1.衡阳师范学院地理与旅游学院,湖南 衡阳 421002
2.水利部水旱灾害防御重点实验室,江苏 南京 210029
3.湖南省 环境教育与可持续发展研究基地,湖南 衡阳 421002
周佳龙(2001—),男,硕士研究生,主要从事水文水资源及防洪减灾方面工作。
杨柳(1986—),女,博士,副教授,主要从事城镇化下水系适应性及防洪减灾等工作。
收稿:2025-12-18,
修回:2026-01-28,
录用:2026-02-03,
网络首发:2026-02-13,
纸质出版:2026-09-25
移动端阅览
周佳龙,杨柳,舒惠,等.基于可解释机器学习的衡阳市洪涝风险评估及影响因子分析[J].人民珠江,2026,47(9):10-24.
ZHOU Jialong,YANG Liu,SHU Hui,et al.Flood Risk Assessment and Driver Analysis in Hengyang City Using Explainable Machine Learning[J].PEARL RIVER,2026,47(09):10-24.
周佳龙,杨柳,舒惠,等.基于可解释机器学习的衡阳市洪涝风险评估及影响因子分析[J].人民珠江,2026,47(9):10-24. DOI: 10.3969/j.issn.1001-9235.2026.09.002.
ZHOU Jialong,YANG Liu,SHU Hui,et al.Flood Risk Assessment and Driver Analysis in Hengyang City Using Explainable Machine Learning[J].PEARL RIVER,2026,47(09):10-24. DOI: 10.3969/j.issn.1001-9235.2026.09.002.
洪涝灾害是地球上发生最频繁且危害性较大的自然灾害,对洪涝风险进行精确评估及其驱动因素的分析在防洪减灾工作中具有重要指示意义。针对传统洪涝风险评价中仅考虑距河远近或单一水系指标对洪涝的影响,而较少对其他水系结构指标对洪涝风险影响作用作深入客观分析的问题,研究构建四维度洪涝风险评价体系,从致灾因子、孕灾环境、承灾体和防灾减灾能力四维度进行洪涝风险评价并验证,同时利用可解释机器学习算法,选取包含河网密度、河流曲度、支流发育系数和水面率的水系指标等影响因子,对洪涝风险影响因子的影响机制进行深入客观分析。结果表明: ①衡阳市洪涝风险呈现“中部、东南东北较高、西部较低”的分布特征,衡阳市市辖区、衡山县和常宁市风险值较高;②基于可解释机器学习算法的影响因子分析来看,湘江流域衡阳段洪涝风险主要受海拔、河网密度和土壤湿度的影响,各影响因子对洪涝风险的影响具有阈值效应,可以为类似地区防洪减灾提供科学依据和参考。
Flooding is among the most frequent and damaging natural hazards worldwide. Accurate flood risk assessment and rigorous analysis of its driving factors are of critical importance for flood prevention and mitigation. Traditional flood risk assessments tend to focus only on proximity to rivers or a single hydrological indicator
lacking in-depth and objective consideration of the influence of other drainage network structure characteristics. To address this gap
this study develops a comprehensive four-dimensional flood risk evaluation framework that integrates hazard factors
hazard-forming environment
exposed elements
and disaster prevention and mitigation capacity. The framework is validated through a consolidated risk mapping procedure
and the resulting indicator set is combined with explainable machine learning algorithms. A suite of drainage network metrics—including drainage density
channel curvature
tributary development coefficient
and water surface ratio—alongside topographic and soil variables
is incorporated to systematically reveal how the water system structure interacts with other controls to shape spatial flood risk.The results indicate a distinct spatial pattern of flood risk across Hengyang City: ① Higher risk concentrates in the central
southeastern
and northeastern parts of the study area
while the western region exhibits relatively lower risk. At the administrative level
the municipal district of Hengyang
Hengshan County
and Changning City show notably elevated risk values. ② Explainable machine learning-based driver analysis reveals that flood risk in the Hengyang section of the Xiangjiang River Basin is predominantly influenced by elevation
drainage density
and soil moisture. Elevation exerts strong control via its effects on runoff generation and flow routing; densely dissected drainage networks accelerate runoff concentration and transmission
increasing inundation potential in downstream lowlands; high soil moisture reduces infiltration capacity and thereby elevates surface runoff and flooding likelihood. Importantly
the relationships between these predictors and flood risk exhibit threshold behavior rather than simple linear trends—risk escalates sharply once critical indicator values are exceeded. Identifying such thresholds enhances the practical utility of the results for flood risk zoning and management.Overall
by integrating a four-dimensional evaluation framework with explainable machine learning and explicit water system metrics
this study quantifies the coupled effects of drainage structure
topography
soil conditions
and human exposure
and pinpoints key thresholds and high-risk spatial patterns. These findings provide scientifically grounded guidance for flood hazard mapping
early warning prioritization
and targeted mitigation measures in the study area and in other basins with similar hydro-geomorphic and socio-environmental characteristics.
