C dSTdt

where dT is the disturbance of the global average surface air temperature; Gf = dT/dQ is the parameter describing the sensitivity of the climatic system to variations of heat input 5Q; c is the heat capacity of the climatic system.

For a step-wise change of SQ, and 5T being equal to zero at the initial time instant, the solution of Equation (6.6.1) has the form ST = <5Feq[l — exp( — i/te)], where 5Teq = Gf5Q is the equilibrium temperature disturbance and te = cGt is the relaxation time. As can be seen, the latter depends not only on the heat capacity and, hence, on whether the most inertial link of the climatic system (ocean) is taken into account or not, but also on its sensitivity.

An analysis of the response of the ocean-atmosphere system to an instantaneously changing C02 concentration within the framework of the three-dimensional global circulation models (GCMs) was carried out by Hansen et al. (1984), Schlesinger et al. (1985), Bryan et al. (1982), Spelman and Manabe (1984), Washington and Meehl (1989) and Stoulfer et al. (1989). In the first of the works mentioned the ocean current velocity in the UML was derived from empirical data, and to describe the vertical heat transport in the deep ocean a one-dimensional (in the vertical direction) diffusive model was used which did not consider heat transport by ordered motions. With such assumptions the reorganization of the ocean circulation under the influence of temperature disturbances and accompanying changes in the meridional heat transport are not taken into account. In the other works mentioned above, these restrictions were excluded. But in the second of them the integration of equations of a coupled ocean-atmosphere GCM was carried out only for the first 16 years, and then the calculation was continued within the framework of a zero-dimensional thermodynamic model of the atmosphere and a diffusive box model of the ocean (in such a model the UML is represented in the form of a well-mixed box, and the deep ocean is considered to be at rest so that the heat transport in it is produced only by the vertical eddy diffusion). Thus, the model is able to estimate the equilibrium response of global average values of surface air temperature and sea water temperature at certain depths, but does not permit the detection of local effects. Bryan et al. (1982), and Spelman and Manabe (1984) ran the GFDL model for 50 years after instantaneously doubling C02. Washington and Meehl (1989) performed a similar experiment over a 30-year period with the NCAR model in which the atmospheric submodel was represented by a spectral GCM, and the ocean submodel was represented by a coarse-grid GCM. Stouffer et al. (1989) also used a spectral atmospheric GCM coupled to a coarse-grid ocean GCM with isopycnal diffusion and heat and fresh water flux corrections for a 60-year period. We discuss their results in detail.

To compare non-equilibrium and equilibrium responses we introduce, according to Bryan et al. (1982), the relative temperature deviation r = (T — T0)/(Teq — T0), where T0 and Teq are equilibrium values of temperature for current and increased concentrations of atmospheric C02. Latitude-altitude distributions of the relative temperature deviation in the troposphere and in the upper one-kilometre ocean layer after 25 years following quadrupling of the concentration of atmospheric C02 are shown in Figure 6.7. It can be seen that the distribution of r in the troposphere and UML remains homogeneous

Figure 6.7 Latitude-altitude distribution of the relative deviation r of the zonal average temperature in the troposphere and in the upper 1 kilometre ocean layer after 25 years from the beginning of the integration, according to Spelman and Manabe (1984).

with latitude. The exception is the polar area where sea ice shields the ocean from the atmosphere and the water temperature is close to freezing point. In the vicinity of the sea ice southern boundary (~75°N) the temperature disturbances are localized in the thin UML separated from the deep ocean by a sharp halocline. The existence of the sea ice and the halocline in the polar ocean and a related decrease in heat exchange between the atmosphere and deep ocean favours a rise in heating of the surface atmospheric layer in high latitudes. To the south of 65 °N the temperature disturbances penetrate to greater depths, with the maximum depth of their penetration being limited to the area of cold deep water formation (about 60 °N). To the south of it the value of r does not depend on latitude at the ocean surface and decreases rapidly with depth, approaching zero in the deep layer. Thus, the meridional distribution of the zonal average temperature after 25 years from an instantaneous increase in the concentration of atmospheric C02 closely resembles the equilibrium distribution in the atmosphere and UML but differs markedly from it in the deep ocean, from which it follows that the use of estimates of the equilibrium response when predicting the zonal average surface air temperature is justified only in the case where the characteristic time scale of the external forcing is more than 25 years.

