Cloud Aerosol and Precipitation Spectrometer (CAPS)
To characterize larger accumulation mode aerosol particles and small to medium-sized
cloud droplets or crystals aboard the WB57-FF, we will deploy the Droplet Measurement
Technologies (DMT) cloud, aerosol and precipitations spectrometer (CAPS, Baumgardner et al.,
2002) that measure in the size range from 0.5 to 1500 m. The CAPS has two optical sensors to
derive the size of individual particles: 1) the cloud imaging probe (CIP) and 2) the cloud aerosol
spectrometer (CAS). The cloud imaging probe (CIP) records the images of particles that pass
through its collimated laser beam. Cloud particle habits and sizes are derived from these images
with a size resolution of 25 m in the range from 50 to 1600 m.
The CAS collects the light scattered in the forward direction (4-12) by particles passing
through a focused laser beam and converts the light intensity to a particle diameter between 0.5
and 50 m using Mie light scattering theory. The CAS also measures backscattered light (168176) in order to derive particle refractive index (Baumgardner et al., 1996) and shape factors
(Baumgardner et al., 2004). Three pieces of information are recorded by the CAS about each
particle, the forward scattered light intensity, backward scattered light intensity, and the time
between each particle. The particle by particle information is used to derive size distributions,
particle refractive index and shapes, and concentration fluctuations at scales of a meter or less.
The plume measurements taken during the PUMA experiment serve to illustrate how the
CAS measurements divulge information about the physical and chemical processes. The time
series in the top panel of Fig. 1 shows the number (solid line) and volume concentration (dashed)
of particles measured by the CAS during the plume pass at 18 Km, averaging in one second
intervals. The bottom panel shows the time series of particle-by-particle sizes as the aircraft
enters the plume during the first 200 ms of penetration, or approximately 36 meters into the
cloud of particles. This shows that the supermicron particles appear to cluster while sub-micron
particles are more homogeneously distributed. The spatial distribution of particles can be
examined also by looking at their interarrival times, a measure of the spacing between particles.
Figure 2 shows the relative frequency of arrival times during the 18 Km cloud pass. If particles
are homogeneously mixed, and uniformly but randomly distributed in space, then the form of
this frequency histogram is a Poisson probability distribution that can be predicted with just the
average arrival time of 800 s (shown as the straight solid line on this figure). It can be seen,
however, that although the general shape of the measured distribution is close to the predicted
curve, there are significant deviations at the short arrival times that indicate that a large number
of the measured particles are parts of clusters with spatial separations smaller than expected.
The size distributions derived from the CAS measurements describe how the plume
particles are stratified between sub and super micron particles, as shown in Fig. 3, and also give
information about their time evolution. Four plume passes were made, the first at 18 km and the
last at 13.6 Km. The 18 km pass is through the most fresh plume and the last pass was through
the oldest. Approximately 10 minutes passed from first to last pass. As seen in Fig. 3, the shape
of the size distribution, where particle concentration is given in frequency relative to total
number, changes very little in the submicron range when the oldest particles are compared with
youngest. The largest difference is an increase in the number of particles in the submicron peak
formed at 0.7 m. As the particles age, the concentration of supermicron particles is seen to
decrease, most likely by evaporation. These are probably contributing to the concentration
increase at 0.7 m.
The particle composition and shape is derived from the ratio of forward to back scattering
(F2BR). Figure 4 compares measured F2BRs with those calculated from Mie theory. In Figs. a-c
the theoretical values, assuming particle sphericity, are shown for particles > 1 m, for refractive
indices of ice (m=1.31), nitric acid (1.40) and aluminum oxide (m=1.60). The measured values
for supermicron particles are shown in Fig. 4d. The comparison of measured and theoretical
F2BRs suggests that the supermicron particles are generally composed of material with higher
refractive indices. These are most likely spherical particles since light scattering models
predicted much higher F2BRs for non-spherical shapes (Noël et al., 2001). A comparison of the
measured submicron F2BRs, shown in Fig. 4e with theoretical distributions of particles with
refractive indices of 1.60 and 1.31 (Figs. 4f and g) indicates that the smaller particles might be
external mixtures with different chemical composition or shapes.