李超超 , 程晓陶 , 申若竹 , 等 . 城市化背景下洪涝灾害新特点及其形成机理 [J]. 灾害学 , 2019 , 34 ( 2 ): 57 - 62 .
Li Chaochao , Chen Xiaotao , Shen Ruozhu , et al . New characteristics and formation mechanism of flood and waterlogging disasters in the context of rapid urbanization [J]. Journal of Catastrophology , 2019 , 34 ( 2 ): 57 - 62 . (in Chinese)
Cai Tian , Li Xinyu , Ding Xiang , et al . Flood risk assessment based on hydrodynamic model and fuzzy comprehensive evaluation with GIS technique [J]. International Journal of Disaster Risk Reduction , 2019 , 35 . DOI: 10.1016/j.ijdrr.2019.101077 http://dx.doi.org/10.1016/j.ijdrr.2019.101077 .
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 .
Xu 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 . (in Chinese)
Avakyan A B , Polyushkin A A . Estimation of flood losses [J]. Hydrotechnical Construction , 1991 , 25 ( 4 ): 181 - 185 .
Allen S K , Linsbauer A , Randhawa S S , et al . Glacial lake outburst flood risk in Himachal Pradesh, India: an integrative and anticipatory approach considering current and future threats [J]. Natural Hazards , 2016 , 84 ( 3 ): 1741 - 1763 .
De Moel H , Jongman B , Kreibich H , et al . Flood risk assessments at different spatial scales [J]. Mitigation and Adaptation Strategies for Global Change , 2015 , 20 ( 6 ): 865 - 90 .
金菊良 , 宋占智 , 周玉良 , 等 . 水旱灾害风险评估方法体系及其实证研究 [J]. 水利水电科技进展 , 2015 , 35 ( 5 ): 142 - 151 .
Jin Juliang , Song Zhanzhi , Zhou Yuliang , et al . Risk assessment method system for flood and drought disasters and its application [J]. Advances in Science and Technology of Water Resources , 2015 , 35 ( 5 ): 142 - 151 . (in Chinese)
曲丽英 . 闽江流域洪涝风险区划快速评估研究 [J]. 中国农村水利水电 , 2023 ( 10 ): 30 - 37,46 .
Qu Liying . Research on the rapid assessment of flood risk zoning in the Minjiang Basin [J]. China Rural Water and Hydropower , 2023 ( 10 ): 30 - 37,46 . (in Chinese)
杨柳 , 冯畅 , 郭哿 , 等 . 基于四维度模型的湖南省湘江流域洪涝灾害风险评估 [J]. 水资源与水工程学报 , 2024 , 35 ( 1 ): 26 - 36 .
Yang Liu , Feng Chang , Guo Ge , et al . Assessment of flood disaster risk based on a four-dimensional model in the Hunan section of the Xiangjiang River Basin [J]. Journal of Water Resources and Water Engineering , 2024 , 35 ( 1 ): 26 - 36 . (in Chinese)
杨柳 , 王晨颖 , 冯畅 , 等 . 考虑水系演变的湘江流域洪水风险四维评价体系构建 [J]. 农业工程学报 , 2023 , 39 ( 3 ): 92 - 101 .
Yang Liu , Wang Chenying , Feng Chang , et al . Four-dimensional assessment system for flood risks with the river network system evolution in Xiangjiang River Basin in Hunan of China [J]. Transactions of the Chinese Society of Agricultural Engineering , 2023 , 39 ( 3 ): 92 - 101 . (in Chinese)
古丽扎尔·莫明 , 杨莲梅 , 刘艳 , 等 . 伊犁河流域近30 a洪水灾害时空分布及孕灾环境特征研究 [J]. 干旱区地理 , 2024 , 47 ( 8 ): 1314 - 1326 .
Gulzar Momin , Yang Lianmei , Liu Yan , et al . Spatial and temporal distribution of flood disasters and characteristics of disaster-prone environment in Ili River Basin in the recent 30 years [J]. Arid Land Geography , 2024 , 47 ( 8 ): 1314 - 1326 . (in Chinese)
王德运 , 张露丹 , 吴祈 , 等 . 基于机器学习算法的洪涝灾害风险评估:以宜昌市为例 [J]. 长江流域资源与环境 , 2023 , 32 ( 8 ): 1710 - 1723 .
Wang Deyun , Zhang Ludan , Wu Qi , et al . Flood risk assessment based on machine learning algorithms: a case study of Yichang City [J]. Resources and Environment in the Yangtze Basin , 2023 , 32 ( 8 ): 1710 - 1723 . (in Chinese)
王金虎 , 张东晖 , 王钰尧 . 城市洪涝风险的组合赋重方法研究与应用: 以淮安市为例 [J]. 水利水电技术(中英文) , 2025 , 56 ( 8 ): 91 - 104 .
