• Alexander, R. C., , and R. L. Mobley, 1976: Monthly average sea-surface temperatures and ice pack limits on a global grid. Mon. Wea. Rev., 104 , 143148.

    • Search Google Scholar
    • Export Citation
  • Arakawa, A., 1975: Modeling clouds and cloud processes for use in climate models. The Physical Basis of Climate and Climate Modeling, GARP Publishing Series, Vol. 16, WMO, 181–197.

    • Search Google Scholar
    • Export Citation
  • Arakawa, A., , and W. H. Schubert, 1974: Interaction of a cumulus cloud ensemble with the large-scale environment, Part I. J. Atmos. Sci., 31 , 674701.

    • Search Google Scholar
    • Export Citation
  • Barkstrom, B. R., 1984: Earth Radiation Budget Experiment (ERBE). Bull. Amer. Meteor. Soc., 65 , 11701185.

  • Boer, G. J., and Coauthors. 1992: Some results from an intercomparison of the climates simulated by 14 atmospheric general circulation models. J. Geophys. Res., 97 , 1277112786.

    • Search Google Scholar
    • Export Citation
  • Bony, S., , and K. Emanuel, 2001: A parameterization of the cloudiness associated with cumulus convection: Evaluation using TOGA COARE data. J. Atmos. Sci., 58 , 31583183.

    • Search Google Scholar
    • Export Citation
  • Cess, R. D., , M. H. Zhang, , B. A. Wielicki, , D. F. Young, , X-L. Zhou, , and Y. Nikitenko, 2001: The influence of the 1998 El Niño upon cloud–radiative forcing over the Pacific warm pool. J. Climate, 14 , 21292137.

    • Search Google Scholar
    • Export Citation
  • Chelliah, M., , and P. Arkin, 1992: Large-scale variability of monthly outgoing longwave radiation anomalies over the global Tropics. J. Climate, 5 , 371389.

    • Search Google Scholar
    • Export Citation
  • Chou, M. D., , M. J. Suarez, , C. H. Ho, , M. M-H. Yan, , and K-T. Lee, 1998: Parameterizations for cloud overlapping and shortwave single-scattering properties for use in general circulation and cloud ensemble models. J. Climate, 11 , 202214.

    • Search Google Scholar
    • Export Citation
  • Coakley Jr., J. A., , and P. Chylek, 1975: The two-stream approximation in radiative transfer: Including the angle of the incident radiation. J. Atmos. Sci., 32 , 409418.

    • Search Google Scholar
    • Export Citation
  • Collins, W. D., 2001: Parameterization of generalized cloud overlap for radiative calculations in general circulation models. J. Atmos. Sci., 58 , 32243242.

    • Search Google Scholar
    • Export Citation
  • Deardorff, J. W., 1972: Parameterization of the planetary boundary layer for use in general circulation models. Mon. Wea. Rev., 100 , 93106.

    • Search Google Scholar
    • Export Citation
  • Dorman, J. L., , and P. J. Sellers, 1989: A global climatology of albedo, roughness length and stomatal resistance for atmospheric general circulation models as represented by the Simple Biosphere model (SiB). J. Appl. Meteor., 28 , 833855.

    • Search Google Scholar
    • Export Citation
  • Dowling, D. R., , and L. F. Radke, 1990: A summary of the physical properties of cirrus clouds. J. Appl. Meteor., 29 , 970978.

  • Fu, Q., 1991: Parameterization of radiative processes in vertically nonhomogeneous multiple scattering atmospheres. Ph.D. dissertation, University of Utah, 259 pp.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., 1996: An accurate parameterization of the solar radiative properties of cirrus clouds for climate models. J. Climate, 9 , 20582082.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., , and K. N. Liou, 1992: On the correlated k-distribution method for radiative transfer in nonhomogeneous atmospheres. J. Atmos. Sci., 49 , 21392156.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., , and K. N. Liou, 1993: Parameterization of the radiative properties of cirrus clouds. J. Atmos. Sci., 50 , 20082025.

  • Fu, Q., , K. N. Liou, , M. C. Cribb, , T. P. Charlock, , and A. Grossman, 1997: Multiple scattering parameterization in thermal infrared radiative transfer. J. Atmos. Sci., 54 , 27992812.

    • Search Google Scholar
    • Export Citation
  • Geleyn, J-F., , and A. Hollingsworth, 1979: An economical analytical method for the computation of the interaction between scattering and line absorption of radiation. Beitr. Phys. Atmos., 52 , 116.

    • Search Google Scholar
    • Export Citation
  • Gu, Y., , and K. N. Liou, 2000: Interactions of radiation, microphysics, and turbulence in the evolution of cirrus clouds. J. Atmos. Sci., 57 , 24632479.

    • Search Google Scholar
    • Export Citation
  • Gu, Y., , and K. N. Liou, 2001: Radiation parameterization for three-dimensional inhomogeneous cirrus clouds: Application to climate models. J. Climate, 14 , 24432457.

    • Search Google Scholar
    • Export Citation
  • Harshvardhan, , R. Davies, , D. A. Randall, , and T. G. Corsetti, 1987: A fast radiation parameterization of atmospheric circulation models. J. Geophys. Res., 92 , 10091016.

    • Search Google Scholar
    • Export Citation
  • Hense, A., , and E. Heise, 1984: A sensitivity study of cloud parameterizations in general circulation models. Beitr. Phys. Atmos., 57 , 240258.

    • Search Google Scholar
    • Export Citation
  • Heymsfield, A. J., , and C. M. R. Platt, 1984: A parameterization of the particle size spectrum of ice clouds in terms of the ambient temperature and ice water content. J. Atmos. Sci., 41 , 846855.

    • Search Google Scholar
    • Export Citation
  • Johnson, D. R., 1997: “General coldness of climate models” and the second law: Implications for modeling the earth system. J. Climate, 10 , 28262846.

    • Search Google Scholar
    • Export Citation
  • Köhler, M., 1999: Explicit prediction of ice clouds in general circulation models. Ph.D. dissertation, University of California, Los Angeles, 167 pp.

    • Search Google Scholar
    • Export Citation
  • Li, D-M., , and K. P. Shine, 1995: A 4-dimensional ozone climatology for UGAMP models. UGAMP Internal Rep. 35.

  • Li, J-L. F., , M. Köhler, , J. D. Farrara, , and C. R. Mechoso, 2002: The impact of stratocumulus cloud radiative properties on surface heat fluxes simulated with a general circulation model. Mon. Wea. Rev., 130 , 14331441.

    • Search Google Scholar
    • Export Citation
  • Liang, X-Z., , and W-C. Wang, 1997: Cloud overlap effects on general circulation model climate simulations. J. Geophys. Res., 102 , 1103911047.

    • Search Google Scholar
    • Export Citation
  • Liou, K. N., 1986: Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev., 114 , 11671199.

  • Liou, K. N., , Q. Fu, , and T. P. Ackerman, 1988: A simple formulation of the delta-four-stream approximation for radiative transfer parameterizations. J. Atmos. Sci., 45 , 19401947.

    • Search Google Scholar
    • Export Citation
  • Lord, S. J., , H. E. Willoughby, , and J. M. Piotrowicz, 1984: Role of a parameterized ice-phase microphysics in an axisymmetric nonhydrostatic tropical cyclone model. J. Atmos. Sci., 41 , 28362848.

    • Search Google Scholar
    • Export Citation
  • Manabe, S., , and R. Strickler, 1964: Thermal equilibrium of the atmospheres with a convective adjustment. J. Atmos. Sci., 21 , 361385.

  • Pan, D-M., , and D. A. Randall, 1998: A cumulus parametrization with a prognostic closure. Quart. J. Roy. Meteor. Soc., 124 , 949981.

  • Ramanathan, V., , E. J. Pitcher, , R. C. Malone, , and M. L. Blackmon, 1983: The response of a spectral general circulation model to refinements in radiative processes. J. Atmos. Sci., 40 , 605630.

    • Search Google Scholar
    • Export Citation
  • Randall, D. A., 1976: The interaction of the planetary boundary layers with large-scale circulations. Ph.D. dissertation, University of California, Los Angeles, 247 pp.

