Rating Tornado Intensity Based on Moblie Radar

[Shawn Schuman]

It came up and I understand there was a great deal of debate, but still, they went with EF5 eventually. How was it okay then but not now? I don't mean to throw anyone under the bus, but it's hard to imagine how this could just be coming about now after a number of tornadoes have already been upgraded and data has been publicly released.

 

[Eric Edgar]

Putting in one more plug for a dual rating system. Everybody stays happy, studies needing to use the data have access to whichever system fits their sampling methods. Maybe it would even reduce the mass hysteria that seems to have evolved with the fact that Central OK has been hit by 3 EF5s in 3 seasons (but only 1 if counting the "old way") compared to 1 F5 from 1983-2010 - I have had several conversations with my non-weather addict friends about this fact and their opinion that it is due to global warming etc etc. The reality of it is that the EF rating probably means very little to the practicing scientist nowadays. But it means a lot to the lay public.


The radar (probe?)-based system could be named after a famous meteorologist who pioneered sampling the immediate tornado environment.

 

[Mike Johnston]

Why not use the best instrument you have? It's kind of like when measuring competitive swimming times changed from stop watches to electronic timing systems. Did that corrupt the database of swimming records? I hardly think so; it just improved the quality of the measurements going forward.

 

[Chance Metz]

In tis case the NWS is missing the boat. Why not use all avalible data to get the most accurate intensity of a tornado you can? Nothing is perfect but it is silly for them to now say the EF rating never should have been increased due to protocol that needs to be updated.

 

[Elaine Spencer]

Speaking of changing, or calling into question, tornado ratings based primarily on damage indicators....

http://www.joplinglobe.com/topstori...ineers-release-study-of-Joplin-tornado-damage

The ASCE report says that, due to lesser quality building construction that made structures vulnerable at lower wind speeds, NO damage truly indicative of EF-5 level winds could be found; that only 4 percent of the damage reached EF-4 proportions; and that 83 percent of damage was caused by winds of EF-2 or less.

Does this mean Joplin was not "really" an EF-5 after all? I wouldn't jump to that conclusion. The NWS does not intend to change its rating, and there were (from what I understand) other DIs besides structural damage (e.g., concrete parking blocks moved/tossed).

But this report does open up a whole new can of worms. If a tornado initially rated EF-3 can be upgraded based on data uncovered later, then I'd think it would also be possible to downgrade an EF-5 or EF-4 based on later data. Whether anyone would actually do this is another story.

 

[Jake Orosi]

If hard data indicates a tornado's EF rating should be lowered, then of course that is what should happen.

The EF rating by damage estimation is frankly dodgy at best, and is severely impacted by independent variables that need to be but can't be controlled, like differing localized construction standards. But that's always been a problem; we've put up with it because we had no other way to determine tornado wind speeds outside of guessing from the visible damage.

I really don't see why people whose motivation is scientific wouldn't jump on direct radar data like a free lunch. Which is what the local office correctly did in the case of El Reno. If the administration of NOAA actually has a policy ordering that the most accurate data is to be disregarded in favor of findings that conform the most to historical determinations - that just seems incredibly backwards to me. But I suppose it's to be expected - the administrators are administrators, not scientists, and are naturally more concerned with smooth sailing than anything else.

Damage estimates aren't going away of course; there's simply not enough money to put a DOW on every single tornadic storm; there are some WFO's that will likely never have one at all. But when it's there and has data to share, raid the commissary! If nothing else, having measured wind data will help us to adjust our damage estimates to be more accurate.

 

[STexan]

I keep hearing "Mobile DOW" and how great that it is where it's implemented. But if we are going to rate tornado damage on radar wind speeds, and we are only going to place real emphasis in mobile dows, then I don't see this is going to work out in terms of providing meaningful data for the future. The numbers will be skewed. If we can't use data from NWS stationary radar at least out to 75% of it's effective range, I don't see future data as being reliable because there will never be mobile dow on all the larger tornadoes across the US. The recorded data will be reliable, but the UNrecorded data that was "ignored" because a mobile unit was not on the storm will only serve to skew the existing newer data. (Oddly, some select twisters in OK and KS [where the mobile dows reside] will no doubt become "more severe" beginning with season x)

There are too many nuts out there looking for any excuse to implement more government regulations and taxes to counter "global climate change" that will be "supported" by the rapid escalation of tornado intensities in OK and KS if the weather science community is not careful in how they rate tornadoes moving forward.

 

[Jeff Duda]

I keep hearing "Mobile DOW" and how great that it is where it's implemented. But if we are going to rate tornado damage on radar wind speeds, and we are only going to place real emphasis in mobile dows, then I don't see this is going to work out in terms of providing meaningful data for the future. The numbers will be skewed. If we can't use data from NWS stationary radar at least out to 75% of it's effective range, I don't see future data as being reliable because there will never be mobile dow on all the larger tornadoes across the US. The recorded data will be reliable, but the UNrecorded data that was "ignored" because a mobile unit was not on the storm will only serve to skew the existing newer data. (Oddly, some select twisters in OK and KS [where the mobile dows reside] will no doubt become "more severe" beginning with season x).

