Meteorological False Alarms in Perspective
The Wall Street Journal Tuesday had a article about screening for prostate cancer. I was struck by some of the statistics:
- 30,000,000 men will have the prostate-specific antigen (PSA) test
- 6,000,000 will have elevated PSA levels (a "positive" result)
- 180,000 will be found to have prostate cancer
That is a false alarm rate of 97%!
I don't for one minute think medicine is easy. However, I wish to reassure meteorologists that the false-alarm rate for general public storm warnings is not unique to our science.
Consider the following: The budget of the National Weather Service is $1.5 billion/year.
The cost of screening those 30 million men is $30,000,000,000 per year!! That is twenty times the budget of the National Weather Service*.
Of course, there are apples, oranges and pineapples here. The point is that meteorology does a very good job with limited resources and that false alarms are not confined to our discipline.
*Note: Weather satellites, because they have uses beyond meteorology, come out of the budget of NOAA and NASA.
At AMS in 2015 I had a chief meteorologist at a station in an active weather market tell me "He wouldn't want to have HIS voice delivering tornado warnings that are wrong 75% of the time".
ReplyDeleteRather than waste my breath or brain cells, I simply chose not to engage in such an arrogant, and frankly ignorant statement. I will 100% stand by this statement. Regardless of whether the circulation of a tornado-warned storm ever reaches the ground or not, DOESN'T MATTER. What DOES matter is if NWS pulled the trigger on that storm, and you're in the polygon, that is one MEAN storm you're likely going to remember. And that will increases the likelihood that the next time you get a WeatherCall tornado warning, you'll remember how scary the last storm was, and take protective action.
So, bottom line is we have no interest in working with a broadcaster harboring such an attitude.