• Bougeault, P., 1985: A simple parameterization of the large-scale effects of cumulus convection. Mon. Wea. Rev.,113, 2108–2121.

  • Cess, R. D., and Coauthors, 1990: Intercomparison and interpretation of climate feedback processes in 19 atmospheric general circulation models. J. Geophys. Res.,95, 16 601–16 615.

  • ——, and Coauthors, 1996: Cloud feedback in atmospheric general circulation models: An update. J. Geophys. Res.,101, 12 791–12 794.

  • Cox, M. D., 1984: A primitive equation, three dimensional model of the ocean. GFDL Tech. Rep. GFDL/Ocean Group No. 1, 69 pp.

  • Cubasch, U., K. Hasselmann, H. Höck, 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.

  • Deardorff, J. W., 1977: A parameterization of ground surface moisture content for use in atmospheric prediction models. J. Appl. Meteor.,16, 1182–1185.

  • Déqué, M., C. Dreveton, A. Braun, and D. Cariolle, 1994: The ARPEGE/IFS atmospheric model: A contribution to the French community climate modelling. Climate Dyn.,10, 249–266.

  • Geleyn, J.-F., and A. Hollingsworth, 1979: An economical analytic method for the computation of the interaction between scattering and line absorption of radiation. Beitr. Phys. Atmos.,52, 1–16.

  • 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.

  • Houghton, J. T., G. J. Jenkins, and J. J. Ephraums, 1990: Climate Change, the IPCC Scientific Assessment. Cambridge University Press, 365 pp.

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

  • Louis, J.-F., M. Tiedtke, and J.-F. Geleyn, 1981: A short history of the operational PBL parameterization of the ECMWF. Proc. Workshop on Planetary Boundary Layer, ECMWF, 59–79.

  • 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-temperature. Climate Dyn.,9, 345–362.

  • Mitchell, J. F. B., and W. J. Ingram, 1992: Carbon dioxide and climate: Mechanisms of changes in clouds. J. Climate,5, 5–21.

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

  • ——, and J. F. B. Mitchell, 1995: Transient response of the Hadley Centre coupled ocean–atmosphere model to increase in carbon dioxide. Part II: Spatial and temporal structure of response. J. Climate,8, 57–80.

  • Roeckner, E., L. Dümenil, E. Kirk, F. Lunkeit, M. Ponater, B. Rockel, R. Sausen, and U. Schlese, 1989: The Hamburg version of the ECMWF model (ECHAM), research activities in atmospheric and oceanic modelling. WMO Tech. Document 322. [Available from World Meteor. Org., Publications Sales Unit, Case Postale 2300, Ch-1211, Geneva 2, Switzerland.].

  • Rockel, B., E. Rischke, and B. Weyres, 1991: A parametrization of broad band radiative transfer properties of water, ice and mixed clouds. Beitr. Phys. Atmos.,64, 1–12.

  • Slingo, A., and D. W. Pearson, 1987: A comparison of the impact of an envelope orography and of parametrization of orographic gravity-wave drag on model simulations. Quart. J. Roy. Meteor. Soc.,113, 847–870.

  • ——, R. C. Wilderspin, and R. N. B. Smith, 1989: The effect of improved physical parameterization on simulation of cloudiness and the earth’s radiation budget in the tropics. J. Geophys. Res.,94, 2281–2301.

  • 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.

  • Tiedtke, M., 1984: The effect of penetrative cumulus convection on the large-scale flow in a general circulation model. Beitr. Phys. Atmos.,57, 216–239.

  • Timbal, B., J.-F. Mahfouf, J.-F. Royer, and D. Cariolle, 1995: Sensitivity to prescribed changes in sea surface temperature and sea-ice in doubled carbon dioxide experiments. Climate Dyn.,12, 1–20.

  • Warrilow, D. A., A. B. Sangster, and A. Slingo, 1986: Modelling of land surface processes and their influence on European climate. Meteorological Office Tech. Note MET 0 20 DCTN 38, 90 pp.

  • View in gallery
    Fig. 1.

