in the data record are the measured and esti-
mated rates of S02 and NO emissions. The measured emission rates are the
measured stack concentrations converted to a mass flow rate based on the
exhaust gas volume flow rate.
110
-------
TABLE 18. PLANT OPERATING CHARACTERISTIC RECORD FORMAT FOR MAGNETIC TAPE
Record
Position
Character
Format
Daily Header Record:
Year 1-2
Month of Year 3-4
Day of Month 5-6
Sulfur Content of Fuel 7-10
Fuel Heat Content, Unit #1 11-15
Heat Rate, Unitfl 16-20
Fuel Heat Content, Unit #2 21-25
Heat Rate, Unit #2 26-30
Fuel Heat Content, Unit #3 31-35
Heat Rate, Unit #3 36-40
41-76
Hourly Records
Stack Temperature, Unit #1 1-3
Air Exhaust Rate, Unit #1 4-6
Net Generator Load, Unit #1 7-9
Theoretical SO2 Emission Rate, #1 10-12
Estimated SO2 Emission Rate, #1 13-15
Theoretical NO Emission Rate, #1 16-18
Estimated NO Emission Rate, #1 19-21
Stack Temperature, Unit #2 22-24
Air Exhaust Rate, Unit #2 25-27
Net Generator Load, Unit #2 28-30
Theoretical SO2 Emission Rate, #2 31-33
Estimated SO2 Emission Rate, #2 34-36
Theoretical NO Emission Rate, #2 37-39
Estimated NO Emission Rate, #2 40-42
Stack Temperature, Unit #3 43-45
Air Exhaust Rate, Unit #3 46-48
Net Generator Load, Unit #3 49-51
Theoretical SO2 Emission Rate, #3 52-54
Measured SO2 Emission Rate, #3 55-57
Estimated SO2 Emission Rate, #3 58-60
Theoretical NO Emission Rate, #3 61-63
Measured NO Emission Rate, #3 64-66
Estimated NO Emission Rate, #3 67-69
Barometric Pressure at Plant 70-74
75-76
12
12
12
F4.2
IS
15
15
IS
IS
15
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
F5. 2
76 or 77
1 to 12
1 to 31
Percent
BTU per pound
BTU per kilowatt-hour
BTU per pound
BTU per kilowatt-hour
BTU per pound
BTU per kilowatt-hour
Blank
Degrees Kelvin
Cubic meters per second
Megawatts
Grams per second
Grams per second
Grams per second
Grams per second
Degrees Kelvin
Cubic meters per second
Megawatts
Grams per second
Grams per second
Grams per second
Grams per second
Degrees Kelvin
Cubic meters per second
Megawatts
Grams per second
Grams per second
Grams per second
Grams per second
Grams per second
Grams per second
In. of mercury
Blank
111
-------
TABLE 19. CORRESPONDING MEASUREMENTS OF
BOILER INTAKE AND STACK EXHAUST GAS FLOW RATES
Unit Number
1
1
mean
2
2
mean
3
3
3
3
3
3
3
3
3
mean
Intake Air (m^/sec)
266.41
268.58
252. 63
252.32
252. 15
252.15
253.79
250.66
253.04
247.22
266.32
238.72
253.64
Exhaust Air (m^/sec)
384.64
395.15
387.96
396. 10
390.27
390.27
388.17
380.17
381.48
373.57
397.42
385.15
363.44
Ratio
1.444
1.471
1.458
1.536
1.570
1.553
1.548
1.548
1.529
1.519
1.508
1.511
1.492
1.613
1.433
1.522
112
-------
S02 Emissions
Theoretical SOo.emission estimates were calculated by the following
"lotion: n GxKxSx F
y - fj
where
Q = stack emission rate, g/sec
G = rate of output by generators, kw
K = thermal conversion rate of boiler/generator units, BTU/kw sec
S = sulfur content of fuel, mass ratio (dimensionless)
F = 1.9 = sulfur dioxide emissions per unit of fuel sulfur, mass
ratio (dimensionless)
H = heat content of fuel, BTU/g.
The thermal conversion rate (K) was computed by the plant for each generator/
boiler unit each week. These estimates for the three generators were
furnished to us by the plant.
In order to estimate SOoemission rates from units 1 and 2, a
multiple linear correlation and regression analysis was performed
using the measured hourly average S02 emission rates for Unit #3, the
generator load data, and the sulfur content data. When the S02 emission
rate was treated as the dependent variable and the load and sulfur content
were as independent variables, the best correlation was obtained when the
analysis was limited to the range of S02 emission rates from 75g/sec to
550g/sec and 14 d of suspicious monitoring data were eliminated from the
analysis. The following regression formula was obtained:
QSQ2 * 485.372 PSULF + 1.185G - 331.007
where PSULF is the percent sulfur content of the coal. The multiple cor-
relation coefficient between the SOg emission rate and the independent
variables PSULF and G is 0.72 and the standard error of estimate is 57.9g/sec.
For typical Clinch River Plant conditions of PSULF = 0.8%, G = 200 MW,
K = 9,200 BTu/KW-hr, and H = 12,000 BTu/lb., the regression formula yields
an S02 emission rate of 294.3 g/sec, while the EPA emission factor from
AP-42 yields 293.9g/sec.
Higher correlations are obtained when the theoretically computed mass
of S02 emitted on a weekly basis is compared to the total measured S02
emitted over a week. Using 30 weeks of these comparisons a linear corre-
lation coefficient of 0.92 was obtained.
The 14 d of suspicious S02 stack monitoring data still remain on the
plant data file since the data are qualitatively useful despite the stack
monitor being out of calibration. The measured S02 emission rates on these
days are higher than the typical rates. The dates are as follows:
113
-------
January 16, 17, 1977
March 21, 22, 1977
June 24, 25, 1977
July 25, 26, 27, 1977
August 2, 3, 1977
August 10, 11, 1977
August 16, 1977
NO Emissions
Theoretical NO emission estimates were calculated by the following
equation:
n _ E x G x K
g H
where
_3
E = 5.87 x 10 = NO emissions per unit of fuel, mass
ratio (dimensionless).
The emission factor (E) is based on the NO molar equivalent to 9 kg of NOo
per m ton of bituminous coal for a dry bottom, pulverized coal furnace
as given in EPA document AP-42, "Compilation of Air Pollution Emission
Factors".
In order to estimate NO emissions from units 1 and 2, linear
regression analysis was also performed on the hourly average NO-measured
emission rates for Unit #3 and the generator load data. When NO was
treated as the dependent variable and the generator load as the indepen-
dent variable the following regression formula was obtained:
QNQ = 1.1036 - 33.152
Fifty-one days of suspicious NO monitoring data were not considered in the
analysis used to form this equation. A linear correlation coefficient of
0.68 and a standard error of estimate of 44.7g/sec were obtained.
The monitored NO emission rates from the 51 days still remain on the
plant data file since the data is qualitatively useful despite the readings
being high compared with typical NO emission rates, these periods gener-
ally began a few days prior to the periodic calibration of the stack moni-
tor by Environmental Data Corporation. The dates are as follows:
December 29, 1976 - January 15, 1977
February 2, 1977 - February 4, 1977
March 1, 1977 - March 15, 1977
May 26, 1977 - June 9, 1977
For typical Clinch River Plant conditions of G = 200 MW, K = 9,200
BTu/KW-hr and H = 12,000 BTu/lb., the regression formula yields an NO emis-
sion rate of 187.4g/sec. When the emission factor (E) is used with these
same conditions, an NO emission rate of 113.5g/sec is obtained.
114
-------
Balloon Data
This data consists of reductions of double and single theodolite
observations of pibal releases, some with associated temperature-sondes, to
give vertical profiles of wind speed, wind direction, and temperature.
Ground-level readings of wet and dry bulb temperatures and relative
humidity were taken with some of the releases and are also included.
Releases are identified by Gregorian Date and Eastern Standard Time.
At each recorded observation height, wind speed and wind direction or
temperature alone, or all three may be reported. The recorded observa-
tion time refers to the elapsed time in minutes since the pibal/t-sonde
release. The temperature corresponds to the recorded height, but the
wind speed and direction are averages over the layer extending from the
height one minute earlier to the height at which they are recorded.
Therefore, the speed and direction should be considered applicable to
a height midway between the height one minute earlier and the height
at which they are recorded.
Data relating to a single release is recorded on the file in records
of three types, including a header record, one or more observation records
and an end of observations record. The formats of each of the three types
of records are defined in Table 20. All of the balloon observation records
are contained in a single data file.
Missing values for integer (I) character variables are indicated by
the value -99; for decimal (F) character variables, by the value -99.9.
It may be noted that all records are 32 bytes long.
