NWS Warning Cancellations

[rdale]

If a warning is cancelled before expiration, can it still count against the scores if nothing verifies?

I notice on a periodic basis that NWS offices will cancel 5 minutes before a 60-75 minute warning expires and don't understand why... Chicago tonight issued a SVR for Kankakee / Iroquois counties at 11:36pm until 12:30am. It was cancelled at 12:29am. Does it make sense? Even stranger - the warning for Newton / Jasper was issued at 10:59pm through 12:15am. It was cancelled at 12:15am!

ILX: Stark at 11:48 til 12:30, cancelled at 12:25.

- Rob

 

[Chauncy Schultz]

This sounds like a case of incorrect wording. Instead of saying "the severe thunderstorm warning for john doe county has expired," the warnining meteorologist apparently used the option of making the text read that the warning was cancelled. That, or in the cases where the warning was cancelled five minutes or so before the warning expired, that may have been very intentional, as the storm could well have been out of the county, and there should be an ending severe weather statement for the warning anyway. For all intensive purposes, the public most likely does not realize the difference anyway.

As for the verification, once the warning is out, it needs to be verified. No matter if it is cancelled two minutes or 58 minutes after issuance, it needs to be verified.

 

[rdale]

"As for the verification, once the warning is out, it needs to be verified. No matter if it is cancelled two minutes or 58 minutes after issuance, it needs to be verified."

Thanks - that's what I was curious about. From the end-user perspective a cancellation doesn't matter, as when most people see the storm pass by they (usually) assume it's done ;>

- Rob

 

[Jay McCoy]

Actually I have seen a number of NWS offices Cancel warning's before they expire. They do this if the storm has moved out of the area warned and no more danger exists. No reason to keep a warning up if there isnt a storm to warn on. This is becoming more common as the wx service moves toward location specific warnings and forecasts. In the text they usually say that the storm no longer poses a threat or has moved out of the area. They have been doing this with watches for a number of years as the storms move east (usually) they will clear the counties from the watch.

As for verification. the way I understand it is if a warning is issued, no matter if its for an hour or 10 minutes it counts in the verification process.

 

[rdale]

"Actually I have seen a number of NWS offices Cancel warning's before they expire."

I understand that - it just seems strange to cancel it one minute (or zero minutes) before expiration.

- Rob

 

[Larry J. Kosch]

Weather Channel

I've seen this type of thing going on when a SWW or a TORN watch box is issued. When the storm system passes thru, the NWS will begin cutting off or dropping the counties on the back side of the storms. Thus when you see a storm loop on a map on the Weather Channel, you will see the watch box pop up, the storm line moves thru and the watch box does a "shrinking" act as it gets trimmed thinner and thinner. Finally all that remains of the watch box is a small sliver left. At times when the storms continue to expand and propagate, the NWS will "piggyback" another watch box right next to the older one to keep ahead of the storms.

Thought I would put in my two bits and change... 8)

 

[Jeff Snyder]

Re: Weather Channel

Originally posted by Larry J. Kosch
I've seen this type of thing going on when a SWW or a TORN watch box is issued. When the storm system passes thru, the NWS will begin cutting off or dropping the counties on the back side of the storms. Thus when you see a storm loop on a map on the Weather Channel, you will see the watch box pop up, the storm line moves thru and the watch box does a \"shrinking\" act as it gets trimmed thinner and thinner. Finally all that remains of the watch box is a small sliver left. At times when the storms continue to expand and propagate, the NWS will \"piggyback\" another watch box right next to the older one to keep ahead of the storms.

Thought I would put in my two bits and change... 8)

Well... this is a considerably different scenario than RDale is inquiring about... Watches cover a large spatial area and are in affect for ~6 hours. Warnings are typcally single-county in area and in affect for 45 minutes. It seems "odd" to cancel a warning 5 minutes before expiration. If that's the case, why not cancel part of the warning... I mean, "the NWS has cancelled the sevre thunderstorm warning in affect for Dakota county for the western half of the county. The eastern half of Dakota co. remains under a warning until 9pm"... I guess it's an attempt to cut down on 'false warning time', which I'll define as time that a warned area doesn't experience severe weather ...

 

[Evan Bookbinder]

Originally posted by rdale
\"Actually I have seen a number of NWS offices Cancel warning's before they expire.\"

I understand that - it just seems strange to cancel it one minute (or zero minutes) before expiration.

- Rob

This was a bug in the last warning generation software, and probably went unnoticed by a lot of people in the heat of battle. The problem has already been fixed and the correct "has been/will be allowed to expire" or similar wording will again become available as the latest AWIPS version is fielded (SGF just got theirs on Monday).

Evan

 

[rdale]

Thanks for the inside info - that explains things much better (but still amazes me at the lack of QC for updates, especially in WWA.)

- Rob