Forecasting models for flow and total dissolved solids in Karoun river-Iran

- We formulate a S-ARIMA model to forecast total dissolved solids.
- We develop a transfer function model to forecast total dissolved solids according to water flow.
- Two variables in TF model in comparison to S-ARIMA model with one variable leads to better forecasting.
- We also present a S-ARIMA model to forecast water flow in Karoun river.

SummaryWater quality is one of the most important factors contributing to a healthy life. From the water quality management point of view, TDS (total dissolved solids) is the most important factor and many water developing plans have been implemented in recognition of this factor. However, these plans have not been perfect and very successful in overcoming the poor water quality problem, so there are a good volume of related studies in the literature. We study TDS and the water flow of the Karoun river in southwest Iran. We collected the necessary time series data from the Harmaleh station located in the river. We present two Univariate Seasonal Autoregressive Integrated Movement Average (ARIMA) models to forecast TDS and water flow in this river. Then, we build up a Transfer Function (TF) model to formulate the TDS as a function of water flow volume. A performance comparison between the Seasonal ARIMA and the TF models are presented.