Predicting of surface water quality index using Discriminant analysis and multiple linear regression

الملخص

This study focuses on evaluation and interpretation of multiple measurements data for surface water quality parameters and pollution sources in the Sourani dam lake - Tartous- Syria, which is permitting us to acquire innovative information about water quality, evaluation and optimal design of monitoring network. Currently, evaluation and forecasting techniques have commonly disseminate, in order to protect and sustain water resources.

The comprehensive pollution index (CPI) was calculated in the seven monitoring stations for all measurements during January 2018 and December 2020, as the source of this pollution outcomes from villages sewage water, agricultural and industrial and tourist facilities nearby to the lake. This index indicates generally that the water quality in the lake is qualified and basically qualified.

In this study, statistical techniques, including discriminant analysis (DA) and multiple linear regression (MLR) were applied to assess the water quality of the lake. The study concerns the analysis of 21 physical, chemical and microbiological parameters in water samples collected monthly over a period of three years (2018-2020) from 7 different sampling sites located around and within the lake. Exploratory analysis of laboratory data initially included statistical description, outliers elimination and pretreatment were done.

It is evidence from the results that the (CPI) calculated through multi-linear regression models were applied to the lake achieves the highest correlation coefficient of (R = 0.98).

These techniques support and offer water management authorities with knowledge associated to the use, adjustment or prediction of a water quality index, and thus contribute to the future development studies of potable water.