    • Search Google Scholar
    • Export Citation
  • Randall, D. A., 1980a: Conditional instability of the first kind upside-down. J. Atmos. Sci., 37 , 125130.

  • Randall, D. A., 1980b: Entrainment into a stratocumulus layer with distributed radiative cooling. J. Atmos. Sci., 37 , 148159.

  • Randall, D. A., , Harshvardhan, , D. A. Dazlich, , and T. G. Corsetti, 1989: Interactions among radiation, convection, and large-scale dynamics in a general circulation model. J. Atmos. Sci., 46 , 19431970.

    • Search Google Scholar
    • Export Citation
  • Rayner, N. A., , C. K. Folland, , D. E. Parker, , and E. B. Horton, 1995: A new global sea-ice and sea surface temperature (GISST) data set for 1903–1994 for forcing climate models. Hadley Centre, Met Office, Internal Note No. 69, 14 pp.

    • Search Google Scholar
    • Export Citation
  • Sasamori, T., 1975: A statistical model for stationary atmospheric cloudiness, liquid water content, and rate of precipitation. Mon. Wea. Rev., 103 , 10371049.

    • Search Google Scholar
    • Export Citation
  • Slingo, A., , and J. M. Slingo, 1988: The response of a general circulation model to cloud longwave radiative forcing. Part I: Introduction and initial experiments. Quart J. Roy. Meteor. Soc., 114 , 10271062.

    • Search Google Scholar
    • Export Citation
  • Slingo, A., , and J. M. Slingo, 1991: Response of the National Center for Atmospheric Research community climate model to improvements in the representation of clouds. J. Geophys. Res., 96 , 1534115347.

    • Search Google Scholar
    • Export Citation
  • Slingo, J. M., 1980: A cloud parameterization scheme derived from GATE data for use with a numerical model. Quart. J. Roy. Meteor. Soc., 106 , 747770.

    • Search Google Scholar
    • Export Citation
  • Smagorinsky, J., 1960: On the dynamical prediction of large-scale condensation by numerical methods. Geophys. Monogr., No. 5, Amer. Geophys. Union, 71–78.

    • Search Google Scholar
    • Export Citation
  • Smith, T. M., , and R. W. Reynolds, 1998: A high-resolution global sea surface temperature climatology for the 1961–90 base period. J. Climate, 11 , 33203323.

    • Search Google Scholar
    • Export Citation
  • Suarez, M. J., , A. Arakawa, , and D. A. Randall, 1983: The parameterization of the planetary boundary layer in the UCLA general circulation model: Formulation and results. Mon. Wea. Rev., 111 , 22242243.

    • Search Google Scholar
    • Export Citation
  • Sundqvist, H., 1978: A parameterization scheme for non-convective condensation including prediction of cloud water content. Quart. J. Roy. Meteor. Soc., 104 , 677690.

    • Search Google Scholar
    • Export Citation
  • Tian, L., , and J. A. Curry, 1989: Cloud overlap statistics. J. Geophys. Res., 94 , 99259935.

  • Tiedtke, M., 1993: Representation of clouds in large-scale models. Mon. Wea. Rev., 121 , 30403061.

  • Trenberth, K. E., , G. W. Branstator, , D. Karoly, , A. Kumar, , N-C. Lau, , and C. Ropelewski, 1998: Progress during TOGA in understanding and modeling global teleconnections associated with tropical sea surface temperatures. J. Geophys. Res., 103 , 1429114324.

    • Search Google Scholar
    • Export Citation
  • Wu, T., , W. R. Cotton, , and W. Y. Y. Cheng, 2000: Radiative effects on the diffusional growth of ice particles in cirrus clouds. J. Atmos. Sci., 57 , 28922904.

    • Search Google Scholar
    • Export Citation
  • Xu, K-M., , and S. K. Krueger, 1991: Evaluation of cloudiness parameterizations using a cumulus ensemble model. Mon. Wea. Rev., 119 , 342367.

    • Search Google Scholar
    • Export Citation
  • Xu, K-M., , and D. A. Randall, 1996a: A semiempirical cloudiness parameterization for use in climate models. J. Atmos. Sci., 53 , 30843102.

    • Search Google Scholar
    • Export Citation
  • Xu, K-M., , and D. A. Randall, 1996b: Evaluation of statistically based cloudiness parameterizations used in climate models. J. Atmos. Sci., 53 , 31033119.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Sample cloud configurations that are consistent with the vertical profile of the partial cloudiness and the overlap assumptions

  • View in gallery

    Jan mean total cloud cover (%) for simulations (a) NRAD, (b) CTRL, and (c) ISCCP observations

  • View in gallery

    Zonal mean total cloud cover differences from observations (%) for NRAD (solid) and CTRL (dashed) for (a) Jan and (b) Jul

  • View in gallery

    Jul mean total precipitation (mm day–1) for simulations (a) NRAD, (b) CTRL, and (c) CMAP observations

  • View in gallery

    Jan mean OLR (W m–2) for simulations (a) NRAD, (b) CTRL, and (c) Arkin observations (Chelliah and Arkin 1992)

  • View in gallery

    Jan horizontal distributions of (a) ice and (b) liquid water contents (mg m–3) in the NRAD simulation

  • View in gallery

    Jan zonal mean of (a) ice and (b) liquid water mixing ratios (10–6 kg kg–1) in the NRAD simulation

  • View in gallery

    Observed SSTs from the DJF dataset provided by Smith and Reynolds (1998) for (a) 1997/98 El Niño, (b) 1998/99 La Niña, and (c) their differences

  • View in gallery

    Differences in DJF precipitation (mm day–1) between 1997/98 El Niño and 1998/99 La Niña for the (a) AGCM NRAD simulation and (b) CMAP observations

  • View in gallery

    Differences in DJF OLR (W m–2) between 1997/98 El Niño and 1998/99 La Niña for the (a) AGCM NRAD simulation and (b) NOAA interpolated data

  • View in gallery

    Differences in the DJF simulated (a) total cloud water mixing ratio (kg kg–1) and (b) cloud cover (%) between 1997/98 El Niño and 1998/99 La Niña. The figure is presented in a longitude–height projection and displays values averaged between 30°S and 30°N

  • View in gallery

    Differences in the DJF simulated (a) shortwave, (b) longwave, and (c) net radiative heating rates (K day–1) between 1997/98 El Niño and 1998/98 La Niña. The figure is presented in a longitude–height projection and displays values averaged between 30°S and 30°N

  • View in gallery

    Jan zonal mean differences of (a) ice, (b) liquid, and (c) total water mixing ratios (kg kg–1) between NRAD and CTRL

  • View in gallery

    Jan zonal mean differences of (a) solar, (b) IR, and (c) net radiative heating rates (K day–1) between NRAD and CTRL

  • View in gallery

    Zonal mean temperature differences (K) between NRAD and CTRL for (a) Jan and (b) Jul

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 30 30 4
PDF Downloads 23 23 2

Parameterization of Cloud–Radiation Processes in the UCLA General Circulation Model

View More View Less
  • 1 Department of Atmospheric Sciences, University of California, Los Angeles, Los Angeles, California
© Get Permissions
Full access

Abstract

A contemporary radiation parameterization scheme has been implemented in the University of California, Los Angeles (UCLA), atmospheric GCM (AGCM). This scheme is a combination of the delta-four-stream method for solar flux transfer and the delta-two-and-four-stream method for thermal infrared flux transfer. Both methods have been demonstrated to be computationally efficient and at the same time highly accurate in comparison with exact radiative transfer computations. The correlated-k distribution method for radiative transfer has been used to represent gaseous absorption in multiple-scattering atmospheres. The single-scattering properties for ice and water clouds are parameterized in terms of ice/liquid water content and mean effective size/radius. In conjunction with the preceding radiative scheme, parameterizations for fractional cloud cover and cloud vertical overlap have also been devised in the model in which the cloud amount is determined from the total cloud water mixing ratio. For radiation calculation purposes, the model clouds are vertically grouped in terms of low, middle, and high types. Maximum overlap is first used for each cloud type, followed by random overlap among the three cloud types. The preceding radiation and cloud parameterizations are incorporated into the UCLA AGCM, and it is shown that the simulated cloud cover and outgoing longwave radiation fields without any special tuning are comparable with those of International Satellite Cloud Climatology Project (ISCCP) dataset and derived from radiation budget experiments. The use of the new radiation and cloud schemes enhances the radiative warming in the mid- to upper tropical troposphere and alleviates the cold bias that is common to many AGCMs. Sensitivity studies show that ice crystal size and cloud inhomogeneity significantly affect the radiation budget at the top of the atmosphere and the earth’s surface.