All instantaneous C02 experiments performed within the framework of the coupled GCMs agree with the earlier mixed-layer model experiments. In particular, they show a larger warming at higher latitudes, as well as enhanced drying in the mid-continental regions in summer and an increase in the soil moisture in winter.

Let us now turn to a discussion of the transient response of the climatic system to a gradual increase in the concentration of atmospheric C02. First we note that the basis of most available estimates of potential climate changes, caused by an increase in the concentration of atmospheric C02, is formed by the a priori assumption of weak interaction between the carbon and thermodynamic cycles in the ocean-atmosphere system. There is no conclusive proof of this assumption. Moreover, simple qualitative arguments point to exactly the contrary. Indeed, if the burning of fossil fuels and the destruction of vegetation is accompanied by an increase in the concentration of atmospheric C02 and a temperature rise in certain subsystems of the climatic system, this must lead to a shift in the chemical equilibrium between carbon dioxide dissolved in the surface ocean layer, on the one hand, and bicarbonate and carbonate ions, on the other. The shift in chemical equilibrium must imply a change in intensity of absorption of anthropogenic C02 by the ocean and, hence, the redistribution of C02 between the atmosphere and ocean. Thus, the interaction between carbon and thermodynamic cycles is not in doubt.

This consideration was taken into account by Kagan et al. (1990) when simulating the evolution of the ocean-atmosphere climatic system from the dawn of the industrial revolution until the end of the twenty-first century. The equations of the 0.5-dimensional model (see Section 5.5) used in this work were supplemented by the appropriate equations for the carbon dioxide budget in atmospheric boxes, and for inorganic carbon in ocean boxes, as well as by expressions for the carbon eddy flux at the low boundary of the UML and for the carbon equivalent flux at the upper boundary of the deep layer, by expressions for the C02 flux at the ocean-atmosphere interface, and by hydrochemical relationships describing the behaviour of C02 in solution. Moreover, it was assumed that the boundary between the northern and southern atmospheric boxes is fixed (coincident with a circle of latitude 60 °N) and that the source Rf of atmospheric C02 is determined only by the burning of fossil fuels, and the influence of the biotic sources and sinks of C02 is manifested only by means of an increase in the sink Rp of atmospheric C02 due to a rise in production of terrestrial biomass, and does not have an effect on the vital functions of the terrestrial and marine biota. Then, assuming that all carbon dioxide emission is concentrated in the southern box we obtain Rn = (PcoJHc)(dN/s$ dt), R{2 = 0, Rpi = (^)0[(1 + P In(cf/4)], i = 1, 2, where dN/dt = rNJ(KJN0) - 1)"1 ert/[l + ((N.JN0) - 1)"1 ert]2, r = 0.03 1/year, N0 = 4.5 x 109 tC, = 5000 x 109 tC, and (R^)0 are the concentration and sink of C02 in the pre-industrial period subject to the influence of the natural seasonal variability, and /? = 0.3. We note that the expression for the production, diV/dt, of anthropogenic C02 was chosen only from considerations of convenience of its realization and the magnitude of the factor /? to provide satisfactory agreement between calculated and observed (during the monitoring period) changes in the annual mean global average concentration of atmospheric C02.

We begin the discussion of the calculated results with an indication of a method of determining the initial conditions. The initial values of the C02 concentration in the atmosphere and the total carbon in the ocean on 1 January 1860 were found from the steady-state time-periodic solution in the absence of an anthropogenic source of C02 in the atmosphere. The initial values of the mass-weighted average air temperature in the northern and southern boxes and sea water temperature in the polar ocean, in the area of cold deep water formation, and in the area of upwelling were prescribed on the same basis. The annual mean values of these and all other characteristics of the carbon and thermodynamic cycles corresponding to time-periodic solution are shown in the first column of Table 6.3. Calculated deviations of the annual mean climatic characteristics from their initial values relating to different years of the period examined (1860-2100) are indicated in the other columns of the table.

