United States
Environmental Protection
Agency
National Exposure
Research Laboratory
Reisearch Triangle Park, NC 27711
Research and Development
EPA/600/SR-95/117
r EPA Project Summary
August 1995
Tracer Studies of Transport and
Transformation in Cumuli
J. L. Stith, A. J. Alkezweeny, and D. A. Burrows
Airborne measurements near
Champaign, IL and Milwaukee, Wl
were made during the summers of
1990 and 1992 to study pollutant trans-
port and transformation by clouds.
Measurements of the aerosol size dis-
tributions, wind, turbulence, cloud mi-
crophysics parameters and trace gases
were made from 31 research flights.
During the 1990 study SF6 was used as
a tracer to determine cloud transport
and entrainment.
In large clouds air from below the
cloud bases was transported without
dilution through the mid-levels of the
clouds. On the other hand, in smaller
clouds a more uniform dilution was
observed as a result of outside air en-
trainment. The dilutions in the lower
levels of the small clouds could be
explained by a simple buoyancy sort-
ing model.
An increase in the relative sizes of
aerosol in the accumulation mode was
observed in an area that was likely af-
fected by the venting of cumuli in the
area. Similar increases in size were not
observed in evaporating regions of
stratiform clouds. A hypothesis is pro-
posed to explain the measurements.
The cumuli activate much smaller aero-
sol which, after aqueous phase reac-
tions and evaporation, have a much
greater relative increase in size than
the larger aerosol activated by stratus
clouds.
The results from the entrainment ex-
periments suggested that, during the
early stages of entrainment, air from
above the rising cloud is carried along-
side the upper cloud region by the cir-
culation present there. Later, the air is
mixed into the main portion of the cloud
and rapidly diluted with the cloud inte-
rior. The observations are consistent
with the hypothesis that entrainment
occurs through a vortex-like circula-
tion that brings air from above the ris-
ing cloud top into the central region of
the cloud.
The eddy correlation method was
used to determine the transfer veloci-
ties of gases and aerosols over Lake
Michigan downwind of Chicago. The
results show downward transfer veloci-
ties (deposition) of 0.15 and 0.86 cm s-1
for 03 and aerosols in the size range of
0.1 to 3.0 jam in diameter and upward
transfer velocities of 0.04 and 0.54 crn
s-1 'for CO2 and water vapor about 7.5
km from the shoreline. At mid-lake
much lower transfer velocities were
measured. The turbulence intensity, in
the subrange, was found to decrease
as the air traveled over the cooler wa-
ter.
This Project Summary was developed
by the National Exposure Research
Laboratory's Atmospheric Modeling Di-
vision, Research Triangle park, NC, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
C ouds cover about half of the earth's
surface and occupy much of the tropo-
sphere. During their formation and dissi-
pation, they interact in several ways with
atmospheric pollution. Cumuliform type
Printed on Recycled Paper
-------
clouds (Cumuli) are especially effective at
vertically redistributing pollutants. The larg-
est variety, cumulonimbus, transport large
amounts of polluted lower level air into
the upper troposphere and bring cleaner
mid- and upper-level air downward.
Smaller cumuli play a similar role, except
that they influence the vertical distribution
of pollutants in the lower atmosphere. Al-
though the effects from these smaller
clouds are not as dramatic as those from
the larger storms, their much greater num-
ber makes them very important in atmo-
spheric chemistry. Both types of clouds
transport pollutants away from the earth's
surface, a major sink for most pollutants.
Dry deposition is reduced, leading to long
range transport of pollutants thus impact-
ing further away receptors. On the other
hand, precipitating clouds remove aero-
sols and trace gases from the troposphere.
Clouds also change the chemical and
physical states of pollutants. For instance,
the formation of acidic species by oxida-
tion of SO2 in clouds is much faster than
in clear air. Clouds modify the size spec-
tra of the atmospheric aerosol, changing
Hs physical properties (e.g. light scatter-
ing, residence time, etc.). Pollutants that
are lifted to higher elevations will experi-
ence higher light intensity which may alter
their photochemistry.
Modeling the fate of pollutants on re-
gional and larger scales is a major area of
research for the US EPA. These models
must properly account for the effects of
clouds. For example, RADM, the Regional
Acid Deposition Model, must account for
cloud processes as well as gas-phase
chemistry and deposition.
This report summarized the work that
was done under a cooperative agreement
between the US EPA and the University
of North Dakota (UNO). The objectives of
the program were to better understand
cloud processes (entrainment, cloud -
aerosol interactions, etc.), and to collect
measurements in support of the EPA re-
gional modeling, especially the cloud mod-
ule for the RADM and related models.
During the period of the program the
RADM and related code was extended by
the EPA to address other concerns be-
sides acid deposition (e.g. the fate of at-
mospheric aerosols). Consequently, the
data from this program also support this
broader effort.
Cloud and Aerosol Interactions
Measurements of in-cloud scavenging
were made using the University of North
Dakota Cessna Citation research aircraft
on June 12, 1992, approximately 60 nau-
tical miles southwest of Green Bay, Wl.
The aircraft was instrumented to measure
several cloud physics and standard me-
teorological parameters. The cloud drop-
let size distributions were measured using
the Particle Measuring System (PMS) for-
ward-scattering spectrometer probe model
100 (FSSP). The FSSP is capable of mea-
suring droplets in the diameter range of
about 3-50 u,m. Another PMSs probe, the
passive cavity aerosol spectrometer probe
model PCASP-100X, was used to mea-
sure the aerosol size distributions in the
range from 0.1 to 3.0 u,m. The probes
complement each other and cover the
range from 0.1 to 50 u.m in a total of 30
channels. A 2DC PMS probe was used to
determine ice crystal concentrations in the
cloud.