115
-------
TABLE 20. BALLOON DATA RECORD FORMAT FOR MAGNETIC TAPE
a. File Organization
First
Release
Second
Release
Last
Release
Header
Record
Observation
Record
Header
Record
Header
Record
End of
File
Observation
Record
Observation
Record
Observation
Record
End
Record
Observation
Record
End
Record
Observation
Record
End
Record
b. Header Record Format
Record
Variable Position
Year 1-2
Month of Year 3-4
Day of Month 5-6
Release Time 7-10
Type of Record 11-11
Baseline 12-13
Relative Humidity, Ground Level 14-16
Dry Bulb Temperature, Ground Level 17-21
Wet Bulb Temperature, Ground Level 22-26
Blank 27-32
Character
Format
12
12
12
14
II
A2
13
F5. 1
F5. 1
Units
76-77
1 to 12
1 to 31
Eastern standard time
11 ^n
Code (see Section 4)
Percent
Degrees centigrade
Degrees centigrade
Missing
Indicator
-99
-99.9
-99.9
(continued)
116
-------
c. Observation Record Format
Variable
Year
Month of Year
Day of Month
Release Time
Type of Record
Observation Time (end)
Height (at observation time)
Wind Speed (average)*
Wind Direction (average)*
Temperature (at observation time)
d. End of Record Format
Variable
Year
Month of Year
Day of Month
Time
Type
Blank
Record
Position
1-2
3-4
5-6
7-10
11-11
12-15
16-19
20-22
23-25
26-30
31-32
Position
1-2
3-4
5-6
7-10
11-11
12-32
Character
Format
12
12
12
14
11
F4-1
14
13
13
F5-1
Character
Format
12
12
12
14
11
Units
76-77
1 to 12
1 to 31
Eastern standard time
"2"
Minutes
Meters
Meters per second
Degrees
Degrees centigrade
Blank
Units
76-77
1 to 12
1 to 31
Eastern standard time
1151!
Blank
Missing
Indicator
-99
-99
-99.9
*Averaged over 1-minute layer below this observation height.
117
-------
SECTION 6
PROJECT CHRONOLOGY
CALENDAR OF EVENTS
Table 21 presents a chronological list of events which describe sig-
nificant field events. These are intended as a guide in interpreting the
data collected during the field project. Major changes in the observing
capability of this sytem are noted and causes for the change are cited.
Also noted are events which may increase or decrease the degree of confi-
dence associated with the data.
118
-------
TABLE 21. CHRONOLOGY OF SIGNIFICANT FIELD EVENTS
Date
Event
June 1, 1976
June 2, 1976
June 3, 1976
June 17, 1976
June 24, 1976
July 1, 1976
July 28, 1976
July 29, 1976
August 16, 1976
August 20, 1976
August 23, 1976
August 24, 1976
September 8, 1976
September 14, 1976
September 18, 1976
September 29, 1976
October 3, 1976
October 5, 1976
Start of field data collection. Valid data being recorded from the
Tower, Munsey and Nash's Ford monitoring sites via telemetering to
the central facility.
First pilot balloon measurements taken.
First run made with mobile van.
Valid data being recorded from Hockey site via telemetering to the
central facility.
Valid data being recorded from Castlewood site via telemetering to
the central facility.
Valid digitized data being recorded from Kent's Ridge site via onsite
cassette tape recorder.
Test run with helicopter over Bristol.
Tower site struck by lightning and no valid data recorded until
August 20.
Began recording 30 m wind speed and wind direction at Hockey site.
Valid SO data from Tower being recorded on strip chart.
Ci
Hi-vol samplers begin collection for sulfates at Munsey, Nash's Ford,
Hockey and Castlewood sites.
Valid data from Tower site being telemetered to central facility.
Full complement of meteorological measurements restored to opera-
tion at Tower site.
Telephone line installed and DAS operating at Kent's Ridge. Cable
failure at Tower site reduced meteorological data to 1 level.
Johnson SO data recorded on strip chart (only until September 24).
Helicopter sampling with power plant precipitators shut down.
Helicopter sampling with power plant precipitators shut down.
Began taking T-Sondes.
(continued)
119
-------
TABLE 21. (continued)
Date
Event
October 8, 1976
October 11, 1976
October 20, 1976
October 21, 1976
October 24, 1976
October 26, 1976
October 27, 1976
November 4, 1976
November 7, 1976
November 8, 1976
November 9, 1976
November 10, 1976
November 11, 1976
November 14, 1976
November 15, 1976
November 16, 1976
November 17, 1976
November 18, 1976
November 24, 1976
December 10, 1976
December 19, 1976
December 30, 1976
Approximately 3|" of rainfall measured in 24-hour period.
All stations taken off auto zero/span.
Power plant unit #2 is down for repair, 4 weeks.
Computer is down until October 26.
Lower level wind speed sensor installed at Hockey site.
Computer back in service.
Lambert site put on-line.
In-stack monitor put in service.
Began converting the NO calibration methodology to provide a more
stable calibration gas.
Helicopter sampling.
Helicopter sampling.
Helicopter sampling.
Helicopter sampling.
Full complement of meteorological sensors back in operation at
Tower site.
Service "A" circuit discontinued.
Helicopter sampling.
Helicopter sampling; Johnson site on line.
All 8440 monitors now operating with stable calibration gas.
Service "C" circuit discontinued.
Hi-Vol sampler from Munsey site installed at Johnson site.
Stack monitor NO channel put in service by EDC; recorded on strip
chart.
Received long-range antenna for T-Sonde receiver.
(continued)
120
-------
TABLE 21. (continued)
Date
Event
January 20, 1977
January 20 - 21,
1977
January 31, 1977
February 9, 1977
February 11, 1977
February 15-23,
1977
March 5, 1977
March 7, 1977
March 18, 1977
March 22, 1977
March 24, 1977
March 30, 1977
Aprils, 1977
April 4, 1977
April 5, 1977
April 6, 1977
April 8, 1977
Changed ADAM System magnetic tape format.
RTI System Audit.
1) Second blazer obtained for use in field operations; 2) sent T-Sonde
receiver to Contel for repairs.
Mobile 1 portable generator taken in for major repairs. Repairs not
completed until March 7 due to lack of parts.
Received repaired T-Sonde receiver.
RTI Performance Audit.
Purchased back-up portable generator for Mobile 1.
Original Mobile 1 portable generator repaired and returned to field.
Attempts at using second surface mobile sampling unit aborted.
Vehicle and personnel involved were switched to intensified upper
air program.
Received EPA Radiosonde equipment at the field site.
Received and installed microbarograph at the field office.
Field personnel constructed T-Sonde antenna mount for the theodolite.
T-Sonde/Pibal program thus reduced from 3 to 2 people.
Heavy rains began in late afternoon.
Continued heavy rains. Severe water leak in trailer roof at Tower
site, but no damage to instruments.
Heavy flooding in Southwest Virginia. Castlewood trailer destroyed
by flood waters from Clinch River. Power plant shut down all
boilers. Moss #3 coal cleaning plant shut down. Stack monitor
damaged. ,
^
Began recovery operations for Castlewood equipment.
Moss #3 coal cleaning plant back in operation.
(continued)
121
-------
TABLE 21. (continued)
Date
Event
April 13, 1977
April 15, 1977
April 17, 1977
April 20, 1977
May 31, 1977
June 6, 1977
June 9, 1977
June 10, 1977
June 11, 1977
June 13, 1977
June 15, 1977
June 16, 1977
June 18, 1977
June 22, 1977
June 24, 1977
June 25, 1977
June 26, 1977
June 30, 1977
Power plant Unit #3 fired up and brought on line.
Stack monitor NO channel repaired. SO channel still not operable.
-------
TABLE 21. (continued)
Date
Event
JulyS, 1977
July 6, 1977
July 7, 1977
July 10, 1977
July 12, 1977
July 15, 1977
July 18, 1977
July 29, 1977
August 1, 1977
August 2, 1977
August 5, 1977
August 12, 1977
August 16, 1977
August 17, 1977
August 20, 1977
August 22, 1977
August 25, 1977
August 28, 1977
September 1, 1977
September 9, 1977
September 30, 1977
Adjustment to NO dilution flow in Kent's Ridge calibrator.
Power failure at stack monitor.
Adjustment to NO dilution flow in Munscy's calibrator.
Lightning strike at Tower site disabling SO analyzer and most equip-
ment as well as data logger. NO on strip chart only.
Performance audit (informal) at Munsey site. Stack monitor returned
to service.
Lambert trailer installed at Castlewood site; Tower site less meteorol-
ogical data, NO , NO.
Begin intensive phase with helicopter sampling.
End intensive phase.
Adjustment to NO dilution in Nash's Ford calibrator.
