Shear Markers

rdale

rdale

EF5
Joined
Mar 1, 2004
Messages
7,562
Location
Lansing, MI
Reposted from Mike Gibson (GR) as he saw it in Nashville and I saw it with Indy coverage last night... Very important. Very.

===

Quick note to any TV people out there: please don't imply the having plenty of "shear markers" in a group indicates a strong rotation! It means that there was noise and/or dealiasing failures, not strength of rotation. I don't know how many times I've heard the Nashville folks say, "wow, look at all those shear markers"!. About the only time you could possibly have multiple valid rotation markers would be something like the Greensburg or Moore tornadoes (>1mi wide) where smaller whirls in the larger rotation might be detectable. AE sorts its 1d rotations by intensity and only keeps the strongest ones in a 5km radius to eliminate all the junk rots.
 
Shear marker question: When a meso marker is up, and the radar is zoomed in, is the maker placed at the center of the circulation or at a point of maximum shear? Since many RAD/ADZ measurements indicate an oval or other irregular shape, how is the meso location determined on a close-in view?

And while I'm at it (perhaps this post should be in the educational area, mods), what is the unit of measurement for shear that reads "/ks" and what is the meaning of this value?

Thanks.
 
I recall on May 12th 2005 with tornado ongoing and driving north into the core afterward Mobile Threatnet was showing about 5 or 6 shear markers near the area of actual tornadic rotation. Following is a shot after I left the storm, but you can still see a few shear markers on the southeast corner of the farthest storm (to the right) in the old tornadic area. It seems with Threatnet that multiple markers is a stronger indicator of rotation, or what do you think here?

May_12_05_0027.jpg
 
Dave Gallaher said:
And while I'm at it (perhaps this post should be in the educational area, mods), what is the unit of measurement for shear that reads "/ks" and what is the meaning of this value?

Thanks.
Click to expand...


It's important to note that the meso and TVS icons lag the refl/vel. data by a few minutes. This is important to realize when storms are moving quickly, resulting in the TVS/meso marker (calculated using the previous scan for the first few minutes of the new scan) being located quite a bit upstream of what the latest reflectivity and velocity data indicate. I've found that the meso/TVS markers will update approx 3-4 minutes after the reflectivity and velocity data come in, until which time the markers will be based on the previous scan.

Re: shear... Remember that shear is really the change in wind velocity with height. Let us consider a situation where wind direction is constant with height. As a crude measurement, suppose the sfc wind is 5 m/s and the wind at 6 km above ground is 30 m/s. The shear is then:

shear = dU/dz = (30 m/s - 5 m/s) / (6000 m) = (25 m/s) / 6000 m = 25/6000 (1/s) = 0.00417 s^-1 = 4.17 per kilosecond (4.17 /ks)

So, the /ks is just the unit meaning "per kilosecond" or "per 1000 seconds". When operational meteorologists
 
Bill Tabor said:
It seems with Threatnet that multiple markers is a stronger indicator of rotation, or what do you think here?
Click to expand...
I was told once that multiple markers mainly mean that the shear was detected by more than one radar - one marker per radar. But I'm sure there are also instances where there are two areas of shear separated by enough distance to create two (or more) markers per radar in the same storm.

Nevertheless, more markers really doesn't mean stronger rotation. I can't recall, but can't you mouse over the marker to get strength info?
 
In my experience, every tornado I have seen has been associated with a storm with Threat Net showing multiple shear markers, with at least one showing a value over 100mph. I have never seen a tornado with a storm that had either a single marker or a marker with a value less than 100mph. So while it may be true in theory that the multiple markers shouldn't indicate a stronger low-level circulation or tornado, in practice they in fact do. Multiple markers don't guarantee a tornado, but ALL tornadic storms I've witnessed have had them.
 
I'm not one for the gizmo crap, but every time we've seen tornadoes when we had the gizmo crap, there were several circles on the screen. I'd have to disagree that multiple shear markers don't mean tornado.

If all else fails, hell...look out the damn window.
 
