Finally It Is Stated in the Published Research: The Original Fujita Scale Is Superior to Its Replacement

Ted Fujita with me in 1993. Ted visited WeatherData, Inc on a 
couple of occasions, this was when he presented some initial findings
of his research on Hurricane Andrew. 

Unquestionably, the greatest and most innovative meteorologist of the 20th Century was Dr. Ted Fujita. Among his discoveries were downbursts, the oscillating tornado alley, wall cloud association with tornadoes, suction vortices and The Fujita Scale, among others. Ted was personally modest and low-key about his accomplishments but he evoked a surprising amount of envy -- which has held weather science back in a number of ways. 

While I cannot speak for others, the envy seemed to stem from his willingness to go into the field and do the very hard work of examining the debris, collecting unconventional data like individuals' photos of storms, and the emotional work of speaking to those affected. Because he was usually the only person willing to take this forensic approach, his success in getting funding for his innovative work was often not appreciated by his peers. 

His discovery of downbursts engendered the most envy with his invention of the Fujita Scale to measure tornado and wind storm intensity coming in second. In spite of the fact it worked well, a number of individuals at a major

The original Fujita Scale.
It allowed for both damage-based wind speed estimates plus
measured wind speed measurements.

government laboratory and at a Texas university (based on comments I heard) didn't like the fact that Fujita used little to no modeling when creating the original scale, which resulted in wind speed estimates that were too high. So, they launched what might be called a coup against the Fujita Scale and got it replaced with the 'Enhanced' Fujita Scale (EFS). While it was supposedly more "scientific" than the original, problems surfaced with the EFS rather quickly. 

NOAA

There are many examples of why the enhanced scale has caused a regression in tornado measurements. Take a look at second-to-the-top line in the table above. As originally intended, direct measurements of tornado wind speeds could be used. For example, mobile Doppler radars measured 313 mph winds in the 2013 El Reno (Oklahoma) Tornado. That measurement could be used in the original scale. But, because the enhanced scale does not allow direct measurements to be included, the "official" measurement, based on damage, is 155 mph. That is, at best, misleading. 

Now, we have a paper in the journal Monthly Weather Review that comes to the formal conclusion the original Fujita Scale may be superior to its replacement. In the words of the authors:

... 3) the legacy Fujita scale may provide better wind speed estimates in stronger tornadoes. These findings contribute to community-wide efforts to improve damage-based estimates of peak tornado wind speeds.

My only criticism of the above is the use of the word, "may." There have been a sufficient number of mobile Doppler measurements of violent tornadoes that it is clear the wind speeds in the original Fujita Scale are superior. 

The good news is that the scientific process, in the end, usually works meaning errors get corrected. Hopefully, our profession will now be able to have a genuine, open conversation about fixing all of this rather than discreet conversations in the hallways of scientific conferences.