On Modification of Global Warming by Sulfate Aerosols

J. F. B. Mitchell Hadley Centre for Climate Prediction and Research, Meteorological Office, Bracknell, Berkshire, United Kingdom

Search for other papers by J. F. B. Mitchell in
Current site
Google Scholar
PubMed
Close
and
T. C. Johns Hadley Centre for Climate Prediction and Research, Meteorological Office, Bracknell, Berkshire, United Kingdom

Search for other papers by T. C. Johns in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

There is increasing evidence that the response of climate to increasing greenhouse gases may be modified by accompanying increases in sulfate aerosols. In this study, the patterns of response in the surface climatology of a coupled ocean–atmosphere general circulation model forced by increases in carbon dioxide alone is compared with those obtained by increasing carbon dioxide and aerosol forcing. The simulations are run from early industrial times using the estimated historical forcing and continued to the end of the twenty-first century assuming a nonintervention emissions scenario for greenhouse gases and aerosols. The comparison is made for the period 2030–2050 when the aerosol forcing is a maximum. In winter, the cooling due to aerosols merely tends to reduce the response to carbon dioxide, whereas in summer, it weakens the monsoon circulations and reverses some of the changes in the hydrological cycle on increasing carbon dioxide. This response is in some respects similar to that found in simulations with changed orbital parameters, as between today and the middle Holocene. The hydrological response in the palaeosimulations is supported by palaeoclimatic reconstructions. The results of changes in aerosol concentrations of the magnitude projected in the scenarios would have a major effect on regional climate, especially over Europe and Southeast Asia.

Corresponding author address: Dr. John F. B. Mitchell, Hadley Centre for Climate Prediction and Research, Meteorological Office, London Road, Bracknell, Berkshire RG12 2SY, United Kingdom.

Email: jfbmitchell@meto.gov.uk

Abstract

There is increasing evidence that the response of climate to increasing greenhouse gases may be modified by accompanying increases in sulfate aerosols. In this study, the patterns of response in the surface climatology of a coupled ocean–atmosphere general circulation model forced by increases in carbon dioxide alone is compared with those obtained by increasing carbon dioxide and aerosol forcing. The simulations are run from early industrial times using the estimated historical forcing and continued to the end of the twenty-first century assuming a nonintervention emissions scenario for greenhouse gases and aerosols. The comparison is made for the period 2030–2050 when the aerosol forcing is a maximum. In winter, the cooling due to aerosols merely tends to reduce the response to carbon dioxide, whereas in summer, it weakens the monsoon circulations and reverses some of the changes in the hydrological cycle on increasing carbon dioxide. This response is in some respects similar to that found in simulations with changed orbital parameters, as between today and the middle Holocene. The hydrological response in the palaeosimulations is supported by palaeoclimatic reconstructions. The results of changes in aerosol concentrations of the magnitude projected in the scenarios would have a major effect on regional climate, especially over Europe and Southeast Asia.

Corresponding author address: Dr. John F. B. Mitchell, Hadley Centre for Climate Prediction and Research, Meteorological Office, London Road, Bracknell, Berkshire RG12 2SY, United Kingdom.

Email: jfbmitchell@meto.gov.uk

1. Introduction

Over the last two decades there has been increasing interest in the influence of human activity on climate. Particular attention has been given to the effects of increases in carbon dioxide resulting from deforestation and the burning of fossil fuels, and other greenhouse gases including chlorofluorocarbons and methane. Greenhouse gases tend to warm climate by reducing the efficiency with which longwave radiation escapes to space. In most cases, their lifetimes are sufficiently long that they are well mixed throughout the troposphere and lower stratosphere, so their concentrations vary little from the global mean. The possible climatic effects of increasing greenhouse gases have been summarized in various reviews (McCracken and Luther 1984; Dickinson 1986; Houghton et al. 1990, 1992, 1996). Results of numerical studies using general circulation models (GCMs) show that the warming is likely to be greater over land than sea, and greater in the Northern Hemisphere than in the Southern Hemisphere. All models produce an increase in global mean evaporation and precipitation consistent with increased radiative heating of the surface, and enhanced longwave cooling from the warmer atmosphere (e.g., Mitchell et al. 1987). The regional changes vary from model to model, though almost all models produce increased precipitation in the southeast Asian monsoon and a reduction in soil moisture over southern Europe in summer. Many models also produce a reduction in soil moisture over much of North America in summer. These projected changes have formed the basis of many impact assessment studies including McTegart and Sheldon (1992).

Burning fossil fuels also leads to the release of sulfur, which oxidizes and forms hydrated sulfate aerosols (see, e.g., Pruppacher and Klett 1978). These particles scatter sunlight and hence tend to cool climate. In contrast to greenhouse gases, aerosols have a lifetime of a week or two, so they tend to be concentrated around or immediately downwind of the main industrial regions. It is only in the last few years that estimates of the geographical distribution of anthropogenic sulfate loading have been made (e.g., by Langner and Rodhe 1991; Taylor and Penner 1994). The estimates have been used, along with various assumptions on the radiative properties of the aerosol, in low resolution climate models to assess the climate effects of sulfate particles (Taylor and Penner 1994; Le Treut et al. 1996; Mitchell et al. 1995a; Roeckner et al. 1996). These studies find that aerosols produce a global-scale cooling that is greatest in northern midlatitudes, where the aerosol loading is greatest, and in high latitudes in winter due to feedbacks between sea-ice and temperature.

There have been other changes in radiative forcing due to human activity. Sulfate particles act as cloud condensation nuclei (CCN); increasing the concentration of CCN may lead to more cloud droplets andhence brighter cloud (Twomey 1974), a phenomenon known as the indirect effect, to distinguish from the direct (scattering) effect noted above. Although the greatest potential for this process is in the areas of low natural CCN concentration over the southern oceans, a preliminary estimate of this effect due to changes in effective radius suggests that the largest radiative cooling to date is again likely to be in and around the industrial regions of the Northern Hemisphere (Jones et al. 1994). Increases in soot and tropospheric ozone, also likely to be largest in the industrial regions, tend to warm (Shine et al. 1994; Haywood and Shine 1995), whereas increases in biogenic aerosols, which may be greatest in the Tropics, tend to cool (e.g., Penner et al. 1994). The magnitude and even the sign of the net effect is uncertain, but it seems that the combined aerosol, soot, and ozone effects could give a net cooling (see Shine et al. 1994), probably concentrated around the main industrial regions. The more extreme estimates of net cooling may be unrealistic, since when combined with the greenhouse gas forcing, they produce a large cooling over the northern continents, which is incompatible with recent observed changes (Mitchell et al. 1995a; Raper et al. 1995).

Warming due to increases in soot and tropospheric ozone will tend to cancel the cooling due to the indirect effect of sulfate aerosols. Thus, the radiative forcing due to direct scattering by sulfate aerosols may crudely approximate the net effect of all these geographically varying forcings. In some models, the combination of greenhouse gas and direct (scattering) aerosol forcing has been shown to give a better representation of the observed pattern of temperature change in recent decades than that simulated with greenhouse gases alone (Santer et al. 1994, 1996; Mitchell et al. 1995b; Johns et al. 1997).

In this paper we compare and contrast the simulated effects of projected increases in greenhouse gases and direct sulfate aerosol forcing on future climate using a high resolution coupled atmosphere–ocean climate model. The main aims of this paper are

  1. to document the changes in surface climatology with and without aerosol effects during the solsticial seasons, and

  2. to explain the response to aerosol forcing during northern summer.

The forcings applied in the experiments are based on actual emissions from 1860 and future emission scenarios from 1990. The scenarios are plausible, though there are large uncertainties in future greenhouse gas concentrations, and even greater uncertainties in those of sulfates and the associated direct radiative forcing. The effects of sulfate aerosols on clouds, and possible changes in other forcing including that due to industrial soot, tropospheric ozone, and biogenic aerosols, are ignored. Bearing in mind these factors and the shortcomings of current climate models, this work should be regarded as a sensitivity study, although we do present evidence that the model’s global-scale response to the aerosol forcing may be at least qualitatively correct.

In the next section we describe the model, the radiative forcing scenario, and the experimental design. In section 3 we present the model results. The mechanism of the response to aerosols in northern summer is investigated in section 4 and compared with the response to changes in orbital forcing since the mid-Holocene. The response to aerosols is then compared with other sensitivity studies. A summary and further discussion are presented in the final section.

2. The model, forcing, and experimental design

a. The model

Both the atmospheric and oceanic components of the model employ a horizontal grid with a spacing of 2.5° lat and 3.75° long and use a realistic geography, orography, and bathymetry.

The atmospheric model is a version of the unified numerical weather prediction/climate model used in the United Kingdom Meteorological Office, including the Hadley Centre (Cullen 1993). The primitive equations are solved on 19 levels. The drag due to gravity waves is parameterized based on Palmer et al. (1986).

Fluxes of heat, moisture, and momentum from the surface are stability dependent, based on Monin–Obukhov similarity theory (Smith 1990). Four soil layers are represented in the land surface temperature scheme (Warrilow et al. 1986). The hydrology scheme is based on a single moist layer and allows for both surface runoff and soil drainage. A vegetative canopy is included to represent the effects of canopy evaporation and interception (Warrilow et al. 1986). The geographical distributions of surface parameters are derived from the observed distribution of soils and vegetation.

A penetrative convective scheme (Gregory and Rowntree 1990), modified to include an explicit downdraught (Gregory and Allen 1991), is used. Convective cloud amounts are related to convective precipitation rate. The large-scale precipitation and cloud scheme is formulated in terms of an explicit cloud water variable following Smith (1990). The dissipation rate of cloud ice is a function of in-cloud ice content following Heymsfield (1977).

Radiative flux calculations follow Slingo and Wilderspin (1986). The longwave cloud emissivity is a simple exponential function of the computed cloud water content following Stephens (1978). There are four spectral bands in the solar scheme. The cloud properties at solar wavelengths are derived from the in-cloud water path assuming an effective radius of 7 μm for water cloud and 30 μm spherical droplets for ice cloud, following Slingo (1989). The surface albedo is a function of snow depth, vegetation type, and, over snow and ice only, temperature.

The ocean model, derived from Cox (1984), has 20 levels with maximum resolution near the surface. Heat and salt are diffused laterally using an isopycnal mixing scheme (Redi 1982). Vertical mixing due to current shear (Pacanowski and Philander 1981) and that due to wind mixing in the mixed layer (Kraus and Turner 1967) are represented. Sea ice is modeled using a “zero-layer” thermodynamic model (Semtner 1976), with modifications to allow a variable lead fraction and to incorporate brine rejection during freezing. Sea ice is advected using the surface current, with ice convergence set to zero when the depth exceeds 4 m, following Bryan (1969).

A 1-day coupling cycle is used, the oceanic component being updated using daily average fluxes of heat, water, and momentum from the atmosphere. Prescribed adjustments are made to the heat and water fluxes to reduce the errors in the simulation of present climate, as in Murphy (1995), except that explicit sea ice adjustments are no longer used.

