1 南京信息工程大学大气科学学院,南京 210044
2 南京信息工程大学气象灾害预报预警与评估协同创新中心/气象灾害教育部重点实验室/气候与环境变化国际合作联合实验室,南京 210044
3 中国气象局国家气候中心,北京100081
1 School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, China
2 Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/Key Laboratory of Meteorological Disaster, Ministry of Education/Joint International Research Laboratory of Climate and Environment Change, Nanjing University of Information Science and Technology, Nanjing 210044, China
3 National Climate Center, China Meteorological Administration, Beijing 100081, China
作者简介 About authors
基于区域气候模式RegCM4对4个全球气候模式动力降尺度模拟（分别记为CdR、EdR、HdR、MdR）以及高分辨率格点观测数据CN05.1的日降水数据,利用“追踪式”客观识别方法,对1981—2005年中国区域性暴雨事件进行了识别,并评估了模式对其气候特征的模拟性能。结果表明：4个动力降尺度模拟以及多模式集合能较好地模拟区域性暴雨事件发生频次、平均持续时间、平均降水量、平均影响范围和综合强度的年内分布特征以及气候平均值。观测的区域性暴雨事件持续时间、平均降水量、平均影响范围和综合强度在不同区间的频率分布特征以及区域性暴雨事件的累计频次、累计持续时间和累计降水量的空间分布特征也能得到很好地再现。模拟值与观测的空间相关系数都在0.9以上,且均方根误差不超过0.4。不过,相对而言,模式模拟的区域性暴雨事件频次略少,主要由对中度区域性暴雨事件低估所致;模拟的平均持续时间和平均降水量略偏高,而平均影响范围略偏小。综合强度方面,除HdR外,其余模拟均有所高估,尤其是MdR。在频率分布特征和空间分布方面,CdR的模拟性能低于其他模拟。多模式集合模拟的平均持续时间、平均降水量、平均影响范围和综合强度的相对误差分别为13%、2%、-11%和3%。关键词：区域性暴雨事件;动力降尺度;模拟评估;气候特征
基于区域气候模式RegCM4对4个全球气候模式动力降尺度模拟（分别记为CdR、EdR、HdR、MdR）以及高分辨率格点观测数据CN05.1的日降水数据,利用“追踪式”客观识别方法,对1981—2005年中国区域性暴雨事件进行了识别,并评估了模式对其气候特征的模拟性能。结果表明：4个动力降尺度模拟以及多模式集合能较好地模拟区域性暴雨事件发生频次、平均持续时间、平均降水量、平均影响范围和综合强度的年内分布特征以及气候平均值。观测的区域性暴雨事件持续时间、平均降水量、平均影响范围和综合强度在不同区间的频率分布特征以及区域性暴雨事件的累计频次、累计持续时间和累计降水量的空间分布特征也能得到很好地再现。模拟值与观测的空间相关系数都在0.9以上,且均方根误差不超过0.4。不过,相对而言,模式模拟的区域性暴雨事件频次略少,主要由对中度区域性暴雨事件低估所致;模拟的平均持续时间和平均降水量略偏高,而平均影响范围略偏小。综合强度方面,除HdR外,其余模拟均有所高估,尤其是MdR。在频率分布特征和空间分布方面,CdR的模拟性能低于其他模拟。多模式集合模拟的平均持续时间、平均降水量、平均影响范围和综合强度的相对误差分别为13%、2%、-11%和3%。
关键词：区域性暴雨事件;动力降尺度;模拟评估;气候特征
Based on the Regional Climate Model Version 4 (RegCM4) simulations (named CdR, EdR, HdR, and MdR, respectively) dynamically downscaling from four global climate models (GCMs) as well as the observed high-resolution grid dataset CN05.1, the regional rainstorm events that occurred in China during 1981-2005 were identified by using a “tracing” objective method. On this basis, the fidelity of the RegCM4 downscaling in simulating the climate features of regional rainstorm events in China was evaluated. The results show that the four RegCM4 simulations and their ensemble can reasonably capture the observed annual cycles and climate mean values of the frequency, duration, rainfall amount, extent, and comprehensive intensity of regional rainstorm events averaged in China. They can also well reproduce the probability of duration, rainfall amount, extent and comprehensive intensity of regional rainstorm events in different bands and the climatologically spatial distribution of accumulative frequency, duration and rainfall amount of regional rainstorm events in the observation. The spatial correlations of the simulations with the observation are all above 0.9 and the root-mean-square errors are generally below 0.4. However, the simulations slightly underestimate the frequency of regional rainstorm events, mainly due to the underestimation of moderate regional rainstorm events. The average duration and average rainfall amount are slightly overestimated, while the average extent is slightly underestimated. The comprehensive intensity is overestimated by the simulations (especially by MdR) with the exception of HdR. For spatial distribution, the performance of CdR is relatively lower than that of other simulations. The relative errors of the ensemble simulation with reference to the observation are 13%, 2%, -11%, and 3% for the average duration, rainfall amount, extent, and comprehensive intensity of regional rainstorm events, respectively.Keywords：Regional rainstorm event;Dynamical downscaling;Model evaluation;Climate feature
