IOCAS-IR  > 海洋环流与波动重点实验室
气候模式中热带海洋系统性偏差的归因及数值模拟试验
张秋实
Subtype博士
Thesis Advisor张荣华
2024-05-16
Degree Grantor中国科学院大学
Place of Conferral中国科学院海洋研究所
Degree Name理学博士
Degree Discipline物理海洋学
Keyword气候模式 模式偏差 次表层温盐场 叶绿素 ENSO 海洋生物引发的加热及反馈
Abstract

气候模式偏差的存在极大地限制了其对于气候模拟、预测和未来预估的能力。因此,识别和减小模式中的系统性偏差一直是气候研究中的重要课题。鉴于海表面温度(Sea Surface Temperature, SST)在海气相互作用中扮演的重要角色,前人的许多研究关注于SST偏差的起源及对气候模拟的影响,而对于海洋次表层的模拟情况研究较少。目前,越来越多的研究表明,次表层海洋在气候变率和可预报性中发挥着重要的作用。然而,在当前的气候模式中,次表层海温偏差的强度甚至超过SST偏差。因此,研究相关海洋次表层系统性偏差的起源及影响十分必要。除此之外,海洋生物过程对于气候模拟同样具有重要影响,但当前的模式对于海洋生物过程的表征仍存在较大偏差,并且海洋生物过程对于气候模式中厄尔尼诺-南方涛动(El Niño-Southern Oscillation, ENSO)模拟的影响仍存在争议。鉴于此,本文使用国际耦合模式比较计划(Coupled Model Intercomparison Project, CMIP)数据识别了当前气候模式中次表层系统性偏差的特征,进一步使用地球系统模式(CESM2)进行敏感性试验来探究相关偏差的起源及影响;分析了CESM2对于海表叶绿素的模拟偏差,并构建了叶绿素统计模型以更好地表征叶绿素的模拟,进一步使用CESM2开展数值试验探究了海洋生物过程对于ENSO模拟的影响。具体的结论如下所示:
(1)评估了最新的海洋环流模式(Ocean Model Intercomparison Project, OMIP)对于热带大西洋次表层海温的模拟情况。显著的次表层暖偏差存在于热带大西洋100-150 m深度范围内,并在赤道外存在两个暖偏差中心。相较于南侧的次表层暖偏差,北侧的暖偏差强度更强,中心深度更深。这些次表层暖偏差全年明显存在,并且出现在几乎所有的OMIP模拟中。通过进行单独海洋试验,本文从大气和海洋两个方面探究了该偏差产生的原因。从大气的角度来看,10°N附近的暖偏差可以部分地归因于海洋模式中给定风场强迫的不确定性。从海洋的角度来看,通过减小海洋模式垂向混合参数化方案中的背景混合系数,北侧的暖偏差明显减小,南侧的暖偏差几乎完全消失。热收支分析表明,由内区垂向混合减小引起的冷却作用加强是次表层暖偏差减小的重要原因。除此之外,赤道大西洋次表层暖偏差减小约50%。随着垂向混合强度的减小,北热带大西洋浅层经圈翻转环流的强度减弱,使得北侧暖偏差向赤道的输运减少,进而引起赤道大西洋次表层海温模拟的改善。