    Zonal average of the temperature anomalies (annual mean): (a) HCe, (b) MPIe, (c) HCm, and (d) MPIm.

  • View in gallery
    Fig. 2.

    As in Fig. 1 but for zonal wind.

  • View in gallery
    Fig. 3.

    As in Fig. 1 but for total cloudiness.

  • View in gallery
    Fig. 4.

    Zonal mean precipitation anomalies obtained in summer (JJA) with MPI forcing (dark line) and HC forcing (gray line). The forced experiments with Arpège-climat are the solid lines, and the coupled models are the dashed lines.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 271 65 6
PDF Downloads 46 18 1

Comparison between Doubled CO2 Time-Slice and Coupled Experiments

B. TimbalMétéo-France/CNRM, Toulouse, France

Search for other papers by B. Timbal in
Current site
Google Scholar
PubMed
Close
,
J-F. MahfoufMétéo-France/CNRM, Toulouse, France

Search for other papers by J-F. Mahfouf in
Current site
Google Scholar
PubMed
Close
,
J-F. RoyerMétéo-France/CNRM, Toulouse, France

Search for other papers by J-F. Royer in
Current site
Google Scholar
PubMed
Close
,
U. CubaschDeutsches Klimarechenzentrum, Hamburg, Germany

Search for other papers by U. Cubasch in
Current site
Google Scholar
PubMed
Close
, and
J. M. MurphyHadley Centre, Bracknell, Berkshire, United Kingdom

Search for other papers by J. M. Murphy in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The production of climate simulations using global coupled ocean–atmosphere models at high resolution is currently limited by computational expense and the long periods of integration that are necessary. A method of increasing the number of experiments that can be performed is the so-called time-slice technique. Using the Arpège-climat atmospheric model three 5-yr integrations of this type were run: a control and two integrations forced with sea surface temperatures derived from coupled model simulations of the transient response to increasing carbon dioxide. These coupled models are the ECHAM1 model of the Max-Planck Institute (Hamburg, Germany) and the U.K. Meteorological Office model of the Hadley Centre. The sensitivity of the response to the oceanic forcing is studied. The results are compared with the 10-yr mean atmospheric response of the coupled models at the time of the doubling of CO2. Global warmings ranging from 1.3 K to 1.9 K are obtained. Special attention is given to the modifications that occur in the hydrological cycle and their sensitivity to the SSTs. Climatic signals related to oceanic forcing, such as the modification of the ITCZ maximum of precipitation, are separated from signals due to the internal feedbacks and physical parameterizations of the models.

* Current affiliation: BMRC, Melbourne, Victoria, Australia.

† Current affiliation: ECMWF, Reading, United Kingdom.

Corresponding author address: Dr. B. Timbal, BMRC, GPO Box 1289K, Melbourne, VIC 3001, Australia.

Email: bxt@bom.gov.au

Abstract

The production of climate simulations using global coupled ocean–atmosphere models at high resolution is currently limited by computational expense and the long periods of integration that are necessary. A method of increasing the number of experiments that can be performed is the so-called time-slice technique. Using the Arpège-climat atmospheric model three 5-yr integrations of this type were run: a control and two integrations forced with sea surface temperatures derived from coupled model simulations of the transient response to increasing carbon dioxide. These coupled models are the ECHAM1 model of the Max-Planck Institute (Hamburg, Germany) and the U.K. Meteorological Office model of the Hadley Centre. The sensitivity of the response to the oceanic forcing is studied. The results are compared with the 10-yr mean atmospheric response of the coupled models at the time of the doubling of CO2. Global warmings ranging from 1.3 K to 1.9 K are obtained. Special attention is given to the modifications that occur in the hydrological cycle and their sensitivity to the SSTs. Climatic signals related to oceanic forcing, such as the modification of the ITCZ maximum of precipitation, are separated from signals due to the internal feedbacks and physical parameterizations of the models.

* Current affiliation: BMRC, Melbourne, Victoria, Australia.

† Current affiliation: ECMWF, Reading, United Kingdom.

Corresponding author address: Dr. B. Timbal, BMRC, GPO Box 1289K, Melbourne, VIC 3001, Australia.

