How I would improve the NHC

[MiamiensisWx]

Disclaimer: This is not a political post or an attempt to bash the National Hurricane Center, but rather a substantive and healthy critique regarding areas where improvements can be made, in my view, to ameliorate communications with various segments of the general public. The issue first arose in the thread on Hurricane Sandy (see my excerpt below, taken from one of my postings regarding TS Warnings that did not verify in SE FL) and in my view is highly relevant in light of reports that many coastal Mid-Atlantic residents are not evacuating due to Hurricane Warnings that did not verify in 2011 as the worst effects of Hurricane Irene remained offshore, leaving weak TS winds on land and immediate coastal waters within <50 n mi of the shore.

To make my critique as comprehensive and fair as possible, I have fully explained my proposed solutions, though beware that the post is long. Enjoy!

http://www.storm2k.org/phpbb2/viewtopic.php?p=2283771#p2283771

These are the types of situations that make me concerned about future complacency due to unnecessary school closings, lost work hours, reduced productivity due to storm mode, et cetera. People may become more finicky in their decisions to prepare and may make incorrect judgments when a more serious event arises due to uncertainties surrounding the track. The same will occur elsewhere if the main impact of Sandy is in Canada and N Maine (as seems increasingly probable) rather than the greater Northeast / Mid-Atlantic--there has been so much coverage on the threat that many people cannot comprehend the possibility that it may not verify as models initially suggested. I know I was a bit guilty as well in terms of elevating the threat, though not as much as some people were.

The legacy warnings + media = perceptions that lead to expectations that may not verify but cause headaches in the process. When the next major disaster strikes the United States, I have a feeling that it will still cause many deaths and contain the element of surprise (even though it may well be warned and communicated quite well), simply due to human nature after all the storm fatigue over non- (or mostly non-) events.

Solutions:

1) Develop separate tropical-storm and hurricane warnings for inland areas (>4 km from the beach), the immediate coastland (<4 km from the beach and including the beach), and nearshore waters <35 km offshore. We already have a working system for areas >35 km offshore and beyond maritime boundaries. The new system would also take into account variations in roughness of terrain such as over more-exposed inland lakes/water bodies over which friction is lower and does not reduce winds as much. Inland wind-decay models have shown enough accuracy (as in remodelings of historical tropical cyclones that have matched observed wind impacts as in Andrew, Katrina, and Hugo, well in accord with equations for inland winds--the inland wind-decay model of KaPlan and DeMaria (2003) based upon the MEOW, or Mean Envelope of Maximum Winds over land--that have generally, though not perfectly, matched the observations, as in the H*Wind grids) and models have improved their resolution so as to make forecasting more precise. There may be errors, of course, but the precision can help people in different industries on land or at sea know better what to expect. For example, specialists at the NHC can reproduce spatial-sequential wind-modelling maps customized based upon the expected size, wind radii, radius of maximum wind, and closest approach according to a timed sequence. The maps could include Google Maps terrain and infrastructure and allow users to type in their address, see their location overlain with the expected winds (pixels) at their location, see the time of strongest winds, and see expected winds all around their area. A separate graphical option would allow viewers to simulate the approach of the cyclone, showing the circulation, isobars, and wind vectors, including different colored wind barbs and lines enclosing the 1-minute, 10-meter 34- (yellow), 50- (orange) and 63-kt (red) radii as in the NHC forecast advisories. In a major hurricane, a 111-kt (dark red) line would enclose the radius of major hurricane winds (dark red barbs). A disclaimer will tell users that the data should not be the determining factor in life-altering decisions but merely reflect the forecast output and are experimental.

b) Pertaining to the above, the system should be accompanied by a graphical system showing simulations of wind impacts based upon the expected sustained 10-m winds, both inland (on land and water bodies), on the immediate coast, just offshore, and well offshore--each of the four zones listed in 1). The system would include a region-specific animation showing local topography, vegetation, and human reference points (i.e., types of buildings and warehouses) and incorporate research by Ted Fujita, Robert Simpson, E. R. Boose, and other wind engineers to simulate the effects of wind stress over time on structures and vegetation, taking into account duration, exposure (assumed to be unobstructed in the animation), structural integrity, general hardiness of regional vegetation (i.e., oaks and sawgrass in the Everglades vs. conifers in Massachusetts), topographical elevations (i.e., <25% above sea level, <50% above sea level, <75% above sea level), and influences of nearby bodies of water like Lake Okeechobee, Narragansett Bay, or Lake Pontchartrain. The animation would show effects on bodies of water and land specific to the strength and duration of winds as modelled in the simulations in 1) and roughly specific to the track of the storm and expected wind headings.