The full model was integrated for a total of 510 simulated years to bring it to near equilibrium. The net flux into the ocean at the start of the experiments is less than 0.2 W m−2, and there is no perceptible drift in global mean surface temperature in the control simulation. A fuller account of the model, the spinup to equilibrium and the simulated control climate is given by Johns et al. (1997), and the simulated variability is assessed by Tett et al. (1997).

b. Specification of greenhouse gas forcing and aerosol distributions and forcing

1) Greenhouse gases

The effect of changes in greenhouse gases is represented by increases in CO2. For the period from 1860 to 1990, the standard CO2 concentration C0 in the model is increased so as to give the change in forcing due to all gases (Table 2.6 of Shine et al. 1990) assuming that CO2 forcing for concentration C is approximated by
0−2

Expression (1) gives an increase of 4.4 W m−2 for doubling of CO2. The model gives an instantaneous change of 4 W m−2 in the net downward flux of longwave and solar radiative flux at the tropopause (cf. Cess et al. 1993), reduced to 3.5 W m−2 after the adjustment of stratospheric temperatures (W. J. Ingram 1995, personal communication, see also Schneider 1975). Hence the change in model forcing from 1860 to 1990 (1.9 W m−2, see Table 1) is about 22% less than in Shine et al. (1990)

From 1990, CO2 is increased by 1% yr−1 (compound) up to 2100, at which time the increase in forcing relative to 1990 (6.5 W m−2) is close to that in IPCC scenario IS92a reported by Mitchell and Gregory (1992). Although the modeled forcing is less than given by (1) for a given CO2 increase, the 1% yr−1 increase in CO2 in the model is faster than the increase in effective CO2 in the IS92a scenario. These two differences almost cancel to give the same change in forcing in the model and IS92a. Unfortunately, the conversion of gaseous emissions to concentrations was revised in Kattenburg et al. (1996), reducing the change in forcing from 1990 to 2100 in IS92a to 5.8 W m−2, so the model forcing at 2100 is about 12% higher than this revised estimate. Note that the observed rate of increase in effective CO2 in recent years is about 0.7% yr−1.

2) Sulfate aerosols

The seasonal variation in sulfate aerosol distribution is ignored. An estimate of the current annual mean pattern of industrial aerosol burden was derived (J. Langner 1993, personal communication) using the sulfur cycle model of Langner and Rodhe (1991, the slow oxidation version), revised to allow for previously neglected sources of sulfur dioxide in the Southern Hemisphere (Fig. 1a). The current pattern was scaled by global and decadal mean industrial sulfate emissions taken from Dignon and Hameed (1989) and Hameed and Dignon (1992) to give a geographical distribution of sulfate aerosol from 1860 to 1990 (see Table 1). This uniform scaling exaggerates the sulfur loading over southeast Asia and underestimates it over North America and Europe in the middle of the period.

A further pattern for 2050 (U. Hansson 1994, personal communication) was derived using the same sulfur model driven by industrial sulfur emissions under IS92a (Fig. 1b). The pattern for the intermediate period was obtained by linearly interpolating in time between the two patterns, and the resulting field was scaled by the global mean emissions under IS92a to give the distribution (Table 1). Thereafter, the 2050 pattern was scaled by global mean emissions.

The model’s radiation scheme does not allow explicitly for scattering of radiation by atmospheric aerosols. Hence we have represented the scattering by sulfate aerosols by an increase in surface albedo, as in Mitchell et al. (1995a). It is assumed that the effect occurs only in the clear-sky fraction of the grid box—we ignore scattering of the diffuse solar beam. The instantaneous cooling rate is independent of solar angle for a globally uniform sulfate layer under clear skies. This is because in our approximation, the increased slant length exactly compensates the reduced downward flux as the solar zenith angle is increased. With clear skies and a uniform sulfate layer, the mean cooling is also proportional to the length of day.

The simulated mean annual global forcing for the current aerosol loading is 0.6 W m−2, which lies within the range of other estimates [0.3 W m−2 by Kiehl and Briegleb (1993), 0.95 W m−2 by Taylor and Penner (1994)], and the pattern of forcing has a spatial correlation of 0.9 with that found using a full aerosol scattering scheme (D. Roberts 1994, personal communication). The estimate of current forcing is discussed further by Mitchell et al. (1995a).

3) The experiments

An experiment, GHG, is started in 1860, forced by increases in carbon dioxide only. A second experiment, SUL, including increases in both carbon dioxide and direct sulfate aerosol forcing is run in parallel. Although the experiments have been run to 2100, the aerosol forcing changes little after 2050 (Fig. 2a, Table 1). The patterns of change in GHG are consistent throughout the twenty-first century, while the contrast between the patterns in SUL and GHG are most pronounced around 2050. Hence we concentrate here on the response averaged over 2030 to 2050 (at which time the greenhouse gas forcing is about 30% greater than that due to doubling carbon dioxide).

3. Results

The changes are computed with respect to 130-yr averages of the control simulation, which represent preindustrial conditions.

a. Global annual mean temperature

GHG warms slowly at first, and then more rapidly to about 1 K by 1990 (Fig. 2b) following the acceleration in the growth of the forcing (Fig. 2a). Thereafter, it warms by over 0.3 K decade−1, consistent with comparable idealized experiments with a 1% yr−1 increase in CO2. SUL warms by only 0.5 K by 1990 due to the presence of cooling by aerosols. Then the rate of warming accelerates to about 0.2 K decade−1 up to 2050 (Fig. 2a), and approaches 0.3 K decade−1 in the final few decades as the aerosol loading levels off, while CO2 continues to increase by 1% yr−1. The comparison of observed and simulated temperatures from 1860 to present is discussed in more detail in Mitchell et al. (1995b) and Johns et al. (1996)

b. Annually averaged changes in zonal mean temperature and winds

The distribution of warming in GHG (Fig. 3a) is generally similar to that found in previous studies with increased CO2 (e.g., Gates et al. 1992). The warming is enhanced in the upper troposphere in the Tropics, and at low levels in polar regions of the Northern Hemisphere. The warming is smaller in the Southern Hemisphere because of the thermal inertia of the high latitude southern ocean. The stratosphere cools, though there is a band of warming at about 100 mb in the Northern Hemisphere. This latter feature is also evident in other models that partly resolve the lower stratosphere (Boer et al. 1992; Cubasch et al. 1992; Rind et al. 1990; Mahfouf et al. 1994). Here, it may be due in part to the use of a centered vertical difference scheme that does not damp two-grid length waves—equilibrium simulations with a positive definite tracer scheme show little evidence of this feature (W. J. Ingram, personal communication).

The changes in tropospheric winds are qualitatively consistent with temperature changes assuming geostrophic balance (Fig. 3b). Westerly flow increases at upper levels in the subtropics associated with the enhanced equator to pole temperature gradient in the upper troposphere. These increases extend to the surface in most regions, especially near 55°S where there is a marked minimum in warming at most levels. Easterly flow increases in high latitudes, associated with the reduced meridional temperature gradient near the surface, and in much of low latitudes, particularly in northern winter (not shown).

In SUL, the addition of aerosols leads to a general reduction in the warming (Fig. 3c). There is a pronounced minimum in warming near 40°N where the aerosol forcing is greatest (Fig. 1b). This reduces the asymmetry between the hemispheres found in GHG, though the warming remains greater in the Northern Hemisphere. The increase in westerly flow in the northern subtropics extends much more noticeably to the surface (Fig. 3d), particularly in winter (not shown), and westerly flow is reduced near 45°N, particularly in summer (not shown).

c. Seasonal changes

In general, the changes in GHG are greater and more significant than those in SUL, because of the greater global mean forcing, and changes in temperature are more significant than those in sea level pressure, precipitation, or soil moisture (Fig. 4). The area of significant change increases with increased forcing, so that by 2100, the temperature changes are significant almost everywhere at the 90% level. The time at which temperature changes over a given fraction of the globe becomes significant generally occurs about two decades earlier in GHG than SUL. The lag in SUL is slightly shorter for temperature than for other variables. The area of changes in precipitation and sea level pressure significant at the 90% level of confidence is no greater than one might expect by chance (about 10%) until the end of this century in GHG, and the beginning of the next in SUL. On the basis of this statistic, one would not yet expect to be able to detect seasonal changes in precipitation or surface pressure in the observed record. [Note that more powerful statistical methods have been developed to test the significance of patterns of change than the simple local t test used here (e.g., Hasselmann 1993; Santer et al. 1993; Mitchell et al. 1995b).] By the period 2030 to 2050, the changes over a substantial portion of the globe have become significant (Fig. 4).

1) Northern winter (December to January)

(i) Temperature

The warming in GHG is generally greater over land than over the ocean, and greatest in the high latitudes of the winter hemisphere (Fig. 5a), as in other transient CO2 experiments (Gates et al. 1992). There are regions of little change or cooling in the southern ocean associated with deep mixing of heat in the ocean and near Greenland due in part to similar reasons (there is also a 10% reduction in the strength of the mean meridional circulation of the Atlantic Ocean).

On adding sulfate aerosols, the global mean forcing is reduced by about 25%, and the warming by 30% (or 0.7 K, Table 2). In the Northern Hemisphere, the maximum radiative cooling is displaced to the south of the region of maximum aerosol loading (Fig. 5b, recall that an annual mean aerosol distribution is used). This is because the length of day decreases and cloud cover generally increases as one moves northward across the latitude of maximum aerosol forcing. Both these factors reduce the impact of the aerosol loading [see section 2b (2)].

There are areas of pronounced cooling relative to GHG over the northern midlatitude continents (Fig. 5c), which include but extend well to the north of the area of maximum radiative forcing (Fig. 5b). This poleward extension of the cooling is due at least in part to increased snow cover and associated feedbacks in the higher latitudes. There is also marked cooling over and around the Arctic although there is no change in radiative forcing in winter. The radiative cooling during summer produces thicker and more extensive sea ice—these changes persist into winter and lead to a pronounced low-level cooling (Ingram et al. 1989; Mitchell et al. 1995a).

The combined greenhouse gas–aerosol forcing is negative over land near 25°N (Fig. 5b). This produces a marked minimum warming around 35°N with isolated regions of cooling (Fig. 5d).

(ii) Sea level pressure

Patterns of change in sea level pressure in greenhouse gas experiments tend to vary more from model to model than do patterns of changes in temperature (Mitchell et al. 1990), particularly in winter. In GHG, much of the change in pressure pattern is similar to that found in Murphy and Mitchell (1995), including the reductions over the Arctic, Antarctica, and other continental landmasses, increases around 45°S and an increase in westerly gradient over western Europe (Fig. 6a). However, in GHG pressure falls rather than rises over much of the northern midlatitude oceans.

