Why Surface Wind Measurements in Hurricanes Don’t Match Satellite Estimates

Dr. Philip Klotzbach writes via email:

Discussion of near impossibility of actually recording strongest winds in a hurricane

Brian McNoldy and I wrote a blog for Capital Weather Gang yesterday where we called out Matt Drudge about his claims that the National Hurricane Center was over-estimating the winds in Hurricane Matthew. As you know, it is nearly impossible to measure the strongest winds in a hurricane due to a variety of factors. Brian and I covered these in some detail in this blog.

Matt Drudge commented on our story last night, and since then, there have been hundreds of comments from individuals claiming that NHC is just a cog in the “global warming” machine. As you know, I’m certainly not a fan of over-hyping climate change and have published several papers showing no trends in intense hurricane activity over the modern observational period.

I feel that it would be incredibly useful for a well-respected climate skeptic website to pick up this story and give some additional credence to it. If the eyewall of Hurricane Matthew had been 50 miles further west, it would have been devastating for the east coast of Florida! We may not be so lucky next time, and I don’t want people to become armchair meteorologists and think the winds are much less than NHC reports.

Happy to help – Anthony


Hurricane intensity is not exaggerated to scare people, and here’s how we know

By Phil Klotzbach and Brian McNoldy

Hurricane Matthew brushed the East Coast as a Category 4 in early October. It scoured the Florida coast with a storm surge that washed out roads and flooded homes and businesses. It dumped over a foot of rain on the Carolinas and triggered deadly flooding. And before all of this, Matthew devastated Haiti and killed at least 600 people — it’s possible we’ll never knowthe final death toll.

But the hurricane’s deadly encounter with Haiti and multibillion-dollar brush with the East Coast wasn’t bad enough for some people — mainly, Matt Drudge, a political commentator and news aggregator, who tweeted skepticism about Matthew’s actual intensity as it barreled toward Florida. Perhaps, he suggested, the National Hurricane Center was overhyping the storm’s maximum winds.

Drudge deletes his tweets regularly, so here’s a screenshot from the morning of Oct. 7:

“The deplorables are starting to wonder if govt has been lying to them about Hurricane Matthew intensity to make exaggerated point on climate,” Drudge mused. (“The deplorables” refers to supporters of Republican presidential nominee Donald Trump.)

Then this (image saved by the BBC):

Drudge argued the point based on data from Caribbean weather stations and buoys that were not reporting winds as strong as what the National Hurricane Center used in its advisories. But the National Hurricane Center uses a lot of different methods to determine a hurricane’s actual peak intensity, and there are some serious issues with relying simply on weather stations and buoys.

The National Hurricane Center — or NHC — uses a variety of satellite-based tools to estimate a hurricane’s winds when the storm is far from land, and as the storm moves closer to land, additional devices are employed.

Hurricane Hunter aircraft are dispatched to investigate the storm in more detail. These planes measure winds both at flight-level, which is approximately 10,000 feet, as well as at the surface through the use of a device called a stepped-frequency microwave radiometer which measures radiation transmitted from the ocean surface back up to the airplane. When high winds churn up the ocean surface, the amount of microwave radiation measured is diffused compared with the amount of radiation measured when the ocean surface is calm.

NHC also utilizes devices called “dropwindsondes” that are dropped from the aircraft at about 10,000 feet and measure temperature, humidity, pressure and wind speed as they descend to the surface. When a storm nears shore, radars can also be utilized to estimate wind velocity.

All of these different measurement devices, combined with surface observations from ships and buoys, are combined to assess a storm’s intensity.

There are several reasons buoys are likely to underestimate a hurricane’s wind.

First, buoys utilize an eight-minute average maximum wind, and NHC uses one-minute average wind. The shorter the wind averaging time, the stronger the reported winds will be. The adjustment factor would be approximately 0.90, which means if a buoy measures 80-mph eight-minute average winds, the one-minute average wind would be about 89 mph.

A more significant bias in the buoy data is the insanely severe ocean conditions that it’s reporting from. Waves in a major hurricane can be as high as 50 feet, and consequently, a buoy experiences very large crests and troughs as the hurricane passes over. When a buoy is on the crest of a wave, it is exposed to the full force of the surface winds, but when it is in the trough of a wave, it is going to be sheltered from the strongest winds. These weaker winds measured in the trough of a wave get averaged into the eight-minute averaging period measured by the buoy, which therefore gives the buoy’s estimated wind speed in a hurricane a low bias.

Surface stations also have significant measurement challenges in hurricanes.

Often during strong wind speeds, the anemometers break. Storm surge can also wash away weather stations before the strongest winds come ashore.

Importantly, a hurricane’s wind field is a wide swath, and the strongest winds may only be found in a small region of the hurricane. The number of surface stations in any area struck by a hurricane is likely not dense enough to accurately measure the strongest winds. For example, one 2014 study found that on average, “a single perfect anemometer experiencing a direct hit by the right side of the eyewall will underestimate the actual peak intensity by 10-20% percent. Even an unusually large number of anemometers (e.g., 3-5) experiencing direct hits by the storm together will underestimate the peak wind speeds by 5%-10%.”