As can be seen, an increase in the concentration of atmospheric C02 causes an enhancement of the absorption of long- and short-wave radiation, a rise in mass-weighted average values of the air temperature and humidity, a decrease in short-wave radiation and an increase in net long-wave radiation at the underlying surface, a domination of the latter over the former and, as a result, a rise in temperature of the surface of land, snow-ice cover and ocean in the area of upwelling, accompanied by an increase in local evaporation and precipitation. As for the temperature in the area of cold deep water formation, this decreases irregularly: sea water temperature falls up to the year 2025, then rises and in the last quarter of the coming century it starts to fall again.

Further, a rise in air temperature in the polar ocean results in a decrease in sensible and latent heat fluxes. This, together with an increase in the net long-wave radiation flux, causes a weakening of the heat transport from the lower to the upper surface of the snow-ice cover, followed by a decrease in the thickness and area of sea ice and the planetary albedo. Snow mass at the sea-ice surface decreases and at the land surface in the northern atmospheric box it increases. The former is connected with a reduction in the sea ice area, the latter with an increase in the precipitation-evaporation difference. An increase in precipitation favours the intensification of run-off during the whole period examined, and growth of the soil moisture content in

Table 6.3. Annual mean values of the climatic characteristics at the beginning of the industrial revolution (1860) and deviations from these in 1985 and as forecast for the twenty-first century

Year

Characteristic

1860

1985

2000

2025

2050

2075

2100

Radiation balance at the upper atmospheric boundary (W/m2):

Absorbed short-wave radiation (W/m2) in the atmosphere:

northern box 45.7 0.2

southern box 91.8 0.3

Net long-wave radiation flux (W/m2) in the atmosphere: northern box southern box

Planetary albedo (%)

Sea ice area (106 km2)

Sea ice thickness (m)

Heat balance (W/m2) of the snow-ice cover in the polar ocean: short-wave radiation flux long-wave radiation flux sensible heat flux latent heat flux heat exchange between upper and lower surfaces heat release due to ice melting phase transitions

Heat balance (W/m2) of the ocean surface in the area of cold deep water formation: short-wave radiation flux 92.6 —1.3 —2.0

long-wave radiation flux —81.7 3.8 5.7

sensible heat flux -189.5 14.0 21.0

heat exchange with lower layers 214.3 —15.5 —23.2

134.6

-1.5

-2.2

— 3.1

-3.6

-6.4

-9.0

148.6

-1.3

-2.0

-3.1

-4.4

-7.5

-10.4

31.40

-0.05

-0.07

-0.08

-0.07

-0.15

-0.29

12.45

-0.66

-0.99

-1.14

-0.53

-1.50

-3.47

2.76

-0.06

-0.10

-0.11

-0.05

-0.15

-0.37

31.2

-0.1

-0.1

-0.2

0.4

0.5

0.2

-24.4

0.9

1.4

2.1

1.5

2.6

4.2

-0.1

0.0

0.0

0.0

-0.1

-0.2

-0.2

-16.3

-0.2

-0.3

-0.5

-0.8

-1.6

-2.7

7.0

-0.3

-0.5

-0.6

-0.7

-1.5

-2.3

2.6

-0.3

-0.5

-0.8

-0.3

0.2

0.8

Heat balance (W/m2) of the ocean surface in the upwelling area: short-wave radiation flux long-wave radiation flux sensible heat flux latent heat flux heat exchange with lower layers

Gas exchange (gC/m2/year) between the ocean and atmosphere: the domain of cold deep water 40.1

formation area the domain of upwelling area — 1.4

Heat balance (W/m2) of the land surface in the northern box; short-wave radiation flux long-wave radiation flux sensible heat flux latent heat flux heat release due to snow melting