The effect of radiative cooling due to aerosols is generally the reverse of the above pattern, with increases in pressure over most landmasses and decreases over the oceans (Fig. 6b). This pattern of sea level pressure change is consistent with the decreased land–sea temperature contrast. In northern midlatitudes, the falls of pressure over the ocean extend to some degree across the continents. The aerosol forcing generally enhances the meridional temperature gradient equatorward of the midlatitude storm track and weakens it to the north (Fig. 3c, cf. Fig 3a). This southward shift in the main baroclinic zone tends to move the main storm tracks southward and may contribute to the falls in pressure at these latitudes. These changes will be investigated in more detail in a separate study.

The net effect of the greenhouse gas and aerosol forcing is to produce a general decrease in pressure in the Northern Hemisphere that is most pronounced over the east of the midlatitude oceans, and marked increases in pressure over the eastern Mediterranean (Fig. 6c)

(iii) Precipitation and soil moisture

Global mean precipitation in GHG is increased by about 5% (Table 2). The patterns of change are similar to those found in earlier CO2 simulations including increases over most of the northern extratropics, along the intertropical convergence zone (ITCZ) and the Southern Hemisphere storm track, and decreases over much of the subtropics and around the Mediterranean (Fig. 7a). Note that the details of changes in tropical precipitation vary considerably from model to model. Changes in soil moisture generally tend to follow the changes in precipitation, though the areas of increase are slightly less extensive, particularly in the Tropics (Fig. 7b).

The changes in precipitation due to the aerosol forcing, at least in the Northern Hemisphere, are generally the reverse of those in GHG and smaller (Fig. 7c). This might be expected given the large global-scale cooling and the strong influence of temperature on many hydrological processes through the Clausius–Clapeyron relation. Again, the changes in soil moisture tend to follow the patterns of changes in precipitation, although the areas of moistening are marginally more widespread (Fig. 7d).

The net effect of the greenhouse gas and aerosol forcing in winter is generally to give similar patterns but smaller changes than those obtained using greenhouse gases alone (Fig. 7e). The most obvious exceptions occur in soil moisture, which shows enhanced moistening over much of Brazil and a net drying rather than a moistening over much of eastern Africa (Fig. 7f). In general, the spatial correlations between the changes in GHG and SUL are about 0.7 to 0.8 for the seasonal changes in temperature, sea level pressure, precipitation, and soil moisture, and higher for temperature in northern winter.

2) Northern summer (June to August)

(i) Temperature

Again, the patterns of warming in GHG are generally similar to those found in earlier studies of increased CO2. The warming is a maximum over land in northern midlatitudes and the subtropics (Fig. 8a), and small over the Arctic and near 45°S.

The global mean aerosol forcing is 25% greater than in winter, the largest increase being over land between 35° and 50°N where it triples. The largest reductions in forcing occur over central Europe, northern India, and China (Fig. 8b). Cloud cover tends to reduce the aerosol forcing in high northern latitudes. The global mean warming in GHG is reduced by 35% (Table 2). A relative cooling exceeding 3 K is found over land around 45°N (Fig. 8c). There is a further zone of maximum cooling around Antarctica associated with increases in sea ice extent.

The net CO2 and aerosol forcing is negative over most of Eurasia from 20° to 55°N (Fig. 8b). As in northern winter, the net warming in SUL shows a pronounced minimum warming in midlatitudes (though here it is farther poleward) with isolated regions of cooling (Fig. 8d). The pattern of warming in the southern hemisphere is similar to that in GHG.

(ii) Sea level pressure

In GHG, pressure is reduced over most land and over most of the Northern Hemisphere (Fig. 9a). The greater warming over the continents than the oceans enhances the land–sea temperature contrast and strengthens the normal monsoon circulations. The Southern Hemisphere anticyclonic belt is displaced poleward, and the pressure falls where Antarctic sea ice has receded. These patterns are remarkably similar to those found in Murphy and Mitchell (1995), perhaps because of the dominant role of the Northern Hemisphere land–sea distribution.

The effect of adding aerosol forcing is largely the opposite of that above—pressure increases over land and reduces in the southern high pressure belts (Fig. 9b). The main exception to this reversal of the response in GHG occurs over the northern midlatitude oceans where pressure reduces. The CO2 forcing is fairly evenly distributed over land and sea (e.g., Mitchell et al. 1995b), whereas aerosol forcing is concentrated over northern midlatitude land (Fig. 1) so one should not expect exactly the reverse response. The changes in flow due to adding aerosol forcing are discussed in section 4.

In SUL, the combined response to CO2 and aerosols removes most of the pressure falls over land and leads to a southward displacement of the Aleutian and Icelandic lows, and a relative low pressure trough extending into the west of North America and Europe (Fig. 9c). Thus, westerly flow is reduced north of about 45°N and increased to the south. The patterns in the Southern Hemisphere are similar to but weaker than those with greenhouse gases alone, consistent with the more uniform radiative forcing found there.

(iii) Precipitation and soil moisture

In GHG, precipitation increases in the southeast Asian monsoons and over much of the oceans in or around the ITCZ but decreases over much of the southern subtropical oceans and around the Mediterranean (Fig. 10a). Over the remainder of the northern extratropical continents, changes are generally small and of either sign. Soil moisture decreases over most of the northern continents, particularly over North America and southern Europe (Fig. 10b). Again, these changes are typical of other studies (e.g., Manabe et al. 1992).

The effect of adding aerosol forcing is generally the reverse of the patterns of change in GHG (Fig. 10c) in the Northern Hemisphere. For example, precipitation reduces over southeast Asia, over much of the oceanic ITCZ, and increases around the Mediterranean and the Caribbean. (The main exception is over North America where there is a significant increase.) The soil moisture changes are also generally the reverse of those in GHG, with increases predominating over much of northern midlatitudes including North America and southern Europe, and decreases over most of southeast Asia (Fig. 10d).

In SUL, precipitation increases much more markedly over the eastern United States than in GHG and increases rather than decreases over southern Europe (Fig. 10e, see also Table 3). Average precipitation decreases rather than increases in the Asian monsoon region (Table 3) and over much of northwest Europe. These anomalies are also evident in the changes in soil moisture (Fig. 10f, Table 3). The reversal of the changes in GHG over Eurasia by the addition of aerosols, if realistic, would be crucial in the assessment of impacts of climate change. Hence, we examine the mechanisms of these changes in more detail in the following section.

4. Mechanisms of change over the northern continents in summer

The aerosol forcing is greatest over Northern Hemisphere land where the aerosol loading is greatest (Fig. 1b) and in summer when the solar angle is greatest. The resulting cooling over the northern continents weakens the land–sea temperature contrast (Fig. 8c), leading to a reduction in low-level convergence over land and increases in surface pressure over the northern extratropical continents. The changes in low-level convergence are evident in the low-level winds (Fig. 11). On adding aerosols, the westerly flow of moist air over southeast Asia from the surrounding ocean is weakened and shifted south (Fig. 11b), reducing rainfall (Fig. 10c, Table 3). Conversely, there is an increase in the westerly flow of moist air from the Atlantic to the Mediterranean and northward over the Mediterranean, fueling the increases in precipitation found in southern Europe. Although there is an enhancement of the large-scale monsoon flow over Eurasia in GHG (Fig. 11a), which produces the opposite effects, the aerosol forcing dominates (Fig. 11c, Table 3). In contrast to northern winter, the net effect over the northern extratropical continents is not merely a weakening of the changes in the hydrological cycle due to increases in greenhouse gases, but also reverses the sign of some of the changes over Eurasia.

Why does the addition of aerosols have a more potent effect on the hydrological cycle in the northern extratropics in summer than in winter? First, aerosol cooling is 50% greater over land than in winter because of the greater solar zenith angle as noted above. Thus, the forcing is greater in summer than winter. Second, in winter the circulation is largely driven by the meridional temperature gradient, which leads to strong westerly flow in midlatitudes, whereas in summer, this flow weakens and the changes in land–sea temperature contrast are more effective in altering the circulation in this case. Thus, the change in circulation for a given forcing is greater in summer. These two factors combined are sufficient to reverse some of the regional changes in precipitation found on increasing greenhouse gases alone.

Reduced insolation in the Northern summer also occurs in simulations where the earth’s orbital parameters are changed from their values six thousand years ago (6kyr BP) to present. Several such studies prescribe sea surface temperatures at their present values (Kutzbach and Guetter 1986; Hewitt and Mitchell 1996; Hall and Valdes 1996). In these studies, land temperatures and hence the land–sea contrast are reduced in summer because the amplitude of the seasonal cycle of insolation is reduced. In fact, the land–sea temperature contrast also reduces in summer in models with an interactive ocean (e.g., Mitchell et al. 1988; Liao et al. 1994) because thermal inertia inhibits the response of the oceans (relative to land) to the reduced solar heating in summer. Felzer et al. (1995) also note that the response to changes in orbital forcing are largest in the summer hemisphere and over the larger continents (Asia and North Africa). In all these studies, simulated surface pressure increases over the northern continents in summer, and decreases over the Pacific1 and Atlantic oceans in midlatitudes with the decrease in summer insolation from the mid-Holocene to present (e.g., Fig. 12a, cf. Fig. 9b).

The spatial correlation between the changes in sea level pressure due to adding aerosols (SUL –GHG) and the change from 6kyr BP to present (Hewitt and Mitchell 1996) is 0.5, and there is a similar correlation between the changes in surface temperature. The corresponding changes in precipitation are also similar, with decreases over southeast Asia and into eastern Africa (Fig. 10c, cf. Fig. 12c) associated with a weakening of the low-level monsoon flow (Fig. 11b, cf. Fig. 12b), and increases over southern Europe associated with increased moist southwesterly flow from the Atlantic. In the palaeoexperiments for the mid-Holocene, in general, soil moisture decreases over much of the northern subtropical continents and increases over much of midlatitudes including southern Europe in summer. Again, these changes are generally similar to those found above on adding aerosols. Note that in such studies the results are usually presented as changes from present—here we discuss the changes from 6kyr BP to present so that the solar forcing is reduced as in SUL –GHG.

The simulated changes in hydrology from mid-Holocene are consistent on a continental scale with palaeoclimatic reconstructions of lake levels [e.g., Street-Perrott and Harrison (1985)] which indicate drier conditions in the northern subtropics, and wetter conditions in midlatitudes in our present climate, and with changes in a monsoon index based on pollen data (e.g., Prell and Van Campo 1986). A similar relationship between summer insolation and reconstructed hydrology is found in other periods (see, e.g., Prell and Kutzbach 1987). Thus, the results here suggest that the summer response to orbital forcing may be similar to that due to aerosol changes.

The extent of the similarity of the response to these two different forcings can be further gauged by comparing different aspects of the changes in heat and water balance. First, both show very similar changes in land–sea contrast in temperature, P –E, and diabatic heating (Table 4). In each case, the reduction in insolation weakens the moisture flux from land to sea and the diabatic heating of the atmosphere. Averaging over all land hides some of the structure of the response over the northern continents. Over land between 0° and 30°N, the decrease in diabatic heating is a result of latent heat release following decreased moisture convergence (Table 5), whereas over 30°–60°N, moisture convergence increases, and the reduction in diabatic heating is mainly due to evaporation increasing at the expense of sensible heat as the surface becomes wetter. In general, the changes in forcing, diabatic heating, and precipitation at this scale are larger in the 6kyr BP experiment, but the cooling is larger in the sulfate experiment because the change in forcing is negative throughout the year, and the sea surface temperatures are free to respond.

