News:

The AARoads Wiki is live! Come check it out!

Main Menu

NWS proposes elimination of advisories

Started by wxfree, June 27, 2020, 09:09:46 AM

Previous topic - Next topic

wxfree

The National Weather Service is proposing to eliminate the use of advisories, short term forecasts, and special weather statements, and replace them with plain language headlines. People traditionally got weather information from television presenters who know how to interpret those products, but now we have easy access to the information, and most people don't really understand it.  They will keep the watch and warning products, which are easier to understand (although some people still manage to misunderstand).

An advisory is an alert about weather that could cause inconvenience and could be hazardous if care is not exercised (dense fog, light snow, abnormal heat, strong but not severe wind). A short-term forecast is a forecast for up to a few hours giving more detailed information about ongoing significant or severe weather, and is used only when needed.  A general forecast might say that there's a 30% chance of thunderstorms this afternoon and evening, some of which may be severe, while a short term forecast might say that a line of strong to severe thunderstorms is moving from City A toward City B.  A special weather statement is either a notification of a significant event, such as a strong but not severe thunderstorm (called a significant weather advisory) or, somewhat confusingly, an update statement for an ongoing severe event for which a warning has already been issued.  They don't re-issue a warning that isn't expiring, but sometimes want to give an update. I think plain and simplified descriptions would convey the information better.  Right now they squeeze it into these larger categories.

The proposal includes changing the headline wording from emphasizing the advisory or other category, to emphasizing the information.  They would also use more bullet points instead of paragraphs. They have a survey to collect views. It also gives examples of the new way they would word the statements.

https://www.surveymonkey.com/r/publichazsimp
I'd like to buy a vowel, Alex.  What is E?


74/171FAN

I took the survey, and my thoughts are that the word "advisory" should stay in order to clarify that we really need to proceed with caution.

I do not mind changing the Special Weather Statements and Short-Term Forecasts though.  I even think that Special Weather Statements for strong thunderstorms should maybe be a "Strong Thunderstorm Advisory".
I am now a PennDOT employee.  My opinions/views do not necessarily reflect the opinions/views of PennDOT.

kphoger

Quote from: wxfree on June 27, 2020, 09:09:46 AM
They will keep the watch and warning products, which are easier to understand (although some people still manage to misunderstand).

A lot of people get watch and warning backwards.  Common sense seems to be that a "watch" means there's actually a storm that they need to "watch out" for, whereas a "warning" is just to "warn" them that one might develop–which is backward to what the terms actually mean.  I'd rather they rename those two terms to something more intuitive.
Keep right except to pass.  Yes.  You.
Visit scenic Orleans County, NY!
Male pronouns, please.

Quote from: Philip K. DickIf you can control the meaning of words, you can control the people who must use them.

Jim

I took the survey and said I don't mind getting rid of that terminology, but added in a comment that the severe threat terminology needs to change.  The marginal/slight/enhanced/moderate/high scales don't effectively convey the intended threat levels very appropriately.
Photos I post are my own unless otherwise noted.
Signs: https://www.teresco.org/pics/signs/
Travel Mapping: https://travelmapping.net/user/?u=terescoj
Counties: http://www.mob-rule.com/user/terescoj
Twitter @JimTeresco (roads, travel, skiing, weather, sports)

wxfree

I agree that advisories have some value.  I think a combination might be a good approach, keeping the category, but rewording the statements to emphasize that threat rather than emphasizing that it's an advisory.

The watch/warning labels are too often misused.  The way I hear them misused is by changing a watch into a warning, as if there's no difference.  I don't have an idea for how to improve that.  Reversing the terms would still be susceptible to using the more severe level incorrectly, and it would be confusing for those who know what they mean but don't keep up with changes.  I think this depends on finding the right words.

I liked that they went from three levels of five, but the categories are nearly worthless.  Like the terrorism threat colors, people don't know what they mean until it's explained.  I would just do away with the categories and use something like level 1 through level 5 risk assessments.
I'd like to buy a vowel, Alex.  What is E?

SectorZ

Where can I take the survey that gets their sister org, the National Hurricane Center, to stop naming/numbering non-tropical systems?

kphoger

Quote from: SectorZ on June 30, 2020, 05:56:07 AM
Where can I take the survey that gets their sister org, the National Hurricane Center, to stop naming/numbering non-tropical systems?

