• Bishop, J. K. B., W. B. Rossow, and E. G. Dutton, 1997: Surface solar irradiance from the International Satellite Cloud Climatology Project 1983–1991. J. Geophys. Res.,102, 6883–6910.

  • Bonan, G., 1998: The land surface climatology of the NCAR land surface model (LSM 1.0) coupled to the NCAR Community Climate Model (CCM3). J. Climate,11, 1307–1326.

  • Boville, B. A., and P. R. Gent, 1998: The NCAR Climate System Model, version one. J. Climate,11, 1115–1130.

  • ——, and J. Hurrell, 1998: A comparison of the atmospheric circulations simulated by the CCM3 and CSM1. J. Climate,11, 1327–1341.

  • Cess, R. D., 1995: Absorption of solar radiation by clouds. Science,267, 496–499.

  • Chou, M.-D., and W. Zhao, 1997: Estimation and model validation of surface solar radiation and cloud radiative forcing using TOGA COARE measurements. J. Climate,10, 610–620.

  • Collins, W. D., J. Wang, J. T. Kiehl, and G. J. Zhang, 1997: Comparison of tropical ocean–atmosphere fluxes with the NCAR Community Climate Model CCM3. J. Climate,10, 3047–3058.

  • Doney, S. C., W. G. Large, and F. O. Bryan, 1998: Surface ocean fluxes and water-mass transformation rates in the coupled NCAR Climate System Model. J. Climate,11, 1420–1441.

  • Frey, H., M. Latif, and T. Stockdale, 1997: The coupled GCM ECHO-2. Part I: The tropical Pacific. Mon. Wea. Rev.,125, 703–720.

  • Gent, P. R., 1991: The heat budget of the TOGA-COARE domain in an ocean model. J. Geophys. Res.,96, 3323–3330.

  • ——, and J. C. McWilliams, 1990: Isopycnal mixing in ocean circulation models. J. Phys. Oceanogr.,20, 150–155.

  • ——, F. O. Bryan, G. Danabasoglu, S. C. Doney, W. R. Holland, W. G. Large, and J. C. McWilliams, 1998: The NCAR Climate System Model global ocean component. J. Climate,11, 1287–1306.

  • Hack, J. J., 1994: Parameterization of moist convection in the National Center for Atmospheric Research Community Climate Model (CCM2). J. Geophys. Res.,99, 5551–5568.

  • Hartmann, D. L., and M. L. Michelsen, 1993: Large-scale effects on the regulation of tropical sea surface temperature. J. Climate,6, 2049–2062.

  • Holtslag, A. A. M., and B. A. Boville, 1993: Local versus non-local boundary layer diffusion in a global climate model. J. Climate,6, 1825–1842.

  • Houze, R. A., 1989: Observed structure of mesoscale convective systems and implications for large-scale heating. Quart. J. Roy. Meteor. Soc.,115, 425–461.

  • Kiehl, J. T., 1994: On the observed near cancellation between longwave and shortwave cloud forcing in tropical regions. J. Climate,7, 559–565.

  • ——, and K. E. Trenberth, 1997: Earth’s annual global mean energy budget. Bull. Amer. Meteor. Soc.,78, 197–208.

  • ——, J. J. Hack, G. Bonan, B. A. Boville, D. Williamson, and P. J. Rasch, 1998: The National Center for Atmospheric Research Community Climate Model: CCM3. J. Climate,11, 1131–1149.

  • Large, W. G., J. C. McWilliams, and S. C. Doney, 1994: Oceanic vertical mixing: A review and a model with a nonlocal boundary layer parameterization. Rev. Geophys.,32, 363–403.

  • ——, G. Danabasoglu, S. C. Doney, and J. C. McWilliams, 1997: Sensitivity to surface forcing and boundary layer mixing in a global ocean model: Annual-mean climatology. J. Phys. Oceanogr.,27, 2418–2447.

  • Lau, N.-C., 1997: Interactions between global SST anomalies and midlatitude atmospheric circulation. Bull. Amer. Meteor. Soc.,78, 21–33.

  • Lindzen, R. S., and S. Nigam, 1987: On the role of sea surface temperature gradients in forcing low-level winds and convergence in the tropics. J. Atmos. Sci.,44, 2418–2436.

  • Lukas, R., and E. Lindstrom, 1997: The mixed layer of the western equatorial Pacific ocean, J. Geophys. Res.,102, 3344–3357.

  • Ma, C.-C., C. R. Mechoso, A. Arakawa, and J. D. Farrara, 1994: Sensitivity of a coupled ocean–atmosphere model to physical parameterizations. J. Climate,7, 1883–1896.

  • ——, ——, A. W. Robertson, and A. Arakawa, 1996: Peruvian stratus clouds and the tropical Pacific circulation: A coupled ocean–atmosphere GCM study. J. Climate,9, 1635–1645.

  • Mechoso, C. R., and Coauthors, 1995: The seasonal cycle over the tropical Pacific in coupled ocean–atmosphere general circulation models. Mon. Wea. Rev.,123, 2825–2838.

  • Oberhuber, J. M., 1988: An atlas based on the “COADS” data set: The budgets of heat, buoyancy and turbulent kinetic energy at the surface of the global ocean. Max-Planck Institute for Meteorology Rep. 15, Max-Planck Institute for Meteorology, Hamburg, Germany, 199 pp.

  • Pilewskie, P., and F. P. J. Valero, 1995: Direct observations of excess solar absorption by clouds. Science,267, 1626–1629.

  • Ramanathan, V., R. D. Cess, E. F. Harrison, P. Minnis, B. R. Barkstrom, E. Ahmad, and D. L. Hartmann, 1989: Cloud radiative forcing and climate: Results from the Earth Radiation Budget Experiment. Science,243, 57–63.

  • ——, R. Dirks, R. Grossman, A. Heymsfield, J. Kuettner, and F. Valero, 1993: Central Equatorial Pacific Experiment design. Center for Clouds, Chemistry and Climate, University of California, San Diego, CA, 54 pp. [Available from NCAR, P. O. Box 3000, Boulder, CO 80307.].

  • ——, B. Subasilar, G. J. Zhang, W. Conant, R. D. Cess, J. T. Kiehl, H. Grassl, and L. Shi, 1995: Warm pool heat budegt and shortwave cloud forcing: A missing physics? Science,267, 499–503.

  • Russell, G. L., J. R. Miller, and D. Rind, 1995: A coupled atmosphere–ocean model for transient climate change studies. Atmos.–Ocean,33, 683–730.