Corresponding author address: Dr. Y. Gu, Department of Atmospheric Sciences, University of California, Los Angeles, 405 Hilgard Ave., Los Angeles, CA 90095. Email: gu@atmos.ucla.edu

Abstract

A contemporary radiation parameterization scheme has been implemented in the University of California, Los Angeles (UCLA), atmospheric GCM (AGCM). This scheme is a combination of the delta-four-stream method for solar flux transfer and the delta-two-and-four-stream method for thermal infrared flux transfer. Both methods have been demonstrated to be computationally efficient and at the same time highly accurate in comparison with exact radiative transfer computations. The correlated-k distribution method for radiative transfer has been used to represent gaseous absorption in multiple-scattering atmospheres. The single-scattering properties for ice and water clouds are parameterized in terms of ice/liquid water content and mean effective size/radius. In conjunction with the preceding radiative scheme, parameterizations for fractional cloud cover and cloud vertical overlap have also been devised in the model in which the cloud amount is determined from the total cloud water mixing ratio. For radiation calculation purposes, the model clouds are vertically grouped in terms of low, middle, and high types. Maximum overlap is first used for each cloud type, followed by random overlap among the three cloud types. The preceding radiation and cloud parameterizations are incorporated into the UCLA AGCM, and it is shown that the simulated cloud cover and outgoing longwave radiation fields without any special tuning are comparable with those of International Satellite Cloud Climatology Project (ISCCP) dataset and derived from radiation budget experiments. The use of the new radiation and cloud schemes enhances the radiative warming in the mid- to upper tropical troposphere and alleviates the cold bias that is common to many AGCMs. Sensitivity studies show that ice crystal size and cloud inhomogeneity significantly affect the radiation budget at the top of the atmosphere and the earth’s surface.

Corresponding author address: Dr. Y. Gu, Department of Atmospheric Sciences, University of California, Los Angeles, 405 Hilgard Ave., Los Angeles, CA 90095. Email: gu@atmos.ucla.edu

1. Introduction

Theoretical, observational, and modeling studies have demonstrated the important role of cloud–radiation interactions in climate variability. The atmospheric general circulation model (AGCM) studies of Ramanathan et al. (1983), Slingo and Slingo (1988, 1991), and Randall et al. (1989) have shown that radiation, latent heat release, and small-scale transport are of equal importance in the cloud–climate problem and that many features of the simulated climate are sensitive to the treatment of clouds and radiation in the model. The cloud–radiation interaction and feedback problem has been identified as the highest priority item in climate research nationally and internationally, as illustrated by a number of composite field and satellite observations, including the International Satellite Cloud Climatology Project (ISCCP), the First ISCCP Regional Experiment (FIRE), the Atmospheric Radiation Measurement Program (ARM), and the Earth Radiation Budget Experiment (ERBE).

The parameterization of cloud–radiation processes in climate models is a complex task. Radiative transfer in the atmosphere is determined by spectrally dependent optical properties. Calculation of the radiative heating/cooling in clouds is complicated due to difficulties in parameterizing their single-scattering properties, especially those of ice clouds due to complexities in the ice crystal size, shape, and orientation, which cannot be determined from the models (Liou 1986). Furthermore, clouds have a typical length scale of perhaps several hundred meters, and display substantial horizontal variability on scales that are generally smaller than the usual AGCM grid box. A specific challenge in studying the cloud–radiation feedback on climate is to estimate the uncertainty in the vertical profile of cloud cover, the information of which is extremely limited in present satellite and surface observations.

Determination of the cloud fraction in climate models is also difficult because condensation processes frequently occur at scales much smaller than an AGCM grid box. Consequently, these processes need to be parameterized in terms of the large-scale variables. There are two approaches to the parameterization of cloudiness in climate models as discussed by Bony and Emanuel (2001). One is to parameterize the cloud fraction as a function of prognostic variables such as relative humidity (e.g., Slingo 1980); the other is to represent the subgrid-scale variability of conserved variables by a probability distribution function (PDF; e.g., Sasamori 1975). Xu and Randall (1996a,b) using a cloud ensemble model (CEM) showed that mean relative humidity is a poor predictor of upper-level cloudiness associated with convection, and that PDF moments in convective regions rely on assumptions not supported by CEM simulations.

The objective of the present research is to improve the computation of radiative transfer processes in the current University of California, Los Angeles (UCLA), AGCM by employing a physically based efficient radiation scheme that can better determine the cloud optical properties and provide more accurate radiative heating fields. In addition, we wish to use the improved AGCM to study cloud–radiation interactions and feedbacks in interannual and intraseasonal climate variations.

A brief description of the UCLA AGCM is given in section 2, followed by a presentation of the new cloud and radiation schemes in section 3. An evaluation of the performance of the new schemes is presented in section 4. In section 5, we investigate the sensitivity of simulated climate to changes in cloud and radiation schemes. Finally, a summary is given in section 6.

2. Overview of the UCLA AGCM

The UCLA AGCM is a state-of-the-art gridpoint model of the global atmosphere. The model prognostic variables are the horizontal wind, potential temperature, mixing ratios of water vapor, cloud liquid water and ice water, planetary boundary layer (PBL) depth, surface pressure, land surface temperature, and snow depth over land. The horizontal finite differencing of the primitive equations is based on a staggered Arakawa “C” grid scheme, while the vertical coordinate employed is the modified sigma coordinate developed by Suarez et al. (1983). For the time integration of prognostic variables, a leapfrog time-differencing scheme is used with a Matsuno step regularly inserted. The PBL is parameterized as a well-mixed layer of variable depth following the work of Deardorff (1972), Randall (1976, 1980a,b) and Suarez et al. (1983) as recently modified by Li et al. (2002). Parameterization of cumulus convection and its interaction with the PBL follows Arakawa and Schubert (1974) and Lord et al. (1984) as modified to predict cloud kinetic energy by Pan and Randall (1998). The geographical distribution of sea surface temperature (SST) is prescribed based on a 31-yr (1960–90) climatology corresponding to the Global Sea Ice and Sea Surface Temperature dataset (GISST) version 2.2 (Rayner et al. 1995). Sea ice thickness and extent are prescribed using data presented by Alexander and Mobley (1976). Surface albedo and roughness length are specified following Dorman and Sellers (1989). Daily values of the surface conditions are determined from the monthly mean values by linear interpolation. Ozone (O3) mixing ratios are prescribed as a function of latitude, height, and time based on the 1985–90 climatology (Li and Shine 1995). In this paper we use the low-resolution AGCM version, which is 4° latitude × 5° longitude with 15 layers from the earth’s surface to 1 hPa.

Two cloud “types” are generated by the model. The first is free atmosphere clouds, whose main sources are grid-scale supersaturation and cumulus detrainment. In the current UCLA AGCM, a grid box of the free atmosphere is assumed to be entirely cloudy (i.e., the cloud fraction is 1) if the total cloud water mixing ratio (qt) satisfies
qtqlqi–10–1
where ql and qi are liquid and ice water mixing ratio, respectively, both of which are prognostic variables of the UCLA AGCM (Köhler 1999); otherwise the grid box is assumed to be cloud free (i.e., the cloud fraction is 0). The selected threshold for cloud formation is several orders of magnitude smaller than the typically observed cloud liquid and ice water mixing ratios (Dowling and Radke 1990). The second cloud type is the PBL clouds, which are generated at the PBL top when this is above the condensation level. PBL clouds are assigned a cloud fraction that increases linearly with pressure thickness to become 1 at and above 12.5 mb.