Adjustment to NO dilution in Tower calibrator.
Begin performance audit.
End performance audit.
Install new style O generators in Munsey* Nash's Ford NO analyzers
analyzers.
Install new style O generator in Kent's Ridge NO analyzer.
Unit #3 shut down until after end of project.
Castlewood site has power.
First data from Castlewood since flood - wind direction on strip chart.
Castlewood site returned to ADAM system - meteorological data only.
Castlewood reporting SO as well as meteorological data.
Reallign lower wind direction sensor at Hockey.
Final day of formal data collection.
123
-------
Section 7
References
Frazier, C. D., and J. M. Williams. 1977. Audit of Clinch River,
Virginia Network. EPA Contract No. 68-02-2725.Research Triangle
Institute, Research Triangle Park, N.C.
Smith, F., C. E. Decker and R. B. Strong. 1977. Audit of Air Monitoring
Network for Clinch River Power Plant Study in Complex Terrain. EPA
Contract No. 68-02-2156. Research Triangle Institute, Research
Triangle Park, N.C.
Engineering Science. 1978. EPA Reference Method Measurements of
Emissions from Clinch River Power Plant"!EPA Contract No. 68-02-2815.
Engineering Science, McLean, VA.
Holzworth, G. C. 1972. "Mixing Heights, Wind Speeds, and Potential
for Urban Air Pollution Throughout the Contiguous United States,"
Environmental Protection Agency, AP-101. Research Triangle Park,
North Carolina.
124
-------
APPENDIX A
SITE LAYOUT DIAGRAMS FOR EIGHT FIXED MONITORING STATIONS
Town
4
MM ^
ra com. x~*N
• -n 226°
KMT
3360M
to
PUNT
TOWER
SITi
Elevation 585 m
125
-------
HILL
UPSLOPE OF HILL STARTS
40' FROM TRAILER
THIN
BRANCHES
4250 M
TO
POWER PLANT
4407 M
TO
COAL PREPARATION
PLANT
MUNSEY
SITE
Elevation 597 m
TOP OF TREE
15' ABOVE
WIND SENSOR
126
-------
SLIGHT
RISE
APR 10"
VALLEY
25"
DEEP
HILL
HILL
SINKHOLE
100'
FROM TRAILER
TO • • TO
COAL PLANT *
PREP. 10820 M
PLANT
84S5M
NASH
FORD
SITE
Elevation 567 m
VALLEY
HILL
VALLEY
HILL
800'
FROM
TRAILER
127
-------
5710 M
TO
POWER PUNT
6038 M
TO
COAL PREPARATION
PLANT
30 METER
TOWER
WALNUT
TREE
HOCKEY
SITE
Elevation 792 m
128
-------
HILL
LJ
1TO8M
TO
COAL PREP.
PLANT
8280 M
TO
POWER
PLANT
20'
HIGH
IS'STACK
106'
CASTLEWOOO
SITE
Elevation 451 m
tt"
THE!
BANNER
HOUSE
300'FROM
TRAILER
CORN
FIELD
CASTLEWOOO
129
-------
30 KM
TO
POWER PLANT
HILL
HORSE & BARN
MOUNTAIN
KENT'S
RIDGE
SITE
Elevation 756 m
130
-------
131
-------
S»9 iH
C9
132
-------
APPENDIX B
MAJOR INSTRUMENT PERFORMANCE CHARACTERISTICS
MONITOR LABS MODEL 8450 SULFUR ANALYZER
The ML Model 8450 operates using the sulphur specific Flame Photo-
metric Detection (FPD) measurement principle. The instrument measures
the change in light intensity in a hydrogen flame burner chamber as
sample air is introduced. The light intensity is measured and converted
to a linear output that is proportional to the amount of sulfur compounds
in the sample air. Since the 8450 responds to most all sulfur compounds,
it was necessary to incorporate a Monitor Labs Model 8740 HgS Scrubber
to remove ambient H2S from the sample air. The scrubber removes HgS by
heating the sample air to 130°C and passing it through a series of scrubber
elements.
The manufacturer's stated specifications for the ML Model 8450 are
as follows:
Ranges: 0-100 ppb, 0-1000 ppb
Maximum Noise: 1 ppb (peak-to-peak)
Minimum Detectable 2 ppb (stated as 2 times maximum
Concentration: peak-to-peak)
Lag Time: Less than 1 second
Precision: 1% full scale
Zero Drift: +_ 2 ppb/day, +_ 5 ppb/3 days
Span Drift: + 5 ppb/day, + 10 ppb/3 days
for 0-1000 ppb range
Linearity: + 1%
Sample Flowrate: T80 ml/min +_ 10 ml/min
Time Constant: Selectable 1, 5, 20, 60, 120
seconds
Maintenance Interval: 45 days
Unattended Operation: 7 days
Support Gas Required: Hydrogen, 99.9% pure
The manufacturer's stated specifications for the ML Model 8740 are
as follows:
Efficiency: 99% 83$ removal
Maximum SO? Depletion: Less than 1%
Scrubber Element Life: 45,000 ppb H2S-hours
Pressure Drop Thru 8740: 0.5" H20
133
-------
The ML Model 8450 and Model 8740 combination performed well during
the field program. Only one or two 8450 units occasionally exceeded the
stated drift specifications, but no data loss was incurred as a result.
Minor instrument malfunction and component failure did result in a small
amount of lost data, but on the whole the 8450 units provided very stable
operation. The 8450 minimum detectable concentration was effectively
raised to between 5 and 8 ppb by the inclusion of the 8740 in the system.
Field tests indicated that the H2S Scrubber suppressed the S02 concentra-
tions by about 3 or 4 ppb as compared to 8450 response without the scrubber
in the system. Only one 8740 scrubber element failure was recorded during
the field project.
MONITOR LABS MODEL 8440 NITROGEN OXIDES ANALYZER
The ML 8440 is a dual channel analyzer that simultaneously measured
NO and NOX using a chemiluminescence measurement system. N02 is then
determined by simple subtraction of the NO signal from the NOX signal.
In the 8440, the sample stream is split into two separate streams.
One stream first passes through a molybdenum converter that converts
all the N02 to NO, and then goes into the NOX reaction chamber. The
second stream passes directly into the NO reaction chamber after it passes
through a length of teflon tubing that delays the sample stream the same
length of time as the NOX sample stream is delayed by the converter.
This delay conserves the time integrity of the split sample streams.
Each of the two reaction chambers are provided a supply of ozone from the
8440 ozone generation system. The reaction between the ozone and any
NO that is in the sample stream produces light. The light intensity is
then measured and electronically converted to a linear signal that is
proportional to the amounts of NO in each reaction chamber. A temperature
controlled critical orifice assembly accurately controls all the flows
in the 8440 system.
The manufacturer's stated specifications for the ML Model 8440 are
as follows:
Ranges: 0 to 200, 500, 1000, 2000, or 5000 ppb
Minimum Detectable
Concentration: 2 ppb
Lag Time: Less than 5 seconds
Zero Instability: Less than 0.1% full scale per year
Span Instability: Less than +_!% full scale per day
Repeatability: + 1% NO and NOX, +1.4% N02
Sample Flowrate: Less than 500 ml/min each channel
nominal
Ozone Flowrate: 80 ml/min each channel nominal
134
-------
In the field, the 8440 units were found to experience a large zero
drift on the order of 10 or more ppb per week. Large span drifts also
occurred somewhat regularly. The outputs of all the 8440 units contained
varying amounts of noise that were independent of the magnitude of the
output signal. The noise level in some of the instruments ranged as high
as 10 or 15 ppb on the 0-2000 ppb range and a 20 second time constant.
The minimum detectable concentration was, therefore, raised to not less
than 10 ppb.
The large span and zero drifts in the 8440's did not result in any
lost data due to the frequency of the multipoint calibrations performed
on the instruments. Some data loss was incurred as a result of component
failure in the 8440. The ozone generation system was a constant source
of problems that resulted in data losses early in the project. A new
ozone generating system developed by Monitor Labs relieved the failure
problems of the earlier system.
MONITOR LABS MODEL 841OA OZONE ANALYZER
The ML Model 8410A measures ozone using the chemiluminescence principle.
Sample air is introduced into a reaction chamber where it is combined with
ethylene from an external source. Reaction between the ethylene and any
ozone that is in the sample air produces light. The intensity of this
light is measured, amplified, and converted to a linear output proportional
to the ozone concentration in the sample air.
