EPA-600/4-77-009
February 1977
Environmental Monitory Series
DIURNAL VARIATIONS IN CARBON MONOXIDE
CONCENTRATIONS, TRAFFIC COUNTS AND
METEOROLOGY
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park,. North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/4-77-U09
February 1977
DIURNAL VARIATIONS IN CARBON MONOXIDE
CONCENTRATIONS, TRAFFIC COUNTS AND METEOROLOGY
Gerard A. DeMarrais
Meteorology and Assessment Division
Environmental Sciences Research Laboratory
Research Triangle Park, N.C. 27/11
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, N.C. 2/711
-------
DISCLAIMER
This report has been reviewed by the Office of Research and Developr
ment, U.S. Environmental Protection Agency, and approved for publication.
Mention of trade names or commercial products does not constitute endorse-
ment or recommendation for use.
AUTHOR'S.AFFILIATION
The author is on assignment with the U.S. Environmental Protection
Agency from the National Oceanic and Atmospheric Administration,
U.S. Department of Commerce.
ii
-------
ABSTRACT
Although pollutant emission patterns play important roles, they cannot
adequately explain the diurnal variations in carbon monoxide concentrations
found in urban areas. In this study, hourly data from a large network of
carbon monoxide monitoring stations, with instrumentation corrected for
moisture interference, are analyzed and compared with traffic flow and
meteorological conditions at several locations in Maryland. The meteor-
ological phenomena that appear to be important in explaining the diurnal
variations involve the ventilation effects resulting from variable wind
speeds and mixing heights.
iii
-------
CONTENTS
Abstract iii
Figures vi
Tables vi
1. Introduction 1
2. Conclusions and Kecommendations , . . 2
3. Data and Method 3
4. Results 5
5. Summary 12
References 13
Appendix
Site description of carbon monoxide monitoring stations . 15
-------
FIGURES
Number Page
1 Locations of CO monitoring stations ............... 17
2 Diurnal variation of CO concentrations by month, Maryland
stations, 1973 ........................ 18-20
3 Diurnal variation of occurrence of selected wind speed ranges
Baltimore, 1951-1960 ..................... 21
4 Diurnal variations of mixing heights, frequency of wind speeds
greater than 12 mph, traffic flow and CO concentrations. ... 22
5 Comparison of diurnal variations in CO concentrations (January). 23
6 Comparison of diurnal variations in CO concentrations (October). 24
TABLES
Number Page
1 Annual average hourly percentage of total vehicular traffic at
four Maryland highway stations 25
2 Calculated urban monthly average mixing heights and ventilation
factors for Washington, DC, area, 1973 26
3 Calculated urban mixing height and ventilation factors for tne
Washington, DC, area for January 1973 27
4 Calculated urban mixing height and ventilation factors for the
Washington, DC, area for October 1973 28
vi
-------
INTRODUCTION
State and local pollution control agencies in complying with federal
regulations (1), gather air quality data at a tremendous rate. The two main
purposes for collecting these data are to determine whether air quality
standards are being exceeded and to provide baseline check points for
assessing changes in air quality. Little use has been made of these data,
which have been gathered for many years, as a resource for determining the
effects of variations in meteorology on the observed air quality.
In this report of a study of data for January through November 197.3,
hourly carbon monoxide (CO) measurements from the air quality network
operated in the State of Maryland are evaluated and compared with specific
meteorological variables. The major sources of the CO emissions in the
area are motor vehicles (2); traffic counts, therefore, are considered.
Major emphasis is given to the diurnal variation of CO concentrations in
each month, and in two cases, to the differences in CO concentrations between
a relatively unpolluted workweek followed by a polluted workweek.
-------
CONCLUSIONS AND RECOMMENDATIONS
1. The average diurnal variation of CO concentrations at the nine Mary-
land stations has a bimodal distribution with peaks in the morning and .
late afternoon.
2. The diurnal variation of CO concentrations at the nine stations is not
well correlated with the local diurnal variation in traffic density. Further
studies should be undertaken to determine whether this Maryland finding is
typical of other locations.
3. The diurnal variations of wind speedi mixing height, and traffic density
explain about 50 percent of the variance in CO concentrations. More detailed
studies, with the traffic count and air quality stations in closer proximity
should be undertaken to determine whether a greater amount of the variance
can be explained by the traffic, wind speed and mixing height.
4. The diurnal variation of CO concentrations consistently shows marked
peaks that persist for 4 hours or less. New studies should be made to
determine whether averaging CO concentrations for 8 hour periods tends to
minimize a CO problem.
5. A curbside station and one remote from traffic but in a metropolitan
area showed very little difference in diurnal variations in CO concentrations.
This indicates that CO is not a very localized problem but readily covers
the nearby area.
-------
DATA AND METHOD
CO INSTRUMENTATION AND DATA
Carbon monoxide concentrations are determined by the nondispersive
infrared absorption method used in the Maryland air quality network. The
instrumentation (3) includes an infrared source, sample and reference cells,
detectors in each cell, a control unit and amplifier, and a recorder.
Because water vapor also absorbs infrared radiation, a standard drying
technique is used to eliminate potential interference. The reference and
sample gas are brought to 60°F and 15 percent relative humidity before being
brought into the cells.
The CO network in Maryland consisted of 19 stations in 1973. A few
stations, however, did not have the drying mechanisms operating properly
early in the year. The data analyzed here are from the file of the National
Air Data Branch (NADB) for the U. S. Environmental Protection Agency, which
maintains pollutant data from throughout the nation. (No December CO data
were available.). The NADB hourly average values are listed at the starting
time of each hour. An individual station record for either a month or work-
week was not used unless each hourly period had at least 60 percent of the
possible measurements; monthly data were not used unless at least 6 months
of information were available. The names and locations of the 12 stations
from which data were used (five in the Baltimore area, five in the
Washington, DC area, and two in the northwest part of the state) are shown
in Figure 1. A brief description of each station can be found in the appen-
dix.
