BCC气候模式对中国近50a极端气候事件的模拟评估

Evaluation of recent 50 years extreme climate events over China simulated by Beijing Climate Center(BCC) climate model

利用中国区域437站1958—2005年逐日气温和降水资料,评估了国家气候中心(BeijingCli-mateCenter,BCC)气候模式对中国近50a极端气候事件空间分布、时间演变等方面的模拟能力。结果表明:1)模式对极端温度和降水多年平均的空间分布具有一定的模拟能力,但尚存在系统性的偏差。暖昼日数的模拟优于冷夜日数,全年冷夜日数的模拟优于冬季和夏季的模拟,而全年和夏季暖昼日数的模拟优于冬季的模拟。极端降水频次的模拟优于极端降水量的模拟;夏季和全年的模拟优于冬季的模拟。夏季极端降水频次的模拟较好,但冬季的模拟在长江中下游和华南偏小、北方偏大,而全年的模拟在长江下游及南部沿海地区系统性偏大;夏季和全年极端降水量的模拟系统性偏低,而冬季在北方偏高、南方偏低。2)模式较好地模拟出了夏季和全年冷夜日数的全国较为一致的减少趋势,再现东北和东南沿海地区冬季冷夜日数的减少趋势,但模拟的趋势较实测偏弱。模式对暖昼日数长期趋势的模拟效果较理想,较好地反映出了大部分地区暖昼事件发生频率显著增加的特征,但冬季的模拟尚有待改进。模式较好地模拟出了夏季和全年极端降水频次的长期趋势,较好地刻画了极端降水频次"南增北减"的特征;模式对冬季极端降水频次的变化趋势几乎无模拟能力。同样,模式也较好地模拟出了极端降水量夏季南增北减的分布形势和冬季的总体增加趋势,但对全年的模拟不理想。3)模式能较好地模拟出冷夜日数和暖昼日数异常变化的主要空间型,对EOF第一模态的时间演变特征具有一定模拟能力;但模式对第二模态时间演变特征的刻画能力较差。模式对极端降水指标的年际变化具有一定的模拟能力,对部分区域极端降水事件的年际变化具有较好的模拟能力;但模拟能力表现出了明显的区域性差异,部分区域极端降水年际变化的模拟结果与实况甚至相反,模式对极端降水年际变化的模拟能力还有待提高。所得结果可为BCC气候模式的改进及极端气候模拟、预估提供一定的参考。

By using 1958-2005 daily air temperature and precipitation data from 437 stations over China,the performance of the Beijing Climate Center（BCC） climate model on simulating the spatial and temporal variations of the extreme temperature and precipitation events is evaluated.Results show：1） BCC model can capture the predominant spatial patterns of the averaged extreme temperature and precipitation events,but produce systematic bias compared to the observations.On the whole,the model performs better in warm days（TX95p） simulation than cold nights（TN5p） simulation.The simulation of annual TN5p is better than those of winter and summer,and the simulations of the annual and summer TX95p are better than that of winter.The simulation of the frequency of extreme precipitation（R95p） is better than that of the extreme precipitation amounts（R95t） ,and the simulations of the annual and summer R95p（R95t） are better than that of winter.BCC model reproduces main features of summer R95p,but overestimates（underestimates） winter R95p in the northern regions of China（the mid-lower reaches of Yangtze River and South China） .Systematically positive bias is found in the annual R95p simulation over the lower Yangtze River and the southern part coastal areas of China.It is found that both summer and annual R95t are underestimated.For winter case,R95t is overestimated（underestimated） to some extent in north（south） part of China.2） Consistently decreasing trends of summer and annual TN5p over China and decreasing trend of winter TN5p in the Northeast and Southeast China are well reproduced,however the simulated trends are somewhat underestimated.The model has very good performance in simulating the long-term trends of TX95p and describes the significant increasing trends over most areas.However,the simulation of winter TX95p needs to be improved in the future.About summer and annual R95p,the observed increasing（decreasing） trends in the southern（northern） region are well simulated.The model shows poor performance in winter R95p simulation.The model also produces satisfactory results for both summer and winter R95t simulation,but gives bad results for the annual case.3） BCC model captures the spatial patterns of leading EOF modes of TN5p and TX95p anomalies,also the main features of the temporal evolution of the first EOF modes,but fails to describe the temporal evolution of the second EOF modes.The model reproduces some basic features about the interannual variation of the extreme precipitation,and has very good performance in some areas.However,the model performance exhibits evident regional differences,and the simulated results over some areas are even contrary to the observation,which suggests that the capability of the model in simulating the interannual variability of extreme precipitation events needs to be greatly improved.This study provides some clues for BCC model improvement as well as its applications in extreme climate events simulation and projection.