It's only a matter of time before we hear about Tropical Depression Ayden.    :rolleyes:
Keep right except to pass.  Yes.  You.
Visit scenic Orleans County, NY!
Male pronouns, please.

Quote from: Philip K. DickIf you can control the meaning of words, you can control the people who must use them.

CNGL-Leudimin

One trouble I have with NWS's watches and warnings is that I don't know how to "translate" them into the European color-coded warning system (yellow, orange, red in order of severity). Further complicating things is the way they are issued: The NWS issues a watch when something might occur in an area and a warning when it is already happening while European meteorological services will issue warnings based on forecasts, e.g. the AEMET (Spanish equivalent of NWS) would issue (for example) an orange thunderstorm warning several hours before the storm actually takes place, while the NWS would only issue a severe thunderstorm watch and only when a storm really forms they would issue a severe thunderstorm warning.
Supporter of the construction of several running gags, including I-366 with a speed limit of 85 mph (137 km/h) and the Hypotenuse.

Please note that I may mention "invalid" FM channels, i.e. ending in an even number or down to 87.5. These are valid in Europe.

Scott5114

Quote from: Jim on June 29, 2020, 02:23:42 PM
The marginal/slight/enhanced/moderate/high scales don't effectively convey the intended threat levels very appropriately.

No, they really don't. Here in Oklahoma, "slight" is bad enough that you're going to want to cancel any evening engagements you might have,  "moderate" can cause school to let out early. Doesn't sound very moderate to me. (And why the heck is 4/5 "moderate", anyway?)

Quote from: CNGL-Leudimin on July 01, 2020, 04:13:28 AM
One trouble I have with NWS's watches and warnings is that I don't know how to "translate" them into the European color-coded warning system (yellow, orange, red in order of severity). Further complicating things is the way they are issued: The NWS issues a watch when something might occur in an area and a warning when it is already happening while European meteorological services will issue warnings based on forecasts, e.g. the AEMET (Spanish equivalent of NWS) would issue (for example) an orange thunderstorm warning several hours before the storm actually takes place, while the NWS would only issue a severe thunderstorm watch and only when a storm really forms they would issue a severe thunderstorm warning.

Your difficulty probably stems from the fact that NWS warning types are not intended to indicate the severity of the storm, only the likelihood or imminence of one and type of threat. A severe thunderstorm warning could cover anything from an isolated cell with soaking rain and lighting to a band of storms the length of a state with 60 mph straight-line winds.  Severity is indicated through the bulletin attached to the warning. If a storm has specific dangerous hazards, an additional alert (such as a tornado warning) will be issued. NWS also communicates hazards with coded language such as "particularly dangerous situation", which is reserved for life-threatening storms.
uncontrollable freak sardine salad chef

Jim

Quote from: Scott5114 on July 01, 2020, 06:32:50 AM
Quote from: Jim on June 29, 2020, 02:23:42 PM
The marginal/slight/enhanced/moderate/high scales don't effectively convey the intended threat levels very appropriately.

No, they really don't. Here in Oklahoma, "slight" is bad enough that you're going to want to cancel any evening engagements you might have,  "moderate" can cause school to let out early. Doesn't sound very moderate to me. (And why the heck is 4/5 "moderate", anyway?)

Up here in the northeast, I take "moderate" to mean "this is likely the day with the greatest chance of damaging storms for the whole year, so I better be on top of this" while the less weather-geeky public reads "moderate" so "yeah, no big deal, it's only a moderate risk, carry on".  Granted, media might take it and overhype it too, but that's a different topic.

Quote
Severity is indicated through the bulletin attached to the warning.

I feel like they've done a good job refining the language issued with the watches and especially warnings to differentiate specific types of threats and their likelihood much better.  Useful, if people read them.

Quote
NWS also communicates hazards with coded language such as "particularly dangerous situation", which is reserved for life-threatening storms.

