The study is aimed in the understanding of groundwater level fluctuations in the Shanmuga Nadhi sub basin due to variations in monthly rainfall. Auto Regressive Integrated Moving Average (ARIMA) model of Box and Jenkins method is used as a stochastic processing time series study. Thirty three years of monthly rainfall and monthly water level records as a time series are statistically analysed for understanding the relationship and influence of rainfall over the natural recharge systems of the Shanmuga Nadhi sub basin in order to evaluate the aquifer behaviour of the study area. 6 rainfall stations and 12 dugwells are selected for the study. The long term data analysis automatically removes any existing bias and the normalisation generalises the relationship in a given point of time. The ARIMA model validated using normalised BIC with minimal standard error on the statistical manipulations are further analysed using regression analysis for goodness of fit on a linear scale. The goodness of fit and higher confidence level of water table fluctuations along with the linear relationship on the algebraic equation provides a stochastic forecast model for groundwater management of the basin.
