扰动变量在强对流天气分析和模式评估中的应用——以苏北里下河地区引发龙卷的扰动系统为例

Application of Anomalous Variables to Severe Convective System Analyses and Model Evaluation:A Case Study on Tornado-Producing Anomalous Systems near Lixia River,Jiangsu Province

2016年6月23日14:00—15:00时,江苏省盐城市阜宁县发生了EF-4级有记录以来的强龙卷风过程,造成了当地30 km长,1.7 km宽的严重气象灾害。受局地环境影响,邻近地区在近50年中发生了大于10人次死亡事件的龙卷风11次,其中6次发生于1979年以后。文章首先是:(1)把6次事件日的再分析变量场分解成瞬时气候和扰动两个部分,发现6次事件都与近地面层出现的负高度扰动轴线和扰动冷暖气团的对峙有关,龙卷发生在暖气团一侧;(2)用模式初始场分解的扰动变量和扰动物理量评估欧美模式对产生阜宁龙卷扰动系统的预报能力。后者的结果也发现,当日14时925 hPa高度扰动槽和850 hPa扰动风辐合线作为环境扰动系统,以及湿涡度扰动和湿散度扰动等环境扰动指标量能够突出地表达有利于龙卷风发生的环境条件。欧洲全球模式能够提前42 h稳定预报出后来扰动槽和环境扰动指标量的位置,美国全球模式也可以提前18 h预报出这样的扰动特征。这些扰动变量和扰动指标量能够快速地帮助预报员判断有可能发生强对流天气的大致时空区域和特征。

During 14 ： 00 - 15 ： 00 Beijing Time （ET） 23 June 2016, an extremely tornado with the EF-4 scale hit the Funing County of Jiangsu Province, causing severe damages. Due to the topographic effect, 11 tornado events were recorded causing more than 10 deaths in nearby region in the past 50 years, and 6 occurred after 1979. By decomposing model initial variables and model predicted outputs into temporal climatic state and anomaly, this paper first analyzed the 6 tornado events using observation data, finding that these tornadoes were associated with low-level negative anomalous height trough axes and clashes of anomalous cold and warm air masses. Tornadoes always occur in the anomalous warm air mass. In the second half of the paper, the performances of ECMWF model and NCEP GFS on the tornado process on 23 June 2016 were evaluated using anomalous variables and indices. Analysis results at 14：00 BT illustrated that the tornado occurred at the intersection between a 925 hPa height anomaly trough and a maximum axis of moist vorticity anomaly and moist convergence anomaly at 850 hPa, respectively. The low-level moisture convergence and moist vorticity anomalies could well indicate the moisture and unstable condition for tornadogenesis. The ECMWF model well predicted the trough-axis feature for leading 42 h in advance while the NCEP GFS predicted for leading 18 h in advance, respectively. These anomalous variables and indices could help forecasters indicate a potential time-space domain of severe convective systems more quickly.