NWS Sandy Service Assessment

[famartin]

Don't kill the messenger...

http://www.nws.noaa.gov/os/assessments/pdfs/Sandy13.pdf

 

[Eskimo Joe]

I have been looking for this for some time.  Thanks for posting it.

[Sunny and Warm]

I've read about 1/2 of it so far, and find it relatively fair and balanced.  Why would you say, "don't kill the messenger"?

[MN Transplant]

 

 

When asked how concerned they were when Sandy was heading for the coast, 58 percent of post-storm survey participants who believed a hurricane watch/warning

was in effect said they were "very concerned."  An almost equal number, 56 percent, who thought another type of NWS warning was in effect (e.g., for coastal flooding or hurricane force winds)

also were very concerned

 

 

 

 

[tdp146]

I think the prototype storm surge inundation map is exactly what is needed to help the public. It's appearance links up well with the usgs sandy mapper done after the storm. I'm interested to see if nhc uses this as an experimental tool this hurricane season.

For me personally, I am still trying to figure out exactly how vertical datum like NAVD88/29 correspond to MLLW/MHHW for my particular location on Long Island. If anyone knows how to do this I would appreciate the help.

[billgwx]

I think the prototype storm surge inundation map is exactly what is needed to help the public. It's appearance links up well with the usgs sandy mapper done after the storm. I'm interested to see if nhc uses this as an experimental tool this hurricane season.

 

Those graphics could be generated on an experimental basis as early as this season...stay tuned.

[N. OF PIKE]

if you are forecasting based on preparing people (perhaps imagining your mom or grand ma) would help some "get it" ,  and want people to take the appropriate precautions for the historical storm  ,  (again not us as wx weenie's) but the pubic,   do you say there is a hurricane warning and the surge will be higher than from any other hurricane that has hit this area or do you say , well public prep be damn'd i'm not changing my tune for the public, we as the NHC will hand this off to the appropriate HPC and local wx offices to handle. I mean again if your priority is NOT getting people prepared in a storm like this, say it's not. Say your priority is sticking to strict definitions , which was a close call anyway, (on how non tropical sandy was) wether it does or doesn't effect public perception, roll the dice (for mom and grandma) and stick to the textbook defintions if that is what your priority is, and if it is I won't fault you, but don't tap dance around it. It's embarrassing.  I mean if a THINK TANK was devoted to figuring out what the public NEEDS to hear in terms of increasing preparation and not falling for the "NHC crying wolf again" they IMO would find out a HURRICANE WARNING would help, they should make the message simple .  Show them this storm surge being consistant with that of a major hurricane that will get there attention, again if you want people prepared. Send a consistent message over and over again from a centralized outlet that any poll or think tank shows the "key words" that are best used to "alert coastal residences" because this was always a coastal flooding/surge threat and that point should have been hammer'd home consistently and repeatedly . I'm not sure what the key words in that message or perhaps accompanied by a slide or picture but dumb it down, if this were to happen next year the NHC would do the same thing and say hey, this is what the text book says, nevermind people don't get what our textbooks definitions mean, its more important to our message to the public that we stick to these destinations bc that is how we talk amongst each other, and were not changing our rules, because we don't do that.

 

I can just say there must be a lot of turmoil inside the NHC about this decision, between those stuck w textbook definitions and those that key words like hurricane is better used to alert people when the surge effects will be very hurricane like, since people know a hurricane =means bad and a subtropical or extra tropical storm means....I dunno is that like a weaker type of storm, its like does the NHC not a have a clue that the general public to a significant degree already thinks they cry wolf too much with storms, why not do your job and get "real" and alert people even if it means going outside (gasp) precious textbook definitions when it will alert people better, seriously.

[famartin]