METEOROLOGICAL DATA
Analysis of the monthly CO data for the individual sites showed signi-
ficant diurnal variations in concentrations as well as month-to-month and
station-to-station variations. These variations are compared with varia-
tions in wind speed for Baltimore later in this report. Mixing height
variations can also be important in evaluating air pollution problems (4,5),
and the available mixing height information (6) for the Washington-Baltimore
area Include calculated morning and afternoon mixing heights and ventilation
factors (depth of the mixing layer multiplied by the average speed in the
layer). These mixing heights are not to be considered absolute numbers but
only relative Indicators of vertical mixing. This condition 1s particularly
true 1n the early morning when, assuming an urban hear island of 9°F, the
calculated mixing height is low (for example, 100 meters or less). Such a
low height indicates the likelihood of a high degree of thermal stability
near the ground 1n urban as well as rural areas.
-------
TRAFFIC PATTERNS
Concurrent data on the diurnal variation of traffic flow at the indivi-
dual pollutant monitoring sites are not available. The Bureau of Traffic
Engineering of Maryland (11) operates 39 continuous-count stations; four
stations were selected as being representative of the areas of concern in
this study. The diurnal variation in 1973 traffic flow, in percentages,
is shown in Table 1. The three counting stations on Interstate-695 (the
Beltway around Baltimore) are representative of the ten CO sampling sites
in the Washington and Baltimore areas. (The counting station west of the
Baltimore-Harrisburg Expressway is only 2 miles west of the CO monitoring
station at Towson and the counting station south of U. S. 1 is 4 miles
northwest of Linthicum). The fourth station shows the traffic pattern in
a Hagerstown suburb and is representative of the CO sampling sites in
Hagerstown and Cumberland.
The traffic data in Table 1 show that the lowest hourly traffic flow,
generally 1 percent or less of the 24-hour flow, occurs from 2 to 5 AM.
Then the traffic increases and attains a peak of about 6.6 percent between
7 and 8 AM. The traffic remains around 5 percent per hour from 9 AM to
3 PM and then increases to a plateau-like peak for about 3 hours. This
afternoon peak is higher than the morning peak. During the early evening
and nighttime, the traffic-flow gradually decreases reaching its lowest
point in the early morning hours. The traffic pattern at the Hagerstown
station differs from the others in that the morning peak is not as high
and the traffic in the evening is a little greater.
-------
RESULTS
DIURNAL VARIATIONS OF CO
Data for nine of the stations shown in Figure 1 met the criteria discussed
earlier for this study. Figure 2 shows time from midnight to midnight as the
abscissa and concentrations in parts per million (ppm, by volume; 1 ppm =
1.146 milligrams per cubic meter, under standard conditions) as the ordinate
for these stations. Average hourly concentrations are plotted as points at
the starting time of the observation. Examination of these data reveals a
more complicated relationship among factors causing diurnal pollutant vari-
ations than the observation (12) that peak concentrations coincide with peak
traffic flows; the correlation between the diurnal variation in CO concen-
trations at Linthicum and the traffic at the 1-695 station south of U. S. 40
is 0.39 for January and only 0.05 for July. The sunrise and sunset marks
in Figure 2 reveal a tendency for the peak concentrations to correlate with
the times of these phenomena. Another interesting feature of the results is
that Bethesda, a station located about one-quarter mile from heavy traffic,
shows relative concentration peaks similar to those of the curbside station
at Cumberland (see station description in Appendix). These detectable peaks
observed away from the sources suggest that CO rapidly spreads over the
urban area.
The patterns of the diurnal concentration curves for the individual
stations over the study period vary from little month-to-month consistency
at Essex (Figure 2) to consistent, double-humped curves with morning and
afternoon peaks at Gaithersburg, Cumberland, Hagerstown, and Hyattsville.
At a majority of the stations, however, there is good month-to-month consist-
ency. Because most stations have individual, unique characteristics, no one
station 1s representative of the area shown in Figure 1.
Another characteristic that is consistently found at a majority of
stations is the progressive shift of the morning peak to earlier times in
the first half of the year and to later times from July to November. A
similar progressive shift at a majority of stations is not seen in the
afternoon data because: (1) peaks are sometimes difficult to detect (Linthicum),
(2) double peaks occur at times (Baltimore), and (3) the month-to-month shift
of the peak is Inconsistent at some places (Linthicum, Hagerstown, and
Hyattsville). A reversal of the shift pattern is seen at three stations
(Gaithersburg, Bethesda, and Cumberland) 1n that the peaks move progressively
te later times 1n the early half of the year and to earlier times from
June through November. A fairly consistent pattern among the stations 1s
for the higher peak values to occur 1n the colder part of the year with the
extreme peak generally occuHng in the morning in January.
-------
Wind Speed and Mixing Height
The general pattern of concentration variation (Figure 2) shows a peak
in the morning followed by a rapid decrease; relatively low concentrations
in the forenoon and most of the afternoon; another high value in the after-
noon or evening; and, finally, a slow decrease in concentrations through
the night. In an attempt to explain this general pattern, the first phenom-
enon sought was one that would allow for diluting the concentrations pro-
duced by the moderate-to-heavy traffic of midday to the same values as those
associated with the light traffic of early morning (see Table I). Wind speed
is an obvious phenomenon that would contribute to this result. The climatol-
ogical wind speed data for Baltimore (10)(Figure 3) show the variation of
low speeds (0-3 mi/hr) and high speeds (13-24 mi/hr and 25 mi/hr and greater)
for each midseason month. The dots in Figure 3 show the 1973 average fre-
quencies of winds in excess of 12 mi/hr (high speed), based on the three-
hourly observations currently available (13;. The 1973 diurnal variation
of high speed differs from the climatological summary but shows the typical
low frequency at night, the increasing frequency in the forenoon with a peak
around noon, and the rather sharp decrease in frequency in the late after-
noon. Thus, the wind speeds allow for a limited amount of horizontal trans-
port in the morning; then in the forenoon, they provide greater volumes of
air for dilution and downwind transport. Finally in the afternoon and evening,
there is progressively less wind for dilution and transport. The higher
frequency of strong winds in April probably contributes to the resulting
low concentrations shown in April graphs in Figure 2.
