Stratus, stratocumulus, altostratus, and cirrus clouds are thought to have the greatest effect on climate due to the large area they cover. In this section we cover recent studies on cirrus and altostratus clouds and their potential impact on climate. Radiative aspects of these clouds are considered in more detail in the review of atmospheric radiation by Stephens [1995] in this issue.
One of the early results of microphysical and radiative studies of cirrus clouds was the large sensitivity of radiative transfer calculations and associated optical depth and cloud reflectance to the ice particle shape and size distribution, with high concentrations of small hexagonal ice crystals leading to more reflectance of the incoming visible solar radiation than equivalent spheres (Stephens et al. [1990]). Prior to this review period, it was felt that the simulation of cirrus cloud feedback on climate was both premature and limited by our lack of understanding of the relationship between size and shape of ice crystals and the gross radiative properties of cirrus (Stephens et al. [1990]). Results during the last four years have increased our understanding of the size and shape of ice crystals in cirrus clouds, as well as our understanding of radiative transfer (Stackhouse and Stephens [1991], Baum et al. [1992], Takano et al. [1992], Kinne et al. [1992], Minnis et al. [1993a], Minnis et al. [1993b], Fu and Liou [1993]). For instance, the radiative transfer parameterization of Fu and Liou [1993] showed that for a given ice water path, cirrus clouds with smaller mean effective sizes reflect more solar radiation, trap more infrared radiation (IR), and produce stronger cloud-top cooling and cloud-base heating. They also show that in most cases the IR greenhouse effect outweighed the solar albedo effect, except when significant numbers of small ice crystals were present. The radiative transfer model developed by Minnis et al. [1993] showed best results when using hexagonal ice crystals of 20 micron diameter and a size distribution of slightly larger hexagonal ice crystals representing cirrostratus cloud. They also showed that the interpretation of cirrus reflectances with water-droplet models leads to biased results. Kinne et al. [1992] compared modeled radiative fluxes with those measured at various altitudes and found that the model results underestimated the solar reflectivity and attenuation, as well as the downward infrared fluxes. Reconciling the model results with the measurements could only be achieved by adding large concentrations of small ice crystals or by altering the backscattering characteristics of the crystals. This study suggests that additional studies on the backscatter characteristics of small crystals of various habits are needed, as well as new aircraft instrumentation to detect ice crystals between 5 and 50 microns diameter.
Gultepe and Rao [1993] calculated the moisture and heat budgets of a cirrus cloud from aircraft measurements during FIRE and showed that the maintenance of cirrus can be attributed to relatively warm and moist advection, radiative cooling at upper levels, and moisture advection in the vertical. Turbulent heat and moisture fluxes were found to be important in the low levels of cirrus. Heymsfield et al. [1991] analyzed aircraft data collected on two altocumulus clouds from FIRE and found that the clouds were structurally and thermodynamically similar to stratocumulus clouds, with extensive cloud top entrainment, a capping temperature inversion, and a dry layer above. A numerical study including radiative transfer calculations suggestsed that radiation played an important role in driving convection in the more dynamically unstable of the two clouds.
As mentioned earlier, Heymsfield and Miloshevich [1993] investigated
ice nucleation mechanisms in cold lenticular wave clouds and found that
homogeneous nucleation was responsible for the ice production in these
clouds at temperatures below -33
C. A similar nucleation mechanism
was hypothesized to exist for cirrus clouds as well.
The effect of volcanic aerosol on cirrus cloud microphysics and radiative
transfer was discussed by Sassen [1992] and Jensen and Toon [1992]. Sassen
[1992] found supercooled droplets in cirrus uncinus cell heads between -40
and -50
C that could only be explained by the freezing point depression
of sulfuric acid droplets typically produced as a result of volcanic
eruptions. Jensen and Toon [1992] examined the potential effect of volcanic
aerosols on cirrus cloud microphysics, and suggested that at temperatures
below -50
C the concentration of ice crystals which nucleate may be
as much as a factor of 5 larger when sulfuric acid droplets of volcanic
origin are present. They also suggested that this effect may increase the
surface warming created by certain types of cirrus clouds near the tropopause
by as much as 8 Wm
.
A number of new remote sensing techniques to observe cirrus clouds were proposed, including multiparameter studies using K-band (8.6 mm) radars (Matrosov [1991], Matrosov and Kropfli [1993]) and multi-wavelength methods using lidar and radars (Intrieri et al. [1993]). Both of these methods showed good potential for characterizing cirrus cloud particle sizes and shapes.
During the past four years, cirrus cloud microphysics and associated radiative transfer has received a significant amount of attention. While a number of issues have been resolved, more work needs to be done especially in characterizing the cirrus cloud particles less than 50 microns in diameter and their associated radiative transfer. This was also identified by Cooper [1991] as a critical issue in the last cloud physics review. As discussed above, the shape and concentration of these small ice particles may be critical in determining the role cirrus clouds play in determining the earth's future climate.