2) Develop an experimental, public SLOSH (Sea, Lake, and Overland Surges from Hurricanes) simulation to model expected storm surges on the NHC Web site. Under this system, users would go to the NHC Web site, look under storm headings, and see a description pointing out the SLOSH modelling for expected local storm-surge levels. Viewers would be able to view spatial-sequential and still images showing expected storm-surge on a location- and address-specific basis, overlain on current elevations and on a Google Maps grid similar to the one in 1). Users would gain an idea as to how expansive the surge will be, what it has the potential to be, when it would occur, and how far inland it may reach. A separate simulation showing the actual surge as modeled in the water-elevation outputs would show the modeled behavior of the surge in the area. The simulation would also include an option to include in the overlay the cyclone itself, modeled exactly the same as in 1) with colored vectors and circular colored lines. In both simulations, viewers would be able to see a Google Maps grid of the area and can click on / type in their location to view water elevations at their location BOTH at a specific point in time and as a progressive simulation as the storm nears. A disclaimer as in 1) would specify that the results are experimental and indicate the POTENTIAL surge values and may not be exact; therefore, residents should not use the graph as a sole tool with which to make life-altering decisions. Under this system, the Saffir-Simpson Scale would not be changed into an IKE-type system based upon Integrated Kinetic Energy (measuring a storm’s destructive potential based approximately upon size, approach, and intensity) or be split into separate scales for wind and surge, nor would it be altered into a scale measuring various destructive impacts on a 1-to-10 scale.

b) However, under the new system, an alternative is to make a dual Saffir-Simpson Scale, which would retain all the features of the current wind-based scale but would also use the IKE Scale to measure the relative, cumulative effects of the cyclone: wind stress, wave damage, storm surge, and wind loads. In this area, covered by the IKE system, a graphical system comparing the IKE of a current tropical cyclone to the IKE of past systems (i.e., Katrina would be compared to Camille in terms of IKE) in the area would give coastal residents an holistic idea of what to expect that would not supplant but rather complement the wind-based, existing Saffir-Simpson Scale. For other impacts such as inland and stream flooding and rainfall, a third graphical system, based partially or wholly upon local NWS and regional NWS / HPC forecasts, would incorporate current saturation and water levels plus expected QPF output to measure the relative hazard of the TC. Part of the graph would show expected, region-specific, inland flash-flood levels on a hazard scale of very likely (lower bound >75%), likely (lower bound >50%), or possible (lower bound >25%) under three categories exceptional flooding (meaning all local streams are expected to be flooded to mean levels >75% of flood stage with extreme likelihood that all areas below a certain elevation [say 15 feet as an arbitrary standard, based upon what is the typical regional river height] will be flooded), severe flooding (meaning most local streams are expected to be flooded to mean levels >50% of flood stage with high likelihood that most areas below a certain elevation will be flooded), and moderate flooding (meaning some local streams are expected to be flooded to mean levels >25% of flood stage with moderate likelihood that most areas below a certain elevation will be flooded).

3) Adopt a more ironclad criterion for watches and warnings, and adhere more strictly to them. Recently, tropical storm warnings were issued for South Florida--meaning Jupiter to South Miami-Dade County--even though observations indicate that none of the stations both on and offshore reported sustained winds of >34 kt at the standard 10-m elevation. However, the excerpt from the following NWS Miami Area Forecast Discussion indicates that the warnings were maintained for a part of the coast and offshore waters within <50 n mi of the shore...even though TS winds were only possible (underlined). A warning is issued when said TS winds are expected, not possible. Based upon this criterion, the warnings should have been made watches for S FL, since models for about a day and a half suggested TS winds were possible--as the NWS Miami AFDs have said consistently--on the immediate coast and just offshore, but did not evince enough evidence that they were expected. Warnings and watches should, in my view, follow the importance of clarity to avoid confusion and future complacency.

The excerpt:

709 FXUS62 KMFL 261845 AFDMFL AREA FORECAST DISCUSSION NATIONAL WEATHER SERVICE MIAMI FL 245 PM EDT FRI OCT 26 2012 .DISCUSSION...HURRICANE SANDY CONTINUES TO SLOWLY DRIFT NORTH OVER THE FAR NORTHERN BAHAMAS. ITS SLOW MOVEMENT WILL CONTINUE TO ALLOW FOR THE OUTER BANDS TO MOVE ACROSS THE EASTERN PORTIONS OF SOUTH FLORIDA THROUGH THIS EVENING. HOWEVER, THE WIND FIELD HAS NOT EXPANDED TO THE EXTENT THAT SOME OF THE MODELS WERE EARLIER INDICATING SO TROPICAL STORM WARNINGS WERE DISCONTINUED FOR COASTAL BROWARD AND MIAMI-DADE COUNTIES AND THE ADJACENT ATLANTIC WATERS INCLUDING BISCAYNE BAY. TROPICAL STORM CONDITIONS WILL REMAIN POSSIBLE IN THE HEAVIER BANDS ACROSS COASTAL REGIONS OF PALM BEACH COUNTY. FOR BROWARD AND MIAMI-DADE COUNTIES, THE WARNING WAS REPLACED WITH A WIND ADVISORY ALONG WITH HIGH SURF ADVISORY AND THE MENTION OF HIGH RISK OF RIP CURRENTS.

http://www.srh.noaa.gov/productview.php?pil=AFDMFL

Questions, comments, and critique are encouraged.

