WMOD00420

• Title: Corrections of Surface Particle Probe Measurements for the Effects of Aspiration
• Date: 8/4/1988
• Weather Modification - Doc Type: Report

548

JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY,

VOLUME 5

(1985) performed numerical studies modeling water
droplet trajectories within the aspirator flow field, while
Holroyd (1986) made comparisons in snowfall with a
2D-C probe operated with and without aspiration.
Norment's calculations indicated that errors, intro-
duced by aspiration into the measurement of particle
concentrations, are a function of particle size. Holroyd's
measurements showed that the instrument overesti-
mates IPC, but found the error to be the same at all
ice particle sizes sampled. The purpose of this paper is
to provide a more complete empirical calibration of
this instrument by presenting simultaneous measure-
ments with an aspirated 2D-C and three other methods
to measure IPC at the surface. For comparison, IPC
measurements were collected with an airborne 2D-C
on a low-flying aircraft, oil-coated slides exposed to an
airstream, and a 2D-C carried on the front bumper of
a truck.

2. Observational equipment

The aspirated 2D-C used in all of the experiments
has been described by Humphries (1985). The size res-
olution of the probe is 25 ~m and the range is 800 ~m.
The probe is always held vertically. The aspirator pro-
vides a constant 13.5 m S-I airspeed at the sampling
section; however, particles only reach a fraction of this
speed before they are sampled. The clock rate of the
2D-C was set for a particle speed of 8 m S-I since this
was found to produce spherical images of 100-300 ~m
water droplets sprayed into the aspirator, and was also
found to produce reasonable snow particle images.
Holroyd (1986), using the "square image" option of
an aspirated PMS gray-scale 2D-C probe, also deter-
mined that the speed of snow particles through the
sample area was about 8 m S-I. The sample volume
with an aspirator is a constant 0.66 L S-l, determined
by the 13.5 m S-I airspeed, and is independent of the
clock rate of the 2D-C. To obtain IPC at the surface,
the other techniques used were measurements by an
airborne 2D-C flown 10 to 20 m above ground, ice
crystals collected on oil-coated slides exposed to an
airstream and counted, and measurements from a 2D-
C carried on a truck. Measurements from the airborne
and truck-mounted probe also provided size distribu-
tions, whereas oil-coated slides were used to provide
only concentration.
Comparisons with the airborne 2D-C were done us-
ing the University of Wyoming's King Air, which has
been described by Cooper et al. (1984). The 2D-C on
the aircraft was similar to the probe used on the ground.
Image data from the airborne and ground 2D-C were
analyzed with the same software to eliminate differ-
ences that could result from different processing al-
gorithms (Walsh 1984). Processing eliminates spurious
particles, counts and sizes acceptable particles, and
calculates sample volume. For these data, software from
the University of Wyoming was used. Sample volume
corrections for small sizes were not used.

Collecting ice crystals on oil-coated slides to measure
IPC has been done by Cooper and Saunders (1980) on
aircraft, and by Vali et al. (1981) in a wind tunnel. For
the experiments presented here, microscope slides were
coated with mineral oil, fixed on the' end of a metal
rod, and spun for 2 to 16 revolutions by an electric
motor. The number of revolutions was selected based
on estimated snowfall rate. After exposure, random
photographs of the slides were taken under a micro-
scope. Particles on the slides were then counted from
photographs. The device used to expose the slides was
built by the University of Wyoming. The glass slides
were 12 X 55 mm and the radius of rotation 611 mm.
The speed of the slides averaged 8.8 m S-I.
Collection efficiencies for the slides were calculated
using Ranz and Wong (1952). At 8,8 m S-I the collec-
tion efficiency for a 12 mm ribbon is 0.3,0.5, and 0.9
for particles with diameters of25, 50, and 100 ~m. For
the purpose of this study, the collection efficiency was
assumed to be 1.0 for particles greater than 100 ~m
and 0.5 for particles less than 100 ~m. Only a few par-
ticles less than 100 ~m were found.
The, IPC were calculated from each slide using

IPC = N;/(A27TLR),

where Ni is the number of ice crystals counted (adjusted
for collection efficiency), A the area photographed, L
the length of the rod, and 1? the number of revolutions.
For comparison, time intervals of 20 sec were used to
obtain concentrations from the 2D-C. This typically
resulted in sample volumes on the 2D-C of 0.5 to 3
times that of the oil-coated slides. Although it is best
to compare sample volumes of similar size, the 'pro-
cessing used for the 2D-C data limited the minimum
sampling interval to 10 s. Thus sample volumes from
the 2D-C were broken into discrete lOs, 6.6 L samples;
whereas sample volumes from the oil-slide data varied
continuously depending on the area photographed and
the number of revolutions. To simplifY data processing,
20 sec was chosen as the sampling interval for the 2D-
C data. This was a compromise between the extremes
of the oil-slide data and insured that the 2D-C mea-
surements coincided closely in time with the oil-slide
data. Approximately half of the oil slides had sample
volumes within 50% of the 13.2 L used for the 2D-C
data. A comparison was made between 6.6 and 13.2
L sample volumes for the 2D-C data and there were
no significant differences.
The truck-mounted system was used by Holroyd
(1986) who described the probe, including the truck
mount. The probe was mounted on a post on the front
bumper of a truck and could be held vertically and
aspirated, or held horizontally and operated in an air-
craft configuration by driving the truck. The only
change in this system was the use of a different ane-
mometer to measure airspeed and control the clock
rate of the 2D-C. The anemometer was calibrated in a
wind tunnel over the range of 0 to 40 m S-I and had