Identifying pollution dynamics using discrete Fourier transform: From an urban-rural river, Central Mexico

Area of Science:

• Environmental Science
• Water Quality Monitoring
• Time Series Analysis

Background:

• Sustainable development and water body sanitation are crucial for human survival.
• River Atoyac in Central Mexico faces pollution challenges impacting water quality.

Purpose of the Study:

• To analyze cyclicity in water quality data from River Atoyac.
• To identify pollution patterns and their sources using advanced analytical techniques.
• To develop a predictive model for pollution events to support public policy.

Main Methods:

• Collected over 750,000 real-time water quality records from monitoring stations.
• Utilized Discrete Fourier Transformation (DFT) for time series analysis.
• Applied multivariate statistical techniques to identify pollution extremes.

Main Results:

• Identified predominant pollution event cycles, including circadian patterns (23-26h) and activity-linked signals (3.3, 5.5, 12-14h).
• Correlated inorganic (metals, metalloids) and organic (pesticides, herbicides, hydrocarbons) pollutants with industrial discharges.
• Detected circadian extremes of polluting compounds at different monitoring stations.

Conclusions:

• Real-time water quality data analysis, particularly DFT, is effective for predicting pollution events.
• Mathematical analysis of time series data can guide pollution prevention strategies.
• Findings support the development of public policies for water pollution supervision and control.