Combined with the wind speed in markedly affecting the concentrations
of pollutants is vertical mixing -- the depth (height) of the surface-based
layer through which the pollutants will be readily dispersed (4). Mixing
height data for 1973 for urban areas in the vicinity of Washington, DC (6)
are shown in Table 2. The monthly values are averages of daily values
calculated according to a standard procedure (14) for each workday. (No
weekend data are available.) Listed in the table are mixing heights,
average wind speed in the mixing layers, and ventilation factors (layer
thickness times speed) for the morning and early afternoon (that is, the
maximum mixing height, corresponding to the maximum surface temperature).
In the morning, the calculated mixing heights were usually low by factors
of 2.5 to 9, compared with the calculated maximum afternoon heights. The
high average mixing heights arid wind speeds on April mornings are probably
prime causes of the lack of marked peaks in early morning concentrations
(see Figure 2).
-------
Unfortunately, these mixing height data give only an indication of how
limited the mixing is in the morning and how extensive it is at its maximum;
the temporal change in vertical mixing is not shown. This temporal change
is directly related to thermal stability. The typical urban pattern (4)
shows that thermal instability is restricted to a relatively shallow surface-
based layer (400 meters or less over most of the nation) until a few hours
after sunrise. Shortly thereafter, as the surface temperatures continue
to Increase, the height of the top of the layer of marked instability
increases rapidly and along with the surface temperature, levels off at
around 2 to 3 PM. Thereafter, the mixing heights increase very little
until about an hour before sunset. Near sunset, as the ground cools,
vertical mixing is inhibited and the condition persists until shortly after
the following sunrise. Enhanced vertical mixing begins latest in the day
during the winter months and is a contributing factor in the extreme concen-
tration peaks (Figure 2) observed on winter mornings. Vertical mixing through
a deep layer persists for only a few daylight hours each day during the
winter months; this is probably a prime cause of high average concentrations
during the mornings of the winter months.
Apparent Dependence of Concentrations on Meteorology and Traffic
The diurnal variations of CO concentrations have been seen to depend on
a combination of the traffic flow and the dilution patterns. In January,
peak concentrations generally occur around 7 to 8 AM, the time of peak traffic
flow. In the forenoon, although the traffic decreases to a moderate flow,
the CO concentrations drop rapidly as the wind speeds increase and vertical
mixing takes place to increasingly greater heights. By the middle of the
day, the increased speeds and mixing heights are sufficient enough to lower
concentrations produced by the moderate mid-morning traffic (about 50 percent
of that during the morning peak) to levels similar to those that occur in
the pre-dawn hours (when the traffic is about 10 percent of that in the
middle of the day). At about 3 PM, the traffic volume increases, the wind
speeds decrease, and shortly thereafter, the layer of enhanced vertical
mixing is reduced. (The change from unstable to stable is marked by a
change from a deep layer with vigorous vertical mixing to a shallow layer
with suppressed mixing). The concentrations increase until they reach a
maximum at around 5 to 6 PM, the time of- peak afternoon traffic flow.
Thereafter, the concentrations decrease gradually as the ventilation is
markedly reduced and the traffic count decreases. The changes from January
to June show the peak concentrations in the morning gradually shifting to
earlier times and those of the afternoon and evening shifting to later times.
From June to November the respective shifts are in the opposite directions.
These shifts follow the times of sunrise and sunset, which in turn govern
the beginning and ending of solar-induced thermal instability and the onset
of relatively strong daytime winds.
-------
Comparison of Parameters in One Area
Simultaneous CO, mixing height, wind speed, and traffic observations for
one site for a month are not available. There are, however, representative
monthly CO, wind speed, and traffic data for the Linthicum area. The CO
data are from Linthicum (EPA data bank); the climatological wind data are
from Friendship International Airport (13), which is 2 miles to the south-
west of Linthicum; and the traffic data (11) are from an Interstate-695
location, which is 4 miles to the northwest of Linthicum. The urban mixing
height data calculated for the Washington-Baltimore area are considered
representative of the Linthicum area.
Figure 4 shows the diurnal variation of the four parameters for January
and October 1973. The curves connect points that are averages for an hour
plotted at the ending time of the hour. Certain assumptions had to be made
with regard to mixing height because only two values are given for each day.
These assumptions were: (l) the morning mixing height value was represen-
tative of the hours from sunset until an hour after sunrise, (2) the maximum
mixing height occurred at 2 PM and remained stationary until an hour before
sunset, (3) the change in mixing height from an hour after sunrise until
the maximum occurred was linear, and (4) the change from the maximum mixing
height to that which prevailed throught the evening and nighttime was
discontinuous. These assumptions are in accord with the diurnal variation
of mixing heights described earlier. (As stated earlier, the mixing height
values should be interpreted qualitatively; they are generally inversely
proportional to atmospheric stability.)