Similarly, I think it's a fairly recent change to be able to replace the standard Tornado Warning, which might be issued when there is just a small chance of a weak tornado indicated by some rotation on the radar, with the Tornado Emergency, which seems to be reserved for a confirmed damaging tornado headed toward a populated area.
Photos I post are my own unless otherwise noted.
Signs: https://www.teresco.org/pics/signs/
Travel Mapping: https://travelmapping.net/user/?u=terescoj
Counties: http://www.mob-rule.com/user/terescoj
Twitter @JimTeresco (roads, travel, skiing, weather, sports)

kphoger

Quote from: CNGL-Leudimin on July 01, 2020, 04:13:28 AM
One trouble I have with NWS's watches and warnings is that I don't know how to "translate" them into the European color-coded warning system (yellow, orange, red in order of severity). Further complicating things is the way they are issued: The NWS issues a watch when something might occur in an area and a warning when it is already happening while European meteorological services will issue warnings based on forecasts, e.g. the AEMET (Spanish equivalent of NWS) would issue (for example) an orange thunderstorm warning several hours before the storm actually takes place, while the NWS would only issue a severe thunderstorm watch and only when a storm really forms they would issue a severe thunderstorm warning.

Technically, a tornado doesn't have to have actually been spotted in order for a tornado warning to be issued.  A warning means that the threat of the condition described is imminent, and it could develop at a moment's notice.  So a tornado warning may be issued with no spotted tornado, if it is determined that the weather conditions are ripe for a tornado to drop down at any time.  A tornado watch, on the other hand, means that those "ripe conditions" haven't developed yet but are likely to soon.
Keep right except to pass.  Yes.  You.
Visit scenic Orleans County, NY!
Male pronouns, please.

Quote from: Philip K. DickIf you can control the meaning of words, you can control the people who must use them.

hbelkins

Quote from: kphoger on July 01, 2020, 10:13:24 AM
Quote from: CNGL-Leudimin on July 01, 2020, 04:13:28 AM
One trouble I have with NWS's watches and warnings is that I don't know how to "translate" them into the European color-coded warning system (yellow, orange, red in order of severity). Further complicating things is the way they are issued: The NWS issues a watch when something might occur in an area and a warning when it is already happening while European meteorological services will issue warnings based on forecasts, e.g. the AEMET (Spanish equivalent of NWS) would issue (for example) an orange thunderstorm warning several hours before the storm actually takes place, while the NWS would only issue a severe thunderstorm watch and only when a storm really forms they would issue a severe thunderstorm warning.

Technically, a tornado doesn't have to have actually been spotted in order for a tornado warning to be issued.  A warning means that the threat of the condition described is imminent, and it could develop at a moment's notice.  So a tornado warning may be issued with no spotted tornado, if it is determined that the weather conditions are ripe for a tornado to drop down at any time.  A tornado watch, on the other hand, means that those "ripe conditions" haven't developed yet but are likely to soon.

Most tornado warnings around here are issued for radar-indicated rotation, and indeed, most of them never turn into funnel clouds or touchdowns. We had an instance several weeks ago not far from here where there was just a strong thunderstorm, not severe, with no associated watch. All of a sudden the radar showed rotation and a tornado warning was issued.


Government would be tolerable if not for politicians and bureaucrats.

Scott5114

I think most of these "radar-indicated rotation" warnings are issued because the NWS office only has radar as a tool, and when you see a hook echo plus the tell-tale red and green signature on base velocity, it's better to issue a warning so municipalities will sound sirens and get people to take shelter, rather than play games trying to figure out if the rotation is at ground level or 50 feet up in the air.

That being said whenever a storm chaser calls into NWS, or there's a clear tornado on the ground being broadcast on local media (through their professional chasers or news chopper), NWS will sub out the "radar-indicated rotation" language for "Spotter confirmed tornado".
uncontrollable freak sardine salad chef

wxfree

#13
This is being rolled out in a strange way.  I've seen no mention of it on the main NWS page, or the Fort Worth office page, which is my local area.  I saw it on the Houston office page.  Random offices around the country have it on their pages.  I was lucky enough to randomly click on the Cleveland office, which was kind enough to give a link to a more informative page, instead of just the survey.  The page links to two more pages with more details.  The text isn't substantially more informative than the PDFs in the survey, but there's a long video I haven't watched yet, so that may be interesting.

https://www.weather.gov/hazardsimplification/

If you're interested in the topic, click through to the reference materials.  They have things like public survey and focus group results.  There's a lot of information there, such as the possibility of adding an "emergency" level above warnings, presumably for things like confirmed severe tornadoes or high-end hurricanes.
I'd like to buy a vowel, Alex.  What is E?