[CourierPR]

MiamiensisWx, you bring much to this blog and I, for one, appreciate your thoughtful input. I, too, fear that when the next big storm hits South Florida, complacency will be rampant if changes are not made with regard to warnings.

[MiamiensisWx]

MiamiensisWx, you bring much to this blog and I, for one, appreciate your thoughtful input. I, too, fear that when the next big storm hits South Florida, complacency will be rampant if changes are not made with regard to warnings.

I was at the beach in Boca Raton today...just after high tide, which occurred at 8:14 a.m., and low tide, which occurred around 3:40 p.m. Even though Sandy was then more than 1,380 n mi / 1,587 statute mi from South FL, I observed wave heights of up to 12-15 feet...the waves broke the stairwells of two lifeguard towers and tore away vegetation by the roots as they came all the way up to the lowest stairs of the boardwalks. At least two signs listing beach conditions were swept away without leaving a trace, while small coconut palms and other shrubs that were growing in the vine zone (lowest edge of the dune facing seaward) were carried south, along with massive amounts of trash and debris (including pieces of a small yellow windjammer!), due to the strong longshore current and immense swells from the northeast. The beach erosion was probably the worst I observed since Hurricane Wilma (2005)...I have photos and video to document the erosion and the size of the waves. The entire beach, even the sections replenished about a year and a half ago, was a total loss due to erosion...in several areas, sand five feet deep that had been placed to elevate the boardwalks was gone. Huge 3-5-foot escarpments were left in its place. Remember: all these effects were caused by a very large storm thousands of miles away.

A direct hit by a similar storm--even a large Category-2 hurricane like Frances, hitting farther south this time (in Miami-Fort Lauderdale, not Stuart)--would cause much higher waves due the direct strike and very long fetch preceding its arrival. A similar case was the 1926 Miami hurricane, a very large hurricane which spent nearly two full days over the Bahamas with winds of 130+ knots / 150+ mph...by the time it struck 10 n mi / 12 statute mi south of Downtown Miami, it had built up immense wave energy due to sustained hurricane-force winds extending 110 mi from the center (confirmed by observations and estimates from the upper Florida Keys to north of Stuart). The storm surge on Miami Beach was relatively small--owing to the deep continental slope offshore--but it still flooded the entire island due to its low elevation and the effects of immense waves. Of course, Downtown Miami suffered more surge (10.9+ feet) plus waves as it fronted shallower Biscayne Bay. The surge in Fort Lauderdale was 6-7 feet on the present Las Olas Riverfront...flooding buildings even on the highest riverfront elevations in downtown...and 12.5+ feet near Port Everglades and Bahia Mar. The beaches from the Miami area to Cocoa Beach were severely eroded.

[Frank2]

What does need immediate improvement is NOAA Weather Radio - it was almost useless during Sandy's Florida approach, filled with endless local products that go on and on and on and on and on - instead of brief and concise products that listeners can use to their benefit...

I've contacted the Miami NWSFO about this issue, and they agree that the products do run very long, but my guess is that the changes need to come from NOAA Headquarters, and again my guess is that they won't change anything - sadly at this point NWR is almost useless because listeners often have to wait an entire tape cycle - often about 10-15 minutes - before hearing a public advisory, and most in today's world are not going to wait that long for something they can find just as easily on their cell phone in seconds...

Frank

[CrazyC83]

Would changing the warning system work? One possibility:

* Create a three-tier Storm Surge warning system:

** Advisory: Storm surge expected but not likely to cause significant flooding (typically 1 to 3 feet, but variable on location - raising bodies of water to Action Stage or Minor Flood Stage)

** Watch/Warning (depending on timing): Storm surge expected to cause significant flooding to Moderate or Major Flood Stage (typically at least 4 feet, but variable on location)

** Emergency: Catastrophic storm surge expected to reach record levels AND above Major Flood Stage

* Eliminate the Tropical Storm and Hurricane warnings, and replace them with marine-level warnings for more clarity:

** Wind Advisory - as now, winds sustained 20 to 33 kt OR gusts at least 34 kt

** Gale Warning - winds sustained 34 to 49 kt (tropical storm force) OR gusts at least 50 kt (severe criteria)

** Storm Warning - winds sustained 50 to 63 kt (strong TS force)

** Hurricane Force Wind Warning - winds sustained 64 to 95 kt (Cat 1-2 force)

** Major Hurricane Force Wind EMERGENCY - winds sustained 96 kt+ (Cat 3-5 force)

These warnings can be issued for both tropical and non-tropical cyclones, and would also replace the High Wind Warning. They would only be issued for areas that are expected to receive those winds, and can take into effect local enhancements such as inland lakes and channels. Few non-beach areas would ever receive an MHE (even in storms like Katrina).

[tolakram]

A single number is, for my money, the only thing that will work. We know how people respond to hurricane designations so make sure 3 or greater is a real dangerous storm and save 1 and 2 for tropical and low end hurricanes. Make it so when people see a 3 they leave.

This is the same discussion as the Saffir Simpson replacement discussion, in my opinion. You can't fix human nature, you have to work with human nature.

