Model tropical Atlantic biases underpin diminished Pacific decadal variability

Model tropical Atlantic biases underpin diminished Pacific decadal variability
复制标题

DOI:
10.1038/s41558-018-0163-4
复制
发表时间:
2018-06-01
影响因子:
30.7
通讯作者:
Collins, Mat
Collins, Mat
中科院分区:
地球科学1区
文献类型:
--
作者:
McGregor, Shayne;Stuecker, Malte F.;Collins, Mat

文献摘要

被引文献

相似文献

近几十年来,太平洋信风表现出前所未有的增强(1)。这种增强与东太平洋海面冷却(2)和世纪初全球表面变暖的减缓(2,3)以及许多其他重大影响(4-9)有关。虽然一些气候模式产生了这些最近观察到的趋势的时间(10),但它们都未能产生趋势幅度(2,11,12)。这可能部分与模式对低频太平洋变率和十年风趋势的明显低估有关(2,11 -13),或者是由于强迫响应的错误表述(1,14 -16),或者两者兼而有之。太平洋和大西洋盆地之间日益突出的联系已被确定为太平洋信风加强的关键驱动力(12,17 -20)。在这里,我们使用有针对性的气候模型实验表明,结合最近的大西洋变暖趋势与典型的气候模型的偏见,导致大大低估了太平洋风和表面温度的反应。这种低估主要是由于大气加热对热带大西洋变暖趋势的反应减少和东移。这一结果表明,最近的太平洋趋势和模式的年代际变率可能会更好地捕捉模式与改进的平均状态的气候。
Pacific trade winds have displayed unprecedented strengthening in recent decades(1). This strengthening has been associated with east Pacific sea surface cooling(2) and the early twenty-first-century slowdown in global surface warming(2,3), amongst a host of other substantial impacts(4-9). Although some climate models produce the timing of these recently observed trends(10), they all fail to produce the trend magnitude(2,11,12). This may in part be related to the apparent model underrepresentation of low-frequency Pacific Ocean variability and decadal wind trends(2,11-13) or be due to a misrepresentation of a forced response(1,14-16) or a combination of both. An increasingly prominent connection between the Pacific and Atlantic basins has been identified as a key driver of this strengthening of the Pacific trade winds(12,17-20). Here we use targeted climate model experiments to show that combining the recent Atlantic warming trend with the typical climate model bias leads to a substantially underestimated response for the Pacific Ocean wind and surface temperature. The underestimation largely stems from a reduction and eastward shift of the atmospheric heating response to the tropical Atlantic warming trend. This result suggests that the recent Pacific trends and model decadal variability may be better captured by models with improved mean-state climatologies.