Enhancing COSMO-DE ensemble forecasts by inexpensive techniques
Enhancing COSMO-DE ensemble forecasts by inexpensive techniques
复制标题
通过廉价技术增强 COSMO-DE 集合预报
DOI：
10.1127/0941-2948/2013/0374
复制
发表时间：
2013
影响因子：
1.2
通讯作者：
C. Gebhardt
中科院分区：
文献类型：
--
作者：
Z. B. Bouallègue;S. Theis;C. Gebhardt
COSMO-DE-EPS是一个基于高分辨率数值天气预报模式COSMO-DE的允许对流的集合预报系统，自2010年12月开始试运行，提供覆盖德国的概率预报。基于侧边界条件、物理参数和初始条件的变化，该系综系统由20个成员组成。为了在计算量较小的情况下增加样本量，COSMO-DE-EPS与其他集成技术相结合：邻域方法和时滞方法。评估了它们对所产生的概率预报质量的影响。客观核实是在六个月的时间内进行的，根据Brier评分及其分解得出的分数显示为2011年6月。集合系统与其他办法的结合改进了对降水的概率预报，特别是对高降水阈值的预报。此外，COSMO-DE-EPS与仅用时滞方法相结合，提高了区域降水概率预报的技巧，而不影响2m温度和阵风预报的技巧。(原名)
COSMO-DE-EPS, a convection-permitting ensemble prediction system based on the high-resolution numerical weather prediction model COSMO-DE, is pre-operational since December 2010, providing probabilistic forecasts which cover Germany. This ensemble system comprises 20 members based on variations of the lateral boundary conditions, the physics parameterizations and the initial conditions. In order to increase the sample size in a computationally inexpensive way, COSMO-DE-EPS is combined with alternative ensemble techniques: the neighborhood method and the time-lagged approach. Their impact on the quality of the resulting probabilistic forecasts is assessed. Objective verification is performed over a six months period, scores based on the Brier score and its decomposition are shown for June 2011. The combination of the ensemble system with the alternative approaches improves probabilistic forecasts of precipitation in particular for high precipitation thresholds. Moreover, combining COSMO-DE-EPS with only the time-lagged approach improves the skill of area probabilities for precipitation and does not deteriorate the skill of 2 m-temperature and wind gusts forecasts. (orig.)