李强,赵铜铁钢

• 1:中山大学水资源与环境研究中心

摘要(Abstract)：

【目的】洪涝灾害统计调查是变化环境下区域防灾减灾和洪涝风险管理的重要依据。大语言模型在水利工程领域展现出强大应用潜力，为充分利用其语义理解和信息抽取能力，提出基于大语言模型的洪涝灾害统计调查方法，全过程实现大语言模型自主的互联网数据统计调查。【方法】面向北京“7·21”“23·7”和珠江流域“22·6”、南方地区“24·4”暴雨洪涝灾害，采用大语言模型统计调查危险程度和损失情况。【结果】结果表明，统计调查准确率随大语言模型温度参数升高而降低，当温度参数为0时准确率最高。北京“7·21”暴雨洪涝灾害的房屋倒塌数量、受灾人口和直接经济损失准确率超过90%,农作物受灾面积和洪峰流量的空检率超过40%;“23·7”暴雨洪涝灾害准确率更高，房屋倒塌数量、死亡失踪人口、受灾人口、农作物受灾面积和洪峰流量的准确率均超过90%。珠江流域“22·6”暴雨洪涝灾害的石角站洪峰流量和飞来峡最大入库流量准确率超过90%,北江流域最大小时降雨量和平均降雨量的准确率分别为83%和61%;南方地区“24·4”暴雨洪涝灾害石角站洪峰流量的准确率为89%,飞来峡最大入库流量、北江流域最大小时降雨量和平均降雨量的空检率超过70%。【结论】大语言模型适用于洪涝灾害数据统计调查，可以为水旱灾害防御工作提供数据支撑。

关键词(KeyWords)： 洪涝灾害;统计调查;大语言模型;互联网数据;信息抽取;气候变化;风险评估;降雨

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划项目(2023YFF0804900);; 国家自然科学基金项目(52379033)