if you are forecasting based on preparing people (perhaps imagining your mom or grand ma) would help some "get it" ,  and want people to take the appropriate precautions for the historical storm  ,  (again not us as wx weenie's) but the pubic,   do you say there is a hurricane warning and the surge will be higher than from any other hurricane that has hit this area or do you say , well public prep be damn'd i'm not changing my tune for the public, we as the NHC will hand this off to the appropriate HPC and local wx offices to handle. I mean again if your priority is NOT getting people prepared in a storm like this, say it's not. Say your priority is sticking to strict definitions , which was a close call anyway, (on how non tropical sandy was) wether it does or doesn't effect public perception, roll the dice (for mom and grandma) and stick to the textbook defintions if that is what your priority is, and if it is I won't fault you, but don't tap dance around it. It's embarrassing.  I mean if a THINK TANK was devoted to figuring out what the public NEEDS to hear in terms of increasing preparation and not falling for the "NHC crying wolf again" they IMO would find out a HURRICANE WARNING would help, they should make the message simple .  Show them this storm surge being consistant with that of a major hurricane that will get there attention, again if you want people prepared. Send a consistent message over and over again from a centralized outlet that any poll or think tank shows the "key words" that are best used to "alert coastal residences" because this was always a coastal flooding/surge threat and that point should have been hammer'd home consistently and repeatedly . I'm not sure what the key words in that message or perhaps accompanied by a slide or picture but dumb it down, if this were to happen next year the NHC would do the same thing and say hey, this is what the text book says, nevermind people don't get what our textbooks definitions mean, its more important to our message to the public that we stick to these destinations bc that is how we talk amongst each other, and were not changing our rules, because we don't do that.

 

I can just say there must be a lot of turmoil inside the NHC about this decision, between those stuck w textbook definitions and those that key words like hurricane is better used to alert people when the surge effects will be very hurricane like, since people know a hurricane =means bad and a subtropical or extra tropical storm means....I dunno is that like a weaker type of storm, its like does the NHC not a have a clue that the general public to a significant degree already thinks they cry wolf too much with storms, why not do your job and get "real" and alert people even if it means going outside (gasp) precious textbook definitions when it will alert people better, seriously.

 

I don't think you read it.

[isohume]

Those graphics could be generated on an experimental basis as early as this season...stay tuned.

 

These will be helpful for briefing purposes. Good addition. 

[N. OF PIKE]

I don't think you read it.

the thing i take issue w is they act like they need recommendation 6 , to have kept up hurricane warnings for the area. I like the paper overall and if they implement recomendation 6 then its a mute point but i will still maintain, they had the ability to keep hurr warnings and watches up since it would serve them better, they or someone ( i would LOVE to know who or how many) decided hey 2hrs before the actual center makes landfall this is "forecast" to be officially post tropical, so despite alot of effects of this hurricane being felt by then, lets bury our heads in the forecast and what textbooks say this maybe , (effectively ignore the official storm is a hurricane up to 50 miles off coast) and pummeling coast line w hurricane effects at that time and act like we were hamstrung by the lack Of having recommendation 6 passd at the time when we were not.

Do you know who or how many were directly responsible for making call to take down hurr warnings when that decision was made.

[famartin]

the thing i take issue w is they act like they need recommendation 6 , to have kept up hurricane warnings for the area. I like the paper overall and if they implement recomendation 6 then its a mute point but i will still maintain, they had the ability to keep hurr warnings and watches up since it would serve them better, they or someone ( i would LOVE to know who or how many) decided hey 2hrs before the actual center makes landfall this is "forecast" to be officially post tropical, so despite alot of effects of this hurricane being felt by then, lets bury our heads in the forecast and what textbooks say this maybe , (effectively ignore the official storm is a hurricane up to 50 miles off coast) and pummeling coast line w hurricane effects at that time and act like we were hamstrung by the lack Of having recommendation 6 passd at the time when we were not.

I still don't think you read it. Maybe just now you read reccommendation 6.

[N. OF PIKE]

I still don't think you read it. Maybe just now you read reccommendation 6.

Do u think they should have decided days before landfall to take down hurricane warnings, that this was a intelligent move if ur goal is to alert a public?

[famartin]

Do u think they should have decided days before landfall to take down hurricane warnings, that this was a intelligent move if ur goal is to alert a public?

I don't disagree with the findings.  So, read them.

[N. OF PIKE]

I don't disagree with the findings. So, read them.

i read the article, in its entirety. Its 60 something pages and it took a substantial amount of time. I dont have the time on a fri nite to re read it many more times, if it is too difficult to directly answer the question i proposed i understand. I can move on. I read the findings, can u answer the question i posed, im very interested in the met communitys feedback, i know a few mets on both sides of fence. If i missd a key finding that disputes the point i bring up about not needing recommendation 6 to keep hurr warnings up, it would be useful. Otherwise no biggie, just being skeptical of a agency having an issue w public perception of the handlin of a storm, then doing a self assessment on themselves and not being bias toward standing up for their decision.

[famartin]

Otherwise no biggie, just being skeptical of a agency having an issue w public perception of the handlin of a storm, then doing a self assessment on themselves and not being bias toward standing up for their decision.