Email: bxt@bom.gov.au

1. Introduction

Three-dimensional atmospheric general circulation models (GCM) have been used worldwide over the past decade to perform climate research. In recent years, this research has expanded to include coupling of atmospheric GCMs to oceanic GCMs and sea-ice models, in an effort to simulate transient, realistic scenarios of greenhouse gas increases (Houghton et al. 1990). In general, such computationally expensive experiments cannot be run often enough to obtain reliable estimates of the statistical significance of the response. An alternative commonly used and less expensive approach is to perform “time-slice” experiments where the atmospheric GCM is forced by sea surface temperature (SST) anomalies. A simple method to compare model sensitivities is to specify spatially and temporally fixed SST anomalies (Cess et al. 1990, 1996). More realistically, time-averaged SST anomalies centered on the time of doubling of CO2 in a coupled model transient simulation can be used. Experiments of this type performed by Mahfouf et al. (1994) showed that the general features of the atmospheric response obtained by integration of a coupled model can be reproduced in a different atmospheric GCM forced by the same SST anomalies. In addition, the internal feedbacks in the forced GCM modify the response relative to that of the coupled model. Because this technique requires relatively modest computer resources, a larger number of sensitivity experiments using high resolution and/or sophisticated physical parameterizations can be performed. Also, climatological SSTs can be imposed in the reference simulation in order to minimize climate drift and ensure the best possible agreement with observations.

In this study we use SST anomalies obtained from two different coupled models, one from the Max-Planck Institute (MPI) and one from the Hadley Centre (HC) to force an atmospheric GCM: the Météo-France Arpège-climat model. Results of these forced experiments are compared with the atmospheric response simulated by the coupled models. In a previous study, Timbal et al. (1995) compared the responses in the two forced experiments in order to identify features sensitive to the oceanic forcing used. This note has the additional purpose of separating the component of the signal driven by the oceanic forcing from the component, which is model dependent. The former component will be common to the forced atmospheric GCM and the coupled model from which the forcing is obtained, while the latter component will clearly depend on the physical parameterization packages of the model used. The results of the coupled models provide a valuable means to analyze the mechanisms operating in the forced experiments.

The purpose of this note is threefold. We wish to 1) briefly introduce the models involved in this comparison: the coupled models from the MPI and the HC, and the atmospheric GCM Arpège-climat. We focus on the main differences in the physics of the models. We also present the experimental designs of both the forced and coupled experiments as well as give a brief summary on the simulated reference (control) climatologies. 2) We wish to describe the climatic changes in the forced experiments and compare them to the results of the coupled models. We identify differences caused by the model physical packages and robust results, which are model independent. 3) Finally, we wish to summarize, in conclusion, the features of both the coupled and forced experiments and the differences found between them.

2. Experimental designs and model climatologies

A comparison of the main features of the models involved in this study is given in Table 1.

The Max-Planck Institute coupled model, ECHAM1, is described in detail by Cubasch et al. (1992). The ice and liquid water contents of clouds are incorporated as additional prognostic variables (Roeckner et al. 1989). Control and perturbed coupled simulations were run for 100 yr. The atmospheric and oceanic components were coupled with flux corrections applied to the surface fluxes of heat, water, and momentum. The CO2 doubling period that we consider is a 10-yr average (2041–2050) of the perturbed run, based on the Intergovernmental Panel on Climate Change (IPCC; Houghton et al. 1990) scenario A. The SST anomalies used to force the Arpège-climat model are the difference between the perturbed and control simulations (in which the CO2 concentration is held fixed at the 1985 level). We also analyzed the results of the atmospheric component of the MPI model for the same period. In this article “MPIm” indicates results from the coupled experiments with ECHAM1 and “MPIe” indicates results from Arpège-climat forced with the MPIm SSTs.