[CrazyC83]

A single number is, for my money, the only thing that will work. We know how people respond to hurricane designations so make sure 3 or greater is a real dangerous storm and save 1 and 2 for tropical and low end hurricanes. Make it so when people see a 3 they leave.

This is the same discussion as the Saffir Simpson replacement discussion, in my opinion. You can't fix human nature, you have to work with human nature.

In that case, 1 would be mostly tropical storms except for the smallest hurricanes.

[KBBOCA]

Here are some related comments I wrote on the Sandy discussion thread (in the context of the potential confusion caused by NHC not having issued hurricane warnings for NJ & NY):

This discussion (about how hurricane watches & warnings were handled) is no longer just limited to weather sites or academics. It's become an issue of interest to the general public too, especially as it affects the question of insurance and hurricane deductibles.

Here's an interesting article at NJ.com

http://www.nj.com/news/index.ssf/2012/1 ... cart_river

One thing that was so striking about the NWS alert maps, was that there were hurricane wind marine warnings just offshore (You can see a saved example of the map from 10/27 here). It was very weird to see that and not have their be corresponding watches / warnings for the coastline etc. since the storm was expected to make direct landfall.

Given that the SSHS categories are now based solely on winds, it *would* seem logical that if a coastal area is expecting hurricane force winds (from a storm that was at some point a tropical system covered by the NHC), there ought to be a hurricane warning, and somehow to avoid confusion, it has to be clearer what PARTS of the area are likely to experience the winds. (e.g. offshore waters only, the immediate coastline or further inland...)

It's a slightly different topic, but I want to give two SHOUT OUTS and give praise for two things that greatly helped me as an amateur have clear information and enabled me to warn / advise friends in the affected areas.

1) the NWS briefings from the Mount Holly office were incredibly clear, concise and helpful and put together in non-technical terms. ABSOLUTELY they saved lives. I hope all NWS offices will consider putting together such a product and doing similar public briefings when they are faced with a significant event.

2) The H*Wind analysis and the IKE scale. It was the IKE scale for storm surge that I found most helpful. All of us here knew this was likely to be a catastrophic event for the NY area when that IKE level for surge reached 5.8 out of 6. Winds were not the primary danger, storm surge was, and that H*Wind analysis communicated that extremely clearly.

So WHY was that information not disseminated in NHC or NWS advisories? I only knew of it because of folks here, and because of Dr. Jeff Masters' blog at Wunderground. I would think being told that storm surge risk was at 5.8 out of 6 would have caused more folks to evacuate Atlantic City, Staten Island, Hoboken, etc.

3) Likewise the CERA storm surge maps. AWESOME. I wish they'd been publicized more in the general media.

[CrazyC83]

If you read the old Service Assessments, they give lots of great points on how to improve the NWS, forecasts and action plans, both immediate and long term. They also give some best practices.

Any thoughts that any of you have on recommendations and best practices? Maybe we should work to submit them to the NWS. Surely there will be a lot of recommendations.

[CrazyC83]

Here are some related comments I wrote on the Sandy discussion thread (in the context of the potential confusion caused by NHC not having issued hurricane warnings for NJ & NY):

This discussion (about how hurricane watches & warnings were handled) is no longer just limited to weather sites or academics. It's become an issue of interest to the general public too, especially as it affects the question of insurance and hurricane deductibles.

Here's an interesting article at NJ.com

http://www.nj.com/news/index.ssf/2012/1 ... cart_river

One thing that was so striking about the NWS alert maps, was that there were hurricane wind marine warnings just offshore (You can see a saved example of the map from 10/27 here). It was very weird to see that and not have their be corresponding watches / warnings for the coastline etc. since the storm was expected to make direct landfall.

Given that the SSHS categories are now based solely on winds, it *would* seem logical that if a coastal area is expecting hurricane force winds (from a storm that was at some point a tropical system covered by the NHC), there ought to be a hurricane warning, and somehow to avoid confusion, it has to be clearer what PARTS of the area are likely to experience the winds. (e.g. offshore waters only, the immediate coastline or further inland...)

It's a slightly different topic, but I want to give two SHOUT OUTS and give praise for two things that greatly helped me as an amateur have clear information and enabled me to warn / advise friends in the affected areas.

1) the NWS briefings from the Mount Holly office were incredibly clear, concise and helpful and put together in non-technical terms. ABSOLUTELY they saved lives. I hope all NWS offices will consider putting together such a product and doing similar public briefings when they are faced with a significant event.

2) The H*Wind analysis and the IKE scale. It was the IKE scale for storm surge that I found most helpful. All of us here knew this was likely to be a catastrophic event for the NY area when that IKE level for surge reached 5.8 out of 6. Winds were not the primary danger, storm surge was, and that H*Wind analysis communicated that extremely clearly.

So WHY was that information not disseminated in NHC or NWS advisories? I only knew of it because of folks here, and because of Dr. Jeff Masters' blog at Wunderground. I would think being told that storm surge risk was at 5.8 out of 6 would have caused more folks to evacuate Atlantic City, Staten Island, Hoboken, etc.

