Collaborative assimilation experiment of Beidou radiosonde and drone-dropped radiosonde based on CMA-TRAMS

Abstract

Based on the China Meteorological Administration's Tropical Regional Atmosphere Model for the South China Sea (CMA-TRAMS), the authors conducted a collaborative assimilation forecasting experiment utilizing both Beidou radiosonde and drone-dropped (HAIYAN-I) radiosonde data in September 2023. Three assimilation experimental groups were designed as follows: Beidou radiosonde assimilation, drone-dropped radiosonde assimilation, and collaborative assimilation of Beidou and drone-dropped radiosonde data (hereinafter referred to as “Beidou-drop”). Additionally, a control group of operational forecasts without these data assimilations was set up. The results indicate that the operational forecast path in the control group deviated northward from the actual path. Besides, the Beidou-drop group showed the most significant improvement in terms of forecasting the typhoon path at 60 to 90 h lead times. Specifically, the 72 h and 90 h path errors were reduced by 66.8 and 82.4 km, respectively, resulting in a much more accurate forecast of Typhoon Haikui's landing point, at the coastal junction of Fujian and Guangdong. Furthermore, the collaborative assimilation revealed a notable impact on improving the forecast of wind and rain associated with Haikui's landfall, aligning more closely with the real case. A marked rise was also seen in the precipitation score of the Beidou-drop group, where the 50 mm TS (threat score) of the 72 h lead time increased from 0.33 in the control experiment to 0.75, and the 100 mm TS rose from 0.18 to 0.39.摘要基于中国气象局南海台风模式CMA-TRAMS, 开展了北斗探空和无人机下投探空协同同化预报试验, 分别开展了北斗探空同化, 无人机下投探空同化, 北斗和下投探空 (beidou-drop) 协同同化试验. 结果表明, CMA-TRAMS业务预报路径较实况偏北, 协同同化了北斗和下投探空的路径预报改进效果最显著, 对60–90小时路径预报均有改进, 72小时和90小时路径误差分别较减小66.8 km和82.4 km, 且台风海葵的登陆区域更趋于向实况 (闽粤沿海交界处); 协同同化了北斗和下投探空对台风“海葵”登陆风雨影响更为显著, 与实况更加吻合, 降水评分提升明显, 72小时预报的50 mm TS评分由对照试验的0.33提高至0.75; 100 mm降水TS评分由对照试验的0.18提高至0.39.