notes and correspondence the effect of doubling the co2

December 1989 A. Noda and T. Tokioka 1057

NOTES AND CORRESPONDENCE

The Effect of Doubling the CO2 Concentration on

Convective and Non-convective Precipitation in

a General Circulation Model Coupled with

a Simple Mixed Layer Ocean Model

By Akira Noda and Tatsushi Tokioka

Meteorological Research Institute, 1-1, Nagarnine, Tsukuba, Ibaraki 305, Japan

(Manuscript received 12 April 1989, in revised form 9 August 1989

1. Introduction

Vast differences exist in the treatment of clouds among the various GCMs that have been used for CO2-sensitivity studies (see e.g., Rutledge and Schlesinger, 1985). However, a common feature has been found among several CO2 sensitivity ex-periments (Hansen et al., 1984; Washington and Meehl, 1984; Wetherald and Manabe, 1986, 1988; Wilson and Mitchell, 1987; also see the review by Schlesinger and Mitchell, 1987). That is, when the CO2 concentration in the model climate is dou-bled, the cloudiness decreases throughout most of the tropical and middle latitude tropospheres of both hemispheres, even though the tropics remains a moist and convectively active region with precipita-tion increasing in these regions. Based on this result, it is inferred that in the semi-arid tropical regions the increase in precipitation is expected to be in the form of convective rainfall, giving an increase in the intensity but not necessarily in the frequency of rain-fall (Jaeger,1988). However, to date no study as yet has investigated the CO2-induced changes in convec-tive and non-convective types of precipitation and their frequency distributions on a global scale. This topic will be examined in the present note, based on a preliminary CO2 sensitivity experiment, per-formed with the use of a general circulation model developed at the Meteorological Research Institute (MRI). Since cloudiness and area of precipitation are closely correlated, the CO2-induced cloud feed-back process will be discussed with respect to the response in the convective and non-convective areasas of precipitation.

C1989, Meteorological Society of Japan

2. Brief description of the model

The model used in the present study is an at-mospheric general circulation model coupled with a 50-m oceanic mixed layer, energy balance sea-ice model that follows the parameterization scheme em-ployed in the zero layer model of Semtner (1976). The atmospheric model consists of a tropospheric

version of the MRI*GCM-I described by Tokioka et al. (1984). A modified a-coordinate with *=(p-Pt)/(Ps - pt) is used in the vertical, where p is the pressure, pt the pressure at the top of the model (=100mb), while p* represents the surface pressure. The model atmosphere is divided into five layers with interfaces located at 1/9, 3/9, 5/9 and 7/9 in

* as shown in Fig. 1. The Arakawa C grid scheme (Arakawa and Mintz,1974) is used, having a regular 4**5* latitude/longitude resolution.

The model determines three types of precipita-tion, which are denoted by LSPREC, MLPREC and CUPREC. The LSPREC is the precipitation due to saturation at a grid point. The MLPREC and CUPREC are the precipitation due to convection originating in the free atmosphere and in the plane-tary boundary layer, respectively.

The LSPREC occurs when the relative humidity at a grid point exceeds 100 %. When condensation occurs in a layer, which is not the lowest layer of the model, the condensed water is transported down to the next lower layer, where it may evaporate. This process is repeated until the lowest layer is reached. When the lowest layer is saturated, the condensed water precipitates to the surface as rain or snow, depending on the surface air temperature.

Moist convective adjustment in the free atmo-sphere occurs whenever the saturation moist static energy h*k in a layer * becomes less than the moist

1058 Journal of the Meteorological Society of Japan Vol. 67, No. 6

Fig. 1. Various types of clouds identified in the MRI.GCM-I. Radiatively interactive

clouds are shaded (from Tokioka et al., 1984)

static energy hk+1 in the next lower layer k+1. That is, an air parcel rising from level, k+1 experi-ences positive buoyancy at level k. If temperature and moisture in the levels k and k+1 are modi-fied in such a way as to reduce hk+1-h*k to zero and supersaturation occurs, the supersaturated Wa-ter precipitates as the MLPREC (Arakawa, 1972).

