国家气候中心短期气候预测模式系统业务化进展

Progress in Developing the Short-range Operational Climate Prediction System of China National Climate Center

该文简要介绍了国家气候中心短期气候预测模式系统的研发成果,并侧重于从海洋资料同化系统、陆面资料同化系统、月动力延伸预测模式系统、季节气候预测模式系统4个方面介绍了第2代短期气候预测模式系统的业务化进展。第2代海洋资料同化系统已初步建成,其对温盐的同化效果总体上优于第1代同化系统;陆面资料同化系统正在研发中,目前已完成其中的多源降水融合子系统的业务建设工作,可为陆面分量提供实时的大气降水强迫分析场;第2代月动力延伸预测系统基于国家气候中心大气环流模式BCC_AGCM2.2建立,已于2012年8月进入准业务运行阶段;第2代季节预测模式系统基于国家气候中心气候系统模式BCC_CSM1.1(m)建立,将于2013年底投入准业务运行。初步评估表明:第2代月动力延伸预测模式系统和季节气候预测模式系统分别对候、旬、月和季节、年际时间尺度的气候变率体现出了一定的预测能力,其对降水、气温、环流等要素的预测技巧总体上要高于第1代预测系统。

The progress in developing the second-generation short-range climate forecast system of National （or Beijing） Climate Center （NCC or BCC） is introduced, focusing on four items, i. e. , the global ocean data assimilation system, the land data assimilation system, the monthly-scale dynamical extended-range fore- cast system （DERF）, and the seasonal climate forecast system. With a better assimilation of temperature and salinity than the first-generation system, the second-generation ocean data assimilation system is now at the quasi-operational level. The land data assimilation system is still under development, but the multi- source precipitation merging subsystem is now quasi-operational and can produce reanalysis of precipitation as a forcing to land system. The atmospheric general circulation model BCC_AGCM2.2 and the climate system model BCC_CSM1. l（m） are the main tools for the second-generation monthly-scale DERF and the second-generation seasonal prediction system, respectively. The former has entered quasi-operational use since middle August of 2012 and conducted four-member real-time forecast jobs and 80 hindcast jobs every day, and the latter will enter its quasi-operational stage by the end of 2013. A preliminary evaluation indi- cates that the second-generation system shows a certain capability in predicting the pentad, ten-day, monthly, seasonal and inter-annual climate variability. It exhibits a higher prediction skill, compared to the first-generation system, in terms of precipitation, surface air temperature, atmospheric circulation and E1 Nifio-Southern Oscillation, and so on. As shown by the hindcasts by two generations of DERF （i. e. , DERF1.0 and DERF2.0） for the monthly mean surface air temperature in January and July, DERF2.0 shows overall higher prediction skill than DERF1.0, especially over the tropical Indian Ocean and Pacific and most mid-high latitude areas in the Northern Hemisphere in January, and most regions in global trop- ics and subtropics in July. Also, the 20-year hindcasts initialized in the end of February of each year by the two generations of seasonal climate prediction system indicate that, the second-generation system shows significant prediction skill of surface air temperature over most areas in spring, especially over the tropical Pacific, Atlantic and Indian Ocean. In contrast, the skills over most areas of the first-generation system are relative lower.