作者(Author): 李强,赵铜铁钢

DOI: 10.13928/j.cnki.wrahe.2025.09.005

参考文献(References)：

• [1] 王浩，赵铜铁钢，田雨，等.考虑时间变化的洪涝灾害损失评估[J].水利学报，2024,55(2):127-136.WANG Hao,ZHAO Tongtiegang,TIAN Yu,et al.Incorporating temporal changes into flood loss assessments[J].Journal of Hydraulic Engineering,2024,55(2):127-136.
• [2] 张建云，舒章康，王鸿杰，等.郑州“7·20”暴雨洪涝几个水文问题的讨论[J].地理学报，2023,78(7):1618-1626.ZHANG Jianyun,SHU Zhangkang,WANG Hongjie,et al.A discussion on several hydrological issues of “7·20” rainstorm and flood in Zhengzhou[J].Acta Geographica Sinica,2023,78(7):1618-1626.
• [3] 中华人民共和国水利部.中国水旱灾害防御公报2023[M].北京：中国水利水电出版社，2024.Ministry of Water Resources of the People’s Republic of China.China Flood and Drought Disaster Prevention Bulletin 2023[M].Beijing:China Water&Power Press,2024.
• [4] ZHANG W,VILLARINI G,VECCHI G A,et al.Urbanization exacerbated the rainfall and flooding caused by hurricane Harvey in Houston[J].Nature,2018,563(7731):384-388.
• [5] 夏军，陈进，佘敦先，等.变化环境下中国现代水网建设的机遇与挑战[J].地理学报，2023,78(7):1608-1617.XIA Jun,CHENG Jin,SHE Dunxian,et al.Opportunities and challenges of national water network construction under changing environment[J].Acta Geographica Sinica,2023,78(7):1608-1617.
• [6] RENTSCHLER J,AVNER P,MARCONCINI M,et al.Global evidence of rapid urban growth in flood zones since 1985[J].Nature,2023,622(7981):87-92.
• [7] 孙晓英，叶冬冬，张建涛，等.北京市“23·7”极端强降雨灾后恢复重建关键问题及对策研究[J].水利发展研究，2024,24(11):22-28.SUN Xiaoying,YE Dongdong,ZHANG Jiantao,et al.Research on key issuesand measures of Beijing“23·7”extreme heavy rainfall disaster recovery and reconstruction planning[J].Water Resources Development Research,2024,24(11):22-28.
• [8] 蓝羽栖，陈学秋，张文明.北江“2024.4”与“2022.6”两场特大洪水比较分析[J].中国防汛抗旱，2024,34(6):8-12.LAN Yuxi,CHEN Xueqiu,ZHANG Wenming.A comparative analysis of two catastrophic floods of Beijiang River Basin in April 2024 and June 2022[J].China Flood & Drought Management,2024,34(6):8-12.
• [9] 康亚静，刘宇，解家毕.超标准暴雨洪水条件下南水北调中线工程沿线各单元风险评估[J].南水北调与水利科技(中英文),2023,21(2):342-351.KANG Y J,LIU Y,XIE J B.Risk assessment of each unit along the middle route of South-to-North Water Transfer Project under over-standard rainstorm flood conditions[J].South-to-North Water Transfers and Water Science & Technology,2023,21(2):342-351.
• [10] 李莹，赵珊珊.2001—2020年中国洪涝灾害损失与致灾危险性研究[J].气候变化研究进展，2022,18(2):154-165.LI Ying,ZHAO Shanshan.Floods losses and hazards in China from 2001 to 2020[J].Climate Change Research,2022,18(2):154-165.
• [11] 王小杰，夏军强，李启杰，等.长江中下游地区洪灾损失率函数构建及洪涝灾害损失评估[J].水资源保护，2023,39(5):88-98.WANG Xiaojie,XIA Junqiang,LI Qijie,et al.Progress on disaster mechanism and risk assessment of urban flood/waterlogging disasters in China[J].Water Resources Protection,2023,39(5):88-98.
• [12] LAZZARIN T,VIERO D P,MOLINARI D,et al.A new framework for flood damage assessment considering the within-event time evolution of hazard,exposure,and vulnerability[J].Journal of Hydrology,2022,615:128687.
• [13] 褚明华，杜晓鹤，何秉顺.我国水旱灾害防御应对[J].水利发展研究，2024,24(8):1-8.CHUMinghua,DUXiaohe,HEBingshun.China’sprevention and responsetoflood and drought disasters[J].Water Resources Development Research,2024,24(8):1-8.
• [14] LAZZARIN T,VIERO D P,MOLINARI D,et al.Flood damage functions based on a single physics-and data-based impact parameter that jointly accounts for water depth and velocity[J].Journal of Hydrology,2022,607:127485.
• [15] 米胤瑜，孔锋.气候变化背景下城市洪水风险管理体系国际比较与启示：以伦敦、纽约、郑州为例[J].水利水电技术(中英文),2023,54(3):21-34.MI Yinyu,KONG Feng.International comparison and enlightenment of urban flood risk management system under background of climate change:Taking London,New York and Zhengzhou as study cases[J].Water Resources and Hydropower Engineering,2023,54(3):21-34.