US 89

Quote from: 74/171FAN on June 27, 2020, 11:43:28 AM
I do not mind changing the Special Weather Statements and Short-Term Forecasts though.  I even think that Special Weather Statements for strong thunderstorms should maybe be a "Strong Thunderstorm Advisory".

Or just use the "Significant Weather Advisory" terminology already used by most NWS offices. I do agree that should be its own product.

Quote from: Jim on July 01, 2020, 08:17:31 AM
Quote from: Scott5114 on July 01, 2020, 06:32:50 AM
Quote from: Jim on June 29, 2020, 02:23:42 PM
The marginal/slight/enhanced/moderate/high scales don't effectively convey the intended threat levels very appropriately.

No, they really don't. Here in Oklahoma, "slight" is bad enough that you're going to want to cancel any evening engagements you might have,  "moderate" can cause school to let out early. Doesn't sound very moderate to me. (And why the heck is 4/5 "moderate", anyway?)

Up here in the northeast, I take "moderate" to mean "this is likely the day with the greatest chance of damaging storms for the whole year, so I better be on top of this" while the less weather-geeky public reads "moderate" so "yeah, no big deal, it's only a moderate risk, carry on".  Granted, media might take it and overhype it too, but that's a different topic.

Originally, there were just three levels: slight, moderate, and high. The problem with that was that the slight category was very broad, and NWS/SPC meteorologists wanted a way to discern the upper-end slights from your general run-of-the-mill slight risk. The solution to that was to create a new category that basically meant "enhanced slight", but that got shortened to just "enhanced". The marginal risk was created in order to show regions with low-end severe probabilities that were worth discussion in the SPC's outlook, but didn't quite meet the criteria for slight risk.

The SPC doesn't seem to care much for anywhere west of the Continental Divide, so here in northern Utah we'll get one or maybe two slight risks at most, despite the fact that we'll be impacted by multiple severe thunderstorms over the course of a year. This might be one of the few places where having a severe thunderstorm warning polygon issued for your location is far more common than a watch.

Quote from: Jim
Quote
NWS also communicates hazards with coded language such as "particularly dangerous situation", which is reserved for life-threatening storms.

Similarly, I think it's a fairly recent change to be able to replace the standard Tornado Warning, which might be issued when there is just a small chance of a weak tornado indicated by some rotation on the radar, with the Tornado Emergency, which seems to be reserved for a confirmed damaging tornado headed toward a populated area.

Tornado Emergencies have been around since the infamous 1999 Moore tornado, but they've become more regular lately especially with the Impact-Based Warning system becoming standardized across the country in the past few years. With the new system, there will be a tag at the end of every tornado warning indicating whether the tornado is radar indicated or observed. If the tornado is likely to be especially destructive, an additional damage threat tag can be added to the end of the warning, which can either be "considerable" (resulting in a PDS tornado warning) or "catastrophic" (tornado emergency).

The same system is used in severe thunderstorm warnings, too. All severe thunderstorm warnings are issued because a storm could produce 60 mph winds and/or 1 inch hail, but the newer warnings have wind and hail tags that communicate more specifically the wind strength and hail size to be expected. There's also a tornado=possible tag that can be added when a storm is showing some rotation, but not quite enough to issue a tornado warning.

wxfree

I think there are (or were) technical descriptions for the marginal, slight, etc. categories.  It's beyond the grasp of most people, but it's something like "X% likelihood of a certain wind or hail severity or any tornado or an intense tornado within Y miles of any point within the risk area."  I remember SPC making maps giving that specific information, but they don't seem to now.  There are charts showing the probabilities in general terms.  "Significant severe" refers to a higher threshold, above basic severe levels.  I think there was an enhanced category for tornadoes EF 3 or above.  The probabilities for tornadoes were lower, and the radii were longer, because of the higher risk they present.  On the charts, you get to the higher levels of risk at lower probabilities for tornadoes.  They may have stopped using a radius because they found that they can't forecast accurately enough to make those meaningful.

https://www.spc.noaa.gov/misc/about.html#Convective%20Outlooks
I'd like to buy a vowel, Alex.  What is E?

