Over the last decades, water quality at the Mthatha River Catchment (MRC) within the Eastern Cape Province of South Africa has been threatened by various pollutants. The continuous effluent concentration discharges from the Mthatha Prison and the Efata School for the Blind and Deaf have caused ineffable damage to the Mthatha River's water quality. Thus, the time series-measured data between 2012 and 2020 were analysed to determine the trends and enable forecasting of selected water quality parameters using the Thomas–Fiering (T–F) stochastic model. The Kendall's τ test trends show an increase in the coefficient of variation of 0.54 and 0.67 at the Mthatha Prison and Efata School, respectively, for abrupt changes, whereas the mean monthly T–F forecasted model shows a good correlation value range from 0.79 to 0.87 for the various predicted variables. The simulated predicted models and trends could serve as a measure to forecast selected water quality parameters' occurrence and a likelihood period where the river pollutants could be controlled. Water managers and researchers would find usefulness in the employed tools for an effective control planning of the river pollutants.

Mthatha town of Eastern Cape Province, South Africa has been challenged to address the pollutant issues that are coming from rampant densification and effluent concentration discharge from the Mthatha Correctional Services Centre and the Efata School for the Blind and Deaf which have caused ineffable impaired damage to the Mthatha River Catchment (MRC). This paper is aimed at identifying drivers of poor water quality in the catchment and classified the River’s water quality into different cluster groups for proper pollutant source control measures. Water quality parameters data comprising of pH; conductivity; Phosphorus; Ammonia (NH4-N); Feacals; and E-coli covering 95 percent and 105 percent of the upstream and downstream sections of the River were available at ten monitored sites of the river catchment. These datasets covering eight years 2012-2020 were analysed in this study. Factor analysis as a choice of principal component analysis (PCA) and Agglomerative Hierarchical Clustering (AHC) was used to deduce inferences for the pollutants’ subsequent classification. The results classified the catchment into three different clusters of lower pollutant (LP), medium pollutant (MP), and high pollutant (HP) areas, with PC1 accounting for 84.54% of the total variance from the three components classification. Adaptive catchment managers would find usefulness in the employed statistical tools in ensuring real-time measures for river non-point pollutants sources control that could offer additional benefits in maintaining a safe life above and below water in the preservation of their public values benefit. The study recommends the issuance of compliance notices and non-point pollutant source control measures to improve the water quality (WQ) parameters.