Outbreak Signatures

[BBauer]

Hi, I saved a bunch of radar images from the April 27th Outbreak like most of you guys do regularly. Anyway, I just had a few questions. I sometimes had the reflectivity side of the dual pane clicked on so the velocity scale is not showing on every one but you get the idea.

1. Are the following holes in the reflectivity scans BWERs?

2. The following is the scan just prior to the tornado entering Tuscaloosa. Why is there a huge orange blob indicating severe wind in one direction? Shouldn't there be a signature of wind in the opposite direction? Where did it go, or what is happening here?

 

[Jeff Snyder]

In the first two images, it's tough to say because I'm not sure the height of the radar beam at the range of the storm (from the radar site). In general, the BWER is a feature you'll see in the mid-levels (or maybe the upper part of what many of us consider the "low-levels"), so, if the storm was a good distance from the radar (indicating the beam was relatively high even on the 0.5 degree scan), it is possible that you were seeing the BWER. You will also occasionally see the hook echo wrap completely around the low-level meso, which may be the case in the 2nd image, and will be seen on the lowest-elevation scans.

The orange velocities on the bottom image are erroneously-unfolded velocity data. The automatic unfolding algorithm typically does a pretty good job, but you'll see errors sometimes when the radial velocities are very strong.

 

[BBauer]

The orange velocities on the bottom image are erroneously-unfolded velocity data. The automatic unfolding algorithm typically does a pretty good job, but you'll see errors sometimes when the radial velocities are very strong.

Thanks for the reply, for the lay person can you explain what you mean by "unfolded"? In one of the threads on the outbreak day I also heard someone mentioning that the tv meteorologist (or the whoever was behind the scenes) was erroneously showing a "stacked" velocity couplet which somehow does not show how much shear is really there, could you explain what "stacked" means too? Thanks a million!

 

[Jeff Snyder]

Because of the way weather radars work, the radar system can only "detect" winds up to a certain speed -- the maximum unambiguous velocity. Suppose that, for a given combination of radar transmitting frequency and pulse repetition frequency (PRF), the maximum unambiguous velocity (here, the same as the so-called Nyquist velocity) is 50 mph. If the outbound velocity component (that is, the component of the wind blowing away from the radar) is 60 mph, the radar will actually see this as a 40 mph wind INbound. In other words, that 60-50 = 10 mph is "wrapped around" the max velocity, ending up as -40 (-50 + 10 = -40), or a 40 mph inbound wind. In essence, the velocity "wraps" around the Nyquist interval; the radar will never calculate winds greater than 50 mph inbound or outbound. The process of "unwrapping" these values is called velocity unfolding or dealiasing.

For the scanning patterns typically used in severe weather situations (VCPs 11 and 12), I think the maximum unambiguous velocities are somewhere in the realm of 50-60 kts. So, if the winds blowing towards or away from the radar are greater than 50-60 kts, the radar will calculate an incorrect radial velocity. The WSR88D radar system uses the Velocity Dealising Algorithm (VDA) to dealias / unfold the velocity data, and it typically does a pretty good job. However, if the radial velocities are significantly greater than the maximum unambiguous velocity (again, typically 50-60 kts, AFAIK, for severe weather operations), if the horizontal shear in radial velocity is particularly high, or if the return radar signal is particularly weak, the data may not be dealiased properly.

That TV met (referring to my post in the 4/27 NOW thread) that was clicking around trying to find the highest radial velocity apparently did not realize that the Level 2 data he was showing was *not* dealiased. As such, it was not possible for the radar to show radial velocities of greater than 50-60 kts. Indeed, even the ABC crew wasn't showing the unfolded / dealiased radial velocity data, but it was apparent that they realized this shortcoming. Some Level 2 radar packages, including GR2 / GR2 AE, perform velocity unfolding. The radial velocity data in the Level 3 data stream are dealiased / unfolded, which you'll see in GR3, on the NWS RIDGE page, and in other locations. The velocity data in your images are unfolded (otherwise, you'd never see orange or blue colors), although not correctly or completely in your 3rd image.

EDIT: I should note that there are a few VCPs (1xx / 2xx) that use different scanning techniques to mitigate some of the effects of velocity and range folding. See this DLOC page and this other one for more information.

 

[BBauer]

Because of the way weather radars work, the radar system can only "detect" winds up to a certain speed -- the maximum unambiguous velocity. Suppose that, for a given combination of radar transmitting frequency and pulse repetition frequency (PRF), the maximum unambiguous velocity (here, the same as the so-called Nyquist velocity) is 50 mph. If the outbound velocity component (that is, the component of the wind blowing away from the radar) is 60 mph, the radar will actually see this as a 40 mph wind INbound. In other words, that 60-50 = 10 mph is "wrapped around" the max velocity, ending up as -40 (-50 + 10 = -40), or a 40 mph inbound wind. In essence, the velocity "wraps" around the Nyquist interval; the radar will never calculate winds greater than 50 mph inbound or outbound. The process of "unwrapping" these values is called velocity unfolding or dealiasing.

For the scanning patterns typically used in severe weather situations (VCPs 11 and 12), I think the maximum unambiguous velocities are somewhere in the realm of 50-60 kts. So, if the winds blowing towards or away from the radar are greater than 50-60 kts, the radar will calculate an incorrect radial velocity. The WSR88D radar system uses the Velocity Dealising Algorithm (VDA) to dealias / unfold the velocity data, and it typically does a pretty good job. However, if the radial velocities are significantly greater than the maximum unambiguous velocity (again, typically 50-60 kts, AFAIK, for severe weather operations), if the horizontal shear in radial velocity is particularly high, or if the return radar signal is particularly weak, the data may not be dealiased properly.

That TV met (referring to my post in the 4/27 NOW thread) that was clicking around trying to find the highest radial velocity apparently did not realize that the Level 2 data he was showing was *not* dealiased. As such, it was not possible for the radar to show radial velocities of greater than 50-60 kts. Indeed, even the ABC crew wasn't showing the unfolded / dealiased radial velocity data, but it was apparent that they realized this shortcoming. Some Level 2 radar packages, including GR2 / GR2 AE, perform velocity unfolding. The radial velocity data in the Level 3 data stream are dealiased / unfolded, which you'll see in GR3, on the NWS RIDGE page, and in other locations. The velocity data in your images are unfolded (otherwise, you'd never see orange or blue colors), although not correctly or completely in your 3rd image.

EDIT: I should note that there are a few VCPs (1xx / 2xx) that use different scanning techniques to mitigate some of the effects of velocity and range folding. See this DLOC page and this other one for more information.

Thanks a million, I really appreciate your response!

 

[George Widener]

Likewise, I also appreciate the detailed response. One of the best things about hanging around this forum is learning new things.

 

[Chip Redmond]

The first radar caption is from KDGX, Jackson, MS and with the city Walnut Grove on the reflectivity, I would say that signature is roughly ~50-60m from the radar. That would likely mean that you are looking at a BWER as Jeff suggested. The same goes for the second image with Fayette, AL from BMX. Good observation. Definitely some impressive radar signatures that day.

Chip