At the Hadley Centre, the coupled experiments described by Murphy (1995) were performed using a gridpoint atmospheric model described by Slingo and Pearson (1987) and an ocean model based on the one described by Cox (1984). Cloud water was a prognostic variable of the model and cloud cover was diagnosed according to the statistical scheme of Smith (1990). Control and perturbed coupled simulations were integrated for a period of 75 yr following a 150-yr preliminary integration performed to provide flux corrections (for heat and water but not momentum) and a quasi-equilibrium initial state for the climate change experiment. The scenario for greenhouse gas increase was 1% yr−1 compound annually. The last 10 yr of each simulation, corresponding to the time of doubling of CO2 in the perturbed run, were used to calculate the average SST anomalies imposed in Arpège-climat (denoted HCe hereafter) and to study the results of the coupled model (denoted HCm hereafter).

The atmospheric GCM forced with imposed SST anomalies is the version 1 of Arpège-climat model described by Déqué et al. (1994). The methodology used for the forced experiments was described in detail by Timbal et al. (1995). Three time-slice integrations were performed: a control and two perturbed ones. It is interesting to use the results of the two coupled models since the SST anomalies imposed were rather different in zonal mean, at least in the Northern Hemisphere (Timbal et al. 1995, Fig. 1). The HC model gives larger positive anomalies between 30° and 60°N. Here the difference displays a strong annual cycle, the maximum difference occurring in summer. In the Southern Hemisphere the zonal mean differences are very small. Over continents, to be consistent with the ocean, the temperature anomalies calculated by the coupled GCMs were added to the deep-soil temperature. The surface temperature is restored toward the new deep-soil temperature with a time-constant of 20 days. Timbal et al. (1995) demonstrated that the relaxation of deep-soil temperatures has little impact on the simulated surface warming patterns. The sea-ice extents used in our perturbed runs were those calculated by the coupled GCMs during the perturbed transient runs. They are averaged over the 10-yr period corresponding to the time of CO2 doubling, whereas the reference simulation uses observed climatological extents. This approach leads to a stronger high latitude warming compared with the coupled models.

The reference climatologies simulated by the coupled models (10-yr average) and the forced atmospheric model (5-yr average) were compared to observations. The main features of the observed climate are reproduced by the three models, which perform in a broadly comparable manner. Nevertheless some limitations in the skill of the models are apparent. Most of the errors occur in areas where the model grid is too coarse to resolve the relevant climatic features adequately. Otherwise it is apparent that using a model with specified climatological SSTs results in a better prediction of surface air temperature. Arpège-climat forced by Center for Ocean–Land–Atmosphere Studies–Climate Analysis Center SST gives, as expected, a good annual cycle of temperature at each latitude. On the other hand, coupled models, which are integrated for several decades or more tend to suffer from climate drift in the ocean: here the tropical ocean becomes too warm in HCm and too cold in MPIm. Some problems are also related to the excessive persistence of sea-ice extent in summer in MPIm. The errors with oceanic surface temperature also influence other climatic variables, such as cloudiness and precipitation through evaporation.

3. Doubled carbon dioxide climate

The globally averaged warming simulated by the models (Table 2) ranged from 1.3°C and 1.9°C, which is within the range given by the IPCC for global coupled experiments (Houghton et al. 1992). Arpège-climat gave a larger warming than the coupled models. The distributions of zonal-mean surface air temperature revealed that the source of this difference was not the midlatitude continental areas, but high latitudes regions over the sea-ice margin. In the forced experiments large temperature anomalies were obtained over areas where sea-ice was present in the control but absent in the 2 × CO2 simulation.

The hydrological cycle intensified in each experiment: the fractional change increased with the magnitude of the warming. The change per unit warming was larger in Arpège-climat than in the coupled models. This is because the simpler approach used to parameterize clouds (i.e., from the specific humidity) is more sensitive than the schemes used in the coupled models, which are based on prognostic cloud water content variables (Table 1). Therefore the anomalies for total cloudiness were different even though the models agreed on the sign of the change. The features simulated by the coupled models and the forced model are comparable at the global scale. It is also of interest to compare the zonal mean results over the entire troposphere, remembering that in the previous section it was noted that the agreement between the zonal means of the simulated reference climatologies is not complete.

