Hurricane Alex

[rdale]

I agree, but it's only like a 25 minute "discrepancy" ... As the satellite and radar positions do indeed have some sort of latency (6 minutes for radar and 30 minutes or so for satellite).

Actually there's very little latency on radar. It's about 15-45 seconds usually from scan complete to display. Satellite takes about 10 minutes.

With supercells we see it too, we see the tornado on the ground, and no "couplet" and the tornado dissapates, then the massive couplet appear on the next scan 6-7 minutes later.

That isn't data latency at all. That means the couplet wasn't visible in radar imagery until 6-7 minutes later. There is no "delay" in radar images.

 

[Jack Beven]

I've been reviewing my footage this morning, which is precisely timestamped, and actually, the lull in Guadalupe Victoria lasted from about 12:25 am to 12:45 am CDT. As I mentioned above, the lull was very clean and pronounced.

Interestingly, this doesn't jibe that well with the NHC's 1 am CDT position, which is literally at Guadalupe Victoria (plug it into Google Maps to see). The center was actually past us by that time and we were getting hammered by hurricane winds on the backside around then. Given this, I'm wondering if the NHC might want to adjust the position at that time in postanalysis.

Josh, please send your obs along when you get the chance. My colleague Richard Pasch will be doing the post-analysis on Alex.

One comment is that if the eye passed over Victoria at about 12:30 AM, it was probably about 0.1 degrees west or west-southwest of there at 1 AM. We probably will be able to tweak the best track a tenth of a degree.

Jack Beven

 

[Josh Morgerman]

Josh, please send your obs along when you get the chance. My colleague Richard Pasch will be doing the post-analysis on Alex.

One comment is that if the eye passed over Victoria at about 12:30 AM, it was probably about 0.1 degrees west or west-southwest of there at 1 AM. We probably will be able to tweak the best track a tenth of a degree.

Jack Beven

Very cool. I'm glad the info could be useful. I'll send it along ASAP.

What was the lowest barometric pressure you recorded and at what precise time was that? Were you able to determine what elevation you were at, so you could adjust that barometric pressure accordingly?

No pressure readings, unfortunately.

 

[Joe Dorn]

Actually there's very little latency on radar. It's about 15-45 seconds usually from scan complete to display. Satellite takes about 10 minutes.

Would you please explain this statement... It is my understanding that a complete WSR-88D volume scan takes about five minutes,varies with mode, and in the fast moving systems this produces a 'graininess' that is very apparent as in crossing a coast line or IH-35 in my case

I believe there is a mode that allows new scans to replace older scans but this can cause confusion since there is a five minute difference between 'adjacent' tilts.

There is probably very little delay once a volume scan is complete to delivery time...

I would like a reference please...

 

[Jeff Snyder]

There is probably very little delay once a volume scan is complete to delivery time...

I would like a reference please...

You get radar data BEFORE the entire volume completes. In other words, you don't need to wait for the entire volume to be collected before the data get delivered to you. For example, GRx downloads data on an elevation-by-elevation basis, AFAIK. Each elevation angle only gets updated every 4.5-5.5 mins (depending upon VCP 11/12/211/212), but the data you get to see may only be as "old" as 45 seconds from the end of that elevation scan.

 

[Josh Morgerman]

A NWS-meteorologist friend of mine just sent me this radar shot from 12:26 am CDT, when we were in the eye (in Guadalupe Victoria). Pretty cool shot.

I had calculated the eye must have been ~3 or 4 mi across, since we were in the middle of the it, the lull lasted 20 mins (12:25 am to 12:45 am), and the system was moving at 10 mph. My estimate corresponds well with this image: based on it, my friend estimates 8 mi (N-S) by 4-5 mi (E-W).

This was definitely the smallest eye I've been in.

I'm so glad the storm was close enough to the border to get picked up by an American radar site!

 

[Josh Morgerman]

Hi, Jack--

From my perspective on the ground, the 90 kt sounds about right.

