洮儿河流域极端降水特征与典型雨型分类研究OA
Study on Extreme Precipitation Characteristics and Typical Rainfall Pattern Classification in the Tao'er River Basin
在全球气候变暖的背景下,极端降水事件日益频发,对流域水资源调控与灾害防控构成严峻挑战.以洮儿河流域为典型区域,基于1993-2024年汛期(5-9月)区域站点实测降水资料,采用K-means聚类算法对1 326场极端降水过程进行分类,识别出4种典型雨型.进一步运用特征参数法系统分析各类雨型的强度、历时与结构特征,并通过趋势检验与线性回归方法,探讨不同雨型的发生频次变化及其时序演化规律.结果表明:洮儿河流域汛期极端降水量在汛期总降水量的占比整体呈波动上升趋势,前期型和中期型发生频次较高,其中前期型呈显著上升趋势;后期型与均匀型发生概率较低,频次变化较为稳定.研究成果可为流域内极端降水过程的识别与演变分析提供方法支撑,对提升区域暴雨灾害的防御能力与流域水资源风险管理水平具有重要参考价值.
Against the backdrop of global warming,extreme precipitation events are occurring with increasing frequency.This poses severe challenges to the management of water resources in watersheds and to disaster prevention and control.This study focuses on the Tao'er River Basin,conducting a systematic,event-based analysis of extreme precipitation using hourly observational data from regional meteorological stations during the flood seasons(May-September)from 1993 to 2024.Independent rainfall episodes were used to identify extreme precipitation events by combining minimum inter-event time intervals and a percentile-based threshold method.This approach provides a clear physical basis for defining extreme precipitation thresholds and yielded 1 326 independent extreme events.Dimensionless cumulative rainfall curves were then constructed for individual events and K-means clustering was applied to classify the temporal structures of precipitation.This initially produced ten objective clusters.These clusters were then consolidated into four typical rainfall patterns—early-peak,mid-peak,late-peak,and uniform—according to the primary concentration stage of rainfall within the event duration.Characteristic parameters,including total precipitation,duration,average intensity,maximum intensity,peak time coefficient and rainfall centroid coefficient,were employed to characterise each rainfall pattern systematically and quantify its intensity,persistence and precipitation concentration features.Trend tests and linear regression were then used to examine the variations in occurrence frequency of the different rainfall patterns over time,and their contribution to total flood-season precipitation.The results indicate that extreme precipitation in the Tao'er River Basin is primarily concentrated between June and August,with July contributing most significantly,while the early and late flood seasons contribute relatively less.The contribution of extreme precipitation to total flood-season rainfall shows an increasing trend,with a pronounced rise after 2009.This suggests a gradual shift towards extreme rainfall dominating the basin's precipitation structure.The four rainfall patterns differ significantly in terms of duration,rainfall concentration,position and intensity evolution.Early-peak events are characterised by short duration and highly concentrated intensity,reflecting abruptness.Mid-peak events display relatively balanced temporal distributions.Late-peak events exhibit a clear lag in rainfall concentration.Uniform events are associated with longer durations and more evenly distributed rainfall.Early-peak and mid-peak events dominated throughout the study period,with early-peak events demonstrating a statistically significant increasing trend.This study systematically reveals the temporal structural characteristics of extreme precipitation in the Tao'er River Basin,providing a sound physical foundation for identifying extreme events,constructing typical rainfall patterns,and assessing basin-scale flood risk.The findings offer valuable scientific support for optimising flood control operations and enhancing regional water resources management in the context of ongoing climate change.
尹莎莎;刘建卫;逄晓腾
大连理工大学建设工程学院,辽宁 大连 116024大连理工大学建设工程学院,辽宁 大连 116024大连理工大学建设工程学院,辽宁 大连 116024
建筑与水利
极端降水降水时程结构雨型分类百分位阈值法洮儿河流域
extreme precipitationprecipitation temporal structurerainfall pattern classificationpercentile-based threshold methodTao'er River Basin
《人民珠江》 2026 (6)
40-49,10
国家自然科学基金(52309079)
评论