The number of surface weather stations reporting hurricane-force winds, even in notoriously strong United States landfalling hurricanes are surprisingly small. For example, in NHC’s post-storm report on Hurricane Katrina (2005), a Category 3 hurricane on the Saffir-Simpson Wind Scale, only five surface weather stations reported sustained hurricane-force winds, with the strongest sustained hurricane-force winds reported being 86 mph. Many of the surface stations reporting the strongest winds were also flagged by NHC as being incomplete, indicating failure of the station before the strongest winds likely occurred. For reference, NHC estimated Katrina’s Louisiana landfall to have had maximum winds of 125 mph and its Mississippi landfall to have had maximum winds of 120 mph.

The peak winds in a hurricane are confined to a small region of the storm called the eyewall, which is the ring of thunderstorms that surround the eye. Beyond the eyewall, the wind speed drops off relatively quickly.

In the case of Hurricane Matthew, while the southeast United States suffered significant flooding damage due to both storm surge and heavy rain, the east coast of Florida dodged a huge bullet. Had Matthew tracked 50 miles farther to the west, the eyewall of the storm would have come ashore, causing much more extensive flooding as well as significant wind damage.

In any case, Matthew’s strongest winds would likely not have been measured by a weather station. The National Hurricane Center provides the best analysis that science can offer.

Advertisements

260 thoughts on “Why Surface Wind Measurements in Hurricanes Don’t Match Satellite Estimates

  1. This is why all real scientists are paying the price for not standing up to the joke that climate science has become. Everyone is painted with suspicion now even if good data is being gathered and presented. I hate to say it but if these guys cried foul when so much bad science was being presented in the last 10 years, we wouldn’t have to now defend against such accusations.

    • I wish they would get of the Deplorables routine. “Deplorable” was a label created by a pathological liar and psychopath, who is amoral, arrogant, elitist, and foul to the core. To use “Deplorables” means you condone the actions of an evil person who looks down her nose at everyone.

      • How does the ” Deplorables” meme even fit into this discussion? As far as I can tell, your use of the term is first, unless there was a deleted comment which is unseen?

      • Well the psychopath is opposed for teh presidency by an economic genius who thinks that he can renegotiate the US national debt and create jobs by initiating a trade war with China. When asked by Sean Hannity how he would defeat ISIS and if he would need ground troops to do it, he replied that he would do it “swiftly and decisively.” The depth of military thought in this fool’s mind cannot be understated. This new Napoleon and embodiment of Julius Caesar noted that he knows more about ISIS than the generals. He absorbed his advanced knowledge in strategy and tactics while learning close order drill at military school. How could generals decades of experience and training stand up to a sharply done presnt arms on a military school parade ground.

        Now think that this fool, whose only idea is whatever pops into his mind at the moment and whose relationship with the truth is so slight that it impossible to fact check the torrents of nonsense that flow from his mouth, has a credible chance of winning the US presidency

    • This is why all real scientists are paying the price for not standing up to the joke that climate science has become. Everyone is painted with suspicion now even if good data is being gathered and presented. I hate to say it but if these guys cried foul when so much bad science was being presented in the last 10 years, we wouldn’t have to now defend against such accusations.

      EXACTLY.

      Thanks to Dr. Philip Klotzbach for the interesting article but the crisis of confidence reflected by Drudge’s comments is totally the fault of the scientific community as a whole.

      Not only were the softy-lefty climate warriors allowed pervert the published literature and act as political activists whilst posing as scientists but when this was all exposed in Climategate emails, rather than being denounced by the majority of responsible scientists there was a resounding silent. They embarrassingly just looked to other way hoping it would all blow over and be forgotten.

      Worse, when there finally were some “investigations” into the malfeasance they were a total whitewash. Fellow academics just circled the wagons and defended the indefensible.

      So, if the science community now finds themselves with little credibility in the public’s eye, they only have themselves to blame.

      • While the science community is partially to blame for its complacency, much of the blame goes onto the universities that no longer educate people in science, but instead substitute activism. One can also point the finger at journalism, but honestly I think that all goes back to universities as well – you have atrocious journalism because journalists are taught to be activists instead of journalists. You have terrible scientific journals because the editors don’t know any better – they never learned what real science is.

        Its amazing to me that good scientists continue to exist despite this backdrop of anti-scientific culture – so I guess I can’t bring myself to blame them too much for hiding in their fox holes and hoping they can continue to follow their work. They aren’t politicians after all. But they should not be surprised at the backlash as people start waking up to the fact they have been misled and exaggerated to.

        Public opinion is like a pendulum – eventually it is going to swing the other way and be far more skeptical of everything no matter what the evidence suggests (like the distrust in common vaccines for example). Its up to people (not just scientists) trained in scientific methods and practicing healthy skepticism to keep the likely truth in the public discourse.

      • Yeah, well, a lot of people just want a get of jail free card to explain away how they rejected reasonable information that they obviously had little understanding of.

      • It is more than just tarred by bad science. Author specifically states they are now correcting via satellite and other data not available in the previous record. So new “estimates” or “adjusted” values are apples and oranges. If (like news media does constantly) the numbers are “unprecedented” and “worst ever”, they are comparing new to old that doesn’t match. Stop justifying this. You should be adjusting new methods Down to compare with historical methods, or ensure comparisons are not made. You sir, have no actual clothes on.

    • “This is why all real scientists are paying the price for not standing up to the joke that climate science has become.” Mr Nail, meet Mr Head.

    • @ David

      Yes, good point.

      I have trouble believing ANY so-called scientist working in the climate field. There must be some honest people working in the field but the dishonest con-artists are such a vast majority that I don’t trust any of them.

