Tornado Data Shows EM-DAT Is Worthless
OCTOBER 23, 2020tags: EMDAT
By Paul Homewood
Back to EM-DAT!
As we know, the UN has been trying to pretend that the number of weather related disasters has rocketed since the 1980s. So I thought I would test their database on one specific area, where there is already well established data – tornadoes in the US.
EM-DAT provides a tool to filter records:
This generates a spreadsheet, which further subdivides into various categories of storms, one of which is tornadoes. And when we download the data for 1980 to 2020, we get this graph:
In short, a sudden jump in 1997, since when the number of tornadoes each year has been substantially higher then in the 1980s and early 90s. Obviously climate change then!!
But the first clue that the EM-DAT database is worthless is that the actual number of tornadoes, as recorded by the NWS, is far higher than EM-DAT say – typically about 500 a year, not half a dozen:
Remember that EM-DAT record even tiny events, which for instance affect only 100 people. Virtually every tornado which goes through a populated area will qualify under this definition.
And, of course, the NOAA graph above confirms that the number of tornadoes has been trending down, not up.
Plainly the EM-DAT database is worthless and should not be used for comparing long term trends.