Severe thunderstorms (those containing large hail, strong wind gusts, or tornadoes) carry with them the potential for damage from several factors, including lightning, hail, extreme precipitation and high winds. So far, however, it has been difficult to express how the frequency of these types of high impact local weather phenomena might change under a changing climate. Work is needed to connect variables reliably reproduced by climate models with the occurrence of severe thunderstorms Relationships between sounding data, re-analyses and severe thunderstorms and tornadoes in the U.S. are being extrapolated to the rest of the world using a comprehensive new database.
The first goal of the project is to develop a large database of NCAR/NCEP reanalysis-derived soundings that can be used to estimate occurrence of severe thunderstorms around the world. To do this, relationships will be developed between sounding parameters and observed severe thunderstorm occurrence in the United States for recent years, when the severe thunderstorm database is most reliable. These relationships will then be applied to the rest of the world.
Earlier, three years of soundings from the late afternoon were created for the land area of the non-polar region. A preliminary set of equations relating sounding parameters and severe weather occurrence were derived and applied to the entire database of soundings. The implied distribution of severe thunderstorm environments appears reasonable, although the tornado distribution is shifted too far to the east in the US.
More recently, work has begun to cover the globe comprehensively. Complete global coverage at all sounding locations for seven years for four times per day was completed in early April, and additional years continue to be run. Selected regions (eastern North America, Europe, and Australia, initially) have been run through a sounding analysis package for the initial seven-year period, to derive important parameters. With the larger database, better relationships between events and sounding parameters will be developed. I anticipate completing that and a preliminary look at the global distribution of some parameters (e.g., convective available potential energy and the shear of the horizontal winds between the surface and mid-troposphere) by the end of FY 2004. Consideration of the annual cycle of combinations of sounding parameters has begun. Early results from the analysis will be presented at the International Conference on Storms (www.stormsconf.org.au), the American Meteorological Conference on Severe Local Storms, and the European Conference on Severe Storms in 2004.