A few questions about radar signatures

[Brandon Schmidt]

Sorry I post a lot of questions here, but I just can't find the answers anywhere else because I sometimes don't know what to search and I know you guys will know but anyways heres a few questions I have about radar:

1) I was thinking about this but couldn't come to a conclusion; the radar starts at 0º and slowly spins to 360º I assume so shouldn't there be a line where products don't line up correctly in between the 0º and 360º radial due to time differences? Iv'e never seen this so it certainly must not be the case but I'm curious as to what is. (looking for a somewhat more detailed account of what goes on during that 4.5 minutes or whatever the VCP time is.)

2) Occasionally I see small areas of lower reflectivity extending from the radar in opposite directions, primarily at night due to greater amounts of lower reflectivity, that are usually perfectly parallel or perfectly perpendicular to the wind fields in the velocity product at the time. Could anyone explain why this is because I can't exactly wrap my head around it.

reflectivity: View attachment 7748
velocity: View attachment 7749

3) I have seen this only a few times and it is easier shown than explained but in short, there are small waves on reflectivity with corresponding alternating velocity directions on the velocity and I have no idea what they are...gravity waves, large horizontal vortices maybe? Does anyone know what they are and what their significance is. Also, are these related to the northward moving bands of low reflectivity seen in the warm sector leading to south-westward moving storms along or ahead of the cold front? (don't know what those are or what they are called either!)

reflectivity: View attachment 7750
velocity: View attachment 7751

I also have a loop but was unsure how to attach a video.

Thanks for any input as it is greatly appreciated!!!

 

[Brandon Schmidt]

Hmmm, also wondering how to post higher quality photos...let me know if you know how, thanks.

 

[mkavulich]

I can give you the answer to question #1:

I can only speak from my experience with a research radar, but it was an S-band radar like NEXRAD and our scan strategy was similar to http://en.wikipedia.org/wiki/NEXRAD's scan strategies. You don't see a discontinuity because it only takes about 10 seconds to make a 360 degree scan. You must remember that scans are taken at a number of different levels, and between each scan the radar is not emitting/receiving for one revolution as it repositions itself vertically for the next scan. That means for a typical 14-elevation scan strategy, the radar must spin 28 times to complete one cycle, which, assuming a rotation rate of 0.1 Hz (one revolution every 10 seconds), takes almost 5 minutes: just about the amount of time between base reflectivity scans for NEXRAD.

I am not sure if it is standard for the NEXRAD network, but our scans began and ended at true north, as you suspected.

For question number 2, I am not a radar expert so it's a bit beyond me.

And as for question 3, I assume they are gravity waves, as they fit the pattern you would expect, but I'm not 100% sure without seeing better imagery or having a better idea of the synoptic pattern. It would make sense with a large area of thunderstorms to the northeast that there could be a substantial cold pool in the area, and the gravity waves traveling on top of this cold pool, but again, this is a bit of conjecture.

 

[Shawn Schuman]

Hmmm, also wondering how to post higher quality photos...let me know if you know how, thanks.

Marcus Diaz posted a nice guide for that.

http://www.stormtrack.org/forum/showthread.php?29833-How-to-post-full-photos-up

I prefer using imgur.com for hosting, but you can use anything.

 

[Jeff Snyder]

Sorry I post a lot of questions here, but I just can't find the answers anywhere else because I sometimes don't know what to search and I know you guys will know but anyways heres a few questions I have about radar:

1) I was thinking about this but couldn't come to a conclusion; the radar starts at 0º and slowly spins to 360º I assume so shouldn't there be a line where products don't line up correctly in between the 0º and 360º radial due to time differences? Iv'e never seen this so it certainly must not be the case but I'm curious as to what is. (looking for a somewhat more detailed account of what goes on during that 4.5 minutes or whatever the VCP time is.)

