青藏高原未来气候变化预估:CMIP5模式结果
Climate Change Projection on the Tibetan Plateau:Results of CMIP5 Models
-
摘要: 本文使用国际耦合模式比较计划第5阶段(CMIP5)中对青藏高原气候模拟较优的气候模式, 在RCP4.5中等偏低辐射强迫情景下对青藏高原未来气候变化进行了预估研究。结果表明, 青藏高原年均地表气温在2006~2100年的线性趋势平均为0.26℃/10a, 增暖幅度与海拔高度大体成正比;相比于1986~2005年参考时段, 2090年代平均升温2.7℃, 21世纪末期增温幅度明显高于早期和中期;在早、中和末期, 年均增温分别为0.8~1.3℃、1.6~2.5℃和2.1~3.1℃;各季节也均为变暖趋势, 其中冬季增温最大。对于年均降水来说, 未来百年将小幅增加, 集合平均趋势为1.15%/10a, 2090年代较参考时段增加10.4%;在早、中和末期的变化范围分别为-1.8%至15.2%、-0.9%至17.8%和1.4%至21.3%;季节降水也呈增加趋势, 夏季增幅明显高于其余三个季节且在21世纪末期较大, 青藏高原未来年均降水增加主要来自于夏季。需要指出的是, 上述预估结果在气候模式间存在着一定的差异, 未来气候变化的不确定性范围较大, 地表气温的可信度相对较高, 而降水的则偏低。Abstract: Climate change for the 21st century over the Tibetan Plateau (TP) is projected using multiple climate models within the phase five of the Coupled Model Intercomparison Project under the Representative Concentration Pathway 4.5 (RCP4.5) scenario.These models have a demonstrated ability to simulate modern climatology.The results show an annual warming trend of 0.26℃ per decade, which correlates positively with the topographical height in 2006-2100.With respect to the reference period 1986-2005, the TP annual temperature increases 2.7℃ in the 2090s, which is stronger than the warming in the early and middle 21st century.In the early, middle, and end periods, annual warming is 0.8-1.3℃, 1.6-2.5℃, and 2.1-3.1℃, respectively.Temperature increases are seen in all seasons, with the strongest warming occurring in winter.On the contrary, overall annual precipitation increases slightly on the TP, with a trend of 1.15% per decade during 2006-2100 and an increase of 10.4% in the 2090s relative to the reference period.Annual precipitation ranges from -1.8% to 15.2% in the early period, from -0.9% to 17.8% in the middle period, and from 1.4% to 21.3% in the end period.Precipitation generally increases in all seasons; the summer increase is larger compared with other seasons, particularly for the end of the 21st century.The annual precipitation increase occurs mainly in summer.It is noted that the above results differ somewhat among models, which indicates a relatively large level of uncertainty and a relatively high (low) reliability of temperature (precipitation) projection.