Rainfall is an important component of the water cycle and which is giving most of the fresh water to the earth. The Amount of rainfall meet various demands of agriculture, industry and irrigation. Jaffna District is under a dry zone and peak rainfall is during the months of October to December. The major source of water for agricultural production is rainfall. The Agriculture field is a leading field in this district. There are 33% of the families in this District solely depending on agriculture. Forecasting rainfall in the Jaffna district plays an important role in the planning and management of agriculture scheme and management of water resource systems. Therefore, it is essential to develop a time series model to analyze the amount of rainfall in the Jaffna district. The main purpose of this study to find a suitable Auto Regression integrated moving average (ARIMA) model to the monthly rainfall data of the Jaffna district. In this study, monthly rainfall of Jaffna district is modeled by Box-Jenkins’ time series approach. The 228 monthly rainfall data got from the department of meteorology, Colombo, Sri Lanka during the period of January, 2000 to 2020 December.  Further, three statistical criteria such as Akaike information criteria, Bayesian information criteria, mean squared error were used in order to select the best model. Through the modeling, it was found that seasonal Auto Regressive Integrated Moving Average: is the best fitting model for the Jaffna district. Moreover, the identified best model has been tested using Ljung-Box chi-squared statistics. An Identified model can be forecast to assist scientists and policymakers in developing strategies for effective scheduling for flood prediction, urban planning, and other environmental purposes. The forecast can be used as a modeling tool to help create better water management practices.

Keywords: Rainfall; ARIMA; Box-Jenkins' Approach