      • That’s not the point, Phil. Have you, or anyone else, ever taken a public stand against Kevin Trenberth’s repeated public misrepresentations that intense hurricanes are the ‘new normal’ due to climate change(sic)? If so, when? We all missed it.

      • Phil, even if you are apolitical and you want your work (and referenced paper) to represent your professionalism, you still need to recognize that the information that you (through the NHC) release to the media, has been hyped and twisted by both the media and the politicians.

        When the media and/or the politicians use you (through your NHC news releases) as a tool to push their separate agenda (be it hype to sell airtime, or an attempt shore up climate followers in an election) you need to protect yourself from that association. You need to say they are wrong for trying to interpret/twist your works in a manner that will ultimately harm your ability to communicate with the public in the future.

      • Pat Frank
        October 28, 2016 at 8:37 am

        That’s not the point, Phil. Have you, or anyone else, ever taken a public stand against Kevin Trenberth’s repeated public misrepresentations that intense hurricanes are the ‘new normal’ due to climate change(sic)? If so, when? We all missed it.

        Please speak for yourself. You don’t appear to have read the Klotzbach/Gray hurricane forecasts, some of which I’ve posted here. As Phil became the lead author of them, Bill started writing more critically about claims that climate change was bringing bigger storms. I think he especially addressed Kerry Emmanuel’s claims. In terms of tropical storms, he was a more prominent voice than Trenberth.

        I’m at work, so I don’t have time to link to one of the Klotzbach/Gray forecast or my posts here, where I probably pointed out Gray’s commentary. Perhaps you can take a little time and do a little reading yourself.

      • Ric Werme, my comments were addressed to Phil. I’m aware of Bill Gray’s critical comments and have watched one or two of his presentations. He’s a hero.

        However, Phil was responding to the posts of Robert of Texas and markstoval, who asked where the honest scientists are who stand up publicly to the falsehoods loudly and incessantly promoted by AGW consensus science-impostures.

        Phil’s reply was that he had published honest science. I’ve no doubt about that. But the field of battle for honesty and integrity in climate science is political and public, not in scientific journalism.

        That was the point Robert and mark made, and that Phil (and you) apparently side-stepped.

      • Not quite sure what you mean here. Publishing in peer-reviewed journals is a scientific “public stand”. For example, I published a paper in 2006 discussing issues with papers published in 2004/2005 that received tons of media attention. Of course, publishing papers saying that TCs aren’t necessarily getting worse isn’t going to get the same kind of media attention. And, let’s just say, at the age of 26 (at the time), I wasn’t going to get the same publicity as someone with the stature of Dr. Gray received. Since I worked very closely with him at the time (and did until his death earlier this year), we decided that it made a lot more sense for him to handle most of the publicity aspects.

        But if you prefer other public means, you can see many recent posts on my Twitter feed (@philklotzbach) where I discuss trends (or lack thereof) in TC activity. Dr. Gray and I also included a lengthy discussion on our seasonal hurricane forecasts for many years about TCs/climate change. Several of our press releases discussing our seasonal hurricane forecasts also delved into this issue. I also recently did an AMA on Reddit where that topic was discussed quite a bit. I’d say that I’ve done a pretty good job of stating my take on the issue publicly. Similarly, Ryan Maue at Weatherbell has also done an excellent job describing the TC/climate change both in published research as well as via his Twitter feed.

    • I remember a bedtime story about this as a kid.
      The Boy Who Cried Wolf used this same storyline.
      Who knew it was really possible to stir up a whole town?

  2. “sustained winds” would indeed average their sustained number whether it was a one minute interval average or 8 minutes…….words do have meanings “sustained” means staying AT or ABOVE the stated wind speed…….the excuses used are an insult to intelligent people…..

    • The shorter the wind averaging time, the stronger the reported winds…of course, they are gabbing gusts..not sustained

      It would have also helped their credibly if they had told people how small the eye was…it spun down, making eye wall winds faster (I hate “stronger”)

      • “Grabbing” is clear in context, Latitude. “Gabbing gusts” is the politicians’ and TV journalists’ speciality.

      • “Stronger” is a legitimate descriptor. The problem with winds is what? They blow things down–with force. So, the relevant parameter of concern is the dynamic pressure of the wind, which is proportional to the SQUARE of the wind velocity. Works every time.

    • These aren’t going to be the same thing. For example, if you have one hour’s worth of wind measurements, you can naturally grab a 1-minute segment in that hour that is going to have a stronger wind speed (or alternatively lower wind speed) than if you grab an 8-minute segment. The World Meteorological Organization recommends a 0.88 correction for a 10-minute to 1-minute average, so for 8-minutes, we approximated 0.90.

    • The thing that caugt my attention was that the hurricane passed “almost” directly over the buoy but still gave “sub-optimal” placement. The excuses are build in even before the reporter finishes. Also other have commented on the adjustments between the 8 minute vs 1 minute data. How can you presume that it is 10-20 % higher wind speed because the lag is 1 minute vs minute, to me that is already suspected adjustments, could it not be lower?. Why not use 3 min data 6 min data and so on. To me this is manipulation As an aside would the storm mot have lost energy if it would have come on land sooner?

      • Directly over is still suboptimal. Optimal would be a combination of being in the eyewall to the right of the path, perfect timing, and a moderate amount of luck.

        Being in the right eyewall the vector addition of the storm’s forward motion and eyewall rotation is maximized. It would be least on the left side (a big break even at Florida’s distance) and in between for a direct hit.