(2)评估了气候模式对于南热带太平洋次表层温盐场的模拟情况。在最新的CMIP6模拟结果中,次表层冷偏差出现在南热带太平洋100-300 m深度范围内,平均温度偏差为-4 °C,最强温度偏差可以达到-10 °C。该冷偏差的出现还伴随着淡的盐度偏差,平均盐度偏差为-0.9 psu,最强盐度偏差可以达到-1.9 psu。相似的次表层温盐偏差在CMIP5模拟中同样显著存在。本文进一步探究了相关偏差产生的原因。回归分析和基于POP2的敏感性试验结果表明,南热带太平洋次表层温盐偏差可以归因于气候模式对于风应力旋度和降水的模拟偏差。通过进行基于CESM2的耦合试验,本文发现耦合模式中风应力旋度和降水偏差的产生与东南热带太平洋SST暖偏差密切相关。东南热带太平洋SST暖偏差可以加强大气对流活动,进而引起低层大气辐合和局地降水增加,从而产生负的风应力旋度偏差和正的降水偏差,之后通过以下过程影响南热带太平洋次表层的模拟:一方面,负的风应力旋度偏差可以加强局地的Ekman抽吸,进而引起次表层冷而淡的偏差;另一方面,由正的降水偏差引起的更淡的海表面盐度偏差信号逐渐向副热带海区延伸,然后该信号通过等密度面过程向赤道次表层传播,进而对南热带太平洋次表层冷而淡的偏差有所贡献。此外,本文发现,南热带太平洋冷而淡的次表层偏差使得纬向等密度面更加平坦,并且减弱了热带-副热带之间的水体交换。同时,该次表层偏差信号可以通过热带太平洋浅层经圈翻转环流向赤道传播,进而引起赤道次表层变冷和温跃层的整体抬升。
(3)构建了叶绿素经验统计模型,并探究了叶绿素年际变率对ENSO模拟的影响。海洋生物场对于ENSO具有一致性的响应特征,同时还可以通过影响太阳短波辐射在上层海洋中的传输对ENSO产生反馈作用。因此,准确地模拟海洋生物场具有重要意义。然而,无论是叶绿素平均态还是其年际变率,当前的地球系统模式对于海洋生物过程的模拟与观测相比仍存在很大的偏差。因此,本文基于观测数据,提取了SST异常和叶绿素异常之间的统计关系,构建了叶绿素经验统计模型。该模型在给定SST异常之后,能够较为准确地刻画出叶绿素异常的响应特征。基于此,本文进一步设计了敏感性试验探究了叶绿素年际变率对于ENSO的调制作用。在海洋试验中,叶绿素年际变率对于ENSO振幅具有削弱作用。通过分析相关加热项的变化,本文发现叶绿素年际变率直接影响了混合层内吸收的太阳辐射(Qabs),同时也对混合层深度(Hm)具有一定的调节作用。以El Niño事件为例,在Qabs减少和Hm加深的共同作用下,太阳短波辐射引起混合层温度的改变率(Rsr)减小,即通过直接的热力作用减弱了ENSO振幅。在Niño4区,SST差异和Rsr差异正相关系数高达0.93,说明叶绿素变化引起热力项的改变对于SST的调制十分重要。除此之外,本文发现叶绿素年际变率对于El Niño强度的削弱强于La Niña,这会使得ENSO非对称性减弱。这可以进一步归因于叶绿素异常对于El Niño和La Niña事件响应强度的差异,从而导致了Qabs和Hm在ENSO暖位相和冷位相时期改变的强度不同,进而使得太阳短波辐射对于混合层的直接加热率存在非对称性。在耦合试验中,叶绿素年际变率对于ENSO的调制作用与海洋试验结果类似。但由于海气耦合的作用,使得叶绿素年际变率对于ENSO的负反馈作用得以放大,在Niño3.4区ENSO振幅减小高达21%。在耦合试验中,叶绿素年际变率同样引起ENSO非对称性进一步减弱。与海洋试验不同的是,耦合试验中SST异常差异的中心并不出现在赤道中西太平洋,而是东移至赤道东太平洋,并且叶绿素年际变率对于El Niño和La Niña强度的削弱更强。
本文系统地评估了最新的海洋模式和气候模式对于海洋次表层温盐场的模拟能力,并探究了相关次表层偏差的产生原因及影响。此外,本文还分析了海表叶绿素模拟偏差,构建了经验统计模型来提高模式对于叶绿素的模拟能力,并探究了叶绿素年际变率对于ENSO模拟的影响。因此,本研究对于认识海洋和气候模式中相关偏差的产生机制具有重要意义,对于提高气候模式的模拟能力具有重要的科学价值和指导作用。

