Modeling of Weather Parameters Using Stochastic Methods

The study was carried out to developed stochastic model for weekly temperature, humidity and precipitation in Solapur (Latitude 17°40′N, Longitude 75°54′E and altitude 483.50 m amsl) station of western part of Maharashtra, India. In the present study, 42 years data (1969–2010) of daily temperature, relative humidity and precipitation of Solapur station have been used for time series analysis. Weekly mean temperature, relative humidity and monthly precipitation values were used to fit the ARIMA class of models for different orders. ARIMA models of first and second orders were selected based on autocorrelation function (ACF) and partial autocorrelation function (PACF) of the time series. The parameters of the selected models were obtained with the help of maximum likelihood method. The diagnostic checking of the selected models was then performed with the help of three tests (i.e. standard error, ACF and PACF of residuals and AIC) to know the adequacy of the selected models. The ARIMA models that passed the adequacy test were selected for forecasting. One year ahead forecast (i.e. for 2010) of temperature, relative humidity and precipitation values were obtained with the help of these selected models and compared with the values of temperature, relative humidity and precipitation obtained from the climatological data of 2010 by root mean square error (RMSE). According to the Seasonal ARIMA model, ACF, PACF and evaluation of all eventual parameters, the results from analysis show that the model fitted is weekly temperature: ARIMA (111) (011) ₅₂ , weekly relative humidity: ARIMA (111) (111) ₅₂ and monthly precipitation: ARIMA (211)(201) ₁₂ and hence are the best stochastic model for generating and forecasting of weekly temperature, relative humidity and monthly precipitation values for Solapur station, Maharashtra, India.

The studies reveal that if sufficient spread and depth of data are used in model building, frequent updating of model may not be necessary. The study also showed the utility of forecast of climatic parameter values in estimating the irrigation quantity and monitoring the insect pest and disease 1 year ahead for pomegranate orchards. It is concluded that seasonal ARIMA model is a viable tool which can successfully be used for generation and forecasting of climatic parameters having inbuilt seasonal patterns.