3) Likewise the CERA storm surge maps. AWESOME. I wish they'd been publicized more in the general media.

1 - BEST PRACTICE by NWS Mount Holly/Philadelphia that needs to be done by all offices in major events. That is especially true in events where breakdowns have happened (like the 2010 Nashville flood)

2 - That was an excellent forecast by the weather people, but again the message didn't get down. Those kind of values need more mention, especially in NHC discussions and by the media.

3 - Again, it seemed S2K and other forums and blogs were most reliable there. We need to reach out to get the bigger message.

[KBBOCA]

re: NWS Best Practices:

I think one thing that hasn't quite fully filtered down into the practices of the NWS and their advisories is the fact that with the internet, their readership has now changed significantly.

The general public may now be turning directly to their site for news and at least one of their advisory products need to reflect that. Rather than making the public read a long text-based forecast statement that is very long with lots of technical terms and probabilities, etc., they need an "end-user" product, which is exactly what the Mount Holly / Philly office did.

That end-user product with several clear simple bullet points was extremely user-friendly for

1) media
2) local officials
3) the general public

and it was very easy to find on their website, and via Twitter.

Gone are the days when it is only media weather forecasters who are relying on the NWS products and interpreting them for local officials and the public.

The Upton NY office did not have nearly the kind of clear info that Mount Holly had. I found it very frustrating to try and find a clear analysis or presentation of the risks on their site. There were so many different watches and warnings active at one time that it made navigating their forecast statement much too difficult for an average user.

[CrazyC83]

Even the more technical discussions get a lot more readers than ever before. They need to be clear, to the point and straight out, like Mount Holly did.

[vbhoutex]

Here are some related comments I wrote on the Sandy discussion thread (in the context of the potential confusion caused by NHC not having issued hurricane warnings for NJ & NY):

This discussion (about how hurricane watches & warnings were handled) is no longer just limited to weather sites or academics. It's become an issue of interest to the general public too, especially as it affects the question of insurance and hurricane deductibles.

Here's an interesting article at NJ.com

http://www.nj.com/news/index.ssf/2012/1 ... cart_river

One thing that was so striking about the NWS alert maps, was that there were hurricane wind marine warnings just offshore (You can see a saved example of the map from 10/27 here). It was very weird to see that and not have their be corresponding watches / warnings for the coastline etc. since the storm was expected to make direct landfall.

Given that the SSHS categories are now based solely on winds, it *would* seem logical that if a coastal area is expecting hurricane force winds (from a storm that was at some point a tropical system covered by the NHC), there ought to be a hurricane warning, and somehow to avoid confusion, it has to be clearer what PARTS of the area are likely to experience the winds. (e.g. offshore waters only, the immediate coastline or further inland...)

It's a slightly different topic, but I want to give two SHOUT OUTS and give praise for two things that greatly helped me as an amateur have clear information and enabled me to warn / advise friends in the affected areas.

1) the NWS briefings from the Mount Holly office were incredibly clear, concise and helpful and put together in non-technical terms. ABSOLUTELY they saved lives. I hope all NWS offices will consider putting together such a product and doing similar public briefings when they are faced with a significant event.

2) The H*Wind analysis and the IKE scale. It was the IKE scale for storm surge that I found most helpful. All of us here knew this was likely to be a catastrophic event for the NY area when that IKE level for surge reached 5.8 out of 6. Winds were not the primary danger, storm surge was, and that H*Wind analysis communicated that extremely clearly.

So WHY was that information not disseminated in NHC or NWS advisories? I only knew of it because of folks here, and because of Dr. Jeff Masters' blog at Wunderground. I would think being told that storm surge risk was at 5.8 out of 6 would have caused more folks to evacuate Atlantic City, Staten Island, Hoboken, etc.

3) Likewise the CERA storm surge maps. AWESOME. I wish they'd been publicized more in the general media.

1 - BEST PRACTICE by NWS Mount Holly/Philadelphia that needs to be done by all offices in major events. That is especially true in events where breakdowns have happened (like the 2010 Nashville flood)

2 - That was an excellent forecast by the weather people, but again the message didn't get down. Those kind of values need more mention, especially in NHC discussions and by the media.

3 - Again, it seemed S2K and other forums and blogs were most reliable there. We need to reach out to get the bigger message.

This is part of why we are here and what we try to do both here and on facebook. We had a ton of registrations coming out of NJ and NY and the NE in general with Sandy's approach and we processed those asap. Hopefully some of those will send their friends, etc. If you have any suggestions on how we can improve our outreach to dispense info we will look into it.

[CrazyC83]

Here are some related comments I wrote on the Sandy discussion thread (in the context of the potential confusion caused by NHC not having issued hurricane warnings for NJ & NY):

This discussion (about how hurricane watches & warnings were handled) is no longer just limited to weather sites or academics. It's become an issue of interest to the general public too, especially as it affects the question of insurance and hurricane deductibles.

Here's an interesting article at NJ.com

http://www.nj.com/news/index.ssf/2012/1 ... cart_river

One thing that was so striking about the NWS alert maps, was that there were hurricane wind marine warnings just offshore (You can see a saved example of the map from 10/27 here). It was very weird to see that and not have their be corresponding watches / warnings for the coastline etc. since the storm was expected to make direct landfall.