Rodwell and Hoskins (1996) found that diabatic heating in the region of the southeast Asian monsoon is associated with descending motion (and by implication reduced precipitation) over the eastern Mediterranean in idealized experiments using an atmospheric GCM. Both increases in aerosols and the changes in orbital parameters between 6kyr BP and present reduce diabatic heating over southeast Asia in our model. These decreases (23 and 14 W m−2, respectively) occur mainly as a result of reduced precipitation (10% and 5%, respectively). In each case, precipitation increases over the eastern Mediterranean whereas the Azores anticyclone and the associated northerly flow on its eastern side is weakened, consistent with Rodwell and Hoskins’s (1996) idealized experiments.

5. Comparison with other studies

We have also examined an experiment SUL1 identical to SUL except that the aerosol forcing over land was enhanced by about 25%, and an equilibrium experiment EQ [the difference between experiment AER3 and 2XCO2 of Mitchell et al. (1995a)] in which three times the aerosol forcing at 1990 was applied. In both cases, the pattern of response to adding aerosols is similar to that found here (Table 6). For example, in winter, pressure increases in high northern latitudes and mostly decreases near 40°N. Precipitation increases over Mexico, southern Europe, and southeast Asia, and decreases over central North America, northern Europe, and southeast Asia. In northern summer, sea level pressure increases over the Arctic and most of the northern continents and decreases over the Pacific and Atlantic near 55°N. Soil moisture and precipitation increase over central north America and southern Europe, and decrease over southeast Asia and northern Europe (precipitation does not increase over central north America in EQ). The changes in hydrology are qualitatively similar in a lower resolution transient experiment MPI (Table 6) carried out by Hasselmann et al. (1995, see also Lal et al. 1995). Thus, the broad response to aerosol forcing located over northern midlatitudes is generally consistent over three models and four experiments.

6. Summary and concluding remarks

The main findings of the current study are as follows.

  1. The response to a gradual increase in CO2 is generally similar to previous studies, including the drying out of the land surface in northern midlatitudes in summer, and the intensification of the Asian summer monsoon.

  2. The result of adding the effect of scattering by sulfate aerosols, represented by an increase in surface albedo, is to reduce the warming, especially in northern midlatitudes, weakening the land–sea temperature contrast and the Asian summer monsoon. Summer precipitation and soil moisture decrease over southeast Asia, and increase over southern Europe and Asia. These regional changes in hydrology are so pronounced that they reverse the changes due to greenhouse gases alone. The effects of aerosols are similar in an experiment with slightly enhanced aerosol forcing, and qualitatively similar to those in an equilibrium experiment with a simpler low-resolution model using a much larger forcing.

  3. The changes in hydrology in summer due to the cooling by aerosols are qualitatively similar to palaeoclimatic simulations of the changes from the mid-Holocene to present. The palaeoclimatic simulations are consistent, at least in a broad sense, with reconstructions from palaeoclimatic data.

There are many uncertainties in the current study. The major sources of uncertainty are the following:

  1. The fidelity of climate models, particularly in representing the magnitude of cloud feedbacks, which have been discussed elsewhere (e.g., Houghton et al. 1990, 1992; Senior and Mitchell 1993).

  2. The aerosol scenarios, including future energy production, and the fuels used. The scenario used gives a threefold increase in sulfate concentrations over southeast Asia —this would lead to severe problems with acid rain. It is possible that the removal of sulfur from emissions will become more widespread, preventing such high levels from being attained.

  3. The derivation of the distribution and radiative properties of sulfate aerosols given an emissions scenario—the forcing derived here is about 50% greater than more detailed radiative calculations (e.g., Kiehl and Briegleb 1993).

  4. Other forcing factors not taken into consideration here. These include the indirect effect of sulfate aerosols as well as changes in soot, biogenic aerosols, and tropospheric ozone, which could, for example, reduce the net cooling over land or give smoother patterns that would result in a less dramatic effect on the land–sea contrast than found here. The projection of these additional factors into the future is even less certain than that of the direct effect of sulfates.

The implication of our findings is that some of the changes in hydrology predicted to occur with increases in greenhouse gases may be less extreme in the short- and medium-term future (several decades). This situation may not prevail in the longer term—sulfate concentrations will be maintained only as long as sulfur emissions continue, whereas CO2 has a lifetime of a century or so, ensuring that concentrations will remain high for many decades to come.

Acknowledgments

We thank Henning Rodhe and Ulf Hansson for providing the aerosol distributions used in this study. William Ingram helped with some of the radiative calculations. Andy Brady, Bob Davies, and Joe Lavery produced most of the diagnostics. J. Waszkewitz kindly provided results from parallel simulations carried out at the Max-Planck-Institute, Hamburg. Chris Hewitt generously made the results from his 6kyr BP experiment available to us before publication. Howard Cattle, James Murphy, and Cath Senior provided useful comments on the text. This work was supported by the United Kingdom Department of the Environment under Contract PECD/7/12/37, with supplementary support from the Commission of the European Community (Contract EV5V-CT92-0123).

REFERENCES

  • Boer, G. J., N. A. McFarlane, and M. Lazare, 1992: Greenhouse-gas induced climate change simulated with the CCC second generation general circulation model. J. Climate,5, 1043–1077.

  • Bryan, K., 1969: Climate and circulation III. The ocean model. Mon. Wea. Rev.,97, 806–827.

  • Cess, R. D., and Coauthors, 1993: Uncertainties in carbon dioxide forcing in atmospheric general circulation models. Science,262, 1252–1255.

  • Cox, M. D., 1984: A primitive equation, three-dimensional model of the ocean. GFDL Ocean Group Tech. Rep. No. 1, 143 pp. [Available from GFDL/NOAA, P.O. Box 308, Princeton, NJ 08542.].

  • Cubasch, U., K. Hasselmann, H. Hock, E. Maier-Reimer, U. Mikolajewicz, B. D. Santer, and R. Sausen, 1992: Time-dependent greenhouse warming computations with a coupled ocean–atmosphere model. Climate Dyn.,8, 55–69.

  • Cullen, M. J. P., 1993: The unified forecast/climate model. Meteor. Mag.,122, 81–94.

  • Dickinson, R.E., 1986: How will future climate change? The Greenhouse Effect, Climate Change, and Ecosystems, B. Bolin, B. R. Doos, J. Jaeger, and R. A. Warrick, Eds., SCOPE Rep. 29, J. Wiley, 206–270.

  • Dignon, J., and S. Hameed, 1989: Global emissions of nitrogen and sulphur oxides from 1860 to 1980. J. Air Waste Manage. Assoc.,39, 180–186.

  • Felzer, B., R. J. Ogelsby, H. Shao, T. Webb III, D. E. Hyman, W. L. Prell, and J. E. Kutzbach, 1995: A systematic study of GCM sensitivity to latitudinal changes in radiation. J. Climate,8, 877–887.

  • Gates, W. L., J. F. B. Mitchell, G. J. Boer, U. Cubasch, and V. P. Meleshko, 1992: Climate modelling, climate prediction and model validation. Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment, J. T. Houghton, B. A. Callendar, and S. K. Varney, Eds., Cambridge University Press, 101–134.

  • Gregory, D., and P. R. Rowntree, 1990: A mass flux convection scheme with representation of cloud ensemble characteristics and stability dependent closure. Mon. Wea. Rev.,118, 1483–1506.

  • ——, and S. Allen, 1991: The effect of convective scale downdraughts upon NWP and climate simulations. Ninth Conf. on Numerical Weather Prediction, Denver, CO, Amer. Meteor. Soc., 122–123.

  • Hall, N. M. J., and P. J. Valdes, 1997: A GCM simulation of the climate 6000 years ago. J. Climate,10, 3–17.

  • Hameed, S., and J. Dignon, 1992: Global emissions of nitrogen and sulfur oxides in fossil fuel combustion, 1970–1986. J. Air Waste Manage. Assoc.,42, 159–163.

  • Hasselmann, K., 1993; Optimal fingerprints for the detection of time dependent climate change. J. Climate,6, 1957–1971.

  • ——, and Coauthors, 1995: Detection of anthropogenic climate change using a fingerprint method. MPI Rep. No. 168, 24 pp. [Available from Max-Planck-Institut für Meteorologie, Bundesstrasse 55, D-20146 Hamburg, Germany.].

  • Haywood, J. M., and K. P. Shine, 1995: The effect of anthropogenic sulfate and soot aerosols on the clear sky planetary radiation budget. Geophys. Res. Lett.,22, 603–606.

  • Hewitt, C. D., and J. F. B. Mitchell, 1996: GCM simulations of the climate of 6kBP: Mean changes and interdecadal variability. J. Climate, 9, 3505–3529.

  • Heymsfield, A. J., 1977: Precipitation development in stratiform ice clouds: A microphysical and dynamical study. J. Atmos. Sci.,34, 367–381.

  • Houghton, J. T., G. J. Jenkins, and J. J. Ephraums, Eds., 1990: Climate Change. The IPCC Scientific Assessment. Cambridge University Press, 366 pp.

  • ——, B. A. Callendar, and S. K. Varney, Eds., 1992: Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press, 200 pp.

  • ——, L. G. Meiro Filho, B. A. Callender, N. Harris, A. Kattenburg, K. Maskell, Eds., 1996: Climate Change. The Science of Climate Change. The Second Assessment of the Intergovernmental Panel on Climate Change. Cambridge University Press, 572 pp.

  • Ingram, W. J., C. A. Wilson, and J. F. B. Mitchell, 1989: Modeling climate change: An assessment of sea-ice and surface albedo feedbacks. J. Geophys. Res.,94, 8609–8622.

  • Johns, T. C., R. E. Carnell, J. F. Crossley, J. M. Gregory, J. F. B. Mitchell, C. A. Senior, S. F. B. Tett, and R. A. Wood, 1997: The second Hadley Centre coupled model ocean–atmosphere GCM: Model description, spinup, and validation. Climate Dyn., in press.

  • Jones, A., D. L. Roberts, and A. Slingo, 1994: A climate model study of indirect radiative forcing by anthropogenic sulphate aerosols. Nature,370, 450–453.

  • Kattenburg, A., F. Giorgi, H. Grassl, G. A. Meehl, J. F. B. Mitchell, R. J. Stouffer, T. Tokioka, A. J. Weaver, and T. M. L. Wigley, 1996: Climate models—Projections of the future. Climate Change. The Science of Climate Change. The Second Assessment of the Intergovernmental Panel on Climate Change, J. T. Houghton, L. G. Meiro Filho, B. A. Callender, N. Harris, A. Kattenburg, and K. Maskell, Eds., Cambridge University Press, 285–357.