Heat balance (W/m2) of the land surface in the southern box; short-wave radiation flux long-wave radiation flux sensible heat flux latent heat flux

Air temperature (K) at the middle level in the atmosphere: northern box 240.67

southern box 258.50

Air humidity (g/kg) at the middle level in the atmosphere: northern box 0.95

southern box 3.28

168.4

-0.2

-0.3

-0.7

-1.0

-1.7

-2.2

-87.1

1.3

1.9

3.7

5.4

8.5

11.2

-0.8

0.0

0.0

0.0

0.0

0.0

0.0

-73.9

-0.9

-1.4

-2.7

-4.0

-6.5

-8.8

-6.6

-0.2

-0.2

-0.3

-0.4

-0.3

227.4

323.2

129.0

227.4

323.2

280.6 30.9

73.7

-0.1

-0.1

-0.2

-0.3

-0.4

0.2

-55.8

0.5

0.7

1.2

1.4

2.7

2.7

-0.6

0.0

0.0

0.0

0.0

-0.1

-0.2

-15.5

-0.4

-0.6

-0.9

-1.0

-2.0

-2.4

1.8

0.0

0.0

0.1

0.1

0.2

0.3

146.9

-0.2

-0.3

-0.6

-0.9

-1.5

-2.0

-99.7

0.8

1.2

2.4

3.5

5.8

7.8

-1.3

0.0

0.0

0.0

0.0

0.0

0.0

-45.9

-0.6

-0.9

-1.8

-2.6

-4.3

Partial pressure (ppm) of C02 in the atmosphere:

northern box 280.3

southern box 280.6

Precipitation rate (mm/day):

northern box 0.96

southern box 2.14

100.5 101.8

203.4 205.9

397.8 401.6

696.5 702.3

1031.5 1038.0

{continued)

Total carbon concentration (10"3 mole COz/l) in the ocean: polar area cold deep water formation area UML in the up welling area deep layer in the upwelling area pH of sea water cold deep water formation area upwelling area

UML thickness (m) in the upwelling area

Heat flux (W/m2) at the upper boundary of the deep layer in the upwelling area

Carbon fluxes (gC/m2 year) at the UML/deep layer interface in the upwelling area: diffusive — 31.3

equivalent —26.7

2.17

0.02

0.02

0.04

0.08

0.13

0.20

2.17

0.02

0.03

0.05

0.09

0.14

0.21

2.05

0.04

0.05

0.10

0.16

0.22

0.26

2.28

0.01

0.01

0.01

0.02

0.03

0.05

8.33

-0.03

-0.04

-0.09

-0.19

-0.35

-0.56

8.31

-0.07

-0.10

-0.18

-0.30

-0.44

-0.56

50.78

-0.03

-0.05

-0.09

-0.14

-0.17

-0.19

16.22

0.08

0.13

0.25

0.59

1.05

1.33

Upwelling velocity (10 7 m/s)

Meridional heat transport (1014 W): heat transport from the upwelling area into the area of cold deep water formation sensible heat transport from the southern into the northern atmospheric box latent heat transport from the southern into the northern atmospheric box

0.73

4.3

6.5

11.7

18.8

24.9

28.3

3.8

5.7

10.2

16.6

22.2

25.4

0.02

0.03

0.04

0.00

-0.01

0.00

9.22

0.38

0.54

0.80

0.33

0.43

0.78

12.73

-0.43

-0.63

-0.51

0.27

0.05

-0.69

3.51

-0.02

-0.03

0.04

0.24

0.31

0.21

Note: positive values of heat due to phase transitions correspond to sea ice formation; negative values correspond to sea ice melting; positive values of gas exchange indicate the absorption of C03 by the ocean; negative values of gas exchange indicate the emission of C02 into the atmosphere.

the northern atmospheric box in the second half of this period. At the end of the current century and in the first quarter of the coming century the soil moisture content in the northern atmospheric box remains practically constant, which is explained by compensation of the effects of evaporation and precipitation. On the other hand, the soil moisture content in the southern atmospheric box reduces due to the predominance of evaporation over precipitation. We also note the tendency to a decrease in temperature of the deep ocean in the area of upwelling and a simultaneous increase in heat transport from the UML into the deep layer. This result, contradicting estimates of the equilibrium response (see Section 6.2), is connected with an increase in temperature contrast between the UML and the deep ocean over time.

