Study of evaluation method on the climate of extreme high temperatures based on dynamic return periods

doi:10.12006/j.issn.1673-1719.2022.008

Abstract

Abstract:

For 754 homogenized daily maximum temperature series across rapid warming China during past 40 years, dynamic return periods and return levels were investigated based on various nonstationary Generalized Extreme Value (GEV) models, and new method of evaluation for high temperature extremes was proposed. By this method, the extremality of extreme climate events can be better explained. This method also enhances the comparability between climate extremes, thus more historical information can be well preserved. Transformed applications of dynamic return periods between stationary and nonstationary models can provide more accurate and reliable assessment about current state of climate extremes in climate change studies. Finally, some long-existing disagreements and mistakes in understanding “once-in-N-year” extreme events (i.e., as a value or a probability) between academic circles and public opinions are clarified effectively via the new method.

Key words: Extreme high temperature, Dynamic return periods, Generalized Extreme Values (GEV), Nonstationary models

TU Kai, YAN Zhong-Wei, FAN Li-Jun, LI Zhen. Study of evaluation method on the climate of extreme high temperatures based on dynamic return periods[J]. Climate Change Research, 2023, 19(1): 11-22.

Figures/Tables 11

Fig. 1 The spatial distribution of 754 stations and partition of regions

Fig. 2 GEV stationary fits demos, according to different shape parameters. (a) Return levels plot (scatters are annual maximums, solid lines are fitted values, dashed lines are 95% confidence intervals), (b) corresponding PDF curves

Fig. 3 The most applicable GEV models for 754 stations across China

Table 1 The numbers and percentages of most suitable GEV model for annual maximum temperatures in China

Fig. 4 Demonstrations of dynamic return levels for 4 different nonstationary GEV models. (Thick horizontal lines indicate fitted medium values in stationary GEV models, thin horizontal lines indicate “once-in-10-year” values (i.e. 90% percentiles) in stationary GEV models. Thick/thin dashed lines indicate corresponding values in nonstationary GEV models. Scatters are annual maximums)

Table 2 Comparisons between the years that the annual maximum temperature reached the “once-in-10-year” level based on dynamic and fixed return periods, for representative stations of four nonstationary models

Fig. 5 The average frequency of “once-in-10-year” extreme temperatures in eight regions and China from 1979 to 2018, in which return levels are based on nonstationary GEV models

Fig. 6 The spatial pattern of the yearly frequency of “once-in-10-year” extreme temperatures, displayed in four different ten-year windows during 1979-2018

Fig. 7 The dynamic indices of “once-in-100-year” and “once-in-10-year” extreme temperatures across China during 1979-2018

Fig. 8 The current (2018) return periods in nonstationary GEV models for the “once-in-100-year” return levels in stationary GEV models

Fig. 9 The number of stations that annual maximum of daily maximum temperatures broke the record (column) and the annual mean daily maximum temperatures (curve)

