[1]Corresponding author address:

Dr. Harold E. Brooks


Norman, OK 73069

[2]NOAA/National Severe Storms Laboratory

Norman, Oklahoma

[3]Northwestern University

Evanston, Illinois

[4]Present affiliation:

Department of Meteorology

The Pennsylvania State University

University Park, Pennsylvania

[5] Note that the conceptual model refers only to the development of low-level mesocyclones, not tornadoes. Brooks et al. (1993) discuss the possibility of non-tornadic, low-level mesocyclones. The existence of such events could introduce a bias into the data set, but in practice, we observe tornadoes and not low-level mesocyclones. Until a large data set of low-level mesocyclones exists, we can only use tornadoes as a proxy for low-level mesocyclones.

[6]For the most part the effect is small. Of the 92 cases presented here, only 13 had values of H which changed by more than 10% and only 7 changed by more than 20% from the values obtained using h = 3 km.

[7]Davies-Jones and Brooks (1993) showed that the evaporation leading to the baroclinically generated low-level mesocyclone occurred in the lowest kilometer of numerically modelled supercells, so that we particularly emphasize low-level evaporation.

[8]EHI = CAPE*H/160000

[9]We note that the current operational mesocyclone detection algorithm in the WSR-88D system is different than that of the NSSL work upon which our study is based. In particular, it has no time continuity constraint and, therefore, "inflates" the number of mesocyclones. This will necessitate some adjustment in future studies.