清江流域长阳段非点源污染源分析及氮磷流失特征研究

发布时间：2018-03-07 18:37

本文选题：SWAT模型　切入点：氮磷流失　出处：《武汉大学》2017年硕士论文　论文类型：学位论文

【摘要】：近年来随着经济的迅猛发展,非点源污染的问题日益凸现出来,产生了许多水环境问题。本文通过在长阳地区展开的污染源调查,确定了流域内氮磷输入大部分来自农业施肥、畜禽养殖以及水产养殖。农业施肥氮磷输入量达到了总氮9781t、总磷3818t,畜禽养殖总氮4381t,总磷1613t,但这部分氮磷在进入水体的过程中会有大幅削减,因此进入水体的数量远不及此;而第三大氮磷输入源水产养殖产生的813t总氮与133t总磷则直接输入水体。本文利用SWAT模型分析了清江长阳段的非点源污染特征。利用2007-2013年隔河岩水库出库流量资料与2011-2013年猫子滩断面水质资料来进行率定与验证。经过率定验证,猫子滩断面的径流、总氮、氨氮、总磷四项指标的率定期与验证期的决定系数R2均在0.6以上,Ens也均高于0.5,说明SWAT模型模拟结果良好,可以达到要求,在研究区域具有一定的适用性。SWAT模拟结果显示:1)研究区域的氮磷流失的高峰时期与降雨时期基本一致,主要出现在4-9月,且其强度在空间上呈现与耕地分布一致的特征。2)支流流域总磷污染负荷呈现输入虽大,输出却小的特征。3)总氮流失形式以硝氮为主;总磷流失形式以吸附态磷为主。4)农业施肥对总氮污染的贡献率最高,占比约36%;水产养殖则对总磷污染的贡献率最高,占比超过50%。年降雨量-氮磷年入河模数的回归分析结果显示,不论全流域尺度还是子流域尺度,氮磷入河模数均和年降水正相关关系十分明显。各子流域年降雨量-总氮、总磷年入河模数回归方程决定系数R2绝大多数高于0.6。而经过多元回归分析后,发现回归方程的斜率K(流失强度系数)、截距B(修正项)与耕地占比及不同土壤类型占比相关性较大。在对流域两种主要土壤(黄棕壤、石灰岩土)为主的子流域分别进行单独分析后,发现区分土壤后得到的回归方程斜率K、截距B与耕地占比之间线性关系的决定系数R2相比不区分土壤时整体上看得到了提升,总氮对应的K、B,其R2均从0.6左右上升到0.9以上;总磷对应的K、B其R2整体上亦有上升,但上升幅度与土壤有关。通过对石灰岩土与黄棕壤的吸附态、溶解态污染物流失能力的对比分析,初步解释了不同土壤对应的氮磷流失强度系数K与耕地比例之间线性方程形式不同的原因,并发现土壤的溶解态氮磷流失能力是影响耕地氮磷流失负荷强度的重要属性。
[Abstract]:In recent years, with the rapid development of economy, the problem of non-point source pollution has become more and more prominent, which has caused many problems of water environment. Through the investigation of pollution sources in Changyang area, it is determined that most of the input of nitrogen and phosphorus in the basin comes from agricultural fertilization. The input amount of nitrogen and phosphorus in agricultural fertilization reached 9781t, total phosphorus 3818t, total nitrogen 4381t, total phosphorus 1613t, but this part of nitrogen and phosphorus will be greatly reduced in the process of entering water body, so the amount of nitrogen and phosphorus entering water body is far less than this; However, 813t total nitrogen and 133t total phosphorus produced by aquaculture, the third largest nitrogen and phosphorus input source, were directly imported into water body. The characteristics of non-point source pollution in Changyang section of Qingjiang River were analyzed by using SWAT model. The discharge data of Geheyan Reservoir from 2007 to 2013 were used to analyze the characteristics of non-point source pollution. With the data of water quality of Maozi Beach section from 2011-2013 to carry out rate determination and verification. The ratio of runoff, total nitrogen, ammonia nitrogen and total phosphorus in the section of Maozi Beach were all higher than 0. 6 and the determining coefficient R2 of validation period were all above 0. 6, which indicated that the simulation results of SWAT model were good and could meet the requirements. The results of SWAT simulation show that the peak time of nitrogen and phosphorus loss in the study area is basically the same as the rainfall period, which mainly occurred in April-September. The intensity of total phosphorus pollution load in tributaries is large, but the output is small. 3) nitrate nitrogen is the main form of total nitrogen loss in the tributaries basin, and its intensity is consistent with the distribution of cultivated land. 2) the total phosphorus pollution load of the tributaries is large, but the output is small. The main form of total phosphorus loss is adsorbed phosphorus. 4) the contribution rate of agricultural fertilization to total nitrogen pollution is the highest, accounting for 36%, while aquaculture contributes the highest to total phosphorus pollution. The results of regression analysis of the annual rainfall, nitrogen and phosphorus entering the river show that there is a significant positive correlation between the nitrogen and phosphorus entering the river and the annual precipitation regardless of the scale of the whole basin or the subbasin. The determination coefficient R2 of the modulus regression equation of the annual total phosphorus is higher than 0.6. However, after multiple regression analysis, It is found that the slope K (loss intensity coefficient), intercept B (correction term) of the regression equation is highly correlated with the percentage of cultivated land and the proportion of different soil types. After separate analysis of the sub-watershed dominated by limestone soil, it was found that the slope of regression equation K, the determining coefficient R2 of linear relationship between intercept B and cultivated land ratio were improved when soil was not distinguished, as a whole, the slope of regression equation K and the coefficient R2 of linear relationship between intercept B and cultivated land ratio were obtained. The R ~ 2 of K _ (2) B corresponding to total nitrogen increased from 0.6 to more than 0.9, and that of K _ (2) to K _ (2) was also increased, but the increase was related to the soil. The adsorption state of limestone soil and yellow brown soil was obtained by means of the adsorption of K _ (2) from limestone soil to yellow brown soil. The comparative analysis of the loss capacity of dissolved pollutants preliminarily explained the different forms of linear equations between the nitrogen and phosphorus loss intensity coefficient K and the proportion of cultivated land in different soils. It was also found that the nitrogen and phosphorus loss capacity of the soil was an important attribute affecting the nitrogen and phosphorus loss load of cultivated land.
【学位授予单位】：武汉大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X522

【相似文献】