The manufacturer's stated specifications for the ML Model 841OA are
as follows:
Ranges: 0 to 200, 500, 1000, 2000, or 5000 ppb
Minimum Detectable
Concentration: 1 ppb
Lag Time: Less than 3 seconds
Zero Instability: Less than 0.1% full scale per year
Span Instability: Less than +_ 1% full scale per day
Repeatability: +. 1%
Sample Flowrate: 300 cc/min nominal
Ethylene Flowrate: 30 cc/min nominal
Support Gas Required: Ethylene, CP grade
The ML Model 841OA provided reliable performance during the field
program. Although not designed as a portable instrument, the 8410A stood
up very well during its use in both the surface and airborne mobile
sampling programs. Zero and span drifts were very small and the outputs
contained little if any noise. A small amount of data loss was incurred
as a result of infrequent minor component failure in the 8410A.
135
-------
THETA SENSORS, INC. SERIES 7111 S02 MONITOR
The TSI Series 7111 S02 Monitor is a single-channel instrument that
measures S02 concentrations through the use of a permselective membrane
permitting selective redox reactions to occur within a sealed electro-
chemical transducer. The permselective membrane permits the passage
of SOg into the transducer cell while inhibiting the passage of interfer-
ing gases. The transducer cell produces a signal that is directly pro-
portional to the S02 concentration in the sample gas and is completely
linear over the instrument range.
The manufacturer's stated specifications for the TSI Series 7111
S02 Monitor are as follows:
Range: 0 to 1000, 3000, or 10,000 ppb
Sensitivity: 1% of full scale
Accuracy: +_ 2% of full scale
Repeatability: +_ 1% of full scale
Zero Drift: +. 2% per day
Span Drift: +_ 1% per day
Sample Flowrate: 0.6 standard cubic feet per hour
Response Time: Less than 60 seconds
Temperature Compensation: +_ 2% over any 10°F range
Temperature Range: 32°F to 113°F
Several problems were encountered with the TSI instrument in its
use in the mobile sample program. Transducer failure was common and
caused various drift problems that led to moderate loss of data. Although
sometimes the instrument would perform as designed, it often became sensi-
tive to the slightest motions encountered in the mobile sampling vehicles.
This sensitivity resulted in full scale spikes, large zero and span drifts
of 20 percent or more, and overall erratic performance. The TSI S02
monitor also experienced large negative zero and span drifts on the order
of several percent of full scale per hour when the sample air became very
cold and the instrument itself was kept within stated operating tempera-
tures. Similar positive drifts occurred when the instrument was operated
in ambient temperatures near 95°F even though this was within the stated
operating range.
Not all drift problems resulted in loss of data since most of the
drifts were linear with time and could be removed during data analysis.
However, the spiking and erratic behavior associated with the motion
sensitivity did result in moderate data losses.
136
-------
MONITOR LABS MODEL 8500 CALIBRATOR
The ML Model 8500 used in the field program contained some or all
of the following features:
• Zero Air production system
• NO dilution system
• N02 permeation system
• S02 permeation system
• Ozone generation system.
The Zero Air production system produces zero air from ambient air
by first oxidizing all NO to N02 then scrubbing out all S02» N02, ozone,
and other unwanted gases with an activated charcoal filter. The zero air
thus produced is then available for use as zero or as dilution air for
the NO, N02, S02, or ozone calibration gases.
The NO dilution system incorporates a cylinder of high concentration
NO in dry nitrogen. The cylinder NO is introduced into the 8500 through
a regulated flow capillary restrictor. This controlled NO flow is mixed
with the dilution air to provide calibration level NO concentrations.
The S02 and N02 permeation tubes are contained in a temperature
controlled oven that stabilizes the tubes' emission rates. Regulated
dilution air is passed over the permeable membranes to provide S02 or N02
calibration gas to the respective analyzers.
Calibration level ozone concentrations are produced by the ultra-
violet irradiation of zero air. The irradiation takes place in a tempera-
ture controlled oven that allows stable, controlled ozone production.
The S02, N02, and ozone systems were calibrated by Monitor Labs using
the following methods:
S02: .Comparison to NBS SRM
N02: Gas Phase Titration per the reference method
Ozone: Bubbling into neutral buffered potassium iodide
per the reference method.
All the dilution and zero air flows are controlled by precision 150 mm
scale rotameters that were initially calibrated by Monitor Labs and
routinely checked by GEOMET using NBS traceable mass flowmeters.
137
-------
The manufacturer's stated specifications for the ML Model 8500 are
as follows:
Accuracy
ML Calibration of Sources:
Flow Calibrations:
Dilution Air Source
Total air flow available:
Total dilution flow/
channel:
Oven Temperature Control:
+. 5% relative to lab standards
+. 2% relative to lab standards
1000-8000 cc/min
400-4000 cc/min
+ 0.1°C
One 8500 pump failure was recorded during the field project; however,
no loss of analyzer data occurred as a result. The pressure regulating
device in the NO dilution system was found to be in error from 10 to
50 percent in three of the 8500 units. As a result, the NO capillary
flows had to be measured with a mass flowmeter to insure the generation
of precise quantities of NO calibration gas. No data losses occurred as
a result of the regulator errors since the analyzer calibration data
could be reconstructed and corrected.
METEOROLOGY RESEARCH, INC. MODEL 1022 WIND SET
The MRI Model 1022 Wind Set consists of a cup anemometer, a wind
vane, and a mounting crossarm. Also incorporated in the system is the
MRI Model 1002 Transmuter.
The MRI cup anemometer works on the light chopper principle. A
light chopper disc causes varying conductivity in a photocell as the
anemometer cups rotate in the wind. The signal generated by the photo-
cell is converted to a linear analog DC output by the Model 1002 Trans-
muter.
The MRI wind vane uses a dual potentiometer - wiper combination to
give a 0° to 540° output that eliminates crossover problems at 360°.
The potentiometer resistance is converted by the transmuter to a linear
analog DC signal that is proportional to the wind direction. The
transmuter also computes sigma azimuth values from the wind vane output.
The manufacturer's stated specifications for the MRI Model 1022
Wind Set are as follows:
Wind Speed
Starting Threshold:
Response Distance:
Flow Coefficient:
0.22 mps
less than 1.5 meters for 63% recovery
1.7 m/rev.
138
-------
Accuracy: ± 0.06 mps or 1% of wind speed
(whichever is greater)
Range: 0.22 mps to 54 mps maximum
540° Wind Direction
Starting Threshold: 0.31 mps
Distance Constant: 1.13 m
Damping Ratio: 0.4 at 10° angle of attack
Accuracy: +_ 2.5°
Range: 0° to 540°
Phasing between
Pot A and Pot B
wipers: 0°
The MRI wind equipment performed well during the field program.
Some bearing deterioration resulting in increased starting thresholds
occurred in a few wind speed sensors. However, periodic lubrication
and bearing replacement kept data losses to a minimum. Infrequent
photocell failure was experienced, but again data losses were kept to
a minimum with the help of spare wind speed sensors and spare photocell
assemblies kept in the field.
The wind direction sensors performed remarkably well with only two
or three potentiometer failures resulting in minimal data loss.
Most data losses that did occur with the MRI wind equipment were
as a result of infrequent circuit board failure in the 1002 transmuter.
Even these losses, however, were minimal because of the large number
of spare circuit boards kept in the field.
METEOROLOGY RESEARCH, INC. MODEL 1053 VECTOR VANE
The MRI Model 1053 Vector Vane measures wind speed, horizontal
wind direction (azimuth), and vertical wind direction (elevation).
The MRI Model 1001 Transmuter was used with the Vector Vane in the
Clinch River project.
Both the wind speed and azimuth measurement principles of the
Vector Vane are the same as the MRI Model 1022 Wind Set. The Vector
Vane uses a single potentiometer and wiper combination to measure
elevation. The Model 1001 transmuter operates in the same manner as the
Model 1002 but is larger and can accommodate up to 10 circuit cards.
139
-------
The manufacturer's stated specifications for the MRI Model 1053
Vector Vane are as follows:
Wind Speed
Starting Threshold: 0.33 mps
Response Distance: 0.9 meters (63% recovery)
Accuracy: +_ 0.09 mps or 1% of wind speed
(whichever is greater)
Range: 0.33 mps to 50 mps maximum
Wind Direction and Elevation
Starting Threshold: 0.33 mps
Delay Distance: 0.9 meters (50% recovery)
Damping Ratio: 0.5
Azimuth Accuracy: ±5.4° (1% of full scale)
Elevation Accuracy: ±2.4° (2% of full scale)
Range
Azimuth: 0° to 540°
Elevation: -60° to +60°
The MRI Vector Vane and 1001 transmuter combination provided reliable
field performance. No bearing failures were recorded, but two Vector
Vane propellers required replacement after locally high winds damaged
them. Data losses as a result of the damaged propellers and infrequent
transmuter circuit board failures were minimal.
