Response to Forcing of the Climate System

Over a time span of a few years the heat balance of the earth can generally be considered to be in balance, which means that the incoming solar radiation, S, is balanced by the outgoing long wave radiation, F. What happens then when suddenly there is a change in either S or F? Let us assume, for example, that there is a sudden increase in C02 concentration to twice the present value.

The immediate response is a reduction in the outgoing longwave radiation at the tropopause of about 3.1 W nT2 and an increase in the downward emission from the stratosphere by about 1.3 W nT2. The sum of the two, 4.4 W nT2, is the net instantaneous forcing at the tropopause.

Following this immediate shock the stratosphere cools. The increased C02 in the stratosphere enhances the thermal emission. Because the stratospheric temperature increases with altitude, this has the effect that the cooling into space is larger than the absorption from layers below. This is in fact the fundamental reason for the CO,-induced cooling in the stratosphere. After stratospheric cooling a new radiative equilibrium develops with the new doubled CO, concentration. This reduces the increased downward emission at the tropopause by about 0.2 W m~2 and the tropopause forcing is adjusted accordingly.

The surface-troposphere system will continue to warm until the entire system reaches a new equilibrium. This may take a considerable time due to the very high heat capacity of the ocean and it will certainly last several decades before an equilibrium is reached, if at all.

Why does the surface-troposphere system warm at all, since in the end the radiation emission from the earth must balance the incoming solar radiation which stays the same? The reason is that the negative vertical temperature gradient in the troposphere has the effect that the equivalent level of outgoing radiation is successively lifted and the levels below are warmed due to hydrostatic influences (Fig. 6). If there were no vertical temperature gradients in the atmosphere, the surface emission would be equal to the outgoing emission at the top of the atmosphere and the greenhouse effect would consequently disappear.

However, this cannot happen in the present atmosphere so the direct warming effect at the surface, assuming no feedback, would amount to about 1.3 K (Ramanathan, 1981). Now it appears that the atmosphere is close to conserving relative humidity, so a warming would increase the water vapor in the atmosphere and hence further increase the warming, thus creating a positive feedback effect. It is interesting to note that even Arrhenius (1896) included the feedback from water vapor.

/ ✓

/ Control (1 x C02 )

\ \ ."'

,,2x C02 raises mean level of emission to space

\\ . Slope ~ 5.5 C km 1

| V

.Surface temperature increase -►

255 288 T F'Kl

FIGURE 6 Illustration of the greenhouse effect. The height of the equivalent outgoing radiation is around 6 km with a temperature of ca. 255 K (global average). A doubling of the C02 will raise the height of the outgoing radiation by a few hundred meters and thus warm the surface accordingly (extrapolated via an averaged lapse rate of 5.5°Ckm_1).

255 288 T F'Kl

FIGURE 6 Illustration of the greenhouse effect. The height of the equivalent outgoing radiation is around 6 km with a temperature of ca. 255 K (global average). A doubling of the C02 will raise the height of the outgoing radiation by a few hundred meters and thus warm the surface accordingly (extrapolated via an averaged lapse rate of 5.5°Ckm_1).

Empirical studies (Hense et al., 1988; Flohn et al, 1989; Raval and Ramanathan, 1989; Gaffen et al., 1991; Inamdar and Ramanathan, 1998) show that temperature and water vapor changes are positively correlated and so are results from model studies (Manabe and Wetherald, 1967; Mitchell, 1989). In summary, it has been shown that both studies by simple models and GCMs and observations from independent sources (Inamdar and Ramanathan, 1998) all converge in the range of a positive feedback factor of 1.3-1.7 from water vapor. The only deviating results are those from Lindzen (1990, 1994), which suggest a negative feedback with water vapor due to a drying out effect of the upper troposphere caused by enhanced deep convection.

Inamdar and Ramanathan (1998) have shown that there are considerable geographical variations in water vapor feedback, with the dominating effect in the equatorial ocean region. In this area the greenhouse feedback exceeds the blackbody emission, reproducing the so-called super-greenhouse effect (Ramanathan and Collins, 1991). The overall results demonstrate the importance of realistically reproducing the three-dimensional atmospheric circulation and the associated water distribution for a credible water vapor feedback.

While models generally agree in reproducing the water vapor feedback, the cloud feedback is much more complex. The overall effect of clouds is to cool the surface and the troposphere since the albedo effect (reflection of solar radiation) is larger than the enhanced absorption of long-wave radiation by clouds. The difference is substantial and amounts to some 20 W itT2. The change in cloud forcing due to enhanced greenhouse forcing is strongly model-dependent, with some models giving positive feedback and others negative (Cess et al., 1997).