The CUPREC is calculated through the cumulus parameterization developed by Arakawa and Schu-bert (1974). The Arakawa Schubert parameteriza-tion takes into account the interaction of a cumu-lus cloud ensemble with the large-scale environment, which is divided into the subcloud mixed layer and the region above. For details, see Arakawa and Schu-bert (1974), Lord and Arakawa (1980), Lord (1982), Lord et al. (1982) and Tokioka ei al. (1984).

Figure 1 shows the various types of clouds iden-tified in the present GCM. However, not all of the clouds are considered in the radiation scheme. Since the cloudiness associated with shallow cumulus con-vection and moist convection in the free atmosphere is much smaller than 10/10, the clouds associated with shallow CUPREC (i.e., cumuli with tops be-low the 400 mb level and an air temperature of above -40*) and the MLPREC are not considered in the

present radiation scheme. This is due to the fact that the shortwave radiation scheme developed by Katayama (1972) is valid for total cloudiness only. This restriction may possibly cause deterioration in the present simulation. For this reason, the authors are developing a new shortwave radiation scheme which permits partial cloudiness.

The diurnal and seasonal cycles are included. Shortwave fluxes are calculated every model hour

while longwave fluxes are determined every three hours. Over snow-free land the surface albedo is specified, using values compiled by Matthews (1983) (see Kitoh et al. (1988) for details). The surface albedo over permanent land ice and snow is the min-imum of (0.85, 0.7+0.15Zs ), where zs is the sur-face terrain height in km. For melting snow and for sea ice without snow cover the albedo is 0.6. The ocean surface albedo is taken from Posey and Clapp (1963). The horizontal oceanic heat transport is assumed

to be zero for simplicity, although it is as impor-tant as the atmospheric heat transport. Therefore, the present model has the same bias as that of the Geophysical Fluid Dynamics Laboratory (GFDL; e.g., see Wetherald and Manabe,1986) and National Center for Atmospheric Research (NCAR; see Wash-ington and Meehl,1984) models in the reproduction of observed values for the control simulation. That is, the simulated temperatures are higher than those observed in the low latitudes, while being lower than those observed in the high latitudes.

3. Experiments

The control (1*CO2) experiment (CO2, 320 ppm) was initialized with model data for July 1 of another experiment, and integrated for 9 years. The doubled CO2 experiment (2*CO2) started from July 1 data of the fifth year of the control exper-iment, and was also integrated for 9 years. The last 2 years of each experiment were used for time aver-aging of the results. In order to investigate the fre-quency distribution of various types of precipitation, a special sampling was made of the precipitation at every time step (one hour) for the calculation of di-abatic heating. These sampling were made for 10 day periods at the beginning of January and July of the last year of each experiment.

The annual global mean surface air temperature for the 1*CO2 experiment was calculated as 14.8* for year 9 and the difference between the 1*CO2 and 2*CO2 experiments (2*CO2 -1*CO2) at-tained the value 4.3*, which is within the range 2.8**5.2* obtained by other studies (see Cess and Potter, 1988). Neither 14.8* nor 4.3* is an equilibrium value since a period of 9 years is too short to reach final equilibrium. However, it can be expected from relaxation curves obtained by the NCAR and GFDL models, that by year 9 roughly 90 % of the final equilibrium value has been attained. Therefore, the results given in the present note are only preliminary, and should be statistically con-firmed by further experiments with longer integra-tions. The deficiencies due to the zero oceanic heat transport were more pronounced for the present ex-periments than for the GFDL and NCAR experi-ments. Temperatures in the low latitudes and sea-


Fig. 2. Latitudinal distribution of the annual mean rates of precipitation; total precipitation (top left), CUPREC (top right), LSPREC (bottom left), MLPREC (bottom right), for the 1*CO2 experiment

(thick solid line) and 2*CO2 experiment (thin solid line).