I was in Guadalupe Victoria, directly in the path of the center, approx. 30 mi inland. (The 1 am CDT advisory position is right over Guadalupe Victoria, and if I remember correctly, we were experiencing a short lull at that time.)

Despite Alex's spectacular radar and satellite presentations, I'd say we had sustained surface winds in the low-Cat-1 range and no higher. This would correspond fairly well with the NHC's landfall intensity estimate of 90 kt, given that Guadalupe Victoria is a bit inland from the coast. (Disclaimer: I'm not professionally trained to estimate wind speeds, and also, it was very dark, so take this for what it's worth.)

And it looks like our estimate was spot on, judging from the HRD's surface-wind swath. Guadalupe Victoria is at 24.1N 98.2W, and the swath indicates that location had sustained surface (1-min/10-m) winds of 65 or 70 kt: http://www.aoml.noaa.gov/hrd/Storm_pages/alex2010/wind.html

 

[Joe Dorn]

You get radar data BEFORE the entire volume completes. In other words, you don't need to wait for the entire volume to be collected before the data get delivered to you. For example, GRx downloads data on an elevation-by-elevation basis, AFAIK. Each elevation angle only gets updated every 4.5-5.5 mins (depending upon VCP 11/12/211/212), but the data you get to see may only be as "old" as 45 seconds from the end of that elevation scan.

I agree that the last scan may be relatively recent but the first scan, the lower angles, will be several minutes old and any given level will be several minutes from the previous one.

I believe that there is a slant correction in some of the radar programs to correct for the system travel during the scan period.

From the NOAA site:
http://weather.noaa.gov/radar/radinfo/radinfo.html

How often are the images updated?
Image updates are based upon the operation mode of the radar at the time the image is generated. The WSR-88D Doppler radar is operated in one of two modes -- clear air mode or precipitation mode. In clear air mode, images are updated every 10 minutes. In precipitation mode, images are updated every four to six minutes. The collection of radar data, repeated at regular time intervals, is referred to as a volume scan.

a little 'Goggling' will bring up the various modes...

 

[Nick Nolte]

I agree that the last scan may be relatively recent but the first scan, the lower angles, will be several minutes old and any given level will be several minutes from the previous one.

I believe that there is a slant correction in some of the radar programs to correct for the system travel during the scan period.

As I understand it, each "slice" is available almost immediately after that tilt is finished scanning, i.e. you can get the 0.5° tilt as soon as it moves on to the 1.5° and so on until the entire volume is completed. So yeah there is a delay between the completion of an entire volume scan, but you can get the individual slices as they are completed.

You can see this happen in GR3 as well: Your 0.5° will update at (arbitrary time for illustration) 1:00:00, then your 1.5° will update at 1:00:40, your 2.4° at 1:01:20 and so on until the new 0.5 is downloaded at 1:06:00 and the process repeats.

Your composite reflectivity, as I understand it, will only update after every volume scan is complete (6 minutes in VCP21) so it will "lag" behind your BR 0.5 by about six minutes.

 

[Josh Morgerman]

Hey, guys!

My video teaser for this chase is now live on my site: www.icyclone.com

* BACKSIDE!!!!!! Of the hurricanes I've been in, this had the severest backside as compared with the front. We were quite surprised at how ferocious it was immediately following the lull. So that's the best part of the video and that's where the fun starts. Looking at the timestamps, the wind starts picking up again at 12:45 am, and it's a full "hurricane scream" by 12:54 am.

* Size. You'll also notice this cyclone had a very small core. The eye was approx. 4 mi wide-- judging by the very short calm (20 mins) and the storm's translational speed (~10 kt)-- and the worst of the backside passed within ~45 mins.

* Shooting Conditions. Of the 'canes I've chased, this was the hardest to shoot. It was at night and the area was totally blacked out. The only light we had to work with was the car headlights. Given this, the quality of the footage is not so hot-- and almost all of it happens through a windshield. But, hey, at least we got something! (It's a shame it was at night, because the wall cloud of the hurricane's tiny eye was visible to the naked eye as it passed overhead.

I'm going to post the video on my YouTube channel as well...