Abstract

In the central region of Switzerland, lying between the Jura Mountains to the north and the Alps to the south, severe hailstorms are a common summertime phenomenon. Eight years of data on these hailstorms show that they are nearly equally divided between left- and right-moving storms. Depending on the exact environmental conditions, the severe hailstorms consist variously of left- or right-moving ordinary-cell storms, left- or right-moving supercell storms, and left-moving storms of an intermediary type (i.e., supercellular in some but not all respects). The left movers of the intermediary type sometimes exhibit a cyclonically rotating echo appendage on the right-rear flank of the storm. This appendage to the left mover resembles a book echo associated with a classic supercell. It is dubbed a false hook, since it has a dynamical configuration substantially different from that of a classic supercell. This difference is demonstrated by the fact that the false hook appears on the wrong side of the left mover for it to be a mirror image of a classic right-moving supercell.

Sounding data show that at bulk Richardson numbers less than 30–50, the right-moving severe hailstorms in central Switzerland tend to be stronger and are more likely to be supercellular, though they are almost never tornadic. The hodograph of the wind in the environment of the storms shows that the winds are about one-half to two-thirds the strength of the winds associated with tornadic storms over the central United States. The wind-shear vector turns generally clockwise with increasing height through the lowest 5–6 km, with a maximum south-westerly wind at about the 3-km level. On days when left-moving storms occur, the shear vector in the lowest 2–3 km of the generally clockwise-turning layer tends to exhibit a slight counterclockwise turning with height.

Model calculations have been carried out for a day on which slight counterclockwise shear was present in the lowest 2–3 km and on which both a right-moving supercell and a left-moving false-hook storm occurred. In addition to rawinsonde data, observations were obtained by three radars, surface stations, and a hailpad network. The model produces splitting storms. The right-and left-moving model storms match the observed storms quite well. The left-hook mover was a false-hook storm, since the separate, cyclonically rotating updraft in the false-hook region does not separate from the left-moving storm. The false-hook appendage is found to consist of updraft and precipitation advected westward and southward in the cyclonically rotating south near flank of the storm. It bounds a cyclonically rotating downdraft on the south side of the storm. When the model simulation is repeated after modifying the environment wind hodograph by reversing the sense of the turning of the shear vector at low levels, so that the environment wind-shear vector turned in the clockwise sense with increasing height throughout the entire lowest 5–6 km, the second split of the left mover occurs much sooner. Consequently, the southern echo appendage is only a transitory feature, and a long-lived false-hook storm is not maintained.

The model simulations indicate that the basic characteristics of thunderstorms in central Switzerland can be realistically reproduced in a numerical model with a flat lower boundary. Hence, the environmental wind and thermodynamic stratification are inferred to be the primary factors determining storm structure. However, the environment supports multiple storm structures, and those storm modes selected by nature at a specific time and location may be determined by very subtle local effects, such as whether the low-level wind hodograph exhibits a slight clockwise or counterclockwise perturbation. Such local variability of the winds is likely related, directly or indirectly, to orography. Such variability is evidently random, though, resulting in the even climatological distribution between left- and right-moving storms.