CAB-4/C0-1
Technical Report
Status Report on the CO "Hot Spot" Project
May, 1978
by
Mark A. Wolcott
Characterization and Applications Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air and Waste Management
U.S. Environmental Protection Agency
-------
Abstract
This report outlines the preliminary information from an EPA study known
as the "CO Hot Spot Project." For this purpose, a "hot spot" is an area
with high ambient concentrations of pollutants which are almost entirely
due to the sources in the immediate area. This investigation centers on
carbon monoxide (CO), a pollutant which in urban areas, has been found .
to be due almost entirely to motor vehicles. The effort is designed to
gather information on meteorological conditions, vehicle characteristics
and ambient CO levels in portions of four metropolitan areas which have
been found to experience exceptionally high levels. The overall objective
of this project is a complete reassessment of the current FTP as a
measurement tool for evaluating light-duty vehicle CO emissions.
The results contained in this report include only early data from the
first city (Phoenix) and are unclear in the determination that ambient
CO levels are strictly a "hot spot" phenomenon.' There are indications
that a "regional" aspect to the problem may exist. More conclusive
information will be available by November, 1978 after the remaining
results are processed and the final report is drafted.
-------
Status Report on the CO "Hot Spot" Project
Background
It is important that a good test procedure accurately predict the on-
the-road percent reduction in emissions that would be.expected to occur
with different mobile source emission standards. It is not as important
that it accurately predict the exact level of the emissions. This is
because ambient concentration levels are assumed to be reduced in the
same proportion as are the emissions from the various sources. Thus,
a test procedure could be non-representative of regional or local
driving conditions and still accurately quantify percent reductions if
vehicle emission control systems uniformly reduce emissions over a wide
portion of the engine map. If, on the other hand, the emission control
systems are optimized to obtain the greatest reduction over the specific
test procedures (as presently occurs with the Federal Test Procedure),
emission reductions can be termed "cycle dependent" and the ambient
concentration reductions estimated would be in error.
The 1975 Federal Test Procedure (FTP) for light duty vehicles is currently
used to measure exhaust emission level hydrocarbons (HC), carbon monoxide
(CO), and oxides of nitrogen (NOx) from passenger cars and light-duty
trucks. This procedure was developed to approximate the emission burden
which can be attributed to a single vehicle during typical daily operation.
The 1975 FTP includes both cold-start and hot-Start testing and the
measured results are weighted in proportion to the number of each type
of start per average vehicle day. Such a procedure should adequately
estimate mobile source emissions which are area dependent. It ought to
therefore accurately predict changes in ambient air quality (op a concentra-
tion basis) as a result of more stringent standards for emissions from
mobile sources. The chemical processes leading to oxidant formation are
known to take place over time. Therefore, both HC and NOx can be
considered to be area pollutants, i.e. all pollutants of these types
emitted in an urban environment will, in time, contribute to the urban
oxidant problem.
Carbon monoxide (CO) is not a reactive pollutant but does pose an
additional analytical problem. Some very high ambient levels are known
to occur at times of low traffic, indicating that under certain meteoro-
logical conditions, CO can act as an area-wide pollutant. A recent
analysis by the Office of Air Quality Planning and Standards (OAQPS)
addressed the time of day that the 8-hour CO standard is exceeded at 30
CO sites throughout the nation. They found that about 50% of CO violations
occur between 6 p.m. and 1 a.m. One cannot be sure of the geographic
spread of the problem from these data, but they do seem to indicate that
the evening peak-hour emissions are more than proportionally influencing
the number of 8-hour CO violations because of a reduced dispersal rate.
-------
-2-
High ambient concentrations of CO are, however, also known to occur as a
result of traffic in the immediate vicinity of a receptor site. On this
basis, CO appears to have both local and regional characteristics. At
the present time, CO emissions for vehicles are measured concurrently
with HC and NOx emissions and the same regional hot/cold weighting
factors are applied. At any given receptor site, the fraction of hot
and cold operation may be significantly different from the respective
fractions which occur over.the "average vehicle-day" defined for the
FTP. Furthermore, driving cycles in locales experiencing excessive
ambient CO levels may differ markedly from the speed-time listing
specified in the 1975 FTP.
CO Hot Spot Program Design
The situation as described above suggested that an evaluation program
was necessary to determine what vehicle mix and operating characteristics
occur at locations with these high CO concentrations so that CO emissions
can be controlled in a way that will allow the National Ambient Air
Quality Standard (NAAQS) for CO to be met. Research ori the CO Hot Spot
problem was initiated in 1974 to address these concerns. The current CO
test procedure can be evaluated with respect to the findings of the
program. If the current FTP does not represent vehicle operating condi-
tions under which CO is a serious problem, the improvements in air
quality which have resulted and will continue to result from the imposition
of CO emission standards may be far less or far more .than the intended
percent reduction. Moreover, the control technology currently applied
to mobile sources to meet exhaust emission standards is highly dependent
on the present test procedure. If it becomes necessary to revise the
FTP to reflect the findings of the program, new control technologies,
possibly more cost effective than the present ones, might be available.