 

Other Abstract

Model biases severely degrade the credibility of climate predictions and future projections in the state-of-the-art coupled general circulation models (CGCMs), and hence it is critically important to identify and rectify the systematic biases in ocean and climate modeling. Considering the key role played by sea surface temperature (SST) in air-sea coupling, realistic SST simulations in CGCMs have become a topic of great concern in the climate modeling communities, while few studies have focused on the simulations of subsurface ocean state. At present, more and more researches show that the subsurface ocean plays an important role in climate variability and predictability. Currently, the subsurface temperature biases are even more pronounced than SST biases. Therefore, it is necessary to study the causes and effects of systematic subsurface biases. In addition, ocean biological processes also play an important role in climate change, but the representations of ocean biological processes are still insufficient in current models, and the feedback and mechanism of ocean biological processes to the El Niño-Southern Oscillation (ENSO) are still controversial. In this paper, Coupled Model Intercomparison Project (CMIP) products are used to identify and analyze the causes and effects of the systematic subsurface biases in the current climate models; and the chlorophyll biases in CESM2 are analyzed, and a statistical model is constructed to characterize the chlorophyll-induced climate effect, and is further used to investigate the modulation effect of ocean biological processes on ENSO. The specific conclusions are as follows:

(1) The subsurface temperature simulations in the tropical Atlantic (TA) from latest ocean models are evaluated in this paper. It is found that significant subsurface warm biases, characterized by two warm bias patches off the equator, exist in 100-150 m over the entire TA basin. Compared with the southern counterpart, the northern subsurface bias is stronger and deeper. These subsurface biases are persistent throughout the year and remain permanent problems in almost all Ocean Model Intercomparison Project (OMIP) simulations. Two potential origins of the subsurface temperature biases are explored by performing POP2-based sensitivity experiments. From an atmospheric perspective, the subsurface bias near 10°N can be related to the uncertainty in prescribed wind forcing. From an oceanic perspective, by constraining vertical mixing in the ocean interior to match what is observed, the northern subsurface bias is significantly reduced and the southern bias is almost eliminated. By conducting heat budget analysis, it is found that the enhanced cooling effect by vertical mixing is primarily responsible for the alleviation of subsurface warm biases. In addition, the equatorial subsurface warm bias is also decreased by ~50%. Following the decrease in vertical mixing, the strength of shallow overturning circulation in the northern TA is weakened. Consequently, the transport of the northern subsurface warm water to the equator is also weakened, leading to the improvement in equatorial subsurface simulation.

(2) The subsurface thermohaline simulations in the southern tropical Pacific (STP) from latest climate models are evaluated in this paper. There exists pronounced subsurface cold bias in 100-300 m over the STP in CMIP6 simulations with an ensemble mean of about -4 °C and an extreme close to -10 °C. This cold bias is accompanied by a fresh subsurface bias of about -0.9 psu in the ensemble mean (-1.9 psu minima). Similar subsurface thermohaline biases also exist in CMIP5 outputs. The causes of these biases are further investigated. By conducting regression analyses and POP2-based sensitivity experiments, it is found that the subsurface thermohaline biases are attributed to the model deficiencies in simulating wind stress curl and precipitation in the STP. By conducting CESM2-based coupled experiments, a warm SST bias in the southeastern tropical Pacific is found to be responsible for the poor simulations in wind stress curl and precipitation. The warm SST bias in the southeastern tropical Pacific acts to strengthen atmospheric convective activity, which induces low-level wind convergence and increases rainfall locally, leading to the negative wind stress curl and excessive precipitation in the STP. Then, the subsurface thermohaline simulations in the STP are affected by the following processes: on the one hand, the negative wind stress curl bias can strengthen the local Ekman pumping, and cause the colder and fresher seawater to be transported to the subsurface; on the other hand, the fresher sea surface salinity bias caused by the positive precipitation bias gradually extends to the subtropical region, and then the subtropical fresh signal propagates to the equator through isopycnal process, contributing to the subsurface cold and fresh biases in the STP. The consequences of these biases are also analyzed. The subsurface thermohaline biases cause density field to increase substantially along 10°S, flattening zonal isopycnal surface and reducing equatorward interior transport. In addition, the anomalously cold and fresh subsurface signals in the STP are seen to propagate to the equator, leading to an overall spurious cooling and an uplift of the thermocline in the equatorial subsurface.

(3) A statistical model for chlorophyll is constructed and the modulation effect of interannual chlorophyll variability on ENSO is investigated. Large interannual variations in ocean biological fields have been observed in the equatorial Pacific during ENSO cycles, and the existence and variability of ocean chlorophyll can modulate ENSO by changing the vertical distribution of penetrative solar radiation in the upper ocean. Therefore, it is of great significance to simulate ocean biological fields accurately. However, whether for the chlorophyll mean state or its interannual variability, the current earth system models still have large biases compared with observations. Therefore, based on the statistical relationship between SST anomalies and chlorophyll anomalies in the observations, an empirical statistical model for chlorophyll is constructed in this paper. After the SST anomalies are given, this statistical model can characterize the response of chlorophyll anomalies accurately. Based on this, this paper further designs sensitivity experiments to investigate the modulation effect of interannual chlorophyll variability on ENSO. In ocean-only experiments, the interannual chlorophyll variability tends to weaken the ENSO amplitude. By analyzing the changes in the relevant heating terms, this paper finds that the interannual chlorophyll variability directly affects the absorbed solar radiation within the mixed layer (Qabs), and also modulates the mixed layer depth (Hm). In the case of El Nino events, under the combined effect of Qabs reduction and mixed layer deepening, the temporal rate of change of the mixed layer temperature induced by solar radiation (Rsr) is reduced, i.e., the ENSO amplitude is weakened through direct thermal effect. In the Niño4 region, the positive correlation coefficient between the SST difference and the Rsr difference is 0.93, indicating that the alteration of the heating term induced by the chlorophyll variability is important for modulating ENSO amplitude. In addition, this paper finds that the interannual chlorophyll variability weakens the intensity of El Niño stronger than that of La Niña, which leads to a weakening of the ENSO asymmetry. This can be further attributed to the difference in chlorophyll anomalies in response to El Niño and La Niña events, which leads to different changes of Qabs and Hm during the warm and cold phases of ENSO, and consequently the asymmetry in Rsr. In the coupled experiments, the effect of interannual chlorophyll anomalies on ENSO simulation is similar to that in the ocean-only experiments. However, the negative feedback effect of interannual chlorophyll anomalies on ENSO is amplified by the sea-air coupling, and the ENSO amplitude decreases by up to 21% in the Niño3.4 region. In the coupled experiments, the interannual chlorophyll variability also induces a further weakening of the ENSO asymmetry. Unlike the ocean-only experiments, the center of the SST anomaly difference in the coupled experiments does not appear in the equatorial west-central Pacific but shifts eastward to the equatorial eastern Pacific, and the weakening of the El Niño and La Niña amplitudes is stronger.