METEOROLOGY RESEARCH, INC. MODEL 815-1 TEMPERATURE SENSOR
The MRI Model 815-1 Temperature Sensor is a naturally aspirated
system that is proportional to the ambient temperature when combined
with the MRI Model 1002 Transmuter. The temperature sensor consists of
a dual thermistor and resistor network to produce a linear resistance
change with an ambient temperature change. The sensor housing is
designed to shield the thermistor elements from both long- and short-
wave radiation while allowing maximum transfer of heat.
The manufacturer's stated specifications for the MRI Model 815-1
Temperature Sensor are as follows:
Sensing Housing
Effect of Radiation: 0.1°C with a wind speed of 1.5 mps
or greater
Air Flow: Naturally aspirated
Temperature Element
Accuracy: + 0.25°C
Output: -15968 to +5835.5 ohms (variable)
Range: -30°C to +50°C
140
-------
No temperature sensor failures were recorded during the entire course
of the field program. However, one or two transmuter circuit board
failures did result in a very minimal amount of lost data.
METEOROLOGY RESEARCH, INC. MODEL 840 SERIES TEMPERATURE SENSORS
The MRI Model 840 Series Temperature Sensor is a shielded, power
aspirated system that operated on the same principle as the MRI Model
815-1. Variations in the Model 840 series allowed the measurement of
A temperature at two levels at the Tower site. Because of the number
of circuit cards involved in this system, the Model 1001 Transmuter was
used.
The manufacturer's stated specifications for the MRI Model 840 Series
Temperature Sensor are as follows:
Sensor Housing
Effect of Radiation: Less than 0.05°C under maximum solar
radiation conditions
Air Flow: 10 to 30 feet per second
Blower: Continuous duty
Temperature Element
Accuracy: +. 0.25°C
Output: -15968 to 5835.5 ohms (variable)
Range: -30°C to +50°C
No temperature sensor failures occurred during the field project;
however, very minimal data losses occurred as a result of infrequent
transmuter circuit board failure.
METEOROLOGY RESEARCH, INC. MODEL 840 SERIES RELATIVE HUMIDITY SENSOR
The MRI Model 840 Series Relative Humidity Sensor is a shielded,
power aspirated system that produces resistance changes proportional to
changes in ambient relative humidity. A pair of thermally matched
strain gauges sense the stress in a system of cellulose crystalite
structures comprised of organic and inorganic crystals. The hygro-
mechanical stress of these crystals related to the moisture content
of the air. The MRI Model 1001 Transmuter was used with the relative
humidity system.
141
-------
The manufacturer's stated specifications for the MRI Model 840 Series
Relative Humidity Sensor are as follows:
Sensor Housing
Air Flow: 10 to 30 feet per second
Blower: Continuous duty
Humidity Element
Accuracy: +3.0% RH
Range: 0% to 100% RH
The failure of the humidity element in the only sensor used in the
field project resulted in considerable loss of data. An MRI spokesman
stated that the elements generally deteriorated gradually so that a pre-
cise failure date would be hard to pinpoint. Because of apparent severe
design problems, MRI discontinued production and marketing of the relative
humidity sensor shortly after the start of the Clinch River project.
METEOROLOGY RESEARCH, INC. MODEL 370 RAIN GAGE
The MRI Model 370 Rain Gage contains an AC powered thermostatically
controlled heater to convert snow and freezing rain to a measurable liquid
state. A deep collecting funnel and a tipping bucket-reed switch combina-
tion provide the means for measuring precipitation. As the bucket is filled
from the collecting funnel, it tips and closes the reed switch. Conversion
from the number of switch closures to a DC signal proportional to precipi-
tation amounts is performed by a circuit card in the MRI Model 1001
Transmuter.
The manufacturer's stated specifications for the MRI Model 370 Rain
Gage are as follows:
Accuracy: +_ 1% at 3 inch/hour rainfall rate
Bucket Capacity: 0.01" rain per tip
Reset Value: 1.00" (resets to 0.00")
No sensor or circuit board failures occurred during the field project.
Periodic cleaning of the collector tube prevented any data losses due to
plugging.
142
-------
INTERNATIONAL LIGHT, INC. MODEL IL 200 UV-VISIBLE PHOTOMETER
The International Light, Inc. Model IL 200 UV-Visible Photometer
measures light intensity with a vacuum photodiode tube. The spectral
range measured is determined by the type of filters used on the optical
input of the photodiode. (For the Clinch River project, response was
limited to wavelengths between 355 and 390 nanometers.) The light
intensity is measured in microamperes, but the instrument output is an
analog DC voltage that is proportional to the light intensity.
The manufacturer's stated specifications for the ILI Model IL 200
UV-Visible Photometer are as follows:
Spectral Response: 355 to 390 nm
Sensitivity Factor: 1.445 x 10~5 (A) (cm?) (iH)
Ranges: 0 to 0.03, 0.1, 0.3, 1.0, 3.0, or
10.0 microamperes
Accuracy: 3% on all ranges
Linearity: 2% on all ranges
No sensor failures occurred during the field project.
MISCO 24-PORT SEQUENTIAL TOTAL PARTICULATE SAMPLER
The Misco 24-Port Sequential Total Particulate Sampler was specially
designed for use in the Clinch River project by Misco Scientific. Each
unit consisted of 24 sampling heads manifolded with electrically operated
solenoids to one blower. A contact closure automatically started a 1-
hour cycle on one sampling head, with a possible total of 24 hours of
sampling utilizing all 24 heads. The units were equipped with a 24-hour
circular recorder that indicated the flow for each sampling head.
ENVIRONMENTAL DATA CORPORATION DIRECT-VIEWING INDUSTRIAL GAS
ANALYZER (DIGA)
The EDC DIGA-Series in-stack monitor operates on the principle of
selective absorption of UV light by S02 and NO gas molecules. With a
UV light source on one side of the flue gas duct and the sensor-receiver
on the other side, the instrument gives an integrated value of the gas
concentration across the path length.
143
-------
The manufacturer's stated specifications for the EDC DIGA-Series
are as follows:
Range
NO: 0 to 1000 ppm
503: 0 to 1000 ppm
Accuracy: 2% of full scale
Repeatability: 1% of full scale
Response Time: 4 seconds
A few component failures resulted in some loss of data from the EDC
monitor. Unreliable data were sometimes obtained when the flue gas
opacity became very high (usually when the plant personnel were blowing
out slag from the boilers).
144
-------
APPENDIX C
DOCUMENTATION FOR CLINCH RIVER
PLUME STUDY DATA TAPE
DATA OVERVIEW
Data collected, processed and archived during the Clinch River Plume
Study in Complex Terrain covers the period from June, 1976 through September,
1977 and is presented in the six following data types.
1. Fixed Station Data
This consists of hourly averaged aerometric data from each of the
eight fixed monitoring stations linked via leased telephone lines
to the ADAM (Aerometric Data Acquisition and Monitoring) recording
system, running on a Data General Nova computer, installed at the
Clinch River Power Plant site. There are eight fixed station data
files, one for each site.
2. Mobile Ground Monitoring Data
These data consist of aerometric measurements collected at
different locations and times throughout the study period by
two specially equipped mobile vans. The measurements supplement
the Fixed Station data by providing measurements below the plume's
ascertained direction of travel.
3. Airborne Monitoring Data
These data were collected during two periods of intensive plume
study, the first in November of 1976 and the second in July of
1977, using a helicopter equipped for aerometric sampling. The
data were recorded approximately four to five times a second
while making a series of plume traverses at different heights
between two given ground based checkpoint locations. An
approximate ground speed of 60 mph was maintained giving a
sampling interval of approximately 20 ft. The reduced data
gives a basis for investigating plume structure and behavior
aloft by such techniques as construction of plume cross
sections.
4. Sulfate Measurements
Sulfate measurements were collected at six fixed stations and
at some locations of the Mobile 1 van. The data were collected
at selected times during the study.
145
-------
5. Plant Operating Data
This consists of recorded hourly operating data for each of
the three boiler units at the plant, collected in the plant
control room together with monitored emissions of S02 and NO
measured by a stack gas analyzer installed in the stack servic-
ing boiler unit #3. Theoretical and estimated hourly emissions
for all units are also included.
6. Upper Air Meteorological Data
This consists of pibal and T-sonde data giving wind directions
and temperature aloft. The data were collected at regular
times and intensified during periods of particular interest
such as the helicopter sampling periods. The upper air data
is augmented by ground level readings of wet and dry bulb
temperatures and relative humidity taken at the pibal launch
site just prior to releases.