Thus, the CO Hot Spot Project will'help to ensure EPA that the reductions
in CO emissions necessary to meet the CO NAAQS can be achieved by an
equivalent reduction in CO emissions as measured by EPA test procedures.
This also helps to ensure that the most effective and economical control
technologies are being employed to control CO emitted by motor vehicles.
In July, 1977, a contract for the CO Hot Spot Program was awarded to
Stanford Research Institute (SRI). The intent of this contract was to
assess the extent of the CO problem in "hot spot" areas and to determine
vehicle operating characteristics (driving cycle and thermal state) in
CO violation areas. For this purpose, a "hot spot" is an area with high
ambient concentrations of pollutants which are almost entirely due to
the sources in the immediate area. The "hot spot" definition was interpreted
to mean that we should concentrate ori determining vehicle operating
characteristics within several* hundred feet of the emissions monitors
and not further than 1/2 mile away.
-------
-3-
Data were collected in San Jose, Seattle, Chicago and Phoenix during the
months of December, 1977 through March, 1978. - All four sites were
chosen because they were established CO hot spots according to OAQPS
analysis. Meteorological conditions, vehicle characteristics and
ambient concentration levels were recorded.
Meteorological conditions included ambient temperature, wind speed and
direction, barometric pressure, humidity and general climatic observations.
Vehicle characteristics included vehicle age and type mix, hot/cold
engine temperature mix, traffic speed and modal behavior, and traffic
volume. Ambient concentration levels of CO, as well as levels of HC and
NOx, were measured and recorded.
Locations for the monitors were chosen after local and state agencies
were contacted to obtain their recommendations as to which areas of the
test city were CO problem areas. Of the areas recommended, the one
chosen contained an intersection at which driver interview and vehicle
temperature measurements could be most easily obtained. . This process
was conducted on the first vehicle in line as it stopped at the traffic
signal. During the interview, the driver was asked the car's make and
model year, how long he had been' driving and how long had the car been
parked with the engine off before the present trip began. At the same
time, a second member of the contractor's staff used an infrared gun to
measure the temperature of the car's hood, trunk and tailpipe. Monitor
sites were grouped into those along the roadway in the vicinity of the
intersection and those either off or above that roadway. The former
group was intended to capture local "hot spot" CO, while the second
group was intended to establish "background CO", as used in the hot-spot
model developed by OAQPS. An investigation of background versus local
CO levels may be used in an evaluation regarding the appropriateness of
the FTP as the CO test procedure.
The data were collected on seven-day, 24-hours per day basis in the four
test cities. In each city, daily collection periods were nine hours
over 15 minute intervals; for the remaining 15 hours each day, the data
were collected over one hour intervals.
Preliminary Results
Although data reduction and analysis are not expected to be complete
until mid-November, 1978, some very preliminary CO data from the Phoenix
test site are available now. Figure 1 is an aerial photograph showing
the pace car route and equipment locations for Phoenix. (The pace car
is used to model the traffic in the vicinity of the receptor sites by
mirroring a typical vehicle's accelerations, decelerations, average
speed and stops). The Phoenix receptor sites can be grouped into those
in the immediate vicinity of a roadway (1-5, 8 and 11) and those either
off or above a roadway (6, 7, 9 and 10). Receptor sites 1, 2, 4 and 5
-------
-4-
are immediately next to the same roadway. Site 3 is in the middle of a
parking lot near another roadway and site 11 is next to a third roadway.
Site 6 is in the middle of a field approximately 30 meters from a roadway.
Site 7 is next to the State of Arizona monitor approximately 33 meters
from a roadway. Site 8 is on.top of the SRI mobile laboratory (6 meters
elevation), approximately 20 meters from a roadway. Sites'9 and 10 are
atop different five- story buildings.
Figures 2 and 3 show CO concentrations in parts per million (ppm)
measured during January 29, 1978. Figures 4 and 5 show CO concentra-
tions measured during February 2nd. For both, days, until the wind picks
up, concentrations are high in the early morning hours, a time when
there, is relatively little traffic. On the 2nd, a Thursday, concentra-
tions. are. higher during the morning rush hours than they are during the
same period on the 29.th, a Sunday. However, by noon on both days, CO
has, for the most part, dispersed. Then, during the evening rush hours,
CO concentrations begin to increase. However, those monitors in the
immediate vicinity of a roadway ("local" monitors) measured a higher .
rate of CO increase during the evening rush..hour than did those monitors
either off or above a roadway (."background" monitors). Figures 3 and 4
also show that on February 2, CO began to disperse after the evening
rush hours but increased again approaching its maximum at 11 p.m. This
lack of CO dispersal could be due to an inversion layer and seems to
indicate that CO has both local hot spot and regional characteristics.
Two other information sources also indicate that there is a regional
aspect to the CO problem. 0A0PS used their Storage and Retrieval of
Aerometric Data (SAROAD) base to determine maximum CO values in areas
with a high number of CO violations and found that those violations
often occur late at night when traffic flow is very low. Furthermore,
the. state environmental agency in Arizona does not view the Phoenix CO
level as a hot spot problem. Special monitoring conducted by that
agency has demonstrated that elevated CO levels occur through the early
a.m. hours on a broad front extending in excess of 10 miles.
