我国作为农业大国，由农业活动引起的水环境问题已日趋严重，根据我国第一次污染普查结果显示，农业面源污染已成为我国水体健康面临的主要威胁，在我国的澜沧江流域，近些年来农业开发加剧，农业活动增多，化肥和农药的施用量明显增多。同时，因澜沧江流域纬度跨度大，上、下游海拔落差多，海拔差异会影响上、下游面源污染的空间分布规律，在一定程度上也会影响澜沧江流域水体健康的管理情况。因此，本研究以澜沧江流域为研究对象，利用SWAT模型模拟出澜沧江流域境内1977~2016年面源污染负荷的时空分布情况，基于SMAP卫星数据及水文监测数据对SWAT模拟结果进行率定与验证以保证模拟的准确性，然后根据模拟结果探究澜沧江流域农业面源污染的分布及变化规律，发现面源污染负荷对海拔的响应情况，最终做出考虑面源污染条件下案例研究区的水体健康指数评估。本研究的主要内容和结论如下： （1）建立SWAT模型数据库，输入DEM、土地利用、气象、土壤、农田管理以及水库的相关数据。开始模拟前，共将流域分成96个子流域，4170个水文响应单元。 （2）利用SWAT-CUP工具对流量模拟结果做出率定与验证，再基于SMAP卫星数据获取澜沧江流域土壤水含量的时空分布状况。经验证证实模拟数值贴合度高，模型模拟结果可靠。 （3）模型模拟结果表明：降雨量、产水量、总氮、总磷的变化趋势一致性比较高，随着降雨量的增加，产水量增加，流域出口处的面源污染负荷变大。比对国家水质标准，各时期流域水质均满足Ⅲ类的水质标准，水质的限制因子为总磷。流域出口处的总氮、总磷负荷量与泥沙负荷、径流量的年内月变化趋势大体一致。 有机氮、总氮、有机磷、总磷的空间分布及变化相似，初始高海拔区负荷量偏高，后续年份中呈减少趋势，逐渐可看出较明显的低海拔负荷量大于高海拔的分布规律；低海拔区硝态氮、可溶性磷负荷量一直明显高于高海拔区；矿物质磷负荷在高海拔区偏小，在时间尺度上高海拔区负荷量逐年减少，低海拔区负荷量逐年增加。 将澜沧江流域分成高、中、低海拔地区，海拔越高降雨量越少，泥沙负荷为低海拔>高海拔>中海拔；总氮、总磷的负荷均是低海拔>中海拔>高海拔，主要因为不同海拔下土壤侵蚀量、降雨量及农田管理措施的差异。另外，降雨量、泥沙、氮、磷负荷具有较为相似的年内变化趋势。 （4）探究不同海拔下两个案例子流域的时空分布变化规律，基本与流域整体规律相似，高海拔地区具有较高的入河系数。通过冗余分析（RDA）对不同海拔条件下的影响因子进行分析，高海拔区不同类别的面源污染及其入河系数与环境因子的相关性大小趋于一致；低海拔区环境因子的影响较为分散。不同海拔下面源污染主要的影响因子不同，影响程度存在较大差异。 （5）选取西双版纳州为例，采用淡水健康指数（FHI）法对研究区水体健康进行评估。西双版纳州境内澜沧江流域的水生态系统健康状况良好，应在继续保持水资源、水质等现有生态环境优势的前提下，继续防治农业面源污染物排放，做好水资源使用及保护规划，整体推进当地水生态系统管理工作的开展。

As a large agricultural country, water environmental problem caused by agricultural activities is becoming more and more serious. According to the national first pollution source survey, nonpoint source pollution has already become the main threat of freshwater health in China. Recently, agricultural development has intensified, agricultural activities have increased, and the amount of fertilizers and pesticides has also increased significantly in Lancang River Basin. The latitudinal span of Lancang River Basin is large, and the drop between upstream and downstream altitudes is high. The altitude difference will affect the spatial distribution of upstream and downstream nonpoint source pollution, then affect the management of the freshwater health of the basin. In this study, we get the temporal and spatial distribution of Lancang River Basin nonpoint source pollution in 1977~2016 with the SWAT model, which was based on Lancang River Basin. The results of SWAT simulation are calibrated and verified to ensure the accuracy of simulation, based on SMAP satellite data and hydrological monitoring data. Then, based on the results of simulation, the distribution and change rule of agricultural nonpoint source pollution in Lancang River Basin are explored. The response of nonpoint source pollution to altitude is found, and the freshwater health index assessment of selected study area is finally considered. The main contents and conclusions of this study are as follows: (1) The SWAT model database is established by preparing the DEM data, the land use data, the meteorological data, the soil type data, the farmland management data and the reservoir data. The model is simulated and divided the study area into 96 sub-watersheds and 4170 HRUs. (2) First, SWAT-CUP is used as a tool to calibrate and verify the simulation results in the study area. Based on SMAP satellite data, the time and space of soil water in the Lancang River Basin is obtained. It is verified that the simulation value is highly consistent with the monitoring value, and the simulation results are reliable. (3) The model simulation results show that, the variation trend of rainfall, water yield, TN and TP is very consistent. With the increase of rainfall, the water yield increases and the pollution load of the outlet of the river basin becomes larger. Comparing with the national water quality standard, it is found that the water quality of each catchment meets the standard of III, and the limiting factor of water quality is TP. In terms of monthly change, the TN and TP load at the outlet of the basin is basically the same as that of sediment load and runoff. The spatial distribution and change of OGRN, TP and ORGP, TP is similar. It is higher at high altitude and it shows a decreasing trend. It can be seen obviously that load at low altitude is larger than the high. The NO3-N and SOLP load at low altitude is obviously higher than the high; the load of SEDP is smaller at high altitude, and its load decreases but it increases at low altitude year by year. The Lancang River Basin is divided into high, middle and low altitudes. The higher the altitude is, the less the rainfall is, the sediment load rule is low> high > middle. The load of TN and TP is low > middle > high, because of the difference of soil erosion, rainfall and farmland management measures at different altitudes. And rainfall, sediment load, TN and TP load has similar trend in one year. (4) The variation of spatial and temporal distribution of the two selected sub-watershed from the high and low altitude, is basically similar to that of the whole basin. River load ratio is higher in high altitude area Based on redundancy analysis (RDA), we get that: the correlation between non-point source pollution, its river load ratio and environmental factors at high altitude is consistent.The influence of environmental factors in low altitude areas is more scattered. The main driving factors of source pollution under different altitudes are different, and the degree of influence is also quite different. (5) Finally, taking Xishuangbanna as an example, the freshwater health index (FHI) method is used to evaluate the freshwater health in the study area. It shows that the freshwater ecosystem in Lancang River Basin in Xishuangbanna is in good health. On the premise of maintaining the existing ecological environment advantages, such as water resources and water environment quality, Xishuangbanna should continue to prevent and control the discharge of agricultural nonpoint source pollutants, plan for the use and protection of water resources, and promote the management of the local freshwater ecosystem.