  • Schneider, E. K., Z. Zhu, B. S. Giese, B. Huang, B. P. Kirtman, J. Shukla, and J. A. Carton, 1997: Annual cycle and ENSO in a coupled ocean–atmosphere general circulation model. Mon. Wea. Rev.,125, 680–702.

  • Schneider, N., T. Barnett, M. Latif, and T. Stockdale, 1996: Warm pool physics in a coupled GCM. J. Climate,9, 219–239.

  • Shea, D. J., K. E. Trenberth, and R. W. Reynolds, 1990: A global monthly sea surface temperature climatology. NCAR Tech. Note NCAR/TN-345, 167 pp. [Available from NCAR, P. O. Box 3000, Boulder, CO 80307.].

  • Stockdale, T., M. Latif, G. Burgers, and J.-O. Wolff, 1994: Some sensitivities of a coupled ocean–atmosphere GCM. Tellus,46A, 367–380.

  • Waliser, D. E., and N. E. Graham, 1993: Convective cloud systems and warm pool sea surface temperatures: Coupled interactions and self regulation. J. Geophys. Res.,98, 12 881–12 893.

  • ——, W. D. Collins, and S. P. Anderson, 1996: An estimate of the surface shortwave cloud forcing over the western pacific during TOGA COARE. Geophys. Res. Lett.,23, 519–522.

  • Weatherly, J., B. Briegleb, W. G. Large, and T. Bettge, 1998: Sea ice and polar climate in the NCAR CSM. J. Climate,11, 1472–1486.

  • Webster, P. J., 1994: The role of hydrological processes in ocean–atmosphere interactions. Rev. Geophys.,32, 427–476.

  • ——, and R. Lukas, 1992: TOGA COARE: The Coupled Ocean–Atmosphere Response Experiment. Bull. Amer. Meteor. Soc.,73, 1377–1416.

  • Zhang, G. J., and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Centre general circulation model. Atmos.–Ocean,33, 407–446.

  • ——, and M. J. McPhaden, 1995: The relationship between sea surface temperature and latent heat flux in the equatorial Pacific. J. Climate,8, 589–605.

  • ——, and R. L. Grossman, 1996: Surface evaporation during the Central Equatorial Pacific Experiment: A climate-scale perspective. J. Climate,9, 2522–2537.

  • View in gallery

    Difference in annual mean SST (a) uncoupled ocean model forced with the observational analysis of Large et al. (1997), (b) uncoupled ocean model forced with CCM3 data, and (c) fully coupled CSM from observed SSTs from Shea et al. (1990) climatology. Contour interval is 1°C.

  • View in gallery

    (Continued)

  • View in gallery

    Annual mean net surface energy budget (W m−2) from (a) forcing of Large et al. (1997) and (b) CCM3. Contour interval is 15 W m−2.

  • View in gallery

    Annual mean net surface shortwave flux (W m−2) from (a) forcing of Large et al. (1997) and (b) CCM3. Contour interval is 10 W m−2.

  • View in gallery

    Annual mean net surface latent heat flux (W m−2) from (a) forcing of Large et al. (1997) and (b) CCM3. Contour interval is 15 W m−2.

  • View in gallery

    Annual mean net surface energy flux (W m−2) from the fully coupled CSM. Contour interval is 30 W m−2.

  • View in gallery

    Annual mean difference (CSM − CCM3) in latent heat flux (W m−2) for the tropical Pacific region. Contour interval is 20 W m−2.

  • View in gallery

    Annual mean difference (CSM − CCM3) in defecit in specific humidity (Kg Kg−1) for the tropical Pacific region. Contour interval is 0.001 Kg Kg−1.

  • View in gallery

    Annual mean difference (CSM − CCM3) in surface zonal wind (m s−1) for the tropical Pacific region. Contour interval is 1 m s−1.

  • View in gallery

    Annual mean difference (CSM − CCM3) in vector surface wind (m s−1) for the tropical Pacific region and sea level pressure (Pa). Contour interval is 30 Pa.

  • View in gallery

    Time evolution of the annual mean sea surface temperature (°C) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

  • View in gallery

    Time evolution of the annual mean SSS (ppt) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

  • View in gallery

    Time evolution of the annual mean net shortwave surface energy flux (W m−2) (– – –) and latent plus sensible plus longwave flux (———) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

  • View in gallery

    Time evolution of the annual mean net surface energy flux (W m−2) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

  • View in gallery

    Time evolution of the annual mean surface zonal wind stress (dyn cm−2) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 132 129 6
PDF Downloads 28 25 5

Simulation of the Tropical Pacific Warm Pool with the NCAR Climate System Model

View More View Less
  • 1 National Center for Atmospheric Research, Boulder, Colorado
© Get Permissions
Full access

Abstract

The simulation of the tropical western Pacific warm pool is explored with the NCAR Climate System Model (CSM). The simulated sea surface temperatures in the Pacific basin have biases that are similar to other coupled model simulations in this region. In particular, an excessive cold tongue of water extends across the Pacific basin, with warm water on either side of this cold tongue. The warm pool region is also too cold. This cold bias exists in spite of an overestimate in surface net energy flux into this region. To understand the source of this bias in SST, simulations from the uncoupled and fully coupled models are analyzed in terms of biases in surface energy budget. These analyses suggest that the strong constraint of little ocean heat transport out of the warm pool region forces a change in SST gradient that leads to an increase in the atmospheric zonal wind. This increase in zonal wind causes an increase in latent heat flux in the warm pool region. The increase in latent heat flux is required to offset a significant (∼35 W m−2) bias in net surface solar flux. The bias in surface solar flux is due to an underestimate of model cloud shortwave absorption.

Corresponding author address: Dr. Jeffrey T. Kiehl, NCAR/CGD, P.O. Box 3000, Boulder, CO 80307-3000.

Email: jtkon@ucar.edu

Abstract

The simulation of the tropical western Pacific warm pool is explored with the NCAR Climate System Model (CSM). The simulated sea surface temperatures in the Pacific basin have biases that are similar to other coupled model simulations in this region. In particular, an excessive cold tongue of water extends across the Pacific basin, with warm water on either side of this cold tongue. The warm pool region is also too cold. This cold bias exists in spite of an overestimate in surface net energy flux into this region. To understand the source of this bias in SST, simulations from the uncoupled and fully coupled models are analyzed in terms of biases in surface energy budget. These analyses suggest that the strong constraint of little ocean heat transport out of the warm pool region forces a change in SST gradient that leads to an increase in the atmospheric zonal wind. This increase in zonal wind causes an increase in latent heat flux in the warm pool region. The increase in latent heat flux is required to offset a significant (∼35 W m−2) bias in net surface solar flux. The bias in surface solar flux is due to an underestimate of model cloud shortwave absorption.