    • The buoy problem is well known. It is a little like have your anemometer behind a building – when the wind is from the wrong direction you miss it altogether.

      The Dropsonds have worked for 30 years at least (unless they have regressed since I used to work with them). The pre-mission calibrations are good assurance they will do what they were designed for. We had a few with dead transmitters, but that is what spares are for. Nothing worse than a crew coming back with no data from a dead transmitter.

      The actual max windspeeds aren’t even throughout the eyewall, but only in a small area of that. A really compact storm’s max speeds can be missed if the station is not within a quarter of mile swath along the track.

      I may have some doubts about the satellite estimates since they really depend on accurate modeling, but the aircraft estimates I am pretty certain of. When the two agree, I assume the model of the environment of the storm is pretty close and have more trust in the satellite estimates, but I am an amateur, although an experienced one, so take that opinion with the appropriate grains of NaCl.

    • Sustained winds as opposed to gusts. That’s all it means.
      As to insulting intelligent people, how would you know?

  3. Buoys provide the most accurate measure of all the tools discussed. 8 minute is the standard! The Storm tracked along the coast barely breaking 20ft seas with winds less than 60 knot 8 min average; which is not what you would get from a Cat 4.. this from buoys that were within 20miles of the center and one that the storm tracked directly over. Even NHC normally uses buoy data when its available. an actual CAT 4 off the coast like that would have killed thousands, which is what happened last time a big storm rolled along that coast. It is very disturbing to see good buoy data being ignored. I personally think that hanging out for any hurricane is a serious mistake but when someone “survives a cat 4” getting them to evacuate becomes virtually impossible.

    • ..exactly right David

      Now they want to move the goal posts again……today’s cat 4 is yesterday’s cat 2….because buoys are no longer the gold standard

      Now they have computer models to tell them the sustained winds.

    • Given that Matthew’s MSLP was ~940 mb at the time when the buoy was measuring those winds, it would require an extraordinarily large wind field (aka Sandy) to have winds that weak in the eyewall. In terms of size, Matthew was not particularly noteworthy for a hurricane with a 940 mb central pressure. If you look at most hurricanes of the past, very few sustained hurricane-force winds above about 85 mph are reported. For example, both the Freeport and Nassau Airport stations stopped reporting just as Matthew was coming ashore. Exuma Airport in the Bahamas did report a very impressive max wind of 119 mph with gusts to 144 mph as Matthew came ashore.

      • Phil, I have been at sea in hurricane force winds, 25ft seas for as long as Mathew was around is extra ordinarily small for a cat 4. A good gale can achieve that given time. My wind anemometer worked just fine and the buoys NOAA uses are not small and they also work fine. I get what your trying to say, there are certainly difficulties in measuring wind speed but Mathew did not leave the kind of damage behind that a prior to Safir Simpson hurricane (1972 or so) would have left. The primary determination of hurricane strength is not wind per se but destruction caused by that wind. Which is typically what the reanalysis will show since most of the time anemometers don’t last out a severe storm. Hype is awful when used over and over. Most of the people who live in hurricane prone areas completely ignore advice. That is the real danger! Bringing some truth to the hype is necessary to restore trust. NHC removed the Sampson part of the scale back in 2010 and they have gotten a bit weird about storm strength since then. They need to get back to a practical calculable means of forecasting the actual strength of a storm. Certainly the reanalysis of Mathew should be interesting unless politics have gotten involved in that as well.

    • Photos of surface damage confirm the buoy. Land based anemometers confirm the buoys and the actual surface damage.
      Mathew was Category 1 over the Bahamas. Mathew was tropical storm at US landfall.

      • One overturned car, the rest are fine. That’s Category 1.
        One photo of light weight aluminum awnings, some tree limbs. All consistent with Category 1.
        I’ve looked at dozens of other photos of the Bahamas that are all consistent with Category 1, no exceptions. Gusts knock over vehicles on highways all the time. I’ve seen decades of midwest tornado damage and plenty of photos. I’ve looked at structural damage assessments by qualified engineers and insurance companies that know exactly and quantitatively what wind speeds and exposure times correlate with observed damage.

        For those who don’t know about Saffir-Simpson Category 1. Here is the short definition.
        Well-constructed frame homes could have damage to roof, shingles, vinyl siding and gutters. Large branches of trees will snap and shallowly rooted trees may be toppled. Extensive damage to power lines and poles likely will result in power outages that could last a few to several days.

        Any photos of Category 2 damage in the Bahamas?? Any evidence or data showing Category 2 winds anywhere in the Bahamas?? Of course not, because there were no sustained winds of that force. The photos are selected by passers-by to show what they thought was significant damage.

      • One consistent theme, a hurricane (cat 1 too) have the potential to spawn tornados which can destroy a small area. so the strength of the storm is demonstrated by widespread damage not just a single car turned over. Tornado wind speed is not what the storm as a whole is rated by. so yes Bahamas were most likely hit with a cat 1. if Mathew had been major when it hit the Bahamas you would get more than roof damage. wide areas of homes would be gone areas would be treeless. When Katrina went north through Mississippi it even though it had calmed down a bit there was a near 30 mile wide area of total destruction that the media ignored because it wasn’t New Orleans.

    • Probably my own bias but it seemed to me the storm held a “cat 3” status well beyond what I would have considered reasonable. Seemed like there may have been a hope to break the cat 3/US landfall drought.

      > cat 2 (by 2004 standards anyway) is where I dust off the generator.