Every PPI or scan at a given elevation angle must start and "end" somewhere. As you noted, this can introduce a discontinuity when two adjacent radials are maximally separated in time. As a previous poster noted, however, the time it takes an 88D to complete a scan in one of the "precip" VCPs isn't very high, so one may not notice the discrepancy. I know that with our mobile radars, however, even with RaXPol doing 180 degrees per second and only taking 2 s to do a full sweep, we see significant discontinuities sometimes between the "starting" radial and the "ending" radial. This is particularly true close to the radar where the radar resolution volume is smallest.

2) Occasionally I see small areas of lower reflectivity extending from the radar in opposite directions, primarily at night due to greater amounts of lower reflectivity, that are usually perfectly parallel or perfectly perpendicular to the wind fields in the velocity product at the time. Could anyone explain why this is because I can't exactly wrap my head around it.

Are you pointing out the area of reduced reflectivity along the 0 kt isodop (i.e. 0 in radial velocity)? If so, this is probably the result of clutter suppression/removal scheme. Since the Doppler spectrum tends to have a max near 0 m/s in these cases, the clutter removal algorithm removes the "energy" near 0 m/s (0 kts, 0 mph, whatever) and, thus, ends up removing much of the signal, reducing the reflectivity factor.
[/QUOTE]

 

[David Biggar]

Doppler velocities are measured through the shift in frequency between the return and the emitted frequency from the radar. This frequency shift (see Doppler Effect) happens with objects that are moving towards or away from the radar. If you don't want some of the science/math behind it, skip to the bottom of the "To Long, Didn't Read" TLDR section...

[TLDR]
The relationship between a target's motion and the radar beam can be expressed as (making up my own variables here):
DiV = TV cos (A)

Where DiV is the doppler indicated velocity, TV is the "true velocity" (if you were actually there measuring how fast the wind is going with an anemometer), and A is angle between the wind direction and the doppler beam.

The cosine in that equation is important. If the wind is moving directly towards/away from the radar:
DiV = TV cos(0)

Cos(0) = 1 which means the velocity detected by radar is the same as the "true velocity".*
*-As a radar beam moves farther from the radar, it gains height, so that true velocity could be a wind speed at 3000ft rather than near the surface.

Now if the wind is perpendicular to the beam, the angle is 90 degrees...cos(90) = 0...so Doppler indicates no velocity, even though the object is moving. An object needs to have some motion towards a radar (an angle < 90 degrees) to create a shift in the frequency, if not doppler paints it as if it was stationary. Now that I have confused you with math, let me try to explain with some pictures...

As mentioned, anything moving perpendicular to the radar beam does not shift the frequency...so if a target was flying in a perfect circle 10 mi around a radar, the radar would never detect the object as having a velocity (since it isn't moving towards/away from the radar...it is always perpendicular to the beam). The moment the target starts slightly moving towards the radar, the frequency starts to shift. By this theory, if there was a perfectly circular tornado, and it was not throwing anything out/in, and was directly over the doppler, you wouldn't see it as it is not creating that doppler shift (it may be ripping the doppler out too, but the velocity image would look glitchy in that case ).


I took the image you uploaded, and circle overlay. On this circle I put arrows representing what doppler "sees." The arrow size relates to the speed being detected by the radar:

Did you notice that all the arrows are pointed at/away from the radar? That's the point...in fact, I am going to bold this since it is the 3rd time I have typed it: doppler measures the velocity of an object if it has some sort of motion towards or away from the radar.

Since we know that the winds in the "low velocity" areas on radar are most likely moving perpendicular to the radar beam, we can start to get an idea of the general flow:

From this point, it is all about visualizing the rest of the wind field, which involves some basic vector work, but since I hate math let's jump to a perfect example of a wind field and the corresponding doppler image:

Notice how the wind arrows along the white portion of the display are perpendicular to the beam.

[/TLDR]

Nutshell, any area of low velocity is due to the target moving perpendicular to the radar beam, which does not create a frequency shift for the doppler to detect. Please forgive any spelling/grammar errors