Corresponding author address: Dr. Jeffrey T. Kiehl, NCAR/CGD, P.O. Box 3000, Boulder, CO 80307-3000.

Email: jtkon@ucar.edu

1. Introduction

The tropical western Pacific warm pool region contains the largest area of water with the warmest sea surface temperatures. The region is also a center of deep convective activity that contributes a significant amount of diabatic heating to the tropical atmosphere, affecting both the meridional Hadley circulation and the Walker circulation (Webster 1994). The importance of this region is widely recognized and was the impetus for the Tropical Oceans Global Atmosphere Coupled Ocean–Atmosphere Response Experiment (TOGA COARE) (Webster and Lukas 1992).

Over the past two decades, coupled atmosphere–ocean models have attempted to simulate the tropical Pacific basin, including the warm pool region. Mechoso et al. (1995) summarize the results from a number of simulations for the tropical Pacific region. Many of these simulations predict sea surface temperatures that are too cold in the western Pacific. More recent studies with a coupled model by Schneider et al. (1997) also indicate that the sea surface temperatures across the equatorial Pacific region are too cold compared to climatology. There has been considerable speculation on the major cause for the simulated cold bias in tropical sea surface temperatures from coupled models. Ma et al. (1994) carried out a series of sensitivity experiments testing the relative importance of different longwave radiation schemes and ocean boundary layer schemes on the simulated sea surface temperatures in the tropical Pacific. They concluded that their coupled simulations were more sensitive to the accuracy of the longwave radiation model. Stockdale et al. (1994) carried out sensitivity studies with the ECHAM–HOPE coupled model. They found that the simulated western Pacific sea surface temperatures were sensitive to the convective cloud fraction in this region. By changing convective cloud fraction in this region to get better agreement with the observed top-of-atmosphere shortwave radiation budget, they, in turn, altered the surface shortwave budget that decreased the climate drift in the western Pacific. Stockdale et al. (1994) and Ma et al. (1996) have also studied the affects of changes in boundary layer clouds on the simulated SSTs in the eastern Pacific.

Although much attention has focused on the coupled models inability to accurately simulate the tropical eastern Pacific “cold tongue,” less focus has been given to the western warm pool simulations. The present work focuses on the simulation of the tropical Pacific sea surface temperature (SST) from the National Center For Atmospheric Research (NCAR) Climate System Model (CSM) (Boville and Gent 1998). The SST biases in the CSM are similar to those found in other coupled modeling efforts. The cause of the SST cold bias in the CSM is explored and a hypothesis for the SST bias, based on a bias in the surface shortwave flux, is proposed.

A recent study by Frey et al. (1997) has direct relevance to the present study. They present results from the coupled Max-Planck Institute at Hamburg’s ECHO-2 model, which uses the ECHAM4 atmospheric general circulation model (GCM) and the HOPE-2 ocean GCM. This study shows simulated SSTs in the tropcial Pacific region that are in much better agreement with observations than the previous version of their coupled model (ECHO-1). Comparing the surface shortwave fluxes from ECHO-2 (see Fig. 4 in Frey et al.) with those from the previous version of the ECHO-1 model (see Fig. 4 of Schneider et al. 1996), there is a significant reduction in shortwave flux at the ocean surface in the latest version of the coupled model. Frey et al. point out that the net flux into the warm pool region in ECHO-2 is 20 W m−2 compared to 60 W m−2 from ECHO-1. The role of shortwave flux in simulating the tropical western Pacific SSTs is the central focus of the present study.

Evaluation of a model’s ability to accurately simulate the energy budget in the tropical western Pacific requires accurate observations of the various components to the net flux of energy into the warm pool region. Two recent field programs, TOGA COARE (Webster and Lukas 1992) and Central Equatorial Pacific Experiment (CEPEX) (Ramanathan et al. 1993), provide observations of the flux components that constrain the net surface energy flux over the Pacific region. These results are used to diagnose biases in the surface energy budget within the CSM. Furthermore, satellite data from the Earth Radiation Budget Experiment (ERBE) are used to constrain the overall energy budget of the atmosphere and ocean systems.

The spinup procedure of the CSM offers the ability to analyze the simulation of the tropical western Pacific in both a fully coupled and uncoupled mode. These results from the spinup procedure provide for a more detailed analysis of the role of the coupled atmosphere–ocean systems ability to respond to biases in the surface energy budget.

The study is organized in the following manner: the models are briefly described in section 2, the model simulations and observational are described in section 3, an analysis of the warm pool energy budget from the uncoupled atmospheric model is given in section 4, an analysis of the coupled model energy budget is given in section 5; furthermore, section 5 explains the cause for the cold SST bias in the western Pacific. Finally, section 6 summarizes the results and discusses the future improvements required to resolve the warm pool bias in the CSM.

2. Model descriptions

Both the atmosphere and coupled models are described in more detail in this issue (see Kiehl et al. 1998;Boville and Gent 1998). Briefly, the atmosphere model is the NCAR Community Climate Model version 3 (CCM3). The model employs a spectral dynamics representation, with a horizontal truncation of T42 (roughly 2.9° × 2.9° equivalent Gaussian grid). There are 18 levels in the vertical. The model includes the deep convection scheme of Zhang and McFarlane (1995) and the Hack (1994) convection scheme to represent shallow and midlevel convection. The radiation model includes a diurnal cycle, with radiation calculations performed hourly. Cloud properties (e.g., fraction, cloud water, particle size) are diagnostic (see Kiehl et al. 1998). The model includes a boundary layer scheme, which is an extension of that developed by Holtslag and Boville (1993). The land surface scheme is represented by the LSM of Bonan (1998), which is a model of the energy, momentum, water, and CO2 exchange between the atmosphere and land surface, accounting for ecological differences among vegetation, hydraulic and thermal differences in soil types, and allowing for multiple surface types within a grid cell.

The NCAR Climate System Model is described in more detail in Boville and Gent (1998). The CSM includes the CCM3 as its atmospheric model, LSM as the land model, the ocean model (Gent et al. 1998) is a version of the Geophysics Fluid Dynamics Laboratory (GFDL) MOM1.1 model that includes the KPP ocean mixed layer of Large et al. (1994), the Gent and McWilliams (1990) parameterization for subgrid-scale eddies, and a fourth-order advection scheme for tracers. The sea-ice model (Weatherly et al. 1998) is a version of the dynamic–thermodynamic model of Flato and Hibler. Simulations of the CSM are described in papers in this special issue.

