FGOALS_g1.1极地气候模拟

对中国科学院大气物理研究所大气科学和地球流体力学数值模拟国家重点实验室发展的气候 系统模式FGOALS-g1.1的极地气候模拟现状进行了较为全面的评估。结果表明,FGOALS- g1.1对南北极海冰的主要分布特征、季节变化和年代际变化趋势具有一定的模拟能力。但也 注意到,与观测相比,模式存在以下几方面的问题：(1) 模拟的海冰总面积北极偏多,而南 极偏少。北极,北大西洋海冰全年明显偏多;夏季,西伯利亚沿海海冰偏多,而波弗特海 海冰偏少。南极,威德尔海和罗斯海冬季海冰偏少。南北极海冰边缘都存在异常的较大范 围密集度很小的碎冰区,夏季尤为显著。(2) 海冰流速在南北极海冰边缘和南极大陆沿岸附 近较大。北极,模式没能模拟出波弗特涡流,并且由于模式网格中北极点的处理问题,造成 其附近错误的海冰流场及厚度分布。这些海冰偏差与模式模拟的大气和海洋状况有着密切的 联系。进一步分析表明,FGOALS-g1.1模拟的冰岛低压和南极绕极西风带明显偏弱, 其通过大气环流和海表面风应力影响向极地的热量输送,在很大程度上导致上述的海冰偏差 。此外,耦合模式中大气-海冰-海洋的相互作用可以放大子模式中的偏差。     

FGOALS_gg1.1极地气候模拟

对中国科学院大气物理研究所大气科学和地球流体力学数值模拟国家重点实验室发展的气候系统模式FGOALS_g1.1的极地气候模拟现状进行了较为全面的评估.结果表明,FGOALS_g1.1对南北极海冰的主要分布特征、季节变化和年代际变化趋势具有一定的模拟能力.但也注意到,与观测相比,模式存在以下几方面的问题:(1)模拟的海冰总面积北极偏多,而南极偏少.北极,北大西洋海冰全年明显偏多;夏季,西伯利亚沿海海冰偏多,而波弗特海海冰偏少.南极,威德尔海和罗斯海冬季海冰偏少.南北极海冰边缘都存在异常的较大范围密集度很小的碎冰区,夏季尤为显著.(2)海冰流速在南北极海冰边缘和南极大陆沿岸附近较大.北极,模式没能模拟出波弗特涡流,并且由于模式网格中北极点的处理问题,造成其附近错误的海冰流场及厚度分布.这些海冰偏差与模式模拟的大气和海洋状况有着密切的联系.进一步分析表明,FGOALS_g1.1模拟的冰岛低压和南极绕极西风带明显偏弱,其通过大气环流和海表面风应力影响向极地的热量输送,在很大程度上导致上述的海冰偏差.此外,耦合模式中大气-海冰-海洋的相互作用可以放大子模式中的偏差.     

CMIP5全球气候模式对中国地区降水模拟能力的评估

使用多种观测资料和43个参加耦合模式比较计划第五阶段（CMIP5）的全球气候模式模拟数据,评估分析了全球气候模式对中国地区1980-2005年降水特征的模拟能力。结果表明：多数CMIP5模式能够模拟出中国降水由西北向东南递增的分布特点,这与耦合模式比较计划第三阶段（CMIP3）的模式模拟结果类似,但华南地区降水模拟偏少,西部高原地区降水模拟偏多。模式能够较好地模拟出降水冬弱夏强的季节变化特征,但降水模拟系统性偏多。从EOF分析结果来看,多数CMIP5模式可以再现中国地区年平均降水的时空变化特征,集合平均的表现优于CMIP3。多模式集合在月、季、年时间尺度下模拟的平均值优于大部分单个模式的结果。CMIP5中6个中国模式的模拟能力与其他模式相当,其中FGOALS-g2、BCC-CSM1-1-m的模拟能力相对较好。     

BCC-CSM2-MR模式对北极海冰和气候的模拟及预估

基于第六次耦合模式比较计划(CMIP6),使用新一代全球模式BCC-CSM2-MR的历史试验和未来共享社会经济路径(SSPs)数据,依据Hadley中心的海表面温度和海冰密集度数据及NCEP/NCAR I再分析资料,评估了BCC-CSM2-MR模式对北极海冰及北极气候的模拟能力,并对未来变化进行了预估.结果表明:BCC...     

