基于STL时序分解的天然径流演变规律与影响因素分析  

Analyses of the evolution and influencing factors of natural runoff in China based on STL decomposition

作　　者：赵轶琳 刘文丰 李政 徐宗学[4,5] 阳坤 吴德丰 ZHAO Yilin;LIU Wenfeng;LI Zheng;XU Zongxue;YANG Kun;WU Defeng(State Key Laboratory of Efficient Utilization of Agricultural Water Resources,China Agricultural University,Beijing 100083,China;National Field Scientific Observation and Research Station on Efficient Water Use of Oasis Agriculture in Wuwei of Gansu Province,Wuwei 733000,China;Center for Agricultural Water Research in China,College of Water Resources and Civil Engineering,China Agricultural University,Beijing 100083,China;College of Water Sciences,Beijing Normal University,Beijing 100875,China;Beijing Key Laboratory of Urban Hydrological Cycle and Sponge City Technology,Beijing 100875,China;Department of Earth System Science,Tsinghua University,Beijing 100084,China)

机构地区：[1]农业水资源高效利用全国重点实验室中国农业大学,北京100083 [2]甘肃武威绿洲农业高效用水国家野外科学观测研究站,甘肃武威733000 [3]中国农业大学中国农业水问题研究中心,北京100083 [4]北京师范大学水科学研究院,北京100875 [5]城市水循环与海绵城市技术北京市重点实验室,北京100875 [6]清华大学地球系统科学系,北京100084

出　　处：《水利学报》2025年第2期216-226,239,共12页Journal of Hydraulic Engineering

基　　金：国家自然科学基金项目(32361143871,52239002);中央高校基本科研业务费专项项目(2024RC033)。

摘　　要：河川径流作为衡量区域水资源量的重要指标,其长期变化呈现出显著的时空变异特征。既往研究多针对实测径流且未剔除短期因素的影响。为揭示天然径流的变化规律,并探究其影响机制,本研究选择1979—2018年我国天然径流格点数据集与区域地面气象要素驱动数据集,针对全国218个三级流域,应用时间序列分解与传统趋势分析相结合的方法,对天然径流序列进行STL分解后分析其时空变化趋势,利用多元线性回归探究天然径流受气候变化影响的主要因素。结果表明:在218个三级流域中,原始天然径流序列呈显著变化趋势的比例仅为35.8%,经STL分解后的天然径流序列更能展现出变化趋势的显著性,趋势显著的流域占比高达75.7%;全国84.4%的三级流域天然径流呈增加趋势,且显著性明显;降雨是影响我国大多数流域天然径流变化的最主要驱动因素,长波辐射和比湿次之;北方流域的影响因素较之南方更为复杂。本研究为变化环境下全面认识我国径流规律特征提供参考。As a key indicator of water resources,river runoff exhibits considerable spatio-temporal variability in its long-term changes.Previous research has primarily focused on measured runoff without excluding the impacts of short-term factors.This article applied a method that combines time series decomposition with traditional trend analysis,based on natural hydrological processes undisturbed by human activities,to discern the dynamics of natural river runoff and explore its influencing mechanisms.Utilizing a gridded natural runoff and regional meteorological driving datasets,this study examined the spatial distribution and trends of natural runoff across 218 tertiary basins in China from 1979 to 2018.The analysis employed the Seasonal-Trend decomposition procedure based on Loess(STL)for decomposing natural runoff series.Additionally,multiple linear regression was employed to identify the primary meteorological factors affecting natural runoff.The findings revealed that,prior to STL decomposition,only 35.8%of the 218 tertiary basins exhibited a significant trend in natural runoff.Post-STL decomposition,however,the changing trend became more obvious,with 75.7%of the basins showing a significant tendency.Following the STL decomposition,84.4%of the tertiary basins demonstrated an increasing trend in natural runoff,with high statistical significance.Precipitation emerged as the most important factor influencing natural runoff variations in most basins,followed by long-wave radiation and specific humidity.The influencing factors varied in complexity between northern and southern China,with the northern basins exhibiting more intricate influencing factors.This study provides reference for comprehensively understanding the characteristics of runoff patterns in China under changing environments.

关 键 词：天然径流 三级流域 STL分解 径流趋势 影响因素 

分 类 号：TV121[水利工程—水文学及水资源]