The coupled model is spunup in four phases. First, the ocean and sea-ice models are spun up to an equilibrium with the prescribed observationally based forcing of Large et al. (1997). Where the phase 1 spinup is carried out without any sea-ice albedo feedback. Second, the sea-ice albedo feedback is made interactive with the same prescribed observationally based forcing as in phase 1. Third, the ocean–sea-ice models are forced from CCM3 state variables obtained from a CCM3 simulation that employed prescribed repeating climatological SSTs. It is important to note that the only forcing term used in flux form from the CCM3 is the downward solar radiation at the surface (i.e., surface insolation), all other forcing terms are imposed as state variables (e.g., surface air temperature, surface air specific humidity, . . .). Fluxes are calculated by the flux coupler based on these state information. In the fourth and final phase of the spinup procedure, the atmosphere model is fully coupled to the ocean model. Results from the last three phases of this spinup procedure are employed in this study to understand biases in the CSM warm pool region.

To summarize the last three phases of uncoupled and coupled models, Fig. 1 presents the SST biases in the tropical Pacific for the uncoupled ocean model forced with the Large et al. data, the uncoupled ocean model forced with CCM3 data, and the fully coupled CSM. The bias in SST is defined as the difference between the model simulated SST and the climatological SSTs from Shea et al. (1990).

Using the observationally based forcing of Large et al. (1997) produces a well-simulated warm pool and tropical Pacific SST distribution. The −1 K bias in the central equatorial Pacific is actually related to a 1 K warm bias in the climatological data for this region (Gent et al. 1998). This bias in the climatological SSTs arises from coarse averaging of finer resolution data across the equatorial region. For the uncoupled ocean model forced with CCM3 data there is a warm bias in tropical western SSTs (between 1 and 2 K), with a cold bias in the eastern equatorial Pacific. The SST bias in the fully coupled CSM model is shown in Fig. 1c. The western Pacific region is too cold by approximately −1 K, while near the date line the cold bias is greater than −2 K. North and south of this cold water there is warm bias in SST between 0 and 1 K. These zonal warm biases, located at 10° north and south, support a zonal precipitation pattern similar to a double ITCZ structure (see Boville and Hurrell 1997). These biases in tropical SSTs and precipitation pattern are generic to fully coupled models (see, e.g., Russell et al. 1996; Mechoso et al. 1995; Ma et al. 1996). The present study will analyze the biases in the uncoupled and coupled models in terms of the surface energy budgets associated with each of these simulations.

3. Model simulations and observational data

The analysis of the CCM3 are based on a 15-yr integration of the model using observed SSTs (1979–93). This climate simulation of CCM3 is described in more detail in Hack et al. (1998), Kiehl et al. (1998), and Hurrell et al. (1998). The warm pool energy budget is based on the annual mean of the full 15-yr simulation, where the warm pool region is defined as 10°N, 10°S; 140°E, 170°E for the present analysis. There is little year-to-year variation in the annual mean energy budget in this region (see, e.g., Ramanathan et al. 1995). Thus, the annual mean budget for the full 15-yr ensemble mean is very similar (<3 W m−2) when compared to any 5-yr period such as the ERBE time period during 1984–89.

The fully coupled CSM results are based on an annual mean ensemble average of 44 yr of the CSM simulation. Boville and Gent (1998) present the geographic distribution of interannual climate variability predicted by the CSM. These results indicate that the 44-yr ensemble average used in this study is very representative of any long (i.e., multidecadel) ensemble average of Pacific SSTs.

The observational estimates of the surface and top-of-atmosphere energy budgets are based on analysis of the TOGA COARE, the Tropical Atmosphere–Ocean (TAO) array, CEPEX, and Earth Radiation Budget Experiment (ERBE) data. Of course these observational data are limited to certain time periods. ERBE data span the time period 1984–89, TOGA COARE took place from November 1992 to March 1993, CEPEX from March 1993 to April 1993, while the TAO array has collected nearly 4 yr of data. A number of investigators (Ramanathan et al. 1995; Waliser et al. 1996; Zhang and McPhaden 1995) have used these data to construct estimates of the annual mean energy budget terms for the warm pool region. Similar to the CCM3 simulations, it is found that variations in the annual mean energy budget of the warm pool region are very small, and thus these various estimates are very representative of the long-term budget of this region. Table 1 provides a summary of this energy budget from the various studies of this region.

This budget indicates that the observed flux of energy into the warm pool ocean is less than 15 W m−2. Ramanathan et al. (1995) argue that an upper bound to the net input of energy into the warm pool region is 20 W m−2. An earlier model study by Gent (1991) indicated that the dynamical heat transport in this region is indeed small (<20 W m−2). These estimates indicate the relative inefficiency of ocean dynamics to transport heat out of the warm pool region. This feature is one of the unique aspects of the warm pool region, that it is a region of very warm SSTs, in spite of the relatively small net input of energy.

With respect to the individual components, at the top of the atmosphere the effect of clouds on the shortwave budget is defined in terms of the shortwave cloud forcing (SWCF), defined as the difference between the all sky net shortwave flux and the clear sky net shortwave flux, SSclr. For the warm pool region this is −64 W m−2, while the longwave cloud forcing is 60 W m−2. Thus, there is a near cancellation in the top-of-atmosphere cloud radiative forcing (see Kiehl 1994). It is important to note that this near cancellation effect does not imply that the radiative effects of clouds in this region are unimportant to the energy budget, since the vertical disposition of the shortwave and longwave radiative effects are quite different.

At the surface, the major fluxes contributing to the energy budget are the net solar flux and the latent heat flux. The SWCF at the surface has been estimated by a number of studies (Ramanathan et al. 1995; Waliser et al. 1996; Chou and Zhao 1997). All of these studies estimate a surface SWCF in the warm pool of around −100 W m−2 and agree to within a few watts per square meter. Table 1 indicates a clear sky net surface flux of 282 W m−2. The clear sky flux value of 282 W m−2 is obtained by adjusting the annual mean clear sky flux measured at American Samoa (Ramanathan et al. 1995) for a 14.25° latitude difference with the warm pool region. This value can also be obtained by taking the surface clear flux observed during the TOGA COARE period and correcting the December–March mean value to an annual mean value by using the amplitude of the surface annual mean flux from the CCM3 (note only the amplitude is used not the absolute flux values). Using a surface SWCF of −100 W m−2 with this clear sky flux implies a net all sky surface flux of 182 W m−2. Note that the Large et al. (1997) net solar surface flux is 185 W m−2, in good agreement with the estimate used in the present study. As pointed out by Large et al. (1997) and Doney et al. (1998), this value is 12.5% (i.e., ∼25 W m−2) less than the Bishop et al. (1997) surface flux based on ISCCP data. Furthermore, this value is in good agreement with the estimate from Oberhuber (1988) of 180 W m−2 (see Schneider et al. 1996).

