Long-term spatiotemporal assessment of water quality in the Karun River, Southern Iran: a novel Python-based approach for rapid processing

ABSTRACT The present study aims to streamline the long-term spatiotemporal river water quality assessment and forecasting utilizing three intuitive Python-based modules: (1) Python toolbox for scalable Outlier Detection (PyOD) to classify significant deviations from the expected water quality norms (outliers), (2) Statsmodels to decompose the river time series data into its trend, seasonal, and residual components, and (3) Automatic Time Series forecasting model (AutoTS) to forecast (and compare) the future water quality state of the Karun River (the case study) in Southern Iran. The findings indicate that the outlier elimination has a remarkable impact on the outcomes of the Karun time series data analysis. Additionally, a significant increase in total dissolved solids (TDS) concentrations and a cyclic pattern were discernable in the decomposed time series. Furthermore, the water quality values were found to be clustered around the median of their datasets. Based on the forecasting validation metrics, the proposed automated forecasting model was found to be promising in predicting the future water quality state of the river.