• [16] SHOYAMA K,CUI Q,HANASHIMA M,et al.Emergency flood detection using multiple information sources:Integrated analysis of natural hazard monitoring and social media data[J].Science of The Total Environment,2021,767:144371.
• [17] KRYVASHEYEU Y,CHEN H,OBRADOVICH N,et al.Rapid assessment of disaster damage using social media activity[J].Science Advances,2016,2(3):e1500779.
• [18] 张谱，张豪，孔锋，等.基于微博数据的暴雨洪涝灾害舆情特征研究：以2021年中国三场暴雨洪涝为例[J].水利水电技术(中英文),2023,54(2):47-59.ZHANG Pu,ZHANG Hao,KONG Feng,et al.A study on public opinion characteristics of rainstorm flooding disasters based on Sina Weibo data:Take the three rainstorm flooding disasters in China in 2021 as an example[J].Water Resources and Hydropower Engineering,2023,54(2):47-59.
• [19] 王艳东，李昊，王腾，等.基于社交媒体的突发事件应急信息挖掘与分析[J].武汉大学学报(信息科学版),2016,41(3):290-297.WANG Yandong,LI Hao,WANG Teng,et al.A study on user querying behavior in generative artificial intelligence environment[J].Geomatics and Information Science of Wuhan University,2016,41(3):290-297.
• [20] 邬柯杰，吴吉东，叶梦琪.社交媒体数据在自然灾害应急管理中的应用研究综述[J].地理科学进展，2020,39(8):1412-1422.WU Kejie,WU Jidong,YE Mengqi.A review on the application of social media data in natural disaster emergency management[J].Progress in Geography,2020,39(8):1412-1422.
• [21] 车万翔，窦志成，冯岩松，等.大模型时代的自然语言处理：挑战、机遇与发展[J].中国科学：信息科学，2023,53(9):1645-1687.CHE Wanxiang,DOU Zhicheng,FENG Yansong,et al.Towards a comprehensive understanding of the impact of large language models on natural language processing:Challenges,opportunities and future directions[J].Bulletin of National Natural Science Foundation of China,2023,53(9):1645-1687.
• [22] NOY S,ZHANG W.Experimental evidence on the productivity effects of generative artificial intelligence[J].Science,2023,381(6654):187-192.
• [23] SPITALE G,BILLER-ANDORNO N,GERMANI F,et al.Large language models encode clinical knowledge[J].Nature,2023,620(7972):172-180.
• [24] 王耀祖，李擎，戴张杰，等.大语言模型研究现状与趋势[J].工程科学学报，2024,46(8):1411-1425.WANG Yaozu,LI Qing,DAI Zhangjie,et al.Current status and trends in large language modeling research[J].Chinese Journal of Engineering,2024,46(8):1411-1425.
• [25] LIN Z,CHEN A,WANG X,et al.Large language models reveal big disparities in current wildfire research[J].Communications Earth & Environment,2024,5(1):1-6.
• [26] 冯钧，畅阳红，陆佳民，等.基于大语言模型的水工程调度知识图谱的构建与应用[J].计算机科学与探索，2024,18(6):1637-1647.FENG Jun,CHANG Yanghong,LU Jiamin,et al.Construction and application of knowledge graph for water engineering scheduling based on large language model[J].Journal of Frontiers of Computer Science and Technology,2024,18(6):1637-1647.
• [27] 孙维亚，邬俊杰，茹兴康，等.数字孪生在南水北调中线高质量发展中的应用实践[J].水利发展研究，2024,24(9):85-89.SUN Weiya,WU Junjie,RU Xingkang,et al.Application and practice of digital twins in high quality development of the middle route of the South-to-North Water Diversion Project[J].Water Resources Development Research,2024,24(9):85-89.
• [28] FOROUMANDI E,MORADKHANI H,SANCHEZ-VILA X,et al.ChatGPT in hydrology and earth sciences:Opportunities,prospects,and concerns[J].Water Resources Research,2023,59(10):e2023WR036288.
• [29] HALLORAN L J S,MHANNA S,BRUNNER P.AI tools such as ChatGPT will disrupt hydrology,too[J].Hydrological Processes,2023,37(3):e14843.
• [30] IRVINE D J,HALLORAN L J S,BRUNNER P.Opportunities and limitations of the ChatGPT Advanced Data Analysis plugin for hydrological analyses[J].Hydrological Processes,2023,37(10):e15015.
• [31] 张谱，张豪，孔锋，等.基于微博数据的暴雨洪涝灾害舆情特征研究：以 2021 年中国三场暴雨洪涝为例[J].水利水电技术(中英文),2023,54(2):47-59.ZHANG Pu,ZHANG Hao,KONG Feng,et al.A study on public opinion characteristics of rainstorm flooding disasters based on Sina Weibo data:take the three rainstorm flooding disasters in China in 2021 as an example[J].Water Resources and Hydropower Engineering,2023,54(2):47-59.