The calculated results for the characteristics of the carbon cycle are not unexpected. It can be seen from Table 6.3 that an increase in the concentration of atmospheric C02 is accompanied by an enhancement of gas exchange between the atmosphere and ocean; this, in turn, causes an increase in the total carbon content and a decrease in the pH in the upper ocean layer, as well as an increase in transfer of carbon excess from the UML to the deep layer and, eventually, an increase in the carbon concentration in the deep layer.

Turning back to the discussion of the characteristics of the thermodynamic cycle we will attempt to explain the above-mentioned change in sign of the temperature variations in the area of cold deep water formation. From Figure 6.8, it follows that, from the end of the first half of the coming century, the sea water temperature in this area will be subject to oscillations. Similar

Figure 6.8 Change in the annual mean temperature in the area of cold deep water formation (a); sea ice area (b); and deviations of the surface air temperature (c); and of precipitation (d) in the northern atmospheric box from the beginning of the industrial revolution to the end of the twenty-first century, according to Kagan et al. (1990).

Figure 6.8 Change in the annual mean temperature in the area of cold deep water formation (a); sea ice area (b); and deviations of the surface air temperature (c); and of precipitation (d) in the northern atmospheric box from the beginning of the industrial revolution to the end of the twenty-first century, according to Kagan et al. (1990).

oscillations manifest themselves in the secular changes in the sea ice area, surface air temperature and precipitation. Furthermore they precede sea water temperature oscillations by about 4-9 years, depending on whether the sea ice area increases or decreases. These features can be interpreted in the following way. Let us recall that, when fixing the boundary between boxes, the extent of the area of cold deep water formation was defined as the difference between the ocean area in the northern box and the sea ice area. Because of this a decrease in the sea ice area, along with an increase in the concentration of atmospheric C02, must result in an increase in the extent of the area of cold deep water formation and, hence, to a fall in sea water temperature in this area owing to trapping of colder waters from the polar ocean. But the fall in sea water temperature favours a decrease in the heat transfer from the area of cold deep water formation into the subice ocean layer and thereby enhances stabilization or even an increase in the sea ice area. With an increase in the sea ice area (which occurs when the heat transfer in the subice ocean layer is less than the vertical heat flux in the snow-ice cover), trapping of colder water from the polar ocean into the area of cold deep water formation terminates. This entails a rise in temperature in the area of cold deep water formation, an increase in the heat transport in the subice ocean layer and, as a result, a further decrease in the sea ice area. Then the cycle of decrease and increase in the sea ice area is repeated.

It is likely that this is the nature of autooscillations appearing in the system of the polar ocean and the area of cold deep water formation at the moment in time when the sea ice area reaches a certain value. Allowing for changes in the secular changes in climatic characteristics generated by these autooscillations has an important consequence: it protects the sea ice from total disappearance. This distinguishes the results obtained here from previous ones, according to which an increase in the concentration of atmospheric C02 determines the total disappearance of northern sea ice in the second quarter of the twenty-first century. The estimates presented do not agree with this conclusion. According to them the annual mean sea ice area in the second quarter of twenty first century is reduced to 1.1 x 107 km2 after which it varies against the background of a slower secular trend. As a result it turns out (see Figure 6.8(b)) that at the end of the twenty-first century the annual mean sea ice area will amount to about 9 x 106 km2, that is, it will decrease as compared with its preindustrial value by 3.4 x 106 km2.