Given that the SSHS categories are now based solely on winds, it *would* seem logical that if a coastal area is expecting hurricane force winds (from a storm that was at some point a tropical system covered by the NHC), there ought to be a hurricane warning, and somehow to avoid confusion, it has to be clearer what PARTS of the area are likely to experience the winds. (e.g. offshore waters only, the immediate coastline or further inland...)

It's a slightly different topic, but I want to give two SHOUT OUTS and give praise for two things that greatly helped me as an amateur have clear information and enabled me to warn / advise friends in the affected areas.

1) the NWS briefings from the Mount Holly office were incredibly clear, concise and helpful and put together in non-technical terms. ABSOLUTELY they saved lives. I hope all NWS offices will consider putting together such a product and doing similar public briefings when they are faced with a significant event.

2) The H*Wind analysis and the IKE scale. It was the IKE scale for storm surge that I found most helpful. All of us here knew this was likely to be a catastrophic event for the NY area when that IKE level for surge reached 5.8 out of 6. Winds were not the primary danger, storm surge was, and that H*Wind analysis communicated that extremely clearly.

So WHY was that information not disseminated in NHC or NWS advisories? I only knew of it because of folks here, and because of Dr. Jeff Masters' blog at Wunderground. I would think being told that storm surge risk was at 5.8 out of 6 would have caused more folks to evacuate Atlantic City, Staten Island, Hoboken, etc.

3) Likewise the CERA storm surge maps. AWESOME. I wish they'd been publicized more in the general media.

1 - BEST PRACTICE by NWS Mount Holly/Philadelphia that needs to be done by all offices in major events. That is especially true in events where breakdowns have happened (like the 2010 Nashville flood)

2 - That was an excellent forecast by the weather people, but again the message didn't get down. Those kind of values need more mention, especially in NHC discussions and by the media.

3 - Again, it seemed S2K and other forums and blogs were most reliable there. We need to reach out to get the bigger message.

This is part of why we are here and what we try to do both here and on facebook. We had a ton of registrations coming out of NJ and NY and the NE in general with Sandy's approach and we processed those asap. Hopefully some of those will send their friends, etc. If you have any suggestions on how we can improve our outreach to dispense info we will look into it.

It's not just us but also other regional-based blogs and other sites like AmericanWx (which has more of a NE focus). I sent an e-mail to NOAA and mentioned several sites. The NHC seemed to believe all of us from the beginning even when the GFS was still out of whack.

[MiamiensisWx]

These are all very good suggestions...but what I continue to fixate on is the need for specificity, provided that it is communicated clearly. I have noticed--and maybe other people have as well--that during disaster after disaster, from Katrina to Rita to Ike to Sandy, human nature has illustrated that people are loath to respond to warnings in return for uncertainties. That is, they will not risk their family members, their jobs, their education, etc. in return for new hassles that may in the end prove, rightly or wrongly, to be unnecessary. (Of course, unnecessary is a subjective factor but is one that people form in their minds according to their specific circumstances in life.)

People need visuals to gain perspective and a better sense of the risk involved should they decide on a particular default option...for example, staying rather than evacuating a certain area due to perceived uncertainties over how the surge or other impacts will hit THEM. They need location-specific, their-centered graphs, visuals, et cetera, rather than abstract blocks of text talking about X ft. of surge between Point Y and Point Z (a distance maybe 25, 50, or more miles long). Otherwise, they will not feel a concrete need to act. I can still remember some S2K members who did not evacuate for Katrina because the NHC surge descriptions mentioned surge up to X ft. between points X and Y--over a broad area--but otherwise told little relatable information that would help the residents make shrewd, timely, life-saving decisions. And of course, some of the same people who survived stronger Category-5 Camille (1969) just could not wrap their minds around the idea that Katrina might surpass Camille at their location. In some regards, this was NOT their fault--the facts were just not relayed to them in a relatable, much less clear, way!

Another pitfall is that, in a fast-paced, technological society, people synthesize information more readily, but unfortunately still mold the data around their own circumstances and prejudices. And technology can often prove to be a crutch or a false sense of security--during the 04/27/2011 and Joplin, MO, tornadoes, people watched their GRLevel radar screens to look for hook echoes and other signs that a tornado was going to hit their house...even though the NWS tornado warning covered the whole supercell / thunderstorm rather than just a single point in the rotation / mesocyclone, as the tornado could reform at any time or place. As a result, the people thought that the tornado would pass a block or more away and did not take action until the tornado was upon them. People are perhaps less likely to feel the fight-or-flight response in the face of a looming disaster if technology gives them an unfounded, or overestimated, sense of certainty. Similarly, if people are not provided with clear disclaimers and sound knowledge, technology will be a danger rather than a source of assistance, for false ideas or ignorance--NOT the same as stupidity!--will remain.

