Fluoride Action Network

Quality assessment and prediction of municipal drinking water using water quality index and artificial neural network: A case study of Wuhan, central China, from 2013 to 2019.

Source: The Science of the Total Environment | June 29th, 2022 | Authors: Xia L, Han Q, Shang L, Wang Y, Li X, Zhang J, Yang T, Liu J, Liu L.
Location: China


The sanitary security of drinking water is closely related to human health, but its quality assessment mainly focused on limited types of indicators and relatively restricted time span. The current study was aimed to evaluate the long-term spatial-temporal distribution of municipal drinking water quality and explore the origin of water contamination based on multiple water indicators of 137 finished water samples and 863 tap water samples from Wuhan city, China. Water quality indexes (WQIs) were calculated to integrate the measured indicators. WQIs of the finished water samples ranged from 0.24 to 0.92, with the qualification rate and excellent rate of 100 % and 96.4 %, respectively, while those of the tap water samples ranged from 0.09 to 3.20, with the qualification rate of 99.9 %, and excellent rate of 95.5 %. Artificial neural network model was constructed based on the time series of WQIs from 2013 to 2019 to predict the water quality thereafter. The predicted WQIs of finished and tap water in 2020 and 2021 qualified on the whole, with the excellent rate of 87.5 % and 92.9 %, respectively.

Except for three samples exceeding the limits of free chlorine residual, chloroform and fluoride, respectively, the majority of indicators reached the threshold values for drinking.

Our study suggested that municipal drinking water quality in Wuhan was generally stable and in line with the national hygiene standards. Moreover, principal component analysis illustrated that the main potential sources of drinking water contamination were inorganic salts and organic matters, followed by pollution from distribution systems, the use of aluminum-containing coagulants and turbidity involved in water treatment, which need more attention.

Keywords: China; Drinking water; Neural network model; Principal component analysis; Spatial-temporal analysis; Water quality.

*Original abstract online at https://pubmed.ncbi.nlm.nih.gov/35779730/