As pointed out by Ramanathan et al. (1995), the surface SWCF is 36 W m−2 greater than the TOA SWCF. This difference in SWCF must be absorbed within the cloudy column, not the clear sky region. The importance of this absorbed shortwave flux to the coupled model climate will be discussed in section 5.

Zhang and McPhaden (1995) and Zhang and Grossman (1996) have carried out extensive analysis of the latent heat budget of the tropical Pacific from both the TOGA TAO array and CEPEX aircraft observations. Their analysis implies a climatological mean latent heat flux of −110 W m−2. Zhang also has pointed out the stability of this flux estimate over the observational period of the TAO array. This value is stable over the 5 yr of data from the TAO array and indicates that the flux is representative of a long-term average for the tropical Pacific region. Note that on shorter timescales (days to a week) the latent heat flux can be much larger than this climatological average. The focus of the present study is the annual mean energy budget averaged over many years (>15).

The net longwave flux term has been studied by Ramanathan et al. (1995) in some detail. Due to the presence of a large column abundance of water vapor in the warm pool region, the downward longwave flux is fairly insensitive to the presence or absence of clouds (unlike the shortwave flux). This is reflected in a surface LWCF that is typically estimated to be 15 W m−2 or less for the warm pool region (see Ramanathan et al. 1995). Thus, this number shows little variation from year to year, indicating that it is climatologically representative. The smallest term in the energy budget is the sensible heat flux from the surface. A number of independent studies find this flux to be 10 W m−2 or less in the warm pool region (see for example Waliser et al. 1996).

As noted above, the sum of the surface fluxes implies a net input of energy into the warm pool region of less than 20 W m−2. This small surface heat flux in conjunction with the fact that the SSTs in this region are high (302.3 K) presents a unique climate regime. Due to the high SSTs and moisture convergence this is also a region of major deep convective activity with associated large-scale stratiform anvil outflow (e.g., Waliser and Graham 1993). Associated with these cloud systems is large diabatic (radiative and latent) heating within the atmosphere (Houze 1989). The influence of this heating is recognized to dynamically extend to the extratropics (Lau 1997), and it accounts for a dynamical transport of 71 W m−2 of energy out of the warm pool region. It is also worth noting that since the SST gradient is very small in this region, there can be little dynamical ocean heat transport in this region (see Gent 1991). The magnitude of these fluxes indicate that the atmospheric heat transport is roughly a factor of 4 times larger than the ocean heat transport in the warm pool region.

4. Uncoupled model results

Figure 1b indicates that the ocean model when forced with the CCM3 climatology produces a warm SST bias in the western Pacific. The cause of this bias can be found by analyzing the warm pool energy budget from the CCM3 simulation forced with observed SSTs. In the uncoupled ocean model simulation forced with CCM3 data, the only flux term used is the surface solar insolation. Other surface fluxes are calculated from state variables through the flux coupler, which use the same bulk formulations employed in the CCM3. Table 1 presents the CCM3 warm pool energy budget, which also includes the observations for comparison.

At the top of the atmosphere, the model shortwave absorption is 9 W m−2 greater than the observations. This implies that the model surface–atmosphere system is receiving 9 W m−2 too much energy. The uncertainty in the regional ERBE TOA fluxes are estimated to be ±10 W m−2 (Ramanathan et al. 1989). So the agreement between model and observations is quite good. There is excellent agreement between the clear sky shortwave flux at the top of the atmosphere, the outgoing longwave flux, and the clear sky longwave flux (<3 W m−2). Thus, the top-of-atmosphere energy budget is generally in very good agreement with the observations. At the surface, the model net flux of energy into the warm pool region is overestimated by 31 W m−2. This is a significant bias in the net flux of energy into the warm pool ocean region (approximately a factor of 2). Of course in the uncoupled atmospheric model (CCM3) this bias in heat flux has no effect on the sea surface temperature. The largest source in the overestimate of surface flux into the warm pool is the cloudy sky net solar flux at the surface (52 W m−2). Note that the clear sky solar flux from the model is in excellent agreement with the observations. Thus, the model SWCF at the surface is −54 W m−2 compared to the observed value of −100 W m−2. The model clouds underestimate the surface shading effect by 46 W m−2, as pointed out in section 3, much of the shading effect is due to shortwave cloud absorption. Current models of shortwave cloud radiative transfer do not produce this magnitude of cloud absorption (see Kiehl and Trenberth 1997); thus climate models will significantly underestimate the cloud shading effect at the ocean surface. Even if we were to address the 9 W m−2 bias in top-of-atmosphere shortwave flux, the model would still underestimate the cloud absorption by 37 W m−2, which is close to the 35 W m−2 cloud absorption reported to exist in the warm pool region by Ramanathan et al. (1995). The CCM3 also overestimates the latent heat flux in this region (see also Collins et al. 1998) by 16 W m−2, which reduces the net flux bias.

The overestimate of heat flux into the uncoupled ocean model causes the SSTs to increase in the western Pacific, and thus the latent heat flux to increase as well. The net heat flux in the uncoupled ocean model forced with CCM3 data is 18 W m−2. This decrease in net heat flux is accomplished through an increase in latent heat flux. The increase in latent heat flux in the uncoupled model can only occur through an increase in local temperature. Thus, the ocean model has forced a net flux closer to the observational estimates.

The geographic distribution of the two dominant flux terms and net energy over the entire Pacific basin are shown in Figs. 2–4. Figure 2 shows the net surface energy flux from the NCOM data and from the 15-yr average of the CCM3 atmospheric simulation. Note that the model overestimate of heat flux extends across the entire Pacific basin. This bias is predominately due to the overestimate of solar flux at the surface (Fig. 3). There is actually better agreement in the latent heat fluxes from the NCOM data and the CCM3 simulation. The good agreement between CCM3 calculated latent heat flux and observations from the TAO array were also found by Collins et al. (1998). The net effect of these two dominant flux terms results, to a large extent, in the overestimate shown in Fig. 2.

