基于Copula函数和随机森林的辽宁省高温干旱复合灾害变化特征

doi:10.13287/j.1001-9332.202605.028

应用生态学报 ›› 2026, Vol. 37 ›› Issue (5): 1595-1604.doi: 10.13287/j.1001-9332.202605.028

基于Copula函数和随机森林的辽宁省高温干旱复合灾害变化特征

刘宇彤^1,2, 刘茂华¹, 陈妮娜^2,3,4*, 米娜^2,3,4, 于文颖^2,3,4, 武晋雯^2,3,4, 冯锐^2,3,4

¹沈阳建筑大学交通与测绘工程学院, 沈阳 110168;
²中国气象局沈阳大气环境研究所, 沈阳 110166;
³沈阳农业与生态气象研究院, 沈阳 110166;
⁴辽宁省农业气象灾害重点实验室, 沈阳 110166

收稿日期:2025-11-28 接受日期:2026-04-01 出版日期:2026-05-18 发布日期:2026-11-18
通讯作者: * E-mail: ninachen.yu@163.com
作者简介:刘宇彤, 女, 2003年生, 硕士研究生。主要从事地理信息系统研究。E-mail: 3443772376@qq.com
基金资助:
辽宁省科技计划联合计划(技术攻关计划)项目(2024JH2/102600166)、国家重点研发计划项目(2022YFF0801301)、辽宁省农业气象灾害重点实验室联合开放基金课题(2024SYIAEKFZD08,2025SYIAEKFZD08)、中国气象局气象能力提升联合研究专项(23NLTSQ012,23NLTSZ006)和国家自然科学基金项目(42275202)

Variation characteristics of high-temperature and drought compound disasters in Liaoning Province based on Copula function and random forest.

LIU Yutong^1,2, LIU Maohua¹, CHEN Nina^2,3,4*, MI Na^2,3,4, YU Wen-ying^2,3,4, WU Jinwen^2,3,4, FENG Rui^2,3,4

¹School of Transportation and Geomatics Engineering, Shenyang Jianzhu University, Shenyang 110168, China;
²Institute of Atmospheric Environment, China Meteorological Administration, Shenyang 110166, China;
³Shenyang Academy of Agricultural and Ecological Meteorology, Shenyang 110166, China;
⁴Liaoning Province Key Laboratory of Agrometeorological Disasters, Shenyang 110166, China

Received:2025-11-28 Accepted:2026-04-01 Online:2026-05-18 Published:2026-11-18

摘要/Abstract

摘要： 全球气候变化背景下,高温干旱复合灾害频发,对粮食安全、生态环境及社会经济可持续发展构成严峻挑战。本研究基于1971—2024年气象观测数据,运用嵌套式Copula模型结合随机森林算法构建强度指数,耦合GIS技术对辽宁省高温干旱复合灾害的时空分异、重现期特征和强度演变规律进行分析。结果表明:研究期间,辽宁省高温干旱复合灾害发生频次呈现西高东低的格局,西部为持续扩张和强化的高频集聚区,而东部及沿海地区则长期稳定为低频区。辽宁省高温干旱复合灾害以短重现期(0～2年)事件为主导,其发生频次高、空间集聚性强,西部是核心高发区;随着重现期延长,灾害发生范围急剧收缩、频次降低,长重现期事件仅在西部零星分布。单次事件强度经历了弱发-初增-剧增-维持-衰减的阶段性演变,于1990s达到峰值;而累计强度则从1970s的西部单极低值集聚,逐步演变为2010s的西部-中部双极高强度集聚格局,且西部核心区强度在研究期末达到峰值。综上,辽宁省西部为高温干旱复合灾害核心影响区域,且灾害强度呈增加趋势。本研究结果可为辽宁省防灾减灾策略制定及风险管控提供科学依据。

