This study was carried out to develop the stream water quality model for the intaking station of Kongju waterworks in the Keum River system. The monthly water quality(total nitrogen and total phosphorus) with periodicity and trend were forecasted by multiplicative ARIMA models and then the applicability of the models was tested based on 7 years of the historical monthly water quality data at Kongju intaking site. The parameter estimation was made with the monthly observed data. The last one year data was used to compare the forecasted water quality by ARIMA model with the observed one. The models are ARIMA(2,0,0)×(0,1,1)_12 for total nitrogen, ARIMA(0,1,1)×(0,1,1)_12 for total phosphorus. The forecasting results showed a good agreement with the observed data. It is implying the applicability of multiplicative ARIMA model for forecasting monthly water quality at the Kongju site.