提前 6 小时以上准确及时预报雷暴，可大大提高空中交通流量管理效率。为实现这一目标，使用天气研究与预报 (WRF) 模型输出变量对 2020 年乌鲁木齐国际机场发生的雷暴进行了调查。尝试研究 WRF 输出变量对雷暴生命周期不同阶段（积云、成熟和消散）的敏感性。本文考虑的变量包括风速 (WSPD)、复合雷达反射率、回波顶部高度、对流有效势能 (CAPE)、对流抑制 (CIN) 和升力指数 (LI)。发现 CIN 对即将到来的雷暴最敏感。WSPD 对雷暴的发生极为敏感，CIN 紧随其后。岬，CIN 和 LI 对消散的雷暴都很敏感。为改进雷暴预报，提出了一种简单实用的客观雷暴预报方法，即基于上述变量组合的雷暴概率（TSP）预报方法。客观 TSP 预测和人工主观预测的比较表明，TSP 预测在 2020 年夏季的表现要好于人工预测。
Accurate and timely forecasts of thunderstorms at lead times of more than 6 h can greatly improve the efficiency of air traffic flow management. To achieve this goal, thunderstorms occurring at Urumqi International Airport in 2020 were investigated using Weather Research and Forecasting (WRF) model output variables. An attempt was made to study the sensitivity of WRF output variables to the different stages (cumulus, mature, and dissipating) of the thunderstorm lifecycle. The variables considered in this paper include the wind speed (WSPD), composite radar reflectivity, echo top height, convective available potential energy (CAPE), convective inhibition (CIN), and lift index (LI). It was found that CIN is the most sensitive to an approaching thunderstorm. WSPD is extremely sensitive to the thunderstorm occurrence, closely followed by CIN. CAPE, CIN, and LI are all sensitive to dissipating thunderstorms. To improve thunderstorm forecasts, a simple and practical objective thunderstorm forecasting method, i.e., the thunderstorm probability (TSP) forecasting method, based on the combination of the above-mentioned variables, was proposed. Comparison of objective TSP forecasts and manual subjective forecasts indicated that TSP forecasts performed much better than did manual forecasts over the summer period of 2020.