Based on the Regional Climate Model Version 4 (RegCM4) simulations (named CdR, EdR, HdR, and MdR, respectively) dynamically downscaling from four global climate models (GCMs) as well as the observed high-resolution grid dataset CN05.1, the regional rainstorm events that occurred in China during 1981-2005 were identified by using a “tracing” objective method. On this basis, the fidelity of the RegCM4 downscaling in simulating the climate features of regional rainstorm events in China was evaluated. The results show that the four RegCM4 simulations and their ensemble can reasonably capture the observed annual cycles and climate mean values of the frequency, duration, rainfall amount, extent, and comprehensive intensity of regional rainstorm events averaged in China. They can also well reproduce the probability of duration, rainfall amount, extent and comprehensive intensity of regional rainstorm events in different bands and the climatologically spatial distribution of accumulative frequency, duration and rainfall amount of regional rainstorm events in the observation. The spatial correlations of the simulations with the observation are all above 0.9 and the root-mean-square errors are generally below 0.4. However, the simulations slightly underestimate the frequency of regional rainstorm events, mainly due to the underestimation of moderate regional rainstorm events. The average duration and average rainfall amount are slightly overestimated, while the average extent is slightly underestimated. The comprehensive intensity is overestimated by the simulations (especially by MdR) with the exception of HdR. For spatial distribution, the performance of CdR is relatively lower than that of other simulations. The relative errors of the ensemble simulation with reference to the observation are 13%, 2%, -11%, and 3% for the average duration, rainfall amount, extent, and comprehensive intensity of regional rainstorm events, respectively.
Keywords：Regional rainstorm event;Dynamical downscaling;Model evaluation;Climate feature
本文引用格式
其中：n为计算频率分布时划分的总区间数,Xo,i和Xs,i分别代表观测和模拟中某一指标在第i个区间的发生频率。S值越接近1,表示模拟效果越好。
表11981—2005年平均的中国区域性暴雨事件发生频次以及不同等级事件发生频次占总频次的百分比
Table 1 Frequency of regional rainstorm events in China averaged from 1981 to 2005 and the percentages of different category of regional rainstorm events to total events
表21981—2005年中国区域性暴雨事件指标的气候平均值
Table 2 Climate mean of the metrics for regional rainstorm events in China averaged from 1981 to 2005
注：1)影响范围用格点数来表示。
图11981—2005年观测和模拟的各月区域性暴雨事件频次(a)、持续时间(b)、平均降水量(c)、平均影响范围(d)和综合强度(e)的占比分布
Fig. 1Monthly distribution of the percentage of frequency (a), duration (b), average rainfall amount (c), average extent (d), and average comprehensive intensity (e) of regional rainstorm events during 1981-2005
图21981—2005年模拟和观测的中国区域性暴雨事件持续时间(a)、平均降水量(b)、平均影响范围(c)和综合强度(d)的频率占比分布
Fig. 2Frequency percentage distributions for duration (a), average rainfall amount (b), average extent (c), and average comprehensive intensity (d) of regional rainstorm events during 1981-2005
表31981—2005年模式模拟的中国区域性暴雨事件各指标频率分布的S评分统计
Table 3 S score of the frequency percentage distribution for each metric of regional rainstorm events in China during 1981-2005
图3观测(a)和MME模拟(b)的1981—2005年平均的区域性暴雨事件年累计频次空间分布以及MME模拟的相对误差(c)
注：(c)图左下角值为相对误差绝对值<0.4的面积比例,下同。
Fig. 3Observed (a) and MME simulated (b) climatological distribution of the accumulative frequency of regional rainstorm events during 1981-2005 and the relative errors of the MME simulation (c) with reference to the observation (the number in the Fig. (c) indicates the proportion of areas where the absolute relative error is less than 0.4)
图4同图3,但为区域性暴雨事件年累计持续时间
Fig. 4Same as Fig. 3, but for the accumulative duration of regional rainstorm events
图5同图3,但为区域性暴雨事件年累计降水量
Fig. 5Same as Fig. 3, but for the accumulative rainfall amount of regional rainstorm events
图61981—2005年平均的中国区域暴雨事件累计频次、持续时间和降水量的泰勒分布图
注：1, MME;2, CdR;3, EdR;4, HdR;5, MdR。
Fig. 6Taylor diagram for accumulative frequency, duration and rainfall amount of regional rainstorm events in China during 1981-2005 with reference to the observation
本文基于区域气候模式RegCM4对4个全球气候模式降尺度模拟输出的逐日降水结果,利用“追踪式”客观识别方法,对1981—2005年中国区域性暴雨事件进行了识别。在此基础上,通过与高分辨率CN05.1降水格点数据识别出的区域性暴雨事件进行对比,评估了模式对中国区域性暴雨事件的模拟性能。总体来讲,RegCM4动力降尺度模式能够较好地模拟出中国区域性暴雨事件的气候态分布特征,这为下一步利用其开展区域性暴雨事件预估研究奠定了基础。具体结论如下。
