文章摘要
梁丽营,高振刚,刘德财,闫小雨,曾鸿鹄,刘峥,覃礼堂,梁延鹏.AnnAGNPS模型在西南岩溶地区奇峰河流域的参数敏感性及适用性分析[J].农业环境科学学报,2020,39(3):590-600.
AnnAGNPS模型在西南岩溶地区奇峰河流域的参数敏感性及适用性分析
Parameter sensitivity and applicability analysis of AnnAGNPS model in Qifeng River watershed in the southwest Karst area of China
投稿时间:2019-10-10  
DOI:10.11654/jaes.2019-1111
中文关键词: 西南岩溶地区  AnnAGNPS模型  参数敏感性  适用性评价  径流模拟
英文关键词: Southwest Karst area  AnnAGNPS model  parameter sensitivity  applicability evaluation  runoff simulation
基金项目:国家自然科学基金项目(51578171)
作者单位E-mail
梁丽营 桂林理工大学环境科学与工程学院, 广西 桂林 541006  
高振刚 桂林理工大学环境科学与工程学院, 广西 桂林 541006  
刘德财 桂林理工大学环境科学与工程学院, 广西 桂林 541006  
闫小雨 桂林理工大学环境科学与工程学院, 广西 桂林 541006  
曾鸿鹄 桂林理工大学环境科学与工程学院, 广西 桂林 541006
桂林理工大学, 岩溶地区水污染控制与用水安全保障协同创新中心, 广西 桂林 541006 
zenghonghu@glut.edu.cn 
刘峥 桂林理工大学化学与生物工程, 广西 桂林 541006 1986034@glut.edu.cn 
覃礼堂 桂林理工大学环境科学与工程学院, 广西 桂林 541006  
梁延鹏 桂林理工大学环境科学与工程学院, 广西 桂林 541006  
摘要点击次数: 2019
全文下载次数: 1673
中文摘要:
      为了探究农业非点源模型(AnnAGNPS)在西南岩溶地区奇峰河小流域地表径流量模拟的适用性,使用差分灵敏度分析(Differential sensitivity analysis,DSA)方法对参数进行敏感性分析,采用敏感指数法评价参数的敏感性等级;利用试错法人工调整模型参数,采用三个评价指标:决定系数(R2)、纳什系数(ENs)和相对偏差(Re)评估模型的性能,检验其在岩溶地区奇峰河小流域的径流模拟的适用性。结果表明:不同临界源区面积(Critical source area,CSA)与最小源区沟道长度(Minimum source channel length,MSCL)取值下,CSA=20 hm2,MSCL=200 m时,能够较为全面地描述奇峰河流域的下垫面的情况。参数敏感性分析结果表明,对径流模拟最敏感的参数是径流曲线数(CN),与非岩溶地区相比,影响地表径流的最敏感的参数相同,取值较非岩溶地区流域的大;对泥沙、总氮、总磷模拟最敏感的参数是土壤可侵蚀性因子与径流曲线数。模拟结果表明:AnnAGNPS模型能很好地模拟流域的地表径流,月尺度下模拟的R2为0.962,ENs为0.831,Re为-14.94%。日尺度下模拟的R2为0.716,ENs为0.946,Re为-13.69%。由于岩溶系统的高渗漏性以及地下水与地表水交换的瞬时性,日尺度下的模拟精度较低。研究表明经过校准与验证的AnnAGNPS模型适用于西南岩溶地区桂林市奇峰河流域的非点源污染负荷模拟。
英文摘要:
      To explore the applicability of the agricultural non-point source AnnAGNPS model in the Qifeng River small watershed in the southwest Karst area, a differential sensitivity analysis(DSA)method was used to analyze the sensitivity of the parameters. The sensitivity index method was used to evaluate the sensitivity level of the parameters. The parameters of the model were adjusted manually by a trialand-error method. The evaluation index, the coefficient(R2), the Nash coefficient (ENs), and the relative deviation(Re)were used to evaluate the performance of the model. The applicability of the model to the Qifeng River small watershed in the Karst area was tested. When the critical source area(CSA)and minimum source channel length(MSCL)values were varied, a CSA of 20 hm2 and MSCL of 200 m were found to comprehensively describe the situation in the Qifeng River Basin. The results of the parameter sensitivity analysis showed that the most sensitive parameter for runoff simulation was the number of runoff curves(CN). Compared with non-Karst areas, the most sensitive parameters affecting surface runoff were the same, although the values were larger than those in non-Karst areas. The most sensitive parameter for sediment, total nitrogen, and total phosphorus simulation was soil erosion factor. The simulation results showed that the AnnAGNPS model could simulate the surface runoff of the Qifeng watershed. The R2, ENs, and Re of the simulation of runoff on the monthly scale were 0.962, 0.831, and -14.94%, respectively, but 0.716, 0.946, and -13.69%, respectively, on the daily scale. Owing to the high leakage of the Karst system and the instantaneous exchange of groundwater and surface water, the simulation accuracy at the daily scale was low. The study indicates that the AnnAGNPS model is suitable for non-point source pollution load simulation in the Qifeng River Basin of Guilin City in the southwest Karst area.
HTML    查看全文   查看/发表评论  下载PDF阅读器