At this time, testing has been completed in all four cities and the data
are being analyzed by the contractor. The focus of the analysis will be
to answer the following questions:
1. UTiat time of day are high CO readings or CO violations occurring?
2. To what extent is CO a local problem rather than a regional problem?
3. Can the vehicle operational characteristics for those vehicles that
contribute to the local CO problem be defined?
4. Can the ambient conditions be characterized for peak CO periods?
5. Was the one week monitoring period in each site representative of
typical operation at that site during CO violation periods?
-------
-5-
This work is currently scheduled to be completed by November, 1978
with the final report due in January, 1979. Additional status reports
will be prepared as data from the CO Hot Spot project becomes available.
-------
m&mwmww
JjfXah
&S|S@MSSS|imM
v"-?"- );-r* f.ijM rtr*¦?•****/".••¦•1 »>"• "> • •>• '"'(¦¦¦¦¦' ¦' > a.-a .j
STATE MONITOR
^Wil'VEHi'CLE* CHECKS
,*ii *<*y*p fn
» P. !»*~«».
1
t /m/vr-* A
iWWfcs;
•it .-. .rtr
i&, 'fift'ri
Ii«rillJ ft?r l- \\
mtxkjj*
.f • Er*,***;;*'?* r: »'* M.* -» st.
t^llWr' <«M!
;< E^lfeAiS^ j ^ ;£§^I5
ftai? * s
>1,. ilrnwtm-.
V.
4*utiU5j:x!
»I«i
jswroes^aa
lliiiSli
tt.'~..'V.
trr.MU
no
k i f. v H1*™ i - r. »-r-*
I % uv t^i I •> •. »**% f,
\ VV¦'* l'] ' '. '">*
V f!-'M! ^"Wwl •?'V V^v*'V :
'•X's ¦'< • V..' Wkv^K*:J?u>'V,
..VI Oji- V- V - tfl1*''..''.! ¦•'¦¦;. ••„%•
Ciir Of citnti.n
?. WASHINGTON
jl.'jTT^ ADAMS
5|; ® ('i| j-.JV.
"/'Y•:' '••.' \l iS -'1 |f
mI i? i ffm'MtfBVi!' •; ¦:••.•. a > & y
l fy/J TOftKSSM $ B#". .' vi j e
,11-j- ¦,,-'i »" <" t* i» • •'" ¦*'' i» * 1 ">r
JEFFERSON
X DAG SAMPLER LOCATIONS
FIGURE 3 PACE CAR ROUTE AND EQUIPMENT-LOCATIONS FOR THE PHOENIX TEST
-------
!TS«i-.'X^*''^,i-«>«!^.«4*'.fc. — —..i.'iin.i J.T^U!Zz.1i±mntv.v:irny**.• J _ . _
P |\0.O^,OV (W<7 h.^C/A \ 6(7 L£vyKjC^,JC\/^ ' <>^
V/mJ ^tfcvd.
-------
'/¦>>'• ¦; Y4"~
/- J^Jxu3^n7^i. .
PKl^^i-vV^^- ( C'C? ' 6^'i C,.,..\,./,I,/I..i-.
3 fVH {*V*',v
20'
E£E
EE
iir1 —
lIT
—i -
~L7_:
—i-
1/7
V1
•4
v/>
10
£"!-:
... i
;!__
-\
rd-
e>
¦ 2b-
iTtr
:im:
. W 4.
• i —
m
V. J ••
SfcJ
— h-
:r.r:\ :r
i
rrr^'v."
ztr.
j— ~!_
J_.. !_,
io
... j.
m
¦—i—
zzfc
.1.-.
-b
_ J
.TT
4
5
v)
ao
._i.
W
.0 J
¦ i
•vj k
vP
&
oi
5.0"
V 10
.•x
i/i
v
1~N~
~tr-
EE
: Jiv.
—I-
**4—
—I
t-j__
ill
j
-i •-. i
Zl\ llf.
_.L n
—rr
„Lj
t;
i-i, -i «.
_L_.
i
_i
12
10
Figure 3
f)A.*> ' Q\hc\l
u
7>?
12
NOON C
,.1-t.L
--I-
-I...
EE
• • i ¦
— J--
¦ ' t '
—
;4~
•~LT..
> >
. . L.._
TTt
4v-
! * ^
;-u;
j
¦I - -
1 2
)0
J! )i
-------
TXT^cTr'
cmtn) $ ,t ,y 3-1
4.-A jWh^
wr" "-""TiT "f ul
~i"S\ "*'" ¦¦"';¦ .... ¦
rTiDOiNv**. Cyic\ UsA^1- ) <—O uo\ -:^-x -*•;
/ / f / C3 x 3 'I t
C*u4U-\ /
Fifiure 4
-------
sx»J~iL~.— .'^i—-.¦¦r":,,,-...'j 'U^i,^^.^-1,,, 11
PWiTvW Baq >«» C_pr^
77. Pc(i V^O'3
4 5 6
Figure 5
II IZ 1.2
MOON Os*^
------- |