The vertical structure of the warming (Fig. 1) shows a maximum in the upper troposphere. This is due to an increase in the depth of vertical motions and to enhanced latent heat, released by moist convection at high altitude (see Mitchell and Ingram 1992; Mahfouf et al. 1994). The magnitude of this maximum differs between models, varying according to the climate sensitivity and the importance of convection in the water cycle. The zonal distribution of the near-surface warming is also important. In HCm, the warming in the Northern Hemisphere below 500 hPa is larger than in MPIe. In this respect HCe is very similar to HCm, showing that the enhanced warming relative to MPIe is robust and caused both by the response of the full coupled ocean (HCm) or the simplest process of SST forcing (HCe). Although the intensity of the tropospheric warming pattern depends on the oceanic warming, the moist convection parameterization also has a strong influence. For example, the tropical maximum temperature increase varies from 1.5 K in MPIm to 3 K in MPIe. The results at the tropopause and in the lower stratosphere depend on the models’ vertical resolution and their ability to reproduce the observed sharp gradient in the thermal structure. In Arpège-climat the warming decreases rapidly with height near the equatorial tropopause and gives way to a cooling that increases rapidly with height in the lower stratosphere. In MPIm, similar behavior is obtained, but with a reduced vertical gradient, while in HCm, the lack of resolution above 100 hPa precludes a study of the structure of the equatorial tropopause and stratosphere, although a stratospheric cooling is observed over 40 hPa in the Tropics.

The dynamics of the troposphere (Fig. 2) also shows similar behavior in each experiment: an increase of the westerly jets at midlatitudes and in the easterlies in the tropics. The magnitude of these changes is proportional to the temperature response (e.g., a smaller increase in MPIm and a larger increase in Arpège-climat, especially HCe). However, the increase in the easterlies varies from a strong signal in HCm where the maximum is located in the band 20°–30°N, to a very weak signal in MPIm where the atmospheric warming is insufficient to produce much response. In HCe, the maximum of the increase is situated in the Northern Hemisphere as in HCm and is stronger than in MPIe, demonstrating the impact of stronger oceanic forcing.

Relatively good agreement is found between models for cloud cover changes (Fig. 3). This is surprising considering the difference in the approaches used to calculate cloud cover in these models. Cloudiness increases near the tropopause (revealing an increase in tropopause height) and decreases in the middle troposphere. The amplitude of the anomalies is related to the magnitude of the global warming. However, the diagnostic cloud scheme used in Arpège-climat gives sharper structures. It also predicts an increase in low cloud in the ITCZ, which differs from the other models which use prognostic schemes for cloud water. Nevertheless, the geographical distributions of changes in cloud cover (not shown) do not show great similarity between the models. This is probably because patterns of change in cloud cover are highly sensitive to those of other model variables, and also because they are too noisy to be statistically robust at the regional scale.

The differences between the models are more significant for zonal mean precipitation in June–August (JJA) (Fig. 4). In the ITCZ, the experiments show an increase in convective precipitation but the amplitude of the signal ranges from 0.2 mm day−1 in MPIm to 0.8 mm day−1 in HCe. A coherent signal appears in HCm and HCe: a strong increase in precipitation over the ITCZ maximum in parallel with a decrease on the equatorial side of the convergence zone. Murphy and Mitchell (1995) argue that in HCm this pattern is driven by the asymmetry in the tropical SST response between the Northern and Southern Hemispheres, while Timbal et al. (1995) suggest that the signal in HCe is related to the midlatitude oceanic forcing through the meridional circulation. In the 40°–60°N latitude band, the decrease in total precipitation in the Arpège-climat experiments is related to continental feedbacks simulated by the model. In the coupled models the increases in precipitation simulated at high latitudes (60°–80°N), which also occur in Arpège-climat, extend southward toward 40°N.

4. Discussion and conclusions

This study demonstrates the benefits of performing time-slice experiments using SST anomalies obtained from transient coupled experiments. When these anomalies are used to force a different atmospheric GCM to that used in the coupled experiment, general features of the atmospheric response driven by oceanic warming are reproduced. The comparison of two forced experiments made with the Arpège-climat model enables us to separate changes driven by the atmospheric part of the climate system (which lead to similar results in MPIe and HCe), from changes driven by the oceans (leading to similar results in MPIe and MPIm or HCe and HCm). The results reveal limitations in the models, which affect the 2 × CO2 minus 1 × CO2 anomalies obtained: these relate to inadequate horizontal or vertical resolution and simple physical parameterization schemes used.