In the control version of the UCLA AGCM we use, the solar and infrared (IR) radiation processes are parameterized following Harshvardhan et al. (1987). This scheme considers Rayleigh scattering, absorption by water vapor and ozone, and the effects of clouds. The calculation of solar radiation neglects the absorption by oxygen (O2), carbon dioxide (CO2), and aerosols. The solar radiative transfer is solved based on a two-stream approximation (Coakley and Chylek 1975), and the cloud optical properties are assumed to be constant for all solar wavelengths, except that the single-scattering albedos of cloud particles for wavelengths shorter than 0.9 µm are taken to be 1. The calculation of IR radiative heating includes the effects of water vapor (H2O), CO2, and O3, and uses one to six bands to resolve the IR spectrum. The IR radiative transfer through clouds is parameterized using a broadband cloud emissivity and the scattering of IR radiation by cloud particles is neglected. For the solar radiation computation, the fractional cloudiness in the PBL is taken into account by simply linearly scaling the cloud optical thickness.

3. The new cloud–radiation parameterizations

a. Radiative transfer calculations

The radiation scheme we implement in the UCLA AGCM is based on the Fu–Liou scheme. The detailed description of the model and the relevant coefficients and parameters used in this scheme are given in Fu and Liou (1992, 1993). Fu et al. (1997) showed that the delta-two-stream method is the most computionally efficient but produces significant errors in fluxes and heating rates under cloudy conditions. High accuracy can be obtained by using the delta-four-stream method, but substantial computer time is required for the calculation of thermal infrared radiative transfer. The delta-two-and-four-stream combination method is sufficiently economical for IR calculations (4 times faster than the delta-four-stream method, and only 50% more than the two-stream method), and at the same time it produces acceptable accuracy under most atmospheric conditions. In view of the above, we implement in the UCLA AGCM a combination of the delta-four-stream approximation for solar flux calcuations (Liou et al. 1988) and delta-two-and-four-stream approximation for IR flux calculations (Fu et al. 1997). This combination has been proven to be computationally efficient and at the same time to produce a high degree of accuracy.

The incorporation of nongray gaseous absorption in multiple-scattering atmospheres is based on the correlated-k distribution method developed by Fu and Liou (1992), in which the cumulative probability of the absorption coefficient, g(k), in a spectral interval is used as an independent variable to replace wavenumber. For different absorbing gases in different spectral regions, a minimum number of quadrature points ranging from 1 to about 10 are required for the integration. The solar and IR spectra are divided into 6 and 12 bands, respectively, according to the location of absorption bands, which are 0.2–0.69, 0.69–1.3, 1.3–1.9, 1.9–2.5, 2.5–3.5, and 3.5–5.0 µm for the solar spectrum, and 2200–1900, 1900–1700, 1700–1400, 1400–1250, 1250–1100, 1100–980, 980–800, 800–670, 670–540, 540–400, 400–280, and 280–10 cm–1 for the IR spectrum. In the former, absorption due to H2O (2500–14 500 cm–1), O3 (50 000–14 500 cm–1), CO2 (2850–5200 cm–1), and O2 (A, B, and γ bands) is taken into account. In the latter, absorption due to H2O (0–2200 cm–1), CO2 (540–800 cm–1), O3 (980–1100 cm–1), CH4 (1100–1400 cm–1), and N2O (1100–1400 cm–1) is included. The continuum absorption of H2O is included in the spectral region 280–1250 cm–1.

The single-scattering properties for cloud particles are required in the radiative transfer calculations. We follow the procedure developed by Fu and Liou (1993) for parameterization of the single-scattering parameters. The extinction coefficient, βe, the single-scattering albedo, ϖ, and the asymmetry factor, g, all dependent on wavelength and position in cloud, are parameterized in terms of ice water content (IWC) and mean effective size (De) in the forms
i1520-0442-16-20-3357-e31
where an, bn, and cn are coefficients determined from numerical fitting based on detailed light-scattering and absorption calculations for a range of ice crystal size distributions and shapes. According to Fu and Liou (1993), N = 1 or 2 is sufficient in these parameterizations to achieve an accuracy within 1% in the fitting. IWC is predicted in the UCLA AGCM. However, De is prescribed as 85 µm based on available ice crystal sizes observed from a number of aircraft observations. Calculation of the single-scattering properties of water clouds requires information regarding the particle shape, particle size distribution, and the particle index of refraction as a function of wavelength. They are parameterized in terms of the prognostic variable liquid water content (LWC) and water effective radius (re), which is fixed to 10 µm, by interpolation using the mean single-scattering properties of eight water cloud types for each spectral band (Fu 1991). To increase computational accuracy, we apply the similarity principle for radiative transfer to each grid point to account for the fractional energy in the diffraction peak of the phase function. Using the correlated-k distribution method, 121 spectral radiation calculations are required for each vertical profile in the AGCM.

b. Cloud fraction and cloud overlapping schemes

In this section, we describe the cloud fraction and cloud overlap schemes that are most suitable for use with the new radiation parameterization.

Two approaches have been developed for the calculation of the horizontal fraction of cloudiness: 1) the formation of clouds treated as a product of a random process in statistical equilibrium with the grid-scale motion (Hense and Heise 1984); and 2) diagnostic equations developed from the predicted relative humidity with different complexities and details (Smagorinsky 1960; Slingo 1980; Xu and Krueger 1991), or from cloud water content that is predicted by the model as a prognostic variable (Sundqvist 1978), while the subgrid-scale source or sink terms are linked to cumulus detrainment (Tiedtke 1993). Xu and Randall (1996a) pointed out that the time change of cloudiness associated with anvil clouds resulting from cumulus activity cannot be realized by using a diagnostic approach involving relative humidity. The prognostic cloud water, on the other hand, is associated with the cloud formation and dissipation processes, which are physically connected to the radiative effects of clouds and latent heat release, as well as to the coupling of convective and stratiform cloud parameterizatons in which the detrained condensed water form cumulus towers is a source of stratiform cloud formation. Xu and Randall (1996a) further analyzed results from CEM simulations and found that the stratiform cloud amount increases with the total cloud water mixing ratio, and that the grid-averaged total cloud water/ice mixing ratio is a good predictor of cloud amount.

It is clear that a comprehensive physical foundation for the determination of cloud fraction has yet to be developed. As a simple extension of the scheme currently used in the model, we define fractional cloudiness (Cf) by linear interpolation in log10qt between qt = 10–10 kg kg–1 corresponding to the upper limit of a cloud-free condition (Cf = 0) and qt = 10–5 kg kg–1 corresponding to an overcast condition (Cf = 1). That is,
i1520-0442-16-20-3357-e34
where qt1 and qt2 are equal to 10–5 and 10–10 kg kg–1 (Dowling and Radke 1990), respectively.

Producing partial cloudiness introduces the cloud vertical overlap problem, which is an important issue in climate model studies. Various parameterizations of cloud overlap effects have been developed, and several AGCM-sensitivity tests have been performed (e.g., Liang and Wang 1997; Chou et al. 1998; Gu and Liou 2001; Collins 2001). The most straightforward approach to dealing with fractional cloud cover is to divide the sky into sectors within which the cloud amount is either 0 or 1. Radiative fluxes are calculated for each sector and then weighted by the respective cloud amount to obtain gridbox fluxes. The most common methods used in contemporary AGCMs are random overlap (Manabe and Strickler 1964) and maximum/random overlap (Geleyn and Hollingsworth 1979; Chou et al. 1998). The latter has been shown to be more consistent with the observed cloud distribution (Tian and Curry 1989).

Here we employ the method of maximum/random overlap, in which clouds are devided into three types according to height: low, middle, and high. Maximum overlap is used for clouds of the same type, while random overlap is subsequently employed for clouds of different types. Scaling of the optical depth in groups of clouds of the same type is performed in terms of the cloud cover based on the method introduced by Chou et al. (1998). According to Collins (2001), all cloud configurations that are consistent with the vertical profile of partial cloudiness and overlap assumption are illustrated in Fig. 1. An atmospheric column, therefore, can be divided into at most eight sectors if clouds are present in all of the three groups. The number of configurations reduces to four if no cloud occurs or if the cloud is overcast in one of the cloud groups. If the cloud amounts in the three groups are c1, c2, and c3, the fractional area for the cloudy case is c1 × c2 × c3, while for the clear case it is (1 – c1) × (1 – c2) × (1 – c3). Radiation calculations can then be performed for each of the cloud configurations, and the all-sky flux can be determined as the weighted sum of the flux computed for each sector.