The ECHAM4/OPYC model discussed below has a negative cloud feedback, with the transient integration having a stronger negative cloud feedback than the equilibrium model (Bengtsson, 1997). The cloud feedback depends, though, to a considerable degree on changes of the lower boundary. Clouds over open water (more common in a warmer climate) have a strong negative forcing, while clouds over sea ice and snow (more common in a cold climate) generate practically no feedback because of similar albedo.

Surface processes such as the melting of snow and ice at higher temperatures will decrease the surface albedo, leading to a positive feedback, while changes in cloud cover and cloud distribution can give rise to either a negative or a positive feedback. Other feedback processes depend on changes in the general circulation, such as those in the dominating storm tracks and in the vertical stability of the atmosphere, affecting the surface temperature. For this reason, as will be demonstrated below, it is not possible to infer from a certain forcing pattern what the climate response would be. This is one of the reasons realistic climate models must be used in such an evaluation. This can be illustrated by comparing the geographical distribution of forcing here taken from the Hamburg climate model (Roeckner et al, 1999) and the corresponding temperature change (Fig. 7 and Table 1). The actual forcing was taken from an equilibrium climate change experiment including the anthropogenic effect

60N-

60N-

Climate 60n And 60s

60N-

30n-

60N-

30n-

30s"

60s"

120e

30s"

60s"

120W

120e

0 0.5 1 1.5 2 [°C] FIGURE 7 (a) Radiative forcing from greenhouse gases, sulfate aerosols (direct and indirect effect), and tropospheric ozone from the anthropogenic emission during 1860-1990. See also Table 1. In the Northern Hemisphere there are widespread areas with negative forcing caused by sulfate aerosols, (b) Equilibrium response calculated from the ECHAM4 coupled to a slab ocean and averaged over 20 years. Note the differences between the forcing and the response pattern. For further information see Roeckner etal. (1999).

TABLE 1 Global Annual Mean Radiative Forcing at the Top of the Tropopause and Equilibrium Response in Global Annual Mean Surface Air Temperature*

Experiment No.

I Iistorical Forcing Experiments

Radiative Forcing

Temperature Response

Climate Sensitivity

(i 860-i 990)

[Wmi

(°C)

(°C/W n-T2)

1

Well-mixed greenhouse gases

2.12

1.82

0.86

(CO,, CHj, N,0, CFCs)

(2.45)"

2

Tropospheric ozone

0.37

0.34

0.91

¡0.2 to 0.6)

3

Direct sulfate aerosol

-0.34

-0.24

0.71

( 0.2 to-0.8)

4

Indirect sulfate aerosol

-0.89

-0.78

0.87

(Oto - 1.5)

Sum (1 to 4)

1.26

1.15

0.91

5

Effects (1 to 4) included

1.26

1.13

0.90

* Forcing data in brackets indicate range of forcing provided by IPCC. 'IPCC value from 1750 to 1994

* Forcing data in brackets indicate range of forcing provided by IPCC. 'IPCC value from 1750 to 1994

of the well-mixed greenhouse gases, sulfate aerosols, and tropos-pheric ozone from the beginning of the industrialization until present. As can be seen, there is practically no correlation between the pattern of forcing and the pattern of temperature response. The areas of net negative forcing over large parts of Eurasia, for example, are becoming significantly warmer. The reason is that warming from other regions, such as from the tropical oceans, transports heat toward the higher latitudes and thus gives rise to a warmer climate.

It follows from this discussion that climate response to external forcing is rather complex and hence, as can be seen from a recent study by Le Treut and McAvaney (1999), strongly model-dependent. Figure 8 shows the equilibrium response in global surface temperature and precipitation to a doubling of C02 for 11 different "state-of-the art" climate models. As can be seen, the temperature increase varies between 2.1 and 4.8 K and the pre

FIGURE 8 Equilibrium response to 2 X CO: for 11 GCM coupled to a mixed layer ocean. For further information see text (after Le Treut and McAvaney, 1999).

Surface warming [ C]

FIGURE 8 Equilibrium response to 2 X CO: for 11 GCM coupled to a mixed layer ocean. For further information see text (after Le Treut and McAvaney, 1999).

cipitation between 1 and 15%. It can further be seen that the increase in precipitation as a function of temperature is significantly less than that from the Clausius-Clapeyrons equation. The reason is that global precipitation must balance global evaporation. Global evaporation in turn is controlled by the net radiative forcing at the ground, which apparently increases more slowly than the availability of moisture in the free atmosphere.

In conclusion, we must still count on considerable inaccuracy even in such general quantities as the change in global average temperature and precipitation — and this is when the forcing of climate is known exactly!

Was this article helpful?

0 0

Post a comment