The forced model represents some aspects of the unperturbed climate better using a computationally cheaper approach, but the simulated 2 × CO2 climate is obtained in a less physically based manner, as the atmosphere–ocean feedbacks are missing.

The main feedback mechanisms operating in global warming experiments are represented in a similar manner by all the models, and hence the basic structure of the zonally and annually (or seasonally) averaged response in each experiment is similar throughout the atmosphere. Nevertheless, the intensity of the simulated anomalies varies according to both the oceanic forcing and the physical parameterization schemes used in the model. The tropospheric warming is a maximum around 200 hPa over the equator. There is a stratospheric cooling that increases with height and also a rise in the height of the tropopause, which is better defined in those models possessing sufficient vertical resolution.

Several aspects of the atmospheric response are sensitive to the magnitude and distribution of the changes in SST. For example, the upper-tropospheric warming in the Tropics increases with the change in global mean SST, while the latitudinal distribution of the SST changes influences the temperature response in the lower troposphere. Warming of the ocean also leads to increased evaporation, which increases specific humidity in the atmosphere. In the warmer, moister 2 × CO2 atmosphere, detrainment of humid air occurs at higher altitude, leading to increases in cloudiness above the tropopause and decreases in the middle troposphere. All models agree on this pattern of change, however, the intensity of the pattern is greater in the Arpège-climat model, due both to the convective parameterisation and its simple diagnostic scheme for cloud cover. All models simulate increases in the midlatitude westerly jets (good agreement between models on the strength of the changes) and in the equatorial trade winds (less agreement).

The geographical distributions of the climate changes simulated in each experiment show less agreement than the planetary-scale features. One exception is the distribution of zonal mean precipitation in northern summer where a marked increase of the ITCZ maximum is obtained in cases where a strong oceanic warming is obtained in the Northern Hemisphere. At present the predicted regional changes in surface climatic variables cannot be regarded as reliable either in forced or coupled GCMs, since the agreement between different models is lower than the agreement between alternative observed climatologies, especially for quantities associated with the hydrological cycle. At current resolution, GCMs are limited in their ability to investigate regional climate change.

Acknowledgments

This work would not have been possible without the help of R. Sausen and M. Windelband at the Max-Planck Institute. Their collaboration is gratefully acknowledged. We thank D. Hess and an anonymous reviewer for their constructive criticism. This work was supported by grants from the PNEDC, the Commission of the European Community (EV5V-CT92-0123, EV5V-CT92-0125, and EV5V-CT94-0505), and Electricité de France.

REFERENCES

  • Bougeault, P., 1985: A simple parameterization of the large-scale effects of cumulus convection. Mon. Wea. Rev.,113, 2108–2121.

  • Cess, R. D., and Coauthors, 1990: Intercomparison and interpretation of climate feedback processes in 19 atmospheric general circulation models. J. Geophys. Res.,95, 16 601–16 615.

  • ——, and Coauthors, 1996: Cloud feedback in atmospheric general circulation models: An update. J. Geophys. Res.,101, 12 791–12 794.

  • Cox, M. D., 1984: A primitive equation, three dimensional model of the ocean. GFDL Tech. Rep. GFDL/Ocean Group No. 1, 69 pp.

  • Cubasch, U., K. Hasselmann, H. Höck, 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.

  • Deardorff, J. W., 1977: A parameterization of ground surface moisture content for use in atmospheric prediction models. J. Appl. Meteor.,16, 1182–1185.

  • Déqué, M., C. Dreveton, A. Braun, and D. Cariolle, 1994: The ARPEGE/IFS atmospheric model: A contribution to the French community climate modelling. Climate Dyn.,10, 249–266.

  • Geleyn, J.-F., and A. Hollingsworth, 1979: An economical analytic method for the computation of the interaction between scattering and line absorption of radiation. Beitr. Phys. Atmos.,52, 1–16.

