土地利用变化对赣江流域径流的影响研究

论文以赣江流域为研究区,基于SWAT分布式水文模型及SUFI-2 算法,利用流域DEM、土地利用、土壤、气象等数据,结合GIS和RS技术,对流域土地利用变化的径流响应进行定量分析与研究。结果表明:影响模型模拟效果前3 位的敏感参数分别为ALPHA_BF、CN2、ESCO,得出率定期和验证期的模拟值与实测流量过程拟合程度较优,其R²和Nash-Sutcliffe 效率系数E_{NS均高于0.90,相对误差|Re|均小于3%;流域内1990 年与2000 年土地利用均以林地、水稻田、草地为主,此3 种类型约占流域总面积的85%;在相同气象条件下,两期土地利用情景下的月、年均径流量变化趋势较为一致,2000 年土地利用情景下的年径流量比1990 年略大;相比1990 年土地利用情景,2000 年产水量增加最为显著的区域主要为赣江流域东部、东南部和西南部,建筑用地、水浇地和裸岩面积的增加、林地面积的减少可能是引起产水量增加的主要因素。}

Study on Runoff Responses to Land Use Change in Ganjiang Basin

Based on SWAT Model and SUFI-2 algorithm, combined with GIS and RS technology, runoff responses to land use change in Ganjiang Basin were analyzed quantitatively using various data derived from watershed DEM, land use, soil and meteorology in this paper. It was concluded that the top three sensitive parameters influencing model were ALPHA_BF, CN2 and ESCO, respectively. The simulation values of calibration and validation periods were in accordance with actual values, R² and Nash-Sutcliffe efficiency coefficient E_{NS were higher than 0.90, the relative errors |Re| were all less than 3%. Forestland, paddyland and grassland took about 85% of the total area of Ganjiang Basin from 1990 to 2000. In the same climate condition, monthly, annual runoff manifest obviously consistent trends under two different land use scenarios, the annual runoff in 2000 is slightly larger than that of 1990 on the whole. Compared with 1990, runoff of the subbasins increases significantly in the east, southeast, southwest region in 2000. The growth of development land, irrigated land and bare rock, reduction of forest area are likely the main factors causing the increase of water production.}