In this paper, we used a nonhomogeneous Gaussian regression model (NGR) as the postprocessing techniques to calibrate probabilistic forecasts that take the form of probability density functions for temperature. We also performed the alternative implementation techniques of NGR, which are stationspecific ensemble model output statistics (EMOS) model. These techniques were applied to forecast temperature over Pyeongchang area using 24-member Ensemble Prediction System for Global (EPSG). The results showed that the station-specific EMOS model performed better than the raw ensemble and EMOS model.

This paper applied the ensemble model output statistics (EMOS) with truncated normal distribution, which are easy to implement postprocessing techniques, to calibrate probabilistic forecasts of wind speed that take the form of probability density functions. We also considered the alternative implementations of EMOS, which were EMOS exchangeable model and reduced EMOS model. These techniques were applied to the forecasts of wind speed over Pyeongchang area using 51 members of the Ensemble Prediction System for Global (EPSG). The performances were evaluated by rank histogram, mean absolute error, root mean square error and continuous ranked probability score. The results showed that EMOS models with truncated normal distribution performed better than the raw ensemble and ensemble mean. Especially, the reduced EMOS model exhibited better prediction skill than EMOS exchangeable model in most stations of study area.

This paper used the Bayesian model averaging (BMA) with gamma distribution that takes the form of probability density functions to calibrate probabilistic forecasts of wind speed. We considered the alternative implementation of BMA, which was BMA gamma exchangeable model. This method was applied for forecasting of wind speed over Pyeongchang area using 51 members of the Ensemble Prediction System for Global (EPSG). The performances were evaluated by rank histogram, means absolute error, root mean square error, continuous ranked probability score and skill score. The results showed that BMA gamma exchangeable models performed better in forecasting wind speed, compared to the raw ensemble and ensemble mean.

This paper considers a homogeneous multiple regression (HMR) model and a non-homogeneous multiple regression model, that is, ensemble model output statistics (EMOS), which are easy to implement postprocessing techniques to calibrate probabilistic forecasts that take the form of Gaussian probability density functions for continuous weather variables. The HMR and EMOS predictive means are biascorrected weighted averages of the ensemble member forecasts and the EMOS predictive variance is a linear function of the ensemble variance. We also consider the alternative implementations of HMR and EMOS which constrains the coefficients to be non-negative and we call these techniques as HMR+ and EMOS+, respectively. These techniques are applied to the forecasts of surface temperature over Pyeongchang area using 24-member Ensemble Prediction System for Global (EPSG). The performances are evaluated by rank histogram, residual quantile-quantile plot, means absolute error, root mean square error and continuous ranked probability score (CRPS). The results showed that HMR+ and EMOS+ models perform better than the raw ensemble mean, HMR and EMOS models. In the comparison of HMR+ and EMOS+ models, HMR+ performs slightly better than EMOS+ model in terms of CRPS, however they had a very similar CRPS and if there exists a ensemble spread-skill relationship, it is seen that EMOS is slightly better calibrated than the homogeneous multiple regression model.

In this study, we analyzed the performance of calibrated probabilistic forecasts of surface temperature over Pyeongchang area in Gangwon province by using Bayeisan Model Averaging (BMA). BMA has been proposed as a statistical post-processing method and a way of correcting bias and underdispersion in ensemble forecasts. The BMA technique provides probabilistic forecast that take the form of a weighted average of Gaussian predictive probability density function centered on the bias-corrected forecast for continuous weather variables. The results of BMA to calibrate surface temperature forecast from 24-member Ensemble Prediction System for Global (EPSG) are obtained and compared with those of multiple regression. The forecast performances such as reliability and accuracy are evaluated by Rank Histogram (RH), Residual Quantile-Quantile (R-Q-Q) plot, Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and the Continuous Ranked Probability Score (CRPS). The results showed that BMA improves the calibration of the equal weighted ensemble and deterministic-style BMA forecasts performs better than that of the deterministic forecast using the single best member.