During the early morning hours in January, the traffic flow and CO con-
centration curves appear to be very similar in shape, with the peak in con-
centrations following the traffic flow peak by 1 hour. In the forenoon, the
concentrations decrease by about 70 percent, whereas the traffic flow decreases
by only 30 percent. Simultaneously with those decreases, the frequency of
high speed (>12 mi/hr) winds increases and mixing heights extend to progress-
ively greater heights; these meteorological factors combine to allow for
substantial dilution of the concentrations. During the early afternoon, the
increase in the mixing height and decrease in the frequency of high speed
winds combine to lower the concentrations only a slight amount. In the late
afternoon and early evening, both the traffic flow and CO concentrations show
marked increases with the start of a prolonged plateau-like peak in concen-
trations, which follows the traffic flow peak by 2 hours. The delay in the
peaking of concentrations occurs because the traffic flow increases to its
maximum at the same time that the mixing height and winds are allowing con-
siderable dilution of the emissions; the peaking in concentrations occurs
after the mixing heights and wind speeds decrease to nighttime low values.
During the last hours of one day and the earliest hours of the next, the CO
concentrations decrease rather slowly although the traffic flow decrease is
rapid. The CO concentrations with minimum traffic flow are about the same
as they are 1n the middle of the afternoon.
8
-------
The results of October are similar to those for January. The major
differences are the displacements of the times of peak CO concentrations.
In the morning, the displacement from January to October is to earlier hours
and in October the peak concentration coincides with the peak traffic flow.
The difference between the January and October meteorology in the morning
occurs because"enhanced vertical mixing, as shown by the initial increase in
mixing heights, starts at an earlier time in October. In the afternoon, the
shift of peak CO concentrations from January to October is to later hours,
and the time delay between peak traffic flow and peak concentrations in
October is 4 hours. This afternoon displacement of the concentration peak
coincides with the displacement toward later hours, from January to October,
of the time when vertical mixing subsides (shown by the time when the discon-
tinuity in mixing height occurs).
To determine the statistical significance of the several variables on
the variation of the CO concentrations, a correlation analysis was applied
to the January data in Figure 4. The simple correlation factors between
concentrations and frequency of winds greater than 12 mi/hr, traffic and
mixing height were -0.27, 0.39, and -0.34, respectively. The multiple
correlation was 0.72, meaning that 52 percent of the variance in CO con-
centrations was explained by these three variables. This correlation is
significant at the 10 percent level.
DETAILED DATA ON POLLUTED PERIODS
A previous report (15) sought out the times of day of the greatest CO
concentrations for 8-hour periods in Baltimore and concluded that the night-
time hours were most frequently associated with the highest 8-hour CO con-
centrations. Because that conclusion is inconsistent with the findings of
this investigation, the meteorology that occurred on the five worst days at
the most polluted station (in the previous report) was evaluated. It was
found that relatively wet weather and high relative humidities prevailed on
all 5 days, indicating that water vapor may have been a cause of the recorded
high concentrations. (All moisture problems in the Maryland network have been
corrected; the data for the 12 stations used in this investigation have been
reported to be free of moisture problems by each local official responsible
for collecting the data).
Meteorological Differences Between Clean and Polluted Workweeks
The data from the Maryland network were examined to find a 2-week period
during which a relatively clean workweek (one with relatively low CO con-
centrations) was followed by a relatively polluted workweek and for which an
abundance of data was available. Two such periods, one in January and the
other 1n October, were found. The Dally Weather Maps (10), three-hourly
weather observations for Baltimore and Washington (9), and the calculated
mixing heights (6) for the periods were gathered and the differences noted.
-------
Data from nine stations are used in the January comparison of the clean
(January 8-12) and polluted (January 15-19) workweeks. The resulting dif-
ferences in hourly concentrations for each combined 5-day period are seen
in Figure 5. Combining the observations for each hour did little to minimize
the marked differences in concentrations. At almost all stations and for a
large majority of hours, the concentrations during the polluted week are
markedly greater and the greatest differences are observed at the times of
the peak concentrations.
A high pressure area centered over the middle of the United States
dominated the weather over Maryland during the clean workweek, January
8-12. Aloft there were westerly winds, and the surface winds at Washing-
ton National Airport and Baltimore's Friendship Airport were generally
light northwesterly through the week, the average temperature for the
period was about 10*F colder than normal. The sky was clear and the
visibility was 10 miles or more for most of the week.
A high pressure area over the eastern United States was the major
feature of the weather during the polluted workweek, January 15-19. On
Friday January 19, a cold front was situated just west of Maryland. The
flow aloft showed that air was being brought in from the southwest. The
surface winds at both airports were northwesterly on Monday but south-
westerly the rest of the week. The average temperatures started out 1°F
to 2°F warmer than normal and rapidly increased to 13°F to 15°F warmer
than normal by the end of the week. Approximately a quarter-inch of rain
fell on Friday afternoon. The visibility was onlv a little less than the
previous week (lowest reported visibility 6 miles) and most days were
cloud-free.
The urban mixing height data are shown in Table 3. The morning mixing
heights and wind speeds are seen to be markedly less in the polluted week
than in the relatively clean week. The ventilation factors at the time of
maximum mixing heights show that there was relatively little relief in the
afternoon during the polluted week, whereas there was much greater ventila-
tion during the clean week.
To determine the statistical significance of the variables on the
variation of CO concentrations, a correlation analysis was applied to the
CO data for Linthicum in Figure 5, the mixing heights in Table 3, the con-
current hourly wind speeds for Baltimore Airport, and the traffic counts for
Station 14 shown in Figure 1. The simple correlations between CO concentra-
tions and wind speed, traffic, and mixing height were -0.63, 0.13, and -0.38
respectively. The multiple correlation was 0.73, indicating that the three
parameters explained 53 percent of the variance in concentrations. This
correlation is significant at the 5 percent level according to an f-test.