关键词: 高温干旱复合灾害, Copula函数, 随机森林, GIS

Abstract: Under the backdrop of global climate change, the frequent occurrence of combined disasters of high temperature and drought poses severe challenges to food security, ecological environment, and sustainable socio-economic development. Based on the meteorological observation data from 1971 to 2024, we constructed an intensity index by combining the nested Copula model with the random forest algorithm, and analyzed the spatiotemporal variations, recurrence interval characteristics and intensity evolution law of compound high-temperature and drought disasters in Liaoning Province by coupling GIS technology. The results showed that the occurrence frequency of compound high-temperature and drought disasters presented a pattern of being high in the west and low in the east. The western region was a continuously expanding and intensifying high-frequency agglomeration area, while the eastern and coastal areas remained a stable low-frequency area for a long time. Compound high-temperature and drought disasters in Liaoning Province were dominated by short recurrence interval (0-2 years) events, which featured with high occurrence frequency and strong spatial agglomeration. The western region as the core high-incidence area. With the extension of the return period, the occurrence scope of disasters shrank sharply and the frequency decreased, and long recurrence interval events were only sporadically distributed in the western region. The intensity of single events experienced a phased evolution of weak occurrence-initial increase-rapid increase-maintenance-attenuation, peaking in the 1990s. The cumulative intensity gradually evolved from a pattern of single low-value agglomeration in the western region in the 1970s to a dual high-intensity agglomeration pattern in the western and central regions in the 2010s. The intensity of the core western region reached its peak at the end of the study period. In summary, the western region of Liaoning Province was the core affected area of compound high-temperature and drought disasters, and the disaster intensity showed an increasing trend. This study could provide a scientific basis for the formulation of disaster prevention and mitigation strategies and risk management in Liaoning Province.

Key words: high-temperature and drought compound disasters, Copula function, random forest, GIS

刘宇彤, 刘茂华, 陈妮娜, 米娜, 于文颖, 武晋雯, 冯锐. 基于Copula函数和随机森林的辽宁省高温干旱复合灾害变化特征[J]. 应用生态学报, 2026, 37(5): 1595-1604.

LIU Yutong, LIU Maohua, CHEN Nina, MI Na, YU Wen-ying, WU Jinwen, FENG Rui. Variation characteristics of high-temperature and drought compound disasters in Liaoning Province based on Copula function and random forest.[J]. Chinese Journal of Applied Ecology, 2026, 37(5): 1595-1604.

