庄照荣,李兴良,刘艳. 2021. 数字滤波初始化方案在高分辨率GRAPES区域模式中的研究[J]. 气象学报, (0):-, doi:10.11676/qxxb2021.021
数字滤波初始化方案在高分辨率GRAPES区域模式中的研究
Study of digital filtering initializations in GRAPES regional model
投稿时间:2020-06-05  修订日期:2020-08-24
DOI:10.11676/qxxb2021.021
中文关键词:  非绝热数字滤波,增量数字滤波,初始化方案,GRAPES区域模式
英文关键词:Diabatic digital filtering, incremental digital filtering, initialization scheme, GRAPES regional model
基金项目:
作者单位E-mail
庄照荣 国家气象中心 zrzhuang@cma.gov.cn 
李兴良 国家气象中心 lixliang@cma.gov.cn 
刘艳 国家气象中心 liuyan@cma.gov.cn 
摘要点击次数: 60
全文下载次数: 57
中文摘要:
      高分辨率同化系统中引入的高密度高频率不规则分布的观测信息会造成分析变量之间的不平衡,而不平衡分析噪音在模式积分中产生的虚假重力波会影响模式预报质量和稳定运行。为了抑制虚假重力波在模式中快速增长,本文通过在水平3km分辨率的GRAPES区域模式中应用非绝热数字滤波和增量数字滤波初始化方案,研究不同滤波截断周期的数字滤波初始化方案对分析预报的影响。试验结果表明,不同初始化方案的分析预报效果依赖于截断周期,截断周期越长,分析时刻模式预报场的权重越低,滤除的高频信息越多。对于3km分辨率模式,2h截断周期的非绝热数字滤波方案(T2h试验)滤除了有气象意义的信息,初始场过于平滑。30min(T30试验)与15min(T15试验)截断周期的初始化方案既滤除了高频噪音,也基本保留了有意义的分析信息。增量数字滤波方案可以保留背景场中快速增长的模态,只滤除分析增量不协调的噪音,其初始场最接近分析场。在GRAPES-MESO系统中采用不同截断周期的数字滤波方案对要素场预报的影响基本可忽略,而对降水预报略有影响;而在GRAPES-RAFS快速分析预报循环系统中,经过多次分析预报循环后,初始化方案对分析和预报的影响会逐渐累积,T15试验对风场分析和12小时风场预报比T2h试验有明显正贡献,同时T15试验的24h内各时段的降水预报质量也整体优于T2h试验。
英文摘要:
      The imbalance of the analysis variables which is created by introducing the irregular high frequency and high density observations in the high-resolution analysis gives rise to unphysical gravity-inertia waves in the model integration, thus it affects the predicted qualities and stability of operational model. In this paper the diabatic digital-filtering initialization and incremental digital filter initialization are applied into 3km resolution GRAPES regional model to suppress noise in the early forecast hours, and the impact on analysis and forecast with the different cutoff periods of initializations are evaluated. The analysis and forecast experiments from 12 June to 28 June 2018 show that the filtering effects relies on the cutoff periods. If the cutoff period is longer, the weight of forecasting field at analysis time is lower and the more high frequency information is removed. In the 3km resolution modelling, a 2h cutoff period leads to damping the meteorological modes and the initial fields are too smooth. And a 30min and 15min cutoff period initializations remove the high frequency noise and retain meaningful analysis information. It should be pointed out that the incremental digital filtering initialization just removes the noise of analysis increments and has smaller initialization increments, so the fast developing modes in background field can be preserved. The impact produced by different cutoff period initialization schemes in GRAPES-MESO system on the other field variables can be neglected rather than the precipitation forecast. After the several analysis and forecast cycles, the accumulated effects of initialization schemes are gradually observed in GRAPES-RAFS system. The results show that the wind analysis qualities and 12h forecast qualities are obviously improved in the T15 experiments in comparison with T2h ones. In addition, ETS scores of precipitation with T15 experiments are higher than those of T2h experiments.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