Data from 11 stations for the October period were available in sufficient
quantity to make comparison of the clean workweek (October 15-19) with the
polluted workweek (October 22-26). The results are seen in Figure 6. At a
majority of stations, practically all hours showed markedly higher concen-
trations, and the differences were largest during the evening. At Riviera
10
-------
Beach, Silver Spring, and Gaithersberg, practically all the differences
were at times of the peaks, and at these times the concentrations were
markedly greater during the polluted week.
A cold front passed through Maryland late on Monday, October 15, 1973,
and a high pressure center dominated the weather from Tuesday through Friday.
The flow aloft over Maryland was generally westerly throughout the period.
The temperature started out warmer thanonormal but became colder with the
passage of the front, averaging about 5°F below normal. There was precipi-
tation at Baltimore and at Washington with the passage of the front on Monday.
Except for one observation during a rain shower, all visibilities were 15
miles or greater, and except for six observations there were no clouds at
Baltimore. The visibility and cloud observations at Washington were very
similar to those of Baltimore except that the visibility was never less
than 10 miles.
A high pressure center remained over the eastern United States through
almost all of the polluted workweek. Late on Friday, October 26, a cold
front passed over Maryland. The flow aloft over the area was initially
from the west, about 10 knots, but then became light and variable. No
precipitation occurred during the period, and although the temperatures
started out colder than normal they became warmer than normal, by about
4*F, most of the week. There were no clouds and the visibilities were
relatively low (but never less than 2 miles) throughout the period.
The urban mixing heights and ventilation factor data (6) are shown'in
Table 4. The morning mixing heights and ventilation factors were much
greater during the clean workweek than during the polluted workweek; there
was generally a four-fold advantage in the ventilation.
The simultaneous observations of concentrations, mixing heights, wind
speeds, and traffic for the Linthicum area for the October period were
analyzed to determine the statistical significance. The simple correlation
with concentrations and wind speed, traffic, and mixing height were -0.51,
0.11, and -0.41 respectively. The multiple correlation was 0.68, indicating
that 46 percent of the variance is explained by the three parameters. This
correlation is significant at the 5 percent level.
11
-------
SUMMARY
1. The average diurnal variation of CO concentrations tends to show a
peak near the beginning of the normal workday, followed by a rapid
decrease and leveling off that persists until about midafternoon;
then a rapid increase to high values that are maintained for about
3 hours in the evening; and then a gradual decrease to low values
in the early morning. The low values of midday and early morning
are comparable.
2. The diurnal variation of CO concentrations is not well correlated
with the diurnal variation in traffic density.
3. The diurnal variation of wind speed or the frequency of winds above
12 mi/hr, mixing height, and traffic density explain about 50 percent
of the variance in CO concentrations on a monthly basis and when a
polluted workweek is compared with a clean workweek.
4. The diurnal variation of concentration consistently shows marked
peaks that persist for 4 hours or less. Averaging CO concentrations
for 8-hour periods minimizes the acute exposure that frequently
persists for several hours.
5. A comparison of the diurnal variation of concentrations between a
cur.bside station and one in a metropolitan area that was remote from
traffic indicated there was little difference between the two sites:
the CO appears to readily move away from the sources and blanket the
entire area.
12
-------
REFERENCES
1. United Stated Congress, Clean Air Act of 1970. P161-604,
42 USC 1857 et seq., Washington, DC, December 1970.
2. Dadiani. J., Air Quality in Montgomery County. Maryland 1969-1973.
Department of EnvironmentaT Protection, Montgomery County
Rockville, MD, 1974. 129 pp.
3. Bowles, A., A. Solomon, and N. Wertheimer, Friendship Heights Air
Quality Study, Bureau of Air Quality Control Technical Memorandum,
Environmental Health Administration, State of Maryland, Baltimore,
MD, 1974. 29 pp.
4. Holzworth, G. C., Mixing Heights, Wind Speeds, and Potential for
Urban Air Pollution Throughout the Contiguous United States, AP-101,
U. S. Environmental Protection Agency, Research Triangle Park, NC,
1972. 118 pp.
5. Holzworth, G. C., "Variations of Meteorology, Pollutant Emissions,
and Air Quality," (presented at 2nd Joint Conference on Sensing of
Environmental Pollutants, Washington, DC, December 10-12, 1973).
6. Hand, J. A., (Air Pollution Meteorologist, National Weather Service
Washington, DC) Inter-office memorandum Mixing Height Data,
November 30, 1973. 2 pp.
7. McCormick, R. A., and C. Xintaras, "Variation in Carbon Monoxide
Concentrations as Related to Sampling Interval, Traffic and
Meteorological Factors," J.. Appl. Meteor., 1:237-243, 1962.
8. Tiao, G. C., G. E. P. Box, and W. J. Hamming, "A Statistical Analysis
of the Los Angeles Ambient Carbon Monoxide Data 1955-1972," J. Air
Poll. Control Assoc. 25:1129-1136, 1975.
9. United States Weather Bureau, Local Climatological Data 1973.
Monthly Summaries for Washington National Airport and Friendship
Airport, Baltimore, U. S. Government Printing Office, Washington,
DC, 1973.
10. United States Department of Commerce, National Oceanic and Atmospheric
Administration, Daily Weather Maps (for selected weeks), U. S.
Government Printing Office, Washington, DC, 1973.
13
-------
11. Bureau of Traffic Engineering, Traffic Trends, State Highway
Administration of the Maryland Dept. of Transportation, Baltimore
MD, 1974. 166 pp.
12. United States Department of Health, Education, and Welfare, Continuous
Air Monitoring Projects in Philadelphia 1962-1965. Pub!. APTD 69-14,
National Air Pollution Control Admin., Cincinnati, Ohio, p. 49, 52,
1969.
