Nisi, L.; Hering, A.; Germann, U.; Schroeer, K.; Barras, H.; Kunz, M.; Martius, O. (2020). Hailstorms in the Alpine region: Diurnal cycle, 4D ‐characteristics, and the nowcasting potential of lightning properties. Quarterly Journal of the Royal Meteorological Society, 146(733), pp. 4170-4194. Royal Meteorological Society 10.1002/qj.3897
![[img]](https://boris.unibe.ch/style/images/fileicons/text.png) |
Text
qj.3897.pdf - Published Version Restricted to registered users only Available under License Publisher holds Copyright. Download (14MB) |
Nowcasting of hailstorms still poses a major challenge to weather services, because of the limited availability of reliable large datasets and the short spatio‐temporal scales involved. Two novel Eulerian and Lagrangian hail climatologies for the Alps are applied to address important aspects of hailstorms in the Alps: the diurnal cycle, their spatio‐temporal development and the lightning properties. The database contains more than 100,000 ordinary and 30,000 hail storms (2002–2017). Based on that large sample of storms, the diurnal cycle of storm initiation and evolution is studied in the context of orographic forcing and cold‐front occurrence statistics. Results show that, during daytime, storms mainly initiate over the foothills (Prealps) and move towards areas with higher terrain elevations. During night‐time, the storms preferably move from the foothills to the plains. Five out of 16 years of the radar‐derived convective storms show a significant yearly positive hail anomaly, from which two years show relative hail‐initiation maxima evenly distributed over the 24 hour without a characteristic diurnal cycle. Relative hail maxima during night‐time cannot always be explained with a higher occurrence of cold fronts. Time series of storm vertically integrated liquid water content are used to separate ordinary and hail storm development. Differences are found between vertically integrated liquid and its density in cold air‐mass storms. Finally, lightning data from a ground‐based network are combined with the radar‐derived hailstreaks and evaluated with respect to their prediction skill as a function of lead time (flash rate, density, peak current, lightning jumps). Results show that lightning data provide only modest skill‐scores in nowcasting hailstorms. Only the sudden increase in lightning rate (referred to as lightning jump) may be used as additional data for hailstorm nowcasting. However, their application in automatic nowcasting systems remains challenging as the lightning jumps occurs at various lead times in the series.
Actions (login required)
 |
Edit item |