      I understand the precautionary principle but for whatever reason this was a blown forecast. It used to be that category was a combination of wind speed and pressure; I thought I read that they’ve dropped pressure from the parameters?

    • I had no idea the 8 minute average was a standard. For NWS ground measurements I was sure it was a 1 minute average, but I’m going to verify that before I make that claim again!

      Thanks Phil – I learned something. :-)

      • Ric, it appears that a 1-minute average is the standard, but buoys are only set up to record 8-minute averages. That seems like a mistake to me – if there exists a standard then buoys should be designed to meet that standard.

        Klotzbach’s point about shelter from waves is interesting. It means that buoys should be principally used to determine gust strength, which itself can give some idea of plausible sustained winds.

        Rich.

      • The standard set by the WMO is 10minute wind average. its been 8 on the buoys for ever so changing the standard means the buoy data is no longer comparable. 8 is the standard. NHC uses one minute because you can’t really get a 8minute reading from an aircraft (moving too fast to read in one area for that long) NHC likes their aircraft which all that info has to be inferred, There are many problems with aircraft data they don’t like to talk about but Phil was nice enough to put some of it out there.

      • David, thanks, that’s interesting, and I didn’t know that. So it becomes more apples and oranges again. Unless…

        In this day and age we can be sophisticated enough to record multiple data streams. So, the 8-minute values could be kept for historical comparisons, and 1-minute values could be recorded as well for comparison with NHC (and with WMO who also use 1 minute values?). There is so much that could be done to get better data, but nobody seems interested.

        Rich.

      • When we recorded weather data for the NWS at the Kennedy Space Center. The sustained wind program was written to record a 2 minute moving average updated every 5 seconds.

  4. My quick read may have missed it but the hurricane quadrant being measured is also important. In the northern hemisphere the Coriolis force assures that the wind around a low will always be counter clockwise. Considering the direction of motion of the hurricane itself to be the ‘front’, the right hand quadrant will always have a greater ground wind speed than any other quadrant because the speed of the storm is added to this sector and subtracted from the opposite measurement. If a speed sensor is not placed there it will not be able to measure the highest wind speed. Similarly, if that portion of the hurricane doesn’t pass over the ground the apparent measured speed will be lower than the actual max sustained wind.

  5. I recall that Nassau, Bahamas received a near direct hit when Matthew was still category 4. Here’s the aftermath:

    http://eleutheranews.com/?p=6765

    Bahamas Marinas Ready And Open For Early Fall-Winter Season
    By Staff2 / in Business, Highlight / on Saturday, 22 Oct 2016 11:44 AM

    NASSAU, THE BAHAMAS – The 47 marina and allied members of the Association of Bahamas Marinas are open for business and are anticipating a busy winter season, almost all members having experienced no significant damage during the recent hurricane Matthew.

    Matthew’s small eye passed over Old Bahama Bay Resort & Yacht Harbour. Even at category 4, the marina and resort did extremely well and suffered little damage. The marina continues to receive and welcome yachts, although power is yet to be restored. Many yachts have their own-power generating capacity. The resort itself escaped damage and will open once power is back.

    ABM members were reporting full restoration of power and water to their docks and resorts within a day or two of the hurricane, the small eye of which traveled over the Exumas, Nassau, Chub Cay and Grand Bahama Island. All the Exuma marinas including Highbourne Cay and The Marina at Emerald Bay (Sandals) reported no or slight damage with Safe Harbour at Cave Cay soon to repair two damaged docks and fully operational with other docks.

    In Nassau and Paradise Island, all marinas escaped unscathed. In the harbour, Bay Street Marina and Nassau Yacht Haven reported no damage and are fully functional. Palm Cay on the southeast coast of New Providence only had a couple of channel markers misplaced while marinas at Lyford Cay and Albany were undamaged. Albany chose to close temporarily to cleanup foliage and make the luxury estate pristine.

    • Photos of damage everywhere confirm that Mathew was not as strong as the NHC claims. Amazing that such a “powerful” hurricane can pass directly over populated areas with little wind damage.
      That’s the point of the Saffir-Simpson scale, even if there were no measurements of wind speeds, the strength of the storm is determined by the surface damage. Trees, poles, roofs, metal signs, etc. all have known structural properties, over the years the known wind speeds are calibrated to the actual damage. Photos of the actual damage confirm the actual wind forces acting to create the damage. The actual wind forces causing the damage are recorded by the photos of the damage. Mathew was Category 1 at most over the Bahamas. The surface damage observed in Florida to North Carolina show less wind force, so the storm was weaker, tropical storm level.

  6. In a sense it doesn’t matter who gets the velocities right. The impact of a revolving tropical storm can be extremely destructive. Certainly, as the good folk of Darwin found at Christmas 1974, a category 2 cyclone can destroy a whole city. I have experienced a category 5 cyclone in a company town that was designed to withstand 135mph winds. There was significant damage but, fortunately, no lives were lost, primarily because of building design and company rules which dictated how their employees protect themselves. I have no wish to ever encounter conditions in the eye of a cyclone again.

    • ‘The northern Australian city of Darwin was devastated early on Christmas morning 1974 when hit by the tropical weather depression that was given the name Cyclone Tracy. As the eye of the cyclone passed over the city between midnight and 7am on Christmas morning, torrential rain fell and the winds were officially recorded at 217 kilometres per hour prior to the Bureau of Meteorology anemometer being destroyed.’
      Hardly a Cat 2, Gary.