Wang Jinhu , Zhang Donghui , Wang Yuyao . Research and application of combined empowerment method for urban flood risk: Take Huai'an City as an example [J]. Water Resources and Hydropower Engineering , 2025 , 56 ( 8 ): 91 - 104 . (in Chinese)
王磊之 , 云兆得 , 胡庆芳 , 等 . 国内外城市雨洪管理指标体系对比及启示 [J]. 水资源保护 , 2022 , 38 ( 2 ): 25 - 31 .
Wang Leizhi , Yun Zhaode , Hu Qingfang , et al . Comparison and enlightenment of urban stormwater management indicator system at home and abroad [J]. Water Resources Protection , 2022 , 38 ( 2 ): 25 - 31 . (in Chinese)
苏广全 , 吕海深 , 朱永华 , 等 . 基于博弈论组合赋权的洪水风险评价:以武威地区为例 [J]. 干旱区研究 , 2022 , 39 ( 3 ): 801 - 809 .
Su Guangquan , Lyu Haishen , Zhu Yonghua , et al . Combined weight method based on game theory for flood risk assessment in the Wuwei Region [J]. Arid Zone Research , 2022 , 39 ( 3 ): 801 - 809 . (in Chinese)
林奕晨 , 周鹏 , 潘悦 , 等 . 荆州市洪涝灾害影响因子探究及风险评估:基于随机森林和XGBoost算法 [J]. 中国农村水利水电 , 2022 ( 6 ): 125 - 132 .
Lin Yichen , Zhou Peng , Pan Yue , et al . Influencing factor research and risk assessment of flood disasters in Jingzhou City: Based on random forest and XGBoost algorithm [J]. China Rural Water and Hydropower , 2022 ( 6 ): 125 - 132 . (in Chinese)
陈军飞 , 张玉琦 , 嵇娟 , 等 . 长三角城市群洪涝韧性时空演化及影响因素 [J]. 水资源保护 , 2024 , 40 ( 6 ): 58 - 68 .
Chen Junfei , ZhanG Yuqi , Ji Juan , et al . Spatiotemporal evolution and influencing factors of flood resilience in Yangtze River Delta urban agglomeration [J]. Water Resources Protection , 2024 , 40 ( 6 ): 58 - 68 . (in Chinese)
项瑾 , 杨柳 , 冯畅 , 等 . 湘江流域中游洪水灾害风险评估及驱动力分析 [J]. 人民珠江 , 2025 , 46 ( 8 ): 13 - 22 .
Xiang Jin , Yang Liu , Feng Chang , et al . Flood risk assessment in midstream Xiangjiang River Basin and its driving force analysis [J]. Pearl River , 2025 , 46 ( 8 ): 13 - 22 . (in Chinese)
Holzinger A . Introduction to machine learning & knowledge extraction (MAKE) [J]. Machine Learning&Knowledge Extraction , 2019 , 1 ( 1 ): 1 - 20 .
Wickramasinghe C S , Marino D , Amarasinghe K . Explainable artificial intelligence [J]. Front Comput Sci-Switz , 2023 , 5 : 3 .
宣润泽 , 杨柳 , 冯畅 , 等 . 城镇扩张背景下湘江中游水系变化及其响应研究 [J]. 水电能源科学 , 2023 , 41 ( 5 ): 30 - 33 .
Xuan Runze , Yang Liu , Feng Chang , et al . Changes of river network system in midstream of Xiangjiang River and its response to urban expansion [J]. Water Resources and Power , 2023 , 41 ( 5 ): 30 - 33 . (in Chinese)
耿晓君 , 赵钟楠 , 李原园 , 等 . 我国极端降水与洪涝灾害事件特点及应对思路和对策 [J]. 中国水利 , 2025 ( 22 ): 47 - 51,72 .
Geng Xiaojun , Zhao Zhongnan , Li Yuanyuan , et al . Characteristics of extreme precipitation and flood disasters in China and response approaches and countermeasures [J]. China Water Resources , 2025 ( 22 ): 47 - 51,72 . (in Chinese)
Liu Yunping , Gao Yuqin , Yuan Feifei , et al . The influence of change in river networks on waterway connectivity and flooding processes [J]. Hydrological Sciences Journal , 2024 , 69 ( 5 ): 997 - 1011 .
Xu Yu , Xu Youpeng , Wu Lei . Changes in river networks and their storage and regulation capacities in the Rapidly Urbanized Taihu Basin, China [J]. Hydrological Processes , 2018 , 32 ( 22 ): 3341 - 3351 .