ice formed during winter were higher and more ex-tensive, respectively, in the present control run than in the GFDL and NCAR control runs. The higher temperatures in low latitudes are partly attributable to the insufficient cloudiness for convective precipi-tation since, as noted in the previous section, clouds associated with shallow CUPREC and MLPREC were neglected in the radiation calculation of the

present experiments, resulting in an overestimation of solar heating in low latitudes. In fact, the tonal and annual averaged total cloudiness in the con-trol run was reduced throughout most of the tropo-sphere except for the lower troposphere of the high latitudes. The explanation as to why the present model simulated more extensive sea ice than the NCAR model is not clear since the snow and ice albedos used in the NCAR model are 0.8 and 0.7, respectively, which are greater than those used in the present model, as shown in the previous section.

4. Results

The annual mean global average changes due to doubled CO2, which were obtained from the means of the last two years of the 2*CO2 and 1*CO2 experiments, are as follows. The global surface tern-

Fig. 3. Latitude-height distribution of the annual mean zonal cloud change (%) due to the doubling of CO2.

perature increased by 4.3*. Evaporation and pre-cipitation showed an increase of 7.4%, while relative humidity at the surface decreased by 2%.

,4.1 Annual Mean Response Figure 2 shows the latitudinal distribution of the

annual mean rates of the various types of precip-itation; CUPREC, LSPREC, MLPREC and total

precipitation. The total precipitatton increases over most latitudes in response to the doubling of CO2.


Fig. 4. Latitude-time cross sections of the zonal mean precipitation (mm/day) for the 1*CO2 experiment

(left) and the differences (0.1mm/day) between the 1*CO2 and 2*CO2 experiments (right).

This is in agreement with the well-known scenario that CO2 warming causes greater evaporation, in-creases the moisture content of the air and results in greater amounts of precipitation. However, Fig. 2 re-veals that this scenario should not necessarily be ap-plied to all the components of the total precipitation. The CUPREC increases over most latitudes in ac-cordance with the above outline. In fact, it is found that the global mean increase in the total precipi-tation for the 2*CO2 model is 0.27mm/day. This value is almost equal to the increase in CUPREC

(=0.28mm/day). On the other hand, the MLPREC exhibits a decrease over the ITCZ while in middle latitudes, its peak values increase and the latitudes of the maximum are shifted poleward. The LSPREC also shows a similar shift in location but its inten-sity is nearly conserved between the 1*CO2 and 2*CO2 experiments. The poleward shift of the rain belt has also been found by Manabe and Wetherald (1980) using a simplified sector general circulation model. It can be concluded from Fig. 3 that the in-crease in CUPREC and the decrease in LSPREC


Fig. 5. Latitude-time cross section of the

change in soil moisture due to the doubling

of CO2. Contour intervals are 1cm. Neg-

ative values are shaded.

in low and middle latitudes results in a decrease in cloudiness. A response similar to Fig. 3 is also found for the relative humidity field (not shown).

4.2 Seasonal Response Latitude-time cross sections of the zonal mean

precipitation for the 1*CO2 experiment and the differences between the 1*CO2 and 2*CO2 exper-iments are presented in Fig. 4. The effect of the pole-ward shift of the middle latitude rain belt (Manabe and Wetherald, 1987) is easily seen in the difference field of the LSPREC in Fig. 4. Here, a decrease in the LSPREC is found equatorward while an increase is seen poleward of the rain belt. As a result, precip-itation is enhanced in the polar latitudes, in partic-ular, in the Southern Hemisphere, suggesting that the Antarctic ice mass would exhibit an increase as opposed to melting in a warmer climate, therefore contributing to the sea level decrease. On the other hand, the decrease of LSPREC equatorward of the rain belt is well correlated with the reduction of soil moisture in the middle latitudes as shown in Fig. 5. Moreover, the soil moisture during the winter of the control run was nearly saturated over exten-sive mid-continental regions of both North Amer-ica and Eurasia in the middle and high latitudes. As a result, in response to the doubling of CO2, soil moisture was reduced over these regions dur-ing the summer. This resulted since most of the enhanced winter precipitation was not used for sup-plying soil moisture but occurred as run off in the early spring, followed by a long spell of enhanced

evaporation (Meehl and Washington, 1988). These results are very similar to those presented in Manabe and Wetherald (1987).