Data Processing Information
The six different types of data are maintained in separate files on one
2400 foot magnetic tape. The tape is a 9-track nonlabelled ASCII tape recorded
at 1600 bpi. The records are blocked but they do not span blocks. The original
data files were produced on a PDF 11/70 IAS system. The data files as they
appear on the tape are as follows:
File Number File Description Record Length Block Size
1 Castlewood fixed-station data 176 3520
2 Hockey fixed-station data 176 3520
3 Johnson fixed-station data- 176 3520
4 Lambert fixed-station data 176 3520
5 Kent's Ridge fixed-station data 176 3520
6 Nash's Ford fixed-station data 176 3520
7 Munsey fixed-station data 176 3520
8 Tower fixed-station data " 176 3520
9 Sulfate data 44 3520
10 Upper-air meteorological data 32 3520
11 Mobile 1 - June 1976-August 1976 76 3040
12 Mobile 1 - September 1976-November 1976 76 3040
13 Mobile 1 - December 1976 - February 1977 76 3040
14 Mobile 1 - March 1977 - May 1977 76 3040
15 Mobile 1 - June 1977 - September 1977 76 3040
16 Mobile 1 - from strip charts for Julian
days 222, 245, 246, 247, 251, 252, 257,
258, 260, 261, 264, 265, 266, 267, 268,
273, 302, 307, 308, of 1976 and days
199-204 of 1977 76 3040
146
-------
File Number File Description Record Length Block Size
17 Mobile 2 - from strip charts for
Julian days 288, 363, 365 of 1976 and days
days 4-7, 11-13, 25, 27, 31-35, 41, 45-48,
52-56, 59-62, 66-70, 74 of 1977 76 3040
18 Helicopter data for November 8, 9, 11, 13,
17, 1976 88 3520
19 Helicopter data for November 10, 1976
Events 9-36 88 3520
20 Helicopter data for July 20, 1977
Events 1-39 88 3520
21 Helicopter data for July 21, 1977
Events 41-45 88 3520
22 Helicopter data for July 23, 1977
Events 47-53 88 3520
23 Helicopter data for July 23, 1977
Events 416-428 88 3520
24 Helicopter data for July 23, 1977
Events 431-475 88 3520
25 Helicopter data for July 24, 1977
Events 501-543 88 3520
26 Helicopter data for July 24, 1977
Events 559-648 88 3520
27 Helicopter data for July 24, 1977
Events 652-662 88 3520
28 Helicopter data for July 26, 1977
Events 749-788, 797-803 88 3520
29 Helicopter data for July 27, 1977
Events 805-861 88 3520
30 Helicopter data for July 27, 1977
Events 866-914 88 3520
31 Helicopter data for July 27, 1977
Events 916-968 88 3520
32 Helicopter data for July 28, 1977
Events 994-077 88 3520
33 Helicopter data for July, 1977
Events 408-412, 545-557, 564, 610, 650,
701-747, 791-795, 970-992 88 3520
34 Plant operating data 76 3040
Fixed Station Data
The fixed station data are hourly averages maintained as separate files
for each of eight stations. Accompanying the hourly average pollutant data
are also the peak concentrations recorded during the hour ending at the given
time. The station names are abbreviated to station codes six characters
or less as TOWER, MUNSEY, NASHS, HOCKEY, CASTLE, KENTS, JOHNSN, and LAMBER.
The table on the following pages describes the parameters and the formats
contained on the fixed station hourly files.
147
-------
I. am her
c
o
!
«
t?
c
U
V
u
J3
Z
u
8
2
5
M
3
5
t
c
1
c
3
C
O
t
s
Q
JD
>
=fc
c
c
u
XX XXX
XX XXX
XX XX XX XX X X X X
XX XX XX XX X X X X
XX XX XX XX X X X X
XX XX XX XX X X X X
XX XX XX XX X X X X
XX XX XX XX X X X X X X X X
&, &* CftCTi O"iC^ CTiCTi C"i CTi rt.' rft fl^ n\ n\ &
M
*c "^
» S£E£E=EE« S
1-4 ooooboobE Uf3
1 >>>>>>>>a-sss«s s
w S <<** cjX^J WJ3£ 0JiZ: 4,^X1 Vy w^ Og ajJJ 0.£ 4,01 „« w g
o « *7 i Ocu&. °CUCL OD-D- ocuo- o*| o iv, N LO i/) f^inm ^t/icrj f^irtt/i f^i/i N in f^i/i ^in f^i/5 **^i/i WLD MI/!
ocm ooom Otsir^ co-* 01-^ vCoo wuTi 5?^ r^-os S*J5 *-ro
bC DC
^H U- U,
£^ ec ai ? " ^ ^
1 ? a xl 1 J 5- I II
DC ^ ^* 'O"O*J*JPcS^*rt ^j'fi'bO Oii
4) ^ >. ^. u t^w'5H<:<£iao^it=c^s
"o ^^^CM ^x ^ro .- l'S'B'aEii'ii:S?»c:E5«'B'S
^ QV- ^Sg'^Sg ^§8 i§8 £S SS ^^ H^ ^^ >> ••S^ gg
1 1 ^ I 8°l 1 § 1 " § 1 1 § 1 I 22 22 22 22 ^§ 11 II 22
< < < <
l-l 1-1 ,-H
rt
5
c
^O
g 5
1 1
c *• *"
'S U ™ "
"2. =3 « 5
.£ .S •£ £
g £ o E
'g -g " t!
•§ « c "
S -° « 0
< < w H
* » *
148
-------
i
s
E
i
V
&
o
V
V
s
s
1
3
1
O
f-
|
1
a
•a
t.
§
e
g
0
Variable Descriptio
tiannel ft
U
X X
en gj
S &!
U
S £
•s & -8 &
U D U C
M in N in
< u, < u.
Oi rf '•S ••-«
*-t N N M
4)
H
10 M Sigma Elevation Flag
10 M Sigma Elevation***
Lower Temperature Differe
Flag
0. 5 to 4 M Temperature
Difference t
ro it
-.
X X
X X
a a
ffl ^
£
0
o
1/5
c s
w fc fti £
-5 ** "0 «
O ^J O ft)
U S U D
(M »4
CJ i/C N i/i
< U, < U-
ro 00 Q Ul
r i II
PJ ^ Oi *-«
30 M Wind Speed Flag
30 M Wind Speed
30 M Wind Direction Flag
30 M Wind Direction
LO *O
" -
XX X
as ?
pi ^ c.
rj
S S S
u S «> S « S
"O DC *D & *B M
Of 0 ft) O O
O O U Q O C
« in N \T) N in
< u. < u- < u.
^f in t?) t/> ic *c
(iii i i
\D oo m irt O M
-^ ^ if. i/i i£ \o
u
30 M Elevadon Angle Flag
30 M Elevation Angle
30 M Sigma Elevation Flag
30 M Sigma Elevation ***
Upper Temperature Differe
Flag
0. 5 to 30 M temperature
D if fere nee 1
N- 00 O\
^* TH 1-(
X
ffi
„.
s
•8 fe |
U C cc
CM ^:
<: u.
(Oh- l~
h~ C V
30 M Sigma Azimuth Flag
30 M Sigma Azimuth***
Filler Characters
o
ro
O f
III
HI
T3 a e z
B- ^ 2 S js
I « i a
II « S
- 3 |sa
< < E 5
* * * +•
t
149
-------
The definitions of the data validity flags are as follows:
W - Measurement is based on less than 22 scan values within
the hour.
M - N02 calibration gas used to adjust instrument was not
properly controlled. Measurement shows changes relative
to other measurements within a few days, but may not be
quantitatively correct.
S - Derived from strip chart.
A - Arithmetic mean rather than vector mean of 2-minute scan
values because wind speed or direction is missing.
L - Wind direction and speed is not reliable because wind
speed is near or below the sensing threshold of the sensor.
E - Estimated zero applied to NOX, NO data. Data is only
qualitatively useful.
U - Data is uncalibrated (no zero/span applied). Data is
only qualitatively useful.
C - Incorrect zero/span values have been applied to data.
Data is only qualitatively useful.
N - NO? not computed since NOX and NO are uncalibrated or
calibrated incorrectly.
K - Kent's Ridge wind direction data unreliable.
The fixed station data files cover the following periods:
TOWER
HOCKEY
MUNSEY
NASHS
KENTS
CASTLE
JOHNSN
LAMBER
Year
76
76
76
76
76
76
76
76
Beginning
Julian Day
153
155
153
165
258
176
320
301
Hour
10
11
10
01
15
14
19
12
Year
77
77
77
77
77
77
77
77
Ending
Julian Day
273
273
273
273
273
273
273
273
Hour
24
24
24
24
24
24
24
24
150
-------
Sulfate Data
Sulfate measurements consist of data for sulfates collected on filters
at six fixed stations, CASTLEWOOD, MUNSEY, HOCKEY, KENTS, NASHS and JOHNSON
and at given locations of the MOBILE 1 van, abbreviated to two characters
as CA, MU, HO, KE, NA, JO and MO respectively.
