On 4 October 2019, giant hailstones of 11 cm were reported in northern parts of Attica in southern Greece. During the same day, multiple large hail reports of hailstones larger than 3 cm and 5 tornadoes were reported in the European Severe Weather Database along the track of a long lived severe thunderstorm that formed over northeastern Peloponnese and moved northeastwards to Attica and Euboea. In this study, we investigate the synoptic and mesoscale weather conditions that led to this rare event by using upper-air measurements from the Athens International Airport, geostationary satellite retrievals, and reanalysis data. Furthermore, the predictability of this rare event is studied through high-resolution simulations performed with three numerical weather prediction models, which are also used operationally by the National Observatory of Athens. The models were able to reproduce the mesoscale environment associated with these severe weather events, showing a highly unstable environment in Saronic gulf with large amounts of convective available potential energy overlapped by a strong deep layer shear. However, the models were not able to fairly reproduce the triggering, track and timing of the severe thunderstorm formation highlighting the great uncertainties associated with the initiation of deep moist convection over areas with complex terrain. Here, we attempt to constrain these uncertainties by applying a diagnostic tool for predicting hail size using an ensemble of high resolution simulations. Our findings reveal similar hail sizes predicted by the numerical weather prediction models to the observed ones and illustrate the usefulness of such hail size diagnostics when applied to a set of high resolution simulations for early warning systems.

2019 年 10 月 4 日，据报道，希腊南部阿提卡北部地区出现 11 厘米的巨大冰雹。同一天，在伯罗奔尼撒半岛东北部形成并向东北移动到阿提卡和欧波亚的长期强雷暴轨迹沿线，欧洲恶劣天气数据库报告了多起大于 3 厘米的冰雹和 5 次龙卷风的大冰雹报告。在这项研究中，我们通过使用雅典国际机场的高空测量、地球静止卫星检索和再分析数据来调查导致这一罕见事件的天气和中尺度天气条件。此外，通过使用三个数值天气预报模型进行的高分辨率模拟研究了这一罕见事件的可预测性，雅典国家天文台也使用了这些模型。这些模型能够重现与这些恶劣天气事件相关的中尺度环境，表明萨罗尼克湾的环境高度不稳定，大量对流可用势能与强烈的深层剪切重叠。然而，这些模型无法公平地再现强雷暴形成的触发、轨迹和时间，突出了与复杂地形区域上深层潮湿对流的启动相关的巨大不确定性。在这里，我们尝试通过应用诊断工具来限制这些不确定性，该诊断工具使用一组高分辨率模拟来预测冰雹大小。