That might work if the WSR-88Ds that comprise the NEXRAD network had 500 m diameter antennae or were also X-band and placed every 30 km apart a la a CASA type network. But that is not likely to happen in our lifetimes. The WSR-88Ds are S-band with 1 deg. beamwidth. This means they pretty much cannot resolve any tornadoes that are not very close to the radar (probably less than 10-20 nmi for the largest of tornadoes...and the smallest of tornadoes will pretty much always be too small to be resolved). I think the El Reno tornado would've been directly sampled had the beam been lower. Given its peak size, there were easily several consecutive azimuths that covered the tornado. However, the beam was above 2000 ft ARL, so likely the beam was sampling the low-level mesocyclone and not the tornado itself.

 

[Jake Orosi]

The decisions on how best to pursue science shouldn't even consider how resulting data would be used for this or that political purpose. It does not belong in the discussion.

What needs to be done:

Continued use of damage estimates and the EF scale

Use of mobile radar to determine which actual measured wind speeds are associated with which damage-derived EF ratings

According revision of EF scale wind speeds

According adjustment (if necessary) of historical ratings

 

[Shawn Schuman]

Speaking of changing, or calling into question, tornado ratings based primarily on damage indicators....

http://www.joplinglobe.com/topstori...ineers-release-study-of-Joplin-tornado-damage

The ASCE report says that, due to lesser quality building construction that made structures vulnerable at lower wind speeds, NO damage truly indicative of EF-5 level winds could be found; that only 4 percent of the damage reached EF-4 proportions; and that 83 percent of damage was caused by winds of EF-2 or less.

Does this mean Joplin was not "really" an EF-5 after all? I wouldn't jump to that conclusion. The NWS does not intend to change its rating, and there were (from what I understand) other DIs besides structural damage (e.g., concrete parking blocks moved/tossed).

But this report does open up a whole new can of worms. If a tornado initially rated EF-3 can be upgraded based on data uncovered later, then I'd think it would also be possible to downgrade an EF-5 or EF-4 based on later data. Whether anyone would actually do this is another story.

You're right, there were concrete parking stops near St. John's ripped up and thrown as far as 50 or 60 yards as well as several manhole covers pulled up and thrown. It makes you wonder how many homes of "superior construction" are actually out there. I'd guess not many at all.

Also keep in mind the direct quote from the EF-scale proposal.

The technology of portable Doppler radar should also be a part of the EF Scale process, either as a direct measurement, when available, or as a means of validating the wind speeds estimated by the experts.

 

[Dan Robinson]

You could also make a convincing analogy to hurricanes, which are sampled and rated based on data from dropsondes and satellites. Even if a Category 5 in the middle of the Gulf weakens to a depression before landfall (doing no damage), it still goes on the books as a 5, simply because it truly was that intensity at some point in its life.

 

[Andy Wehrle]

What we have here is a clash between the desire to use the best data possible (using direct measurements where available) versus the desire to achieve climatological consistency.

Grazulis said it best: "With consistency, one can at least understand the problems with the data." (The Tornado: Nature's Ultimate Windstorm, Ch. 7, p.144)

 

[STexan]

I just don't want skewed data that may affect future actions and policies.

 

[Sean Ramsey]

This is a good debate and it's kind of a fun problem to try to solve. I lean more to appending the current formula using the best info we have and maybe developing a population impact rating so older data doesn't get tarnished and it gives you the ability to use newer data sources like the mobile radar when it is applicable, plus shows the impact the tornado had on the populus as a whole.

Example: Use the higher of the EF damage rating or radar scanned rating (if available) and add the population impact rating (scale of 1 to 5) then divide it by two. So if you have a gigantic tornado in the middle of nowhere and somehow gets an EF-5 damage rating or radar indicated winds rating, but does relatively little damage and gets a population impact rating of 1 then it is rated a 3. If it tears through a city or town and does EF-3 damage, maybe it gets a population impact rating of 5, so the overall rating would be 4.

 

[Shawn Schuman]

You could also make a convincing analogy to hurricanes, which are sampled and rated based on data from dropsondes and satellites. Even if a Category 5 in the middle of the Gulf weakens to a depression before landfall (doing no damage), it still goes on the books as a 5, simply because it truly was that intensity at some point in its life.

The only difference, of course, is that all hurricanes can be monitored. That isn't at all the case with tornadoes. But still, I think accuracy is more important than consistency with a scale that isn't consistent to begin with. Not to mention that the EF-scale proposal explicitly stated that radar data should be used when available.