References 32

[1]	Sousa P M, Barriopedro D, García-Herrera R, et al. Distinct influences of large-scale circulation and regional feedbacks in two exceptional 2019 European heatwaves[J]. Communications Earth and Environment, 2020, 1: 48
[2]	严中伟, 丁一汇, 翟盘茂, 等. 近百年中国气候变暖趋势之再评估[J]. 气象学报, 2020, 78 (3): 370-378.
	Yan Z W, Ding Y H, Zhai P M, et al. Re-assessing climatic warming in China since the last century[J]. Acta Meteorologica Sinica, 2020, 78 (3): 370-378 (in Chinese)
[3]	中国气象局气候变化中心. 中国气候变化蓝皮书(2021)[R]. 北京: 科学出版社, 2021.
	Climate Change Center, China Meteorological Administration. Blue book on climate change in China 2021[R]. Beijing: Sciences Press, 2021 (in Chinese)
[4]	李庆祥, 黄嘉佑. 对我国极端高温事件阈值的探讨[J]. 应用气象学报, 2011, 22 (2): 138-144.
	Li Q X, Huang J Y. Threshold values on extreme high temperature events in China[J]. Journal of Applied Meteorological Science, 2011, 22 (2): 138-144 (in Chinese)
[5]	Karl T R, Nicholls N, Ghazi A. CLIVAR/GCOS/WMO workshop on indices and indicators for climate extremes: workshop summary[J]. Climatic Change, 1999, 42: 3-7
[6]	周雅清, 任国玉. 中国大陆1956—2008年极端气温事件变化特征分析[J]. 气候与环境研究, 2010, 15 (4): 405-417.
	Zhou Y Q, Ren G Y. Variation characteristics of extreme temperature indices in mainland China during 1956-2008[J]. Climatic and Environmental Research, 2010, 15 (4): 405-417 (in Chinese)
[7]	翟盘茂, 潘晓华. 中国北方近50年温度和降水极端事件变化[J]. 地理学报, 2003, 58(S): 1-10.
	Zhai P M, Pan X H. Change in extreme temperature and precipitation over northern China during the second half of the 20th century[J]. Acta Geographica Sinica, 2003, 58(S): 1-10 (in Chinese)
[8]	叶殿秀, 尹继福, 陈正洪, 等. 1961—2010年我国夏季高温热浪的时空变化特征[J]. 气候变化研究进展, 2013, 9 (1): 15-20.
	Ye D X, Yin J F, Chen Z H, et al. Spatiotemporal change characteristics of summer heatwaves in China in 1961-2010[J]. Climate Change Research, 2013, 9 (1): 15-20 (in Chinese)
[9]	Parey S, Malek F, Laurent C, et al. Trends and climate evolution: statistical approach for very high temperatures in France[J]. Climatic Change, 2007, 81 (3): 331-352
[10]	Cooley D. Return periods and return levels under climate change[C]//AghaKouchak A, Easterling D, Hsu K, et al. Extremes in a changing climate. Dordrecht: Springer, 2013: 97-114
[11]	李洋, 杨赤. 中国区域气候极值重现水平的非平稳模型及趋势分析[J]. 气候与环境研究, 2015, 20 (3): 347-355.
	Li Y, Yang C. Non-stationary modeling and trend analysis of return levels of climate extremes in China[J]. Climatic and Environmental Research, 2015, 20 (3): 347-355 (in Chinese)
[12]	陈创买, 蒋光达, 周文. 广东年最大日降水量重现期的计算[J]. 人民珠江, 1998, 1: 15-19.
	Chen C M, Jiang G D, Zhou W. Computation of recurrence interval of maximum daily precipitation in the year in Guangdong[J]. Pearl River, 1998, 1: 15-19 (in Chinese)
[13]	吴佳, 高学杰, 石英. 中国20年一遇气温和降水极值变化的高分辨率模拟[J]. 气候变化研究进展, 2012, 8 (4): 243-249.
	Wu J, Gao X J, Shi Y. Changes of 20-year return temperature and precipitation extremes over China simulated by RegCM3[J]. Climate Change Research, 2012, 8 (4): 243-249 (in Chinese)
[14]	Coles S G. An introduction to statistical modeling of extreme values[M]. London: Springer, 2001
[15]	Kharin V, Zwiers F. Estimating extremes in transient climate change simulations[J]. Journal of Climate, 2005, 18 (8): 1156-1173
[16]	Zhang X, Wang J, Zwiers F W, et al. The influence of large-scale climate variability on winter maximum daily precipitation over North America[J]. Journal of Climate, 2010, 23 (11): 2902-2915
[17]	Yee T W, Stephenson A G. Vector generalized linear and additive extreme value models[J]. Extremes, 2007, 10: 1-19
[18]	Yang C, Li Y, Xu J. Bayesian geoadditive modelling of climate extremes with nonparametric spatially varying temporal effects[J]. International Journal of Climatology, 2016, 36: 3975-3987
[19]	Rohmer J, Thieblemont R, Le Cozannet G, et al. Revisiting the link between extreme sea levels and climate variability using a spline-based non-stationary extreme value analysis[J]. Weather and Climate Extremes, 2021, 33: 100352
[20]	Tu K, Yan Z. Non-stationary climate changes in summer high-temperature extremes in Shanghai since the late 19th century[J]. International Journal of Climatology, 2021, 41 (S1): 718-733
[21]	Li Z, Yan Z, Wu H. Updated homogenized Chinese temperature series with physical consistency[J]. Atmospheric and Oceanic Science Letters, 2015, 8 (1): 17-22
[22]	杜勤勤, 张明军, 王圣杰, 等. 中国气温变化对全球变暖停滞的响应[J]. 地理学报, 2018, 73 (9): 1748-1764. doi: 10.11821/dlxb201809011
	Du Q Q, Zhang M J, Wang S J, et al. Changes in air temperature of China in response to global warming hiatus[J]. Acta Geographica Sinica, 2018, 73 (9): 1748-1764 (in Chinese) doi: 10.11821/dlxb201809011
[23]	Yee T W. Vector Generalized Linear and Additive Models: with an implementation in R[M]. New York: Springer, 2015: 447-468
[24]	Katz R W. Statistical methods for nonstationary extremes[C]// AghaKouchak A, Easterling D, Hsu K, et al. Extremes in a changing climate. Dordrecht: Springer, 2013: 15-38
[25]	Olsen J, Lambert J H, Haimes Y. Risk of extreme events under nonstationary conditions[J]. Risk Analysis, 1998, 18: 49-510
[26]	Min S K, Cai W, Whetton P. Influence of climate variability on seasonal extremes over Australia[J]. Journal of Geophysical Research: Atmospheres, 2013, 118: 643-654
[27]	Akaike H. A new look at the statistical model identification[J]. IEEE Transactions on Automatic Control, 1974, 19 (6): 716-723
[28]	中国科学院. 地球大数据支撑可持续发展目标报告(2020)[R/OL]. 2020 [2022-01-17]. http://sdgs.casearth.cn/zw/zy/bg/202009/P020200927321587522760.pdf.
	Chinese Academy of Sciences. Big Earth data in support of the sustainable development goals (2020)[R/OL]. 2020 [2022-01-17]. http://sdgs.casearth.cn/zw/zy/bg/202009/P020200927321587522760.pdf (in Chinese)
[29]	Fan L, Chen D. Trends in extreme precipitation indices across China detected using quantile regression[J]. Atmospheric Science Letters, 2016, 17 (7): 400-406
[30]	周波涛, 钱进. IPCC AR6报告解读:极端天气气候事件变化[J]. 气候变化研究进展, 2021, 17 (6): 713-718.
	Zhou B T, Qian J. Changes of weather and climate extremes in the IPCC AR6[J]. Climate Change Research, 2021, 17 (6): 713-718 (in Chinese)
[31]	Tu K, Yan Z, Dong W. Climatic jumps in precipitation and extremes in drying North China during 1954-2006[J]. Journal of The Meteorological Society of Japan, 2010, 88 (1): 29-42
[32]	Xia J, Tu K, Yan Z, et al. The super-heat wave in eastern China during July-August 2013: a perspective of climate change[J]. International Journal of Climatology, 2016, 36: 1291-1298. DOI: 10.1002/joc.4424 doi: 10.1002/joc.4424 URL