海洋模式初始化同化方案对IAP近期气候预测系统回报试验技巧的影响

doi:10.11867/j.issn.1001-8166.2017.04.0342

地球科学进展 ›› 2017, Vol. 32 ›› Issue (4): 342 -352. doi: 10.11867/j.issn.1001-8166.2017.04.0342

吴波 ¹(

), 周天军 ^1, ², 孙倩 ³

1. 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,北京 100029
2.中国科学院大学,北京 100049
3.成都信息工程大学大气科学学院,四川成都 610225

收稿日期:2016-10-23 修回日期:2017-01-10 出版日期:2017-04-20
基金资助:
公益性行业(气象)科研专项项目“基于FGOALS-s、CMA和CESM气候系统模式的年代际集合预测系统的建立与研究”(编号:GYHY201506012);国家自然科学基金项目“以大西洋多年代际振荡作为主要预报因子的夏季北半球热带外气候年代际预测研究”(编号:41675089)资助

Impacts of Initialization Schemes of Oceanic States on the Predictive Skills of the IAP Near-Term Climate Prediction System

Bo Wu ¹( ), Tianjun Zhou ^1, ², Qian Sun ³

1.LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029,China
2.University of Chinese Academy of Sciences, Beijing 100049, China
3.College of Atmospheric Science,Chengdu University of Information Technology, Chengdu 610225,China

Received:2016-10-23 Revised:2017-01-10 Online:2017-04-20 Published:2017-04-20
About author:
First author:Wu Bo (1982-), male,Hefei City, Anhui Province, Associate professor, Research areas include climate dynamics and climate modeling.E-mail:wubo@mail.iap.ac.cn
Supported by:
Foundation item:Project supported by the R&D Special Fund for Public Welfare Industry (Meteorology) “Development and research of ensemble decadal climate prediction system based on global climate models FGOALS-s, CMA and CESM” (No.GYHY201506012);The National Natural Science Foundation of China“A study of decadal prediction for extratropical Northern Hemisphere climate during summer:Using Atlantic multidecadal oscillation as a predictor”(No.41675089)

全文 ( 12 ) 下载PDF ( 915 )

基于耦合气候系统模式FGOALS-s2的中国科学院大气物理研究所(IAP)近期气候预测系统(以下简称IAP DecPreS系统)发展了2种初始化方法:第一种采用Incremental Analysis Update(IAU)方案同化格点化的海洋温度和盐度客观分析资料EN3;第二种采用集合最优差值(EnOI)结合IAU(EnOI-IAU)的方案同化原始海洋温度和盐度观测廓线资料EN4。主要目的是比较基于2种初始化方案开展的年代际预测试验的技巧。时间相关系数、均方根技巧评分等指标均表明,基于EnOI-IAU方案的回报试验对与太平洋年代际振荡(PDO)有关的北太平洋海表面温度(SST)的回报技巧显著高于基于IAU方案的回报试验。而对于大西洋多年代际振荡(AMO),EnOI-IAU方案回报试验的技巧低于IAU方案回报试验。AMO存在副极地和热带北大西洋2个活动中心,EnOI-IAU方案在热带中心的技巧仅略低于IAU方案,但是它在热带外区域模拟出了虚假的降温趋势,因此技巧远低于后者。

关键词: 年代际预测, 耦合气候系统模式, 太平洋年代际振荡, 大西洋多年代际振荡

Based on the near-term climate prediction system of the Institute of Atmospheric Physics (hereafter IAP-DecPreS), we developed two distinct initialization schemes for the Coupled Global Climate Models (CGCM), FGOALS-s2. The first scheme used the Incremental Analysis Update (IAU) to assimilate gridded oceanic temperature and salinity data derived from the EN3 dataset. The second scheme used the merge of the ensemble optimal interpolation (EnOI) and IAU scheme (hereafter EnOI-IAU) to assimilate raw observational oceanic temperature and salinity profiles. The predictive skills of the decadal prediction experiments based on the two schemes were compared. Several metrics including temporal correlation and root mean square skills score indicate that the experiment based on the EnOI-IAU shows significantly higher predictive skills in the Sea Surface Temperature (SST) anomalies in the North Pacific associated with the Pacific Decadal Oscillation (PDO), than the experiment based on the IAU. In contrast, for the Atlantic Multi-Decadal Oscillation (AMO), the predictive skills of the experiment based on the EnOI-IAU are lower than that based on the IAU. The AMO has two activity centers, located in the subpolar and tropical North Atlantic. The skills of the experiment based on the EnOI are close to that based on the IAU in the tropical North Atlantic, while much lower than the latter in the extratropical region due to a false simulation of the warming trend in the region.

