NWS Dissemination Re-architecture

[brunota2003]

Just ran across this (the link is posted at the top of the NWS main page):

Welcome to the NWS Dissemination Re-architecture

In this community, you can submit ideas, vote on existing ideas, or add comments.

As a current National Weather Service (NWS) data user, we at NWS are reaching out to you to request your comments as we begin, in the upcoming months, to develop a new service delivery architecture designed to meet your needs. Our primary goal is to fulfill our mission to ensure the protection of life, property, and the enhancement of the national economy, and we seek to improve how we do this by disseminating the necessary data using strategies and methods that make use of new and emerging technology. Users of NWS dissemination services are the most important group affected by this process, and we want you to be involved.

Seeing how we have a topic dealing with modification of watches/warnings, I thought it would be appropriate to post

http://nwsrearch.ideascale.com/

*stickied by vbhoutex*

[brunota2003]

Also of note (for anyone who has not seen this), here is iNWS...which is short for InteractiveNWS:

http://inws.wrh.noaa.gov/

Apparently, you can receive customized text and e-mail alerts for NWS products.

InteractiveNWS (iNWS) is the home of new mobile and desktop innovations of the National Weather Service. This application suite allows NWS partners to receive National Weather Service products in new and innovative ways, such as text messaging and mobile-enabled webpages. iNWS strives to fulfill our mission of protecting life and property by using new technology to reach out to our customers.

But it also says:

PLEASE NOTE:
iNWS is intended for members of community emergency planning and response management (i.e. emergency managers, law enforcement, fire and emergency responsers, transportation and safety managers, public officials), Skywarn Net Control operators, government partners of NWS offices, and members of the electronic media. Do not register if you do not fit this description.

[CrazyC83]

I wrote this proposal up:

While all severe weather is dangerous, some are more so than others, and the higher levels suggest extremely dangerous events. This is similar to the Slight, Moderate and High Risk levels that the Storm Prediction Center uses for the coverage and degree of the forecasted threat. This would initially apply only to convective systems, but could be expanded for winter storms, flooding and other events.

My thought for Severe Thunderstorm Warnings and general thunderstorm advisories:

Code Green (non-severe) - wind gusts below 39 mph, hail less than 0.50 inches, infrequent or in-cloud lightning

Code Blue (strong, covered by Special Weather Statement) - Wind gusts 39-57 mph, hail 0.50 to 0.88 inches, frequent cloud-to-ground lightning

Code Yellow (severe) - Wind gusts 58-74 mph, hail 1.00 to 1.75 inches

Code Orange (significant severe) - Wind gusts 75-89 mph, hail 2.00 to 3.75 inches

Code Red (extremely severe) - Wind gusts 90-109 mph, hail 4.00 inches or higher

Code Black - Wind gusts 110 mph or higher (should be covered by an Extreme Wind Warning)

For tornado warnings, the color coding should be as follows:

Code Green (not issued, no tornadoes)

Code Blue (weak rotation, covered in Severe Thunderstorm Warning or Special Weather Statement)

Code Yellow - Radar indicated with rotation, no ground truth

Code Orange - Radar indicated intense rotation or tornado confirmed on the ground with little or no damage (likely EF0-EF1)

Code Red (Tornado Emergency if headed for a community of 5,000 or more in the next 10 minutes, or at the discretion of the WFO) - Radar indicated extreme rotation or confirmed on the ground with significant damage or over 200 yards wide (possible EF2-EF3)

Code Black (automatic Tornado Emergency which must be issued as well) - Radar indicated debris ball or confirmed on the ground with extreme damage or fatalities or over 800 yards wide (possible EF4-EF5)

If a situation changes, the WFO is obligated to issue a new update within 1 minute for tornadoes and 3 minutes for severe thunderstorms with an upgraded warning, or within 5 minutes for a downgraded warning.

These are just my ideas to try to get the general public to understand warnings better and the degree of them to separate routine warnings where normal precautions should be taken from those where extreme measures must be taken to save lives (like in Joplin, Tuscaloosa and other places).

[vbhoutex]

I know you put a lot of thought into your suggestions. I understand and like what you have done, but I fear it is more complicated than what the general public will/can absorb. Maybe I underestimate the general public, but over the last 50 years I have seen so much laziness or "it won't happen here" attitude that I wonder what it will take or if the public can be educated well enough to get rid of the negative attitudes. JMHO

[TYNI]

For the severe thunderstorm and tornado warnings, I would leave out the code green and maybe blue. Basically, it is saying there is nothing significant happening. Simple, simple, simple is the way to go. I'm not sure it's the message that needs fixing, but more getting the message out there quick enough.