  • Kiehl, J. T., and B. P. Briegleb, 1993: The relative roles of sulfate aerosols and greenhouse gases in climate forcing. Nature,260, 311–314.

  • Kraus, E. B., and J. S. Turner, 1967: A one-dimensional model of the seasonal thermocline Part II: Tellus,19, 98–105.

  • Kutzbach, J. E., and P. J. Guetter, 1986: The influence of changing orbital parameters and surface boundary conditions on climate simulations for the past 18 000 years. J. Atmos. Sci.,43, 1726–1759.

  • Lal, M., U. Cubasch, R. Voss, and J. Waskewitz, 1995: Effect of transient increase in greenhouse gases and sulphate aerosols on monsoon climate. Curr. Sci.,69, 752–763.

  • Langner, J., and H. Rodhe, 1991: A global three-dimensional model of the tropospheric sulfur cycle. J. Atmos. Chem.,13, 225–263.

  • Le Treut, H., M. Forichon, O. Boucher, and Z. X. Li, 1996: Aerosol indirect effect, greenhouse gases forcing, and cloud feedback associated to the climate response. Climate Sensitivity: Physical Mechanisms and their Validation, H. Le Trent, Ed., NATO ASI Series 1, Vol. 34, Springer-Verlag, 267–280.

  • Liao, X., A. F. Street-Perrott, and J. F. B. Mitchell, 1994: Two GCM experiments for 6000 years BP: Comparisons with Palaeoclimatic reconstructions. Palaeoclim.—Data Modelling,1, 99–123.

  • MacCracken, M. C., and F. M. Luther, Eds., 1984. Detecting the climatic effects of increasing carbon dioxide. Rep. DOE/ER0235, 198 pp. [Available from U.S. Dept. of Energy, Washngton, DC 20545.].

  • Mahfouf, J. F., D. Cariolle, J.-F. Royer, J.-F. Geleyn, and B. Timbal, 1994: Response of the Meteo-France climate model to changes in CO2 and sea-surface temperatures. Climate Dyn.,9, 345–362.

  • Manabe, S., M. J. Spelman, and R. J. Stouffer, 1992: Transient responses of a coupled ocean–atmosphere model to gradual changes in CO2. Part II: Seasonal response. J. Climate,5, 105–126.

  • McTegart, W. J., and G. W. Sheldon, Eds., 1992: The Supplementary Report to the IPCC Impacts Assessment. Australian Government Publishing Service, 112 pp.

  • Mitchell, J. F. B., and J. M. Gregory, 1992: Climatic consequences of emissions and a comparison of IS92a and SA90. Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment, J. T. Houghton, B. A. Callendar, and S. K. Varney, Eds., Cambridge University Press, 171–182.

  • ——, C. A. Wilson, and W. M. Cunnington, 1987: On CO2 sensitivity and model dependence of results. Quart. J. Roy. Meteor. Soc.,113, 293–322.

  • ——, N. S. Grahame, and K. J. Needham, 1988: Climate simulations for 9000 years before present: Seasonal variations and effect of the Laurentide ice sheet. J. Geophys. Res.,93, 8283–8303.

  • ——, S. Manabe, V. Meleshko, and T. Tokioka, 1990: Equilibrium climate change and its implications for the future. Climate Change. The IPCC Scientific Assessment, J. T. Houghton, G. J. Jenkins, and J. J. Ephraums, Eds., Cambridge University Press, 131–172.

  • ——, R. A. Davis, W. J. Ingram, and C. A. Senior, 1995a: On surface temperature, greenhouse gases, and aerosols: Models and observations. J. Climate,8, 2364–2386.

  • ——, T. C. Johns, J. M. Gregory, and S. F. B. Tett, 1995b: Climate response to increasing levels of greenhouse gases and sulphate aerosols. Nature,376, 501–504.

  • Murphy, J. M., 1995: Transient response of the Hadley Centre coupled model to increasing carbon dioxide. Part I: Control climate and flux adjustment. J. Climate,8, 36–56.

  • ——, and J. F. B. Mitchell, 1995: Transient response of the Hadley Centre coupled model to increasing carbon dixide. Part II: Temporal and spatial evolution of patterns. J. Climate,8, 57–80.

  • Pacanowski, R. C., and S. G. Philander, 1981: Parameterization of vertical mixing in numerical model of the tropical oceans. J. Phys. Oceanogr.,11, 1443–1451.

  • Palmer, T. N., G. J. Shutts, and R. Swinbank, 1986: Alleviation of a systematic westerly bias in general circulation and numerical weather prediction models through an orographic gravity wave drag parameterization. Quart. J. Roy. Meteor. Soc.,112, 1001–1039.

  • Penner, J. E., and Coauthors, 1994: Quantifying and minimizing the uncertainty of climate forcing by anthropogenic aerosols. Bull. Amer. Meteor. Soc.,75, 375–400.

  • Prell, W. L., and E. Van Campo, 1986: Coherent response of Arabian sea upwelling and pollen transports to late Quaternary monsoonal winds. Nature,329, 526–528.

  • ——, and J. E. Kutzbach, 1987: Monsoon variability over the past 150 000 years. J. Geophys. Res.,92, 8411–8425.

  • Pruppacher, H. R., and J. D. Klett, 1978: Microphysics of Cloud and Precipitation. D. Reidel, 714 pp.

  • Raper, S. C. B., T. M. L. Wigley, and R. Warrick, 1995: Global sea-level rise: Past and future. Rising Sea Level and Subsiding Coastal Areas, J. D. Milliman and B. U. Haq, Eds., Kluwer Academic, 11–45.

  • Redi, M. H., 1982: Oceanic isopycnal mixing by coordinate rotation. J. Phys. Oceanogr.,12, 1154–1158.

  • Rind D., R. Suozzo, N. K. Balachandran, and M. J. Prather, 1990: Climate and the middle atmosphere. Part I: The doubled CO2 climate. J. Atmos. Sci.,47, 475–494.

  • Rodwell, M. J., and B. J. Hoskins, 1996: Monsoons and the dynamics of deserts. Quart. J. Roy. Meteor. Soc.,122, 1385–1404.

  • Roeckner, E., T. Siebert, and J. Feichter, 1996: Climatic response to anthropogenic sulfate forcing simulated with a general circulation model. Aerosol Forcing of Climate, R. J. Charlson and J. Heintzenberg, Eds., J. Wiley and Sons, 349–362.

  • Santer, B. D., T. M. L. Wigley, and P. D. Jones, 1993: Correlation methods in fingerprint detection studies. Climate Dyn.,8, 265–276.

  • ——, W. Bruggemann, U. Cubasch, K. Hasselmann, H. Hock, E. Maier Reimer, and U. Mikolajewicz, 1994: Signal–to–noise analysis of time-dependent greenhouse warming experiments. Part I: Pattern analysis. Climate Dyn.,9, 267–285.

  • ——, and Coauthors, 1996: A search for human influences on the thermal structure of the atmosphere. Nature,384, 39–46.

  • Schneider, S. H., 1975: On the carbon-dioxide climate confusion. J. Atmos. Sci.,32, 2060–2066.

  • Semtner, A. J., 1976: A model for thermodynamic growth of sea-ice in numerical investigations of climate. J. Phys. Oceanogr.,6, 379–389.

  • Senior, C. A., and J. F. B. Mitchell, 1993: CO2 and climate: The impact of cloud parameterization. J. Climate,6, 393–418.

  • Shine, K. P., R. G. Derwent, D. J. Wuebbles, and J.-J. Morcrette, 1990: Radiative forcing of climate. Climate Change. The IPCC Scientific Assessment, J. T. Houghton, G. J. Jenkins, and J. J. Ephraums, Eds., Cambridge University Press, 41–68.

  • ——, Y. Fouquart, V. Ramaswamy, S. Solomon, and J. Srinivasan, 1994: Radiative forcing. Climate Change 1994. Radiative Forcing of Climate Changes and an Evaluation of the IPCC IS92 Emission Scenarios, J. T. Houghton, L. G. Meiria Filho, J. Bruce, H. Lee, B. A. Callendar, E. Haites, N. Harris, and K. Maskell, Eds., Cambridge University Press, 163–203.

  • Slingo, A., 1989: A GCM parametrization for the shortwave radiative properties of water clouds. J. Atmos. Sci.,46, 1419–1427.

  • ——, and R. C. Wilderspin, 1986: Development of a revised longwave radiation scheme for an atmospheric general circulation model. Quart. J. Roy. Meteor. Soc.,112, 371–386.

  • Smith, R. N. B., 1990: A scheme for predicting layer clouds and their water content in a general circulation model. Quart. J. Roy. Meteor. Soc.,116, 435–460.

  • Stephens, G. L., 1978: Radiation profiles in extended water clouds. II: Parametrization schemes. J. Atmos. Sci.,35, 2123–2132.

  • Street-Perrott, A. F., and S. P. Harrison, 1985: Lake levels and climate reconstruction. Palaeoclimatic Analysis and Modelling, A. D. Hecht, Ed., John Wiley, 291–340.

  • Taylor, K., and J. E. Penner, 1994: Climate system response to aerosols and greenhouse gases: A model study. Nature,369, 734–737.

  • Tett, S. F. B., T. C. Johns, and J. F. B. Mitchell, 1997: Global and regional variability in a coupled AOGCM. Climate Dyn., in press.

  • Twomey, S., 1974: Pollution and planetary albedo. Atmos. Environ.,8, 1251–1256.

  • Warrilow, D. A., A. B. Sangster, and A. Slingo, 1986: Modelling of land surface processes and their influence on European climate. Meteor. Office 20 DCTN 38, 92 pp. [Available from Meteorological Office, Bracknell RG12 2SZ, United Kingdom.].

Fig. 1.
Fig. 1.

Industrial sulfate aerosol loading derived from the sulfur cycle model of Langner and Rodhe (1991) using the “slow oxidation” parameterization. (Contours every 4 mgm m−2, extra contour at 2 mgm m−2.) (a) Present day (J. Langner, personal communication). (b) For 2050 (U. Hansson 1994, personal communication), using sulfur emissions based on IPCC scenario IS92a (see Houghton et al. 1992).

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 2.
Fig. 2.

(a) Global mean radiative forcing in GHG (dashed line) and SUL (solid line)(W m−2). Radiative cooling due to aerosols is shown as a dotted line. (b) Global annual mean temperature changes (K) in GHG (dashed line) and SUL (solid). The observed change is dotted (see Houghton et al. 1996).

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 3.
Fig. 3.

Zonally averaged changes in atmospheric temperature (contours every 1 K) and zonal wind (contours every 1 m s−1), meaned over 2030–2050. Areas of decrease are stippled. (a) Temperature (GHG), (b) zonal wind (GHG), (c) temperature (SUL), (d) zonal wind (SUL).

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 4.
Fig. 4.

Fractional area of the globe covered by significant changes at the 90% level in a two-tailed t test for temperature (solid line), pressure at mean sea level (long dashes), precipitation (dot–dashed line), and soil moisture (dotted line, as fraction of total land area). The test is applied to decadal means of the solsticial seasons, relative to the unperturbed population given by 23 decadal means from the control simulation, assumed to be independent: (a) GHG, December to February; (b) GHG, June to August; (c) SUL, December to February; (d) SUL, June to August.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 5.
Fig. 5.