    • No way cyclone Tracey was a Cat 2 storm. They barely had low resolution sat pics back then and the destruction of the buildings and port were more consistent with a major cat 3 or 4 storm. My next door neighbour lived through it and told me how they survived by lying in a cast iron bath tub whilst watching their house explode around them. My sister lived through Cat 5 cyclone Vance in Exmouth Western Australia. She described watching a big 4×4 vehicle blowning down the road like a toy smashing into a house. She left town and never went back.

    • No way cyclone Tracey was a Cat 2 storm. They barely had low resolution sat pics back then and the destruction of the buildings and port were more consistent with a major cat 3 or 4 storm. My next door neighbour lived through it and told me how they survived by lying in a cast iron bath tub whilst watching their house explode around them. My sister lived through Cat 5 cyclone Vance in Exmouth Western Australia. She described watching a big 4×4 vehicle blowning down the road like a toy smashing into a house. She left town and never went back. Darwin post cyclone Tracey.
      http://www.nma.gov.au/__data/assets/image/0011/448526/Wagaman-685w.jpg

  7. We may not be so lucky next time …

    This is a true, but, a highly-likely-to-mislead statement. It strongly implies, falsely, to the average reader, that hurricanes ARE getting worse, but, that we just “got lucky,” this time.

    The truth of the matter is: regardless of how strong the winds were, there have not been any >= Cat. 4 hurricanes to make landfall in the United States for over 11 years. That fact is the significant one. To say, “we just got lucky,” is to say essentially, “Yes, hurricanes are ramping up, it’s just that they are all out at sea.”

    THAT THEY ARE ALL OUT AT SEA (v.a v. the U.S.) IS THE POINT! :)

    Language is important — always remember, we are fighting a battle for TRUTH (give the enemies of truth NO foot hold to launch an attack at you):

    The AGWers run with such statements and will use it in their propaganda that hurricanes are worsening and that it is due to human CO2 emissions.

    All you lukewarmers who like to talk about how human CO2 “causes” “some” warming — heads up:

    if you do not at the same time, every time you say something like that, carefully (as a good scientist always does) clarify what you mean (i.e., that such “causation” is theoretical only, not measured (in the climate system called “earth”) AND that such theorized “warming” would be, if it is not, indeed, overwhelmed by natural warming, so slight as to be negligible vis a vis a climate shift on earth) YOU ARE HELPING THE ENEMIES OF SCIENCE.

    • The term “lucky” was in the colloquial email that I sent to Anthony, not in the article I wrote for CWG. But, if Matthew had come ashore, the results would have been much different for the state of Florida. That’s the challenge of forecasting a hurricane tracking parallel to the coastline. Very small changes in track have very large differences in impacts.

      • Dear Dr. Klotzbach,

        You were BRILLIANT in your forecasting of Matthew. We, here, were so very impressed.

        Thank you for taking the time to respond. I think you may have (due to my not writing clearly enough) not quite seen my point, but, that’s okay. I am often misunderstood around here!

        Yes, indeed. It would have been a very bad situation if it had been farther west.

        Dr. Gray would be so proud of you, carrying on the torch with such painstaking care to be accurate, and with such a bold willingness to stand up for what you know is right.

        With admiration,

        Janice

      • very early on, the climate catastrophists started prancing around with their rattles in great anticipation of what was to come- chanting numbers and predicting more freakin alpaca lips. and then, AS USUAL, it didn’t happen.
        was anybody not expecting the fallback to be ‘we were so lucky to dodge the bullet but we will get our heads smashed like pumpkins dropped from the parking ramp NEXT TIME for sure!

        hype is hype. no apology is acceptable.

      • Matthew would have certainly been worse had it come ashore. We *were* lucky it stayed out to sea.

        The Governor of Florida was criticized for supposedly hyping the threat, but it wouldn’t have been hype had the storm moved far enough west, no matter what category of hurricane it was. The Governor was properly cautious in such a situation.

        What I glean mostly from this article, is measuring wind speeds inside hurricanes is not as precise as we might prefer.

      • I think that the Bahamas tell us what Matthew would have been like if it had made landfall in Florida. Namely, Category 1 damage (as per bw’s comments upstream) in a few areas.

        Richard.

  8. I enjoyed reading this. Most folks think you attach an instrument and get a measurement … end of story. Instrumentation is an art and a science. If you don’t understand what you’re doing, you can be led seriously astray. The consequences range from embarrassment to death.

    Dr. Klotzbach tells it like it is and gets my respect. The people who don’t get my respect insist that their results are +/- 0.1% when the data that they are using doesn’t even support +/- 10%.

    • +100 commieBob. I’ve had to try to figure out what the measurements meant from flight tests for a couple decades – instrumentation, IMHO, is much more art than science. I took a one-semester undergrad course in instrumentation – one of the most practical course I ever took.

      Dr Klotzbach, thank you for your detailed article, and please recognize that for many here it has been so long since they’ve seen real science that they tend to suspect everything and everyone. As a lad on Long Island (NY), I spent several days in the basement, on several occasions, when major hurricanes passed over. I have gone out during the passage of the eye on two occasions and enjoyed how calm they were, especially since we then knew that the worst of the storm had passed.

      Keep up the good work, keep warning people of the storms and the possible damage and loss of life, and keep providing regular updates to the storm track.

      • “I have gone out during the passage of the eye on two occasions and enjoyed how calm they were, especially since we then knew that the worst of the storm had passed.”