The only reason I can see for keeping them, is to peak people's interest early enough to start PAYING ATTENTION in case the situation worsens. You and I and others on the board may read the mesoscale discussions and convective outlooks every day, and know what to expect (in fact, the media had done a really good job during the past two outbreaks to not only communicate the high probability for the next day or that day, but to provide consistent, updated coverage leading to the events and during the events).

I only go through severe warnings here in Canada, so maybe I don't have a right to comment on warnings where lives are at stake all the time.
JMHO

[CrazyC83]

The reasoning I have in the color coding is that all warnings can be treated by the public similarly and seen that way regardless of the level of danger, and while all are dangerous, some are much more so than others.

For example, a Severe Thunderstorm Warning has the same general headline (not text) whether it is for localized 60 mph winds (enough to knock down tree limbs) or if it is for a massive derecho with widespread 110 mph winds (enough to destroy mobile homes and cause significant structural damage across entire communities). Likewise for hail, the same wording is used whether it is 1 inch (enough to dent car windows) or 5 inches (enough to go through roofs and destroy vehicles).

The same is true for tornado warnings - the same headline would be used for radar-indicated rotation which is only enough to put down a small, brief tornado if there is one, and for a long-lived, extremely large confirmed wedge tornado over a heavily populated area with an obvious debris ball signature. That may be why many didn't realize the danger in situations like in Joplin and Tuscaloosa if they were not watching television at the time.

Faster warning updates would be required (1 minute for tornadoes, 3 minutes for severe thunderstorms) as soon as data suggests an upgraded warning level is necessary.

[supercane]

Sorry to rain on your parade, but this interesting, well-written proposal, but unfortunately doomed due to complexity.
Currently people understand and respond poorly to warnings, and yet the proposal above splits each type of severe warning into six different levels. In the face of an urgent situation remembering the distinction among the different levels would just add confusion.
Another weakness is that the split levels require much better ground truth and imply much better specificity in predicting the storm effects than are currently available.
Thirdly, the different levels convey a message that different actions are needed for each level, when in fact, this isn't necessarily the case. For instance, the scheme above would imply that people need to take a radar-signature only tornado less seriously than other tornadoes, when in fact if a warning is issued you want to emphasize the message "take cover now." For the Springfield, MA, tornado earlier this month, the system would require an awkward update from Code Yellow to Red within 4 minutes, which causes undesirable mixed messages from an emergency/risk management standpoint. Such a system will likely cause some complacency (yellow's not a big deal, there's no evidence now, so I just wait and see). Similarly, the call to action for a severe storm is similar whether the winds are 50 kt or 80 kt: "take cover now."
More updates are always nice, but the requirements above run into both technological constraints(radar scans take generally more than 1 min, interpretation takes time, trends may not be evident early, uncertainty in storm evolution remains) and psychosocial constraints (changing your message too often may add to confusion).
Also, such a proposal would seem counter to the recent NWS movement to simplify advisories, as we saw with the recent consolidation of the winter weather suite (which was a good move) and the desire to change the current advisory map appearing on the main NWS and each WFO's homepage into a simpler three-tier watch/warning/advisory map a la the Alaska Region (which I do not like).
For these reasons, I find the proposal above, while well intentioned, is too complex yet does not adequately address the initial purpose given as its motivation.

[dolebot_Broward_NW]

Sorry to rain on your parade, but this interesting, well-written proposal, but unfortunately doomed due to complexity.

Sorry, this, and exactly this is what is wrong. When you deal with delivering life and death instructions to the General Public you really do have to think about the lowest common denominator. Late night talk show hosts and comedians have murdered the now obsolete Terrorist Warning System, for its similar complexity.

Think about what people need to know.

Do I need to go huddle in the basement, or just stay inside?

I almost want to say that the current severe thunderstorm watch/warning system, and the tornado watch/warning system is workable. It tells people what and where, but I think perhaps too many/too wide of a warned area leads to complacency. You always hear "They keep giving us tornado warnings but never any tornados". That complacency is dangerous.

[senorpepr]

I agree... keep it simple. Albiet different criteria, I frequently get calls from our customers with confusion over the difference between a watch and warning. The other issue is complacency. I agree that different storms have different levels of severity, but let's look at the Saffir-Simpson hurricane scale. I've seen folks here and elsewhere dismiss a storm for "only" being a category two. It would be bad business to make certain severe thunderstorm/tornado warnings "less" serious.