1,58
R. Stuhlmann
As shown in Table 7.2, the quality of clouds, to almost perfectly absorb longwave radiation
and at the same time to excellently reflect shortwave radiation, produce opposite effects on
the radiation balance TOA. This effect of cancellation can be illstrated by a very simplified
model taking into account the decrease in temperature with altitude in the atmosphere and
the increase in reflection as a function of cloud albedo. On the condition of global annual
average solar insolation and assuming an unrealistic atmosphere without longwave atmospheric
absorption above the cloud tops, C F may be approximated as follows:
C F = - So ~a + OLR elear - arTs _ rZCT )4
4
(7.6)
Here, ~a is the difference in albedo between the overcast- and the clear-sky and a(Ts - r ZCT)4
the approximated outgoing longwave radiation for an overcast cloud situation with its top at
a height ZCT, where Ts and r are the surface temperature and the atmospheric lapse rate,
respectively. For numeric calculations a value of 265 Wm- 2 can be taken from Table 7.2 for
OIRciear, where the surface temperature and lapse rate can be set to climatic values of 288 K
and 6.5 K/km, respectively.
Figure 7.5 presents results calculated from Equation 7.6. As can be seen, the longwave effect
is dominating above regions where the change in albedo is small, as for instance above deserts.
The opposite is true for dark surfaces like oceans, where the change in albedo between an
overcast and a clear-sky is largest. In generaL Figure 7.5 shows that the effect of an increase in
cloud albedo can be compensated by an increase in cloud top height. Since the assumption of
a non absorbing atmosphere above the cloud top is unrealistic, in particular for low cloud tops,
the range of validity of Figure 7.5 starts at cloud tops which are approximately higher than 5
km.
The regional distributions of the longwave, shortwave and net cloud forcing, determined for the
month April 1985 from ERBE data, are shown in Figures 7.6 to 7.8. The longwave cloud forcing
is largest with heatings of C FLW > 60 Wm- 2 above the continents within the ITCZ. Secondary
maxima of C F LW > 40 Wm- 2 are found in the regions of the mid-latitude storm tracks. These
all are areas where upper-level clouds are present, which strongly modulate the longwave part
of the energy balance TOA as discussed with Figure 7.5. Since these clouds are also optical
thick and, thus, bright, the general pattern of the shortwave cloud forcing is similar to the
longwave one, but with the opposite sign. Areas of strongest cooling with values of C Fsw <
-70 Wm- 2 are found for the mid-latitude storm tracks within the northern hemisphere and
for the ICTZ, while for the southern storm track a somewhat smaller cooling of C Fsw < -50
Wm- 2 is found. Since both spectral components have opposite signs and, thus, tend to cancel
each other the general pattern ot the net cloud forcing is different from the two former ones.
We still see that the largest net cloud forcing with a cooling of C F < -60 Wm- 2 is connected
to the northern storm tracks. Other striking areas of negative net cloud forcing are those with
the prevailing low-level stratus clouds located off the west coast of South America and Africa
with CF < -30 Wm- 2 , caused by the strong albedo but small temperature contrast between
clouds and ocean. On the contrary, above the Sahara desert C F is positive with values larger
than 5 Wm- 2 , which can be attributed to the small albedo contrast above deserts.
Figure 7.9, Hartmann (1993), presents the zonal and seasonal averages of C FLW , C Fsw and
C F. The curves of the zonal average net cloud forcing show the importance of mid-latitude
cloudiness, which especially during the summer months produce a strong shortwave cooling
much larger than the corresponding longwave heating. As a result a strong seasonal dependence
of C F is found polewards of about. ± 30° with peak values at about ± 60° forced by the seasonal
variation in solar insolation. In the convective regions of the tropics, the large shortwave and
longwave components more of less cancel each other. The resulting net cloud forcing is negative
R. Stuhlmann
As shown in Table 7.2, the quality of clouds, to almost perfectly absorb longwave radiation
and at the same time to excellently reflect shortwave radiation, produce opposite effects on
the radiation balance TOA. This effect of cancellation can be illstrated by a very simplified
model taking into account the decrease in temperature with altitude in the atmosphere and
the increase in reflection as a function of cloud albedo. On the condition of global annual
average solar insolation and assuming an unrealistic atmosphere without longwave atmospheric
absorption above the cloud tops, C F may be approximated as follows:
C F = - So ~a + OLR elear - arTs _ rZCT )4
4
(7.6)
Here, ~a is the difference in albedo between the overcast- and the clear-sky and a(Ts - r ZCT)4
the approximated outgoing longwave radiation for an overcast cloud situation with its top at
a height ZCT, where Ts and r are the surface temperature and the atmospheric lapse rate,
respectively. For numeric calculations a value of 265 Wm- 2 can be taken from Table 7.2 for
OIRciear, where the surface temperature and lapse rate can be set to climatic values of 288 K
and 6.5 K/km, respectively.
Figure 7.5 presents results calculated from Equation 7.6. As can be seen, the longwave effect
is dominating above regions where the change in albedo is small, as for instance above deserts.
The opposite is true for dark surfaces like oceans, where the change in albedo between an
overcast and a clear-sky is largest. In generaL Figure 7.5 shows that the effect of an increase in
cloud albedo can be compensated by an increase in cloud top height. Since the assumption of
a non absorbing atmosphere above the cloud top is unrealistic, in particular for low cloud tops,
the range of validity of Figure 7.5 starts at cloud tops which are approximately higher than 5
km.
The regional distributions of the longwave, shortwave and net cloud forcing, determined for the
month April 1985 from ERBE data, are shown in Figures 7.6 to 7.8. The longwave cloud forcing
is largest with heatings of C FLW > 60 Wm- 2 above the continents within the ITCZ. Secondary
maxima of C F LW > 40 Wm- 2 are found in the regions of the mid-latitude storm tracks. These
all are areas where upper-level clouds are present, which strongly modulate the longwave part
of the energy balance TOA as discussed with Figure 7.5. Since these clouds are also optical
thick and, thus, bright, the general pattern of the shortwave cloud forcing is similar to the
longwave one, but with the opposite sign. Areas of strongest cooling with values of C Fsw <
-70 Wm- 2 are found for the mid-latitude storm tracks within the northern hemisphere and
for the ICTZ, while for the southern storm track a somewhat smaller cooling of C Fsw < -50
Wm- 2 is found. Since both spectral components have opposite signs and, thus, tend to cancel
each other the general pattern ot the net cloud forcing is different from the two former ones.
We still see that the largest net cloud forcing with a cooling of C F < -60 Wm- 2 is connected
to the northern storm tracks. Other striking areas of negative net cloud forcing are those with
the prevailing low-level stratus clouds located off the west coast of South America and Africa
with CF < -30 Wm- 2 , caused by the strong albedo but small temperature contrast between
clouds and ocean. On the contrary, above the Sahara desert C F is positive with values larger
than 5 Wm- 2 , which can be attributed to the small albedo contrast above deserts.
Figure 7.9, Hartmann (1993), presents the zonal and seasonal averages of C FLW , C Fsw and
C F. The curves of the zonal average net cloud forcing show the importance of mid-latitude
cloudiness, which especially during the summer months produce a strong shortwave cooling
much larger than the corresponding longwave heating. As a result a strong seasonal dependence
of C F is found polewards of about. ± 30° with peak values at about ± 60° forced by the seasonal
variation in solar insolation. In the convective regions of the tropics, the large shortwave and
longwave components more of less cancel each other. The resulting net cloud forcing is negative