北极海冰融化影响东亚冬季天气和气候的研究进展以及学术争论焦点问题

北极历来是影响东亚冬季天气、气候的关键区域之一。北极表面增暖要比全球平均快2~3倍,即所谓北极的放大效应。随着全球增暖的持续以及北极海冰的持续融化,北极的生态环境正在发生显著的变化,进而可能对北半球中、低纬度的天气、气候产生影响。本文概述了有关北极海冰融化影响冬季东亚天气、气候的主要研究进展,特别是自2000年以来,北极海冰异常偏少影响东亚冬季气候变率以及极端严寒事件的可能途径、存在的科学问题,以及学术界的争论焦点。秋、冬季节是北极海冰快速形成时期,此时北极海冰对大气环流的影响要强于大气对海冰的影响。近二十年来的研究结果表明,北极海冰异常偏少,不仅影响北冰洋局地的气温和降水变化,而且通过复杂的相互作用和反馈过程,对北半球中、低纬度的天气、气候产生影响。北极海冰通过以下两个可能机制来影响东亚冬季的天气、气候:（1）北极海冰的负反馈机制;（2）由海冰异常偏少引起的平流层-对流层相互作用机制。秋、冬季节北极海冰持续异常偏少,特别是,巴伦支海-喀拉海海冰异常偏少,既可以加强冬季西伯利亚高压（东亚冬季风偏强）,也可以导致冬季风偏弱。导致海冰影响不确定性的部分原因是:（1）夏季北极大气环流状态影响北极海冰异常偏少对冬季大气环流的反馈效果;（2）冬季大气环流对北极海冰异常偏少响应的位置、强度不同造成的。秋、冬季节北极海冰持续异常偏少,在适宜的条件下（例如,前期夏季北极大气环流的热力和动力条件,有利于加强北极海冰偏少对冬季大气的反馈作用）,可以激发出有利于冬季亚洲大陆极端严寒过程的大气环流异常。目前学术界争论焦点主要集中在以下两个方面:（1）关于北极增暖、北极海冰融化对中纬度区域影响的争论;（2）关于1980年代后期以来,冬季欧亚大陆表面气温呈现降温趋势的原因。目前,有关北极海冰融化影响冬季欧亚大陆次季节变化以及极端天气、气候事件的过程和机制,我们认知非常有限,亟需开展深入细致的研究。     

CICE海冰模式中反照率相关参数对北极海冰模拟的影响

国家气候中心气候系统模式BCC_CSM2.0最新耦合了美国Los Alamos国家实验室发展的海冰模式CICE5.0,为试验模式中与反照率相关参数的敏感性及其对模拟结果的影响,提高模式对北极海冰的模拟能力,选取海冰模式中3个主要参数进行了敏感性试验。利用以BCC_CSM2.0耦合框架为基础建立的海冰-海洋耦合模式,选取CORE资料为大气强迫场开展试验,试验的3个参数分别为冰/雪表面反射率、雪粒半径和雪粒半径参考温度。结果表明,参数取值的不同对北极海冰的模拟有显著的影响,优化后的取值组合极大提高了模式的模拟能力,主要表现在:（1）改善了对北极冬季海冰厚度的模拟,海冰厚度增大,与观测资料更为吻合;（2）显著提高了对北极夏季海冰密集度的模拟能力,从而模拟的北极海冰范围年际循环与观测更为一致。参数取值的优化改进了模式对海冰反照率的模拟,进而影响了冰面短波辐射的吸收和海冰表层的融化,最终提高了模式对海冰密集度和厚度的模拟效果。      

BCC_CSM北极海冰模拟性能的改进对东亚冬季气候模拟的影响

国家气候中心气候系统模式(BCC_CSM)将美国Los Alamos国家实验室发展的海冰模式CICE5.0替代原有的海冰模式SIS,形成一个新版本耦合模式,很好地提高了模式对北极海冰和北极气候的模拟能力.在此基础上,本文评估新耦合模式对1985-2014年东亚冬季气候的模拟性能,检验北极海冰模拟性能的改进对东亚冬季气候...     

Comparing the Arctic climate in Chinese and other CMIP6 models

本研究采用来自耦合模式相互比较项目第六阶段(CMIP6)模式,评估和对比了10个中国模式和27个其他国际模式对北极冬季气候的历史模拟性能.本文的主要目的是展现中国模式对北极气候的模拟能力,并了解其在国际上的模拟水平.结果表明,对于气候态的模拟,中国模式在模拟北极温度场和大气场这些气候学方面与其他国际模式相当.而在趋势方面,中国模式同样和其他国际模式都能很好地模拟出北极变暖的特征.此外,与观测到的环流相比,CMIP6多模式集合平均值(MME)并没有显著的正趋势,这可能是因为外部强迫的作用.     

