基于SWAT模型的涪江流域土地利用变化的水沙响应研究OA北大核心CSTPCD
Response of Runoff and Sediment Yields to Land Use Change in Fu River Watershed Based on SWAT Model
[目的]分析涪江流域水沙对土地利用类型变化的响应情况,为流域水土资源规划管理和生态环境建设提供科学依据.[方法]应用SWAT分布式水文模型,采用情景模拟方法构建3期自然情景以及3种极端情景,分析流域1990-2010年的土地利用变化情况以及量化不同土地利用类型对流域水沙的作用程度.[结果](1)1999-2010年,流域上游以林草地为主,约占流域总面积的46％;中下游主要为农业用地和居民用地,耕地占流域总面积的51％左右;居民用地与未利用地扩张率分别为95.86％和434.43％.自然情景下,流域径流变化率小于5％,侵蚀产沙变化率小于15％.(2)SWAT模型在涪江流域具有较强的适用性,径流率定期和验证期的R2和NSE分别为0.86,0.85;0.80,0.78,泥沙分别为0.68,0.67;0.68,0.43.情景模拟分析得出具体土地利用类型对径流的影响程度:林地＞草地＞未利用地,相对于基准期,林地覆被情景减少了 35.66％的产流量.对流域侵蚀产沙的影响程度:未利用地＞林地＞草地,未利用地情景下增加了 56.74％的侵蚀产沙量,林地和草地情景分别减少了 43.57％和30.04％.(3)基于随机森林算法的自变量影响程度重要性排序结果表明:影响产流量的前3个土地利用类型是耕地、林地、草地,权重系数分别为2.057,2.030,1.148;影响侵蚀产沙量的是草地、耕地、林地,权重系数分别为1.753,1.208,1.003.[结论]除耕地外,径流对林地变化的响应最显著,侵蚀产沙量对草地变化的响应最显著,当土地利用类型向林草地转化时,对流域具有显著减水减沙效果.该研究可为涪江流域未来土地利用结构优化和水土资源的合理规划提供科学依据.
[Objective]The aims of this study are to quantitatively analyze the response of runoff and sediment yields to land use change in Fu River Watershed,and to provide scientific evidences for the planning and management of water and soil resources and ecological construction in the river basins.[Methods]The SWAT distributed hydrological model was applied.Three natural scenarios and three extreme scenarios were set up to qualify the variation of land use change during 1990 and 2010 and corresponding contribution rates to watershed runoff and sediment discharge.[Results](1)From 1999 to 2010,the upper reaches of the watershed was dominated by woodland and grassland,comprising approximately 46％of the total watershed area.The middle and lower reaches of the watershed were primarily composed of agricultural and urban areas,with cultivated land accounting for approximately 51％of the total area.As a result of rapid urbanization,urban land and bare land expanded significantly at rates of 95.86％and 434.43％,respectively.Under the natural scenarios,the change rate of water yield was less than 5％,and the change rate of sediment yield is less than 15％.(2)SWAT model had strong applicability in Fu River Watershed,the R2 and NSE values for runoff at calibration and verification periods were 0.86,0.85 and 0.80,0.78,respectively,and for sediment discharge were 0.68,0.67 and 0.68,0.43,respectively.Scenario simulation analyses illustrated the effects of specific land use types on runoff decreased in the order:woodland＞grassland＞bare land.Compared with the base period,the woodland cover scenario reduced runoff by 35.66％.Impact of different land use types on sediment yields decreased in the order:bare land＞woodland＞grassland.Compared with the base period,the bare land scenario increased the sediment yield by 56.74％,and the woodland and grassland decreased sediment yield by 43.57％and 30.04％,respectively.(3)The importance ranking results of independent variables based on random forest algorithm showed that the first three land use types affecting runoff were farmland,woodland and grassland,and the random forest weights were 2.057,2.030 and 1.148,respectively,while grassland,farmland and woodland affected sediment yield most significantly and the random forest weights were 1.753,1.208 and 1.003,respectively.[Conclusion]In addition to farmland,the response of runoff to forestland change is the most significant,and the response of erosion and sediment yield to grassland change is the most significant.The conversion of land use type to grassland and woodland significantly reduces water and sediment yields in the basin,while the bare land has a significant impact on sediment yield in the watershed.This research can provide scientific basis for future land use structure optimization and rational planning of water and soil resources in the Fu River Watershed.
刘雪妍;张会兰;罗泽宇;张嘉琪;安妮
北京林业大学水土保持学院,北京 100083北京林业大学水土保持学院,北京 100083||重庆缙云山三峡库区森林生态系统国家定位观测研究站,重庆 400711
地球科学
土地利用变化SWAT模型情景分析随机森林算法涪江流域
land use changeSWAT modelscenario analysisRandom ForestFu River Watershed
《水土保持研究》 2024 (003)
79-89,100 / 12
国家自然科学基金项目(52279056);"十四五"重点研发项目(2022YFF130009505)
评论