Ž .
Atmospheric Research 52 2000 303–314 www.elsevier.comrlocateratmos
Short communication
Cloud drop spectra at different levels and with respect to cloud thickness and rain
S.K. Paul
Indian Institute of Tropical Meteorology, Pune 411 008, India Received 27 April 1999; received in revised form 23 August 1999; accepted 6 October 1999
Abstract
Ž .
Measurements on drop size were made in cumulus clouds over Pune inland region during the summer monsoon seasons. In this paper, the measurements of the cloud drop spectra made in
non-raining clouds at different levels and for different thickness have been studied. Also, those on Ž
. the days with rain and with no rain the rain being observed within the clouds have been
compared. The average spectra broadened with height. The concentration of drops 50 mm Ž
. Ž
. Ž
. N , liquid water content LWC , mean volume diameter MVD and dispersion increased with
L
Ž .
Ž .
height. The concentration of drops - 20 mm N and total concentration N
decreased with
S T
height. The spectra were broader, while N and N are smaller and the other parameters are
S T
greater for thicker clouds as compared to those for thinner. The spectra were broader, while N
S
and N are smaller and the other parameters are greater on the days with rain with respect to those
T
on the days with no rain. The distributions were bimodal at higher levels, for thicker clouds and on the days with rain, while they were unimodal at lower levels, for thinner clouds and on the
days with no rain. The variations of the cloud drop spectra, preceding rain, at initial stage of rain and following rain are discussed. q 2000 Published by Elsevier Science B.V. All rights reserved.
Keywords: Indian clouds; Cloud drop spectra; Liquid water content
1. Introduction
A knowledge of cloud drop size distribution is important in the study of cloud physics, cloud development and rain formation. The processes which determine droplet
size distribution in cumulus clouds are not completely understood although cloud Ž
. condensation nuclei CCN spectrum and the entrainment between the cloud and the
Ž .
surrounding clear air are known to be important factors Raga and Jonas, 1993 .
Ž .
Weickmann and Aufm Kampe 1953 showed drop spectra ranging from 6 to 66 mm diameter in fair weather cumulus and ranging from 6 to 200 mm in cumulonimbus
0169-8095r00r - see front matter q 2000 Published by Elsevier Science B.V. All rights reserved. Ž
. PII: S 0 1 6 9 - 8 0 9 5 9 9 0 0 0 3 7 - X
Ž .
clouds. Squires 1958 obtained drop spectra 4–92 mm and 4–165 mm diameter in dark Ž
. stratus and Hawaiian orographic clouds, respectively. Skhirt Tadze 1984 measured
drops in the diameter range 2–70 mm by photoelectric counter in cumulus clouds over Cuba. Cloud droplet spectra were also measured with a forward scattering spectrometer
Ž .
probe Raga and Jonas, 1993; Hallberg et al., 1997 . The growth rate of raindrops during fall depends upon the cloud type which, in turn, is determined by the vertical wind shear
Ž .
Ž .
Takahashi, 1981 . Takahashi et al. 1989 observed that the summits of trade wind cumuli are regions of active raindrop growth.
The author, in an earlier paper, obtained broader droplet spectra over the Pune Ž
. inland region than over the Arabian sea and this was mentioned to be due to the
Ž presence of an industrial complex belt upwind of the experimental site at Pune Paul and
. Pillai, 1992 . Here, the author makes a comparative study of the cloud drop spectra at
different levels, for different cloud thickness and on rain and non-rain situations.
2. Measurements
Measurements of drop size spectra were made in cumulus clouds from a DC-3 Ž
X X
. aircraft at different levels over Pune 18832 N, 73851 E, 559 m MSL region during
different summer monsoon seasons. The experimental site near Pune is about 150 km from the nearest West Coast on the Arabian Sea. Winds are mostly westerly. The clouds
over the Pune region are modified maritime. In this paper, the measurements of the spectra in cumulus clouds, 1–3 km thick, made in the years 1984, 1985 and 1986 in
Ž .
non-raining clouds at different levels 100–1000 m above cloud base and for different Ž
. cloud thicknesses ranges of vertical extent 0.3–1.2 km and 1.4–3.0 km have been
studied. Also, those measured on the days with rain and on the days with no rain in the experimental area in the years 1976, 1979 and 1980 have been compared, the occur-
Ž rencernon-occurrence of rain being noticed visually within the clouds on board the
. aircraft . The cloud base level was about 1400 m MSL. The measured updraftsrdown-
drafts were 5 m s
y1
and the lapse rate 0.4–0.6r100 m in the clouds. In general, the samples were collected towards the center of a horizontal cross-section of the cloud that
was 6–10 km. For different levels, 200 m averages were taken, as the number of samples was limited.
Ž .
A mechanical semi-automatic cloud drop sampler Srivastava and Kapoor, 1960 Ž
. containing magnesium oxide-coated glass slides effective size 4.4 cm = 1.0 cm was
used for sampling the cloud drops. Each slide was exposed for 14.8 ms in the cloud at an impact speed of 54 m s
y1
. The crater sizes were measured using an optical microscope of magnification 80 = and 400 = for cloud drop diameters 20 mm and
Ž .
F 20 mm, respectively. The exposure was made in cloud air away about 50 cm from the aircraft. The volume of cloud sampled for each exposure was about 300 cm
3
. True drop sizes were obtained from proper calibrations made earlier of the ratio of actual to
Ž .
crater size, for different cloud drop sizes, in the oxide layer 15–20 mm thick . The Ž
. details of calibrations were reported elsewhere Paul et al., 1980 . The craters with mean
diameters 11.5–241.2 mm corresponded to cloud drops 5.7–90.0 mm diameter. Correc- Ž
tions were made in respect of collection efficiencies for different drop sizes Ranz and
. Wong, 1952 . The collection efficiencies were 43–98 for cloud drops 4–18 mm at an
air speed of 54 m s
y1
.
3. Results and discussion
