Prediction of the Type and Amount of Surface Water Pollutants using Time Series Models (ARIMA) and L-THIA Model (Case Study: Namrood Sub-Basin, Hablehrood Watershed)

Due to the important role of non-point source pollution in water resources management, in this study time series modeling was applied to forecast water quality parameters and L-THIA model (one type of non-point source pollution models) was applied to estimate water pollutants. The purpose of this study was to compare results of L-THIA model and ARIMA models in Namrood sub-basin located in the Hablehrood watershed. At first, land use changes were studied from the years 1974 till 2017 that showed increase in agricultural lands and expansion of cities and roads. Then, using L-THIA model for both land use categories, the amount of pollutant and the volume of runoff were calculated that showed high growth. In the end, using ARIMA models were estimated water quality parameters for 30 years. Among the different ARIMA models, a model with a lowest error and akaike (AIC) criterion was selected as an optimal model for TDS, total of cations and anions. Desirable models for TDS, total of cations and anions were (0,1,1), (1,1,2) and (1,1,1), respectively. The end, diagrams of Trend Analysis and Time Series were performed for three parameters that indicated high growth in amount of pollutant. The results showed efficiency of time series modeling in water resources studies in order to forecast water quality parameters.