论文部分内容阅读
评估了中国科学院大气物理研究所大气科学和地球流体动力学数值模拟国家重点实验室海洋环流模式L30T63和海气耦合模式FGCM 0模拟的热带太平洋年平均状态 ,资料取自L30T63由观测的大气强迫驱动的Control试验、由NCARCCM3大气强迫驱动的Spinup试验、以及相应的海气耦合模式FGCM 0。主要的结论是 :( 1 )在“准确”的海表强迫下 ,Control模拟的海面温度和温跃层与观测结果相当接近 ,模式的固有误差是赤道冷舌过分西伸和东南太平洋温跃层偏浅。 ( 2 )Spinup能模拟出合理的热带太平洋上层海洋环流 ,但存在两个问题 ,即 :暖池区海面温度显著偏高、沿赤道的梯度过大 ;赤道温跃层偏浅、东西向坡度偏小 ,它们分别与CCM3提供的海表短波辐射通量和风应力的系统误差有关。这两个问题很可能是海气耦合模式FGCM 0运行初期误差迅速发展的重要原因。 ( 3)FGCM 0模拟的赤道暖池区上层 1 0 0m的平均温度比观测低 3℃。分析表明FGCM 0夸大了暖池区海洋动力过程的降温作用 ,使得模拟的“暖池”在一定程度上具有冷舌的属性。FGCM 0模拟的热带南太平洋温跃层比观测结果偏浅数十米到 1 0 0m ,以致赤道两侧的上层海洋温度分布趋于对称 ,成为“doubleITCZ”现象在上层海洋中的表现。风应力旋度的系统误差和垂直混合随深度衰减过快?
The annual mean state of the tropical Pacific simulated by ocean circulation model L30T63 and air-sea coupling model FGCM 0 of State Key Laboratory of Numerical Modeling of Atmospheric Sciences and Geodynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences was evaluated. Data were taken from L30T63 driven by observed atmospheric forcing Control test, the Spinup test driven by the NCARCCM3 atmosphere, and the corresponding air-sea coupling model FGCM0. The main conclusions are as follows: (1) Under the “accurate” sea surface pressure, the sea surface temperature and thermocline controlled by Control are quite close to the observation results. The inherent errors of the model are the over-extensional equatorial cold tongue and the thermocline over the southeastern Pacific shallow. (2) Spinup can simulate a reasonable upper tropical Pacific Ocean circulation, but there are two problems: the sea surface temperature in the warm pool area is significantly higher and the gradient along the equator is too large; the equatorial thermocline is shallow and the east-west gradient Small, they are respectively related to the systematic errors of short-wave radiant flux and wind stress of the sea surface provided by CCM3. These two problems are likely to be the main reason for the rapid development of the error in the early stage of operation of the air-sea coupling FGCM 0. (3) The average temperature of the upper 10 000 m in the simulated equatorial warm pool area is 3 ° C lower than the observation. The analysis shows that FGCM 0 exaggerated the cooling effect of the marine dynamic process in the warm pool area, making the simulated “warm pool” have the cold tongue property to a certain extent. The simulated tropical South Pacific thermocline in FGCM 0 is slightly more than a few tens of meters above the observed value, resulting in a symmetrical upper ocean temperature distribution on both sides of the equator, becoming a manifestation of the “doubleITCZ” phenomenon in the upper ocean. Systematic error of wind stress curl and vertical mixing decay too fast with depth?