Temperature and Precipitation Evaluation of Monthly Dynamic Extended Range Forecast Operational System DERF2.0 in China

On the basis of the data of 669 observed weather stations supplied by the National Meteorological Information Center and hindcast data of the National Climate Centre second-generation monthly Dynamic Extended Range Forecast operational system (DERF2.0) from 1982 to 2010, temperature and precipitation in the prediction performance were evaluated and analyzed by using the anomaly correlation coefficient (ACC), mean square skill score (MSSS), anomaly sign consistency rate (R), and short-term climate prediction operational grading evaluation scores (Pg). The results indicated that the temperature prediction performance of DERF2.0 was significantly better than that of the DERF1.0 operational system in current usage and that the ACC skill score of temperature was noticeably higher than the operational score of the short-range climate forecast. Compared with temperature, the precipitation prediction performance of DERF2.0 was relatively poor. The ACC skill score of precipitation of DERF2.0 was close to that of DERF1.0. DERF2.0 was somewhat skillful in extreme drought and flood years such as 1998 and 2006. Furthermore, the prediction performance of temperature was significantly better than that of precipitation in extreme drought and flood years. From space, the prediction performance of DERF2.0 on the deterministic prediction was poor in the southwest. Thus, DERF2.0 should be improved.