Tornado History Project - Over 50,000 tornado maps. Statistics from 1950-2007.

The SPC (Storm Prediction Center) has recently added 2006 data to their historical tornado archive . A quick glance through the documentation reveals 2 new data elements have been added:

• Pl - A categorization of path length calculated as Pl=2*log(Length) + 1
• Pw - A categorization of path width caculated as Pw=2*log(Width) + 5

Additionally, 2 data elements present in past releases have been removed:

• DPI (Destruction Potential Index)
• Code (the meaning of this element was unknown)

Over the next few weeks I’ll be comparing the 2006 file with the 2005 file to see if any changes have been made to past (pre 2006) data. Once everything is squared away, I’ll add 2006 data to the TornadoHistoryProject website. Hopefully, this will be done by the end of September (or before), so keep checking back.

I have also obtained a separate historical archive from the NOAA (National Oceanic and Atmospheric Administration) that includes more detailed path information for multi-county and multi-state tornadoes (I am not yet sure of the derivation of this archive; I’ll post more when I have more information…) Essentially, lat/lon coordinates are given when a tornado passes from one county or state to another. This will result in a more accurate representation of tornado paths (particularly for long paths.) Having said that, there are a number of discrepancies between the two historical archives that need to be resolved. I may end up adding the expanded path data in separate year by year releases as each year’s data is cross-checked between the two databases. However I do it, this will definitely be a labor intensive project so it may take a while to fully complete it. I’ll post more details as I go along.