风暴尺度集合预报最优侧边界条件扰动方法设计:个例分析

考虑到全球预报模式与风暴尺度预报模式在分辨率上的显著差异,在构造风暴尺度集合预报系统的时候需要用一个中间分辨率的中尺度区域模式为风暴尺度模式提供侧边界条件扰动,但如何构造侧边界扰动才能更为有效地提高风暴尺度集合预报系统的预报能力目前仍然未知。本文基于WRF模式,通过一次个例试验设计了风暴尺度集合预报中的3种不同侧边界扰动方案,结果表明:直接通过0.5°水平分辨率全球集合预报扰动插值所得到的侧边界扰动(LBC_DOWN)在预报中可以获得较高的大尺度扰动能量,而在中尺度区域模式(本文中为模式外层)中通过ETKF循环所构造的侧边界条件扰动(LBC_CYCLE)包含较高的中小尺度能量,而将LBC_CYCLE中的中尺度扰动信息与LBC_DOWN中的大尺度扰动信息相混合所得到的混合侧边界扰动(LBC_BLEND)在大尺度能量上更接近于LBC_DOWN,在中小尺度能量上更接近于LBC_CYCLE;LBC_BLEND较前两种方案有着更好的离散度技巧表现;在降水概率预报技巧方面LBC_BLEND与LBC_CYCLE较为接近,且均优于LBC_DOWN。

Optimal design of lateral boundary condition perturbation method in storm-scale ensemble forecast: A case study

Considering the apparent difference between global forecasting model and convection-permitting model in resolution ratio, an intermediate limited-area model should be used to provide Lateral Boundary Condition Perturbations(LBCPs) for Storm-Scale Ensemble Forecast(SSEF), but the problem that what kind of LBCPs are the most suitable for SSEF still remains unknown. Three different LBCPs schemes are tested in a case study based on WRF model. The results show that LBC_DOWN has better large scale perturbation energy while LBC_CYCLE has more in smaller scale. The blended method LBC_BLEND is much closer to LBC_DOWN at large scale and much closer to LBC_CYCLE at smaller scale; the blended method LBC_BLEND also perform better at spread-skill than the other two experiments; meanwhile, LBC_BLEND has almost the same skill as LBC_CYCLE has in precipitation probability forecast.