5. Coupled model results

In the final spinup phase, the CCM3 is fully coupled to the ocean model. The fully coupled model produces an SST bias (Fig. 1c) that is significantly different from the uncoupled ocean model forced with CCM3 output (Fig. 1b). In the fully coupled model the warm pool is now too cold. This is in spite of the initial overestimate in heat flux in the previous simulation. How does this adjustment take place? To understand this process, the warm pool energy budget from the fully coupled CSM simulation is given Table 1, along with the observational estimates.

At the top of the atmosphere there has been little change in the solar absorbed flux, only 1 W m−2. There is a larger change in the longwave flux at the top of the atmosphere, 10 W m−2, which is associated with a decrease in very high cloud over the warm pool region in the model. At the surface there is also little change in solar flux between the CCM3 and the CSM simulations, only 3 W m−2. The largest change in the surface energy budget is a 19 W m−2 increase in the latent heat flux. This increase in loss of energy from the surface leads to a reduction in the net surface flux to 25 W m−2. This net flux is much closer to the estimated upper bound in ocean heat transport of 20 W m−2 (Gent 1991; Ramanathan et al. 1993). This increase in latent heat flux occurs in spite of a 0.7-K decrease in the warm pool SST. The cold bias in tropical SSTs extends across the equatorial Pacific (Fig. 1c) with warm water on either side of these cold waters. This warm water supports a double ITCZ structure that is unrealistic (Boville and Hurrell 1998).

The net surface flux is shown in Fig. 5 and looks quite similar to the NCOM net surface flux (see Fig. 2), which is based on observational data (see Large et al. 1997). The geographic distribution in the change in latent heat flux is shown in Fig. 6. The large increase (>40 W m−2) in the warm pool region is quite evident. The latent heat flux is determined by a bulk flux formulation that depends on three factors: an exchangecoefficient, the specific humidity deficit between the surface and surface air, and the wind speed. The change in exchange coefficient is very small, due to little change in the near-surface static stability. The change in the deficit in specific humidity is shown in Fig. 7. Changes in the warm pool region are less than 1 g kg−1. The change in the zonal wind (the dominant component in this region) is shown in Fig. 8. The zonal wind has increased by nearly 3 m s−1 in the warm pool region. This acceleration of the easterly wind component accounts for the large increase in latent heat flux. The acceleration of the easterlies is accomplished through a change in the gradient of SST in the Pacific basin. This strong coupling between gradient in SST and surface winds was discussed by Lindzen and Nigam (1987). Figure 9 shows the change in vector wind superimposed on the change in sea level pressure. Note the increase in sea level pressure centered along the date line. Thus, the change in the warm pool is a result of a nonlocal, nonlinear dynamical adjustment process in the coupled system. This response is imposed on the system due to the inefficiency of the ocean dynamical heat transport in the warm pool region. It is interesting to note that this cooling response to an initial overestimate of heat into the warm pool is exactly the opposite of the response of the uncoupled ocean model (Fig. 1b). In the uncoupled ocean simulation, there is no possibility for the atmospheric dynamics to respond to changes in SST patterns, the only response possible is a local thermodynamic one, that is, a warming of the warm pool SSTs. Thus, the fully coupled model allows for more degrees of freedom in response to biases in surface heat flux.

The time for the adjustment between phase three and phase four of the spinup procedure is very fast for the warm pool region. Figure 10 shows the simulation of SST for the last 16 yr of the phase three spinup procedure and the first 60 yr of the fully coupled model. Prior to fully coupling the SST varies between 29.8° to 30.1°C. The pattern repeats every 5 yr due to the use of a 5-yr climatology from the CCM3 simulations. Within the first year of coupling, the SSTs have adjusted to a much colder state (∼28.3°C). The time evolution of the warm pool sea surface salinity (SSS) (in ppt) is shown in Fig. 11. The horizontal line denotes the measured value reported in Lukas and Lindstrom (1991). Prior to fully coupling the SSS is too low, while within the first year of coupling the modeled SSS has increased to agree closer to the observed value of 34.2 ppt. This agreement in SSS occurs in spite of the cold bias in SST. Note that the fully coupled mean mixed layer depth of 26.9 m agrees well with the observed value of 29 m.

The temporal evolution of the net surface solar flux (Fig. 12) indicates the small change shown in Table 1. The standard deviation in this flux is small, 2.7 W m−2. Variations in the turbulent plus net longwave flux are larger than those in the net surface solar flux. Figure 13 shows the evolution of the net surface flux in the warm pool region. The gray area denotes the observational estimate of this flux. These results indicate that the uncoupled model (see years −16 to −1) has adjusted to a reasonable value compared to the observational estimates and that there is perhaps a small increase in this flux. However, the standard deviation, 7 W m−2, of this flux is large compared to the mean value. Finally, Fig. 14 shows the time evolution of the zonal component of the surface wind stress. There is a 33% increase in the zonal wind stress when the atmosphere and ocean models are fully coupled (year 0). This increase in zonal wind is responsible for the increase in latent heat in the warm pool region.

6. Summary and discussion

The SST bias is a result of a change in the gradient in SST that increases the zonal surface wind. This increase in zonal wind is required to reduce the net heat flux into this region, since the ocean dynamics are incapable of transporting more than ∼20 W m−2 from this region. The importance of SST gradients in determining local SSTs has also been discussed by Hartmann and Michelsen (1993) in the context of a simple energy balance model.

The overestimate of net heat flux is a result of excessive solar flux reaching the ocean surface, despite good agreement in solar flux at the top of the atmosphere. This indicates that the bias cannot be addressed through an increase in model cloud cover, since this would degrade the top-of-atmosphere energy budget. The model shortwave bias is similar in magnitude to the shortwave cloud absorption missing in models discussed by Ramanathan et al. (1995), Cess et al. (1995), and Pilewskie and Valero (1995). Cess et al. (1995) point out that this absorption exists at a number of geographic locations. The implications of this study is that if the clouds in the NCAR CSM absorbed roughly 35 W m−2 more radiation in the western Pacific, while preserving good agreement with the top-of-atmosphere energy budget, then the surface fluxes in the Pacific would be sufficient to insure a realistic net heat flux for this region. Results from the first phase of the spinup procedure suggest that given these fluxes, the SST biases in the tropical warm pool would be greatly reduced, since the gradient in SST would not need to drastically change to alleviate the bias in surface energy flux.

In a recent study, J. Tribbia and Y.-H. Lee (1998, personal communication) have carried out a series of simulations with a tropical Pacific basin model coupled to CCM3. Their investigations into the main cause of the biases in the mean SST and seasonal to interannual SST variability in this coupled system are a result of surface energy flux biases in this region of 20–30 W m−2. They have shown that adjusting for these flux biases not only improves the mean SST bias, but also greatly improves the simulated seasonal and interannual variability in the tropical Pacific SSTs.