参考文献

[1] Lamptey B, Kolli RK, Amo-Boateng M. Climate change information for regional impact and risk assessment. Frontiers in Environmental Science, 2023, 11: 1207320
[2] Afroz M, Chen G, Anandhi A. Drought- and heatwave-associated compound extremes: A review of hotspots, variables, parameters, drivers, impacts, and analysis frameworks. Frontiers in Earth Science, 2023, 10: 914437
[3] Zscheischler J, Martius O, Westra S, et al. A typology of compound weather and climate events. Nature Reviews Earth & Environment, 2020, 1: 333-347
[4] 李瑞英, 吕桂恒, 郝晓雷, 等. 鲁西南区域性高温干旱复合事件特征及危险性分析. 中国农业气象, 2024, 45(6): 657-668
[5] Xie HR, Tian SC, Li F, et al. Contrasting diurnal impacts of vapor pressure deficit on water use efficiency in two semiarid steppe ecosystems. Ecological Processes, 2025, 14: 68
[6] Zhou YR, Li HX, Sun B, et al. Predicting maize production in Northeastern China: Unraveling the influence of summer compound heat-drought events through physical mechanisms. Agricultural and Forest Meteorology, 2025, 372: 110709
[7] 孙博, 王会军, 黄艳艳, 等. 2022年夏季中国高温干旱气候特征及成因探讨. 大气科学学报, 2023, 46(1): 1-8
[8] Zeng HL, Xiao C, Chen XY, et al. State of China’s climate in 2018. Atmospheric and Oceanic Science Letters, 2019, 12: 349-354
[9] 苏小琁, 阎琦, 陆井龙, 等. 1961—2023年辽宁区域性高温过程时空特征及其环流分型. 干旱气象, 2025, 43(4): 531-539
[10] 王艺霖, 王蕊, 单凤君, 等. 基于SPEI指数的辽宁省玉米干旱特征分析. 辽宁工业大学学报:自然科学版, 2024, 44(4): 273-280
[11] 曹永强, 李可欣, 任博, 等. 基于SPEI的辽宁省气象干旱特征及驱动因素分析. 水利水电科技进展, 2022, 42(5): 28-36
[12] 杨雨潇, 赵峻池, 张琪, 等. 变暖环境下华北地区农业气象灾害风险评估. Ⅲ. 京津冀地区高温事件持续日数与强度联合重现期分析. 中国农业气象, 2024, 45(7): 798-808
[13] Pourhagverdi F, Pakdaman Z, Bazrafshan O, et al. Uncertainty quantification in drought severity-duration-frequency analysis: A maximum entropy copula framework for arid region. Theoretical and Applied Climato-logy, 2025, 156: 1-23
[14] 苏春, 汪千程. 基于改进多层级嵌套Copula方法的机电产品可靠性评估. 东南大学学报:自然科学版, 2022, 52(5): 981-989
[15] 唐吉喆, 徐梦冉, 莫宇, 等. 生态地理分区视角下的辽宁省植被归一化植被指数的时空变化. 应用生态学报, 2023, 34(12): 3271-3278
[16] Vicente-Serrano SM, Beguería S, López-Moreno JI, et al. A multiscalar drought index sensitive to global warming: The standardized precipitation evapotranspiration index. Journal of Climate, 2010, 23: 1696-1718
[17] Lee S, Moriasi DN, Mehr AD, et al. Sensitivity of standardized precipitation and evapotranspiration index (SPEI) to the choice of SPEI probability distribution and evapotranspiration method. Journal of Hydrology: Regional Studies, 2024, 53: 101761
[18] 中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. 气象干旱等级(GB/T 20481—2017). 北京: 中国标准出版社, 2017
[19] 丁司丞, 方超, 卓玛拉姆, 等. 基于机器学习模型的农田土壤容重模拟及其影响因素. 应用生态学报, 2025, 36(9): 2827-2835
[20] Shi PH, Li Y, Biswas A, et al. Spatial-temporal evolution and intrinsic drivers of compound drought and heatwave events in Mainland China. Science of the Total Environment, 2024, 948: 174834
[21] 魏琳, 杨学军, 王旭东, 等. 基于三维Copula函数的滦河流域干支流丰枯遭遇研究. 海河水利, 2023(1): 87-91
[22] 任立良, 万仪欣, 袁山水, 等. 基于网格聚合理论和Copula函数的流域极端降水三维识别及时空动态演变研究. 水资源保护, 2025, 41(5): 151-159
[23] Schmitt J, Offermann F, Ribeiro AFS, et al. Drought risk management in agriculture: A Copula perspective on crop diversification. Agricultural Economics, 2024, 55: 823-847
[24] 龚郑洁, 雷勇, 钟露露, 等. 干旱胁迫下珠江流域植被响应的滞后效应及损失概率评估. 应用生态学报, 2024, 35(4): 1083-1091
[25] Bai JW, Li YM, Li JW, et al. Multinomial random forest. Pattern Recognition, 2022, 122: 108331
[26] Ganaie MA, Tanveer M, Suganthan PN, et al. Oblique and rotation double random forest. Neural Networks, 2022, 153: 496-517
[27] 王晓楠, 苏文浩, 董灵波, 等. 基于随机森林的兴安落叶松天然林单木年龄预估模型. 应用生态学报, 2024, 35(4): 1055-1063
[28] 贾子玉, 庄前友, 红梅, 等. 利用随机森林与遥感数据估算内蒙古东部草地土壤有机碳含量. 草业科学, 2025, 42(8): 1901-1910
[29] Peres RAP, Pereira CAB, De Souza Junior CR, et al. Analysis of learning curves in predictive modeling using exponential curve fitting with an asymptotic approach. PLoS One, 2024, 19(4): e0299811
[30] Mohr F, van Rijn JN. Fast and informative model selection using learning curve cross-validation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45: 13133-13145
[31] 安敏, 韦雅倩, 何伟军, 等. 气候变化和人类活动对长江经济带生态敏感性的影响分析. 环境科学, 2024, 45(10): 5833-5843
[32] 姜珊, 张淑杰, 张菁, 等. 基于SPI和SPEI指数的辽宁省大豆生长季干旱时空变化特征. 气象与环境学报, 2023, 39(4): 130-137
[33] 任嘉欣, 王卫光, 魏佳, 等. 中国夏季复合干热事件的时空演变规律及其与气候模态的关系. 中国农村水利水电, 2024(9): 173-180
[34] 张义桐, 郝增超, 陈阳, 等. 全球多时间尺度复合高温干旱增加. 中国科学: 地球科学, 2025, 55(10): 3321-3332
[35] Li WH, Li LF, Ting MF, et al. Intensification of Nor-thern Hemisphere subtropical highs in a warming climate. Nature Geoscience, 2012, 5: 830-834
[36] 林益同, 周晓宇, 赵春雨, 等. 1961—2020年辽宁省气候年景分析及综合气候年景评价. 气象与环境学报, 2023, 39(6): 105-111
[37] Ye FC, Wang P, Yang Y, et al. Anthropogenic forcing dominates changes in compound long-duration dry and heat extremes in China. Climatic Change, 2025, 178: 30
[38] 路畅, 马龙, 刘廷玺, 等. 1951—2018年中国年降水量及气象干旱的时空变异. 应用生态学报, 2022, 33(6): 1572-1580
[39] 刘洁, 王宁练, 花婷, 等. 1960—2016年中国北方半干旱区盛夏降水时空变化及其水汽输送特征分析. 气候变化研究进展, 2019, 15(3): 257-269
[40] 陈东东, 栗晓玮, 张璐阳, 等. 西南地区水稻生育期复合高温干旱事件时空演变特征. 生态学杂志, 2025, 44(11): 3864-3872
[41] 苗正伟, 徐利岗, 刘璐, 等. 京津冀地区夏季复合干旱高温事件变化特征分析. (2025-12-08)[2026-01-09]. 中国农村水利. https://link.cnki.net/urlid/42.1419.TV.20251205.1957.024
[42] 雷智雯, 陈东东, 栗晓玮, 等. 基于Copula函数的西南地区一季稻高温干旱复合事件演变特征. (2025-11-17)[2026-01-09]. 生态学杂志. https://link.cnki.net/urlid/21.1148.q.20251114.1424.004