On the other hand, it can be seen in Fig. 4 that the MLPREC and CUPREC exhibit responses quite dif-ferent from those for the LSPREC. In fact, the mag-nitudes of the increase of the MLPREC around the middle latitude rain belt are significantly reduced during summer, while those of the CUPREC are enhanced in the early summer. In the ITCZ, where both the CUPREC and MLPREC are intense, the CUPREC increases in the 2*CO2 experiment, but suppressed values of the MLPREC are found adja-cent to the peak increases of the CUPREC. This suggests the following interrelationship between the CUPREC and MLPREC responses. Surface warm-ing due to the doubling of CO2 preferably enhances deep penetrative convection, transporting surface heat energy to the upper troposphere. This heat transport reduces the lapse rate, and the deep heat-ing induces a downdraft in the surrounding region. These conditions are not conducive for middle level convection. However, it is not clear as to why the CUPREC increases on the northern side of the ITCZ while the MLPREC decreases on the southern side. A reversed response in the total precipitation is found in the results of the NCAR model but not in the GFDL model (see Fig. 19 of Schlesinger and Mitchell, 1987).

I.3 Response in precipitation area Figures 6 and 7 show scatter diagrams of the pre-

cipitation rate (mm/day) versus the ratio of the pre-cipitation area to the global domain. In order to obtain a frequency distribution of precipitation, the data were sampled at every time step (one hour) from January 1 to 10 (Fig. 6) and July 1 to 10 (Fig. 7) of the final years of the 1*CO2 and 2*CO2 experiments. The mean scattering is approximated by ellipses in each panel of the figures, with a thick lined ellipse representing the 1*CO2 and the thin lined ellipse denoting the 2*CO2 experiment.

It can be seen from Figs. 6 and 7 that the total precipitation increases but the precipitation area de-creases in both months when the CO2 concentration is doubled. The CUPREC rate increases as the area of precipitation increases. A clear reduction in the area of precipitation is found for the LSPREC, al-though little change can be seen in the precipitation rate. This feature is more significant during January than during July, reflecting a hemispherically differ-ent response. No definite changes are found for the MLPREC. Therefore, the reduction in the area and the increased rate of the total precipitation are at-tributable to the LSPREC and CUPREC responses, respectively. Similar scatter diagrams are illustrated in Figs. 8

(January) and 9 (July), representing three latitudi-


Fig. 6. Scatter diagrams of the precipitation rate (mm/day) versus the ratio (%) of the precipitation area to the global domain for January 1 to 10. Ellipses drawn with thick solid lines and thin solid lines denote the root mean square scattering for 1*CO2 and 2*CO2, respectively. Data points for 1*CO2

and 2*CO2 are denoted by crosses and dots, respectively.

nal zones, 90*S*28*S, 28*S*28*N and 28*N*90*N. Again, responses similar to the global domain can be seen in each zone except for the tropical zone dur-ing July, where convective precipitation (CUPREC and MLPREC) dominates the total precipitation. For responses in zones narrower than those shown in Figs. 8 and 9, similar scatter diagramss for smaller latitude bands were made (not shown). The dia-grams reveal an increase in the LSPREC area in the high latitudes of the Northern Hemisphere. This is consistent with the increase in cloudiness found over this region in the present (Fig. 3) and other experi-ments (e.g., see Fig. 23 in Schlesinger and Mitchell, 1987). Comparing Figs. 8 with 9, one can find a good

qualitative agreement in the area response between summer minus winter and 2*CO2-1*CO2. In addition to this, Hansen et al. (1984) and Wether-ald and Manabe (1986) found good similarity be-

tween the latitude-height distribution of the differ-ence in tonal mean cloud amount obtained from an increase of the solar constant and that obtained from an increase of CO2 concentration. These facts suggest that the main factor that determines the cloud area response on global scales is tem-

