全球变化知识资源中心

Rainfall is the significant parameter for climate change, meteorological and hydrological process. The present paper describes the seasonal rainfall patterns in the Rayalaseema region of Andhra Pradesh state, India using the unobserved component model (UCM) with the hidden components like trend, seasonal, cyclical and irregular. The seasonal rainfall data were provided by India Meteorological Department (IMD), using daily gridded rainfall data with 10 automatic weather stations spread over the Rayalaseema region and the study deals with four seasons of rainfall as classified by IMD, namely winter, pre-monsoon, southwest monsoon and northeast monsoon. Basic Structural Model (BSM) with the components of constant trend, deterministic trigonometric seasonal, deterministic cycle and irregular is selected from the parsimonious models of UCM based on Akaike's information criteria (AIC), Bayesian information criteria (BIC), significant tests and statistical fit. The model parameters are obtained using maximum likelihood method; the adequacy of the selected model is determined through correlation and normal diagnostics. The forecast of the seasonal rainfall patterns during the years 2016-2018 has been noticed with the help of selected UCM. From the model forecast, it is observed that the pre-monsoon season receives rainfall of 96.7 mm in the years 2016 and 2018, whereas 71.3 mm in the year 2017; the southwest monsoon season receives the rainfall of 396.0 mm in the year 2016, 424.2 mm in 2017 and 419.2 mm in 2018; the northeast monsoon season receives the rainfall of 286.8 mm in 2016, 261.3 mm in 2017 and 286.8 mm in 2018.