NWS "Storm Based" Warning System (and Issues with)

[twarner]

Mike Smith up a point about NWR that I wanted to expand on but didn't want to hijack another thread.

The problem with the $20 NWR is that it quickly ends up in the trash due to all of the overnight warnings that don't pertain to or of no interest to the user.

I believe that the NWS "Stormed Based" warning system is actually a step back as far as weather radios. I'm sure many can remember the weather radio before they had SAME codes. Continuious alerts all night for storms moving through counties in and around you all through the night. You were either be sleep deprived or you turned it off.

The NWS basically has taken a step back to these old days with the "Storm Based" alert system. There were times when there were multiple warnings out for the county (but wouldn't be relevant for your area). Then you would be subjected to a reissue of these warnings when they move into another portion of the county (that were not relevant).

I had a relative that I gave one of my old weather radios (w/SAME) who finally turned it off due to the excessive mounts of non-relevant warnings. One night I even turned it off because I believe the risk was low and got tire of it going off for marginal svr storms that were not going to affect me.

I suppose the "Storm Based" alert system overall is a good step forward for Emergency Managers and for certain segments of the population. But it would seem that there needs to be an advancement in the weather radio that knows your location and whether you are in the warning box.(either you pre-program you Lat/Lon) or has some type of internal GPS.

2nd I think the local NWS service needs to rethink how it issue their warnings. At times it seemed like they would be issuing 3 successive warnings for 1 storm as it moved through the county. I think that it would be better to issue 1 warning for those portion of the county that are being affected and for those they believe will be affected by this storm - especially when they know that a majority of people are likely asleep.

 

[rdale]

The NWS basically has taken a step back to these old days with the "Storm Based" alert system. There were times when there were multiple warnings out for the county (but wouldn't be relevant for your area). Then you would be subjected to a reissue of these warnings when they move into another portion of the county (that were not relevant).

Just because NWR hasn't kept up with the times -- means that this is a step back. Siren notification can now be restricted to those in the path. Sending spotters out can now be vectored in towards the polygon without needing to interpret radar.

But it would seem that there needs to be an advancement in the weather radio that knows your location and whether you are in the warning box.(either you pre-program you Lat/Lon) or has some type of internal GPS.

I'm not sure that the SAME code stream includes the polygon, but I agree it's something that is needed.

2nd I think the local NWS service needs to rethink how it issue their warnings. At times it seemed like they would be issuing 3 successive warnings for 1 storm as it moved through the county.

That's much better than one big, LONG, warning covering the whole county. Issuing new ones, especially when you are talking TOR, means people are getting the latest warning information too.

 

[Chris Nuttall]

The NWS basically has taken a step back to these old days with the "Storm Based" alert system. There were times when there were multiple warnings out for the county (but wouldn't be relevant for your area). Then you would be subjected to a reissue of these warnings when they move into another portion of the county (that were not relevant).

Suppose you live in the SE corner of a county, but a storm is only affecting and going to affect the NW corner of the county. Under the old system, the entire county would be placed under a warning...whether you were under a threat from the storm or not. How is this type of warning more relevant than the current system? Under storm-based warnings, you may receive the warning via the SAME codes, but technically, you may not actually be in the warning. As previously mentioned, hopefully, this is an area where SAME technology can be improved. However, IMO, a warning is being issued for any part of a county that you live in is relevant, and if you're not paying attention to what's going on for whatever reason, turn on the TV and find out what's happening.

2nd I think the local NWS service needs to rethink how it issue their warnings. At times it seemed like they would be issuing 3 successive warnings for 1 storm as it moved through the county. I think that it would be better to issue 1 warning for those portion of the county that are being affected and for those they believe will be affected by this storm - especially when they know that a majority of people are likely asleep.

Current NWS Directives state that SVRs should only have a valid time of 30-60 minutes. We have not yet gotten to the point that we can precisely predict the exact evolution of a storm beyond this time frame without a rather sizeable amount of uncertainity. Isn't it a better idea to allow forecasters the flexibility to change the warning box and/or reissue to meet the situation rather than blanket a large area that has no reason to be under a warning?

Storm-based warnings also provide a great benefit during MCS situations. You can draw one large polygon to encompass an entire bow echo (just an example) and only have to issue updates for the one box rather than issue separate warnings and follow-up statements for every county along the line.

 

[Jeff Snyder]

The Huntsville NWSFO was (or still is?) using the first number in the FIPS codes to break up a single county into different sections, thereby allowing more precision in the NWR. See http://www.srh.noaa.gov/hun/nwr/Morgan_Sub_County_NWR.php for more info. In my opinion, this would be a good step forward, since it will help bring NWR more into line with polygon-based warnings. In Chris's example above, then, the people in the SE part of the county wouldn't hear the warning tone since that part of the county isn't in the polygon. It seems to be implemented such that the current FIPS code for each county would end up being the "whole county" code, so that those who do NOT change to a new part-specific code will still end up being warned. Regardless, this allows for counties to be broken into 9 sections, something that would be useful for some of the huge counties in the southwestern US (and Cherry Co, NE).

It sounds like a relatively big GIS undertaking given the number of counties in the US, but perhaps each NWSFO could do the county break-up for every county in their CWA.

 

[rdale]

It's a good start - however getting that map to people, and having them understand it enough, seems unlikely. Remember NWR is the method for less than 10% of the population getting warnings.

 

[Mike Smith]

I'm going to get myself in trouble again expressing "politically incorrect" thoughts in meteorology, but here goes:

1. To my knowledge, you cannot program the sub-county codes into a WR-SAME system (please correct me if I am wrong) at the low end receiver (the $19 discussed previously) and, perhaps, the high end receiver. I believe they use full counties only.

2. Even if one can program them, almost no one does (based on substantial anecdotal evidence) and it would take a HUGE education program to get this message across.

3. It is not worth doing #2 because almost no one gets their warning from NWR. It is less than 5% of the population, which has been reconfirmed from recent studies. Television continues to be the dominant source of storm warnings. Every study I have seen confirms this.

4. There is an ongoing thread on stormtrack.org ( http://stormtrack.org/forum/showthread.php?t=18030 ) about new software solutions for storm chasing. Over time, these will migrate into new solutions for the public.

If people in any substantial numbers used NWR, I would be in favor of seeking a solution to this problem but, if after 40 years (first NWR's were installed in 1968) they still have <5% market penetration, it is time to move on to better solutions to the alerting problem.

 

[Bobby Prentice]

The NOAA Weather Radio SAME codes have always been a stupid idea. They cover too large an area (tied to counties) and nobody knows what their SAME code is. SAME codes have the potential to cover areas smaller than whole counties, but have never been implemented for whatever reason. Why couldn't they have used zip codes instead? Zip codes cover smaller geographic areas tied to population density (~43,000 of them) and, most importantly, virtually everyone knows their zip code.

NOAA Weather Radio is an antiquated technology that has been usurped by cell phone and other wireless technologies. As of 2009, nearly 85% of the U.S. population subscribes to a cell phone service. A 2005 federal law requires every cell phone sold to have a GPS chip. A cell phone with a GPS chip would enable warnings down to street level, even if the phone moved. This is the easiest, most effective way to reach the masses. The Federal Government just hasn't realized or reacted this fact yet.