论文部分内容阅读
海洋同化系统为年代际预测试验提供初值,其性能可能会影响年代际预测技巧,因此评估其对重要年代际变化现象的模拟能力非常必要。观测发现,东亚夏季风(EASM)和前冬厄尔尼诺—南方涛动(ENSO)的关系在1970s末加强,随后在1990s中期后减弱。基于FGOALS-s2耦合气候模式的海洋同化系统评估了其对这2次年代际变化的模拟能力。结果表明,决定模式能否再现EASM和前冬ENSO关系的年代际变化有2个重要因素:(1)与前冬ENSO有关的夏季印度洋—太平洋海温型的年代际变化;(2)模式西北太平洋反气旋对热带海温的响应偏差。模式中西北太平洋反气旋与东北印度洋的暖海温关系稳定,当1970s末前冬ENSO对夏季印度洋海温影响显著增强时,模式能够模拟出北印度洋降水以及赤道东印度洋至海洋大陆上空Kelvin波的增强,从而可再现EASM与前冬ENSO关系的增强;而1990s中期后模式中与前冬ENSO有关的东北印度洋海温异常进一步增强,与观测相反,使得模式未能再现观测中EASM与前冬ENSO关系的减弱。此外,1990s中期后模式对夏季中太平洋冷海温异常的Rossby波响应存在较大偏差,是其未能再现此次年代际变化的另一个原因。研究表明,与ENSO有关的热带印太海温的年代际变化预测水平和模式对海温的响应偏差将在一定程度上制约模式对EASM与ENSO关系的年代际预测能力。
Marine assimilation systems provide initial values for interdecadal prediction experiments and their performance may affect interdecadal prediction skills, so assessing their ability to simulate important decadal variability is necessary. Observations revealed that the relationship between the East Asian Summer Monsoon (EASM) and the Pre-winter El Niño-Southern Oscillation (ENSO) intensified in the late 1970s and then weakened after the mid-1990s. The ocean assimilation system based on the coupled climate model of FGOALS-s2 assesses its ability to simulate these two interdecadal variations. The results show that there are two important factors that determine whether decisional patterns can reproduce the interdecadal variation of the EASM-Pre-ENSO relationship: (1) the interdecadal variation of summer Indian-Pacific SST in relation to the pre-winter ENSO; and (2) the model northwest Pacific anticyclone response to tropical SST. Mode, the Northwest Pacific anticyclone has a stable relationship with the northeastern Indian Ocean’s warm sea temperature. When the pre-1970s ENSO has a significant impact on the summer Indian Ocean SST, the model can simulate the North Indian Ocean Precipitation and the Kelvin waves over the equatorial East Indian Ocean to the ocean Which can reproduce the enhancement of the relationship between EASM and pre-winter ENSO. However, the SST anomalies in the northeastern Indian Ocean associated with the pre-winter ENSO increase further in the mid-1990s, which make the model fail to reproduce the observed EASM and pre-winter ENSO Weakening the relationship. In addition, there is a large deviation of the Rossby wave response to the summer solonospheric anomalies in the mid-1990s, which is another reason why it failed to reproduce the interdecadal variation. Studies have shown that the deviation of the ENSO-related interdecadal variability forecast of the tropical Indian Ocean temperature from the SST response will to a certain extent constrain the interdecadal predictive power of the EESM-ENSO relationship.