Simulated changes (2030–2050) averaged over December to February. Contours every 1 K (temperature) and 2 W m−2 (radiative forcing), with negative areas stippled. (a) Temperature change due to greenhouse gases only (GHG). Changes over 99% of globe are significant at the 90% level using a two-tailed t test using yearly seasonal means, assumed to be independent, with 20 years from GHG and 230 years from the control. Only the changes on the fringes of the areas of cooling are not significant. (b) Net radiative forcing due to CO2 and aerosols in SUL. The CO2 forcing is calculated from the instantaneous change at the tropopause when CO2 is doubled, calculated every radiation time step in a 5-yr integration, and scaled logarithmically to allow for the mean change in CO2 in the period. The aerosol forcing is evaluated in SUL directly by evaluating the solar fluxes every radiation time step. (c) Temperature change due to aerosol forcing (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 69% of the globe. They are not significant over most of the southern extratropics, south of Greenland, and much of central North America. (d) Temperature change due to greenhouse gases and aerosol (SUL). Changes are significant at the 90% level (two-tailed test) over 96% of the globe. They are not significant around the areas of cooling.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 5.
Fig. 6.
Fig. 6.

Simulated changes in sea level pressure (2030–2050) during December to February. Contours every hPa, areas of decrease are shaded. (a) Greenhouse gases (GHG). Changes are significant at the 90% level (two-tailed test) over 62% of the globe. The changes in regions around the zero contour and the regions of small changes over the east of China and much of the subtropical Pacific are not significant. (b) Aerosols (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 20% of the globe. Increases over the Canadian Arctic, much of Antarctica, India, and the Arabian Sea and decreases near Japan and much of southern midlatitudes are significant. (c) Greenhouse gases and aerosols (SUL). Changes are significant at the 90% level (two-tailed test) over 57% of the globe. The decreases over the northeastern Pacific and Atlantic Oceans and central Asia, and the increases over eastern Europe are significant. The increases in the southern subtropical anticyclones, and the reduction south of New Zealand are also significant.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 7.
Fig. 7.

Simulated changes (2030–2050) from December to February in precipitation (contours at 0 and +/−1, 2, and 5 mm day−1) and soil moisture (contours at 0 and +/− 1, and 2 cm). Areas of decrease are shaded. (a) Precipitation in GHG. Increases are significant at the 90% level (two-tailed test) over 39% of the globe including most extratropical land in the Northern Hemisphere and much of the ITCZ and Southern Hemisphere storm track. Decreases are significant over 19% of the globe including much of the eastern subtropical oceans and the Mediterranean. (b) Soil moisture in GHG. Increases are significant at the 90% level (two-tailed test) over 38% of land including most of the northern extratropics and eastern Africa. Decreases are significant over 17% of land including parts of southern Europe, southern Africa, and south America. (c) Precipitation (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 21% of the globe including decreases over northern and southeast Asia and east Africa, and increases over south and east of the Mediterranean. (d) Soil moisture (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 24% of the globe. Decreases over parts of northern high latitudes, east Africa, and southeast Asia and local increases around the Mediterranean are significant. (e) Precipitation in SUL. Increases are significant at the 90% level (two-tailed test) over 31% of the globe including much of the northern extratropical continents and the ITCZ. Decreases are significant over 20% of the globe including much of the eastern subtropical oceans and the eastern Mediterranean. (f) Soil moisture in SUL. Increases are significant at the 90% level (two-tailed test) over 38% of land including much of North America, northern Europe, and central Asia. Decreases are significant over 16% of land including much of Mexico, southern Africa, and Australia.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 7.
Fig. 7.
Fig. 8.
Fig. 8.

As Fig. 5 but for June to August. (a) Temperature (GHG). Changes are significant at the 90% level over more than 99% of the globe using a two-tailed t test based on yearly means. Changes are not significant over parts of the Arctic. (b) Net radiative forcing due to greenhouse gases and aerosols (SUL). (c) Temperature change due to aerosol forcing (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 76% of the globe. They are not significant over much of the southern extratropics, northern high latitudes, and the northeastern Atlantic and Pacific basins. (d) Temperature change due to greenhouse gases and aerosol (SUL). Changes are significant at the 90% level (two-tailed test) over 96% of the globe. They are not generally significant in and around the areas of cooling.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 8.
Fig. 9.
Fig. 9.

As Fig. 6 but for June to August. (a) Greenhouse gases (GHG). Changes are significant at the 90% level (two-tailed test) over 64% of the globe. Changes are not significant in the vicinity of the zero contour. (b) Aerosols (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 32% of the globe including increases over the Arctic and most of Eurasia and decreases over the northeastern Atlantic and Pacific. (c) Greenhouse gases and aerosols (SUL). Changes are significant at the 90% level (two-tailed test) over 65% of the globe. Changes are not significant around the zero contour, over the midlatitude southern Atlantic, and much of the eastern equatorial Pacific.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 10.
Fig. 10.

As Fig. 7 but for June to August. (a) Precipitation in GHG. Increases are significant at the 90% level (two-tailed test) over 33% of the globe including most of southern high latitudes, parts of northwest Europe, and southeast Asia, and decreases are significant over 21% of the globe, including much of the subtropical oceans, western India, and around the Mediterranean. (b) Soil moisture in GHG. Increases are significant at the 90% level (two-tailed test) over 6% of land, and decreases are significant over 33% including most of Brazil and parts of India, East Africa, and immediately north of the Mediterranean. (c) Precipitation (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 24% of the globe including increases over central North America and around the Mediterranean and decreases over India and northern Europe. (d) Soil moisture (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 26% of the globe including most of the increases over North America and around the Mediterranean and the decreases over India. (e) Precipitation in SUL. Increases are significant at the 90% level (two-tailed test) over 29% of the globe including much of central North America and southern Europe, and decreases are significant over 19% of the globe including most of India and part of East Africa. (f) Soil moisture in SUL. Increases are significant at the 90% level (two-tailed test) over 15% of land including much of North America, the western Sahara and part of central Asia, and decreases significant over 28% of land including India, Brazil, and much of East Africa.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 10.
Fig. 10.
Fig. 11.
Fig. 11.

Simulated changes in wind at eta level 0.87 (typically 1500 m) in June to August (2030–2050). Regions with changes larger than 0.75 m s−1 are stippled, and the direction of the change is indicated by the arrows of uniform length: (a) GHG; (b) SUL –GHG; (c) SUL.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Fig. 12.
Fig. 12.

Simulated changes from 6kyr BP to present in June to August (from Hewitt and Mitchell 1996). In the 6kyr BP simulation, only the orbital parameters and CO2 concentrations were changed. (a) Sea level pressure. Contours every hPa; areas of decrease are shaded. (b) Wind at eta level 0.87 (typically 1500 m). Regions with changes larger than 0.75 m s−1 are stippled, and the direction of the change is indicated by the arrows of uniform length. (c) Precipitation (contours at 0 and +/−1, 2 and 5 mm day−1). Areas of decrease are shaded.

Citation: Journal of Climate 10, 2; 10.1175/1520-0442(1997)010<0245:OMOGWB>2.0.CO;2

Table 1.

Evolution of forcing and related factors. The forcing is the global annual average. Up to 1990, the aerosol field is obtained by scaling the 1986 field by the weight given in the table, after 1990 it is the weighted sum of the 1986 and 2050 fields.

Table 1.
Table 2.

Changes in global mean forcing, temperature, and precipitation.

Table 2.
Table 3.

Mean changes over selected regions (land only), averaged over 2030–2050; Southeast Asia, 5°–30°N, 70°–105°E; southern Europe, 35°–50°N, 10°W–40°E; and central North America, 35°–50°N, 80°–105°W.

Table 3.
Table 4.

Changes in land–sea contrast in (SUL –GHG) and 6kyr BP [the difference between simulations for present and 6000 years ago, from Hewitt and Mitchell (1996)].

Table 4.
Table 5.

Changes in selected parameters over Northern Hemisphere land (June, July, and August) averaged over 0°–30°N (first two columns) and 30°–60°N (last two columns) in (SUL –GHG) and 6kyr BP [the difference between simulations for present and 6000 years ago from Hewitt and Mitchell (1996)].

Table 5.
Table 6.

Mean changes over selected land areas, as in Table 2, but for the effect of aerosol forcing in (SUL –GHG) and (SUL1 –GHG) for the period 2030–2050 (see text), at equilibrium in the case of EQ (Mitchell et al. 1995a) and between the decades 1880–89 and 2040–49 for MPI (Hasselmann et al. 1995; J. Waszkewitz 1995, personal communication). Note that the sensitivity of the model used in EQ is about twice that used in the other experiments.

Table 6.
Save
  • Boer, G. J., N. A. McFarlane, and M. Lazare, 1992: Greenhouse-gas induced climate change simulated with the CCC second generation general circulation model. J. Climate,5, 1043–1077.

  • Bryan, K., 1969: Climate and circulation III. The ocean model. Mon. Wea. Rev.,97, 806–827.

  • Cess, R. D., and Coauthors, 1993: Uncertainties in carbon dioxide forcing in atmospheric general circulation models. Science,262, 1252–1255.

  • Cox, M. D., 1984: A primitive equation, three-dimensional model of the ocean. GFDL Ocean Group Tech. Rep. No. 1, 143 pp. [Available from GFDL/NOAA, P.O. Box 308, Princeton, NJ 08542.].

  • Cubasch, U., K. Hasselmann, H. Hock, E. Maier-Reimer, U. Mikolajewicz, B. D. Santer, and R. Sausen, 1992: Time-dependent greenhouse warming computations with a coupled ocean–atmosphere model. Climate Dyn.,8, 55–69.

  • Cullen, M. J. P., 1993: The unified forecast/climate model. Meteor. Mag.,122, 81–94.

  • Dickinson, R.E., 1986: How will future climate change? The Greenhouse Effect, Climate Change, and Ecosystems, B. Bolin, B. R. Doos, J. Jaeger, and R. A. Warrick, Eds., SCOPE Rep. 29, J. Wiley, 206–270.

  • Dignon, J., and S. Hameed, 1989: Global emissions of nitrogen and sulphur oxides from 1860 to 1980. J. Air Waste Manage. Assoc.,39, 180–186.

  • Felzer, B., R. J. Ogelsby, H. Shao, T. Webb III, D. E. Hyman, W. L. Prell, and J. E. Kutzbach, 1995: A systematic study of GCM sensitivity to latitudinal changes in radiation. J. Climate,8, 877–887.

  • Gates, W. L., J. F. B. Mitchell, G. J. Boer, U. Cubasch, and V. P. Meleshko, 1992: Climate modelling, climate prediction and model validation. Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment, J. T. Houghton, B. A. Callendar, and S. K. Varney, Eds., Cambridge University Press, 101–134.