The droplet concentration just above the
cloud base is^on|y about half pfjhe aero-
sol concentration measured in the cloud
inflow. This indicates that only about half
of the aerosols act as cloud condensation
nuclei active at the supersaturation of the
cloud under study. To further investigate
this problem we have examined aerosol
size distributions from all sampled alti-
tudes. The distribution in the cloud inflow
(where the relative humidity was 96%)
shows the normalized concentration de-
creasing with increasing aerosol diameter.
However, within the cloud the size distri-
butions are relatively flat. When compared
to the inflow region, the in-cloud distribu-
tions show a marked increase in the con-
centrations of aerosols greater than about
0.4 mm. This means that a significant
portion of the aerosols did not grow larger
than 3.0 mm. The cloud droplet size distri-
bution is slightly shifted toward larger drop-
lets at the higher altitudes.
The results show that for cumulus clouds
aerosol sizes are shifted toward slightly
larger sizes after evaporation. On the other
hand, no significant change in the aerosol
sizes was observed after the aerosols were
processed by stratiform-clouds.- We_be-
lieve that this result is due primarily to the
differences in cloud supersaturation in
stratiform cloud droplets compared to drop-
lets in cumuli.
Transfer Velocities of Gases
and Aerosols Across the Lake
Michigan Surface
On June 18, 1992 a constant altitude
flight was made at 300 m above Lake
Michigan near the Chicago shoreline and
about 50 km downwind of it. The eddy
correlation method was used to calculate
the fluxes of CO?, O3, water vapor and
aerosols in the diameter range of 0.1 to
3.0 urn. The fluxes near the shoreline
were found to be significantly higher than
those in the middle of the lake. The turbu-
lence intensities, as measured by e"3, were
2.92+0.75 cm^s-' and 1.34±0.39 cm^s"'
near the shoreline and mid-lake respec-
tively. Fluxes measured near the shore-
line were likely to be representative of
those at the surface of the lake, because
of the strong turbulence during the mea-
surement. However, this is not true for the
fluxes measured at mid-lake.
The fluxes near the shoreline for O3
and aerosols were directed toward the
surface and corresponding to transfer
(deposition) velocities of 0.15 cm s-l and
0.86 cm s'1, respectively. For CO2, and
water vapor, the fluxes were directed up-
ward and corresponding to transfer ve-
locities of 0.04 cm s'1 and 0.54 cm s"',
respectively.
A west-to-east constant altitude flight
over the lake, starting from Chicago,
showed that turbulence, as measured by
e1'3, decayed slowly along the flight track.
The O concentration steadily increased
from 39 ppb to about 52 ppb as the air
moved away from the shoreline.
Transport of Air by Convective
Clouds
Large volumes of air from the mixed
layer were lifted by a severe conyective
storm with very little mixing with air from
the mid and lower levels of the free tropo-
sphere. Thus, these very strong storms
may be best modeled with little or no
entrainment in their lower levels. On a
much smaller scale, the small clouds that
were sampled contained a more uniform
distribution of mixtures of air from below
the cloud with air from near the level of
entrainment. The upper portions of these
clouds seem to be more dilute than their
lower part, but this result may also be a
function of the age of the cloud.
The techniques of buoyancy sorting are
used in recent convective parameteriza-
tion techniques. This approach was tested
against observations and used to explain
- the-behavior of a cloud-base region that
was tagged with a tracer. The results of
the buoyancy sorting model are encour-
aging in that they offer a simple explana-
tion for the behavior of the tagged region
and they are able to explain the distribu-
tion of conserved parameters fairly well in
the lower levels of the clouds that were
studied, and less well in the upper re-
gions. This may be improved when more
information on the dynamics of the clouds
are included. The results to-date suggest
that a comparison of our results with the
results from a more dynamic model—that
also includes the buoyancy sorting
method—should be conducted.
Entrainment
SF6 tracer was released during single
aircraft passes above the top of growing
-------
turrets associated with three different cu-
muli to study the entrainment of air by the
cloud. The clouds ranged in size from a
vigorous convective turret associated with
a small thunderstorm, to a small, fair-
weather cumulus.
The results suggest that, during the early
stages of entrainment, the tracer remained
mostly out of the cloud and was carried
outward and down alongside the upper
cloud regions by the circulation present
there. In each experiment, concentrated
SF6 was first found on the edges of the
cloud turrets. Later, the tracer mixed into
the main portion of the turrets and rapidly
diluted. The observations are consistent
with the hypothesis that entrainment oc-
curs through a vortex-like circulation that
brings air from just above the rising cloud
top into the central region of the cloud.
Analysis of buoyancy considerations in
cumuli suggest that much of the entrained
air should remain relatively close to, or
just slightly lower than, the altitude of en-
trainment. This helps explain the behavior
of the tracer in the above study; a portion
of the entrained tracer should be found
just below the altitude where it was re-
leased after being entrained into the cloud.
The tracer was indeed found in these lo-
cs.tions (although it may have also been
located in other parts of the cloud that
wore not sampled).
-------
Jeffrey L Stfth, AbdulJ. Alkezweeny, and D.A. Burrows are with the University of
North Dakota, Dept. of Atmospheric Sciences, Grand Forks, ND 58201-9007.
EPA Project Officers, Dr. Jason K.S. Ching and Dr. Jonathan E. Pleim are on
assignment to the National Exposure Research Laboratory (formerly Atmo-
spheric Research and Exposure Assessment Laboratory, see below), from the
National Oceanic and Atmospheric Administration.
Tha complete report, entitled "Tracer Studies of Transport and Transformation in
Cumuli," (Order No. PB95-255717; Cost: $19.50, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Modeling Division
National Exposure Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC27711
United States
Environmental Protection Agency
Technology Transfer and Support Division (CERI)
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/600/SR-95/117
------- |