How the tone of SVR/TOR forecasts has changed

[Elaine Spencer]

Today I was poking around on the internet reading some blog posts and articles about the 1974 Super Outbreak (42 years ago today) and also the 1965 Palm Sunday outbreak. The material that I read today enabled me to see, for the first time, exactly how the forecasts issued by SELS/NSSFC prior to each of these outbreaks was worded.

Here is the service assessment done by NOAA after the 1965 outbreak, which contains the text of every severe weather statement issued by SELS and by local Weather Bureau offices during the outbreak:

http://www.nws.noaa.gov/om/assessments/pdfs/palmsunday65.pdf

And here is a blog post with the text of NSSFC's morning outlook on 4/3/74:

http://www.alabamawx.com/?p=100979

What really struck me about these forecasts is how low-key the wording was compared to what one would see today in advance of a major tornado outbreak. For example, the 4/3/74 outlook said that "scattered severe thunderstorms" were expected in all or parts of about 16 states, "with activity spreading eastward during the (forecast) period." That was it -- no mention of the word "tornado". Even that, however, was apparently enough to alert experienced mets that something big was brewing, although to me it doesn't sound all that scary.

Or take the tornado forecasts (they weren't called "watches" yet) issued on 4/11/65: "A few severe thunderstorms with large hail, damaging winds, and one or two tornadoes are expected..." One or two tornadoes? I presume that if similar conditions happened today SPC would have MDT and HIGH risk flags flying well ahead of time, openly mentioning the possibility of violent or long tracked tornadoes, and every storm chaser, and Facebook or Twitter user would be screaming "There's a huge tornado outbreak coming!!" days in advance.

My question is: was the low-key wording in "vintage" TOR forecasts/outlooks due to the forecasters not wanting to freak out the public, or was it more because they really didn't have the tools available yet to forecast just how severe a tornado outbreak was going to be? Or was it less difficult to capture the public's attention back in the days before 24-hour saturation news/weather coverage and social media? It's a topic I find intriguing.

 

[Donnie Beasley]

I would say it's a little of both. There was a time up until about the 50's where weather forecasters were forbidden or strongly discouraged from forecasting tornadoes due to panicking the public. I imagine it took a little while to get away from that way of thinking. The other part is obviously we have come a long way with technology and also what we have learned as far as what conditions are favorable for tornadoes compared to what we knew back then.

 

[Kem Poyner]

Like Donnie said tornado forecasts in the 50's was forbidden. Only military bases were notified if tornados were predicted. In 1948 Robert C Miller and Major E.J. Fawbush started making some very accurate predictions from Tinker Air Force Base using a radar unit taken from an old B-29 Bomber. The "tornado forecasts" were classified. This went on for some time. Then a power struggle started about which Weather Service could make could issue such a statement. I am not sure how long this went on, but the super-outbreak of 1974 was a big game changer. Another contributing factor about that time was Dr. Fujita. He really helped fine tune tornadogenisis.

 

[GPhillips]

While the articles in these links may not directly address the question in the original post, they do give a perspective on the past history of tornado/severe storm forecasting.

http://www.spc.noaa.gov/history/early.html

http://celebrating200years.noaa.gov/magazine/tornado_forecasting/

http://www.cimms.ou.edu/~doswell/TSIII/TSIII_forecast.html

 

[Paul Knightley]

I think this demonstrates the huge leaps and bounds the science of meteorology has taken over the last 50 years or so, in both our understanding of severe weather events, and our understanding of how to communicate with the 'user'. Both these events occurred before the modern era of detailed storm studies - yes, chasing was already underway, especially by 1974, but the visual clues of supercells, for example, was only really in its infancy. And then you have the tools for forecasting - supercomputers in the 70s were less powerful than the smartphone in your pocket now, and that gives a clue about their ability to predict the atmosphere. It would be interesting to see what modern output would show of these events now, and then compare side-by-side with what was available at the time.

 

[Nick Smith]

I can't say for certain about the reason for the brevity of the 1974 forecast, but Allen Pearson (then director of NSSFC) has said in the past that they anticipated it would be a big day. This was after noticing some similarities to the 1965 outbreak.

The state-of-the-art in tornado forecasting (for the time) and the limitations on the teletype circuits they used to transmit these messages may be partially responsible for the length of the messages.According to a paper written about the "Evolution of Severe Thunderstorm Criteria within the Weather Service" (by Joseph Galway) the FAA did not allow plain language transmissions on their Service A teletype circuits until about 1973. You'll notice on discussions before the 90s that there is a significant amount of contractions used. Forecasting techniques tended to be more empirical, comparing a forecast event to similar ones in the past. Information at the time would of included analyzing the surface observations, upper air heights, temperature, moisture and wind. Plotting the Skew-T and hodograph to look for the same things we see today. Some forecast techniques or products were not yet invented, ones that we take for granted today.

The Weather Bureau actually lifted the ban on the use of "tornado" in its forecasts in 1938, but didn't officially begin forecasting them until 1952. That's when the Severe Weather Unit was established at the Bureau's analysis center in Washington, DC. The name was changed to SELS in 1953 and moved to Kansas City, MO in 1954. You can read more about that in some of the links posted earlier and I can point you to some others if you're interested.