        Hang on, if you are in the eye of the storm doesn’t that mean that you have suffered the winds in one quadrant of the eyewall, but are soon to suffer the winds in a different quadrant?

        Rich.

  9. OK, if they are going to do this they need to change the stated standards for classification. They says Buoys are not accurate because of masking by waves. And common sense rules out a dropsonde hanging at 10 meters above the surface for 1 minute to get a sustained wind reading. And satellite measurements at 10 meters above the surface are corrupted from spray, spume, and waves. Wind speeds at higher altitudes above the surface in these storms are much higher because of the lack or friction with the surface but there is no scale which grades the winds to say that if the wind is at a certain velocity at a particular level it will be at a particular velocity at another level. Their “estimates” are guesses and not hard data. So we see that they have positioned themselves to declare the strength of a tropical cyclone without anyone else being able to validate their claims.

    • It is simply impossible to state with definitive certainty the intensity of a hurricane at any point and time. The max winds typically are in only a small portion of the eyewall. NHC forecasts estimate that their error is about +- 8 mph with aircraft reconnaissance. With tropical cyclones out in the open ocean and estimations from satellite, these uncertainties are even larger. There is much more confidence in measurements of the central pressure of a hurricane than there are in the max winds given the relative ease of measuring this parameter.

      • yet that doesn’t seem to stop anybody from stating with lurid conviction the horrible devastation in store, does it?
        first it was doom and now we’re gonna get precious about eyewalls and confidences?
        what a swift pivot from ‘State of Emergency’, eh?
        if nothing else, you suckers can dance a fandango like a burning spastic.

    • I suspect, that much like tornadoes, the true relative intensity should be from analysis of on ground destruction (or lack thereof), after the passing. From the number of people remaining behind after being required to leave by authorities, it would appear that some locals had also been “fooled” one too many times by the weather “experts” wild guesses and decided to take their chances. I remember when the predicted path of Katrina changed so many times that when they did decide that New Orleans was finally truly at risk, it was essentially too late. Would anyone argue from watching regular storm forecasts that they routinely heighten the risk just in case they are wrong? Another thing I hate is the propensity to play with weather radar color intensity so that when it’s supposed to be “pouring” there is not even a sprinkle (or the radar area location is fouled up which is also believable). Much like the 7 day (or even 3 day) forecasts, it’s no wonder that many who actually pay attention treat these predictions as numerology..

      • Correct, surface damage directly measures wind force. Just test a stop sign in a wind tunnel. The sign bends over consistently because the bending forces are known for the type of steel pole. Same for a wood wall, damage is proportional to wind forces. Over years of photographic evidence in different storms, with nearby anemometers, the strength of the storm is observed. Mathew was not a hurricane over the southeast US. The damage is entirely consistent with tropical storm force winds.

      • If we determine the “intensity” of a hurricane based solely upon ground destruction that leaves a few holes.
        How do you measure intensity where the is no ground to be effected such as out-at-sea?
        How do you measure destructive damage? Reverse engineer the force needed to create such damage?
        Surely one cannot categorize ground damage based upon monetary damages or else Haiti will never see anything larger than a Cat 1 as there isn’t enough valuable property to damage.

    • Other than in the Atlantic (and very rarely in the NE Pacific), all estimates of hurricane intensity at sea are derived from satellite imagery. Aircraft reconnaissance are a luxury for us in the Atlantic basin, and in the NW Pacific through 1987. On a positive note, the NW Pacific will start up limited aircraft reconnaissance again next year.

      • In the early 80s we used to fly them out of Guam. Even when we closed that squadron the unit at Keesler used to deploy a plane out to fly the peak of the season. Of course that all wrapped up by the late 80s

      • I am sorry, but your analysis and report have one really massive glaring problem.
        Historical Data.
        The Hurricanes/Typhoons/Tornadoes that DID devestate areas of the US and the Pacific had ground based Wind Speed measurements comensurate with the damage that they caused.
        So basically you are bringing in to question all that historic data which was measured using just ground based instruments and the very Grading Systems currently in use.

        It is therefore incumbent on the Scientists to try and ensure that other methods match those ground based instruments for Grading Purposes.
        Modern Techniques may be better for understanding Weather Events but should not be used to Hype (compared to ground measurements) current and future events.

        It is much like modern Electronic Thermometers that pick up Transient temperatures that the old Mercury ones could not, should not be used for establishing “Record” temperatures unless they can be verified by the old technique.
        Otherwise we are trying to compare apples & oranges.

      • A C Osborn; that was exactly my thoughts as I was reading. While I don’t see any errors in the essay, what I got from it is that historic hurricane data needs to be padded up, which would suggest that hurricanes are actually trending down.

  10. “If the eyewall of Hurricane Matthew had been 50 miles further west, it would have been devastating for the east coast of Florida,”

    This statement looks like hype to me. Are we being no asked to consider that if the storm would have cost great havoc across the entire coast? If the eyewall of Hurricane Matthew had been 50 miles further west, it would have been devastating for “some portion of the” east coast of Florida – but hitting land would have slowed it down. It would not have traversed the entire length of Florida at the same speed and intensity if it were 50 miles further west. Am I. Issuing something?

    • Sorry-most glaring typos fixed.