These results indicate the sensitivity of a coupled atmosphere–ocean model to biases in surface energy flux. They also illustrate that the response of a fully coupled model to such biases can be quite different than the response of an uncoupled ocean model. The added complexity of the coupled system allows for nonlocal, nonlinear responses that are prohibited in the uncoupled model. Continued efforts to reduce biases in surface energy fluxes in a physically based manner should help reduce biases in the simulated SSTs of the fully coupled model.

Acknowledgments

This study has benefited from discussions with members of the NCAR Oceanography Section and with Dr. B. Boville. I wish to thank Dr. F. Bryan for providing me with results from the ocean model simulations. I wish to thank Dr. W. Collins for discussions concerning the observational estimates of the warm pool energy budget. I thank J. Dunn for producing the figures for this study. This study was supported in part by the DOE Atmospheric Radiation Measurements Program Grant DE-FG05-93ER61376.

REFERENCES

  • Bishop, J. K. B., W. B. Rossow, and E. G. Dutton, 1997: Surface solar irradiance from the International Satellite Cloud Climatology Project 1983–1991. J. Geophys. Res.,102, 6883–6910.

  • Bonan, G., 1998: The land surface climatology of the NCAR land surface model (LSM 1.0) coupled to the NCAR Community Climate Model (CCM3). J. Climate,11, 1307–1326.

  • Boville, B. A., and P. R. Gent, 1998: The NCAR Climate System Model, version one. J. Climate,11, 1115–1130.

  • ——, and J. Hurrell, 1998: A comparison of the atmospheric circulations simulated by the CCM3 and CSM1. J. Climate,11, 1327–1341.

  • Cess, R. D., 1995: Absorption of solar radiation by clouds. Science,267, 496–499.

  • Chou, M.-D., and W. Zhao, 1997: Estimation and model validation of surface solar radiation and cloud radiative forcing using TOGA COARE measurements. J. Climate,10, 610–620.

  • Collins, W. D., J. Wang, J. T. Kiehl, and G. J. Zhang, 1997: Comparison of tropical ocean–atmosphere fluxes with the NCAR Community Climate Model CCM3. J. Climate,10, 3047–3058.

  • Doney, S. C., W. G. Large, and F. O. Bryan, 1998: Surface ocean fluxes and water-mass transformation rates in the coupled NCAR Climate System Model. J. Climate,11, 1420–1441.

  • Frey, H., M. Latif, and T. Stockdale, 1997: The coupled GCM ECHO-2. Part I: The tropical Pacific. Mon. Wea. Rev.,125, 703–720.

  • Gent, P. R., 1991: The heat budget of the TOGA-COARE domain in an ocean model. J. Geophys. Res.,96, 3323–3330.

  • ——, and J. C. McWilliams, 1990: Isopycnal mixing in ocean circulation models. J. Phys. Oceanogr.,20, 150–155.

  • ——, F. O. Bryan, G. Danabasoglu, S. C. Doney, W. R. Holland, W. G. Large, and J. C. McWilliams, 1998: The NCAR Climate System Model global ocean component. J. Climate,11, 1287–1306.

  • Hack, J. J., 1994: Parameterization of moist convection in the National Center for Atmospheric Research Community Climate Model (CCM2). J. Geophys. Res.,99, 5551–5568.

  • Hartmann, D. L., and M. L. Michelsen, 1993: Large-scale effects on the regulation of tropical sea surface temperature. J. Climate,6, 2049–2062.

  • Holtslag, A. A. M., and B. A. Boville, 1993: Local versus non-local boundary layer diffusion in a global climate model. J. Climate,6, 1825–1842.

  • Houze, R. A., 1989: Observed structure of mesoscale convective systems and implications for large-scale heating. Quart. J. Roy. Meteor. Soc.,115, 425–461.

  • Kiehl, J. T., 1994: On the observed near cancellation between longwave and shortwave cloud forcing in tropical regions. J. Climate,7, 559–565.

  • ——, and K. E. Trenberth, 1997: Earth’s annual global mean energy budget. Bull. Amer. Meteor. Soc.,78, 197–208.

  • ——, J. J. Hack, G. Bonan, B. A. Boville, D. Williamson, and P. J. Rasch, 1998: The National Center for Atmospheric Research Community Climate Model: CCM3. J. Climate,11, 1131–1149.

  • Large, W. G., J. C. McWilliams, and S. C. Doney, 1994: Oceanic vertical mixing: A review and a model with a nonlocal boundary layer parameterization. Rev. Geophys.,32, 363–403.

  • ——, G. Danabasoglu, S. C. Doney, and J. C. McWilliams, 1997: Sensitivity to surface forcing and boundary layer mixing in a global ocean model: Annual-mean climatology. J. Phys. Oceanogr.,27, 2418–2447.

  • Lau, N.-C., 1997: Interactions between global SST anomalies and midlatitude atmospheric circulation. Bull. Amer. Meteor. Soc.,78, 21–33.

  • Lindzen, R. S., and S. Nigam, 1987: On the role of sea surface temperature gradients in forcing low-level winds and convergence in the tropics. J. Atmos. Sci.,44, 2418–2436.

  • Lukas, R., and E. Lindstrom, 1997: The mixed layer of the western equatorial Pacific ocean, J. Geophys. Res.,102, 3344–3357.

  • Ma, C.-C., C. R. Mechoso, A. Arakawa, and J. D. Farrara, 1994: Sensitivity of a coupled ocean–atmosphere model to physical parameterizations. J. Climate,7, 1883–1896.

  • ——, ——, A. W. Robertson, and A. Arakawa, 1996: Peruvian stratus clouds and the tropical Pacific circulation: A coupled ocean–atmosphere GCM study. J. Climate,9, 1635–1645.

  • Mechoso, C. R., and Coauthors, 1995: The seasonal cycle over the tropical Pacific in coupled ocean–atmosphere general circulation models. Mon. Wea. Rev.,123, 2825–2838.

  • Oberhuber, J. M., 1988: An atlas based on the “COADS” data set: The budgets of heat, buoyancy and turbulent kinetic energy at the surface of the global ocean. Max-Planck Institute for Meteorology Rep. 15, Max-Planck Institute for Meteorology, Hamburg, Germany, 199 pp.

  • Pilewskie, P., and F. P. J. Valero, 1995: Direct observations of excess solar absorption by clouds. Science,267, 1626–1629.