Baumgardner, D., J.E. Dye, B. Gandrud, K. Barr, K. Kelly, K.R. Chan, 1996: Refractive indices of
aerosols in the upper troposphere and lower stratosphere, Geophys. Res. Lett, 23, 749-752.
Baumgardner, D., H. Jonsson, W. Dawson, D. O’Connor and R. Newton, 2002: The cloud,
aerosol and precipitation spectrometer (CAPS): A new instrument for cloud investigations,
Atmos. Res., 59-60, 251-264.
Baumgardner, D., H. Chepfer, G.B. Raga, G.L. Kok, 2004: The Shapes of Very Small Cirrus
Particles Derived from In Situ Measurements, Geophys. Res. Lett.,Submitted.
Noël V., G. Ledanois, H. Chepfer, P. H. Flamant, Computation of single scattering matrix for
non-spherical particles randomly or horizontally oriented in space , Appl. Opt, 40, 43654375, 2001.
2000
160
1600
120
1200
80
800
40
400
0
81228
3
81229
81230
81231
81232
81233
81234
Volume (m3cm-3)
Concentration (cm-3)
200
0
81235
Seconds in Plume
Size (m)
2.5
2
1.5
1
0.5
81228
81228.05
81228.1
81228.15
81228.2
Seconds in Plume
Figure 1:
The time series in the top panel shows the of number concentration (solid line)
and volume concentration (dashed) of particles measured by the CAS during the plume
pass at 18 Km. The bottom panel shows the time series of particle-by-particle sizes as the
aircraft enters the plume during the first 200 ms of penetration, or approximately 36
meters into the cloud of particles.
Frequency of Occurence
0.1
0.01
0.001
0
400
800
1200
1600
2000
2400
2800
3200
Interarrival time (s)
Figure 2: This graph is the relative frequency of arrival times during the 18 Km cloud pass. If
particles are homogeneously mixed, and uniformly but randomly distributed in space, then the
form of this frequency histogram is a Poisson probability distribution that can be predicted with
just the average arrival time of 800 s (shown as the straight solid line on this figure).
Fraction of Total
0.5
13.6 Km
0.4
0.3
0.2
0.1
Fraction of Total
0
0.5
14.2 Km
0.4
0.3
0.2
0.1
Fraction of Total
0
0.5
0.4
16 Km
0.3
0.2
0.1
Fraction of Total
0
0.5
18 Km
0.4
0.3
0.2
0.1
0
0
0.5
1
1.5
2
2.5
3
Particle Size (m)
Figure 3: The size distributions derived from the CAS measurements describe how the plume
particles are stratified between sub and super micron particles and give information about their
time evolution. Four plume passes were made, the first at 18 km and the last at 13.6 Km. The 18
km pass is through the most fresh plume and the last pass was through the oldest. Approximately
10 minutes passed from first to last pass. The particle concentration is given in frequency relative
to total number.
(a)
1
(b)
Spheres > 1 m
m = 1.31
(c)
Spheres > 1 m
m=1.40
Spheres > 1 m
m=1.60
0.1
0.01
0.001
1
(d)
(g)
(e) Measurements
< 2 m
Measurements
> 1 m
Spheres < 1 m
m=1.31
0.1
(f)
Spheres < 1 m
m=1.60
0.01
0
20 40 60 80 100 120 140 160 180 200
Forward to Back Scatter Ratio
0
20 40 60 80 100 120 140 160 180 200
Forward to Back Scatter Ratio
0
20 40 60 80 100 120 140 160 180 200
Forward to Back Scatter Ratio
Figure 4: The measured forward to back scattered ratios (F2BR) are compared with those
calculated from Mie theory. In Figs. a-c the theoretical values, assuming particle sphericity, are
shown for particles > 1 m, for refractive indices of ice (m=1.31), nitric acid (1.40) and
aluminum oxide (m=1.60). The measured values for supermicron particles are shown in Fig. 4d.
The measured submicron F2BRs, shown in Fig. 4e are shown with theoretical distributions of
particles with refractive indices of 1.60 and 1.31 (Figs. 4f and g).