38
A.D. Del Genio
a) Regional cloud regimes.
The globe can be divided into regions in which different cloud types dominate. This is really a
segregation into different climate regimes, because cloud types typically follow properties of the
temperature and water vapor profiles and the dynamics. Cloud regimes are conveniently identified by two-dimensional frequency distributions of cloud top pressure and optical thickness,
such as those shown in Figure 2.4. Each regime defines a particular set of controlling physical
processes, so if we can understand why a particular cloud type prevails in a given regime, we
may be able to anticipate how it should change as the climate is perturbed. Such regions
need not be geographically contiguous; they merely must share common important physical
processes. At a minimum, probably 5 different cloud regimes can be identified in this way (see
Figures 2.4 and 2.5):
- Tropical convection zones.
This regime includes the tropical West Pacific, South Pacific Convergence Zone, Amazon Basin,
central Africa, and the maritime continent. As indicated by the right panel in Figure 2.4, it is
characterized by high, optically thick clouds (deep convective towers and associated anvils) and
low-level boundary layer stratus. The anvil clouds generally control the TOA radiative fluxes
in this regime. The central question for climate change is whether the strength of convection
will increase in a warmer climate, and if so, how it will affect anvil radiative properties (cf. Ye
et al., 1996). Some insight might be gained by comparing the radiative properties of maritime
and continental convective clusters, which probably differ in the strength of cumulus updrafts.
Anvil properties depend on detrainment of liquid and ice from convective updrafts, but the
dynamics and microphysics of this process is represented crudely if at all in current GCMs.
Another issue for these clouds is the correct particle size and phase function for ice crystals
(Minnis et al., 1993), and whether the microphysical properties of these ice clouds differ from
thin cirrus.
-Subtropical/tropical ocean subsidence regions.
These areas, off the west coasts of North America, South America, Africa, and Australia, are
dominated by low-level marine stratus, capped by a strong trade inversion. The radiative
and microphysical properties of these clouds have been relatively well-defined by several field
programs ( FIRE, ASTEX), and their typical single layer, pure liquid nature allows them to
be monitored more easily from space than the other regimes. The great uncertainty is the
question of the mechanisms that cause the transition from these almost unbroken cloud decks
to the more scattered, deeper fields of trade cumuli that prevail equatorward and west of
these zones. Possible contributors to marine stratus breakup include detachment of the cloud
layer from the surface due to stabilization of the sub cloud layer by solar heating or drizzle
evaporation, and cloud top entrainment instability (cf. Wang and Albrecht, 1994). A related
question is the cause of the ISCCP-observed tendency for the optical thickness of these clouds
to decrease with temperature (Tselioudis et al., 1992). If marine stratus breakup and thinning
at higher temperatures are indicative of what to expect in a climate change, then a region of
positive cloud feedback may exist. Most GCMs underpredict cloudiness in these regions (cf.
Fig. 2.2, bottom).
-Subtropical/midlatitude continents.
This may actually be several separate regimes, encompassing both the relatively cloud-free
subtropical deserts and the cloudier midlatitude continents, which themselves vary with the
seasons. In winter, midlevel optically thick and moderate optical thickness clouds (e.g., nimbostratus and altostratus) are common, along with thin frontal cirrus and low stratus, in the
United States and Europe. In summer, both shallow and deep cumulus increase in importance, but the other cloud types are still evident. Some continental regions such as Siberia
in winter have more in common with the subtropical deserts than with the more populated
A.D. Del Genio
a) Regional cloud regimes.
The globe can be divided into regions in which different cloud types dominate. This is really a
segregation into different climate regimes, because cloud types typically follow properties of the
temperature and water vapor profiles and the dynamics. Cloud regimes are conveniently identified by two-dimensional frequency distributions of cloud top pressure and optical thickness,
such as those shown in Figure 2.4. Each regime defines a particular set of controlling physical
processes, so if we can understand why a particular cloud type prevails in a given regime, we
may be able to anticipate how it should change as the climate is perturbed. Such regions
need not be geographically contiguous; they merely must share common important physical
processes. At a minimum, probably 5 different cloud regimes can be identified in this way (see
Figures 2.4 and 2.5):
- Tropical convection zones.
This regime includes the tropical West Pacific, South Pacific Convergence Zone, Amazon Basin,
central Africa, and the maritime continent. As indicated by the right panel in Figure 2.4, it is
characterized by high, optically thick clouds (deep convective towers and associated anvils) and
low-level boundary layer stratus. The anvil clouds generally control the TOA radiative fluxes
in this regime. The central question for climate change is whether the strength of convection
will increase in a warmer climate, and if so, how it will affect anvil radiative properties (cf. Ye
et al., 1996). Some insight might be gained by comparing the radiative properties of maritime
and continental convective clusters, which probably differ in the strength of cumulus updrafts.
Anvil properties depend on detrainment of liquid and ice from convective updrafts, but the
dynamics and microphysics of this process is represented crudely if at all in current GCMs.
Another issue for these clouds is the correct particle size and phase function for ice crystals
(Minnis et al., 1993), and whether the microphysical properties of these ice clouds differ from
thin cirrus.
-Subtropical/tropical ocean subsidence regions.
These areas, off the west coasts of North America, South America, Africa, and Australia, are
dominated by low-level marine stratus, capped by a strong trade inversion. The radiative
and microphysical properties of these clouds have been relatively well-defined by several field
programs ( FIRE, ASTEX), and their typical single layer, pure liquid nature allows them to
be monitored more easily from space than the other regimes. The great uncertainty is the
question of the mechanisms that cause the transition from these almost unbroken cloud decks
to the more scattered, deeper fields of trade cumuli that prevail equatorward and west of
these zones. Possible contributors to marine stratus breakup include detachment of the cloud
layer from the surface due to stabilization of the sub cloud layer by solar heating or drizzle
evaporation, and cloud top entrainment instability (cf. Wang and Albrecht, 1994). A related
question is the cause of the ISCCP-observed tendency for the optical thickness of these clouds
to decrease with temperature (Tselioudis et al., 1992). If marine stratus breakup and thinning
at higher temperatures are indicative of what to expect in a climate change, then a region of
positive cloud feedback may exist. Most GCMs underpredict cloudiness in these regions (cf.
Fig. 2.2, bottom).
-Subtropical/midlatitude continents.
This may actually be several separate regimes, encompassing both the relatively cloud-free
subtropical deserts and the cloudier midlatitude continents, which themselves vary with the
seasons. In winter, midlevel optically thick and moderate optical thickness clouds (e.g., nimbostratus and altostratus) are common, along with thin frontal cirrus and low stratus, in the
United States and Europe. In summer, both shallow and deep cumulus increase in importance, but the other cloud types are still evident. Some continental regions such as Siberia
in winter have more in common with the subtropical deserts than with the more populated