The net sulfate determined by analysis of the filters together with the
volume of air that the filter was exposed to in the one-hour period starting
at the given time was used in calculating a sulfate concentration. The
record format is as follows:
Variable
ST
DATE
TIKE
FILTER #
NETS04
AIRFLOW
S04CONC
ELEV
DIST
BEAR
Position
1-2
3-8
9-12
13-16
17-20
21-25
26-30
for MOBILE 1
31-34
35-38
39-42
43-44
Format
A2
16
14
14
14
F5.1
F5.1
Description
First 2 characters of station name.
CA, MU, HO, KE, NA, JO, MO (Mobile 1)
Date in Year (YY), Month (MM) and
Day (DD) (i.e., YYMMDD)
Time at start of sample in EST
Filter number
Net sulfate (yg)
Airflow (cubic meters)
Sulfate concentration (ug/cu m)
(MO) location is also recorded (as follows):
14
F4.1
14
Elevation above sea level (ft)
Distance from plant (km)
Bearing from plant (deg)
Blank
Upper Air Meteorological Data
These data consist of vertical profiles of wind direction and speed,
some with temperature, based on single- and double-theodolite pibal mea-
surements, some with temperature-sondes. Ground level readings of wet and
dry bulb temperatures and relative humidity taken with some of the releases
are also included. Releases are identified by Gregorian date and time in
EST.
At each height there may be wind speed and wind direction or temperature
or all three. Minute refers to the elapsed time in minutes since the pibal/
t-sonde release and is used to interpolate heights to map significant
temperature structures as well as to compute wind speeds and wind directions.
151
-------
The wind speed and direction are actually averages over the layer
extending from the height one minute earlier to the height at which they
are recorded. Therefore, the speed and direction should be considered
applicable to a height midway between the height one minute earlier and
the height at which they are recorded.
Example:
Minute
0. 0
1.0
2.0
Height
0
160
320
Wind Speed
Missing
2
4
Wind Direction
Missing
275
292
Temperature
5.7
5.0
3.8
In the above example the wind speed of 2 m/sec and direction of 275° should
be applicable to a height of 80 m and the wind speed of 4 m/sec and direc-
tion of 292° should apply to 240 m. The temperatures, however, apply to
the heights given on the records.
Data relating to a single release is recorded on the file in records
of three types as follows:
1. HEADER
2. OBSERVATION
1 release
3. END
Header
OBS1
OBS2
--
END
The format of each of the three different types of records within a
release follows.
1. HEADER RECORD
This is always the first record of each sounding and contains data
relating to the release as a whole. The record format is:
Variable
DATE
TIME
7-10
TYPE
BASE
HUMIDITY
DRYBULB
WETBULB
11-11
12-13
14-16
17-21
22-26
27-32
14
II
A2
13
F5.1
F5. 1
Description
Gregorian date in YYMMDD form (i.e., YY .- Year,
MM = Month, DD = Day)
Time of release in EST in form of HHMM (i. e.,
HH = Hour, MM = Minute)
A "1" identifies this record as a header record
Baseline
Relative humidity (%)
Dry bulb temp (°C)
Wet bulb temp(OG)
Blank
152
-------
Baselines are AB, BC, AC for double theodolite reductions and SA, SB
for single theodolite releases from release points A and B respectively.
Before October 19, 1976, releases were also made using a baseline near the
Tower site. These are designated as baseline TW. Single theodolite soundings
from the Tower site can be identified by the constant rate of rise of the
balloon after the second minute.
2. OBSERVATION RECORD
This record contains data for the heights within the sounding
designated by the previous HEADER record. Data are given successively
by increasing height.
The record format is:
Variable Position Format Description
DATE 1 -6 16 Gregorian date in YYMMDD form
TIME 7-10 14 Time of release in EST
TYPE 11-11 II A "2" identifies this record as an observation record
MINUTE 12-15 F4. 1 Minutes since release
HEIGHT 16-19 14 Height of observation (meters) (see example, p. 152)
SPEED 20-22 13 Wind speed for layer from this height to previous
height (m/sec)
DIRECTION 23-25 13 Wind direction for layer from this height to
previous height (degrees)
TEMP 26-30 F5.1 Temperature at this height (°C)
31-32 Blanks
Missing values are as follows:
SPEED
DIRECTION
TEMP
-99
-99
-99.9
3. END RECORD
This record indicates the end of a release and is followed by another
HEADER record or an end of file. The record format is:
Variable
DATE
TIME
TYPE
Position
1-6
7-10
11-11
12-32
Format
16
14
II
Description
Gregorian date in YYMMDD form
Time of release in EST
A "3" identifies this record as an END record
Blank
153
-------
Ground Mobile Monitoring Data
The ground mobile monitoring data taken during the Clinch River field
study are contained in files 11 through 17 of the tape.
Ground mobile monitoring data consists of data collected at different
locations by the two mobile ground monitoring vans, designated Mobile 1
and Mobile 2. The majority of the data was originally recorded every 30
seconds on cassettes in the mobile units and then transcribed to 9 track
ASCII reels. Each location sampled was identified by an event marker and
time which were also recorded on the cassette. The azimuth, distance and
elevation of each location sampled was then abstracted from the operating
log and merged with the data on tape using the time and event flag to
match records. The data were also recorded on strip charts, and for the
periods when the cassette recorders were not operating, the data was trans-
cribed from the strip charts, keypunched and processed into files of the
same format as the tape data. The data from strip charts is in the form
of average and peak pollutant concentrations for each event, contained on
separate, consecutive records.
Data recorded were S02 for Mobile 2 and S02, NOX, N02, NO, 03, wind
speed, wind direction and temperature for Mobile 1. Each channel's data
were converted to engineering units using the appropriate scale for each
channel and zero span calibrations applied using values taken before each
run. The calibrated data was recorded on a file in the following format.
Variable
UNIT
YEAR
JDAY
TIME
SEC
FLAG
EVENT
EASTING
NORTHING
ELEV
S02
NOX
NO 2
NO
03
WS
WD
TEMP
Position
2-2
4-5
6-8
10-13
14-15
16-16
17-19
20-27
28-35
36-40
44-47
48-51
52-55
56-59
60-63
64-67
68-71
72-75
76
Format Description
II Mobile unit # (1 or 2)
12 Last 2 digits of year (76 or 77)
13 Julian day
14 Time of scan (Eastern Standard Time)
12 Second of scan
Al A or P (indicating average or peak for
the event)
13 Event marker
F8. 2 Easting distance from plant (decameters)
F8. 2 Northing distance from plant (decameters)
15 Elevation (ft above sea level) at the start
of event
14 SO2 concentration (ppb)
14 NOX concentration (ppb)
14 NO2 concentration (ppb)
14 NO concentration (ppb)
14 Og concentration (ppb)
14 Wind speed (mph)
14 Wind direction (degrees)
14 Temperature (°C)
Blank
154
-------
Missing value indicators of -999 are used for all pollutants and meteor-
ological data, except calm winds are indicated by WS=0 and WD=90.
Airborne Monitoring Data
Measurements of the plume taken by instrumentation aboard a helicopter
during two sampling periods (November 1976 and July 1977) are contained
in files 18 through 33 on the tape.
Helicopter data from the two airborne monitoring periods were recorded
approximately 4 to 5 times per second on cassette tape while making a series
of plume traverses at different heights between two given ground based check-
point locations. A ground speed of approximately 60 mph was maintained
giving a sampling interval of approximately 20 ft. The pollutant data were
also recorded on stip charts. For periods during which the cassette recorder
was not operating the strip chart data were transcribed, keypunched and
processed into files of the same format as the tape data. The significant
points (maxima and minima) of the chart trace were coded for each helicopter
traverse. Since the maxima and minima occurred at slightly different times
for each pollutant, each record was filled in with values for all pollutants
by linearly interpolating over time between the successive maxima and minima.
The calibrated data was recorded on a file in the following format:
Variable Position Format Description
UNIT 2-2 II Mobile unit # (3)
YEAR 4-5 12 Last 2 digits of year (76 or 77)
JDAY 6-8 13 Julian day
TIME 10-13 14 Time of scan (Eastern Standard Time)
SEC 14-15 12 Second of scan
EVENT 17-19 A3 Event marker
EASTING 20-27 F8. 2 Easting distance from plant (decameters)
NORTHING 28-35 F8. 2 Northing distance from plant (decameters)
ELEV 36-40 15 Elevation (ft above sea level) of
helicopter
SC>2 44-47 14 SC>2 concentration (ppb)
NOX 48-51 14 NOX concentration (ppb)
NO 2 52-55 14 NO 2 concentration (ppb)
NO 56-59 14 NO concentration (ppb)
03 60-63 14 03 concentration (ppb)
RH 64-67 14 Relative humidity (%)
68-83 Blank
TEMP 84-87 14 Temperature (°C)
88 Blank
All event markers are designated by numbers except on November 8, 1976,
some events are designated by numbers and the letter "A". Missing value
indicators of -999 are used for the pollutants and relative humidity data.