13. United States Department of Commerce, Weather Bureau, Decennial
Census of_ United States Climate - Summary of Hourly Observations
Baltimore. Maryland 1951-1960. Climatograpfiy of the United States
No. 82-18, Weather Bureau, Washington, DC, 1962.
14. Gross, E., The National Air Pollution Potential Forecast Program,
ESSA Tech Memo WBTM NMC 477 National Meteorological Center, Washington,
DC, 1970. 28 pp.
15. GCA Corporation, Transportation Controls to Reduce Motor Vehicle
Emissions in Baltimore, Maryland, Report APTD-1443, U. S. Environ-
mental Protection Agency, Research Triangle Park, NC, 1972. 47 pp.
14
-------
APPENDIX
SITE DESCRIPTIONS OF CARBON MONOXIDE
MONITORING STATIONS
Each site description, except Cumberland, is based on a site visit. The
Average Daily Traffic (ADT) counts are based on 1973 data that was provided
by the Traffic Inventory Section of the State Highway Division of the Mary-
land Department of Transportation. All the sampling intakes were 10 feet
above the ground except where noted otherwise.
1. Gaithersburg-Sampling intake is 35 feet above the street level on the
side of a building, about 75 feet from the nearest curb of Highway 355
(ADT-27,000 vehicles) and 150 feet from an intersection with a traffic
light.
2. Silver Spring-Sampling device is in a park about 50 feet north of the
nearest edge of Highway 1-495 (ADT 89,800 vehicles).
3. Bethesda-Sampling device is in a meadow about 100 yards from the nearest
local traffic (a few hundred cars a day), 0.25 mile from Old Georgetown
Road (ADT 35,000 vehicles) and 0.5 mile from Rockville Pike (ADT 42,900
vehicles).
4. Linthicum-Sampling device is in a 20-car, school yard, parking field
about 80 feet from the nearest local traffic (probably less than 100
cars a day) and 200 yards from Highway 1-695 (ADT 60,000 vehicles).
Friendship Airport is about 2.0 miles to the southwest of this station.
5. Baltimore (Calvert and 22nd Sts)-Samp!ing device is in a 100-car parking
field about 90 feet from Calvert St. (No ADT) and 50 feet from 22nd St.
(No ADT). Calvert St. is a major, one-way throughfare that carries
heavy outbound traffic in the afternoon.
6. Essex-Sampling device is in a 10-car parking field about 10 feet from
the nearest local traffic (a few hundred cars a day) and 100 yards north
of Eastern Blvd. (ADT 36,000 vehicles).
7. Cumberland-Sampling device is in a grassy field 20 feet from the curb
of Industrial Blvd. (ADT 21,500 vehicles').
8. Hagerstown-Sampling device is at rear portion of a 50-car parking
field about 175 feet from U. S. Highway UN. (ADT 10,000 vehicles)
and 175 feet from Maryland Highway 64 (ADT 10,000).
15
-------
9. Hyattsville-Sampling intake is 10 feet above the ground but below the
level of the nearby road. Sampling device is in a meadow 75 feet from
the nearest edge of Highway 410 (ADT 30,000 vehicles).
10. Towson-Sampling intake is 12 feet above ground on the top, rear end of
a service building on a college campus, next to a large wooded area.
The closest road, Highway 1-695 (ADT 76,500 vehicles) is about 150 yards
to the north of the sampling station.
11. Suitland-Sampling device is adjacent to a 500-car parking field 150 feet
southwest of Suitland Road (ADT 18,900 vehicles). Station is 0.3
mile from Silver Hill Road (ADT 34,000 vehicles) and 0.5 mile from
Suitland Parkway (ADT 18,500 vehicles).
12. Riveria Beach-Sampling devices is in a 20-car, school yard, parking field
about 10 feet from the nearest local traffic ( a few hundred cars a day)
and 0.25 miles from a major throughfare, Fort Smallwood Road (ADT 14,734
vehicles).
16
-------
PENNSYLVANIA
MARYLAND
7 CUMBERLAND
8 HAGERSTOWN
9 HYATTSVILLE
1 GAITHERSBURG
2 SILVER SPRING
3 BETHESDA
4 LINTHICUM
5 BALTIMORE 22"* 6 CALVERTSTS.
6 ESSEX
12 RIVIERA BEACH
13 1-695 WEST OF BALTI-
MORE HARRISBURG
EXPRESSWAY
14 1-695 SOUTH OF U.S.
40
15 1-695 SOUTH OF U.S.I
16 U.S.40WESTOF
HAGERSTOWN
/ \s I
Figure 1. Locations of CO monitoring stations (numbers corresponding to names and numbers shown in Table 1).
-------
3.0
1.5
0
4.5
2.5
0.5
3.1
1.6
0.1
3.0
1.5
0
3.5
2.0
0.5
6.0
3.0
0
4.0
2.5
1.0
12
SUNRISE
SUNSET
6 9 NOON 3
SILVER SPRING
9 12
6 9 NOON 3 6
BETHESDA ' 3')
12
Figure 2. Diurnal variation of CO concentrations by month, Maryland stations, 1973.
18
-------
3.0
5 1.5
0
3.0
jjj 1.5
0
3.0
I'-5
SUNRISE
SUNSET
E
3.0
5"
0
3.0
| 1.5
UU
o
O
o
SUNRISE
SUNSET
3.0
12
9 NOON 3 6
LINTHICUM (4)
12
4.0
2.0
0
3.0
1.5
3.0
1.5
0
3.0
1.5
0
4.0
2.0
0
3.0
1.5
SUNRISE
SUNSET
SUNRISE
SUNSET
12 3 6 9 NOON 3 6
CALVERT & 221"1 STS. (£l
BALTIMORE
3.0
1.5
0
3.0
1.5
0
3.0
1.5
0
3.0
1.5
0
4.0
2.5
1.0
4.0
2.5
1.0
12
SUNRISE
SUNSET
SUNRISE
SUNSET
9 NOON
ESSEX (
3 6 9 12
Figure 2. (continued) Diurnal variation of CO concentrations by month, Maryland stations, 1973.
