Prediction of occurrence of daily summer monsoon precipitation over Karnataka

Abstract

AbstractKarnataka, a State in south peninsular India, receives 73% of its annual rainfall during the southwest monsoon season. Because of the complex physiographical features, the rainfall pattern over the State shows large spatial variation from 50 to 350 cm. The coefficient of interannual variation of the monsoon rainfall is about 15% over coastal Karnataka (CK) and between 20 and 30% over interior Karnataka. It is, therefore, a difficult task to predict the location specific daily rainfall over Karnataka. In this study, an attempt has been made to develop an objective tool for forecasting the occurrence and non‐occurrence of precipitation during a 24 h period for the 19 stations in Karnataka during the monsoon season. The probability of precipitation (POP) model is developed using forward stepwise regression with the available surface and upper air parameters from synoptic and radiosonde and radio wind stations in and around Karnataka as potential predictors. The POP model has been developed based on the data from 1981 to 1996 and verified with the data from 1997 to 2002. Different skill scores are computed using a yes/no contingency table. The POP model performs very well, with percentages of correct (PC) forecasts for occurrence/non‐occurrence of precipitation being 57–91% for the independent data. Comparing the results of the POP model with that of the conventional method of forecast for Bangalore City, the PC forecasts improves from 44 to 56% with the use of the POP model. Copyright © 2011 Royal Meteorological Society