两种吸附模型对阿特拉津在壤质砂土中的模拟效果分析  

Analysis of simulated migration of atrazine in a type of loamy sand based on two adsorption models

作　　者：李晓宇 任仲宇[1,2] 李芳春 井琦 魏明海[4] LI Xiao-yu;REN Zhong-yu;LI Fang-chun;JING Qi;WEI Ming-hai(College of Architectural and Civil Engineering,Beijing University of Technology,Beijing 100124,China;Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering,Beijing 100124,China;China National Knowledge Infrastructure,Beijing 100192,China;Chinese Academy of Environmental Planning,Beijing 100012,China)

机构地区：[1]北京工业大学建筑工程学院,北京100124 [2]水质科学与水环境恢复工程北京市重点实验室,北京100124 [3]同方知网(北京)技术有限公司,北京100192 [4]生态环境部环境规划院,北京100012

出　　处：《农业环境科学学报》2020年第1期191-200,共10页Journal of Agro-Environment Science

基　　金：国家重点研发计划项目(2016YFC0401404)~~

摘　　要：研究阿特拉津在土壤中的迁移转化规律,为阿特拉津的地下水污染防治提供依据。以阿特拉津为研究对象,构建室内一维土柱试验模拟阿特拉津在连续注入和清水淋洗条件下,在饱和壤质砂土中的迁移过程;基于单点非平衡吸附模型和吸附-解吸模型,采用HYDRUS-1D软件模拟了阿特拉津在土柱中的迁移过程。结果表明:整体调参拟合下,单点非平衡吸附模型模拟阿特拉津连续注入饱和壤质砂土的Nash-Suttcliffes模拟效果系数(NSCs)为0.909~0.922,均方根误差(RMSEs)为2.752~3.167,而淋洗条件下的NSCs为负值,RMSEs为8.703~12.424;吸附-解吸模型模拟阿特拉津连续注入、淋洗过程中,在饱和壤质砂土的NSCs为0.901~0.954,RMSEs为2.037~3.289。分段调参拟合后,单点非平衡吸附模型模拟阿特拉津连续注入饱和壤质砂土的NSCs提高到0.919~0.941,RMSEs减小到2.405~2.986,淋洗条件下的NSCs提高到0.927~0.940,RMSEs减小到2.036~2.309;吸附-解吸模型模拟阿特拉津连续注入、淋洗过程中,在饱和壤质砂土中的NSCs提高到0.904~0.956,RMSEs减小到2.037~3.247。研究表明:整体调参下,采用单点非平衡吸附模型模拟阿特拉津连续注入饱和壤质砂土时拟合度高,而淋洗条件下拟合度低;吸附-解吸模型在模拟阿特拉津连续注入、淋洗过程时拟合度均较高;分段调参可提高单点非平衡吸附模型模拟阿特拉津在淋洗阶段迁移的拟合度。To better understand the migration of atrazine in soil and provide foundations for groundwater pollution prevention, a one-dimensional soil column experiment was constructed to simulate the migration of atrazine in a type of saturated loamy sand under the conditions of consecutive atrazine injection and soil column leaching with deionized water, respectively. The software HYDRUS-1 D was used to simulate the migration process of atrazine in the soil column based on the one-site non-equilibrium adsorption model and the adsorption-desorption model. The results indicated that the simulation quality was influenced by the fitting methods. In the case of overall parameter fitting, the Nash–Suttcliffe coefficients(NSCs)and root mean squared errors(RMSEs)of the one-site non-equilibrium adsorption model were0.909~0.922 and 2.752~3.167, respectively, for the simulation of the consecutive injection of atrazine;under the leaching condition, the NSCs of the simulation were negative and the RMSEs were 8.703~12.424. Under both experimental conditions, the NSCs and RMSEs of the adsorption-desorption simulation were 0.901~0.954 and 2.037~3.289, respectively. By using segmental parameter fitting for the one-site non-equilibrium adsorption model simulation, the NSCs improved to 0.919~0.941 while the RMSEs decreased to 2.405~2.986 for the consecutive injection of atrazine, and the NSCs improved to 0.927~0.940 while the RMSEs decreased to 2.036~2.309 for leaching. In the adsorption-desorption model simulation, the NSCs improved to 0.904~0.956 and RMSEs decreased to 2.037~3.247 for both processes. The fitting results indicated that in the case of overall parameter fitting, the one-site non-equilibrium adsorption model demonstrated a good fitting performance in simulating the migration of atrazine under the consecutive injection conditions but a poor fitting performance under leaching conditions, whereas the adsorption-desorption model could effectively simulate the migration of atrazine in saturated loamy sand under both experimental

关 键 词：阿特拉津 吸附 解吸 HYDRUS-1D 模拟 

分 类 号：X592[环境科学与工程—环境工程]