perature. Since the precipitation area and cloudi-ness are correlated, they also invoke the hypothesis that, on global scales, precipitation increases but cloudiness decreases as the temperature increases. There are some observations which support this hy-

pothesis. Table 1 shows differences in total cloudi-ness and precipitation between summer.- and win-ter for three zones, 90*S*30*S, 30*S*30*N and 30*N*90*N . These values were obtained by sub-tracting the monthly averaged cloudiness and pre-cipitation in January from the corresponding values in July in the Northern Hemisphere, while subtract-ing the July from the January values for the South-


Fig. 7. As in Fig. 6 but for July 1 to 10.

Table 1. Differences in cloudiness and precipitation between summer and winter, obtained by subtracting the monthly averaged cloudiness and precipitation values for January (July) from the corresponding

values for July (January) in the Northern (Southern) Hemisphere.

emn Hemisphere. The cloudiness given by Dopplick (1979) and the International Satellite Cloud Cli-matology Project (ISCCP; July 1983 and January 1984) decreases during summer in the middle and high latitudes of both hemispheres, while the precip-itation reported by Jaeger (1976) increases in these latitudes in the Northern Hemisphere. However, op-

posite responses are found in data of cloudiness given by Berlyand and Strokina (1980) and those of precipitation by Jaeger (1976) for the Southern Hemisphere. Moreover, all of the data shown in Ta-ble 1 indicate that both the cloudiness and precip-itation are enhanced over the low latitudes during summer.


Fig. 8. As in Fig. 6 but for the latitudinal zones, 28*N*90*N (top), 28*S*28*N (middle) and 90*S*28*S

(bottom). T denotes the total precipitation, C the CUPREC, M the MLPREC and L the LSPREC.

It should be noted that in the present, the NCAR and the Goddard Institute for Space Flight (GISS) models predicted decreases in cloudiness throughout the tropical troposphere when the CO2 concentra-tion was doubled, which is not consistent with the observations of decreasing cloudiness as temperature increases. In this respect, Wetherald and Manabe (1988) stressed the importance of the vertical res-olution of the tropopause and lower stratospheric clouds on simulating the response of cloud amount

in the tropics. Actually, clouds were not permit-ted to form near the tropopause or lower strato-sphere in the low latitudes of the NCAR, GISS and the tropospheric version of the MRI models. It is speculated that if clouds had been allowed to form at these higher levels, then these models might have obtained, at least qualitatively, similar results as the Wetherald and Manabe (1988) and Wilson and Mitchell (1987) studies. They found a gen-eral decrease of middle and upper tropospheric cloud


Fig. 9. As in Fig. 8 but for July 1 to 10.

but a significant increase of high cloud cover around

the tropopause in low latitudes, while both the

NCAR and GISS models produced an increase of

high clouds around the tropopause in the middle

to higher latitudes. The discrepancy of the MRI

model in this respect, along with others, may be attributable to the use of the tropospheric model.

Indeed, it appears that GCMs which have been run

more recently at the GFDL, without the above re-

striction on cloud formation, generally reproduce an

increase of high clouds at all latitudes in response to

a global warming (Wetherald, 1989; personal com-munication).

4.4 Response in Me frequency distribution Figures 10 and 11 show the cumulative frequency

distribution of the precipitation for the same data used in Fig. 8. Plotted is the log-frequency of pre-cipitation exceeding a given precipitation rate per one hour (Fig. 10) or per one consecutive precipita-tion (Fig. 11) at each grid point. It can be seen that the functional form of the cumulative frequency dis-


Fig. 10. Cumulative frequency distribution for precipitation rates of mm per hour for January 1 to 10.

Plotted is the log-frequency of precipitation exceeding a given rate per one hour at each grid point.

Data points for 1*CO2 and 2*CO2 are denoted by crosses and dots, respectively.


