To investigate and assess the effects of land use and its changes on concentrations of heavy metals (Pb, Zn, Cd, Cu, Mn, Ni, Fe) in the tributary of drinking water reservoir catchment, soils of different land use types (forest, arable land, meadows and pastures, residential areas), suspended sediment and bottom sediment were collected. Heavy metals were analyzed using atomic absorption spectrophotometry (AAS). The metal distribution pattern was observed, where Zn and Cd could be considered as main metal contaminants. The variation in the concentration level of Zn and Cd in studied soils showed the impact of pollution from anthropogenic activities. Also some seasonal variations were visible among the suspended sediment and bottom sediment samples which could be associated with land agricultural practices or meteorological conditions. The sediment fingerprints approach used for determining sources of the suspension in the catchment showed (Kruskal-Wallis H test, p<0.05), that only Mn and Ni were not able to be distinguished among the potential sediment sources. A multiple linear regression model described the relationship between suspended sediment and 4 types of soil samples. The results related suspended composition mostly to the samples from the residential land use. Considering the contemporary trend of observed changes in land use resulting in conversion of agricultural areas into residential and service structures these changes can be essential for the contamination of aquatic environment. This situation is a warning sign due to the rapid industrialization, urbanization and intensive agriculture in this region what can significantly affect the drinking water quality.
When high precision modelling is required, for example, with the estimation of suspended sediment load (SSL), data-driven models are preferred over physically-based numerical models for their real-time, short-horizon prediction ability. The investigation of SSL, as an important index in engineering practices assessment, like design and operation of the hydraulic structures not only shows the hydrological behaviour of the river, but also illustrates the valuable information about the water quality deterioration, surface-groundwater interaction and land-use changes of the watershed. The following data-driven methods were compared in order to predict SSL at the Seyra gauging station on the Karaj River in Iran: Fuzzy logic (FL), two adaptive neuro-fuzzy inference systems (i.e., ANFIS-GP and ANFIS-FCM models), an artificial neural network (ANN), and least squares support vector machine (LSSVM). Monthly average river flow and SSL data for 50 years were obtained from the Tehran Regional Water Authority (TRWA). The data was first divided into training, validation and test sets and the SSL was then predicted using the ANN, FL, ANFIS, and LSSVM models. The reliability of the applied models was evaluated by the correlation coefficient (R), root mean square error (RMSE), and mean absolute error (MAE). The results showed that the ANFIS models outperformed the ANN, FL, and LSSVM models for predicting SSL using the given input and output data. Overall, the performances of the artificial intelligence models used in the present study were satisfac-tory in predicting the non-linear behaviour of the SSL.