US 89

Quote from: wxfree on July 16, 2020, 05:36:45 PM
I think there are (or were) technical descriptions for the marginal, slight, etc. categories.  It's beyond the grasp of most people, but it's something like "X% likelihood of a certain wind or hail severity or any tornado or an intense tornado within Y miles of any point within the risk area."  I remember SPC making maps giving that specific information, but they don't seem to now.  There are charts showing the probabilities in general terms.  "Significant severe" refers to a higher threshold, above basic severe levels.  I think there was an enhanced category for tornadoes EF 3 or above.  The probabilities for tornadoes were lower, and the radii were longer, because of the higher risk they present.  On the charts, you get to the higher levels of risk at lower probabilities for tornadoes.  They may have stopped using a radius because they found that they can't forecast accurately enough to make those meaningful.

https://www.spc.noaa.gov/misc/about.html#Convective%20Outlooks

SPC does still make those individual tornado/wind/hail probability risk maps - you can get to them from the outlook page if you mouse over the tornado/wind/hail tabs above the map. The technical definition of those probability forecasts still involves radii - 25 miles, to be exact. If your area is highlighted in a 15% area of severe wind, for example, that correlates to a slight risk, and it means there's a 15% chance that a 58+ mph wind gust will occur within 25 miles of you.

The probability forecasts do include "hatched areas" to illustrate areas where significant severe weather - defined as EF2+ tornadoes, 75+ mph wind, or 2+ inch hail - is more likely. Technically, a hatched area is where there is at least a 10% chance that significant severe weather will be observed within 25 miles of a point. Those are weird because once you start getting up into the higher probabilities, they actually influence the ultimate categorical risk that winds up on the main map. For example, a 30% tornado probability only correlates to a high risk if a significant tornado risk area is also present - otherwise, the ultimate risk level is just moderate.

wxfree

Quote from: US 89 on July 16, 2020, 08:28:14 PM
Quote from: wxfree on July 16, 2020, 05:36:45 PM
I think there are (or were) technical descriptions for the marginal, slight, etc. categories.  It's beyond the grasp of most people, but it's something like "X% likelihood of a certain wind or hail severity or any tornado or an intense tornado within Y miles of any point within the risk area."  I remember SPC making maps giving that specific information, but they don't seem to now.  There are charts showing the probabilities in general terms.  "Significant severe" refers to a higher threshold, above basic severe levels.  I think there was an enhanced category for tornadoes EF 3 or above.  The probabilities for tornadoes were lower, and the radii were longer, because of the higher risk they present.  On the charts, you get to the higher levels of risk at lower probabilities for tornadoes.  They may have stopped using a radius because they found that they can't forecast accurately enough to make those meaningful.

https://www.spc.noaa.gov/misc/about.html#Convective%20Outlooks

SPC does still make those individual tornado/wind/hail probability risk maps - you can get to them from the outlook page if you mouse over the tornado/wind/hail tabs above the map. The technical definition of those probability forecasts still involves radii - 25 miles, to be exact. If your area is highlighted in a 15% area of severe wind, for example, that correlates to a slight risk, and it means there's a 15% chance that a 58+ mph wind gust will occur within 25 miles of you.

The probability forecasts do include "hatched areas" to illustrate areas where significant severe weather - defined as EF2+ tornadoes, 75+ mph wind, or 2+ inch hail - is more likely. Technically, a hatched area is where there is at least a 10% chance that significant severe weather will be observed within 25 miles of a point. Those are weird because once you start getting up into the higher probabilities, they actually influence the ultimate categorical risk that winds up on the main map. For example, a 30% tornado probability only correlates to a high risk if a significant tornado risk area is also present - otherwise, the ultimate risk level is just moderate.

The radius is given in the convective outlook description.  I didn't see it earlier.  I think that information was displayed with the maps years ago, so I was looking for it there again.  The "significant severe" terminology is defined there, too.  I hadn't even looked at the maps in several years.  I needed some catching up.
I'd like to buy a vowel, Alex.  What is E?

michravera

Quote from: wxfree on June 27, 2020, 09:09:46 AM
The National Weather Service is proposing to eliminate the use of advisories, short term forecasts, and special weather statements, and replace them with plain language headlines. People traditionally got weather information from television presenters who know how to interpret those products, but now we have easy access to the information, and most people don't really understand it.  They will keep the watch and warning products, which are easier to understand (although some people still manage to misunderstand).

