青岛近海海雾预报多模式对比分析及误差订正-海洋预报

青岛近海海雾预报多模式对比分析及误差订正

单位：青岛市气象局, 山东青岛 266003

分类号：P732.2

出版年·卷·期（页码）：2025·42·第三期（1-8）

摘要：

利用2018—2021年4—8月青岛气象自动观测站资料、ECMWF细网格模式资料、CMA-SH9模式资料及中国海洋大学区域大气与海洋短期实时预报系统模式，对比分析不同月份和不同天气形势下青岛近海各模式能见度产品效果预报。结果表明：在青岛雾季的不同月份，各模式的能见度预报效果有差距，整个雾季预报中，中国海洋大学模式最优，其次为CMA-SH9模式，ECMWF模式的预报效果最差。ECMWF模式的能见度预报较实况偏大，漏报率较高，与另外两个模式的差距明显，CMA-SH9模式和中国海洋大学模式的能见度预报较实况偏低，预报的雾日较实况偏多，漏报率较低但空报率较高。出雾时，当天气形势为低压倒槽型、锋面型、均压场型时，中国海洋大学模式的能见度总体预报效果最优，当天气形势为高压底部型时，CMA-SH9模式最优。运用线性回归方法、非线性回归方法和对数回归方法分别对各模式进行误差订正，对比发现非线性回归方法的订正效果最好，修正后ECMWF模式的准确率提高最为明显。

This paper uses the data of Qingdao Meteorological Automatic Observation Station, ECMWF fine grid model, CMA-SH9 model and Ocean University of China Regional Atmospheric and Oceanic Short term Real time Forecast System from April to August during 2018—2021 to compare and analyze the visibility forecasting products in the Qingdao coastal area in different months and under different weather situations. The conclusions are as follows: The visibility forecasts of each model vary in different months of fog season. The Ocean University of China model has the best forecasting effect in the whole fog season, followed by the CMA-SH9model, and the ECMWF model has the worst forecasting effect. The visibility forecasts of the ECMWF model are larger than the actual situation, and there is a high rate of missing reports, which is significantly different from the other two models. The visibility forecasts of the CMA-SH9 model and Ocean University of China model are lower than the actual situation, and the forecast fog days are more than the actual situation. Although the rate of missing reports is low, the rate of false reports is high. The overall visibility forecasting effect of the Ocean University of China model is the best when the weather situation is low pressure inverted trough type, front type and pressure equalizing field type, and the CMA-SH9 model is the best when the weather situation is high pressure bottom type. The linear regression method, nonlinear regression method and logarithmic regression method are used to correct the errors of each model. The comparison shows that the nonlinear regression method has the best correction effect, and the accuracy of the modified ECMWF fore cast has the most obvious improvement.

