Interdecadal change in the Northern Hemisphere seasonal climate prediction skill: part I. The leading forced mode of atmospheric circulation
Interdecadal change in the Northern Hemisphere seasonal climate prediction skill: part I. The leading forced mode of atmospheric circulation
复制标题
DOI：
10.1007/s00382-013-1988-1
复制
发表时间：
2014-09
期刊：
影响因子：
4.6
通讯作者：
Xiaojing Jia;June‐Yi Lee;Hai Lin;A. Alessandri;K. Ha
中科院分区：
文献类型：
--
作者：
Xiaojing Jia;June‐Yi Lee;Hai Lin;A. Alessandri;K. Ha
利用6个大气-海洋耦合气候模式的观测资料和1个月的超前后发资料，研究了1970年代末前后厄尔尼诺Niño和南方涛动（ENSO）的变化对大气环流超前最大协方差分析模式（MCA1）的年代际变化响应。利用从1960-1980年（P1）和1981-2001年（P2） 11月1日开始的12 - 1 - 2月预报，重点研究了北太平洋-北美（NPNA）地区冬季气候的变率和可预测性。观测分析表明，从P1到P2， ENSO变率、相关的热带对流活动以及MCA1均显著增强。结果表明，NPNA表层气候异常在P2比P1与MCA1的相关性更显著，特别是在北美。6个耦合模式及其多模式集合不仅能够捕获NPNA地区MCA1的年代际变化及其与地面气温和降水的关系，而且对P2的MCA1和地面气候异常的预测能力也显著高于P1。然而，模型在MCA1的空间分布上存在系统偏差。结果表明，MCA1的年代际变化有助于提高NPNA近世气候的预报能力。
Using observations and 1-month lead hindcast data from six coupled atmosphere–ocean climate models, this study investigates the interdecadal change in the leading maximum covariance analysis mode (MCA1) of atmospheric circulation in response to the changes in the El Niño and Southern Oscillation (ENSO) occurred around late 1970s. We focus on boreal winter climate variability and predictability over the North Pacific–North American (NPNA) region using December–January–February prediction initiated from November 1st in the period of 1960–1980 (P1) and 1981–2001 (P2). Observed analysis reveals that ENSO variability, the related tropical convective activity, and thus the MCA1 are considerably enhanced from P1 to P2. As a result, surface climate anomalies over the NPNA are more significantly correlated with the MCA1 in P2 than P1, particularly over North America. The six coupled models and their multi-model ensemble not only are capable of capturing the interdecadal change of the MCA1 and its relationship with surface air temperature and precipitation over the NPNA regions but also have significantly higher forecast skills for the MCA1 and the surface climate anomalies in P2 than P1. However, models have systematic biases in the spatial distribution of the MCA1. It is demonstrated that the interdecadal change in the MCA1 should contribute to the improved forecast skill of the NPNA climate during recent epoch.