Key words: Decadal prediction, Coupled GCM, PDO, AMO.

中图分类号:

P467

图1 基于均方根误差的FGOALS-s2 2组年代际预测试验的技巧评估
(a)DP-EnOI-IAU试验回报的年平均SST异常(回报时间段6~9年平均)的RMSSS,白点代表通过5%显著性检验;(b)同(a),但为DP-IAU试验结果;(c)DP-EnOI-IAU与DP-IAU试验RMSE的比较,计算公式:[1-

RMSE (DP - EnOI - IAU) RMSE (DP - IAU)

]×100

Fig.1 Skill evaluations of two decadal prediction experiments by FGOALS-s2 based on root mean square error
(a) RMSSS of the annual mean SST anomalies in the DP-EnOI-IAU experiments for the predictions averaged over the hindcate years 6~9,white dots denote the values reaching the 5% significance level;(b) as in (a), but for the DP-IAU experiments;(c) Comparison of the RMSE between the DP-EnOI-IAU and DP-IAU experiments calculated as[1-

RMSE (DP - EnOI - IAU) RMSE (DP - IAU)

] ×100

RMSE (DP - EnOI - IAU) RMSE (DP - IAU)

]×100

RMSE (DP - EnOI - IAU) RMSE (DP - IAU)

] ×100

图2 基于时间相关系数的FGOALS-s2 2组年代际预测试验的技巧评估
(a)DP-EnOI-IAU试验回报的年平均SST异常(回报时间段6~9年平均)与对应观测的时间相关系数的空间分布,白点代表通过5%显著性检验;(b)同(a),但为DP-IAU试验结果;(c)为(a)与(b)之差;(d)~(f)同(a)~(c),但为去趋势的SST异常

Fig.2 Skill evaluations of two decadal prediction experiments by FGOALS-s2 based on temporal correlation
(a) Spatial distributions of the temporal correlations of the annual mean SST anomalies for the predictions averaged over the hindcate years 6~9 derived from the DP-EnOI-IAU and corresponding observations,white dots denote the values reaching the 5% significance level;(b) as in (a),but for the temporal correlation;(c) Differences between (a) and (b);(d)~(f) as (a)~(c), but for the detrended SST anomalies

图3 全球平均(60°N~60°S)表面温度随时间的演变
红线和蓝线分别为DP-EnOI-IAU和DP-IAU试验结果;阴影代表不同集合成员的离散度;黑线为对应观测的结果;所有点在时间轴上的位置是回报年的第7年,代表回报年6~9年的平均

Fig.3 Time series of the nearly global mean surface temperature (60°N~60°S)
Red and blue lines are derived from the DP-EnOI-IAU and DP-IAU experiments, respectively. Shadings represent the spreads of the ens-emble members;Black line is corresponding observational references.All the points on the lines are marked on the hindcast year 7 and represent the averages over the hindcast years 6~9

图4 观测中提取的太平洋年代际振荡模态
(a)对20°N以北太平洋区域4年滑动平均SST异常做EOF分析得到的第一模态的空间型,左上角括号内为该模态的方差贡献;(b)红线为EOF1对应的标准化的主成分时间序列;黑线为标准化的中纬度北太平洋PDO关键区(30°~42°N,150°E~155°W,(a)中黑框区域)区域平均SST异常

Fig.4 Pacific decadal oscillation mode derived from the observation
(a) The spatial pattern of the first EOF mode of the 4 year-running averaged SST anomalies in the North Pacific to the north of the 20°N;The value at the upper left corner is the variance contribution of the mode;(b) Red line is the normalized principal component time series of the EOF1 mode;Black line is the normalized area-averaged SST anomalies in the 30°~42°N,150°E~155°W (black box in (a))

图5 30°~42°N,150°E~155°W区域平均海表面温度异常( 图4 a中黑框)随时间的演变
红线和蓝线分别为DP-EnOI-IAU和DP-IAU试验结果;阴影代表不同集合成员的离散度;黑线为对应观测的结果;所有点在时间轴上的位置是回报年的第7年,代表回报年6~9年的平均