• [32] MIAO C,HU J,MORADKHANI H,et al.Hydrological research evolution:A large language model-based analysis of 310,000 studies published globally between 1980 and 2023[J].Water Resources Research,2024,60(6):e2024WR038077.
• [33] 龚世诚，艾萍，熊传圣，等.基于ChatGPT的网络空间中水文文本信息抽取技术[J].水力发电，2024,50(10):28-32.GONG Shicheng,AI Ping,XIONG Chuansheng,et al.Hydrological object information extraction technology in cyberspace based on ChatGPT[J].Water Power,2024,50(10):28-32.
• [34] 杨柳，姚葳，马辉，等.LLM在水利政府网站公共服务中的应用研究[J].水利信息化，2024(2):58-62.YANG Liu,YAOwei,MA Hui,et al.Research on the application of LLM in public services on water conservancy government websites[J].Water Resources Informatization,2024(2):58-62.
• [35] 钱峰，成建国，夏润亮，等.水利大模型的建设思路、构建框架与应用场景初探[J].中国水利，2024(9):9-19.QIAN Feng,CHENG Jianguo,XIA Runliang,et al.A preliminary exploration of the construction ideas,framework,and application scenarios of water conservancy large models[J].China Water Resources,2024(9):9-19.
• [36] RAY P P.Leveraging deep learning and language models in revolutionizing water resource management,research,and policy making:A case for ChatGPT[J].ACS ES&T Water,2023,3(8):1984-1986.
• [37] XUE Z,XU C,XU X.Application of ChatGPT in natural disaster prevention and reduction[J].Natural Hazards Research,2023,3(3):556-562.
• [38] 赵鑫，窦志成，文继荣.大语言模型时代下的信息检索研究发展趋势[J].中国科学基金，2023,37(5):786-792.ZHAO Xin,DOU Zhicheng,WEN Jirong.The development of information retrieval in the era of large language model[J].Bulletin of National Natural Science Foundation of China,2023,37(5):786-792.
• [39] 王若佳，范科鸣，刘智锋，等.生成式人工智能环境下用户信息检索式行为研究[J].数据分析与知识发现，2024,8(z1):20-30.WANG Ruojia,FAN Keming,LIU Zhifeng,et al.A study on user querying behavior in generative artificial intelligence environment[J].Data Analysis and Knowledge Discovery,2024,8(z1):20-30.
• [40] 孙晓华，车天琪，伊珺瑶，等.人工智能大模型、范式冲击与科技创新变革[J/OL].大连理工大学学报(社会科学版),2024:1-13.SUN Xiaohua,CHE Tianqi,YI Junyao,et al.Large generative artificial intelligence models,paradigm shift and science and innovation revolution[J/OL].Journal of Dalian University of Technology (Social Sciences),2024:1-13.
• [41] 宋晓猛，徐楠涛，董子然.变化环境下城市洪涝风险评估与调控策略[J].水利发展研究，2025,25(1):56-64,85.SONG Xiaomeng,XU Nantao,DONG Ziran.Risk assessment and control strategies of urban flooding under changing environments[J].Water Resources Development Research,2025,25(1):56-64,85.
• [42] 张建云，王银堂，贺瑞敏，等.中国城市洪涝问题及成因分析[J].水科学进展，2016,27(4):485-491.ZHANG Jianyun,WANG Yinting,HE Ruimin,et al.Discussion on the urban flood and waterlogging and causes analysis in China[J].Advances in Water Science,2016,27(4):485-491.
• [43] 刘培，黄鹏飞，许劼靖，等.西、北江三角洲洪水位对洪水归槽与河床下切的响应[J].南水北调与水利科技(中英文),2023,21(3):501-511.LIU P,HUANG P F,XU J J,et al.Response of flood level to flood returning to main channel and channel deepning in the West and North River Delta[J].South-to-North Water Transfers and Water Science & Technology,2023,21(3):501-511.
• [44] 江帆，王东海，曾智琳.近20年季风爆发前后珠江三角洲前汛期短时强降水的时空演变特征与成因[J].大气科学，2024,48(2):715-734.JIANG Fan,WANG Donghai,ZENG Zhilin.Spatiotemporal evolution characteristics and causes of short-duration heavy rainfall in the presummer rainy season over the Pearl River Delta Region before and after the South China sea summer monsoon onset in the past 20 Years[J].Chinese Journal of Atmospheric Sciences,2024,48(2):715-734.
• [45] 张启义.北京房山区“23·7”特大暴雨灾害的成因及启示[J].中国防汛抗旱，2023,33(10):43-47.ZHANG Qiyi.The cause and enlightenment of the “23.7” extremely rainstorm disaster in Fangshan District of Beijing City[J].China Flood & Drought Management,2023,33(10):43-47.
• [46] 荆浩，亢妍妍，吴宏议，等.北京“23·7”极端强降雨特征和成因分析[J].气象，2024,50(5):616-629.JIN Hao,KANG Yanyan,WU Hongyi,et al.Characteristics and causes of the July 2023 extremely torrential rain in Beijing[J].Meteorological Monthly,2024,50(5):616-629.