  • Gregory, D., and P. R. Rowntree, 1990: A mass flux convection scheme with representation of cloud ensemble characteristics and stability dependent closure. Mon. Wea. Rev.,118, 1483–1506.

  • ——, and S. Allen, 1991: The effect of convective scale downdraughts upon NWP and climate simulations. Ninth Conf. on Numerical Weather Prediction, Denver, CO, Amer. Meteor. Soc., 122–123.

  • Hall, N. M. J., and P. J. Valdes, 1997: A GCM simulation of the climate 6000 years ago. J. Climate,10, 3–17.

  • Hameed, S., and J. Dignon, 1992: Global emissions of nitrogen and sulfur oxides in fossil fuel combustion, 1970–1986. J. Air Waste Manage. Assoc.,42, 159–163.

  • Hasselmann, K., 1993; Optimal fingerprints for the detection of time dependent climate change. J. Climate,6, 1957–1971.

  • ——, and Coauthors, 1995: Detection of anthropogenic climate change using a fingerprint method. MPI Rep. No. 168, 24 pp. [Available from Max-Planck-Institut für Meteorologie, Bundesstrasse 55, D-20146 Hamburg, Germany.].

  • Haywood, J. M., and K. P. Shine, 1995: The effect of anthropogenic sulfate and soot aerosols on the clear sky planetary radiation budget. Geophys. Res. Lett.,22, 603–606.

  • Hewitt, C. D., and J. F. B. Mitchell, 1996: GCM simulations of the climate of 6kBP: Mean changes and interdecadal variability. J. Climate, 9, 3505–3529.

  • Heymsfield, A. J., 1977: Precipitation development in stratiform ice clouds: A microphysical and dynamical study. J. Atmos. Sci.,34, 367–381.

  • Houghton, J. T., G. J. Jenkins, and J. J. Ephraums, Eds., 1990: Climate Change. The IPCC Scientific Assessment. Cambridge University Press, 366 pp.

  • ——, B. A. Callendar, and S. K. Varney, Eds., 1992: Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press, 200 pp.

  • ——, L. G. Meiro Filho, B. A. Callender, N. Harris, A. Kattenburg, K. Maskell, Eds., 1996: Climate Change. The Science of Climate Change. The Second Assessment of the Intergovernmental Panel on Climate Change. Cambridge University Press, 572 pp.

  • Ingram, W. J., C. A. Wilson, and J. F. B. Mitchell, 1989: Modeling climate change: An assessment of sea-ice and surface albedo feedbacks. J. Geophys. Res.,94, 8609–8622.

  • Johns, T. C., R. E. Carnell, J. F. Crossley, J. M. Gregory, J. F. B. Mitchell, C. A. Senior, S. F. B. Tett, and R. A. Wood, 1997: The second Hadley Centre coupled model ocean–atmosphere GCM: Model description, spinup, and validation. Climate Dyn., in press.

  • Jones, A., D. L. Roberts, and A. Slingo, 1994: A climate model study of indirect radiative forcing by anthropogenic sulphate aerosols. Nature,370, 450–453.

  • Kattenburg, A., F. Giorgi, H. Grassl, G. A. Meehl, J. F. B. Mitchell, R. J. Stouffer, T. Tokioka, A. J. Weaver, and T. M. L. Wigley, 1996: Climate models—Projections of the future. Climate Change. The Science of Climate Change. The Second Assessment of the Intergovernmental Panel on Climate Change, J. T. Houghton, L. G. Meiro Filho, B. A. Callender, N. Harris, A. Kattenburg, and K. Maskell, Eds., Cambridge University Press, 285–357.

  • Kiehl, J. T., and B. P. Briegleb, 1993: The relative roles of sulfate aerosols and greenhouse gases in climate forcing. Nature,260, 311–314.

  • Kraus, E. B., and J. S. Turner, 1967: A one-dimensional model of the seasonal thermocline Part II: Tellus,19, 98–105.

  • Kutzbach, J. E., and P. J. Guetter, 1986: The influence of changing orbital parameters and surface boundary conditions on climate simulations for the past 18 000 years. J. Atmos. Sci.,43, 1726–1759.

  • Lal, M., U. Cubasch, R. Voss, and J. Waskewitz, 1995: Effect of transient increase in greenhouse gases and sulphate aerosols on monsoon climate. Curr. Sci.,69, 752–763.

  • Langner, J., and H. Rodhe, 1991: A global three-dimensional model of the tropospheric sulfur cycle. J. Atmos. Chem.,13, 225–263.

  • Le Treut, H., M. Forichon, O. Boucher, and Z. X. Li, 1996: Aerosol indirect effect, greenhouse gases forcing, and cloud feedback associated to the climate response. Climate Sensitivity: Physical Mechanisms and their Validation, H. Le Trent, Ed., NATO ASI Series 1, Vol. 34, Springer-Verlag, 267–280.

  • Liao, X., A. F. Street-Perrott, and J. F. B. Mitchell, 1994: Two GCM experiments for 6000 years BP: Comparisons with Palaeoclimatic reconstructions. Palaeoclim.—Data Modelling,1, 99–123.

  • MacCracken, M. C., and F. M. Luther, Eds., 1984. Detecting the climatic effects of increasing carbon dioxide. Rep. DOE/ER0235, 198 pp. [Available from U.S. Dept. of Energy, Washngton, DC 20545.].

  • Mahfouf, J. F., D. Cariolle, J.-F. Royer, J.-F. Geleyn, and B. Timbal, 1994: Response of the Meteo-France climate model to changes in CO2 and sea-surface temperatures. Climate Dyn.,9, 345–362.

  • Manabe, S., M. J. Spelman, and R. J. Stouffer, 1992: Transient responses of a coupled ocean–atmosphere model to gradual changes in CO2. Part II: Seasonal response. J. Climate,5, 105–126.

  • McTegart, W. J., and G. W. Sheldon, Eds., 1992: The Supplementary Report to the IPCC Impacts Assessment. Australian Government Publishing Service, 112 pp.

  • Mitchell, J. F. B., and J. M. Gregory, 1992: Climatic consequences of emissions and a comparison of IS92a and SA90. Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment, J. T. Houghton, B. A. Callendar, and S. K. Varney, Eds., Cambridge University Press, 171–182.

  • ——, C. A. Wilson, and W. M. Cunnington, 1987: On CO2 sensitivity and model dependence of results. Quart. J. Roy. Meteor. Soc.,113, 293–322.

  • ——, N. S. Grahame, and K. J. Needham, 1988: Climate simulations for 9000 years before present: Seasonal variations and effect of the Laurentide ice sheet. J. Geophys. Res.,93, 8283–8303.

  • ——, S. Manabe, V. Meleshko, and T. Tokioka, 1990: Equilibrium climate change and its implications for the future. Climate Change. The IPCC Scientific Assessment, J. T. Houghton, G. J. Jenkins, and J. J. Ephraums, Eds., Cambridge University Press, 131–172.

  • ——, R. A. Davis, W. J. Ingram, and C. A. Senior, 1995a: On surface temperature, greenhouse gases, and aerosols: Models and observations. J. Climate,8, 2364–2386.

  • ——, T. C. Johns, J. M. Gregory, and S. F. B. Tett, 1995b: Climate response to increasing levels of greenhouse gases and sulphate aerosols. Nature,376, 501–504.

  • Murphy, J. M., 1995: Transient response of the Hadley Centre coupled model to increasing carbon dioxide. Part I: Control climate and flux adjustment. J. Climate,8, 36–56.

  • ——, and J. F. B. Mitchell, 1995: Transient response of the Hadley Centre coupled model to increasing carbon dixide. Part II: Temporal and spatial evolution of patterns. J. Climate,8, 57–80.

  • Pacanowski, R. C., and S. G. Philander, 1981: Parameterization of vertical mixing in numerical model of the tropical oceans. J. Phys. Oceanogr.,11, 1443–1451.

  • Palmer, T. N., G. J. Shutts, and R. Swinbank, 1986: Alleviation of a systematic westerly bias in general circulation and numerical weather prediction models through an orographic gravity wave drag parameterization. Quart. J. Roy. Meteor. Soc.,112, 1001–1039.

  • Penner, J. E., and Coauthors, 1994: Quantifying and minimizing the uncertainty of climate forcing by anthropogenic aerosols. Bull. Amer. Meteor. Soc.,75, 375–400.

  • Prell, W. L., and E. Van Campo, 1986: Coherent response of Arabian sea upwelling and pollen transports to late Quaternary monsoonal winds. Nature,329, 526–528.

  • ——, and J. E. Kutzbach, 1987: Monsoon variability over the past 150 000 years. J. Geophys. Res.,92, 8411–8425.

  • Pruppacher, H. R., and J. D. Klett, 1978: Microphysics of Cloud and Precipitation. D. Reidel, 714 pp.

  • Raper, S. C. B., T. M. L. Wigley, and R. Warrick, 1995: Global sea-level rise: Past and future. Rising Sea Level and Subsiding Coastal Areas, J. D. Milliman and B. U. Haq, Eds., Kluwer Academic, 11–45.

  • Redi, M. H., 1982: Oceanic isopycnal mixing by coordinate rotation. J. Phys. Oceanogr.,12, 1154–1158.

  • Rind D., R. Suozzo, N. K. Balachandran, and M. J. Prather, 1990: Climate and the middle atmosphere. Part I: The doubled CO2 climate. J. Atmos. Sci.,47, 475–494.

  • Rodwell, M. J., and B. J. Hoskins, 1996: Monsoons and the dynamics of deserts. Quart. J. Roy. Meteor. Soc.,122, 1385–1404.

  • Roeckner, E., T. Siebert, and J. Feichter, 1996: Climatic response to anthropogenic sulfate forcing simulated with a general circulation model. Aerosol Forcing of Climate, R. J. Charlson and J. Heintzenberg, Eds., J. Wiley and Sons, 349–362.

  • Santer, B. D., T. M. L. Wigley, and P. D. Jones, 1993: Correlation methods in fingerprint detection studies. Climate Dyn.,8, 265–276.

  • ——, W. Bruggemann, U. Cubasch, K. Hasselmann, H. Hock, E. Maier Reimer, and U. Mikolajewicz, 1994: Signal–to–noise analysis of time-dependent greenhouse warming experiments. Part I: Pattern analysis. Climate Dyn.,9, 267–285.

  • ——, and Coauthors, 1996: A search for human influences on the thermal structure of the atmosphere. Nature,384, 39–46.

  • Schneider, S. H., 1975: On the carbon-dioxide climate confusion. J. Atmos. Sci.,32, 2060–2066.

  • Semtner, A. J., 1976: A model for thermodynamic growth of sea-ice in numerical investigations of climate. J. Phys. Oceanogr.,6, 379–389.

  • Senior, C. A., and J. F. B. Mitchell, 1993: CO2 and climate: The impact of cloud parameterization. J. Climate,6, 393–418.