The reliability of the estimates can be judged on the basis of a comparison of predicted and actual changes in the concentration of atmospheric C02 during the monitoring period (from 1958 until the present time). As can be seen from Figure 6.9 they show satisfactory agreement between themselves, and their

Figure 6.9 Change in the partial pressure of atmospheric C02, according to Kagan et al. (1990). (a) calculated results in the southern atmospheric box; (b) observational data from the Mauna Loa observatory.

agreement extends not only to inter-annual changes but also to seasonal oscillations of the characteristics examined.

Detailed information on the spatial structure of the climatic system response to a gradual increase in the concentration of atmospheric C02 may be obtained only on the basis of experiments with coupled ocean-atmosphere GCMs. To date, four such experiments have been performed. The first was realized with the NCAR model (Washington and Meehl, 1989); the second, with the modified version of the GFDL model (Stouffer et al., 1989); the third, with the UKMO model (Murphy, 1990); and the fourth, with the MPI model (Cubasch et al, 1991). We note, first, that the equilibrium response of each of these models to doubling C02 is different: when the ocean is represented as a specified-depth mixed layer with no heat transport, the equilibrium sensitivity is 4.5 °C for the NCAR model, 4.0 °C for the GFDL model, and 2.6 °C for the MPI model. Further, these models have different C02 doubling times (100 years for the NCAR model, 70 years for the GFDL and UKMO models, and 60 years for the MPI model). Finally, these models differ among themselves in parametrizations of various physical processes, spatial resolution and the use of some unphysical devices (e.g. correction of heat, fresh water and momentum fluxes at the ocean-atmosphere interface) when coupling the ocean and atmospheric submodels.

But despite these differences all of the coupled ocean-atmosphere GCMs exhibit a number of similar overall features in their transient and spatial responses. So, in all cases the annual mean globally averaged increase in surface air temperature is approximately 60% of the equilibrium warming due to thermal inertia of the deep ocean. All of the models demonstrate markedly smaller warming over oceans than over land in the Southern Hemisphere, compared to the Northern Hemisphere. That could be a result of the stronger heat uptake by the deep ocean. In the Northern Hemisphere the largest warming is found over middle continental areas in summer, and over the Arctic Ocean in winter. Accordingly, this leads to an increase in seasonality over the southern half of North America and south-east Europe and to a reduction in seasonality over the Arctic region. All of the models predict pronounced warming in the deep ocean adjacent to Antarctica and Greenland where the deep convective mixing occurs. In the other parts of the ocean the warming is largely confined to the upper 0.5 km layer. As a consequence, sea surface temperature changes in the northern North Atlantic and around Antarctica are small and even may be negative during the first decade of the simulation.

There is no consensus about changes in the deep water production and ocean circulation regime. Thus, Stouffer et al. (1989), and Washington and Meehl (1989) concluded that increasing surface temperature and precipitation would be accompanied by a decrease in the deep water production as the increased precipitation plus runoff exceed the increased evaporation thereby freshening the water. This does not take place in the Southern Ocean where the deep water formation is governed by a wider set of determining factors. In addition, Stouffer et al. (1989) noted significant weakening of the North Atlantic thermohaline circulation, while Washington and Meehl (1989) did not find this. Both models disagree on the ocean thermohaline circulation response for the Southern Hemisphere.

As regards the hydrological cycle changes, they are similar to those in comparable equilibrium experiments using models with a specified-depth upper mixed layer ocean. In winter, precipitation is generally higher over land areas north of 50 °N and is little changed or decreased further south. In summer, precipitation is lower over most mid-latitude continental areas and increases over Central America and south-east Asia. Correspondingly, soil moisture is greater over mid- to high-latitude northern continents in winter and is lower in summer. Exceptions are the Indian subcontinent and the Mediterranean region. They are described by a summer increase and a winter decrease in soil moisture respectively. The pattern of changes in the Southern Hemisphere is less well defined because of these changes, and their seasonal variations are small, but decreases in precipitation over most of the southern subtropical oceans in winter are apparent.