So we need to quickly and accurately present information in such a way that the well-meaning but ignorant public will not make the wrong interpretations or judgments. They need to be able to relate their situation to the message. And they need to keep a sense of perspective and treat the visuals and technology as aids, not final arbiters of judgment. Otherwise, as society grows larger and more complex, with various competing interests, we will continue to see larger death tolls in significant natural disasters like the recent tornado outbreaks, the 2010 Nashville flood, Sandy, and whatever looms on the horizon. And in the process, the NHC and the NWS may be, fairly or unfairly, criticized for not relating the message to the public. Fair or not, that seems to be the heart of the larger matter at stake.

This is my real worry, for Sandy, while horrendous, was probably NOT the upper limit of what is possible for Nature in the Northeast, as the 1938 hurricane showed. The next time another region of the nation gets its Sandy or worse--particularly areas that have not seen them in a long time, like Savannah, GA, Cedar Key, FL, the FL Keys, and the Miami metro area--unless the system of communication is changed, I would, sadly, not be surprised to see many well-meaning, ignorant (NOT stupid!) people make unfortunate, wrong decisions.

[MiamiensisWx]

In regard to future Sandy-type scenarios at the locations I mentioned in the preceding post, here are some potential historical reincarnations.

1) Savannah, GA--Historically hit by major hurricanes in 1804, 1824, 1854, 1893, 1896, and 1898, Savannah is highly vulnerable to storm surge. The 1824 hurricane, whose eye passed over Darien, GA, completely inundated St. Simon Island (10 ft. above mean sea level) and nearby Sapelo Island, indicating a storm surge of over 10 ft. at landfall. Savannah experienced Category-1 winds and much flooding occurred that damaged the cotton crops--the worst such flooding on the Savannah River since 1796. The 1804 hurricane, which made landfall on St. Simon Island, was even more severe...with a storm surge of over 20 ft. on the Savannah River, St. Catherine’s Island, St. Simon Island, and the bays between Port Royal and St. Helena Sounds, near Beaufort, SC. The storm was extremely large, as hurricane winds occurred at Charleston, SC, about 150 mi NE of the landfall point. The August 1893 hurricane, which produced a storm surge up to 18 ft. along the GA / SC border (with 15 ft. measured at Isle of Hope, GA), was extremely large, with an outer closed 1010-mb radius of 180 n mi, larger than that for the 1010-mb contour of Hurricane Hugo in 1989. A double eyewall was present, maximizing the surge and the wind field. The 1896 hurricane was a rapidly moving, near-Category-4 hurricane at landfall near Cedar Key, FL...110 kt / 960 mb with a very small, tight eye...and was still a major hurricane in NE FL and SE GA. It caused significant wind damage at Darien and other locations along its path. Finally, the 1898 hurricane, the last major to strike GA, was an intensifying Category-4 at landfall...115 kt / 938 mb...produced a 16-ft. storm surge at Brunswick, GA, despite having a smaller-than-average radius of maximum wind, 20 n mi. A large hurricane like Sandy, even a Category 1/2, would cause significant surge along the GA coast.

[mitchell]

What does need immediate improvement is NOAA Weather Radio - it was almost useless during Sandy's Florida approach, filled with endless local products that go on and on and on and on and on - instead of brief and concise products that listeners can use to their benefit...

Frank

AMEN Brother! Sometimes I think that computer voice must get paid by the minute! What is the longest possible way to say that four different counties are all under the same EXACT set of warnings? Repeat the same exact warning statement four times with all of the same information with just the county name changed!

We've got a hurricane bearing down on us and 2-3 minutes per cycle gets devoted to the previous days's climatic summary because in the middle of a hurricane I clearly need to know about yesterday's heating degree days at the various local airports, but the local hourly buoy obs get deleted from the cycle.

[CrazyC83]

In regard to future Sandy-type scenarios at the locations I mentioned in the preceding post, here are some potential historical reincarnations.

1) Savannah, GA--Historically hit by major hurricanes in 1804, 1824, 1854, 1893, 1896, and 1898, Savannah is highly vulnerable to storm surge. The 1824 hurricane, whose eye passed over Darien, GA, completely inundated St. Simon Island (10 ft. above mean sea level) and nearby Sapelo Island, indicating a storm surge of over 10 ft. at landfall. Savannah experienced Category-1 winds and much flooding occurred that damaged the cotton crops--the worst such flooding on the Savannah River since 1796. The 1804 hurricane, which made landfall on St. Simon Island, was even more severe...with a storm surge of over 20 ft. on the Savannah River, St. Catherine’s Island, St. Simon Island, and the bays between Port Royal and St. Helena Sounds, near Beaufort, SC. The storm was extremely large, as hurricane winds occurred at Charleston, SC, about 150 mi NE of the landfall point. The August 1893 hurricane, which produced a storm surge up to 18 ft. along the GA / SC border (with 15 ft. measured at Isle of Hope, GA), was extremely large, with an outer closed 1010-mb radius of 180 n mi, larger than that for the 1010-mb contour of Hurricane Hugo in 1989. A double eyewall was present, maximizing the surge and the wind field. The 1896 hurricane was a rapidly moving, near-Category-4 hurricane at landfall near Cedar Key, FL...110 kt / 960 mb with a very small, tight eye...and was still a major hurricane in NE FL and SE GA. It caused significant wind damage at Darien and other locations along its path. Finally, the 1898 hurricane, the last major to strike GA, was an intensifying Category-4 at landfall...115 kt / 938 mb...produced a 16-ft. storm surge at Brunswick, GA, despite having a smaller-than-average radius of maximum wind, 20 n mi. A large hurricane like Sandy, even a Category 1/2, would cause significant surge along the GA coast.