参加CMIP5的四个中国气候模式模拟的东亚冬季风年际变率

本文比较了中国参加“国际耦合模式比较计划”（CMIP5）的四个大气环流模式（即FGOALS-g2、FGOALS-s2、BCC-CSM1-1、BNU-ESM大气模式）在观测海温驱动下,对东亚冬季风（EAWM）气候态和年际变率的模拟能力。结果表明,在气候态上,四个模式均合理再现了EAWM高低层环流系统（包括低层西伯利亚高压（SH）、阿留申低压、异常偏北风、和中高层东亚大槽、西风急流）,其中对2 m气温和500 hPa高度场的模拟技巧最高,四个模式模拟的结果与再分析资料的空间相关系数都达到0.99。在年际变率上,分别对东亚北部地区（30°N～60°N,100°E～140°E）和东亚南部地区（0°～30°N,100°E～140°E）的2 m气温进行经验正交函数分解（EOF）,提取变率主导模态。结果表明,在东亚北部地区,四个模式对2 m气温第一模态（简称“北部型”）的空间分布均有很高的模拟技巧,但只有BNU-ESM能够较好再现其对应的年际变率,其模拟的时间序列与观测的相关系数为0.69。四个模式均能模拟出观测中的3.1 a主导周期,但只有FGOALS-s2和BNU-ESM能模拟出观测中的2.5 a主导周期。在东亚南部地区,模式模拟的前两个主模态共同解释观测中第一模态（简称“南部型”）的特征,其中FGOALS-g2、FGOALS-s2和BNU-ESM的综合模拟技巧较高,但只有BNU-ESM成功再现了观测中2.5 a和3.1 a的主导周期。机理分析表明,FGOALS-g2、FGOALS-s2、BNU-ESM三个模式能合理再现菲律宾海反气旋,同时对南部型有较高的模拟能力,而BCC-CSM1-1则未能有效再现菲律宾海反气旋,使得 BCC-CSM1-1对南部型模拟技巧较低。观测和四个模式模拟的结果一致表现出北极涛动（AO）与北部型PC1呈显著相关,影响大于SH。     

北极夏季无海冰状态时的东亚气候变化数值模拟研究

利用CMIP3模式在IPCC SRES A1B情景下对未来气候的预测结果,得到北极夏季无海冰的一种情况,即"free Arctic".利用相应的海温场和CO2含量驱动全球大气环流模式,模拟北极夏季无海冰时的东亚气候.试验结果表明,夏季北极无海冰时,全球表面气温有不同程度的明显升高,高纬地区升温幅度大干低纬地区,同纬度地...     

参加CMIP5计划的四个中国模式模拟的东亚地区降水结构特征及未来变化

在全球变暖的背景下, 降水特征的改变体现为降水总量和降水结构的变化。由于缺乏较为长期、覆盖范围广的较高分辨率逐日降水资料, 过去对东亚降水的研究多关注其降水量的长期趋势和时空变率, 较少涉及降水结构的变化。本文利用当前最新且分辨率最高、覆盖范围最广的逐日亚洲陆地降水数据集(简称APHRODITE)以及四个中国参加第五次国际耦合模式比较计划(简称CMIP5)的模式(BCC-CSM1-1, BNU-ESM, FGOALS-g2和FGOALS-s2), 研究了东亚地区降水结构的观测特征及四个模式的模拟能力。基于此, 通过分析四个模式的未来预估试验, 探讨东亚地区降水结构在全球变暖背景下的变化。结果表明:整个东亚地区的累积降水量呈现出随着降水强度的增加先增加后减小的分布形态;降水频率则是随着强度的增加显著减小。小雨(中雨)呈现出南少北多(南多北少)的形态;强降水则较多分布在华南沿海以及日本南部地区。长期趋势上, 整个东亚地区大体上呈现小雨和30 mm/d以上的大雨增加, 而中等强度降水减少的变化趋势。四个模式对东亚降水结构的气候态模拟能力较好。BCC-CSM1-1和FGOALS-g2能够合理再现观测中各个强度降水的变化趋势, 而其他模式模拟不出中雨的减小趋势。四个模式的未来预估表明, 在全球变暖的背景下, 东亚地区30 mm/d以上的强降水会增加。且降水强度越大, 增加越明显。以30°N为界, 小雨(中雨)在变暖背景下呈现南部增加北部减少(南部减少北部增加)的变化趋势。     

