基于改进LAM模型的河流污染源解析方法与例证

River pollution source apportionment based on improved LAM modelling and its application

污染源精准解析是河流水环境治理的基础和关键技术.针对基于流量-浓度关系的LAM模型(负荷分摊模型)存在参数不确定性高的问题,本研究提出基于流量-通量关系的新型LAM源解析模型,并以广东省北江为案例,解析了2018年北江COD_(Mn)的污染来源特征.结果表明:(1)改进模型与原始模型的计算结果存在显著差异,基于改进模型的解析结果显示2018年北江COD_(Mn)主要来源于非点源,其负荷占比高达95.1%,点源仅占4.9%,而原始模型的结果为非点源占比100%;(2)改进后的LAM模型可实现拟合优度的提升,相对于原始模型,判定系数由0.89提升到0.92,主要参数的标准误差显著下降;(3)原始模型无法识别负荷占比较小的污染源,改进后的模型识别能力显著提升,结论与研究区实际污染源结构更相符.研究提出的改进模型与应用可为多源河流污染源精准解析提供方法和案例借鉴.

Accurate identification of pollution sources is the basis and one of the key technologies for efficient catchment management.Due to the high uncertainty of parameters and large analytical error of traditional LAM(Load Apportionment Model)in terms of flow-flux relationships,this study proposed a novel LAM source identification model based on flow-flux relationships,and took the Beijiang River,Guangdong Province as an example to identify the source characteristics of COD_(Mn)pollution in 2018.Our results show that:(1)Significant differences were observed between the results of the improved model and the original model.The improved model indicated that the COD_(Mn)in the Beijiang River in 2018 was mainly from non-point sources,accounting for 95.1%of the total load,whereas the original model suggested that non-point sources accounted for all the load.(2)The improved LAM model show better simulation results.Compared with the original model,the resolution coefficient increased from 0.89 to 0.92,and the standard error of the main parameters substantially reduced.(3)The original model could not identify the pollution sources with a relatively small load contribution,but the identification ability of the improved model increased significantly,and the conclusion was more consistent with the actual pollution sources in the study area.The proposed model and its application in this study can provide insights for accurate analysis of river multiple pollution sources.