To implement the cloud overlap scheme in the radiation calculation, the optical properties for clear and overcast conditions for each layer are first determined. A combination of the optical properties for clear or overcast conditions corresponding to each layer is then configured throughout the entire atmospheric column. Radiative transfer calculations using the delta-four-stream approximation for solar and the delta-two-and-four-stream combination for IR can subsequently be applied to each configuration to obtain the radiation fluxes at each layer. The computational cost of employing the new cloudiness and radiation schemes is only slightly more than the original Fu–Liou scheme. This is because in many cases the number of columns required for radiative transfer calculations reduces significantly due to clear (0%) or overcast (100%) cloud cover by one or more of the cloud types.

4. Performance of the new cloud and radiation schemes

a. Simulated January and July climatologies

We now evaluate the updated UCLA AGCM with the newly implemented cloud and radiation schemes. This is done by comparing a 5-yr-long simulation using the new radiation scheme (NRAD) with another using the original scheme (CTRL). The two simulations use the same input data, which include a 0.7 factor introduced to adjust the cloud optical depth in order to account for the cloud inhomogeneity effect, a uniform effective ice crystal size of 85 µm, and a uniform water droplet effective radius of 10 µm. Results are presented in terms of the January means, with the global means for July shown in Table 1.

We start by comparing the total cloud cover from the two simulations with data compiled by the ISCCP. In terms of global means, the total cloudiness simulation in NRAD is closer to ISCCP values than that in CTRL by 4.9% and 5.3% for January and July, respectively (Table 1). The differences between CTRL and NRAD are statistically significant at the 95% level according to the t test. Figure 2 shows the geographical distribution of the simulated total cloud cover and observations. There is general agreement between simulations and observations. NRAD produces a better simulation of total cloudiness in North Africa, the South Atlantic, and southeast Pacific Oceans, where CTRL overestimates the cloudiness. In mid- to high latitudes, both simulations underestimate the cloudiness. Thus, the incorporation of the new cloud scheme has improved the cloudiness simulation in some regions, particularly in low latitudes.

To better see this improvement, we display the differences between the zonal mean total cloud cover simulated by NRAD and CTRL and the observations for January (Fig. 3a) and July (Fig. 3b). In January, there is a deep drop in observed cloudiness around 10°–30°N, where CTRL overestimates cloudiness by about 15% and NRAD produces a much closer agreement. Around 30°S and 30°–50°N, CTRL results are closer to observations. In July, NRAD produces a better simulation of cloud cover in the regions 0°–20°S and 20°–30°N, but underestimates the cloudiness in latitudes 0°–15° and 30°–50°N. Although NRAD produces better global mean cloud cover, further improvement in mid- to high latitudes is required. It appears that cloud generation schemes might have to be location dependent according to the dominant cloud type. Note that the difference in simulated cloudiness is produced both by the new cloud scheme and the new radiation calculations (the NRAD simulation). We have found that a simulation with the new radiation scheme but without the inclusion of the new cloud scheme produces better cloud cover almost everywhere than that generated by CTRL (not shown). However, the cloudiness is inferior to that produced by NRAD, particularly in the low latitudes of the winter hemisphere (not shown).

Figure 4 shows the simulated July mean total precipitation rate and the observation from the Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP). The two simulations and the observation exhibit similar large-scale distributions. A strong intertropical convergence zone (ITCZ) over the central-east Pacific and Atlantic Oceans is well simulated. Both simulations, however, yield less extensive precipitation in the west Pacific Ocean, where NRAD shows a slightly better agreement than CTRL. The region of high precipitation corresponding to the east Asian summer monsoon is well captured in the simulations. The simulations also generate realistic rainfall in Central America and Africa. The differences in global mean values (Table 1) between CTRL and NRAD are statistically significant at 95% level based on the t test, which shows that NRAD produces a better agreement.

We next verify the simulated outgoing longwave radiation (OLR). Since OLR is significantly affected by high clouds, this comparison indicates to some extent the quality of simulated ice clouds. In January, CTRL and NRAD underestimate global mean OLR values by 4.3 and 3.1 W m–2, respectively. In July, NRAD also produces a slightly better agreement with the observed OLR than CTRL (Table 1). The corresponding root-mean-square differences are 14.72 and 15.65 for January, and 15.27 and 16.56 for July for NRAD and CTRL, respectively. Figure 5 shows the geographic distribution of OLR from the two simulations. There are slight shifts of the convection zones over South America and Africa in both simulations, but overall the results are more accurate in NRAD. While both OLR simulations show overestimates in the southeast Pacific and South Atlantic Oceans, the NRAD results are closer to the observed data.

Figure 6 shows the horizontal distributions of ice and liquid cloud water content in NRAD. The ice and liquid water distribution patterns are similar to that of precipitation. The 2D zonal mean of ice water mixing ratio (Fig. 7a) shows that the main sources of ice water are from mid- to low latitudes. Ice water appears in higher regions in the Tropics due to the higher tropopause in the equatorial area. Liquid cloud water (Fig. 7b) mainly occurs in the lower region of the troposphere. Observed global data for these two parameters are not available at present.

The preceding comparisons illustrate that a satisfactory performance of the UCLA AGCM is obtained using the new cloud and radiation schemes. We emphasize that this was achieved without application of any “tuning.” Simulation with the new schemes has produced a more accurate radiation budget and more realistic cloud fields than those generated from the control run, as compared with observations.

b. Application to the 1997/98 El Niño and 1998/99 La Niña

The El Niño–Southern Oscillation (ENSO) phenomenon is a primary component of global interannual climate variability. The El Niño event of 1997/98 was one of the strongest on record by most standards and was followed by a La Niña event in 1998/99. We apply the updated UCLA AGCM to examine the atmospheric response to the SST anomalies during these events. Cess et al. (2001) showed that during the strong 1997/98 El Niño there was a substantially greater tendency toward net radiative cooling over the warm pool due to changes in cloud vertical structure as compared to the mean climatology and even to the weaker 1987 El Niño.

With this motivation, we perform 6-month-long runs using the updated AGCM and SSTs from the Smith and Reynolds (1998) dataset for the winters of 1997/98 and 1998/99 from initial conditions corresponding to 1 October prepared by interpolating pressure-level data from an analysis of observations to the model’s sigma surfaces. Figure 8 displays the December–January–February (DJF) mean SST for 1997/98, 1998/99, and their differences. During the 1997/98 El Niño event, SSTs were anomalously warm in the central to eastern equatorial Pacific, and anomalously cold to the north and south of the warmer area.

Figure 9 shows the differences of DJF mean precipitation for the 1997/98 El Niño and 1998/99 La Niña, as provided by the CMAP observations and the AGCM simulations. The precipitation increasing along the central and eastern equatorial Pacific over the warmer SSTs in the observed data is well captured in the simulation. The South Pacific convergence zone (SPCZ) shifts north and east of its average position. The reduction of precipitation in the western Pacific is simulated by the model. There are positive values of anomalous precipitation over the western Indian Ocean in both the simulation and the observation, indicating that the ITCZ is stronger and closer to the equator. The responses of CTRL to El Niño and La Niña are similar to those of NRAD (results not shown). This is expected since other parameterizations are more directly important for this response than the radiation process.

Figure 10 compares the differences of DJF mean OLR between El Niño and La Niña in observations (NOAA interpolated OLR) and the simulations. OLRs are correlated with the amount of convection in the Tropics. During the Tropical Ocean Global Atmosphere (TOGA) program, OLR information by the National Oceanic and Atmospheric Administration (NOAA) satellites since 1974 (Chelliah and Arkin 1992) has become the most widely used satellite-based measurement used for prediction of convective precipitation (Trenberth et al. 1998). In general, regions with positive and negative differences in the observations have counterparts in the model results. The most notable OLR differences are of about –60 W m–2 over the enhanced convection in the central-equatorial Pacific. In addition, there are high positive values associated with reduced convection north of that region, west of Indonesia, and in southwest Australia, in both simulation and observed data.

