This paper aims to investigate the effects of natural variables, including precipitation and flow rate, on the quality of the Zarjoub River in Guilan province, Iran. The new hydrological insight in this study is a lack of national research focused on the dual effects of rainfall and flow rate on river water quality in coastal areas along the Anzali Wetland. To investigate the effect of precipitation and flow rate on river water quality, nine water quality variables were monitored during the 10‐year period. In this article, (a) the existence of trends and the best fitted models of water quality parameters and the discharge and precipitation were analysed using statistical techniques and (b) the relationships between concentration of constituents with the discharge and precipitation on the up‐stream and middle station were also examined. Box plots, for explaining the distribution of a data collection, were used. The results showed the existence of trend of water quality parameters with river flow and rainfall. As presented in Section 3, with increasing precipitation and flow rate, concentration of all constituents, except pH and SO₄2, decreased. On the contrary, the maximum amount of water quality elements was observed in low precipitations; therefore, the maximum concentration occurred in less than 15 mm precipitation. Simple regression was used to evaluate the discharge concentration and precipitation concentration. According to the correlation coefficient (r), the relationship between concentration and precipitation is weaker than (0.238) discharge concentration (0.699). The results further showed climate change and river water quality to be related.

Quality and quantity of streamflow are crucial components in the management and control of water resources according which are challenging due to their nonstationarity and uncertainty path. This paper presented an ensemble data pre-processing-based machine learning (ML) algorithm for the decision-support of water resource management and water pollution control at the watershed scale due to the nonlinear path of streamflow. In the proposed hybrid model, a new time–frequency analysis algorithm, variational mode decomposition (VMD), is implemented to deal with the nonlinearity and nonstationary of a streamflow process. The VMD is exploited to decompose the original water quality and quantity series into a series of intrinsic mode functions (IMFs) with different frequencies. Therefore, an ensemble algorithm, bootstrap aggregating (bagging) algorithm is coupled with two common ML, reduced error pruning tree (REPT) and random forest (RF), to predict all the decomposed modes using VMD. Then, in order to reduce the variance among the base classifiers of the proposed ML, a bootstrap aggregation technique was recruited. Finally, the predicting value of the original water quality and quantity series is obtained by adding up the predicting results of all the decomposed modes. The proposed hybrid decomposition–ensemble model has been applied to two stations in Karoon River, Iran. Results obtained from this study indicate that the proposed hybrid decomposition–ensemble model can capture the nonlinear characteristics of a streamflow process in terms of water quality and quantity simultaneously and thus provide more accurate predicting results compared with those models without data frequency decomposing.