Fig.5 Time series of the SST anomalies in 30°~42°N,150°E~155°W (black box in Fig.4 a)
Red and blue lines are derived from the DP-EnOI-IAU and DP-IAUexperiments, respectively;Shadings represent the spreads of the ensemble members. Black line is corresponding observational references;All the points on the lines are marked on the hindcast year 7 and represent the averages over the hindcast years 6~9

图6 观测中提取的大西洋多年代际振荡模态
(a)将SST异常向4年滑动平均AMO指数回归,代表AMO的空间型;AMO指数定义为北太平洋区域平均(0°~60°N,80°W~0°)SST异常减去全球平均(60°S~60°N)SST异常;(b)红线为标准化的4年滑动平均AMO指数;黑线为标准化的4年滑动平均的热带北大西洋区域平均SST异常(0°~25°N,80~10°W,(a)中黑框区域);蓝线为标准化的4年滑动平均的副极地北大西洋区域平均SST异常(40°~65°N,70°~30°W,(a)中蓝框区域)

Fig.6 Atlantic multidecadal oscillation mode derived from the observation
(a) SST anomalies regressed against the four years running averaged AMO index. The AMO is defined as the area-averaged SST anomalies in the North Atlantic(0°~60°N, 40°W~0°) minus nearly global mean SST anomalies;(b) Red line is normalized four years running averaged AMO index;Black line is the normalized four years running averaged area-averaged SST anomalies in the tropical Atlantic (0°~25°N, 80°~10°W,black box in (a)); Blue line is the normalized four years running averaged area-averaged SST anomalies in thesubpolar gyre Atlantic (40°~65°N, 70°~30°W, blue box in (a))

图7 大西洋区域预报技巧
(a)0°~25°N,80°~10°W( 图6 a中黑框)区域平均海表面温度异常随时间的演变;(b)40°~65°N,70°~30°W( 图6 a中蓝框)区域平均海表面温度异常随时间的演变;红线和蓝线分别为DP-EnOI-IAU和DP-IAU试验结果;阴影代表不同集合成员的离散度;黑线为对应观测的结果;所有点在时间轴上的位置是回报年的第7年,代表回报年6~9年的平均

Fig.7 Predictive skills in Atlantic
(a)Time series of the SST anomalies in 0°~25°N, 80°~10°W (black box in Fig. 6 a); (b) Time series of the SST anomalies in 40°~65°N,70°~30°W (blue box in Fig. 6 a); Red and blue lines are derived from the DP-EnOI-IAU and DP-IAU experiments, respectively; Shadings represent the spreads of the ensemble members; Black line is corresponding observational references; All the points on the lines aremarked on the hindcast year 7 and represent the averages over the hindcast years 6~9

