| 摘要: |
| 为了改进温带气旋数值预报的精度，基于WRF （Weather Research and Forecasting）模式，利用GSI （Gridpoint Statistical Interpolation）-EnKF （Ensemble Kalman Filter）系统，设计了一套温带气旋集合预报方法，其具有的2种选择方案通过滤掉质量较差的集合成员从而将集合成员数目控制在10以内，达到了大幅降低集合预报计算量的目的。针对2020年7月一次影响黄海的温带气旋个例，开展了一系列决定性预报与集合预报的数值对比试验。分析结果如下：1）不采取任何择优方案的集合预报效果就已经明显优于决定性预报，而采取择优方案使得预报效果进一步得到提升；2）预报初始时刻择优（直接择优方案）的集合预报效果远不如短时积分3 h后才进行择优（积分择优方案）的预报效果；3）积分择优方案优于直接择优方案的原因是，初始场集合体中的成员经过短时积分后其误差得以放大而使得择优更加准确。多个例的应用结果进一步表明，本文提出的积分择优方案温带气旋集合预报方法具有较好的业务预报应用前景。 |
| 关键词: 黄渤海 温带气旋 集合预报 GSI-EnKF 初始场择优 |
| DOI：10.11759/hykx20220218001 |
| 分类号: |
| 基金项目:国家重点研发计划项目（2022YFC3004200）；国家自然科学基金项目（42075069） |
|
| Numerical ensemble forecasting experiments on extratropical cyclones affecting the Yellow and Bohai Seas |
|
GUAN Qin-yu1,2, GUO Jing-tian3, GAO Shan-hong1,2
|
|
1.College of Oceanic and Atmospheric Sciences, Ocean University of China, Qingdao 266100, China;2.Key Laboratory of Physical Oceanography, Ocean University of China, Qingdao 266100, China;3.Shandong Provincial Key Laboratory of Marine Ecological Environment and Disaster Prevention and Mitigation, Qingdao 266100, China
|
| Abstract: |
| Based on the weather research and forecasting model and gridpoint statistical interpolation-ensemble Kalman filter system, an ensemble forecast method for extratropical cyclones was designed to improve its numerical forecast. This method has two optimization schemes and can reduce ensemble members to less than ten by filtering out worse members (i.e., choosing good members), thereby greatly reducing the computational cost. A series of numerical experiments was conducted on an extratropical cyclone affecting the Yellow Sea in July 2020, including a deterministic forecast and ensemble forecasts. The results can be outlined as follows:1) the ensemble forecast without an optimization method is much better than the deterministic forecast, and optimization methods can further improve forecast skill; 2) the forecast skill of the optimization method operated in the initial condition (termed as direct-optimization scheme) outperforms that of the optimization operated after a 3-h short-range forecast (termed as forecast-optimization scheme); 3) this outperformance is due to a magnification of deviations in the initial condition through a short-range forecast that leads to a more accurate choice of good members. The results of the multicase application suggest that the ensemble forecast method with the forecast-optimization scheme has a good application prospect in an operational forecast of an extratropical cyclone. |
| Key words: Yellow and Bohai Seas extratropical cyclone ensemble forecast GSI-EnKF initial condition optimization |
