Collaborative Research: Dynamics of Seasonal Forecast Uncertainty: Cross-Basin Ocean-Atmosphere Interactions

每隔几年，太平洋沿着赤道的一条狭长地带就会急剧变暖，这条狭长地带大致从南美海岸延伸到日期变更线。 这种变暖被称为厄尔尼诺事件，厄尔尼诺和它们的冷拉尼娜对立统称为厄尔尼诺/南方涛动（ENSO）事件。 虽然厄尔尼诺/南方涛动事件大致相似，但也有重要的差异，特别是在东太平洋或中太平洋海表温度变化最大方面。 它们的强度也不同，厄尔尼诺事件往往比拉尼娜事件更强。 厄尔尼诺/南方涛动的后果在全世界都能感受到，从印度季风的变化到加拿大冬季的严寒，这些影响因事件的模式和规模而异。 ENSO事件的动力学机制及其全球影响的多样性是一个重要的实际问题，也是一个科学感兴趣的话题。本项目研究的观点是，ENSO事件的多样性主要是由于发展中的ENSO事件与全球海洋上发生的各种不太突出的气候变率模式之间的相互作用。 例如，太平洋赤道模态（PMM）是一种变化模式，其中北太平洋阿留申低压的波动产生温暖的SST（或相反相位的冷SST），通过海气相互作用缓慢地向赤道太平洋传播。 PMM和ENSO之间的重叠域使得PMM可能是ENSO事件多样化的嫌疑人。 同样，印度洋偶极子（IOD）模式中印度洋的阶段性变暖影响了西太平洋的信风，鉴于信风波动在ENSO演变中的核心作用，对ENSO有潜在的影响。为了研究次级模式对ENSO多样性的影响，该奖项的主要研究人员（PI）利用了用于预测ENSO事件的集合方法。 ENSO预测是指使用气候模式，根据观测到的初始条件（大气和海洋在给定时间的状态）预测ENSO，但执行多个预测模拟而不是一个，并以略有不同的初始条件开始每个预测模拟。 由此产生的扰动初始条件包络（PICE）给出了ENSO演变的最佳估计，并估计了预测中的不确定性。 这个项目的想法是，在一个PICE中的个人预测模拟包含不同的次级模式，和次级模式和ENSO之间的相互作用可以通过检查模拟之间的ENSO演变的差异进行评估。 这种方法的一个优点是，一个大型的数据库的PICE模拟已经创建的地球系统预测（ESP）工作组的社区地球系统模型（CESM）。 PICE数据集提供了比观测记录大得多的样本量，因此可以获得统计上可靠的结果。这项工作具有社会重要性，因为它与ENSO预测直接相关。 除了在预测模拟中审查厄尔尼诺/南方涛动的演变外，研究还利用PICE模拟来了解厄尔尼诺/南方涛动事件之间的差异如何导致厄尔尼诺/南方涛动对世界人口稠密地区影响的差异。 该项目还支持两名研究生，并为本科生提供实习机会。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

Every few years the Pacific ocean warms dramatically in a narrow strip along the equator extending roughly from the South American coast to the dateline. This warming is called an El Nino event, and El Ninos and their cold La Nina opposites are referred to collectively as El Nino/Southern Oscillation (ENSO) events. While ENSO events are broadly similar there are important differences, in particular they differ in whether the sea surface temperature (SST) change is greatest in the eastern or central Pacific. They also differ in their magnitude, and El Nino events are often stronger than La Nina events. The consequences of ENSO are felt worldwide, from changes in the Indian monsoon to the severity of winters in Canada, and these impacts vary according to the pattern and magnitude of the event. The dynamical mechanisms that cause diversity in behavior among ENSO events and their global impacts are thus an important practical problem as well as a topic of scientific interest.This project examines the idea that much of the diversity of ENSO events occurs because of interactions between the developing ENSO event and a variety of less prominent climate variability modes occurring over the global oceans. For example the Pacific meridional mode (PMM) is a variability pattern in which fluctuations of the Aleutian Low over the North Pacific generate warm SSTs (or cold, in the opposite phase) which propagate slowly toward the equatorial Pacific through air-sea interactions. The overlap between the domains of the PMM and ENSO makes the PMM a likely suspect in diversifying ENSO events. Likewise, the episodic warming of the Indian Ocean in the Indian Ocean Dipole (IOD) mode affects the trade winds in the western Pacific, with potential consequences for ENSO given the central role of trade wind fluctuations in ENSO evolution.To examine the effect of secondary modes on ENSO diversity the Principal Investigators (PIs) of this award take advantage of the ensemble method used to predict ENSO events. Ensemble prediction means using a climate model to predict ENSO based on observed initial conditions (the state of the atmosphere and ocean at a given time) but performing several forecast simulations instead of one, and starting each forecast simulation with slightly different initial conditions. The resulting Perturbed Initial Condition Ensemble (PICE) gives a best estimate of the evolution of ENSO and also estimates the uncertainty in the prediction. The idea of this project is that the individual forecast simulations in a PICE contain different secondary modes, and the interactions between secondary modes and ENSO can be assessed by examining differences in ENSO evolution among the simulations. An advantage of this method is that a large database of PICE simulations has been created by the Earth System Prediction (ESP) Working Group of the Communtiy Earth System Model (CESM). The PICE dataset provides a much larger sample size than the observational record, thus statistically robust results can be obtained.The work is of societal importance given its direct connection to ENSO prediction. In addition to its examination of ENSO evolution in prediction simulations the research uses the PICE simulations to understand how differences among ENSO events lead to differences in the impacts of ENSO in populous parts of the world. The project also supports two graduate students and provides internship opportunities for undergraduates.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.