Ternary forecast of heavy snowfall in the Honam area,Korea

The objective of this study is to improve the statistical modelingfor the ternary forecast of heavy snowfall in the Honam area in Korea. Theternary forecast of heavy snowfall consists of one of three values, 0 for lessthan 50 mm, 1 for an advisory (50--150 mm), and 2 for a warning (more than150 mm). For our study, the observed daily snow amounts and the numerical modeloutputs for 45 synoptic factors at 17 stations in the Honam area during 5 years(2001 to 2005) are used as observations and potential predictors respectively.For statistical modeling and validation, the data set is divided into trainingdata and validation data by cluster analysis. A multi-grade logistic regressionmodel and neural networks are separately applied to generate the probabilitiesof three categories based on the model output statistic (MOS) method. Twomodels are estimated by the training data and tested by the validation data.Based on the estimated probabilities, three thresholds are chosen to generateternary forecasts. The results are summarized in 3×3 contingency tablesand the results of the three-grade logistic regression model are compared tothose of the neural networks model. According to the model training and modelvalidation results, the estimated three-grade logistic regression model isrecommended as a ternary forecast model for heavy snowfall in the Honam area.