Scholarlink

<p>Chemical pollution is a serious threat to river ecosystems. A systemic understanding of chemical co-occurrence, emission dynamics and sources is required by the Water Framework Directive to mitigate freshwater pollution. Non-target screening is key for a holistic assessment of chemicals in river water, both known and as-yet-unknown. Here, we show that its combination with multivariate statistics and advection-reaction modeling enables reconstructing emissions from their various sources, including timing and location of their release. We combine untargeted GC-MS measurements at eight sampling sites along the river Rhine with a source apportionment model that separates co-occurring chemicals by their spatiotemporal dynamics. We find that the primary pollution sources of volatile chemicals in the Rhine are industrial and gasoline spills and tentatively identify as-yet-unmonitored chemicals showing similar dynamics to priority hydrocarbons, thus warranting further investigation. By considering historical low-resolution GC-MS data, we show that this modeling framework allows characterizing nontargeted emission dynamics even for legacy datasets with limited resolution. This is particularly relevant for water authorities, as it allows extracting source-specific information from existing monitoring data to support effective decisions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Tracking pollution sources in a river network by spatiotemporal modeling of untargeted GC-MS measurements

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月26日

查看全文
  • Maria Cairoli,
  • Maikel Engels,
  • Antoni Ginebreda,
  • Gerard Stroomberg,
  • Joanne de Jonge,
  • Lutgarde Buydens,
  • Roma Tauler,
  • Jeroen Jansen

摘要

Chemical pollution is a serious threat to river ecosystems. A systemic understanding of chemical co-occurrence, emission dynamics and sources is required by the Water Framework Directive to mitigate freshwater pollution. Non-target screening is key for a holistic assessment of chemicals in river water, both known and as-yet-unknown. Here, we show that its combination with multivariate statistics and advection-reaction modeling enables reconstructing emissions from their various sources, including timing and location of their release. We combine untargeted GC-MS measurements at eight sampling sites along the river Rhine with a source apportionment model that separates co-occurring chemicals by their spatiotemporal dynamics. We find that the primary pollution sources of volatile chemicals in the Rhine are industrial and gasoline spills and tentatively identify as-yet-unmonitored chemicals showing similar dynamics to priority hydrocarbons, thus warranting further investigation. By considering historical low-resolution GC-MS data, we show that this modeling framework allows characterizing nontargeted emission dynamics even for legacy datasets with limited resolution. This is particularly relevant for water authorities, as it allows extracting source-specific information from existing monitoring data to support effective decisions.