  • Ramanathan, V., R. D. Cess, E. F. Harrison, P. Minnis, B. R. Barkstrom, E. Ahmad, and D. L. Hartmann, 1989: Cloud radiative forcing and climate: Results from the Earth Radiation Budget Experiment. Science,243, 57–63.

  • ——, R. Dirks, R. Grossman, A. Heymsfield, J. Kuettner, and F. Valero, 1993: Central Equatorial Pacific Experiment design. Center for Clouds, Chemistry and Climate, University of California, San Diego, CA, 54 pp. [Available from NCAR, P. O. Box 3000, Boulder, CO 80307.].

  • ——, B. Subasilar, G. J. Zhang, W. Conant, R. D. Cess, J. T. Kiehl, H. Grassl, and L. Shi, 1995: Warm pool heat budegt and shortwave cloud forcing: A missing physics? Science,267, 499–503.

  • Russell, G. L., J. R. Miller, and D. Rind, 1995: A coupled atmosphere–ocean model for transient climate change studies. Atmos.–Ocean,33, 683–730.

  • Schneider, E. K., Z. Zhu, B. S. Giese, B. Huang, B. P. Kirtman, J. Shukla, and J. A. Carton, 1997: Annual cycle and ENSO in a coupled ocean–atmosphere general circulation model. Mon. Wea. Rev.,125, 680–702.

  • Schneider, N., T. Barnett, M. Latif, and T. Stockdale, 1996: Warm pool physics in a coupled GCM. J. Climate,9, 219–239.

  • Shea, D. J., K. E. Trenberth, and R. W. Reynolds, 1990: A global monthly sea surface temperature climatology. NCAR Tech. Note NCAR/TN-345, 167 pp. [Available from NCAR, P. O. Box 3000, Boulder, CO 80307.].

  • Stockdale, T., M. Latif, G. Burgers, and J.-O. Wolff, 1994: Some sensitivities of a coupled ocean–atmosphere GCM. Tellus,46A, 367–380.

  • Waliser, D. E., and N. E. Graham, 1993: Convective cloud systems and warm pool sea surface temperatures: Coupled interactions and self regulation. J. Geophys. Res.,98, 12 881–12 893.

  • ——, W. D. Collins, and S. P. Anderson, 1996: An estimate of the surface shortwave cloud forcing over the western pacific during TOGA COARE. Geophys. Res. Lett.,23, 519–522.

  • Weatherly, J., B. Briegleb, W. G. Large, and T. Bettge, 1998: Sea ice and polar climate in the NCAR CSM. J. Climate,11, 1472–1486.

  • Webster, P. J., 1994: The role of hydrological processes in ocean–atmosphere interactions. Rev. Geophys.,32, 427–476.

  • ——, and R. Lukas, 1992: TOGA COARE: The Coupled Ocean–Atmosphere Response Experiment. Bull. Amer. Meteor. Soc.,73, 1377–1416.

  • Zhang, G. J., and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Centre general circulation model. Atmos.–Ocean,33, 407–446.

  • ——, and M. J. McPhaden, 1995: The relationship between sea surface temperature and latent heat flux in the equatorial Pacific. J. Climate,8, 589–605.

  • ——, and R. L. Grossman, 1996: Surface evaporation during the Central Equatorial Pacific Experiment: A climate-scale perspective. J. Climate,9, 2522–2537.

 Fig. 1.
Fig. 1.

Difference in annual mean SST (a) uncoupled ocean model forced with the observational analysis of Large et al. (1997), (b) uncoupled ocean model forced with CCM3 data, and (c) fully coupled CSM from observed SSTs from Shea et al. (1990) climatology. Contour interval is 1°C.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 2.
Fig. 2.

Annual mean net surface energy budget (W m−2) from (a) forcing of Large et al. (1997) and (b) CCM3. Contour interval is 15 W m−2.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 3.
Fig. 3.

Annual mean net surface shortwave flux (W m−2) from (a) forcing of Large et al. (1997) and (b) CCM3. Contour interval is 10 W m−2.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 4.
Fig. 4.

Annual mean net surface latent heat flux (W m−2) from (a) forcing of Large et al. (1997) and (b) CCM3. Contour interval is 15 W m−2.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 5.
Fig. 5.

Annual mean net surface energy flux (W m−2) from the fully coupled CSM. Contour interval is 30 W m−2.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 6.
Fig. 6.

Annual mean difference (CSM − CCM3) in latent heat flux (W m−2) for the tropical Pacific region. Contour interval is 20 W m−2.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 7.
Fig. 7.

Annual mean difference (CSM − CCM3) in defecit in specific humidity (Kg Kg−1) for the tropical Pacific region. Contour interval is 0.001 Kg Kg−1.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 8.
Fig. 8.

Annual mean difference (CSM − CCM3) in surface zonal wind (m s−1) for the tropical Pacific region. Contour interval is 1 m s−1.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 9.
Fig. 9.

Annual mean difference (CSM − CCM3) in vector surface wind (m s−1) for the tropical Pacific region and sea level pressure (Pa). Contour interval is 30 Pa.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 10.
Fig. 10.

Time evolution of the annual mean sea surface temperature (°C) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 11.
Fig. 11.

Time evolution of the annual mean SSS (ppt) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 12.
Fig. 12.

Time evolution of the annual mean net shortwave surface energy flux (W m−2) (– – –) and latent plus sensible plus longwave flux (———) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 13.
Fig. 13.

Time evolution of the annual mean net surface energy flux (W m−2) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Fig. 14.
Fig. 14.

Time evolution of the annual mean surface zonal wind stress (dyn cm−2) for the warm pool region. Negative times denote the uncoupled model simulation. Positive times denote the fully coupled CSM simulation.

Citation: Journal of Climate 11, 6; 10.1175/1520-0442(1998)011<1342:SOTTPW>2.0.CO;2

Table 1.

Observational CCM3 and CSM energy budgets for the tropical Pacific warm pool region (10°S, 10°N; 140°E, 170°E). Negative sign denotes energy loss at either the top of the atmosphere or at the surface. Here S denotes the net shortwave flux, Sclr denotes the net clear sky shortwave flux, F is the net longwave flux, Fclr is the net clear sky longwave flux, LH is the latent heat flux, SH is the sensible heat flux, and NET is the net energy flux. Source of surface observations is discussed in section 3.

Table 1.

* An electronic supplement to this article may be found on the CD-ROM accompanying this issue or at http://www.ametsoc.org/AMS.

+

The National Center for Atmospheric Research is sponsored by the National Science Foundation.

Save