Cess et al. (2001) used two datasets from ERBE and the Clouds and the Earth’s Radiant Energy System (CERES) to investigate the influence of the 1997/98 El Niño on the cloud–radiative forcing over the Pacific warm pool. They showed that both the cloud height and cloud fractional coverage decreased in 1998. To examine these impacts we plot in Fig. 11 the differences between the DJF mean simulated total cloud water mixing ratio and cloud cover averaged between 30°S and 30°N during the 1997/98 El Niño and the 1998/99 La Niña. Over the warm pool, both the water mixing ratio and high cloud cover are reduced during the 1997/98 El Niño, in agreement with the analysis presented by Cess et al. (2001). On the contrary, the cloud water and cloud cover increase over the central and eastern Pacific, associated with the sea surface warming. The corresponding differences in radiative heating rates are displayed in Fig. 12. A net cooling over the warm pool and a net heating in the equatorial eastern Pacific are produced in the simulation, in general agreement with the results reported in Cess et al. (2001).

5. Effects of cloud–radiation on simulated climate

Clouds significantly interact with dynamics, radiation, and hydrological processes in a nonlinear fashion, and their formation represents one of the important mechanisms for the vertical redistribution of large-scale momentum, latent heat, and sensible heat (Arakawa 1975; Xu and Randall 1996a). In this section, we investigate the effects of updating the cloud and radiation schemes on the AGCM simulations.

a. Response to new cloud and radiation schemes

We first examine the effect of the new schemes on the generation of clouds in the model. Figure 13 shows the January zonal mean differences between ice, liquid, and total cloud water mixing ratios obtained in NRAD and CTRL. The distributions of ice water mixing ratio have similar patterns in both simulations (see Fig. 7 for NRAD). However, NRAD produces less ice cloud water in the mid- to upper troposphere, particularly in the midtroposphere over the subtropics in both hemispheres, and more ice water in the upper troposphere above the ITCZ. Also NRAD produces more liquid water in the lower troposphere over the ITCZ and subtropics of the summer hemisphere. Figure 14 shows the corresponding differences in radiative heating rates. Increase (decrease) in cloud amount normally produces more (less) solar heating within the cloud, while it generates more (less) IR cooling at the cloud top and more (less) IR warming at the cloud bottom. For the clear atmosphere area, changes in cloud distributions can have two opposite effects in shortwave heating: a warming (cooling) effect caused by the absorption of more available downward solar flux and a cooling (warming) effect due to the decreased (increased) absorption of the upward flux reflected by clouds in the regions where cloud water decreased (increased). For longwave radiation, less high cloud generally produces less downward flux below, while less low cloud yields enhanced upward flux above due to a reduced blocking of surface emission. As a result, in the cloud layers, the patterns of heating rate difference are similar to those of ice and liquid water cloud differences. The heating/cooling effect produced by IR radiation overwhelms that produced by its solar counterpart, resulting in warming of the mid- to upper and the lower troposphere over the ITCZ. The differences in heating rates due to grid scale and cumulus condensation (not shown) are of the same magnitude as those in radiative heating. Less cloud implies less latent heat release and vice versa, indicating that radiation and latent heat are of equal importance in the simulations. The simulated results for July are similar to those for January.

Figure 15 presents the January and July zonal mean temperature differences between NRAD and CTRL for the whole atmosphere. Related to the net radiative warming of the mid- and upper troposphere and the stronger subsidence due to reduced cloudiness, most regions in the troposphere are warmer in NRAD. A cold bias in the lower troposphere in both summer and winter and a polar upper troposphere in the summer is a common feature in most AGCMs (Boer et al. 1992; Liang and Wang 1997), including the UCLA AGCM. The warming in the troposphere produced in the NRAD experiment tends to correct this deficiency. However, the general coldness of climate models is not simply a radiation problem. Some studies (e.g., Johnson 1997) have suggested a sensitivity in a climate model’s temperature response to aphysical entropy sources introduced by spurious numerical dispersion/diffusion, Gibb’s oscillation, physical parameterizations, and other factors. Therefore, we do not expect to solve the coldness problem using the new cloud–radiation scheme alone. In fact, despite the warming in almost the whole vertical domain, NRAD produces lower atmosphere temperatures in the lower stratosphere. The surface evaporation and precipitation are slightly reduced in NRAD, as shown in Table 1. This is consistent with an increase in the tropospheric mean temperature (0.8 K) that is larger than the increase in the surface temperature (0.3 K), which leads to an enhancement of the atmospheric stability at lower levels.

b. Ice crystal size and cloud inhomogeneity effects

In NRAD the ice crystal size was uniformly prescribed as 85 µm. The ice crystal size distribution, however, is a function of temperature (Heymsfield and Platt 1984), and significantly interacts with the radiation field (Gu and Liou 2000; Wu et al. 2000). Therefore, it would be desirable to develop an interactive mean effective ice crystal size parameterization in connection with radiative transfer calculations in climate models. To explore this issue, we examine the effects of changing the mean effective ice crystal size from the 85 µm prescribed in NRAD to 50 µm, which is also in the ice crystal size range from 20 to 130 µm based on 28 sample values from six observational campaigns reported by Fu (1996). Table 2 shows the difference in radiative fluxes at the top of the atmosphere (TOA) and surface between NRAD and the simulation with a smaller ice crystal size. Use of a smaller value in the model reduces OLR by 4.5 W m–2, and enhances the reflected solar radiation at TOA by 4.8 W m–2. This is because for a given ice water content or ice water path, smaller ice crystals reflect more solar radiation, and at the same time trap more thermal infrared radiation.

To investigate the cloud inhomogeneity effect in the AGCM, we carried out another simulation by removing the cloud inhomogeneity (i.e., by setting the inhomogeneity factor to 1). This modification reduces OLR by 4.6 W m–2, and simultaneously enhances the reflected solar radiation by about 12 W m–2. The inhomogeneity effect on solar radiation clearly exceeds that on IR radiation. The corresponding planetary albedo increases from 27.4% to 31.31%, indicating that the cloud inhomogeneity has a significant impact on the simulation of planetary albedo.

6. Conclusions

An advanced yet efficient radiation parameterization has been implemented in the UCLA AGCM. In this parameterization, the extinction coefficient, single-scattering albedo, and asymmetry factor are parameterized in terms of ice water content or liquid water content. Gaseous absorption in multiple-scattering atmospheres is included by means of the correlated-k distribution approach. The delta-four-stream approximation for solar radiation and the delta-two-and-four-stream combination method for IR radiation are employed to compute the transfer of spectral radiation in the atmosphere. Fractional cloud cover is parameterized in terms of the total cloud condensate. Clouds are grouped according to height into three types: high, middle, and low. Maximum overlap is assumed for each cloud type, whereas random overlap is applied to clouds of different types.

Simulations using the original UCLA AGCM and the updated version that incorporates the new cloud and radiation schemes were carried out and the results were compared. We showed that the new schemes work well in the UCLA AGCM and produce more satisfactory radiation budget and cloud field simulations. The simulation of global mean cloud cover was improved by about 5%. The cloud cover was overestimated in the original version (in which cloud cover is either clear or overcast.). The results reveal that an improved cloud cover parameterization scheme for mid- to high latitudes is required to better capture the cloud features in those areas.

The updated AGCM was further applied to simulations with SST distributions corresponding to the 1997/98 El Niño and the 1998/99 La Niña events. The general agreement between the simulated and observed data indicates that the model atmosphere with the new cloud and radiation schemes can produce a realistic response to the SST forcing in the tropical Pacific. The results also show reduced high clouds and a net radiative cooling over the warm pool. All of these features are in general agreement with the recent analyses of the observations for this El Niño event.

Analyses of the results show that the new schemes tend to generate less cloud in the mid- to upper troposphere, resulting in increased radiative heating in these areas. The troposphere then becomes warmer due to radiative heating and the stronger subsidence related to reduced cloud cover. This warming effect corrects to some extent a cold bias in the mid- to upper troposphere of the original model.

We also investigated the impact of varying the ice crystal size and cloud inhomogeneity parameters in the model. A reduction in the ice crystal size reduced OLR at the TOA, while increasing the planetary albedo. The effect of ice crystal size on solar and IR radiation was of the same magnitude. The cloud inhomogeneity primarily affects the transfer of solar radiation. Homogeneous clouds resulted in an increase in the reflected solar radiation at TOA as compared to the inhomogeneous cloud with the same amount of cloud water. The sensitivity of model simulations to ice crystal size and cloud inhomogeneity suggests that interactive parameterizations for these two parameters would be a significant step toward a more reliable and physically based radiation calculation.