This paper systematically evaluates the simulations of the latest ocean and climate models for the ocean subsurface thermohaline fields, and investigates the causes and effects of the associated subsurface biases. In addition, this paper also analyzes the chlorophyll biases, constructs a empirical statistical model to improve the simulations of climate model for chlorophyll, and investigates the modulation effect of interannual chlorophyll variability on ENSO. Therefore, this study is of great significance in understanding the generation mechanism of the relevant biases in ocean and climate models, and has important scientific value and guiding role in improving the simulations of climate models.

Subject Area物理海洋学
MOST Discipline Catalogue海洋科学
Language中文
Table of Contents

第1章 绪论 1

1.1 研究背景及意义 1

1.2 研究现状 2

1.2.1 气候模式中平均态模拟偏差 2

1.2.2 气候模式中气候变率模拟偏差 7

1.3 关键科学问题与主要研究内容 10

第2章 数据、模式和方法 13

2.1 相关数据简介 13

2.1.1 观测和再分析数据简介 13

2.1.2 CMIP数据简介 14

2.2 CESM2地球系统模式介绍 20

2.3 分析方法 23

第3章 热带大西洋次表层海温偏差的成因及影响 25

3.1 引言 25

3.2 数据和模式试验 26

3.3 热带大西洋次表层海温偏差的时空特征 27

3.4 热带大西洋次表层海温偏差的成因 28

3.4.1 大气风场强迫的作用 28

3.4.2 海洋垂向混合的贡献 31

3.5 对于赤道模拟的影响 36

3.6 本章小结 38

第4章 南热带太平洋次表层温盐偏差的成因及影响 41

4.1 引言 41

4.2 数据和模式试验 41

4.3 南热带太平洋次表层温盐偏差的特征 47

4.4 南热带太平洋次表层温盐偏差的成因 49

4.4.1 大气强迫的作用 49

4.4.2 SST偏差对于风应力和降水模拟偏差的贡献 56

4.5 对于赤道次表层模拟的影响 59

4.6 本章小结 60

第5章 热带太平洋海表叶绿素偏差及叶绿素年际变率对ENSO模拟的影响 63

5.1 引言 63

5.2 数据、方法和模式试验 64

5.3 CESM2中海表叶绿素模拟偏差 67

5.4 对叶绿素经验统计模型的评估 68

5.5 单独海洋试验 71

5.5.1 叶绿素年际变率对于ENSO振幅模拟的影响 71

5.5.2 叶绿素年际变率对于ENSO非对称性模拟的影响 74

5.6 耦合试验 78

5.7 讨论与比较 84

5.8 本章小结 86

第6章 全文总结与展望 89

6.1 本文的主要结论 89

6.2 论文的创新点 91

6.3 未来工作展望 92

参考文献 95

致  谢 107

作者简历及攻读学位期间发表的学术论文与其他相关学术成果 109

Document Type学位论文
Identifierhttp://ir.qdio.ac.cn/handle/337002/185244
Collection海洋环流与波动重点实验室
Recommended Citation
GB/T 7714
张秋实. 气候模式中热带海洋系统性偏差的归因及数值模拟试验[D]. 中国科学院海洋研究所. 中国科学院大学,2024.
Files in This Item:
File Name/Size DocType Version Access License
张秋实-气候模式中热带海洋系统性偏差的归(54824KB)学位论文 暂不开放CC BY-NC-SA
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张秋实]'s Articles
Baidu academic
Similar articles in Baidu academic
[张秋实]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张秋实]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.