摘要
以闽江流域水口水库(SR)及上游来流(T1~T3)为研究对象,采用贝叶斯同位素混合模型(MixSIAR),通过传统硝酸氮氮氧稳定同位素和构建的水氢氧稳定同位素-硝酸氮浓度两种方法对水库非汛期和汛期上游来流来水的硝酸氮贡献进行溯源分析,采用完全混合模型和MixSIAR的偏差信息准则(DIC)值对两种方法结果进行验证.结果表明:区域非汛期硝酸氮浓度高于汛期,上游来流沙溪T3毗邻工业区,周边建设用地比例最高,具有最大的硝酸氮污染.汛期T3和SR的δ^(18)O-NO_(3)^(-)和所有点位δ^(18)O-H_(2)O、δD-H_(2)O均出现降低,表明水氢氧同位素低的降雨携带低δ^(18)O-NO_(3)^(-)的污染源进入水体.两种溯源方法在汛期溯源结果相当,最大误差仅为13.04%,但非汛期溯源结果存在显著差异,0分馏条件下硝酸氮氮氧稳定同位素结果为:T2(49.4%)>T3(26.1%)>T1(24.5%);水氢氧稳定同位素-硝酸氮浓度为:T3(44.8%)>T1(29.6%)>T2(25.6%).本研究构建的氢氧稳定同位素-硝酸氮浓度方法DIC均低于不同分馏条件下传统硝酸氮氮氧稳定同位素方法,并采用完全混合模型验证水库的硝酸氮浓度误差在汛期和非汛期仅为1.22%和0.89%,表明新建的溯源方法具有最高的可信度,同时该方法可推广至其他污染物的溯源分析,为流域水量、水质混合及污染物流域管控提供了一种解决方法.
In this study,the ShuiKou Reservoir(SR)in the Min River Basin and its upstream inflows(T1~T3)were selected as the research area.The Bayesian Isotope Mixing Model(MixSIAR)was employed to conduct source apportionment of nitrate contributions during both non-flood and flood seasons.Two approaches were implemented:conventional nitrate nitrogen-oxygen isotopes and a newly developed"water hydrogen-oxygen stable isotope-nitrate concentration"method.The results from both methods were validated through a complete mixing model and the deviation information criterion(DIC)values from MixSIAR.The findings revealed that nitrate nitrogen concentrations in non-flood seasons exceeded those in flood seasons.Among the upstream inflows,the Shaxi River(T3),which is adjacent to an industrial area and has the highest proportion of construction land in its vicinity,demonstrated the most severe nitrate pollution.During flood seasons,a decline ofδ^(18)O-NO_(3)^(-)values in T3and SR were observed,alongside a reduction in both waterδ^(18)O-H_(2)O andδD-H_(2)O values within the entire research area.The results indicated that rainfall with lowerδ^(18)O-H_(2)O andδD-H_(2)O signatures brought pollutants with lowerδ^(18)O-NO_(3)^(-)to the research area.Secondly,both methods showed comparable source apportionment results during flood seasons(maximum error 13.04%),but exhibited significant discrepancies in non-flood seasons.Conventional isotope analysis under 0fractionation yielded contributions:T2(49.4%)>T3(26.1%)>T1(24.5%),whereas the new method indicated:T3(44.8%)>T1(29.6%)>T2(25.6%).Thirdly,the novel method demonstrated superior reliability,with lower DIC than those obtained using conventional methods under different fractionation.The nitrate concentration errors validated by the complete mixing model at merely 1.22%(flood season)and 0.89%(non-flood season).Those further confirmed its highest credibility.This methodology shows potential for extension to other pollutant source tracking studies,providing solution for watershed water quantity-quality integration and pollution control management.
作者
魏健
陈莉莉
陈君君
谢蓉蓉
李家兵
马晓丹
吴琼瑶
石成春
WEI Jian;CHEN Li-li;CHEN Jun-jun;XIE Rong-rong;LI Jia-bing;MA Xiao-dan;WU Qiong-yao;SHI Cheng-chun(Carbon Neutral Modern Industry College,College of Environmental Science and Engineering,Fujian Normal University,Fuzhou 350007,China;Fujian Provincial Key Laboratory of Environmental Engineering,Fujian Academy of Environmental Sciences,Fuzhou 350013,China;Fujian Provincial Key Laboratory of Pollution Control and Resource Recycling,Digital Fujian Internet-of-things Laboratory of Environmental Monitoring,Fujian Normal University,Fuzhou 350007,China)
出处
《中国环境科学》
2026年第3期1543-1553,共11页
China Environmental Science
基金
福建省公益类科研院所专项(2023R1014001)。