19
-------
3.2
I "
0.2
3.0
S '-5
u.
0
3.9
| 2.4
0.9
SUNRISE
SUNSET
2
Q.
CL
J.U
1.5
n
-
.
-~_/^
-
' ^
3.0
1.5
0
3.0
1.5
0
12
SUNRISE
SUNSET
SUNRISE
SUNSET
3.0
1.5
0
3.0
1.5
0
3.0
1.5
0
3.0
1.5
0
3.0
1.5
0
SUNRISE
SUNSET
i
SUNRISE
SUNSET
369 NOON 3 6 9 12 12 3 6 9 NOON 3 6 9 12 12 3 6 9 NOON,
CUMBERLAND (7) HAGERSTOWN (?) HYATTSVULE
3 6 9 12
Figure 2. (continued) Diurnal variation of CO concentrations by month, Maryland stations, 1973.
20
-------
70
60
SO
40
30
20
10
T~T
v>
EC
" 0
V
369 NOON 3 6 9 12
JANUARY
70
60
50
40
30
20
10
/;
/ v \
u I
<2 /
o
u
o
369 NOON 3 6 9 12
APRIL
369 NOON 3 6 9 12
OCTOBER
03 MILES PER HOUR
13-24 MILES PER HOUR
25 MPH AND OVER
Figure 3. Diurnal variation of occurrence of selected wind speed ranges Baltimore, 1951 1960.
("FREQUENCY OF SPEEDS GREATER THAN 12 IN 1973
BASED ON 3 HOURLY OBSERVATIONS)
21
-------
12
9 NOON
JANUARY
MH
MIXING
HEIGHT
(m)
%OF
WINDS
>12 mph
TR
HOURLY
%OF
TRAFFIC
CO
CONC.
(ppm)
- 900 60 12 3.0 -
- 800 50 10 2.5 -
-700408 2.0 -
-60030 6 1.5 -
-500204 1.0
-400102 0.5 -
-3000 0 0.0
- 1300 60 12 3.0 -i
1100 50 10 2.5 -
- 900 40 8 2.0 -
-70030 6 - 1.5 -
-50020 4 1.0 -
-30010 2 - 0.5 -
L- 100 0 0 0.0 '
NOON
OCTOBER
Figure 4. Diurnal variations of mixing heights, frequency of wind speeds greater than 12 mph,
traffic flow and CO concentrations (hourly data plotted at hour ending time - eastern standard)
1973, Linthilum Area.
22
-------
IU.3
9.0
7.5
6.0
4.5
3.0
1.5
1 1 1 1 1 1 1
-
\ S~\ DIRTY
\ / \ WORKWEEK
_ \-X JANUARY 22-26 1973 _
_
\ X \ CLEAN
- v^ ^- WORKWEEK -
JANUARY 15-19 1973
1 1 1 1 1 1 1
i i i I I i r
©BALTIMORE
CALVERT AND 22nd STS. _
10.5
9.0
7.5
6.0
4.5
3.0
i 1.5
I I I T I I I
(f3) TOWSON
i i i i i i
n i r T i r
(7)CUMBERLAND
o 9.0
u
7.5
6.0
4.5
3.0
1.5
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
I T I I I I I
©HYATTSVILLE
n i i i i r
-------
10.5
9.0
7.5
6.0
4.5
3.0
1.5
1IIII
MARYLAND CO DATA
DIRTY
WORKWEEK
-OCTOBER 15-19
1973
CLEAN
^ WORKWEEK
/ OCTOBER 8-12
' 1973
I I I I I
V-M I
0BALTIMORE
CALVERT&22"''STS
9.0
7.5
6.0
4.5
I3'"
a.
z 1.5
4
CC
I I I
(TtfJTOWSON
I I
T I T II
^CUMBERLAND
I I I I I I
(T)BETHESDA
I I I I I
(Tl)SUITLAND
I I I I I I I
(12) RIVIERA BEACH
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
i i r \ T
7)SILVER SPRING
12 3 6 9 NOON 3 6 9 12
9 NOON 3
TIME, hour
I T I I \
(T)HAGERSTOWN
12 3
9 NOON 3 6 9 12
Figure 6. Comparison of diurnal variations in CO concentration. A clean workweek vs
a dirty workweek, October 1973, Maryland stations.
-------
Table 1. ANNUAL AVERAGE HOURLY PERCENTAGE OF TOTAL VEHICULAR TRAFFIC
AT FOUR MARYLAND HIGHWAY STATIONS, 1973a.
Location
1-695 1-695
West of Baltimore South of
Hour Harrisburg Expressway U. S. 40
of day (13) (14)
12-1 AM
1-2
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-1PM
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
1.6
1.1
0.5
0.5
1.1
3.6
7.0
6.8
5.2
4.9
5.1
5.3
5.3
5.7
6.8
8.2
7.8
5.7
4.6
3.8
3.5
2.8
2.3
1.7
1.1
0.6
0.6
1.0
3.6
6.7
6.3
5.0
4.8
5.1
5.3
5.3
5.6
6.8
8.4
7.8
5.9
4.9
4.0
3.6
2.8
2.3
1-695
South of
U. S. 1
(15)
1.9
1.3
0.6
0.6
1.4
4.4
7.5
6.0
4.4
4.3
4.5
4.8
5.0
5.5
6.6
8.3
7.9
5.7
4.8
4.0
3.8
3.1
2.6
U. S. 40
West of
Hagerstown
(16)
2.0
1.5
0.5
0.5
1.2
4.0
5.3
4.7
4.6
4.7
5.1
5.5
5.5
5.9
7.0
8.1
7.0
6.2
5.4
4.5
4.3
3.1
2.6
The Maryland Department of Transport describes the first three stations as
roads with consistent yearly traffic (little difference among the 12 months)
and the Hagerstown road as one with a consistent yearly traffic load and a
moderate seasonal traffic peak.