基于Copula函数和随机森林的辽宁省高温干旱复合灾害变化特征

Variation characteristics of high-temperature and drought compound disasters in Liaoning Province based on Copula function and random forest.

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王宗星, 沈炜霞, 袁蔡燕, 许建秀, 颉洪涛. 2000—2020年钱塘江流域土地利用格局及水源涵养功能变化 [J]. 应用生态学报, 2026, 37(6): 1799-1807.
[2]	罗世纪, 张弥, 贾磊, 肖薇, 乔珩, 张神宝, 石婕, 葛培, 杨甫禹, 何洋. 基于机器学习模型的太湖浮叶植物区CO₂通量模拟 [J]. 应用生态学报, 2026, 37(5): 1651-1664.
[3]	王花兰, 王卓雅, 沈志源, 沈铚诚. 河西走廊地区旅游景区可达性及影响机制 [J]. 应用生态学报, 2026, 37(4): 1227-1238.
[4]	多玲花, 段雅卿, 王俊淇. 不同格网尺度下城市形态对北京四环内地表温度的季节性影响 [J]. 应用生态学报, 2026, 37(3): 683-694.
[5]	刘薇, 曹腾飞, 于福鑫, 黄可乐, 郑涵致, 牛百成. 黄河源园区归一化植被指数的时空格局与预测模型 [J]. 应用生态学报, 2025, 36(7): 2121-2130.
[6]	赵艺丰, 赵天亚, 王鹏, 张崇良, 徐宾铎, 薛莹, 纪毓鹏. 青岛近岸海域带鱼群体资源分布及其与环境因子的关系 [J]. 应用生态学报, 2025, 36(6): 1907-1914.
[7]	欧阳逸云, 李春辉, 倪荣雨, 赵平欣, 曾爱聪, 郭福涛. 基于双变量统计和多准则决策分析的小尺度森林火险区划 [J]. 应用生态学报, 2025, 36(1): 187-196.
[8]	刘宁, 鱼舜尧, 张采月, 赵允格. 黄土高原典型维管植物冠层下生物结皮的分布与预测 [J]. 应用生态学报, 2024, 35(9): 2571-2580.
[9]	丁启东, 王怡婧, 张俊华, 贾科利, 黄华雨. 利用CARS算法联合协变量估算盐碱农田土壤水分和有机质含量 [J]. 应用生态学报, 2024, 35(5): 1321-1330.
[10]	龚郑洁, 雷勇, 钟露露, 武传号. 干旱胁迫下珠江流域植被响应的滞后效应及损失概率评估 [J]. 应用生态学报, 2024, 35(4): 1083-1091.
[11]	赵允格, 吉静怡, 张万涛, 明姣, 黄琬雲, 高丽倩. 黄土高原生物土壤结皮分布时空分异特征 [J]. 应用生态学报, 2024, 35(3): 739-748.
[12]	欧阳逸云, 苏漳文, 李春辉, 曾爱聪, 郭福涛. 基于模糊逻辑和网络层次分析法的森林火险区划 [J]. 应用生态学报, 2024, 35(2): 354-362.
[13]	袁佳玉, 熊立, 吴志伟, 朱诗豪, 康平, 李顺. 江西省赣州市南康区松材线虫病发生特征 [J]. 应用生态学报, 2024, 35(2): 507-515.
[14]	来洒洒, 陈文波, 危小建, 程钰颖. 基于生态韧性评价的鄱阳湖流域自然保护地预测与空缺区识别 [J]. 应用生态学报, 2024, 35(12): 3461-3468.
[15]	焦鹏华, 牛健植, 苗禹博, 李君宜, 王迪. 2001—2020年全球植被对极端气候的响应 [J]. 应用生态学报, 2024, 35(11): 2992-3004.