  • Shine, K. P., R. G. Derwent, D. J. Wuebbles, and J.-J. Morcrette, 1990: Radiative forcing of climate. Climate Change. The IPCC Scientific Assessment, J. T. Houghton, G. J. Jenkins, and J. J. Ephraums, Eds., Cambridge University Press, 41–68.

  • ——, Y. Fouquart, V. Ramaswamy, S. Solomon, and J. Srinivasan, 1994: Radiative forcing. Climate Change 1994. Radiative Forcing of Climate Changes and an Evaluation of the IPCC IS92 Emission Scenarios, J. T. Houghton, L. G. Meiria Filho, J. Bruce, H. Lee, B. A. Callendar, E. Haites, N. Harris, and K. Maskell, Eds., Cambridge University Press, 163–203.

  • Slingo, A., 1989: A GCM parametrization for the shortwave radiative properties of water clouds. J. Atmos. Sci.,46, 1419–1427.

  • ——, and R. C. Wilderspin, 1986: Development of a revised longwave radiation scheme for an atmospheric general circulation model. Quart. J. Roy. Meteor. Soc.,112, 371–386.

  • Smith, R. N. B., 1990: A scheme for predicting layer clouds and their water content in a general circulation model. Quart. J. Roy. Meteor. Soc.,116, 435–460.

  • Stephens, G. L., 1978: Radiation profiles in extended water clouds. II: Parametrization schemes. J. Atmos. Sci.,35, 2123–2132.

  • Street-Perrott, A. F., and S. P. Harrison, 1985: Lake levels and climate reconstruction. Palaeoclimatic Analysis and Modelling, A. D. Hecht, Ed., John Wiley, 291–340.

  • Taylor, K., and J. E. Penner, 1994: Climate system response to aerosols and greenhouse gases: A model study. Nature,369, 734–737.

  • Tett, S. F. B., T. C. Johns, and J. F. B. Mitchell, 1997: Global and regional variability in a coupled AOGCM. Climate Dyn., in press.

  • Twomey, S., 1974: Pollution and planetary albedo. Atmos. Environ.,8, 1251–1256.

  • Warrilow, D. A., A. B. Sangster, and A. Slingo, 1986: Modelling of land surface processes and their influence on European climate. Meteor. Office 20 DCTN 38, 92 pp. [Available from Meteorological Office, Bracknell RG12 2SZ, United Kingdom.].

  • Fig. 1.

    Industrial sulfate aerosol loading derived from the sulfur cycle model of Langner and Rodhe (1991) using the “slow oxidation” parameterization. (Contours every 4 mgm m−2, extra contour at 2 mgm m−2.) (a) Present day (J. Langner, personal communication). (b) For 2050 (U. Hansson 1994, personal communication), using sulfur emissions based on IPCC scenario IS92a (see Houghton et al. 1992).

  • Fig. 2.

    (a) Global mean radiative forcing in GHG (dashed line) and SUL (solid line)(W m−2). Radiative cooling due to aerosols is shown as a dotted line. (b) Global annual mean temperature changes (K) in GHG (dashed line) and SUL (solid). The observed change is dotted (see Houghton et al. 1996).

  • Fig. 3.

    Zonally averaged changes in atmospheric temperature (contours every 1 K) and zonal wind (contours every 1 m s−1), meaned over 2030–2050. Areas of decrease are stippled. (a) Temperature (GHG), (b) zonal wind (GHG), (c) temperature (SUL), (d) zonal wind (SUL).

  • Fig. 4.

    Fractional area of the globe covered by significant changes at the 90% level in a two-tailed t test for temperature (solid line), pressure at mean sea level (long dashes), precipitation (dot–dashed line), and soil moisture (dotted line, as fraction of total land area). The test is applied to decadal means of the solsticial seasons, relative to the unperturbed population given by 23 decadal means from the control simulation, assumed to be independent: (a) GHG, December to February; (b) GHG, June to August; (c) SUL, December to February; (d) SUL, June to August.

  • Fig. 5.

    Simulated changes (2030–2050) averaged over December to February. Contours every 1 K (temperature) and 2 W m−2 (radiative forcing), with negative areas stippled. (a) Temperature change due to greenhouse gases only (GHG). Changes over 99% of globe are significant at the 90% level using a two-tailed t test using yearly seasonal means, assumed to be independent, with 20 years from GHG and 230 years from the control. Only the changes on the fringes of the areas of cooling are not significant. (b) Net radiative forcing due to CO2 and aerosols in SUL. The CO2 forcing is calculated from the instantaneous change at the tropopause when CO2 is doubled, calculated every radiation time step in a 5-yr integration, and scaled logarithmically to allow for the mean change in CO2 in the period. The aerosol forcing is evaluated in SUL directly by evaluating the solar fluxes every radiation time step. (c) Temperature change due to aerosol forcing (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 69% of the globe. They are not significant over most of the southern extratropics, south of Greenland, and much of central North America. (d) Temperature change due to greenhouse gases and aerosol (SUL). Changes are significant at the 90% level (two-tailed test) over 96% of the globe. They are not significant around the areas of cooling.

  • Fig. 5.

    (Continued)

  • Fig. 6.

    Simulated changes in sea level pressure (2030–2050) during December to February. Contours every hPa, areas of decrease are shaded. (a) Greenhouse gases (GHG). Changes are significant at the 90% level (two-tailed test) over 62% of the globe. The changes in regions around the zero contour and the regions of small changes over the east of China and much of the subtropical Pacific are not significant. (b) Aerosols (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 20% of the globe. Increases over the Canadian Arctic, much of Antarctica, India, and the Arabian Sea and decreases near Japan and much of southern midlatitudes are significant. (c) Greenhouse gases and aerosols (SUL). Changes are significant at the 90% level (two-tailed test) over 57% of the globe. The decreases over the northeastern Pacific and Atlantic Oceans and central Asia, and the increases over eastern Europe are significant. The increases in the southern subtropical anticyclones, and the reduction south of New Zealand are also significant.

  • Fig. 7.

    Simulated changes (2030–2050) from December to February in precipitation (contours at 0 and +/−1, 2, and 5 mm day−1) and soil moisture (contours at 0 and +/− 1, and 2 cm). Areas of decrease are shaded. (a) Precipitation in GHG. Increases are significant at the 90% level (two-tailed test) over 39% of the globe including most extratropical land in the Northern Hemisphere and much of the ITCZ and Southern Hemisphere storm track. Decreases are significant over 19% of the globe including much of the eastern subtropical oceans and the Mediterranean. (b) Soil moisture in GHG. Increases are significant at the 90% level (two-tailed test) over 38% of land including most of the northern extratropics and eastern Africa. Decreases are significant over 17% of land including parts of southern Europe, southern Africa, and south America. (c) Precipitation (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 21% of the globe including decreases over northern and southeast Asia and east Africa, and increases over south and east of the Mediterranean. (d) Soil moisture (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 24% of the globe. Decreases over parts of northern high latitudes, east Africa, and southeast Asia and local increases around the Mediterranean are significant. (e) Precipitation in SUL. Increases are significant at the 90% level (two-tailed test) over 31% of the globe including much of the northern extratropical continents and the ITCZ. Decreases are significant over 20% of the globe including much of the eastern subtropical oceans and the eastern Mediterranean. (f) Soil moisture in SUL. Increases are significant at the 90% level (two-tailed test) over 38% of land including much of North America, northern Europe, and central Asia. Decreases are significant over 16% of land including much of Mexico, southern Africa, and Australia.

  • Fig. 7.

    (Continued)

  • Fig. 7.

    (Continued)

  • Fig. 8.

    As Fig. 5 but for June to August. (a) Temperature (GHG). Changes are significant at the 90% level over more than 99% of the globe using a two-tailed t test based on yearly means. Changes are not significant over parts of the Arctic. (b) Net radiative forcing due to greenhouse gases and aerosols (SUL). (c) Temperature change due to aerosol forcing (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 76% of the globe. They are not significant over much of the southern extratropics, northern high latitudes, and the northeastern Atlantic and Pacific basins. (d) Temperature change due to greenhouse gases and aerosol (SUL). Changes are significant at the 90% level (two-tailed test) over 96% of the globe. They are not generally significant in and around the areas of cooling.

  • Fig. 8.

    (Continued)

  • Fig. 9.

    As Fig. 6 but for June to August. (a) Greenhouse gases (GHG). Changes are significant at the 90% level (two-tailed test) over 64% of the globe. Changes are not significant in the vicinity of the zero contour. (b) Aerosols (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 32% of the globe including increases over the Arctic and most of Eurasia and decreases over the northeastern Atlantic and Pacific. (c) Greenhouse gases and aerosols (SUL). Changes are significant at the 90% level (two-tailed test) over 65% of the globe. Changes are not significant around the zero contour, over the midlatitude southern Atlantic, and much of the eastern equatorial Pacific.

  • Fig. 10.

    As Fig. 7 but for June to August. (a) Precipitation in GHG. Increases are significant at the 90% level (two-tailed test) over 33% of the globe including most of southern high latitudes, parts of northwest Europe, and southeast Asia, and decreases are significant over 21% of the globe, including much of the subtropical oceans, western India, and around the Mediterranean. (b) Soil moisture in GHG. Increases are significant at the 90% level (two-tailed test) over 6% of land, and decreases are significant over 33% including most of Brazil and parts of India, East Africa, and immediately north of the Mediterranean. (c) Precipitation (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 24% of the globe including increases over central North America and around the Mediterranean and decreases over India and northern Europe. (d) Soil moisture (SUL –GHG). Changes are significant at the 90% level (two-tailed test) over 26% of the globe including most of the increases over North America and around the Mediterranean and the decreases over India. (e) Precipitation in SUL. Increases are significant at the 90% level (two-tailed test) over 29% of the globe including much of central North America and southern Europe, and decreases are significant over 19% of the globe including most of India and part of East Africa. (f) Soil moisture in SUL. Increases are significant at the 90% level (two-tailed test) over 15% of land including much of North America, the western Sahara and part of central Asia, and decreases significant over 28% of land including India, Brazil, and much of East Africa.

  • Fig. 10.

    (Continued)

  • Fig. 10.

    (Continued)

  • Fig. 11.

    Simulated changes in wind at eta level 0.87 (typically 1500 m) in June to August (2030–2050). Regions with changes larger than 0.75 m s−1 are stippled, and the direction of the change is indicated by the arrows of uniform length: (a) GHG; (b) SUL –GHG; (c) SUL.

  • Fig. 12.

    Simulated changes from 6kyr BP to present in June to August (from Hewitt and Mitchell 1996). In the 6kyr BP simulation, only the orbital parameters and CO2 concentrations were changed. (a) Sea level pressure. Contours every hPa; areas of decrease are shaded. (b) Wind at eta level 0.87 (typically 1500 m). Regions with changes larger than 0.75 m s−1 are stippled, and the direction of the change is indicated by the arrows of uniform length. (c) Precipitation (contours at 0 and +/−1, 2 and 5 mm day−1). Areas of decrease are shaded.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1312 330 31
PDF Downloads 657 129 13