An advisory is an alert about weather that could cause inconvenience and could be hazardous if care is not exercised (dense fog, light snow, abnormal heat, strong but not severe wind). A short-term forecast is a forecast for up to a few hours giving more detailed information about ongoing significant or severe weather, and is used only when needed.  A general forecast might say that there's a 30% chance of thunderstorms this afternoon and evening, some of which may be severe, while a short term forecast might say that a line of strong to severe thunderstorms is moving from City A toward City B.  A special weather statement is either a notification of a significant event, such as a strong but not severe thunderstorm (called a significant weather advisory) or, somewhat confusingly, an update statement for an ongoing severe event for which a warning has already been issued.  They don't re-issue a warning that isn't expiring, but sometimes want to give an update. I think plain and simplified descriptions would convey the information better.  Right now they squeeze it into these larger categories.

The proposal includes changing the headline wording from emphasizing the advisory or other category, to emphasizing the information.  They would also use more bullet points instead of paragraphs. They have a survey to collect views. It also gives examples of the new way they would word the statements.

https://www.surveymonkey.com/r/publichazsimp

It sounds like they need a three-dimensional model. Time, probability that the event will occur at all, and a probability of the severity. It would seem that they could do this in plain language that at least a statistician would understand. "(By) 0430Z (21:30PDT), 30% (chance of) hail, 40% (chance of the hail that does occur reaching) 20 mm (diameter)" Could be rendered:
Advisories:
HELL 0430Z 30%HAIL 40% to 20 mm

TXtoNJ

I'd change "watch" to something like "tornado aware", then update the warning system to "yellow warning" meaning radar-indicated only, "red warning" meaning tornado spotted, and "tornado emergency" for a powerful storm entering a populated area.

zachary_amaryllis

the 'winter weather advisory' drives me up the wall..

'gee, its january.. i wonder what sort of weather i should expect?'
clinched:
I-64, I-80, I-76 (west), *64s in hampton roads, 225,270,180 (co, wy)

I-55

Quote from: zachary_amaryllis on August 11, 2020, 01:59:58 AM
the 'winter weather advisory' drives me up the wall..

'gee, its january.. i wonder what sort of weather i should expect?'

Indiana says hold my beer
Let's Go Purdue Basketball Whoosh

zachary_amaryllis

Quote from: I-55 on August 11, 2020, 01:16:34 PM
Quote from: zachary_amaryllis on August 11, 2020, 01:59:58 AM
the 'winter weather advisory' drives me up the wall..

'gee, its january.. i wonder what sort of weather i should expect?'

Indiana says hold my beer

sweet  :clap:
clinched:
I-64, I-80, I-76 (west), *64s in hampton roads, 225,270,180 (co, wy)

SSOWorld

https://www.spc.noaa.gov/products/watch/ww0426.html

The watch of the Derecho in Iowa/Illinois last Monday.  PDS? Yes it was.  (No, I don't mean Particuarly damn stupid)
Scott O.

Not all who wander are lost...
Ah, the open skies, wind at my back, warm sun on my... wait, where the hell am I?!
As a matter of fact, I do own the road.
Raise your what?

Wisconsin - out-multiplexing your state since 1918.

SectorZ

Quote from: zachary_amaryllis on August 11, 2020, 01:59:58 AM
the 'winter weather advisory' drives me up the wall..

'gee, its january.. i wonder what sort of weather i should expect?'

They've also dumbed it down severely over the years. In Massachusetts it used to mean 4-6 inches of snow or a tenth to quarter of an inch of ice accretion. Over the past few years it has been invoked for such things as freezing fog causing a coating of ice on the roads or a mere 2 inches of snow, and specifically citing that as the need for the advisory.

It's fun to watch them ruin a meaningful product, ask the masses whether the product is needed anymore, then come out and say "the public says it's not a good product" after they used it for target practice for years.

We need to bring back Ice Storm Watch/Warnings, Snow Advisories, Freezing Rain Advisories, and Heavy Snow Watch/Warnings. Winter Weather Advisories and Winter Storm Watch/Warnings should be for mutli-precip events in the same area. Annoyingly, the Norton (Boston) NWS never used those specific watches and warnings.



Opinions expressed here on belong solely to the poster and do not represent or reflect the opinions or beliefs of AARoads, its creators and/or associates.