ECMWF数值模式对中韩航线海域灾害性大风的预报效果检验
DOI:
作者:
作者单位:

1.山东省气象防灾减灾重点实验室;2.威海市气象局;3.山东省威海市气象局

作者简介:

通讯作者:

中图分类号:

基金项目:

山东省气象局面上课题(2021sdqxm08);山东省自然科学基金创新发展联合基金项目(ZR2024LQX009)


请扫码阅读

Verification of ECMWF Numerical Model's Forecast Effect on Disastrous Strong Winds in the China-Korea Shipping Route Area
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    利用2016-2021年黄海中部的11个观测站(沿海站2个、浮标站9个)风场资料、ECMWF细网格风场预报资料,运用相关性分析、TS评分以及误差检验方法,系统评估中韩航线海域灾害性大风数值模式预报的偏差特征。研究结果表明:(1)ECMWF模式与实况风速相关性较高(R=0.81,通过99%的信度检验),但风速预报系统性偏强1 m/s,TS评分值随风力增大呈显著下降趋势,8级以上大风空报率、漏报率均超过50%。(2)风向预报呈逆时针偏转特征,风综合预报准确率较风向、风速单独准确率低50%以上。(3)空间分布显示:近海站点风速误差显著高于远海,南部沿海站点更加明显。

    Abstract:

    Using wind field data from 11 observation stations (2 coastal stations and 9 buoy stations) in the central Yellow Sea and ECMWF fine-mesh wind field forecast data during 2016–2021, this study systematically evaluates the bias characteristics of numerical model forecasts for disastrous gales in the China-Korea route sea area by applying correlation analysis, TS score, and error verification methods. The results show that: (1) The ECMWF model has a relatively high correlation with the observed wind speed (R=0.81, passing the 99% confidence test), but the wind speed forecast is systematically overestimated by approximately 1 m/s. The TS score shows a significant downward trend as the wind force increases, and both the false alarm rate and missing report rate of gales above force 8 exceed 50%. (2) The wind direction forecast exhibits a counterclockwise deflection feature, and the accuracy of the comprehensive wind forecast is more than 50% lower than the individual accuracy of wind direction and wind speed forecasts. (3) The spatial distribution indicates that the wind speed error at offshore stations is significantly higher than that at offshore stations, with the southern coastal stations being more pronounced.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
文章历史
  • 收稿日期:2025-09-09
  • 最后修改日期:2025-12-15
  • 录用日期:2025-12-15
  • 在线发布日期:
  • 出版日期: