Abstract: On the basis of the time series analysis,the changes of nitrate-N concentrations in Gaoyao section of Xijiang River are studied in this paper for the analysis of the integrated catchment's response. It is found that the nitrate-N time series has a linear increase trend. Besides,its peak nitrate-N seasonal index occurs in May,almost the same time as the peak precipitation seasonal index does. It can be concluded that the soil erosion from precipitation,as well as the increase use of nitrogen fertilizer in May are the main cause of the peak nitrate export in the Xijiang catchment. In order to predict the changes of nitrate-N concentration in the Xijiang River,an integrated autoregressive and moving average model of ARIMA(1,0,1)(2,1, 1)¹² is developed in this paper. The model's responses are verified by the comparison to the measurements of nitrate-N concentration in Gaoyao station,and a favorable match is found between them. The model simulation is conducted for the prediction of the future nitrate-N export in Xijiang River. It is clear that this paper is beneficial for the water environment protection in the Xijiang catchment.