[1]	Kirtman B, Power S B, Adedoyin J A, et al.Near-term climate change: Projections and predictability[C]∥Climate Change 2013: The Physical Science Basis, Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, 2013.
[2]	Meehl G A, Goddard L, Murphy J, et al.Decadal prediction: Can it be skillful?[J]. Bulletin of the American Meteorological Society, 2009, 90(10):1 467-1 485.
[3]	Meehl G A, Goddard L, Boer G, et al.Decadal climate prediction: An update from the trenches[J]. Bulletin of the American Meteorological Society, 2014, 95(2):243-267.
[4]	Liu Z.Dynamics of interdecadal climate variability: A historical perspective[J]. Journal of Climate, 2012, 25(6):1 963-1 995.
[5]	Doblas-Reyes F J, Andreuburillo I, Chikamoto Y, et al. Initialized near-term regional climate change prediction[J]. Nature Communication, 2013, 4(7):1 078-1 090.
[6]	García-Serrano J, Guemas V, Doblas-Reyes F J. Added-value from initialization in predictions of Atlantic multi-decadal variability[J]. Climate Dynamics, 2015, 44(9/10):25-30.
[7]	Meehl G A, Teng H.Case studies for initialization decadal hindcasts and predictions for the Pacific region[J]. Geophysical Research Letters, 2012, 39(22):20 813-20 821.
[8]	Mochizuki T, Masuda S, Ishikawa Y, et al.Multi-year climate prediction with initialization based on 4D-Var data assimilation[J]. Geophysical Research Letters, 2016, 43(8): 3 903-3 910.
[9]	Taylor K E, Stouffer R J, Meehl G A.An overview of CMIP5 and the experiment design[J]. Bulletin of the American Meteorological Society, 2012, 93(4):485-498.
[10]	Xin X G, Wu T W, Zhang J.Introduction of CMIP5 experiments carried out with the Climate system models of Beijing Climate Center[J]. Advances in Climate Change Research, 2013, 4(1):41-49.
[11]	Gao Feng, Xin Xiaoge, Wu Tongwen.A study of the prediction of regional and global temperature on decadal time scale with BCC_CSM1.1 Model[J]. Chinese Journal of Atmospheric Sciences, 2012, 36(6):1 165-1 179.
	[高峰, 辛晓歌, 吴统文. BCC_CSM1.1对10年尺度全球及区域温度的预测研究[J]. 大气科学, 2012, 36(6):1 165-1 179.]
[12]	Wang B, Liu M, Yu Y, et al.Preliminary evaluations of FGOALS-g2 for decadal predictions[J]. Advances in Atmospheric Sciences, 2013, 30(3):674-683.
[13]	Wu B, Zhou T J.Prediction of decadal variability of sea surface temperature by a coupled global climate model FGOALS_gl developed in LASG/IAP[J]. Chinese Science Bulletin, 2012, 57(19):2 453-2 459.
[14]	Bao Q, Lin P F, Zhou T J, et al.The flexible global Ocean-Atmosphere-Land System model Version: FGOALS-s2[J]. Advances in Atmospheric Sciences, 2013, 30(3):561-576.
[15]	Collins W D, Bitz C M, Blackmon M L, et al.The Community Climate System Model version 3 (CCSM3)[J]. Journal of Climate, 2006, 19(11):2 122-2 143.
[16]	Liu H L, Lin P F, Yu Y Q, et al.The baseline evaluation of LASG/IAP Climate system Ocean Model (LICOM) version 2.0[J]. Journal of Meteorological Research, 2012, 26(3):318-329.
[17]	Lin P F, Yu Y Q, Liu H L.The mean state of oceanic circulation simulated by the Coupled Model FGOALS-s2[J]. Advances in Atmospheric Sciences, 2013, 30(3):175-192.
[18]	Oleson K W, Dai Y J, Bonan G B, et al.Technical Description of the Community Land Model (CLM)[M]∥Colorado: National Center for Boulder, Atmospheric Research,2004:256-265.
[19]	Guinehut S, Coatanoan C, Dhomps A L, et al.On the use of satellite altimeter data in Argo quality control[J]. Journal of Atmospheric and Oceanic Technology, 2009, 26(2):395-402.
[20]	Wu B, Chen X L, Song F F, et al.Initialized decadal predictions by LASG/IAP climate system model FGOALS-s2: Evaluations of strengths and weaknesses[J]. Advances in Meteorology, 2015, (6):1-12.
[21]	Bloom S C.Data assimilation using incremental analysis updates[J]. Monthly Weather Review, 1996, 124(6):1 256.
[22]	Good S A, Martin M J, Rayner N A.EN4: Quality controlled ocean temperature and salinity profiles and monthly objective analyses with uncertainty estimates[J]. Journal of Geophysical Research: Oceans, 2013, 118(12): 6 704-6 716.
[23]	Rayner N A, Parker D E, Horton E B, et al.Global analyses of SST, sea ice and night marine air temperature since the late nineteenth century[J]. Journal of Geophysical Research Atmospheres, 2003, 108(D14):1 063-1 082.
[24]	Evensen G.The ensemble Kalman filter: Theoretical formulation and practical implementation[J]. Ocean Dynamics, 2003, 53(4):343-367.
[25]	Oke P R, Allen J S, Miller R N, et al. Assimilation of surface velocity data into a primitive equation coastal ocean model[J]. Journal of Geophysical Research Oceans, 2002, 107(C9):5-1-5-25.
[26]	Eifuku R, Yoshino I, Pohlmann I S.A comparison of full-field and anomaly initialization for seasonal to decadal climate prediction[J]. Climate Dynamics, 2013, 41(11/12):3 325-3 338.
[27]	CLIVAR. Data and Bias Correction for Decadal Climate Predictions[R].Qingdao: International CLIVAR Project Office, 2011.
[28]	Doblas-Reyes F J, Andreuburillo I, Chikamoto Y, et al. Initialized near-term regional climate change prediction[J]. Nature Communications, 2013, 4(7):1 078-1 090.
[29]	Wilks D S.Statistical methods in the atmospheric sciences[M]∥Statistical Methods in the Atmospheric Sciences. Elsevier, 2011:183-197.
[30]	Morice C P, K ennedy J J, Rayner N A, et al. Quantifying uncertainties in global and regional temperature change using an ensemble of observational estimates: The HadCRUT4 dataset[J]. Journal of Geophysical Research Atmospheres, 2012, 117(D8):393-407.
[31]	Easterling D R, Wehner M F.Is the climate warming or cooling?[J]. Geophysical Research Letters, 2009, 36(8):262-275.
[32]	Kosaka Y, Xie S P.Recent global-warming hiatus tied to equatorial Pacific surface cooling[J]. Nature, 2013, 501(7 467):403-407.
[33]	Newman M, Alexander M A, Ault T R, et al.The Pacific decadal oscillation, Revisited[J]. Journal of Climate, 2016, 29(12):4 399-4 427.
[34]	Mantua N J, Hare S R, Zhang Y, et al.A Pacific interdecadal climate oscillation with impacts on salmon production[J]. Bulletin of the American Meteorological Society, 1997, 78(6):1 069-1 079.
[35]	Delworth T, Zhang L R, Mann M E.Decadal to centennial variability of the Atlantic from observations and models. Ocean circulation: Mechanisms and impacts[J]. Geophysical Monograph Series, 2007, 173:131-148.
[36]	Kilbourne K H, Quinn T M, Webb R, et al.Paleoclimate proxy perspective on Caribbean climate since the year 1751: Evidence of cooler temperatures and multidecadal variability[J]. Paleoceanography, 2008, 23(3):764-767.
[37]	Trenberth K E, Shea D J.Atlantic hurricanes and natural variability in 2005[J]. Geophysical Research Letters, 2006, 33(33):285-293.
[38]	Gastineau G, Frankignoul C.Influence of the North Atlantic SST Variability on the Atmospheric Circulation during the Twentieth Century[J]. Journal of Climate, 2015, 28(4):1 396-1 416.
[39]	Smith D, Eade R, Pohlmann H.A comparison of full-field and anomaly initialization for seasonal to decadal climate prediction[J]. Climate Dynamics, 2013, 40:3 325-3 338.

