Distribution patterns and source contributions of emerging contaminants in urban wastewater systems: from pumping...

Sewage pumping stations and wastewater treatment plants (WWTPs) serve as critical nodes through which emerging pollutants (ECs) migrate from urban and industrial sources into aquatic environments. However, research on the co-occurrence, source dynamics, and fate of ECs within integrated pumping station-WWTP system remains limited, particularly spatial heterogeneity and source-sink relationship. This study addresses this gap by proposing a multi-model analytical framework based on the "fingerprint features" of pollutants, integrating self-organizing maps (SOM), positive matrix factorization (PMF), and the XGBoost machine learning algorithm. This framework is applied to investigate the distribution patterns, coexistence characteristics, and potential sources of over 80 ECs in 7 WWTPs and 47 pumping stations in a coastal city in China. Results show high consistency in EC abundance between WWTPs and pumping station, with co-occurrence frequency strongly influenced by their physical-chemical properties and spatial distribution. Acetaminophen (893 ng/L), caffeine (278 ng/L), and norfloxacin (269 ng/L) exhibited the highest average concentrations in pumping stations. Despite incomplete removal of ECs during treatment, wastewater processes effectively mitigate the complex co-occurrence of ECs. Interactions such as competitive adsorption led to fluctuations in their transport behavior, particularly over long distances. The distinctive "fingerprint" characteristics of ECs, which reflect their unique distribution patterns, were leveraged in source apportionment using advanced techniques such as self-organizing mapping, positive matrix factorization, and the XGBoost model. These methods provided critical insights into the spatial heterogeneity of pollution sources, which are mainly driven by human activities and regional economic development. The results emphasized the impact of local wastewater infrastructure, human activities and regional economic development on EC spatial distributions, providing a basis for informed environmental management planning for ECs.