Assessing the results of the transient C02 experiments with coupled ocean-atmosphere GCMs it should be remembered that three of these (GFDL, MPI and UKMO models) use the procedure of flux correction lest the ocean temperature and salinity be essentially different from those presently observed. This correction causes substantial changes of fluxes comparable in magnitude to the fluxes themselves. Moreover, these changes tend to be of opposite signs to the signs of the fluxes (IPCC, 1992). Therefore, the use of the flux correction entails, in principle, significant distortions in the model response to small perturbations and particularly to perturbations with inter-annual and decadal timescales created by a gradual increase of the atmospheric C02. On the other hand, if the flux correction is not applied (as in the NCAR model) the solution drift introduces large systematic errors in the coupled ocean-atmosphere simulation. Again, we are facing the choice: which of the evils is the lesser? The answer to this question did not wait: as has been shown by Manabe et al. (1991, 1992) within the framework of the GFDL model with and without flux corrections, the simulated changes in the ocean-atmosphere climate induced by a gradual increase of atmospheric C02 are similar and may not be substantially affected.

We now compare available estimates of the increase in the global average surface temperature during the coming century obtained with an allowance for the thermal inertia of the climatic system. Analysis of the data presented in Table 6.4 shows that these estimates do not differ very much from each other, despite the differences in the models and in the scenarios for the production of industrial C02. This, and the fact that the detected long-term tendency to change in climatic characteristics agrees in general with empirical estimates are encouraging. In particular, judging from the analyses of instrumental record data (IPCC, 1990,1992), the linear trend in land surface air temperature for the period 1881-1989 gives a rate of warming of 0.53 °C/100 year in the Northern and 0.52 °C/100 year in the Southern Hemisphere. According to ship data the increase in sea surface temperature between the late ninteenth century and the latter half of the twentieth century was less than 0.3 °C in the Northern and 0.5 °C in the Southern Hemisphere, so that the overall increase in global average sea surface temperature during the period indicated amounts to about 0.4 °C (0.43 °C between the periods 1861-1900 and 1981-1990). Combined land and marine observation data yield an increase in global average temperature of 0.47 °C between the 20-year period 1881-1900

Table 6.4. Possible changes in the annual mean global average surface air temperature (°C) in the twenty-first century relative to its preindustrial value, generated by the increase in the concentration of atmospheric C02

Author

2000

2025

Year 2050

2075

2100

Schneider and

0.45-0.80

0.90-1.50

1.40-2.25

2.20-3.45

_

Thompson (1981)

Hansen et al. (1981)

0.2-0.3

0.5-1.0

0.7-2.4

0.9-3.4

1.2-4.4

Dickinson (1982)

0.40

0.65

1.05

1.50

-

Schlesinger (1983)

0.30-0.55

0.65-0.95

1.05-1.45

1.45-1.80

1.80-2.0

Hansen et al. (1984)

0.75-0.90

-

-

-

-

Harvey and

0.7

1.3

2.1

3.1

-

Schneider (1985)

Kagan et al. (1986)

0.46

0.72

1.02

1.73

2.43

Peng et al. (1987)

0.55

0.90

1.40

1.90

-

Budyko and Israel

1.0

1.5

1.75-2.25

-

-

(1987)

IPCC (1990)

1.1

1.75

2.6

3.25

4.25

and the latest decade 1981-1990. Comparable increases in the Northern and Southern Hemispheres are 0.47 and 0.48 °C respectively. There is also evidence, though not yet conclusive on global or even hemispheric scales due to limitations in the quantity and quality of the available information, of an enhancement of precipitation in high and middle latitudes and an increase in the frequency of droughts (a decrease in soil moisture) in lower latitudes of the continents. The estimates given in Table 6.3 do not contradict these conclusions. Moreover, if we take into account that the observed changes in the climatic characteristics are determined not only by an increase in the concentration of the atmospheric C02 but also by an increase in the concentration of other greenhouse gases, then the discrepancies between calculated and observed estimates can be recognized as acceptable, or at least explainable. Of course, this does not exclude the necessity for further perfection of prognostic models and improvement in the quality of forecasts.

0 0

Post a comment