Acknowledgments

Research reported in this paper has been supported by DOE Grant DE-FG03-00ER62904 and NSF Grant ATM-9907924.

REFERENCES

  • Alexander, R. C., , and R. L. Mobley, 1976: Monthly average sea-surface temperatures and ice pack limits on a global grid. Mon. Wea. Rev., 104 , 143148.

    • Search Google Scholar
    • Export Citation
  • Arakawa, A., 1975: Modeling clouds and cloud processes for use in climate models. The Physical Basis of Climate and Climate Modeling, GARP Publishing Series, Vol. 16, WMO, 181–197.

    • Search Google Scholar
    • Export Citation
  • Arakawa, A., , and W. H. Schubert, 1974: Interaction of a cumulus cloud ensemble with the large-scale environment, Part I. J. Atmos. Sci., 31 , 674701.

    • Search Google Scholar
    • Export Citation
  • Barkstrom, B. R., 1984: Earth Radiation Budget Experiment (ERBE). Bull. Amer. Meteor. Soc., 65 , 11701185.

  • Boer, G. J., and Coauthors. 1992: Some results from an intercomparison of the climates simulated by 14 atmospheric general circulation models. J. Geophys. Res., 97 , 1277112786.

    • Search Google Scholar
    • Export Citation
  • Bony, S., , and K. Emanuel, 2001: A parameterization of the cloudiness associated with cumulus convection: Evaluation using TOGA COARE data. J. Atmos. Sci., 58 , 31583183.

    • Search Google Scholar
    • Export Citation
  • Cess, R. D., , M. H. Zhang, , B. A. Wielicki, , D. F. Young, , X-L. Zhou, , and Y. Nikitenko, 2001: The influence of the 1998 El Niño upon cloud–radiative forcing over the Pacific warm pool. J. Climate, 14 , 21292137.

    • Search Google Scholar
    • Export Citation
  • Chelliah, M., , and P. Arkin, 1992: Large-scale variability of monthly outgoing longwave radiation anomalies over the global Tropics. J. Climate, 5 , 371389.

    • Search Google Scholar
    • Export Citation
  • Chou, M. D., , M. J. Suarez, , C. H. Ho, , M. M-H. Yan, , and K-T. Lee, 1998: Parameterizations for cloud overlapping and shortwave single-scattering properties for use in general circulation and cloud ensemble models. J. Climate, 11 , 202214.

    • Search Google Scholar
    • Export Citation
  • Coakley Jr., J. A., , and P. Chylek, 1975: The two-stream approximation in radiative transfer: Including the angle of the incident radiation. J. Atmos. Sci., 32 , 409418.

    • Search Google Scholar
    • Export Citation
  • Collins, W. D., 2001: Parameterization of generalized cloud overlap for radiative calculations in general circulation models. J. Atmos. Sci., 58 , 32243242.

    • Search Google Scholar
    • Export Citation
  • Deardorff, J. W., 1972: Parameterization of the planetary boundary layer for use in general circulation models. Mon. Wea. Rev., 100 , 93106.

    • Search Google Scholar
    • Export Citation
  • Dorman, J. L., , and P. J. Sellers, 1989: A global climatology of albedo, roughness length and stomatal resistance for atmospheric general circulation models as represented by the Simple Biosphere model (SiB). J. Appl. Meteor., 28 , 833855.

    • Search Google Scholar
    • Export Citation
  • Dowling, D. R., , and L. F. Radke, 1990: A summary of the physical properties of cirrus clouds. J. Appl. Meteor., 29 , 970978.

  • Fu, Q., 1991: Parameterization of radiative processes in vertically nonhomogeneous multiple scattering atmospheres. Ph.D. dissertation, University of Utah, 259 pp.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., 1996: An accurate parameterization of the solar radiative properties of cirrus clouds for climate models. J. Climate, 9 , 20582082.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., , and K. N. Liou, 1992: On the correlated k-distribution method for radiative transfer in nonhomogeneous atmospheres. J. Atmos. Sci., 49 , 21392156.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., , and K. N. Liou, 1993: Parameterization of the radiative properties of cirrus clouds. J. Atmos. Sci., 50 , 20082025.

  • Fu, Q., , K. N. Liou, , M. C. Cribb, , T. P. Charlock, , and A. Grossman, 1997: Multiple scattering parameterization in thermal infrared radiative transfer. J. Atmos. Sci., 54 , 27992812.

    • Search Google Scholar
    • Export Citation
  • Geleyn, J-F., , and A. Hollingsworth, 1979: An economical analytical method for the computation of the interaction between scattering and line absorption of radiation. Beitr. Phys. Atmos., 52 , 116.

    • Search Google Scholar
    • Export Citation
  • Gu, Y., , and K. N. Liou, 2000: Interactions of radiation, microphysics, and turbulence in the evolution of cirrus clouds. J. Atmos. Sci., 57 , 24632479.

    • Search Google Scholar
    • Export Citation
  • Gu, Y., , and K. N. Liou, 2001: Radiation parameterization for three-dimensional inhomogeneous cirrus clouds: Application to climate models. J. Climate, 14 , 24432457.

    • Search Google Scholar
    • Export Citation
  • Harshvardhan, , R. Davies, , D. A. Randall, , and T. G. Corsetti, 1987: A fast radiation parameterization of atmospheric circulation models. J. Geophys. Res., 92 , 10091016.

    • Search Google Scholar
    • Export Citation
  • Hense, A., , and E. Heise, 1984: A sensitivity study of cloud parameterizations in general circulation models. Beitr. Phys. Atmos., 57 , 240258.

    • Search Google Scholar
    • Export Citation
  • Heymsfield, A. J., , and C. M. R. Platt, 1984: A parameterization of the particle size spectrum of ice clouds in terms of the ambient temperature and ice water content. J. Atmos. Sci., 41 , 846855.

    • Search Google Scholar
    • Export Citation
  • Johnson, D. R., 1997: “General coldness of climate models” and the second law: Implications for modeling the earth system. J. Climate, 10 , 28262846.

    • Search Google Scholar
    • Export Citation
  • Köhler, M., 1999: Explicit prediction of ice clouds in general circulation models. Ph.D. dissertation, University of California, Los Angeles, 167 pp.

    • Search Google Scholar
    • Export Citation
  • Li, D-M., , and K. P. Shine, 1995: A 4-dimensional ozone climatology for UGAMP models. UGAMP Internal Rep. 35.

  • Li, J-L. F., , M. Köhler, , J. D. Farrara, , and C. R. Mechoso, 2002: The impact of stratocumulus cloud radiative properties on surface heat fluxes simulated with a general circulation model. Mon. Wea. Rev., 130 , 14331441.

    • Search Google Scholar
    • Export Citation
  • Liang, X-Z., , and W-C. Wang, 1997: Cloud overlap effects on general circulation model climate simulations. J. Geophys. Res., 102 , 1103911047.

    • Search Google Scholar
    • Export Citation
  • Liou, K. N., 1986: Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev., 114 , 11671199.

  • Liou, K. N., , Q. Fu, , and T. P. Ackerman, 1988: A simple formulation of the delta-four-stream approximation for radiative transfer parameterizations. J. Atmos. Sci., 45 , 19401947.

    • Search Google Scholar
    • Export Citation
  • Lord, S. J., , H. E. Willoughby, , and J. M. Piotrowicz, 1984: Role of a parameterized ice-phase microphysics in an axisymmetric nonhydrostatic tropical cyclone model. J. Atmos. Sci., 41 , 28362848.

    • Search Google Scholar
    • Export Citation
  • Manabe, S., , and R. Strickler, 1964: Thermal equilibrium of the atmospheres with a convective adjustment. J. Atmos. Sci., 21 , 361385.

  • Pan, D-M., , and D. A. Randall, 1998: A cumulus parametrization with a prognostic closure. Quart. J. Roy. Meteor. Soc., 124 , 949981.