  • 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.

  • Houghton, J. T., G. J. Jenkins, and J. J. Ephraums, 1990: Climate Change, the IPCC Scientific Assessment. Cambridge University Press, 365 pp.

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

  • Louis, J.-F., M. Tiedtke, and J.-F. Geleyn, 1981: A short history of the operational PBL parameterization of the ECMWF. Proc. Workshop on Planetary Boundary Layer, ECMWF, 59–79.

  • 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-temperature. Climate Dyn.,9, 345–362.

  • Mitchell, J. F. B., and W. J. Ingram, 1992: Carbon dioxide and climate: Mechanisms of changes in clouds. J. Climate,5, 5–21.

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

  • ——, and J. F. B. Mitchell, 1995: Transient response of the Hadley Centre coupled ocean–atmosphere model to increase in carbon dioxide. Part II: Spatial and temporal structure of response. J. Climate,8, 57–80.

  • Roeckner, E., L. Dümenil, E. Kirk, F. Lunkeit, M. Ponater, B. Rockel, R. Sausen, and U. Schlese, 1989: The Hamburg version of the ECMWF model (ECHAM), research activities in atmospheric and oceanic modelling. WMO Tech. Document 322. [Available from World Meteor. Org., Publications Sales Unit, Case Postale 2300, Ch-1211, Geneva 2, Switzerland.].

  • Rockel, B., E. Rischke, and B. Weyres, 1991: A parametrization of broad band radiative transfer properties of water, ice and mixed clouds. Beitr. Phys. Atmos.,64, 1–12.

  • Slingo, A., and D. W. Pearson, 1987: A comparison of the impact of an envelope orography and of parametrization of orographic gravity-wave drag on model simulations. Quart. J. Roy. Meteor. Soc.,113, 847–870.

  • ——, R. C. Wilderspin, and R. N. B. Smith, 1989: The effect of improved physical parameterization on simulation of cloudiness and the earth’s radiation budget in the tropics. J. Geophys. Res.,94, 2281–2301.

  • 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.

  • Tiedtke, M., 1984: The effect of penetrative cumulus convection on the large-scale flow in a general circulation model. Beitr. Phys. Atmos.,57, 216–239.

  • Timbal, B., J.-F. Mahfouf, J.-F. Royer, and D. Cariolle, 1995: Sensitivity to prescribed changes in sea surface temperature and sea-ice in doubled carbon dioxide experiments. Climate Dyn.,12, 1–20.

  • Warrilow, D. A., A. B. Sangster, and A. Slingo, 1986: Modelling of land surface processes and their influence on European climate. Meteorological Office Tech. Note MET 0 20 DCTN 38, 90 pp.

Fig. 1.
Fig. 1.

Zonal average of the temperature anomalies (annual mean): (a) HCe, (b) MPIe, (c) HCm, and (d) MPIm.

Citation: Journal of Climate 10, 6; 10.1175/1520-0442(1997)010<1463:CBDCTS>2.0.CO;2

Fig. 2.
Fig. 2.

As in Fig. 1 but for zonal wind.

Citation: Journal of Climate 10, 6; 10.1175/1520-0442(1997)010<1463:CBDCTS>2.0.CO;2

Fig. 3.
Fig. 3.

As in Fig. 1 but for total cloudiness.

Citation: Journal of Climate 10, 6; 10.1175/1520-0442(1997)010<1463:CBDCTS>2.0.CO;2

Fig. 4.
Fig. 4.

Zonal mean precipitation anomalies obtained in summer (JJA) with MPI forcing (dark line) and HC forcing (gray line). The forced experiments with Arpège-climat are the solid lines, and the coupled models are the dashed lines.

Citation: Journal of Climate 10, 6; 10.1175/1520-0442(1997)010<1463:CBDCTS>2.0.CO;2

Table 1.

The main characteristics of the models and the references that describe the physical parameterizations of each model.

Table 1.
Table 2.

The global anomalies obtained for 2 × CO2 climate.

Table 2.
Save