LAUSR

Abstract Dissolved organic matters (DOM) in wastewater are the key fraction to be treated, and affect various technologies inversely. Ultrahigh-resolution mass spectrometry (URHMS) offers a molecular level insight into the DOM transformation during wastewater treatment processes. However, owing to the complex and high dimensional dataset, data mining is the knowledge gap to widely apply molecular information method. This review summarises the framework of data acquisition, data treatment and data mining processes. The data acquisition and treatment include sample pre-treatment, instrumental analysis and raw data conversion. Typical data mining strategies are classified into Venn diagram and correlation analysis, which can be applied in three wastewater treatment scenarios. Venn diagram is expanded to a generalized classification model and semiquantitative calculation methods. Additionally, correlation analysis can link DOM molecules with various parameters, including time, special, environmental and microbial factors. These data mining strategies are suitable for both batch and continuous wastewater treatment reactors in lab or engineering scale. Recommends are proposed to model the DOM transformation in the future. This review offers a framework to support UHRMS applications in wastewater treatment, which can be applied in other environmental monitoring scenarios as well.