辐射参数化方案对一个海气耦合模式云和辐射模拟的影响

比较Morcrette辐射方案和Fu_Liou辐射方案对NCC/IAP T63海气耦合模式云和辐射模拟的影响, 结果表明:两种方案模拟的大气顶入射辐射存在明显的差异; 晴空大气Fu_Liou方案的短波吸收能力在全球普遍较Morcrette方案低; 在60°S~60°N之间, Fu_Liou方案模拟的行星反照率更接近于ERBE卫星观测; 在对大气顶净辐射的模拟上, 除了冬季的太平洋和大西洋东岸云量明显减少的部分地区外, Fu_Liou方案对大气顶净辐射的模拟总体上较Morcrette方案有了较为明显的改善; Fu_Liou方案模拟的海洋低层云显著减少, 而热带地区高云的模拟明显增加; 由于采用了“二元云量”算法, 尽管云量有所减少, Fu_Liou方案模拟的云短波吸收作用仍有所增强, 一定程度上改进了Morcrette方案云的短波吸收作用偏弱的现象。

The Effects of Radiative Parameterization Scheme on Simulation of Cloud and Radiation in an AOGCM

Radiation is an important process in atmosphere. Currently the radiative parameterization schemes used in climate models are mainly simplified arithmetic, which show great uncertainty in climate change projection. To better understand the effects of radiative parameterizations on cloud and radiation simulation, a new parameterization scheme, Fu_Liou scheme, is introduced into NCC/IAP T63 AOGCM to replace the old scheme, Morcrette scheme. NCC/IAP T63 AOGCM is a spectral model with triangular truncation at wave number 63. The horizontal resolution is 1.875°×1.875° for both atmosphere and ocean component, and there are 16 and 30 vertical layers for atmosphere and ocean respectively. The main differences between two radiative schemes are in such aspects as division of waveband, approximation for radiative transfer, and treatment of cloud layer overlap in radiation calculation. In Fu_Liou scheme, the waveband is divided into 18 sub-wavebands, 6 for shortwave and 12 for long wave, while there are only 8 sub-wavebands (2 for shortwave and 6 for long wave) in Morcrette scheme. Fu_Liou scheme adopts δ-four-stream approximation which has high calculating accuracy, and uses the binary cloud when calculating the cloud effects on radiation. Two 20-year integrations are processed by use of the two schemes respectively. The first 10 years are used for coupling adjustment, and the last 10 years for analysis. Two types of monthly mean data, 40-year reanalysis data of European Centre for Medium-range Weather Forecasts (ECMWF)(ERA-40) and data from Earth Radiation Budget Experiment (ERBE), are used to verify the results. The effects of the two schemes on the simulation of cloud and radiation are analysed in detail. The main results are as follows:Firstly, the incident radiation at TOA simulated by two schemes differs obviously, which is mainly caused by the difference in radiative arithmetic. Secondly, in clear sky the shortwave absorption of FuLiou scheme is generally lower than Morcrette scheme, especially in middle and high latitudes of winter hemisphere. The planetary albedo simulated by Fu_Liou scheme is closer to ERBE data than Morcrette scheme between 60°S and 60°N, mainly due to the improvement of simulation on clear-sky albedo. Thirdly, the simulated net radiation at TOA in Fu_Liou scheme is improved obviously as compared with Morcrette scheme except for part of the east Pacific and east Atlantic where the cloud cover decreases remarkably in winter. Finally, the low stratocumulus simulated by Fu_Liou scheme decreases remarkably, which is caused by the weakening of static stability over ocean. The high cloud increases obviously, mainly due to the intensifying of convection in tropical region, as compared with Morcrette scheme. Although the cloud cover decreases, the use of binary cloud increases the cloud shortwave absorption in Fu_Liou scheme which improves the weak absorption in Morcrette scheme in some degree.