Any negative number can be considered missing for the temperature data.
155
-------
Records containing "333" in the event field are used to separate traverses
on files 18-32. The airborne monitoring data in these files are uncorrected
for instrument response time and the time lag involved in the gas reaching
the instruments from the base of the helicopter. These adjustments are
considered in the data analysis phase of the project.
Plant Operating Data
The plant operating data for the coal-fired Clinch River Power Plant
is contained in one file on the tape. This data covers the period
June 1, 1976 through September 30, 1977. The data consists of weekly
average coal heat content and heat rate for each of the three units, a
daily coal sulfur content (actually a 7-d running average of sulfur con-
tent of coal deliveries to the plant), and hourly S02 and NO emission
rates, stack temperatures, exhaust airflow rates, and generator loads for
each unit. Also included are hourly plant control room readings of baro-
metric pressure. The S02 and NO emissions data consist of monitored emis-
sion rates for Unit #3 for a significant portion of the study, along with
emissions for all three units calculated by two separate methods. Emis-
sion estimates were made using EPA emission factors (labelled "theoretical")
and also through the use of regression equations (labelled "estimated")
derived from hour by hour comparisons of generator load, sulfur content
and measured emissions for Unit #3.
The file is structured as follows. The data for one day is contained
in 25 consecutive records. The first record which is the daily header
record contains the date, sulfur content, heat content and heat rate.
This is followed by 24 hourly records, each containing emission rates,
stack temperatures, exhaust airflow rates, and generator loads. The file
format is presented on the following page. Missing value indicators of
999 are used for all the hourly plant data.
Fourteen days of suspicious S02 monitoring data exist on the plant
data file. Fifty-one days of NO monitoring data are also suspicious.
These data are qualitatively useful despite the stack monitor being out
of calibration. The dates of the suspicious data are contained in Section
5.
156
-------
PLANT OPERATING CHARACTERISTIC RECORD FORMAT FOR MAGNETIC TAPE
Record
Position
Character
Format
Daily Header Record:
Year 1-2
Month of Year 3-4
Day of Month 5-6
Sulfur Content of Fuel 7-10
Fuel Heat Content, Unit #1 11-15
Heat Rate, Unit#l 16-20
Fuel Heat Content, Unit #2 21-25
He at Rate, Unit #2 26-30
Fuel Heat Content, Unit #3 31 -35
Heat Rate, Unit #3 36-40
41-76
Hourly Records
Stack Temperature, Unit #1 1-3
Air Exhaust Rate, Unit#l 4-6
Net Generator Load, Unit #1 7-9
Theoretical SO2 Emission Rate, #1 10-12
Estimated SO2 Emission R ate, # 1 13-15
Theoretical NO Emission Rate, #1 16-18
Estimated NO Emission Rate, #1 19-21
Stack Temperature, Unit #2 22-24
Air Exhaust Rate, Unit #2 25-27
Net Generator Load, Unit #2 28-30
Theoretical SO2 Emission Rate, #2 31-33
Estimated SO2 Emission Rate, #2 34-36
Theoretical NO Emission Rate, #2 37-39
Estimated NO Emission Rate, #2 40-42
Stack Temperature, Unit #3 43-45
A ir Exhaust R ate, Unit #3 46 -48
Net Generator Load, Unit #3 49-51
Theoretical SO2 Emission Rate, #3 52-54
Measured SO2 Emission Rate, #3 55-57
Estimated SO2 Emission Rate, #3 58-60
Theoretical NO Emission Rate, #3 61-63
Measured NO Emission Rate, #3 64-66
Estimated NO Emission Rate, #3 67-69
Barometric Pressure at Plant 70-74
75-76
12
12
12
F4. 2
15
15
15
15
15
IS
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
F5. 2
76 or 77
1 to 12
1 to 31
Percent
BTU per pound
BTU per kilowatt-hour
BTU per pound
BTU per kilowatt-hour
BTU per pound
BTU per kilowatt-hour
Blank
Degrees Kelvin
Cubic mete is per second
Megawatts
Grams per second
Grams per second
Grams per second
Grams per second
Degrees Kelvin
Cubic meters per second
Megawatts
Grams per second
Grams per second
Grams per second
Grams per second
Degrees Kelvin
Cubic meters per second
Megawatts
Grams per second
Grams per second
Grams per second
Grams per second
Grams per second
Grams per second
In. of mercury
Blank
157
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
i. REPORT NO.
EPA-600/7-79-0103
3. RECIPIENT'S.ACCESSIOI*NO.
4. TITLE AND SUBTITLE
5. REPORT DATE
POWER PLANT STACK PLUMES IN COMPLEX TERRAIN.
Description of an Aerometric Field Study
January 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Robert C. Koch, W. Gale Biggs, Douglas Cover, Harry
Rector. Paul F. Stenberg. and Kenneth E. Pickering
EF-678 (5125)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
GEOMET, Incorporated
15 Firstfield Road
Gaithersburg, MD 20760
10. PROGRAM ELEMENT NO.
INE625 EA-20 (FY-78)
11. CONTRACT/GRANT NO.
68-02-2260
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Environmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park. NC 27711
TYPE OF REPORT AND PERIO
Interim 4/76-10/78
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
REFERENCE: POWER PLANT STACK PLUMES IN COMPLEX TERRAIN.
Data Collected During An Aerometric Field Study. EPA-600/7-79-010b
16. ABSTRACT
A monitoring network consisting of eight fixed sites was set up and operated in
the vicinity of the 712-megawatt, coal-fired, Clinch River Virginia Steam Plant
during the period of June 1, 1976 to September 30, 1977. At most of the monitoring
stations sulfur dioxide, nitric oxide, total nitrogen oxides, wind direction, wind
speed, temperature, and sulfates were measured. Two of the stations had 30-meter
towers from which meteorological variables were measured. On four or five days per
week a mobile van was operated to supplement the fixed site monitoring, and once or
twice a day balloons were used to measure winds and temperatures aloft in the vicinity
of the plant site. A monitor for continuous measurement of sulfur dioxide and nitric
oxide was operated in one of the two plant stacks. These data were supplemented by
helicopter measurements taken during ten-day periods in November 1976 and July 1977.
The system used to collect the data, the operational procedures used to run
the system, and its performance record are described in this report. The data that
were collected have been edited and recorded on magnetic tape and are available
from the National Technical Information Service. An appendix in this report describes
the formats required for reading this tape.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
*Air pollution
*Terrain
*Field tests
*Sulfur dioxide
*Plumes
*Nitrogen oxides
*Electric power plants
*Atmospheric diffusion
*Sulfates
*Meteorological data
13B
08F
14B
07B
21B
10B
04A
04B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
168
20. SECURITY CLASS (This page)
MTY CLASS (Thispag
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
158
-------
Z TJ
m
m
I
01
O
O
O
QJ
8-
Q
<
CD
ft)
1
CD
^i
Co
1 -
1 o
^
_t/)
O
»i
O
^
•"•»
^
§
!>l
D
i-*
CD
0)
->
Q
-•*
~-S
oT
o
CD
Q)
^
Q.
3
C
3
0
3-
CD
C
§•
§
r~+
8-
CO
^t
CD
0
0
o
a
3
c
CD
^i
CD
o
CD
5'
I'
3-
ffc
Co
fb
cS"
3-
a
o'
Q)
>••*
CD
OJ
>
o
55
s
Q.
^
CD
?
^.
3
0
3-
CD
Qj
&•
O
<
CD
Cu
1
CD
CO
CO
--K
x;
Q
C
Q)
1
CD
Co
CO
CO
5
o
Q
-*
>
CD
0
-Q
CD"
CB
CO
CD
o
Q)
eg
^D
o
a
3-
CD
8-
o
<
CD
oT
Cr
CD
D
z
c >
> H
0
1 <• T,
O Ti Tl
-o O o
T) -% -
OT, ^
**a
c < c
z > c/)
m
5 0)
^ m
TJ •
r (fi
O co
< °
m O
7J
m
0)
0)
5
§ 0) X
« S
"2
CO QJ
CD ^
Q3 O
O
^
o'
s
0°
=r Q3
E Q.
O
o m
a> O
-
o o 0
CD
m
z
o
C
a)
m
Z
O O
I H
1 s;
> >
f~ Z
TJ 0
3 -n
O m
H m
m en
3 J
0 R
------- |