Yeah that PDF's been mentioned on here a couple times before....I like it because it's the only time I've been able to find a photograph that clearly shows the two separate trenches from the Philadelphia tornado.
Other interesting thing of note regarding Andover; it was in its stovepipe phase for quite a while, over a half hour until it hit the AFB. Most violent tornadoes intensify extremely quickly and rapidly (especially if they are multivortex), typically within seconds to minutes of touchdown. Andover was on the ground for at least 15-20 minutes before it exploded into a wedge at McConnell AFB, again if not for that downward draft of air it probably would have dissipated by the time it hit the base.
In the photo of the crumpled car you can see that the grass is bent at an angle.
One thing about New Richmond is that its fatality-to-injury ratio (117:200) was very high. It was even higher than those for the Pomeroy (71:200) and Sherman (73:200) events. According to various accounts the New Richmond tornado was quite visible to onlookers and therefore, per Thomas P. Grazulis (Significant Tornadoes 1680–1991, p. 688), many residents were able to seek shelter in time and thus reduce the potential death toll somewhat. So the fact that New Richmond still produced so many fatalities, along with a high fatality-to-injury ratio, despite being well anticipated is certainly suggestive of an extremely violent tornado.
talkweather.com
The death toll from this is certainly way higher as slaves would not have been included among the fatalities.
I have noticed that there have been instances of tornadoes riding valleys, such as with the Shoal Creek - Ohatchee Valley EF4 Tornado on April 27, 2011 (the second EF4 that was spawned from the Tuscaloosa supercell after the Tuscaloosa Tornado dissipated). Some of these valleys may hold rivers within them. However, I'm not sure if it's just me but the tornado, given that map, didn't appear to track along with the river for a long time, and just happened to pass over the worst possible spot and the worst possible time.The death toll from this is certainly way higher as slaves would not have been included among the fatalities.
This tornado is an example of a worst-case scenario, a tornado moving along a major mode of transport (Mississippi River) and encountering heavy traffic (steamboats); the modern-day equivalent would be a rain-wrapped EF5 following a freeway corridor in Dallas or Chicago during rush hour.
Interesting detail is that a tornado in the 1908 Dixie Outbreak took a path not that far off from this one.
This is perhaps the earliest significant Dixie event that we have on record
2 gendisasters articles on it:
Another article on it: https://www.ustornadoes.com/2017/05/07/natchez-ms-tornado-1840/
I've always found it interesting that this thing appears to have followed the river directly and was right on it for much of its path, I can't think of another instance of a tornado following a large body of water for a significant time; it'd be interesting to study what happens underneath large bodies of water when violent tornadoes go over it or follow it for long periods of time.
storymaps.arcgis.com
Come to think there was a tornado (F4) in the 1953 Flint-Worcester outbreak that spent most of its life traveling over Lake Michigan which makes it impossible to ascertain its exact path length; it'd be interesting if such a thing occurred today and were documented, if a massive tornado traveling over a body of water has any unusual effects underneath, if the vortex extends to the bottom of the lake or not, etc.