On the eighth anniversary of the notorious Joplin Tornado, I have a number of thoughts about weather forecasting, storm warnings and related issues.
As many of you know, the Joplin Tornado, the subject of my second book, is the most deadly tornado since the government’s tornado warning program began in the late 1950’s. Scores of excess deaths occurred because of terrible flaws in the warning system that day.
Unfortunately, we have not learned some of the lessons from Joplin. As discussed in When the Sirens Were Silent, one of the issues was their overuse of the tornado sirens. When interviewed by KMOV TV in St. Louis a few days later, I criticized St. Louis’ emergency managers (EM) for sounding the tornado sirens the day before 50 miles behind a tornado!! When they interviewed the EM for their story his reply was, “We’d rather be safe that sorry.” Nonsense! What they are unwittingly doing is training people to disregard the sirens.
|Yesterday, via Twitter. Tornado was headed for St. Louis|
So, yesterday evening, when I was covering the St. Louis area tornado on Twitter, I turned on KMOV and they were commenting on EM’s sounding the sirens outside the tornado warnings! It is a shame things have not changed in the intervening eight years.
Much has been made of Monday’s tornado overforecasts (including mine). I’m going to cover that in a separate posting. What I want to comment on now is the over-coverage of the tornado forecasts and the under-coverage of the flood forecasts. As I walked into the barbershop yesterday I believe I overheard someone say, “I wasn’t expecting all this flooding.” I’ve often wondered why meteorologists under-cover floods. My best guess is that tornadoes threaten lives immediately, often occur during the day, and are “photogenic” on television.
Floods, on the other hand, take at least hours to develop, often develop at night and – because they close roads – are tough for news crews to get to. Helicopters don’t like to fly in extremely heavy rain. Yet, when all is said and done, it is almost certain the floods caused by the rains of the last three days will do more damage – and displace more people – than the tornadoes. Finally, it takes a basic knowledge of hydrology to forecast floods and that is something most meteorologists don't learn in school. Flood forecasting and flood coverage is an area weather science and broadcast meteorologists need to improve.
This last item is addressed to meteorologists: Challenge yourself to stop over-relying on the computer models. I strongly recommend divorcing yourself from the models for 24 hours and then try to make short-term forecasts (24 hours) in a changing weather situation completely without them in differing weather conditions. My willingness to disregard the models allowed me to make a superior heavy rain forecast Monday. It can be done – and, it will save humiliation when the models are wrong.