As regards me I've to agree with Dan,multiple markers don't guarantee a tornado, and what about the value of the shear? Generally when you see the maximum of the scale(orange color) you have high probability to have a tornado. Anyone can confirm?
 
If we were all privy to the details behind the MTN shear algorithm, then we would know for sure. Perhaps they do "add" more markers for stronger rotation, but that would be artificial, and impractical IMO when you can just instead just color-code for stronger shear?

Given my experience with algorithms designed to detect cyclonic shear in Doppler velocity data, there is no magic threshold of shear strength that can discriminate between non-tornadic and tornadic mesocyclones (or tornado "EF" scale, for that matter). Statistics show that there is a higher percentage of tornadic mesocyclones with stronger shear. But they also show that you can get tornadoes with weak shear, and non-tornadic mesos with very strong shear. A lot of this depends on how the radar is sampling the shear, and this is a function of beam-width to core-radius ratio, vortex-center offset from the radar beam center, altitude of the radar beam, separation of multiple radar beams in the vertical, and data quality (e.g., proper dealiasing).

That said, in general, the stronger the meso, the higher likelihood it is tornadic, but this is no guarantee, and don't discount weak rotation. I've seen several cases where there were tornadoes with no rotation couplet in the velocity data.

This is a handy reference: http://www.wdtb.noaa.gov/modules/twg02/index.html
 
Thanks, Rob and Jeff (and all),

Following the link that rdale posted to the GRLevelX trial forum, the site admin made a post in 2004 indicating that eventually the shear markers would be changed: "Later on I'll make their size be driven by the RAD and ADZ."

Does anyone know what happened to this upgrade? It would be a most beneficial enhancement, IMO. I'm still not clear on what the location of the marker means when zoomed in to the point that the marker is physically smaller on the map than the meso it is indicating. Point of greatest shear? Center of meso? Boundary of meso?
 
The markers resulting from multiple radar sites makes a lot of sense - I never thought of that before. Basically you get a marker from each radar site's couplet then, depending on how many 88Ds can 'see' the mesocyclone. So if a strong couplet is showing up on multiple radars, that means there is rotation showing up at several different levels ('slices') of the storm. Closer radars would show low-level circulations, farther radars would show higher ones.

At any rate, a strong storm-scale circulation is inevitably going to show up on all radars in range, which will result in the multiple markers. In most cases, only the strongest mesos with circulations through all levels will do that. The final result being that the multiple markers, while not originally intended as such, are reliable indicators of a dangerous storm with a higher probability of being tornadic.
 
Doesn't the radar angle of inclination, distances to the radars and dealiasing play a factor in the "MYSTERY" of shear marker inconsistency too?...
 
Dan Robinson said:
At any rate, a strong storm-scale circulation is inevitably going to show up on all radars in range, which will result in the multiple markers. In most cases, only the strongest mesos with circulations through all levels will do that. The final result being that the multiple markers, while not originally intended as such, are reliable indicators of a dangerous storm with a higher probability of being tornadic.
Click to expand...
But if this is the case, then the maximum number of markers that any location could receive is dependent on the number of radars that can effectively scan that location. If radar coverage is poor, than a similarly strong mesocyclone may not receive many "wheels of fortune".

There's always a danger in trusting an algorithm without being able to drill down to the base data that is used to create the algorithm output. You cannot do this with MTN.
 
I think it's also important to note that Mike G. was talking about the TV guys who probably use their own radars...TV radars spin pretty quick to get the rapid updates that they do, so the data is quite noisy (as Mike noted in his note passed through rdale). That's probably another reason for the large number of shear markers. For those with access to the AMS journals, this paper is a good one on the subject:
Wood, V.T., and R.A. Brown, 1997: Effects of Radar Sampling on Single-Doppler Velocity Signatures of Mesocyclones and Tornadoes. Wea. Forecasting, 12, 928–938.
http://ams.allenpress.com/perlserv/?request=get-document&doi=10.1175%2F1520-0434%281997%29012%3C0928%3AEORSOS%3E2.0.CO%3B2
 
Last edited by a moderator:
You must log in or register to reply here.
Back
Top