Water vapor transport and cross-equatorial flow over the Asian-Australia monsoon region simulated by CMIP5 climate models

The summer mean water vapor transport (WVT) and cross-equatorial flow (CEF) over the Asian-Australian monsoon region simulated by 22 coupled atmospheric-oceanic general circulation models (AOGCMs) from the World Climate Research Programme’s Coupled Model Intercomparison Project Phase 5 (CMIP5) were evaluated. Based on climatology of the twentieth-century simulations, most of models have a reasonably realistic representation of summer monsoon WVT characterized by southeast water vapor conveyor belt over the South Indian Ocean and southwest belt from the Arabian Sea to the East Asian. The correlation coefficients between NCEP reanalysis and simulations of BCC-CSM1-1, BNU-ESM, CanESM2, FGOALS-s2, MIROC4h and MPI-ESM-LR are up to 0.8. The simulated CEF depicted by the meridional wind along the equator includes the Somali jet and eastern CEF in low atmosphere and the reverse circulation in upper atmosphere, which were generally consistent with NCEP reanalysis. Multi-model ensemble means (MME) can reproduce more reasonable climatological features in spatial distribution both of WVT and CEF. Ten models with more reasonable WVT simulations were selected for future projection studies, including BCC-CSM1-1, BNU-ESM, CanESM2, CCSM4, FGOALS-s2, FIO-ESM, GFDL-ESM2G, MRIOC5, MPI-ESM-LR and NorESM-1M. Analysis based on the future projection experiments in RCP (Representative Concentration Pathway) 2.6, RCP4.5, RCP6 and RCP8.5 show that the global warming forced by different RCP scenarios will results in enhanced WVT over the Indian area and the west Pacific and weaken WVT in the low latitudes of tropical Indian Ocean.     

Evaluation of the performance of CMIP5 and CMIP6 models in simulating the South Pacific Quadrupole-ENSO relationship

先前的观测研究表明,南太平洋四极子海温模态(SPQ)可以有效地作为ENSO的前兆信号.本文利用20个CMIP6模式及其对应的20个先前的CMIP5模式的工业化前气候模拟试验数据,评估和比较了CMIP6以及CMIP5模式对SPQ与ENSO的关系的模拟能力.结果表明,大多数CMIP5和CMIP6模式可以合理地模拟SPQ的基本特征.与早期的CMIP5模式相比,CMIP6模式能够更加真实地模拟SPQ与ENSO之间的关系.进一步分析表明,CMIP6模式模拟SPQ与ENSO关系的能力提高,是因为CMIP6模式能够更好地模拟出在副热带/热带太平洋上与SPQ相关的表面海气热力耦合过程,以及在赤道太平洋上与SPQ相关的次表层海温的异常相应.     

Sea-ice and its response to CO2 forcing as simulated by global climate models

The simulation of sea-ice in global climate models participating in the Coupled Model Intercomparison Project (CMIP1 and CMIP2) is analyzed. CMIP1 simulations are of the unpertubed control climate whereas in CMIP2, all models have been forced with the same 1% yr^–1 increase in CO₂ concentration, starting from a near equilibrium initial condition. These simulations are not intended as forecasts of climate change, but rather provide a means of evaluating the response of current climate models to the same forcing. The difference in modeled response therefore indicates the range (or uncertainty) in model sensitivity to greenhouse gas and other climatic perturbations. The results illustrate a wide range in the ability of climate models to reproduce contemporary sea-ice extent and thickness; however, the errors are not obviously related to the manner in which sea-ice processes are represented in the models (e.g. the inclusion or neglect of sea-ice motion). The implication is that errors in the ocean and atmosphere components of the climate model are at least as important. There is also a large range in the simulated sea-ice response to CO₂ change, again with no obvious stratification in terms of model attributes. In contrast to results obtained earlier with a particular model, the CMIP ensemble yields rather mixed results in terms of the dependence of high-latitude warming on sea-ice initial conditions. There is an indication that, in the Arctic, models that produce thick ice in their control integration exhibit less warming than those with thin ice. The opposite tendency appears in the Antarctic (albeit with low statistical significance). There is a tendency for models with more extensive ice coverage in the Southern Hemisphere to exhibit greater Antarctic warming. Results for the Arctic indicate the opposite tendency (though with low statistical significance).A list of the CMIP modeling groups is included in the Acknowledgements section.     

