Water quality prediction, a well-established field with broad implications across various sectors, is thoroughly examined in this comprehensive review. Through an exhaustive analysis of over 170 studies conducted in the last five years, we focus on the application of machine learning for predicting water quality. The review begins by presenting the latest methodologies for acquiring water quality data. Categorizing machine learning-based predictions for water quality into two primary segments—indicator prediction and water quality index prediction—further distinguishes between single-indicator and multi-indicator predictions. A meticulous examination of each method’s technical details follows. This article explores current cutting-edge research trends in machine learning algorithms, providing a technical perspective on their application in water quality prediction. It investigates the utilization of algorithms in predicting water quality and concludes by highlighting significant challenges and future research directions. Emphasis is placed on key areas such as hydrodynamic water quality coupling, effective data processing and acquisition, and mitigating model uncertainty. The paper provides a detailed perspective on the present state of application and the principal characteristics of emerging technologies in water quality prediction.

Context: Anaerobic ammonium oxidation coupled with iron (III) reduction (feammox) is an important nitrogen loss pathway in natural ecosystems. This is the first research to investigate the feammox process in the Yellow River Delta wetlands. Aims: To reveal the nitrogen (N) loss caused by feammox in the Yellow River Delta wetland. Methods: Slurry incubation, N isotope tracer technology, and metagenomic sequencing were used to study the feammox process and nitrogen-cycling functional microorganisms and genes. Key results: Feammox was not detected in the Phragmites australis-covered (Phragmites) and mudflat (Mudflat) sites, and the feammox rate in the willow-covered site (Willow) was 0.02 mg N kg−1 d−1, accounting for 0.2% of the total N loss, which was lower than in other types of wetlands. The contributions of anammox to N loss were 32% (Phragmites), 17% (Mudflat), and 31% (Willow). Denitrification gene abundance was the highest, accounting for more than 40%, and anammox genes were not detected. The dominant genus of feammox functional microorganisms at Willow site was Anaeromyxobacter with an abundance of 62%, and the dominant genus at Phragmites and Mudflat sites was Pseudomonas, with abundances of 79% and 82%, respectively. Conclusions: Feammox in the Yellow River Delta Wetland was weak, and denitrification was the main N removal pathway. Implications: Further research is required to study why feammox rate is low in the Yellow River Delta wetlands.

With the rapid urbanization of China, environmental risks posed by informal landfills, particularly those dominated by agricultural waste, are an urgent yet understudied concern. This study systematically monitored groundwater quality surrounding five typical informal agricultural waste landfills in a coastal Chinese city. Eight major pollutants were analyzed using pollution index evaluation, the health risk model and multivariate statistical methods. The results indicate one landfill as a high-priority concern, exhibiting a combined multi-index pollution pattern with an exceedance rate of 87.5%, where NO₃⁻-N, F⁻, COD_Mn, and total hardness are the dominant indicators. Another landfill showed high background levels and anthropogenic impacts. Total non-carcinogenic risk of all landfills is below 1 (negligible). Children face approximately twice the health risk of adults. The exposure risk through drinking water ingestion is three orders of magnitude higher than that from dermal contact, with NO₃⁻-N contributing >90% of the total risk. Groundwater deterioration is primarily affected by geological conditions and seawater intrusion (52.31%), followed by agricultural activities and soil characteristics. Given these findings, priority attention should be directed to nitrogen-driven landfill and multi-index composite pollution landfill, with reinforced source tracing and control of NO₃⁻-N, alongside long-term monitoring for regional groundwater protection.

Bacterial taxa of potential public-health relevance and predicted antimicrobial-resistance functions are important in effluent management. This study compared bacterial communities, environmental associations, and predicted functions in aquaculture effluents (Group A) and wastewater treatment plant effluents (Group W) from the Mi River Basin, China, using a 16S rRNA-based framework, with the WHO 2024 Bacterial Priority Pathogens List used for contextual interpretation. relatively consistent profiles, with Acinetobacter, Limnohabitans, and Pseudomonas among abundant genera. Group W showed site-specific patterns, including high relative abundances of Legionella, Flavobacterium, and Mycobacterium in W3 and Pseudarcobacter and Arcobacter in W4 and W5. Among 400 genus–factor correlations, only the negative Sphingorhabdus–COD association remained significant after BH correction. The aggregate predicted abundance of six ARG-related KOs was 2.22-fold higher in Group A. The tetA-type KO was higher in Group A, whereas erm-type was higher in Group W. Seven samples had abundance-weighted NSTI values below 0.10, and directional patterns remained after excluding W5. Although 16S-based analyses cannot directly confirm pathogens or ARGs, the framework distinguished effluent-specific microbial and predicted resistance-related patterns, supporting large-scale routine monitoring and tiered risk management for water reuse and receiving waters.