Fig. 11. As in Fig. 10 but for precipitation rates of mm per one consecutive precipitation event. Plotted is

the log-frequency of precipitation exceeding a given rate per one consecutive precipitation at each grid

point.


tribution depends on the type of precipitation, lat-itude and sampling (one hour or consecutive) but it is almost independent of the CO2 concentration. In general, when the CO2 concentration is doubled, convective precipitation (CUPREC and MLPREC) of high precipitation rates increase, but the reverse is true for non-convective precipitation (LSPREC).

Most of the cumulative frequency distributions in the range for high precipitation rates decrease more rapidly than power law curves, i, e., na (*), where a is a constant and n(*) is the cumulative frequency exceeding the precipitation * (mm). The only distri-bution that obeys the power law is the consecutive MLPREC in low latitudes. It is of interest to inves-tigate what processes determine the functional form of the distribution but this is beyond the scope of the present note.

5. Summary

The MRI*GCM-I coupled with a simple mixed ocean model has been used to investigate the ef-fect of doubling the CO2 concentration on convec-tive and non-convective types of precipitation. The following changes in precipitation have been found when the CO2 concentration was doubled.

1) The increase of the global mean total precipitation was attributed to the increase in

the CUPREC (precipitation due to cumulus convection rooted in the planetary boundary

layer). 2) In the ITCZ, the MLPREC (precipitation due

to convective adjustment in the free atmosphere) decreased significantly at the latitudes

adjacent to where the CUPREC showed an increase. In the mid-latitude rain belt, the increased rate of the CUPREC had a peak dur-

ing early summer, but that of the MLPREC was minimized in the summer.

3) The decrease in the area of total precipitation was attributed to the reduction in the LSPREC

(precipitation due to large-scale condensation) area. The area response obtained by increasing

the CO2 concentration is similar to that found between summer and winter, suggesting that

the main factor that determines the response is temperature.

4) The frequency of intense convective rainfall increases but that of intense non-convective rain-

fall decreases. The present model does not take into account the horizontal oceanic heat transport and also neglects cloudiness due to shallow cumulus convection and stratus in the planetary boundary layer. These re-strictions may have affected the above results to some extent. A new experiment excluding the re-strictions is now under way.

Acknowledgements

The authors are grateful to K. Kiriyama, the head of the climate research division of the Meteorologi-cal Research Institute (MRI), for his encouragement, and to R.T. Wetherald, T. Nitta and Y. Hayashi for valuable comments on the manuscript. Thanks are extended to T. Nagai for processing the ISCCP data and to members of the MRIGCM group for helpful discussions. Computations were made on the HI-TAC S810/10 at the MRI.

References

Arakawa, A., 1972: Design of the UCLA Atmospheric General Circulation Model. Numerical Simulation of

Weather and Climate. Tech. Rept. No. 7, Dept. of Meteorology, UCLA, 116pp.

Arakawa, A. and W.H. Schubert, 1974: Interaction of a cumulus cloud ensemble with the large-scale environ-

ment. Part I. J. Atmos. Sci., 31, 674-701. Arakawa, A. and Y. Mintz, 1974: The UCLA Atmo-

spheric General Circulation Model, Notes distributed at the workshop. 25 March-4 April 1974, UCLA.

Berlyand, T.G. and L.A. Strokina, 1980: Global Distri- bution of Total Cloud Amount. Leningrad Gidrome-

teoizdat. Cess, R.D. and G.L. Potter, 1988: A methodology for

understanding and intercomparing climate feedback processes in general circulation models. J. Geophys. Res., 93, 8305-8314.

Dopplick, T.G., 1979: Radiative heating of the global atmosphere: Corrigendum. J. Atmos. Sci., 36,1812-

1817. Hansen, L., A. Lacis, D. Rind, G. Russell, P. Stone,

I. Fund, R. Ruedy and J. Lerner, 1984: Analysis of feedback mechanisms, in Climate Processes and

Climate Sensitivity, Geophys. Monogr. Ser., vol. 29, edited by J.E. Hansen and T. Takahashi, pp.130-163,

AGU, Washington, D.C. Jaeger, L., 1976: Monatskarten des Niederschlags fur die

ganze Erde, Ber. Dtsch. Wetterdienstes, 139, 1-38. Jaeger, J., 1988: Developing policies for responding to cli-

matic change, (to be published in the World Climate Programme publications series).