That sounds like a personal problem.

[Turtle]

I quickly read most of the report, but one HUGE theme in it is the fact that NWS forecast positions (known as FTEs) are ESSENTIAL to do what is needed to update NHC and WFO products.  NWS needs to create new, more easily understandable products to help the public on WFO and NHC websites.  Also, NWS needs to embrace social media more readily, as well as social sciences.  

 

I felt like I was being hit over the head with the absolute need to fill NWS and NHC vacancies ASAP, but how do they do that with a HIRING FREEZE and threat of furloughs?   NWSEO has been saying this for MONTHS if not YEARS!!!  Hmmm....

 

Just my own opinion.

 

--Turtle  

[MGorse]

Do u think they should have decided days before landfall to take down hurricane warnings, that this was a intelligent move if ur goal is to alert a public?

 

Hurricane warnings were never taken down for this event here, since they were not issued (non-tropical warnings were issued instead, which many articles I have read seem to not even mention). It was coordinated ahead of time on what headlines to use so all involved were on the same page. There have been hurricanes in the past (i.e. Katrina) where hurricane warnings were issued well in advance and people still did not leave, and that resulted in fatalities. What is important is not what label is used, but the actual impacts that will be felt and being sure that the public is aware of this.

 

Based on what was stated in the service assessment, the warning type overall did not really matter. In New Jersey, the emergency managers in the coastal counties all stated that they believe it would not have mattered if a hurricane warning was issued.

[MGorse]

I quickly read most of the report, but one HUGE theme in it is the fact that NWS forecast positions (known as FTEs) are ESSENTIAL to do what is needed to update NHC and WFO products.  NWS needs to create new, more easily understandable products to help the public on WFO and NHC websites.  Also, NWS needs to embrace social media more readily, as well as social sciences.  

 

I felt like I was being hit over the head with the absolute need to fill NWS and NHC vacancies ASAP, but how do they do that with a HIRING FREEZE and threat of furloughs?   NWSEO has been saying this for MONTHS if not YEARS!!!  Hmmm....

 

Just my own opinion.

 

--Turtle  

 

What we also need to do is utilize wording in our products to get people to take the appropriate action. The wording, especially the canned statements, in our products I think are stale and they typically say the same thing over and over again. And I agree, the NWS needs to involve social science much more.

[WeatherQ]

I read the report. Since I am under Mt. Holly's coverage area, I am a bit biased towards the excellent work and products we always get from this exceptional WFO. Also being a volley FF and an ER nurse, Mt. Holly kept us up to date on the storms likely severity and impact. I will not soon forget the gravity of the PowerPoints and the MIC, Gary Sz's 'personal plea' to coastal residents. Our WFO in Mt. Holly saved lives. Nuff said.

[billgwx]

Hurricane warnings were never taken down for this event here, since they were not issued (non-tropical warnings were issued instead, which many articles I have read seem to not even mention). It was coordinated ahead of time on what headlines to use so all involved were on the same page. There have been hurricanes in the past (i.e. Katrina) where hurricane warnings were issued well in advance and people still did not leave, and that resulted in fatalities. What is important is not what label is used, but the actual impacts that will be felt and being sure that the public is aware of this.

 

Based on what was stated in the service assessment, the warning type overall did not really matter. In New Jersey, the emergency managers in the coastal counties all stated that they believe it would not have mattered if a hurricane warning was issued.

 

Emergency managers might have done much the same--but what would people have done differently had hurricane warnings been issued? I think the response would have been different for many, but still biased for some by the experience of Irene the year before.

[MGorse]

Emergency managers might have done much the same--but what would people have done differently had hurricane warnings been issued? I think the response would have been different for many, but still biased for some by the experience of Irene the year before.

 

Having hurricane warnings up does not always matter. Look at the past, for example Hurricane Katrina. It seems no matter what, there are always some people that stay.

[Amped]

Just got around to reading this. Looks like the Sandy was a nonhurricane that still managed to be the 2nd most damaging hurricane to hit the US.

 

Also I didn't realize the GFS did better with the track starting 72hrs out.  So the complaint that the GFS blew the forecast iwhile the Euro nailed it is somewhat over-hyped.

 

But, If anyone saved images from the classic run where the GFS took Sandy east of Bermuda and then changed its mind and hooked it back into Maine, feel to post it. 