胡嘉玮 , 徐宗学 , 陈浩 , 等 . 基于精细化GDP格局的城市洪涝风险评估:以济南市黄台桥流域为例 [J]. 南水北调与水利科技(中英文) , 2025 , 23 ( 6 ): 1480 - 1490 .
Hu Jiawei , Xu Zongxue , Chen Hao , et al . Urban flood risk assessment based on refined GDP spatial pattern:A case study of Huangtaiqiao River basin, Jinan City [J]. South-to-North Water Transfers and Water Science & Technology , 2025 , 23 ( 6 ): 1480 - 1490 . (in Chinese)
吴京戎 , 蒋翔 . 基于随机森林和组合赋权法的武汉市洪涝灾害风险评估研究 [J]. 给水排水 , 2024 , 60 ( 7 ): 162 - 168,174 .
Wu Jingrong , Jiang Xiang . Research on flood risk assessment of Wuhan city based on random forest and combined weighting [J]. Water & Wastewater Engineering , 2024 , 60 ( 7 ): 162 - 168,174 . (in Chinese)
李红艳 , 郝景开 , 刘大为 , 等 . 基于元启发式算法优化的洪水风险评价模型 [J]. 水资源保护 , 2024 , 40 ( 6 ): 104 - 112,164 .
Li Hongyan , Hao Jingkai , Liu Dawei , et al . An optimized flood risk evaluation model based on meta-heuristic algorithm [J]. Water Resources Protection , 2024 , 40 ( 6 ): 104 - 112,164 . (in Chinese)
苑希民 , 苏智 , 王丽娜 , 等 . 基于AHP-CRITIC-灰色关联分析的城市洪水风险评估研究 [J]. 水资源与水工程学报 , 2025 , 36 ( 1 ): 11 - 17,28 .
Yuan Ximin , Su Zhi , Wang Lina , et al . Urban flood risk assessment based on AHP-CRITIC-grey relational analysis [J]. Journal of Water Resources and Water Engineering , 2025 , 36 ( 1 ): 11 - 17,28 . (in Chinese)
Guo Guizhen , Gao Yang , Sun Kun . Whether higher risk indicates severe loss in the flood disaster assessment: A comparative analysis in the Hubei province of central China [J]. Journal of Hydrology: Regional Studies (DEC 2024 ), 2024 , 56. DOI: 10.1016/j.ejrh.2024.102002 http://dx.doi.org/10.1016/j.ejrh.2024.102002 .
邓雪 , 李家铭 , 曾浩健 , 等 . 层次分析法权重计算方法分析及其应用研究 [J]. 数学的实践与认识 , 2012 , 42 ( 7 ): 93 - 100 .
Deng Xue , Li Jiaming , Zeng Hao Jian , et al . Research on computation methods of ahp weight vector and its applications [J]. Mathematics in Practice and Theory , 2012 , 42 ( 7 ): 93 - 100 . (in Chinese)
兰继斌 , 徐扬 , 霍良安 , 等 . 模糊层次分析法权重研究 [J]. 系统工程理论与实践 , 2006 ( 9 ): 107 - 12 .
Lan Jibin , Xu Yang , Huo Liang’an . research on the priorities of fuzzy analytical hierarchy process [J]. Systems Engineering-Theory & Practice , 2006 ( 9 ): 107 - 112 . (in Chinese)
程启月 . 评测指标权重确定的结构熵权法 [J]. 系统工程理论与实践 , 2010 , 30 ( 7 ): 1225 - 8 .
Cheng Qiyue . Structure entropy weight method to confirm the weight of evaluating index [J]. Systems Engineering-Theory & Practice , 2010 , 30 ( 7 ): 1225 - 1228 . (in Chinese)
Cai Xitian , Li Luyi , Fisher J B , et al . The responses of ecosystem water use efficiency to CO 2 , nitrogen deposition, and climatic drivers across China [J]. Journal of Hydrology , 2023 , 622 . DOI: 10.1016/j.jhydrol.2023.129696 http://dx.doi.org/10.1016/j.jhydrol.2023.129696 .
Jiang Shijie , Tarasova Larisa , Yu Guo , et al . Compounding effects in flood drivers challenge estimates of extreme river floods [J]. Science Advances , 2024 , 10 ( 13 ). DOI: 10.1126/sciadv.adl4005 http://dx.doi.org/10.1126/sciadv.adl4005 .
李占山 , 姚鑫 , 刘兆赓 , 等 . 基于LightGBM的特征选择算法 [J]. 东北大学学报(自然科学版) , 2021 , 42 ( 12 ): 1688 - 1695 .
Li Zhanshan , Yao Xin , Liu Zhaogeng , et al . Feature selection algorithm based on LightGBM [J]. Journal of Northeastern University(Natural Science) , 2021 , 42 ( 12 ): 1688 - 1695 . (in Chinese)
0
浏览量
709
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621