ICT to get Dual Pol. radar

[rdale]

Another delay.


SITE MOD START DATA START DATE
Wichita, KS (ICT) Jan 4 Jan 16
Phoenix, AZ (KIWA) Jan 24 Feb 6
Fort Polk, LA (KPOE) Feb 14 Feb 27
Morehead City, NC (KMHX) Feb 28 Mar 13
Chicago, IL (KLOT) Feb 28 Mar 13

 

[ahaberlie]

Well excuse my enthusiasm but I am particularly interested in the higher angle Zdr values.

The ZDR columns, as the presentors called them, will be able to show liquid (oblong) precip above the freezing layer, suggesting high positive vertical velocities which could be associated with an updraft.

What this means is you can now see reflectivity, velocity, and vertical/aloft evidence of an updraft/tornado.

This would be extremely helpful in pinpointing a possible tornado's location, no?

 

[Bart_Comstock]

Well excuse my enthusiasm but I am particularly interested in the higher angle Zdr values.

The ZDR columns, as the presentors called them, will be able to show liquid (oblong) precip above the freezing layer, suggesting high positive vertical velocities which could be associated with an updraft.

What this means is you can now see reflectivity, velocity, and vertical/aloft evidence of an updraft/tornado.

This would be extremely helpful in pinpointing a possible tornado's location, no?

If you are trying to find a tornado on radar by looking above the freezing layer then you are looking too high. Finding rotation at that height is just a sign that there is mid level rotation in the storm but it does not mean that there is a tornado on the ground or even low level rotation. Just finding the updraft on radar doesn't mean you have found the tornado's "pinpoint" location.

Honestly the real problem with trying to "see a tornado" with a fixed radar station is that more often than not the storm will be to far away from the radar site, or there will be to many obstacles blocking the radar's view of the ground. As you get farther from the radar site the lowest scan the site can make increases in altitude. The beam that a site makes does not curve with the surface of the earth, it travels in a straight line. Take the Hammon, OK tornado from this year. While the storm was producing a tornado its radar signature did not look impressive. This is because the storm was not very tall and was so far from the radar site that the radar beam was passing well over the top of the tornado and it's parent meso. Bellow is a radar snapshot of when the tornado had just touched down. Before this point the storm had been producing numerous funnels and showed visual signs of rapid rotation.

VELOCITY - > http://www.srh.noaa.gov/images/oun/wxevents/20100308/05srm_2321.jpg
REFLECIVITY - > http://www.srh.noaa.gov/images/oun/wxevents/20100308/05refl_2321.jpg
In the images above the tornado would be SW of Hammon or just bellow the very center of of the image.

Another thing to note is that when we see a mesocyclone on radar it is already pretty easy to pinpoint where in the storm the tornado will be. We can already find the updraft and rotation in a storm with the current WSR-88 radar sites.

 

[Patrick Marsh]

Another thing to note is that when we see a mesocyclone on radar it is already pretty easy to pinpoint where in the storm the tornado will be. We can already find the updraft and rotation in a storm with the current WSR-88 radar sites.

Hate to nitpick, but you'll have an idea of where the supercell tornado might be located. All thunderstorms, including those without mesocyclones have a non-zero probability of generating a non-mesocyclone tornado, such as along the lines of the leading edge of gust front, etc.

(There is debate in scientific circles about whether non-mesocyclone tornadoes are *actual* tornadoes, but that's a different discussion.)

 

[Bart_Comstock]

Well of course it is just part of the bigger picture. I was actually just relaying information I heard from the NWS presentation on the website. They said that this could be used to find evidence of a convective updraft.. now it may be tilted away from the possible ground location of the updraft.. but in conjunction with regular reflectivity and velocity data.. couldn't it be helpful?

I don't see it being to much more helpful beyond what we already can see when it comes to trying to locate a convective updraft. We can do that pretty well as of now. Rdale knows more on the subject than I do. IIRC the big advantage of being able to the shape of hydrometers will be in the detection of hail and what size it is. Differential Reflectivity (ZDR) is going to help us greatly in the identification of hail shafts and where exactly in the storm the large stones are and how big they may be.