[Ed Lizard]

Just got around to reading this. Looks like the Sandy was a nonhurricane that still managed to be the 2nd most damaging hurricane to hit the US.

 

Also I didn't realize the GFS did better with the track starting 72hrs out.  So the complaint that the GFS blew the forecast iwhile the Euro nailed it is somewhat over-hyped.

 

But, If anyone saved images from the classic run where the GFS took Sandy east of Bermuda and then changed its mind and hooked it back into Maine, feel to post it. 

 

As noted elsewhere, a small part of the OKX CWA in Suffolk County, NY, did probably experience sustained hurricane force winds just before the system was reclassified as post-tropical, meaning that Sandy does count as a NY state hurricane impact.

[mempho]

 

As noted elsewhere, a small part of the OKX CWA in Suffolk County, NY, did probably experience sustained hurricane force winds just before the system was reclassified as post-tropical, meaning that Sandy does count as a NY state hurricane impact.

Oops.

Sent from my DroidX using Tapatalk 2

[billgwx]

Just got around to reading this. Looks like the Sandy was a nonhurricane that still managed to be the 2nd most damaging hurricane to hit the US.

 

Also I didn't realize the GFS did better with the track starting 72hrs out.  So the complaint that the GFS blew the forecast iwhile the Euro nailed it is somewhat over-hyped.

 

But, If anyone saved images from the classic run where the GFS took Sandy east of Bermuda and then changed its mind and hooked it back into Maine, feel to post it. 

 

In the longer term, the ECMWF was clearly superior to and more consistent than the GFS. Even its forecast 10 days prior to landfall showed the anomalous high latitude blocking and suggested an East Coast threat. The GFS did catch on from time to time at that time range, but overall kept trying to undercut the block more often than not.

 

Nearer term, the ECMWF had a left-of-track bias, while the GFS went right of track. I also recall seeing this with Irene, post-tropical Noel in 2007, and post-tropical Ernesto 2006, to name a few other EC storms. In all cases except Ernesto (where I recall some ECMWF runs slinging the system as far west as Montana!), an ECMWF/GFS blend at this time range would have been close to perfect, and as a result knew 3-4 days out that an NJ landfall was the most likely scenario.

 

My recollection of GFS ensemble member performance was that the more westward-leaning members (including the operational run) were also better than the ensemble mean. Ongoing research at SUNY Stony Brook is showing that some of this sensitivity in the GFS can be traced to initialization of features over the north central Pacific! Some IMO has to do with model resolution, especially where it decreases after T+192 hours.

 

One final note--funny how the same forecasting rules of thumb we often apply to winter storm low tracks also apply to East Coast hurricanes!

[Amped]

In the longer term, the ECMWF was clearly superior to and more consistent than the GFS. Even its forecast 10 days prior to landfall showed the anomalous high latitude blocking and suggested an East Coast threat. The GFS did catch on from time to time at that time range, but overall kept trying to undercut the block more often than not.

Nearer term, the ECMWF had a left-of-track bias, while the GFS went right of track. I also recall seeing this with Irene, post-tropical Noel in 2007, and post-tropical Ernesto 2006, to name a few other EC storms. In all cases except Ernesto (where I recall some ECMWF runs slinging the system as far west as Montana!), an ECMWF/GFS blend at this time range would have been close to perfect, and as a result knew 3-4 days out that an NJ landfall was the most likely scenario.

My recollection of GFS ensemble member performance was that the more westward-leaning members (including the operational run) were also better than the ensemble mean. Ongoing research at SUNY Stony Brook is showing that some of this sensitivity in the GFS can be traced to initialization of features over the north central Pacific! Some IMO has to do with model resolution, especially where it decreases after T+192 hours.

One final note--funny how the same forecasting rules of thumb we often apply to winter storm low tracks also apply to East Coast hurricanes!

Ecmwf usually favors the left side and gfs takes the right side.

Agreed that compermise is usually the best because there's no saying which is right. They both have an alongtrack slow bias even in the short range but if they get the landfall point right most of us don't even notice it was 70 miles ahead of schedule.

[BrianLaverty]

Can't believe how bad Sandy destroyed parts of the Rhode Island coast, yet "Rhode Island" was not mentioned even once in the report by the NWS.

[Sunny and Warm]

Can't believe how bad Sandy destroyed parts of the Rhode Island coast, yet "Rhode Island" was not mentioned even once in the report by the NWS.

the report was an assessment of NWS performance(improvements), not a recap of Sandy per se.