Skip to main navigation Skip to search Skip to main content

Integration of methane oxidation with anammox for enhancing nitrogen removal: Performance, mechanisms, and machine learning analysis

  • Xueqin Lu
  • , Mengting Xia
  • , Yingxiang Tang
  • , Yibo Sun
  • , Shiliang Heng
  • , Xue Chen
  • , Weijie Hu
  • , Guangyin Zhen*
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Anaerobic ammonia oxidation (anammox) is a promising process for nitrogen removal, yet nitrate accumulation limits its application. This study developed a methane (CH4)-driven system integrating denitrification and anammox for efficient nitrogen removal with CH4 recycling. In a long-term operation of a CH4-fed up-flow anaerobic sludge blanket reactor (UASB), a total nitrogen removal efficiency of 93 % was achieved under the operational condition of a hydraulic retention time (HRT) of 6 h and an influent nitrogen concentration of 450 mg N L−1. CH4 enhanced nitrogen removal by irritating the growth of denitrifying bacteria (the abundance of Pseudomonas increased to 8 %) without requiring anaerobic methanotrophic (ANME) archaea or denitrifying anaerobic methane oxidation (DAMO) bacteria. Futhermore, the abundance of anaerobic oxidizing bacteria (AnAOB) rose from 0.5 % to 18.6 %. Random Forest modeling identified AnAOB activity, effluent NH4+ and NO3 as key factors governing removal efficiency. This study demonstrates that CH4-driven denitrification coupled with anammox significantly enhances wastewater nitrogen removal efficiency, providing a novel strategy for CH4 utilization in bioremediation.

Original languageEnglish
Article number127719
JournalJournal of Environmental Management
Volume395
DOIs
StatePublished - Dec 2025

Keywords

  • Anammox
  • Denitrifying methane oxidation
  • Machine learning
  • Nitrogen removal

Fingerprint

Dive into the research topics of 'Integration of methane oxidation with anammox for enhancing nitrogen removal: Performance, mechanisms, and machine learning analysis'. Together they form a unique fingerprint.

Cite this