Simulation of sea ice in FGOALS-g2: Climatology and late 20th century changes

Sea ice is an important component in the Earth’s climate system. Coupled climate system models are indispensable tools for the study of sea ice, its internal processes, interaction with other components, and projection of future changes. This paper evaluates the simulation of sea ice by the Flexible Global Ocean-Atmosphere-Land System model Grid-point Version 2 (FGOALS-g2), in the fifth phase of the Coupled Model Inter-comparison Project (CMIP5), with a focus on historical experiments and late 20th century simulation. Through analysis, we find that FGOALS-g2 produces reasonable Arctic and Antarctic sea ice climatology and variability. Sea ice spatial distribution and seasonal change characteristics are well captured. The decrease of Arctic sea ice extent in the late 20th century is reproduced in simulations, although the decrease trend is lower compared with observations. Simulated Antarctic sea ice shows a reasonable distribution and seasonal cycle with high accordance to the amplitude of winter-summer changes. Large improvement is achieved as compared with FGOALS-g1.0 in CMIP3. Diagnosis of atmospheric and oceanic forcing on sea ice reveals several shortcomings and major aspects to improve upon in the future: (1) ocean model improvements to remove the artificial island at the North Pole; (2) higher resolution of the atmosphere model for better simulation of important features such as, among others, the Icelandic Low and westerly wind over the Southern Ocean; and (3) ocean model improvements to accurately receive freshwater input from land, and higher resolution for resolving major water channels in the Canadian Arctic Archipelago.     

全球海气耦合模式对中国区域年代际气候变化预测能力的评估

利用4个海气耦合模式对1960～2005年的多年代际回报结果,评估模式对中国区域年代际气候变化(温度和降水)的预测潜力,并初步给出2005～2015年的气候预测结果。与CMIP3多模式集合1960～2000年结果以及观测实况比较的结果表明:融入观测资料进行同化的年代际气候预测模式,对中国区域温度和降水的模拟能力总体好于CMIP3模式。年代际气候预测模式对温度气候场的模拟仍以"冷偏差"为主,但较之CMIP3模式已有显著改进,中国区域平均的冷偏差减少1.3°C;对降水气候场的模拟仍以"湿偏差"为主,但在华南沿海和西北内陆降水的模拟能力优于CMIP3模式,中国区域平均的湿偏差降低了20%。年代际模式和CMIP3模式都能较好地模拟出中国区域尤其是北方20世纪后期的增暖信号;但CMIP3模式对20世纪后期中国东部降水的旱涝结构演变的模拟与观测相反;而年代际气候预测模式未能再现华北偏旱的变化,但能成功地模拟江淮流域和华南沿海的旱涝演变。2005～2015年的10年预测表明中国区域将继续增暖0.3～0.7°C,且增温幅度北方大于南方,增幅中心位于西北内陆和青藏高原;而降水的变化趋势不显著,黄淮地区、西北内陆和青藏高原的降水略有增加,而西南地区降水将减少。但需要指出的是,这种预测的不确定性是相当大的。     

Asian summer monsoon onset in simulations and CMIP5 projections using four Chinese climate models

The reproducibility and future changes of the onset of the Asian summer monsoon were analyzed based on the simulations and projections under the Representative Concentration Pathways(RCP) scenario in which anthropogenic emissions continue to rise throughout the 21 st century(i.e. RCP8.5) by all realizations from four Chinese models that participated in the Coupled Model Intercomparison Project Phase 5(CMIP5). Delayed onset of the monsoon over the Arabian Sea was evident in all simulations for present-day climate, which was associated with a too weak simulation of the low-level Somali jet in May.A consistent advanced onset of the monsoon was found only over the Arabian Sea in the projections, where the advanced onset of the monsoon was accompanied by an increase of rainfall and an anomalous anticyclone over the northern Indian Ocean. In all the models except FGOALS-g2, the enhanced low-level Somali jet transported more water vapor to the Arabian Sea, whereas in FGOALS-g2 the enhanced rainfall was determined more by the increased wind convergence. Furthermore,and again in all models except FGOALS-g2, the equatorial SST warming, with maximum increase over the eastern Pacific,enhanced convection in the central West Pacific and reduced convection over the eastern Indian Ocean and Maritime Continent region, which drove the anomalous anticyclonic circulation over the western Indian Ocean. In contrast, in FGOALS-g2, there was minimal(near-zero) warming of projected SST in the central equatorial Pacific, with decreased convection in the central West Pacific and enhanced convection over the Maritime Continent. The broader-scale differences among the models across the Pacific were related to both the differences in the projected SST pattern and in the present-day simulations.