• [47] 王京晶，聂红海，蔡斯龙.“22·6”北江流域特大洪水分析[J].水文，2024,45(1):106-112.WANG Jingjing,NIE Honghai,CAI Silong.Investigation and analysis of “22·6” Beijiang River Basin extreme flood[J].Journal of China Hydrology,2024,45(1):106-112.
• [48] 卢康明，陈学秋.珠江流域北江“2022.6”特大洪水预报分析[J].中国防汛抗旱，2023,33(9):56-61.LU Kangming,CHEN Xueqiu.Analysis the catastrophic flood forecast of Beijiang River of Pearl River Basin in June 2022[J].China Flood & Drought Management,2023,33(9):56-61.
• [49] 孔锋.2012年北京“7·21”特大暴雨洪涝灾害应对及启示[J].中国减灾，2022(9):42-45.KONG Feng.Response to and enlightenment from the “7·21” catastrophic rainstorm and flood disaster in Beijing,2012[J].Disaster Reduction in China,2022(9):42-45.
• [50] 中华人民共和国应急管理部.应急管理部发布2022年全国十大自然灾害 [EB/OL].(2023-01-12)[2024-11-03].https://www.mem.gov.cn/xw/yjglbgzdt/202301/t20230112_440396.shtml.Ministry of Emergency Management of the People’s Republic of China.The Ministry of Emergency Management has released the top 10 natural disasters in 2022 [EB/OL].(2023-01-12)[2024-11-03].https://www.mem.gov.cn/xw/yjglbgzdt/202301/t20230112_ 440396.shtml.
• [51] 国家防灾减灾救灾委员会办公室，中华人民共和国应急管理部.国家防灾减灾救灾委员会办公室应急管理部发布2024年4月全国自然灾害情况[EB/OL].(2024-05-17)[2024-11-03].https://www.mem.gov.cn/xw/yjglbgzdt/202405/t20240517_488741.shtml.National Disaster Prevention and Reduction Commission Office,Ministry of Emergency Management of the People’s Republic of China.National Disaster Prevention and Reduction Commission Office,Ministry of Emergency Management of the People’s Republic of China released the national natural disaster situation for April 2024 [EB/OL].(2024-05-17)[2024-11-03].https://www.mem.gov.cn/xw/yjglbgzdt/202405/t20240517_488741.shtml.
• [52] 赵浜，曹树金.国内外生成式AI大模型执行情报领域典型任务的测试分析[J].情报资料工作，2023,44(5):6-17.ZHAO Bang,CAO Shujin.Test analysis of typical tasks in the information field performed by generative AI large models at home and abroad[J].Data Analysis and Knowledge Discovery,2023,44(5):6-17.
• [53] 唐永兰，徐桂荣，唐国瑛，等.“23·7”华北特大暴雨过程小时强降水时空分布特征[J].大气科学学报，2024,47(5):778-788.TANG Yonglan,XU Guirong,TANG Guoying,et al.Temporal and spatial distribution characteristics of hourly heavy rainfall of the “23·7” heavy rainstorm event in North China[J].Transactions of Atmospheric Sciences,2024,47(5):778-788.
• [54] 符娇兰，权婉晴，麦子，等.“23·7”华北特大暴雨过程雨强精细化特征及动力和热力条件初探[J].气象，2023,49(12):1435-1450.FU Jiaolan,QUAN Wanqing,MAI Zi,et al.Preliminary study on the refined characteristics of rainfall intensity and dynamic and thermodynamic conditions in the July 2023 severe torrential rain in North China[J].Meteorological Monthly,2023,49(12):1435-1450.
• [55] HUANG Y,HUANG J X.Exploring ChatGPT for next-generation information retrieval:Opportunities and challenges[J].Web Intelligence,2024,22(1):31-44.
• [56] 徐宗学，陈浩，任梅芳，等.中国城市洪涝致灾机理与风险评估研究进展[J].水科学进展，2020,31(5):713-724.XUE Zongxue,CHEN Hao,REN Meifang,et al.Progress on disaster mechanism and risk assessment of urban flood/waterlogging disasters in China[J].Advances in Water Science,2020,31(5):713-724.
• [57] 马俊学，高会然，许冲.北京市昌平区韩台村“23·7”暴雨山洪泥石流灾害特征分析[J].水利水电技术(中英文),2024,55(7):1-18.MA Junxue,GAO Huiran,XU Chong.Characteristics of flash flood-debris flow disaster induced by the ‘23·7’ rainstorm in Hantai Village,Changping District,Beijing[J].Water Resources and Hydropower Engineering,2024,55(7):1-18.
• [58] KREIBICH H,VAN LOON A F,SCHR?TER K,et al.The challenge of unprecedented floods and droughts in risk management[J].Nature,2022,608(7921):80-86.
• [59] 李永坤，胡小红，张焜，等.全球极端暴雨常态下的超大型城市洪涝应对策略研究：以北京市为例[J].中国水利，2023(12):31-36.LI Yongkun,HU Xiaohong,ZHANG Kun,et al.Flood response strategies for megacities in the context of global extreme rainstorms patterns:taking Beijing as an example[J].China Water Resources,2023(12):31-36.