You were right to be calling this out for so long. I know I’m not a met, but as someone who’s been fascinated with violent tornadoes since I was a kid and has dived into many of these higher-end tornadoes, it always seemed intuitive to me that this kind of contextual damage clearly indicated a violent tornado occurred regardless of the structures it impacted.
I think I’ve said this in this forum before that it should be a focus area of an in-depth study if it isn’t already being looked at. We have enough cases of it happening that I suspect that you would at least find a statistical correlation between the two, especially when it’s matched up to tornadoes that were actually rated EF-4+ based on structural impacts. I imagine for more modern cases that you could also match it up with DOW data or radar data in general?
The part about statistical correlation is the crux of this whole thing. I have severe problems with mathematics, so I am unfortunately not the right guy to do the number crunching, but basically when it comes to true, extreme, “bare soil exposed” grass scouring, I essentially have been doing a long term rough statistics analysis of this phenomenon since 2011, and have looked into many incidents of this occurring prior to that year. What I look at is the structural damage within in the immediate vicinity of the scouring, correlation with other extreme contextual phenomena, the official rating, and the general consensus of how strong the tornado actually was.
While I can’t give you the actual, exact percentages and statistics, I have found a very high undeniable correlation between extreme grass scouring and:
-Buildings that have been completely and totally obliterated. This correlation is almost absolute in consistency. Unless I’m forgetting one, I have never found a single instance of extreme, bare soil grass scouring in the immediate vicinity of homes that have any walls left standing. It simply does not happen, and therefore means it cannot occur at EF3 or lower intensity. In addition to that, I’d estimate that about 90% of the time or more I find bare soil grass scouring, I find slabbed and completely obliterated homes within the immediate vicinity. The few outliers to this are homes that are leveled but not slabbed, and are almost seen in situations involving extreme urban/suburban debris loading, or are within what I call the “scour prone zone” (OK and TX). So while outliers do sometimes occur, the causation of these outliers is obvious, and do not present enough troubling statistical variability to make the correlation between bare soil grass scouring and DOD 10 homes loose or tenuous in any way. This essentially means there is an undeniable link between this degree of damage to homes, and this type of scouring, which equates to an undeniable link between this type of scouring and high-end EF4/EF5 intensity.
-High-end EF4 and EF5 rated events, or suspected high-end EF4 and EF5 events with lower ratings either due to a lack of DIs, poor constructions, or failure of the survey team to assign an appropriate rating. When you only factor in official high-end EF4s and EF5s, the correlation here is admittedly looser. But the key point here is once you factor in tornadoes that are strongly suspected of being high-end events by officials or the “court of public opinion”, suddenly the correlation becomes much, MUCH stronger. The key is knowing when you have a high-end violent tornado that has been underrated. So just because you have an EF3 RATED tornado that only swept away unanchored farmhouses but scoured the ground to bare soil, in no way means that this is actually an example of an EF3 INTENSITY tornado that scoured the ground to bare soil. It simply means it is a default lower rating due to bad construction, and still falls within the extremely high correlation between DOD 10 homes and bare soil grass scouring discussed in the previous bullet point.
-Other extreme high-end contextual damage indicators. When you have bare soil grass scouring, it is almost always correlated with two or more hallmarks associated with high-end EF4 to EF5 intensity. This includes extreme debris granulation, total disassembly of vehicles, stripping of vehicles down to the chassis, vehicle carried a half-mile or more, movement or lofting of massive multi ton-objects (metal storage tanks, train cars, oil drilling equipment, etc), extreme wind rowing, concrete scouring, extreme debarking of low-lying shrubs, and extreme high-end tree damage (total stubbing and/or root balling with 90% or more bark removed). The fact that this type of ground scouring essentially never happens without other extreme contextual phenomena in the immediate vicinity is extremely statistically significant.
I know this is a wall of text and probably isn’t easy to read or decipher all at once, but I hope this further clarifies why I feel so strongly about this. If I had a better math brain, I’d love to do an actual paper and have real numbers to cite for each bullet point, and actually scientifically back this once and for all. Unfortunately, because I can’t do that, the correlation between bare soil grass scouring and high-end EF4/EF5 tornado events remains something that I know to be true, but cannot prove. That’s why I get so frustrated when people try to dismiss it.
