国家自然科学基金项目（51779245, U19A2050, 41961124004）；江苏省水利科技项目（2018051）
1.College of Agricultural Sciences and Engineering, Hohai University;2.Changshu Agro-eclological National Field Scientific Observation and Research Station, Institute of Soil Science, Chinese Academy of Sciences
National Natural Science Foundation of China (51779245, U19A2050, 41961124004); Water Science and Technology Project of Jiangsu Province (2018051).
空间尺度上，种植业面源污染负荷估算可用分为田块尺度、流域尺度和区域尺度。国家尺度属于区域尺度范畴，是掌握整个国家种植业源污染负荷概况、检验种植业面源污染防治措施成效、预测面源污染发展趋势的重要尺度。本文综述了国家尺度上种植业面源污染负荷估算的研究方法，主要包括输出系数法、改进输出系数法、多元回归法、贝叶斯递归回归树模型、过程模型模拟法等，并利用以上方法对我国种植业面源污染负荷进行估算。其中氮素径流损失估算结果在0.30~2.40 Tg之间，氮素淋洗损失为0.36~2.03 Tg，磷素径流损失为3.5~6.37 Mg。指出了我国国家尺度种植业面源污染流失量估算方法存在以下主要问题：源头排放未区分背景排放和肥料排放，也未考虑南方一年多熟制排放差异，过程削减未充分考虑沿程消纳。为此，提出我国国家尺度种植业面源污染负荷估算方法应进一步明细各土地利用输出系数、区分背景排放和肥料排放，考虑我国熟制的区域差异、考虑田块到流域出口的沿程削减。
At the spatial scale, non-point source pollution load estimation of planting system can be divided into field scale, watershed scale, and regional scale. The national scale belongs to regional scale, which is an important scale for understanding the general situation of non-point source pollution from the planting system in the entire country, testing the effectiveness of non-point source pollution control measures, and predicting the trend of non-point source pollution. This paper reviews the research methods of non-point source pollution load estimation from planting system at the national scale. These methods mainly include export coefficient method, improved export coefficient method, multiple regression method, Bayesian recursive regression tree, and progress module. Then the estimated results of these methods were compared. The estimated nitrogen runoff was between 0.30-2.40 Tg, nitrogen leaching was between 0.36-2.03 Tg, and phosphorus runoff was between 3.5-6.37 Mg in planting system of China. There are two main problems existed: 1) for source emissions process, these methods do not distinguish between background emissions and fertilizer emissions, and between multiple crops a year in South China; 2) for retention process, these models do not fully consider removal process along the pathway. Finally, it is proposed that the non-point source pollution load estimation for planting system at national scale should further detail the output coefficients of each land use, distinguish between background emissions and fertilizer emissions, take into account the regional differences in ripening system, and consider removals along the pathway from field to watershed outlet.