Extended Abstract (176K)2.2
Ensemble based probabilistic forecast verification
Yuejian Zhu, NOAA/NWS/NCEP/EMC, Camp Springs, MD; and Z. Toth
NCEP ensemble verification system has been developed to evaluate ensemble based probabilistic forecast since 90s. This system is mainly to focus on two attributes (reliability and resolution) for NCEP ensemble based probabilistic forecast in additional to the traditional measures, such as Pattern Anomaly Correlation (PAC) and Root Mean Square (RMS) error for ensemble mean, rank histogram, outlier and etc. The events definitions for probabilistic scores are based on 1) defined thresholds, 2) climatological percentiles and 3) defined by ensemble members. The probabilistic skill scores are based on NCEP/NCAR 40-year reanalysis climatology. Currently, this system generates Brier Skill Scores (BSS) with reliability and resolution, Ranked Probability Skill Score (RPSS), Continuous Ranked Probability Skill Score (CRPSS), Relative Operational Characteristics (ROC), Relative Economic Value (REV) and etc to apply to upper atmospheric variables, such as 500hPa height, 850hPa temperature, and near surface variables, such as 1000hPa height, 2-meter temperature, 10-meter wind. Recently, this system has been upgraded to apply for Northern American Ensemble Forecast System (NAEFS), too.
Session 2, Ensemble Forecasting Including Post Processing II
Monday, 21 January 2008, 10:45 AM-11:45 AM, 219
Previous paper Next paper
Recorded presentation