The numbers in parentheses correspond to the site identification numbers
in Figure 1.
25
-------
Table 2. CALCULATED URBAN MONTHLY AVERAGE MIXING HEIGHTS AND VENTILATION
FACTORS FOR WASHINGTON, DC, AREA, 1973.
Morning conditions
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Morning
mixing
height,
m
332
346
369
556
359
395
312
431
482
360
389
Speed in
mixing
layer,
m/sec
6.0
5.7
5.2
.7.4
4.6
2.8
3.3
3.3
4.2
, 3.9
5.6
a
Ventilation3
mz/sec
1,992
1,972
1,919
4,114
1,651
1,106
1,030
1,422
2,024
1,404
2,178
Afternoon conditions
Maximum
mixing
height,
m
882
1,045
969
1,618
1,883
1,483
1,704
1,788
1,517 ,
1,568
1,470
Speed in
mixing
layer,
m/sec
7.8
7.8
7.5
9.0
9.0
4.5
4.8
4.4
5.8
6.5
9.2
a
Ventilation
m2/sec
6 ,880
8,151
7,268
14,562
16,947
6,674
8.T79
7,867
8,799
10,192
13,524
Mixing height times wind speed.
26
-------
Table 3. CALCULATED URBAN MIXING HEIGHT AND VENTIALTION FACTORS FOR
THE WASHINGTON, DC, AREA FOR JANUARY 1973.
Morning conditions
Afternoon conditions
Day
of
Month
Morning
mixing
height,
m
Speed
mixing
layer,
m/sec
in
a
Ventilation*
m2/sec
Maximum
mixing
height,
m
Speed i
mixing
layer,
m/sec
n
a
Ventilation
.nr/sec
8
9
10
11
12
15
16
17
18
19
420
480
120
100
480
90
310
50
40
80
6.7
5.7
4.9
3.6
7.7 ,
(0.5)'
4.1
2.1
0.5
5.1
2,814
2,736
588
360
3,696 ,
(45)1
1,271
105
20
408
650
1,350
1,320
1,290
1,470
355
230
210
130
280
5.7
8.6
6.4
9.6
8.8
8.6
7.9
5.3
4.6
6.5
3,705
11,610
8,448
12,384
12,936
3,050
1,817
1,113
598
1,820
aMixing height times wind speed.
Value in parentheses indicates estimated speed. Reported speed was zero.
27
-------
Table 4. CALCULATED URBAN MIXING HEIGHT AND VENTILATION FACTORS FOR
THE WASHINGTON, DC, AREA FOR OCTOBER 1973.
Morning conditions
Day
of
Month
15
16
17
18
19
22
23
24
25
26
Morning
mi xi ng
height,
m
170
1,040
1,020
270
540
180
140
100
170
170
Speed
mixing
layer,
m/sec
4.3
11. 6
S.O
3.4
4-° u
(0.5)b
1.0
1.0
3.1
3:3
in
Ventilation3
mvsec
731
12,064
-8,160
918
2,160
(90)b
140
TOO
527
561
Afternoon conditions
.Maximum
mi xi ng
height,
m
2,300
1,380
.2,300
1,970
2,000
1,200
1,220
1,150
1,550
1,800
Speed in
mixing
1 ayer ,
m/sec
11 .5
11.8
10.4
10.9
5.0
4.0
2.0
4.4
4.3
7.1
Ventilation9
m2/sec
26,450
16,284
23,920
21 ,473
10,000
4,800
2,440
5,060
6,665
12,780
Mixing height times wind speeds.
"Value in parentheses indicates estimated speed. Reported speed was zero.
.28
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/4-77-009
2.
3. RECIPIENT'S ACCESSIOI*NO.
4. TITLE AND SUBTITLE
DIURNAL VARIATIONS IN CARBON MONOXIDE CONCENTRATIONS,
TRAFFIC COUNTS AND METEOROLOGY
5. REPORT DATE
February Iy77
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Gerard A. DeMarrais*
9. PERFORMING ORGANIZATION NAME AND ADDRESS
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
10. PROGRAM ELEMENT NO.
1AA603
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
In-house 9/75-2/76
14. SPONSORING AGENCY CODE
EPA - ORD
15. SUPPLEMENTARY NOTES
*0n assignment from the National Oceanic and Atmospheric Administration,
U.S. Department of Commerce.
16. ABSTRACT
Although pollutant emission patterns play important roles, they cannot adequate-
ly explain the diurnal variations in carbon monoxide concentrations found in urban
areas. In this study, hourly data from a large network of carbon monoxide
monitoring stations, with instrumentation corrected for moisture interference, are
analyzed and compared with traffic flow and meteorological conditions at several
locations in Maryland. The meteorological phenomena that appear to be important
in explaining the diurnal variations involve the ventilation effects resulting from
variable wind speeds and mixing heights.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
* Air pollution
*Carbon monoxide
* Meteorological data
* Traffic surveys
* Diurnal variations
* Wind velocity
* Mixing
* Height
~T3~B~
07 B
04 B
13 B
13 M
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report!
UNCLASSIFIED
21. NO. OF PAGES
34
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
29
------- |