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

The quality of life and human survival is dependent on sustainable development and sanitation of water bodies in an environment. The present research focuses on cyclicity data of more than 750,000 records of parameters associated with the water quality from a rural-urban river monitoring stations in real-time from River Atoyac in Central Mexico. The events detected in the instrumental records correlated with 2528 laboratory and instrumental determinations. The 64 polluting compounds were grouped into inorganic compounds (metals and metalloids) and organic compounds (pesticides, herbicides, hydrocarbons). Metal associated compounds were grouped along mechanical, pharmaceutical and textile industries which associates itself with the entry of polluting components. The cyclicity of the events was detected through Discrete Fourier Transformation time series analysis identifying the predominant events in each station. These highlight the events at 23-26 h corresponding to a circadian pattern of the metabolism of the city. Likewise, pollution signals were detected at 3.3, 5.5, and 12-14 h, associated with discharges from economic activities. Multivariate statistical techniques were used to identify the circadian extremes of a regionalized cycle of polluting compounds in each of the stations. The results of this research allow pollution prevention using a mathematical analysis of time series of different quality parameters collected at monitoring stations in real-time as a tool for predicting polluting events. The DFT analysis makes it possible to prevent polluting events in different bodies of water, allowing to support the development of public policies based on the supervision and control of pollution.