(1) MME和4个单模式模拟的中国区域性暴雨事件的发生频次、持续时间、降水量、影响范围和综合强度的气候平均值与观测比较吻合。不过,相对来讲,模式模拟的区域性暴雨事件频次略偏少,主要因低估了中度区域性暴雨事件所致;模拟的平均持续时间和平均降水量略偏多,而平均影响范围略偏小;综合强度方面,除HdR外,其余模式均有所高估,尤其是MdR。
(2) MME和4个单模式能较好地模拟出区域性暴雨事件的发生频次、平均持续时间、平均降水量、平均影响范围和综合强度的逐月分布特征。观测的区域性暴雨事件持续时间、平均降水量、平均影响范围和综合强度在不同区间的频率分布特征也能得到很好的模拟,除CdR模拟的平均降水量分布的S评分为0.78外,其余模拟的S评分均>0.85。
(3) MME和4个单模式能很好地模拟多年平均的区域性暴雨事件年累计频次、累计持续时间和累计降水量的空间分布特征。模拟与观测的空间相关系数都在0.9以上,且均方根误差不超过0.4。相比而言,CdR的模拟能力略低于其他模式。
Observed trends in extreme precipitation events in China during 1961-2001 and the associated changes in large-scale circulation
Trends in total precipitation and frequency of daily precipitation extremes over China
Trend of extreme precipitation events over China in last 40 years
Changes in temperature and precipitation extreme indices over China: analysis of a high-resolution grid dataset
Projected change in extreme rainfall events in China by the end of the 21st century using CMIP5 models
Projected changes in temperature and precipitation extremes in China by the CMIP5 multimodel ensembles
Comparison of CMIP6 and CMIP5 models in simulating climate extremes
Projected risk of flooding disaster in China based on CMIP5 models
Guidelines on analysis of extremes in a changing climate in support of informed decisions for adaptation
An objective identification technique for regional extreme events
Changes in regional heavy rainfall events in China during 1961-2012
The characteristics of regional heavy precipitation events over eastern monsoon China during 1960-2013
中国区域性极端降水事件及人口经济暴露度研究
A study on regional extreme precipitation events and the exposure of population and economy in China
1961—2016年我国区域暴雨过程的客观识别及其气候特征
Objective identification and climatic characters of the regional rainstorm event in China from 1961 to 2016
Evaluation of a high-resolution historical simulation over China: climatology and extremes
Performance of RegCM4 over major river basins in China
Near future (2016-40) summer precipitation changes over China as projected by a regional climate model (RCM) under the RCP8.5 emissions scenario: comparison between RCM downscaling and the driving GCM
Improved performance of high-resolution atmospheric models in simulating the East-Asian summer monsoon rainbelt
RegCM4: model description and preliminary tests over multiple CORDEX domains
Comparison of convective parameterizations in RegCM4 experiments over China with CLM as the land surface model
Future changes in thermal comfort conditions over China based on multi-RegCM4 simulations
Projected changes in haze pollution potential in China: an ensemble of regional climate model simulations
一套格点化的中国区域逐日观测资料及与其它资料的对比
A gridded daily observation dataset over China region and comparison with the other datasets
Bias correction of temperature and precipitation over China for RCM simulations using the QM and QDM methods
Temporal-spatial monitoring of extreme precipitation event: determining simultaneously the time period it lasts and the geographic region it affects
Evaluation of the AR4 climate models’ simulated daily maximum temperature, minimum temperature, and precipitation over Australia using probability density functions
Summarizing multiple aspects of model performance in a single diagram
East Asian monsoon change for the 21st century: results of CMIP3 and CMIP5 models
Reliability of climate models for China through the IPCC Third to Fifth assessment reports
A diagram for evaluating multiple aspects of model performance in simulating vector fields
Present day bias and future change signal of temperature over China in a series of multi-GCM driven RCM simulations
Multimodel ensemble projection of precipitation in Eastern China under A1B emission scenario
Evaluation of global climate models in simulating extreme precipitation in China
Analyses of extreme climate events over China based on CMIP5 historical and future simulations
Development of climate and Earth System Models in China: past achievements and new CMIP6 results