Katayama, A., 1972: A simplified scheme for computing radiative transfer in the troposphere. Tech. Rept. No.

6, Dept. Met. UCLA, 77pp. Kitoh, A., K. Yamazaki and T. Tokioka, 1988: Influence

of soil moisture and surface albedo changes over the African tropical rain forest on summer climate inves-

tigated with the MRI.GCM-I. J. Meteor. Soc. Japan, 66, 65-86.

Lord, S.J., 1982: Interaction of a cumulus ensemble with the large-scale environment. Part III: Semi-

prognostic test of the Arakawa-Schubert cumulus parameterization. J. Atmos. Sci., 39, 88-103.

Lord, S.J. and A. Arakawa, 1980: Interaction of a cumulus ensemble with the large-scale environment. Part

II. J. Atmos. sci., 37, 2677-2692. Lord, S.J., W.C. Chao and A. Arakawa 1982: Interaction

of a cumulus ensemble with the large-scale environment. Part IV: The discrete model. J. Atmos. Sci.,


39, 104-113. Manabe, S. and R.T. Wetherald, 1980: On the distri-

bution of climate change resulting from a increase in CO2 content of the atmosphere. J. Atmos. Sci., 37,

99-118. Manabe, S. and R.T. Wetherald, 1987: Large-scale

changes of soil wetness induced by an increase in atmospheric carbon dioxide. J. Atmos. Sci., 44, 1211-

1235. Matthews, E., 1983: Global vegetation and land use: new

high-resolution data bases for climate studies. J. At- mos. Sci., 22, 474-487.

Meehl, G.E. and W.M. Washington, 1988: A compari- son of soil-moisture sensitivity in two global climate

models. J. Atrnos. Sci., 45, 1476-1492. Posey, J.W. and P.F. Clapp, 1964: Global distribution

of normal surface albedo. Geofisica Irnternacional, 4, 33-48.

Rutledge, S.A. and M.E. Schlesinger, 1985: The treatment of clouds in general circulation models: current

status and more physically based parameterizations. Report No. 63, Climatic Research Institute, Oregon

State University, Corvallis, OR, 49pp. Semtner, A.J., 1976: A model for the thermodynamicc

growth of sea ice in numerical investigations of cli-

mate. J. Phys. Oceanogr., 6, 379-389. Schlesinger, M.E. and J.F.B. Mitchell, 1987: Climate

model simulations of the equilibrium climatic response to increased carbon dioxide, Rev. Grophysi.,

25, 760-798. Tokioka, T., K. Yamazaki, I. Yagai and A. Kitoh, 1984:

A description of the Meteorological Research Institute atmospheric general circulation model (MRI.GCM-I).

Technical Report of the Meteorological Research Insti- tute, No. 13, MRI, Tsukuba, 249pp.

Washington, W.M. and G.A. Meehl,1984: Seasonal cycle experiment on the climate sensitivity due to a doubling of CO2 with an atmospheric general circulation

model coupled to a simple mixed-layer ocean model. J. Atmos. Sci., 89, 9475-9503.

Wetherald, R.T. and S. Manabe, 1986: An investigation of cloud cover change in response to thermal forcing.

Gum. Change, 8, 5-23. Wetherald, R.T. and S. Manabe, 1988: Cloud feedback

processes in a general circulation model. J. Atmos. Sc., 45, 1397-1415.

Wilson, C.A. and J.F.B. Mitchell, 1987: A doubled COz climate sensitivity experiment with a global climate

model including a simple ocean. J. Atmos. Sci., 92, 13315-13343.

大気・海洋混合層モデルで得られた二酸化炭素倍層が

対流性・非対流性降水に及ぼす効果

野田彰・時岡達志

(気象研究所)

notes and correspondence the effect of doubling the co2

Documents