Climate Simulations At Ccsr

CCSR was established in 1991. Its objectives are the development of climate models and achieving an understanding of the dynamics of the climate system by using numerical models. When we began to discuss the implementation plan for the center, we defined the following strategies for model development:

1. Because the center belongs to the university, emphasis should be placed on understanding climate rather than simulating and predicting climate.

2. Much attention should be focused on attacking problems by using numerical models.

3. More emphasis should be placed on the interactions between modeling and remote sensing, because global modeling requires global data for many currently unobserved quantities, which may be obtained in the future through satellite remote sensing with surface validation sites. This idea is the core of our model evaluation strategy.

4. We have to educate students to focus not on the atmosphere or ocean alone, but rather to dedicate themselves to climate issues involving both the atmosphere and ocean.

In the following subsections, we briefly present our models and introduce research results that have been obtained by using these models in global warming experiments and simulations of important climate phenomena such as the quasi-biennial oscillation.

A. The CCSR Atmospheric General Circulation Model

Ever since CCSR was established, development of our atmospheric general circulation model (AGCM) has been conducted jointly with the National Institute for Environmental Studies (NIES). CCSR has received computer time and manpower from the NIES. The first version of our AGCM was completed in 1995 (Numaguti et al., 1997). The key features of the model can be summarized as follows:

1. It is a three-dimensional hydrostatic primitive global spectral model with a sigma coordinate. Finite-difference methods are used to represent the vertical structure of the atmosphere. The standard model resolution is T21 or T42, with L20.

2. The prognostic variables of the CCSR AGCM are vorticity and divergence, temperature, surface pressure, specific humidity, soil temperature, soil moisture, snow amount, and river water storage.

3. The radiation code is based on the two-stream discrete ordinate method and the k distribution method (Nakajima and Tanaka, 1986). The radiative flux is calculated in 18 wavelength bands. Band absorptions by H20, C02, 03, N20, CH4, and 16 species of CFCs are considered. Continuum absorption by H20, 02, and 03 are included. Rayleigh scattering by gases and particle scattering and absorption by cloud and aerosol particles are considered.

4. Convection is parameterized using a simplified Arakawa-Schubert scheme. Large-scale condensation is parameterized based on the scheme of Le Treut and Li (1991). The cloud water mixing ratio is a prognostic variable.

5. Surface fluxes are based on the bulk formula (Louis, 1979). The planetary boundary layer is parameterized using a closure model developed by Mellor and Yamada (1974, 1982).

6. The effects of sub-grid-scale orographic gravity waves are included following McFarlane (1987).

7. The land-surface model is a modified bucket scheme (Manabe et al., 1965).

B. The CCSR Ocean General Circulation Model

The CCSR ocean general circulation model (OGCM) has been developed by members of the CCSR staff. Its finite-difference scheme is almost the same as that of the Geophysical Fluid Dynamics Laboratory (GFDL) model (Bryan, 1969), except that a weighted upcurrent scheme is adopted for the advection of temperature and salinity (Suginohara and Aoki, 1991). The OGCM has exceptional computational efficiency.

C. An AMIP Run

Using these models, we participated in various comparison projects, e.g., AMIP (the Atmospheric Model Intercomparison Project), CMIP (the Coupled Model Intercomparison Project), and PMIP (the Paleo-climate Model Intercomparison Project). Unfortunately, we were late in sending our results, and so our results are not included in the AMIP reports. Our AMIP results are presented here, however. Figure la shows the zonal mean temperature and Fig. lb the zonal wind velocity for June, July, and

100-

300 400 500

700 850-I 1000

Zonal mean U-component JJA

Zonal mean temperature JJA

Zonal mean U-component JJA

100-

1

9

\\vit

\\1\1

|

\ \ \\\ \\4p'//,

u //

\ \ \ \\ \35'//

// \ — s

\ \\VW/

/ j \ — J

V / / \10 /

\ \v2oyjj

\

V \ \ ^ ///

\ 5

) Vin'/A'

' > »

60S 30S

Zonal mean temperature JJA

700 850-I 1000

Figure 1 Height-latitude cross section for the zonal mean climatology based on the CCSR/NIES AGCM AMIP results, (a) Zonal wind (U) and (b) temperature (T) for summer (June, July, and August), (c) Zonal wind and (d) temperature for winter (December, January, and February). The contour interval for the winds is 5 m s ~1. Broken lines indicate negative values. The contour interval for the temperature is 10 K.

Figure 1 Height-latitude cross section for the zonal mean climatology based on the CCSR/NIES AGCM AMIP results, (a) Zonal wind (U) and (b) temperature (T) for summer (June, July, and August), (c) Zonal wind and (d) temperature for winter (December, January, and February). The contour interval for the winds is 5 m s ~1. Broken lines indicate negative values. The contour interval for the temperature is 10 K.

August. The zonal mean temperature is shown in Fig. lc and the zonal wind velocity in Fig. Id for December, January, and February. The corresponding figures based on NCEP data are presented in Fig. 2. Comparison shows that the performance of our model is average, although there are systematic errors. For example, a cold bias and westerly bias exist in the

Zonal mean temperature JJA

Zonal mean temperature JJA

Renewable Energy Eco Friendly

Renewable Energy Eco Friendly

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable.

Get My Free Ebook


Post a comment