  • Ramanathan, V., , E. J. Pitcher, , R. C. Malone, , and M. L. Blackmon, 1983: The response of a spectral general circulation model to refinements in radiative processes. J. Atmos. Sci., 40 , 605630.

    • Search Google Scholar
    • Export Citation
  • Randall, D. A., 1976: The interaction of the planetary boundary layers with large-scale circulations. Ph.D. dissertation, University of California, Los Angeles, 247 pp.

    • Search Google Scholar
    • Export Citation
  • Randall, D. A., 1980a: Conditional instability of the first kind upside-down. J. Atmos. Sci., 37 , 125130.

  • Randall, D. A., 1980b: Entrainment into a stratocumulus layer with distributed radiative cooling. J. Atmos. Sci., 37 , 148159.

  • Randall, D. A., , Harshvardhan, , D. A. Dazlich, , and T. G. Corsetti, 1989: Interactions among radiation, convection, and large-scale dynamics in a general circulation model. J. Atmos. Sci., 46 , 19431970.

    • Search Google Scholar
    • Export Citation
  • Rayner, N. A., , C. K. Folland, , D. E. Parker, , and E. B. Horton, 1995: A new global sea-ice and sea surface temperature (GISST) data set for 1903–1994 for forcing climate models. Hadley Centre, Met Office, Internal Note No. 69, 14 pp.

    • Search Google Scholar
    • Export Citation
  • Sasamori, T., 1975: A statistical model for stationary atmospheric cloudiness, liquid water content, and rate of precipitation. Mon. Wea. Rev., 103 , 10371049.

    • Search Google Scholar
    • Export Citation
  • Slingo, A., , and J. M. Slingo, 1988: The response of a general circulation model to cloud longwave radiative forcing. Part I: Introduction and initial experiments. Quart J. Roy. Meteor. Soc., 114 , 10271062.

    • Search Google Scholar
    • Export Citation
  • Slingo, A., , and J. M. Slingo, 1991: Response of the National Center for Atmospheric Research community climate model to improvements in the representation of clouds. J. Geophys. Res., 96 , 1534115347.

    • Search Google Scholar
    • Export Citation
  • Slingo, J. M., 1980: A cloud parameterization scheme derived from GATE data for use with a numerical model. Quart. J. Roy. Meteor. Soc., 106 , 747770.

    • Search Google Scholar
    • Export Citation
  • Smagorinsky, J., 1960: On the dynamical prediction of large-scale condensation by numerical methods. Geophys. Monogr., No. 5, Amer. Geophys. Union, 71–78.

    • Search Google Scholar
    • Export Citation
  • Smith, T. M., , and R. W. Reynolds, 1998: A high-resolution global sea surface temperature climatology for the 1961–90 base period. J. Climate, 11 , 33203323.

    • Search Google Scholar
    • Export Citation
  • Suarez, M. J., , A. Arakawa, , and D. A. Randall, 1983: The parameterization of the planetary boundary layer in the UCLA general circulation model: Formulation and results. Mon. Wea. Rev., 111 , 22242243.

    • Search Google Scholar
    • Export Citation
  • Sundqvist, H., 1978: A parameterization scheme for non-convective condensation including prediction of cloud water content. Quart. J. Roy. Meteor. Soc., 104 , 677690.

    • Search Google Scholar
    • Export Citation
  • Tian, L., , and J. A. Curry, 1989: Cloud overlap statistics. J. Geophys. Res., 94 , 99259935.

  • Tiedtke, M., 1993: Representation of clouds in large-scale models. Mon. Wea. Rev., 121 , 30403061.

  • Trenberth, K. E., , G. W. Branstator, , D. Karoly, , A. Kumar, , N-C. Lau, , and C. Ropelewski, 1998: Progress during TOGA in understanding and modeling global teleconnections associated with tropical sea surface temperatures. J. Geophys. Res., 103 , 1429114324.

    • Search Google Scholar
    • Export Citation
  • Wu, T., , W. R. Cotton, , and W. Y. Y. Cheng, 2000: Radiative effects on the diffusional growth of ice particles in cirrus clouds. J. Atmos. Sci., 57 , 28922904.

    • Search Google Scholar
    • Export Citation
  • Xu, K-M., , and S. K. Krueger, 1991: Evaluation of cloudiness parameterizations using a cumulus ensemble model. Mon. Wea. Rev., 119 , 342367.

    • Search Google Scholar
    • Export Citation
  • Xu, K-M., , and D. A. Randall, 1996a: A semiempirical cloudiness parameterization for use in climate models. J. Atmos. Sci., 53 , 30843102.

    • Search Google Scholar
    • Export Citation
  • Xu, K-M., , and D. A. Randall, 1996b: Evaluation of statistically based cloudiness parameterizations used in climate models. J. Atmos. Sci., 53 , 31033119.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

Sample cloud configurations that are consistent with the vertical profile of the partial cloudiness and the overlap assumptions

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 2.
Fig. 2.

Jan mean total cloud cover (%) for simulations (a) NRAD, (b) CTRL, and (c) ISCCP observations

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 3.
Fig. 3.

Zonal mean total cloud cover differences from observations (%) for NRAD (solid) and CTRL (dashed) for (a) Jan and (b) Jul

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 4.
Fig. 4.

Jul mean total precipitation (mm day–1) for simulations (a) NRAD, (b) CTRL, and (c) CMAP observations

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 5.
Fig. 5.

Jan mean OLR (W m–2) for simulations (a) NRAD, (b) CTRL, and (c) Arkin observations (Chelliah and Arkin 1992)

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 6.
Fig. 6.

Jan horizontal distributions of (a) ice and (b) liquid water contents (mg m–3) in the NRAD simulation

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 7.
Fig. 7.

Jan zonal mean of (a) ice and (b) liquid water mixing ratios (10–6 kg kg–1) in the NRAD simulation

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 8.
Fig. 8.

Observed SSTs from the DJF dataset provided by Smith and Reynolds (1998) for (a) 1997/98 El Niño, (b) 1998/99 La Niña, and (c) their differences

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 9.
Fig. 9.

Differences in DJF precipitation (mm day–1) between 1997/98 El Niño and 1998/99 La Niña for the (a) AGCM NRAD simulation and (b) CMAP observations

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 10.
Fig. 10.

Differences in DJF OLR (W m–2) between 1997/98 El Niño and 1998/99 La Niña for the (a) AGCM NRAD simulation and (b) NOAA interpolated data

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 11.
Fig. 11.

Differences in the DJF simulated (a) total cloud water mixing ratio (kg kg–1) and (b) cloud cover (%) between 1997/98 El Niño and 1998/99 La Niña. The figure is presented in a longitude–height projection and displays values averaged between 30°S and 30°N

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 12.
Fig. 12.

Differences in the DJF simulated (a) shortwave, (b) longwave, and (c) net radiative heating rates (K day–1) between 1997/98 El Niño and 1998/98 La Niña. The figure is presented in a longitude–height projection and displays values averaged between 30°S and 30°N

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 13.
Fig. 13.

Jan zonal mean differences of (a) ice, (b) liquid, and (c) total water mixing ratios (kg kg–1) between NRAD and CTRL

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 14.
Fig. 14.

Jan zonal mean differences of (a) solar, (b) IR, and (c) net radiative heating rates (K day–1) between NRAD and CTRL

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Fig. 15.
Fig. 15.

Zonal mean temperature differences (K) between NRAD and CTRL for (a) Jan and (b) Jul

Citation: Journal of Climate 16, 20; 10.1175/1520-0442(2003)016<3357:POCPIT>2.0.CO;2

Table 1.

Global mean values of total cloud cover (%), OLR (W m–2 ), precipitation (mm day–1 ), and planetary albedo (%) in observations and the simulations NRAD and CTRL for Jan and Jul. Measurement datasets are as follows. Total cloud cover: ISCCP; OLR and planetary albedo: ERBE (Barkstrom 1984); precipitation: CMAP

Table 1.
Table 2.

Simulated changes in Jan global mean shortwave (SW) and longwave (LW) radiative fluxes (W m–2 ) at the TOA and the earth’s surface between NRAD and two simulations with an ice crystal size of 50 µm and inhomogeneity factor (inho) of 1.0, respectively

Table 2.
Save