Properties of Clouds and Cloud Systems
195
been completed (Greenwald et al., 1993; Liu and Curry, 1992, 1993; Lin and Rossow, 1994). Ice
crystals in clouds are too small to affect microwave emissions; but combining reflected sunlight
(or IR spectra) with microwave can be used to estimate ice water path from satellites (Lin and
Rossow, 1994). Lidar and millimeter radar scattering can also be used to estimate ice water
path (Sassen et al., 1989; Kropfli, 1995): lidar is most sensitive to very low optical thicknesses
« 0.5), whereas radar remains useful up to very large values (> 60). These techniques are
more sensitive to particle sizes, however. Radar has not been used on satellites yet; one lidar
has flown on the Shuttle (McCormick et al., 1995).
Cloud particle sizes (dimensions in the range from 6-60 {tm) have been determined for liquid
water clouds on Earth from satellite measurements of the wavelength dependence of reflected
sunlight (Han et al., 1994). IR spectra can also be used, especially for the larger size particles
occurring in ice clouds (Carlson et al., 1993), but this has not been done for Earth observations.
Size information for precipitation-sized particles (dimensions> 500 {tm) can be determined from
centimeter radar scattering, but this has not been done from satellites yet (an experimental
mission is planned for 1997: the Tropical Rainfall Measuring Mission).
Retrieval of particle sizes is especially sensitive to particle shape when using observations of
reflected sunlight; hence, measurements of the angular distribution of reflected intensity and
polarization can be used, in principle, to infer particle shape but this technique has not been
tried before.
There are many different methods for determining cloud top location from satellite measurements. Two that have not been used for Earth observations but have been used on other planets
are measuring the strength of gas absorption features, giving the amount of gas over the cloud,
and measuring the amount of Rayleigh scattering from the polarization of reflected sunlight,
giving the pressure at cloud top (Travis et al., 1978). Timing the return pulses from lidar and
radar gives a more direct estimate of cloud top height, but only the former has recently been
attempted from space (LITE mission on Shuttle). Two more extensively used methods are
determining heights at which sunlight is extinguished when the Earth's limb occults the sun
(Woodbury and McCormick, 1986; Kent et al., 1993; Wang et al., 1995) and determining cloud
top temperature from IR emissions (Rossow et al., 1989; Wylie and Menzel, 1989; Wylie et al.,
1994).
Cloud base location can sometimes be determined from the spectrum of IR emission, if cloud
optical thickness is low enough (Minnis et al., 1995). Also, lidar can determine cloud base if it
can penetrate the whole cloud layer (Sassen et al., 1989). Millimeter radar can penetrate much
thicker clouds and would provide a more general survey of cloud vertical structure, but such
an instrument has never been flown on a satellite.
9.3 Summary of Observed Cloud Properties
Clouds exhibit a very wide range of properties from the thinnest wisps of ice cirrus clouds at
15-20 km altitude or small clumps of liquid cumulus clouds near the surface to the violent
storm clouds that extend throughout the whole depth of the troposphere and produce heavy
rainfall. Cirrus have very little water content, occupy thin layers, and may have patchy horizontal coverage. Storm cloud systems have nearly the maximum possible water content, occupy
multiple or very deep layers, and can cover areas more than 1000 km across. Table 9.2 lists
cloud properties in order of decreasing significance in determining microphysical processes in
clouds, including precipitation, and in determining radiative transfer in clouds. Figures 9.1-9.4
summarize the distributions of some of these properties obtained from extensive surveys of
weather observations and satellite remote sensing.
195
been completed (Greenwald et al., 1993; Liu and Curry, 1992, 1993; Lin and Rossow, 1994). Ice
crystals in clouds are too small to affect microwave emissions; but combining reflected sunlight
(or IR spectra) with microwave can be used to estimate ice water path from satellites (Lin and
Rossow, 1994). Lidar and millimeter radar scattering can also be used to estimate ice water
path (Sassen et al., 1989; Kropfli, 1995): lidar is most sensitive to very low optical thicknesses
« 0.5), whereas radar remains useful up to very large values (> 60). These techniques are
more sensitive to particle sizes, however. Radar has not been used on satellites yet; one lidar
has flown on the Shuttle (McCormick et al., 1995).
Cloud particle sizes (dimensions in the range from 6-60 {tm) have been determined for liquid
water clouds on Earth from satellite measurements of the wavelength dependence of reflected
sunlight (Han et al., 1994). IR spectra can also be used, especially for the larger size particles
occurring in ice clouds (Carlson et al., 1993), but this has not been done for Earth observations.
Size information for precipitation-sized particles (dimensions> 500 {tm) can be determined from
centimeter radar scattering, but this has not been done from satellites yet (an experimental
mission is planned for 1997: the Tropical Rainfall Measuring Mission).
Retrieval of particle sizes is especially sensitive to particle shape when using observations of
reflected sunlight; hence, measurements of the angular distribution of reflected intensity and
polarization can be used, in principle, to infer particle shape but this technique has not been
tried before.
There are many different methods for determining cloud top location from satellite measurements. Two that have not been used for Earth observations but have been used on other planets
are measuring the strength of gas absorption features, giving the amount of gas over the cloud,
and measuring the amount of Rayleigh scattering from the polarization of reflected sunlight,
giving the pressure at cloud top (Travis et al., 1978). Timing the return pulses from lidar and
radar gives a more direct estimate of cloud top height, but only the former has recently been
attempted from space (LITE mission on Shuttle). Two more extensively used methods are
determining heights at which sunlight is extinguished when the Earth's limb occults the sun
(Woodbury and McCormick, 1986; Kent et al., 1993; Wang et al., 1995) and determining cloud
top temperature from IR emissions (Rossow et al., 1989; Wylie and Menzel, 1989; Wylie et al.,
1994).
Cloud base location can sometimes be determined from the spectrum of IR emission, if cloud
optical thickness is low enough (Minnis et al., 1995). Also, lidar can determine cloud base if it
can penetrate the whole cloud layer (Sassen et al., 1989). Millimeter radar can penetrate much
thicker clouds and would provide a more general survey of cloud vertical structure, but such
an instrument has never been flown on a satellite.
9.3 Summary of Observed Cloud Properties
Clouds exhibit a very wide range of properties from the thinnest wisps of ice cirrus clouds at
15-20 km altitude or small clumps of liquid cumulus clouds near the surface to the violent
storm clouds that extend throughout the whole depth of the troposphere and produce heavy
rainfall. Cirrus have very little water content, occupy thin layers, and may have patchy horizontal coverage. Storm cloud systems have nearly the maximum possible water content, occupy
multiple or very deep layers, and can cover areas more than 1000 km across. Table 9.2 lists
cloud properties in order of decreasing significance in determining microphysical processes in
clouds, including precipitation, and in determining radiative transfer in clouds. Figures 9.1-9.4
summarize the distributions of some of these properties obtained from extensive surveys of
weather observations and satellite remote sensing.