Water quality of Danube Delta systems: ecological status and prediction using machine-learning algorithms

Area of Science:

• Environmental Science
• Ecology
• Aquatic Biology

Background:

• Global environmental issues impact European wetlands like the Danube Delta.
• The Water Framework Directive (2000/60/EC) aims to mitigate these environmental problems.
• Altered water quality and habitat heterogeneity affect aquatic biocenosis, particularly macroinvertebrate communities.

Purpose of the Study:

• To evaluate the ecological status of the Danube Delta's Southern Branch (Saint Gheorghe).
• To utilize benthic fauna and computational methods for real-time water quality monitoring.
• To assess spatial and temporal variations in aquatic system status.

Main Methods:

• Analysis of spatial and temporal variability of unicriterial and multicriterial indices.
• Characterization of chemical status, including coliform bacteria and chemical parameters.
• Application of machine learning (ML) algorithms for real-time water quality classification.

Main Results:

• Ecological status assessment revealed moderate status based on biological quality elements.
• Chemical status assessment indicated a good ecological status.
• ML algorithms also suggested a good ecological status for the water body.

Conclusions:

• Benthic fauna and ML algorithms offer viable methods for assessing aquatic ecosystem health.
• Real-time monitoring using ML can complement traditional biological assessments.
• The Danube Delta's ecological status varies, with biological indicators suggesting moderate health while chemical and ML approaches indicate good health.