[1]	夏松, 刘鹏, 江志红, 程军. CMIP5和 CMIP6模式在历史试验下对 AMO和 PDO的模拟评估[J]. 地球科学进展, 2021, 36(1): 58-68.
[2]	满文敏, 周天军. IAP年代际预测试验中火山活动对太平洋海温预测技巧的影响[J]. 地球科学进展, 2017, 32(4): 353-361.
[3]	陈晓龙, 吴波, 周天军. FGOALS-s2海洋同化系统中东亚夏季风和前冬厄尔尼诺—南方涛动关系的年代际变化[J]. 地球科学进展, 2017, 32(4): 362-372.
[4]	郭准, 周天军. IAP近期际气候预测系统海洋初始化试验中海表温度和层积云的关系[J]. 地球科学进展, 2017, 32(4): 373-381.
[5]	容新尧, 刘征宇, 段晚锁. 耦合模式中北太平洋和北大西洋海表面温度年代际可预报性和预报技巧的季节依赖性[J]. 地球科学进展, 2017, 32(4): 382-395.
[6]	韩振宇, 吴波, 辛晓歌. BCC_CSM1.1气候模式对全球海表温度年代际变化的回报能力评估[J]. 地球科学进展, 2017, 32(4): 396-408.
[7]	张丽霞, 张文霞, 周天军, 吴波. ENSEMBLES耦合模式对全球陆地季风区夏季降水的年代际预测能力评估[J]. 地球科学进展, 2017, 32(4): 409-419.
[8]	陈幸荣, 蔡怡, 谭晶, 黄勇勇, 汪雷. 全球变暖hiatus现象的研究进展[J]. 地球科学进展, 2014, 29(8): 947-955.

阅读次数

全文

摘要