参考文献：

[1] 王彬华.海雾[M].北京:海洋出版社, 1983.WANG B H. Sea fog[M]. Beijing:China Ocean Press, 1983.
[2] 李建华,崔宜少,李爱霞.山东半岛及其近海大雾的统计与分析[J].海洋预报, 2010, 27(6):51-56.LI J H, CUI Y S, LI A X. Statistics and analysis of heavy fog over the Shandong Peninsula and its adjacent seas[J]. Marine Forecasts,2010, 27(6):51-56.
[3] 张苏平,鲍献文.近十年中国海雾研究进展[J].中国海洋大学学报, 2008, 38(3):359-366.ZHANG S P, BAO X W. The main advances in sea fog research in China[J]. Periodical of Ocean University of China, 2008, 38(3):359-366.
[4] ZHANG S P, XIE S P, LIU Q Y, et al. Seasonal variations of Yellow Sea fog:observations and mechanisms[J]. Journal of Climate, 2009, 22(24):6758-6772.
[5] 丁做尉,张苏平,白慧,等.黄海雾季开始日期的确定及其年际变化[J].中国海洋大学学报, 2011, 41(4):11-18.DING Z W, ZHANG S P, BAI H, et al. Identification of the beginning of the sea-fog season over Yellow Sea and its annual variation[J]. Periodical of Ocean University of China, 2011, 41(4):11-18.
[6] 刘志杰,王炜,赵玥,等.一种能见度融合方法在黄渤海区域雾数值预报中的应用[J].海洋预报, 2021, 38(3):59-67.LIU Z J, WANG W, ZHAO Y, et al. The application of an enhanced fusion algorithm for visibility in assessing the sea fog level in the Bohai Sea and Yellow Sea[J]. Marine Forecasts, 2021,38(3):59-67.
[7] 刘慧,饶晓琴,张恒德,等.全国环境气象业务模式对比检验分析[C]//2016年全国重大天气过程总结和预报技术经验交流会.北京:中国气象学会,中国气象局, 2016.LIU H, RAO X Q, ZHANG H D, et al. Comparative verification of national environmental meteorological operational models[C]//Beijing:Chinese Meteorological Society, China Meteorological Administration, 2016.
[8] 傅刚,王菁茜,张美根,等.一次黄海海雾事件的观测与数值模拟研究--以2004年4月11日为例[J].中国海洋大学学报, 2004,34(5):720-726.FU G, WANG J Q, ZHANG M G, et al. An observational and numerical study of a sea fog event over the Yellow Sea on 11 April,2004[J]. Periodical of Ocean University of China, 2004, 34(5):720-726.
[9] 陆雪,高山红,饶莉娟,等.春季黄海海雾WRF参数化方案敏感性研究[J].应用气象学报, 2014, 25(3):312-320.LU X, GAO S H, RAO L J, et al. Sensitivity study of WRF parameterization schemes for the spring sea fog in the Yellow Sea[J]. Journal of Applied Meteorological Science, 2014, 25(3):312-320.
[10] 尹晓惠,时少英,邓长菊.北京一次大雾天气成因分析及模式预报性能检验[J].安徽农业科学, 2015, 43(1):149-152.YIN X H, SHI S Y, DENG C J. Analysis of heavy fog process and model prediction test in Beijing[J]. Journal of Anhui Agricultural Sciences, 2015, 43(1):149-152.
[11] 赵诗童,时晓曚,吴晓京,等.三种经典夜间陆地雾遥感反演方法的适用性对比分析[J].海洋气象学报, 2021, 41(1):45-57.ZHAO S T, SHI X M, WU X J, et al. Applicability comparison of three classical remote sensing retrieval methods for nighttime land fog[J]. Journal of Marine Meteorology, 2021, 41(1):45-57.
[12] 杨巧兰,夏冬,王静. 2014-2016年珠海能见度的模式预报检验[J].广东气象, 2021, 43(4):33-36.YANG Q L, XIA D, WANG J. Verification of visibility model forecasts in Zhuhai from 2014 to 2016[J]. Guangdong Meteorology, 2021, 43(4):33-36.
[13] 田云菲,杨悦,高山红.一个黄渤海海雾大气水平能见度算法[J].海洋气象学报, 2019, 39(2):24-33.TIAN Y F, YANG Y, GAO S H. An algorithm of atmospheric horizontal visibility associated with sea fog over the Yellow and Bohai Seas[J]. Journal of Marine Meteorology, 2019, 39(2):24-33.
[14] TARDIF R M. Characterizing fog and the physical mechanisms leading to its formation during precipitation in a coastal area of the northeastern United States[D]. Denver:University of Colorado, 2007.
[15] 戚非,闫勇,田应.基于MATLAB的多项式拟合[J].实验室科学, 2006(5):63-65.QI F, YAN Y, TIAN Y. Polynomial fit and interpolation of physical experiment data[J]. Laboratory Science, 2006(5):63-65.
[16] SCHOENFELD D. Partial residuals for the proportional hazards regression model[J]. Biometrika, 1982, 69(1):239-241.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】