From that analysis, the 1893 hurricane was likely only a Cat 2 (~85 kt winds), yet still produced that intense surge. That is another example that it doesn't take a really high-intensity storm to cause a major disaster. The pressure was probably about 948mb at landfall (despite the wind estimate I have of 85 kt) which was likely just south of Savannah based on the 957 recorded there, but the RMW was ENORMOUS in that case - maybe even larger than the site's estimate! As for that filling rate, it might have not been as fast as suggested since the RMW caught Charleston well before landfall.

The 1896 and 1898 storms were definitely more intense - Cat 3 or 4 per HURDAT seem reasonable - but smaller.

IMO for Savannah, the 1893 scenario is the worst case for surge even though the storm was much weaker.

[MiamiensisWx]

2) Tampa Bay / Pinellas County, FL--Historically, although few major hurricane have directly struck Tampa Bay--one in September 1848 and another, the last, in October 1921--the area is much more vulnerable to hurricane impacts than many people suppose. Since 1851, 12 hurricanes have passed within 40 n mi / 45 s mi of the Tampa Bay region, representing a frequency of one hurricane every 14 years. Of these 12 cases, four were major hurricanes--one Category 4 (the 1935 Labor Day hurricane), three Category 3--as they passed closest to the region, in 1921, 1935, 1949, and 1950 (Easy) each. Each of these storms produced sustained hurricane-force winds on the beaches at Saint Petersburg and winds of at least strong tropical-storm intensity at Tampa International Airport / MacDill Air Force Base. The 1921 hurricane, with a radius of maximum wind near 20 n mi and sustained winds of 100 kt at landfall, struck Tarpon Springs at low tide but still completely flooded Bayshore Boulevard, Davis and Harbor Islands, Palmetto Beach, and DeSoto Park, along with areas near the mouth of the Hillsborough River and near the shoreline sections of Downtown Tampa (storm surge: 10.5 ft). A similarly sized storm at high tide would cause a surge of 15 to 18 ft. in the same area(s). The tiny 1935 Labor Day hurricane passed about 30 n mi / 35 s mi offshore while moving north-northwest, yet caused a storm surge of 5 to 6 ft., locally even higher, along beaches in Pinellas County and wave heights well over 16 ft. Severe beach erosion and hurricane-force winds affected the coastline there. A very large hurricane in 1944, reanalyzed to have been 90 kt at landfall, hit near Nokomis but caused sustained winds of 45 to 50 kt in the Tampa Bay region. Hurricane Easy in 1950 paralleled the coast and struck Cedar Key, but it produced sustained hurricane-force winds in Pinellas County and a storm surge up to 8 ft. along portions of Tampa Bay--the highest surge since the 1921 hurricane. So the idea that Tampa Bay is protected may be comforting on a relative scale, but the historical record says otherwise...that the region is far, far, far from immune.

The worst documented hurricane to affect the area, however, likely occurred in late September of 1848. That hurricane likely came at or near high tide, but it was also larger and much more intense than the 1921 hurricane...as Fort Brooke, near Downtown Tampa, measured a pressure of 28.18 inHg / 954.3 mb concomitant with hurricane-force winds shifting to the southwest. As a ship near Cedar Key was dismasted by hurricane-force winds around the same time, the hurricane likely made landfall a bit north of Tarpon Springs, perhaps near New Port Richey, with a radius of maximum wind near 25 n mi / 30 n mi or even larger. The central pressure was likely close to 940 mb, as Tampa may have been just outside the maximum winds, for the hurricane winds just shifted to southwest without diminishing...even strengthening, according to the observer, after they shifted. The measured storm surge at Tampa was at least 15 ft., and the entire Interbay Peninsula was submerged. Egmont Key and all the offshore islands near Pinellas County were submerged entirely...the center of Pinellas County, probably from Clearwater to near downtown Saint Petersburg, was also flooded. The geography of the area and Southwest FL was so drastically changed that navigational charts were made out-of-date by the storm...ships could recognize few navigable inlets or recognizable islands in the Tampa Bay area. The real surge in this storm was likely close to 20 ft. on the Hillsborough River, for observers reported that only the tops of large pine trees stood above the water. Winds were definitely stronger than in 1921, for many pine trees actually snapped just above ground level, and even large oaks snapped or toppled...so the storm may have been closer to 110 kt than 100 kt--similar to the intensity of Katrina in LA.

[CrazyC83]

If a Katrina-type storm were to make landfall at, say, Tarpon Springs (or even up towards Spring Hill), what would the surge be in Tampa Bay?