第一作者:Teofana Chonova
通讯作者:Teofana Chonova,Heinz Singer
通讯单位:瑞士联邦水产科学与技术研究所
DOI:10.1016/j.watres.2024.122745
尽管为改善河流水质做出了巨大努力,但化学污染仍然是一个重要问题。除了众所周知的污染物外,近年来,工业源污染物也日益受到关注,因为人们对其发生、归宿和环境风险的认识存在巨大差距。此外,这类污染物通常会随着时间的推移而出现高浓度波动,这使得它们的可预测性和可测量性较低。
本研究基于来自单一地点的时间性高频 LC-HRMS 监测数据,深入分析了莱茵河化学污染的不同来源。一项新开发的优先排序策略选出了近 3000 种潜在的主要污染物。基于时间行为的新型分类分析确定了其中 53% 的化合物(占数据集中记录的时间积分强度的 62%)来自非正常排放源。非正常排放可能来自工业生产周期。在对其他潜在的非正常排放源进行界定后,我们有确凿证据表明,相当一部分非正常排放可能来自工业活动。对 16 种不规则排放物质的结构阐释证实了这一结论,并成功确认了这些物质的工业来源。这些化合物包括 3-氯-5-(三氟甲基)吡啶-2-羧酸和 4-(二甲基氨基)-2,2-二苯基戊腈。此外,还有 40 种化合物的时间排放模式与这 16 种工业化合物相似,这有力地说明了存在共同的污染源。最后,选出了 100 种排名靠前的化合物,用于进一步的结构阐释和减排措施。本研究中概述的计算方法可有效地应用于其他大型河流流域,以识别来自工业污染源的未知污染物。
Fig. 1. The Rhine River sampling station at Basel (grey circle) with outlines for the river catchment and subcatchment at Basel. The land cover types and the WWTPs (Capacity – population > 5 000) within the catchment are also shown. Please note that time patterns of contaminants discharged upstream from large lakes may be smoothed out, potentially masking emission dynamics of chemical pollutants.
Fig. 2. (A) Barplots visualizing percentage of profiles that meet priority criteria in terms of profile numbers (top) and time-integrated intensities (bottom); (B) Contribution of single profiles to the time-integrated intensity in the dataset; (C) Horizontal barplots presenting total number of profiles that meet each of the priority criteria (set size) and vertical barplots presenting number of profiles that meet a single or several priority criteria (intersection size).
Fig. 3. (A) Barplots visualizing number (counts) and time-integrated intensities (intens) of prioritized profiles that were classified as belonging to the constant and irregular emission type, respectively. (B) t-SNE (t-distributed Stochastic Neighbor Embedding) plot of prioritized profiles, showing relationships between profiles regarding their emissions dynamics and field of application. Each symbol represents the time-profile of one contaminant (PPP = plant protection products, PPTP = transformation products of plant protection products, DPPO = diphenylphosphine oxide, TPPO = triphenylphosphine oxide, DEET = N,N-Diethyl-meta-toluamide, MCPA = 2-methyl-4-chlorophenoxyacetic acid). The bigger symbols represent the application of the known target compounds.
Fig. 4. Examples of newly identified substances and non-targets with similar temporal patterns. Non-target stands for profiles of unknown substances.
在这项研究中,对莱茵河畔魏尔监测站的长期、每日 LC-HRMS 测量结果进行了分析,以确定污染物并阐明其来源。结果表明,结合使用数据科学工具,对单个地点进行时间性高频监测,可以全面了解人为污染的潜在来源。这为河流监测和管理提供了关键信息。这些成果不仅与莱茵河的治理有关,也与其他承受类似人类压力的河流有关,如欧洲的多瑙河、美国的密西西比河和哈德逊河,或中国的长江和黄河。莱茵河中新检测到的化合物可能与其他具有类似重工业活动的流域相关。所建议的化合物优先排序和确定化学污染主要来源的策略很容易适用于其他类似的集水区,并可满足制定管理措施以减少工业污染的迫切需要。
Teofana Chonova, Steffen Ruppe, Ingrid Langlois, Dorrit S. Griesshaber, Martin Loos, Mark Honti, Kathrin Fenner, Heinz Singer, Unveiling industrial emissions in a large European river: Insights from data mining of high-frequency measurements, Water Research, 2025, https://doi.org/10.1016/j.watres.2024.122745
声明:本公众号仅分享前沿学术成果,无商业用途。如涉及侵权,请立刻联系公众号后台或发送邮件,我们将及时修改或删除!
邮箱:Environ2022@163.com
欢迎大家将《水处理文献速递》加为星标
即时获取前沿学术成果
若有帮助,请点击“在看”分享!
投稿、转载请扫描下方二维码联系小编吧