Another big advantage with the dual-pol radars will be with identifying winter precip types and whether or not its making it to the ground.

With correlation coefficient it is possible to see tornado debris signature, but again, the further away from the storm the radar is the harder it is for it to see low and thus this will not come into play unless the storm is close to the radar site.

Dual-Pols also will give us a better handle on rainfall rate estimates and lead to better flash flood warnings with an enhanced lead time.

The WSR-88 radars only sends out and listens to radio waves of a horizontal orientation where as a Dual-Pol does both horizontal and vertical polarization.

 

[Bart_Comstock]

Hate to nitpick, but you'll have an idea of where the supercell tornado might be located. All thunderstorms, including those without mesocyclones have a non-zero probability of generating a non-mesocyclone tornado, such as along the lines of the leading edge of gust front, etc.

(There is debate in scientific circles about whether non-mesocyclone tornadoes are *actual* tornadoes, but that's a different discussion.)

Thanks Patrick. I was thinking way too narrow minded and completely forgot about landspouts and non-mesocyclonic waterspouts.

 

[Patrick Marsh]

Thanks Patrick. I was thinking way too narrow minded and completely forgot about landspouts and non-mesocyclonic waterspouts.

Easy to do...these are the ones we spend most of our time worrying about!

 

[Jeff Snyder]

As Bart noted, the primary benefits of the dual-pol upgrade for most of the public will be better accumulated precipitation estimates (R-Kdp or R-Zdr/Kdp are much better than R-Z, in general) and better hail identification. The latter may more broadly be expressed as better hydrometeor identification.

There are a number of polarimetric signatures associated with storms and supercells that have been identified (e.g. Kumjian and Ryzhkov 2008, Romine et al. 2008, Kumjian et al. 2008) -- Zdr columns, rho_hv and Zdr rings (or half-rings), Zdr arcs, Zdr and Kdp columns, and the tornado debris signature (TDS). As Bart noted, the TDS is only really seen when the tornado is very near the radar -- we see it quite a bit in our mobile radar data, but we're also always <30 km from the tornado. There are several cases in WSR-88D data (from KOUN), as well as some C-band radars (the 5/10/10 OU-PRIME dataset has some good examples). The Zdr arc is currently thought to be caused by drop size sorting resulting from rain falling through an environment with non-negligible storm-relative helicity (Kumjian and Ryzhkov 2009).

What I think many on this board are looking for, myself included, are signatures that provide some prognostic ability. Joey Picca presented some preliminary work at the 2010 SLS in Denver a couple of weeks ago looking at the volume of the Zdr column relative to hail and tornado occurrence (see Picca and Ryzhkov 2010); Matt Kumjian also showed some results from an examination of the polarimetric characteristics of hail (see Kumjian et al. 2010). My dissertation research is focused on the time evolution of polarimetric signatures (essentially examining how the Kdp columns, mid-level rho_hv rings, etc., are affected by changes in CAPE and shear), though I'm only in the relatively early stages of this. Some images of nice examples of some of these polarimetric signatures (primarily those that are seen >3 km AGL) are available in our conference paper (extended abstract) from SLS - see Snyder et al. 2010

FWIW, I also contend that most people away from the warning desk (heck, perhaps even at the warning desk) will find it quite difficult to make sense of Kdp, Zdr, and rho_hv data in a real-time situation (e.g. while chasing, arm-chair chasing, etc.) . In isolation, save for perhaps Kdp, it's difficult to know what the data mean. For example, to get the most value from Zdr, I think most folks will want to see Zdr AND ZH, or rho_HV AND ZH. Again, the rainfall estimates should be considerably improved, and the hydrometeor classification likely will be popular, but I think the individual products will be considerably less popular with most chasers. When you have the time to thoroughly examine all products, certainly, you can learn a lot more about the hydrometeors / scatterers in and around the storm, however!

EDIT: Freezing level identification may also be useful, particular in winter precipitation situations -- Zdr and Kdp are very useful in this case!