| 基于冗余分析与正定矩阵因子模型的水质特征与流域污染源解析 |
| Water Quality Characterization and Pollution Source Identification in the Minjiang River Based on Redundancy Analysis and Receptor Model |
| 投稿时间：2024-03-15 修订日期：2024-04-15 |
| DOI：10.15928/j.1674-3075.202403150076 |
| 中文关键词:水环境 流域污染 PMF模型 冗余分析 源解析 |
| 英文关键词:water environment watershed pollution positive matrix factorization (PMF) model redundancy analysis source identification |
| 基金项目:国家自然科学基金面上项目(No.52170104, 51979237)；四川省科技厅自然科学基金项目(23NSFSC0838)；长江生态环境保护修复联合研究二期项目（2022-LHYJ-02-0509-10）。 |
|
| 摘要点击次数: 434 |
| 全文下载次数: 57 |
| 中文摘要: |
| 精准识别水体潜在污染源, 为流域水环境污染防治提供一种新的污染溯源方法和思路。基于地表水水质数据与土地利用指标, 利用冗余分析（RDA）阐明水质对土地利用的响应机制, 并将此作为辅助信息优化正定矩阵因子（PMF）模型污染源解析过程。结果表明, 研究区以氮磷和有机污染为主, 各水质指标间存在不同程度关联性；土地利用指标对水质指标的影响方式和强度不同, 耕地、建设用地、人口密度、化肥施用量和单位面积工业GDP表现为对水质不利的因素, 林地和草地表现为对水质保护有利的因素；污染源贡献依次为企业点源污染排放（23.13%）>农业面源污染（18.71%）>季节因素（16.67%）>生活污水排放（15.56%）>城市面源污染（15.26%）>自然源（10.67%）。 |
| 英文摘要: |
| Accurate identification of water pollution sources is a prerequisite for effective pollution control and sustainable watershed management. In this study, the relatively polluted Meishan section of the middle Minjiang River was selected for research. We elucidated the response mechanism of water quality to land use based on surface water quality data at 11 monitoring sites in the Meishan section and the distribution of each land use type in 2019 using redundancy analysis (RDA). The response mechanism of water quality to land use was then used as auxiliary information to optimize the pollution source analysis process of the positive matrix factorization (PMF) model. The study area was primarily affected by nitrogen, phosphorus, and organic pollution, and there were varying degrees of correlation among different water quality parameters. The impact of land use indicators on water quality parameters varied in mode and intensity. Cultivated land, construction land, population density, chemical fertilizer application, and industrial GDP per unit area were factors detrimental to water quality, while forest and grassland were beneficial for water quality. The identification and quantitative analysis of the pollution sources was completed by combining RDA and PME model analysis, and the results shows that the contribution rates of different pollution sources were as follows: industrial point source pollution (23.13%) > agricultural non-point source pollution (18.71%) > seasonal factors (16.67%) > domestic sewage discharge (15.56%) > urban non-point source pollution (15.26%) > natural sources (10.67%). The primary contributing indicators of industrial point source pollution were F- (48.70%), TN (45.42%), and CODCr (36.52%). The primary contributing indicators of agricultural non-point source pollution were NH3-N (65.99%), TN (25.92%), and TP (22.70%). Organic pollution-related parameters such as CODMn, CODCr, BOD5, and petroleum were the main loading indicators for domestic sewage and urban non-point source pollution. The primary contributing indicator for natural sources was As, with a contribution rate exceeding 60%, but the average concentration was low and the overall impact of natural sources was relatively small. This study provides a new method and approach for tracing pollution sources in watershed water pollution prevention and control. |
| 任兴念,陈斯恺,郭珊珊,高东东,王 春,张 涵.2024.基于冗余分析与正定矩阵因子模型的水质特征与流域污染源解析[J].水生态学杂志,45(5):142-150. |
| 查看全文 查看/发表评论 下载PDF阅读器 HTML |
|
|
|