      “If the eyewall of Hurricane Matthew had been 50 miles further west, it would have been devastating for the east coast of Florida,”

      This statement looks like hype to me. Are we being asked to consider that the storm would have caused great havoc across the entire Florida coast? If the eyewall of Hurricane Matthew had been 50 miles further west, it would have been devastating for “some portion of the” east coast of Florida – but hitting land would have slowed it down. It would not have traversed the entire length of Florida at the same speed and intensity if it were 50 miles further west. Am I missing something?

      • only thing i think you left out is IT DID NOT HAPPEN- which, after all, is the reality that brings forth the apologia for the sick fantasies.
        and next time a kitten sneezes it will have a name like HURRICANE TROGDOR because they are such careful scientists, doing their best. they only need more resources, you know. some airplanes would be great. learjets, maybe. and some lavish funding. maybe a supercomputer. because
        HEAP BIG WARMY.

      • Why not also announce what would have happened if the hurricane had been 50 mi further east too? I agree the 50 mi west statement and the “luck” statement are hype.

      • “If the eyewall of Hurricane Matthew had been 50 miles further west, it would have been devastating for “some portion of the” east coast of Florida – but hitting land would have slowed it down. It would not have traversed the entire length of Florida at the same speed and intensity if it were 50 miles further west. Am I missing something?”

        That “some portion” of Florida would have been hit very hard. As far as I know, noone has claimed a hurricane would maintain its intensity the full length of Florida, once it hit land.

        Of course the hurricane will lose intensity as it moves inland, but the people who get the direct hit from a hurricane will get the full intensity of the hurricane. Since you don’t know the exact path of the hurricane, you have to warn the whole area because somewhere in that area the hurricane will hit that “some portion”.

        I don’t think the hurricane warnings were overblown at all. They were reasonable under the circumstances.

        What was overblown was the supposed connection between hurricanes and human-caused climate change, that Hillary Clinton and other alarmists were trying to make.

      • If, as stated, the hurricane was moving parallel to the coastline, a little distance makes a big difference.

        Think of it this way: cyclones are vortices, and the winds subject to vortex flow have a velocity that is inversely proportional to the distance from the center of the vortex (eye of the hurricane). The closer to the eye, the faster they go–up to a point where vortical flow breaks down and the core of the vortex must actually rotate like a phonograph record (velocity linearly declines with radius). But the blowing force of a wind is simply its dynamic pressure, which is proportional to the square of its velocity. This means that the wind force is inversely proportional to the SQUARE of the distance from the center of the hurricane. I don’t know how far that 50 mile point would have been, but this means that if the hurricane came close enough that the distance to its center dropped in half, the force of the winds would have been 4 times greater.

        To press the point, a 40 mph wind has 4 times the dynamic pressure of a 20 mph wind. And an 80 mph wind has 4 times the dynamic pressure of a 40 mph wind. It grows rather fast.

  11. I am not qualified to post an opinion here — but when did that ever stop me?

    It seems to me that searching out the highest wind speed of a storm is sort of “cherry picking’. That is made plain in the article where the highest wind speed of Matthew was located in a small area of the storm. So the methodology seems to be — keep looking till you find what you want to find.

    Such a small area was affected by this “highest wind speed” that you have to wonder how highest wind speed has come to define the potential destructive power of a hurricane? Has anyone ever shown that this metric, highest wind speed, in any way relates to the storm’s average intensity? Would not the “average wind speed” of a hurricane be a better metric for determining its potential destructive power?

    I think “highest wind speed” is a bit of useless data that tells very little about a hurricane. I think Phil Klotzback is just defending the “old ways of doing business” rather than questioning the old ways and looking for better. He needs to rethink his approach.

    And that is my unqualified opinion.

    Eugene WR Gallun

    • This would pose very challenging. You would have to decide a specific radius in which to average wind speeds. Some hurricanes may have very strong winds in the eyewall (like Hurricane Seymour in the NE Pacific recently), but these hurricane-force winds may only extend ~10-20 miles out from the center. Also in highly sheared situations, hurricanes may have strong winds on one side with much weaker winds on the other side. The average wind would represent a wind that was actually unlikely to occur in that region.

      • There is also a great variance in the actual diameter of a tropical cyclone even within a given category. Andrew was a very compact CAT V for example. With that in mind and making it clear that I am now sure how well pressures correlate to wind velocities I’m thinking your correct about pressure being the metric for classification but not just minimum pressure as measured at a point in the eye wall. Perhaps pressures taken at set of standard distances from the eye wall?

      • Dr. Phil Klotzbach — Thank you for your reply. Almost everybody just ignores me.

        But “highest wind speed” is likely to occur almost no place doing a hurricane.

        Giving your reply some very quick thought — Deciding on a specific radius in which to average wind speeds
        seems wrong for the reasons you stated.

        !) The destruction caused by a hurricane cannot be used post facto to measure its potential destructive level. When I was a kid when a hurricane was going to hit Florida this question was always asked — was it going to hit a city or come ashore somewhere on the undeveloped and unpopulated coastline? That question is no longer askable. For Katrina a levee broke in New Orleans and the flooding caused most of the damage, New Orleans being below sea level. Hurricane Sandy and the Hurricane of 1938 (the worser hurricane) had one thing in common. They both came straight out of the Atlantic and headed due west and drove straight inland. (Sandy, the weaker storm, just picked a terrible time and place to come ashore.) All other hurricanes in that area came up along the coast, half of the storm still over the water and not affecting the land.

        2) So maybe what we need is the size of the round area in which hurricane level winds blow.. This would be broken down into lesser circles of higher wind speeds as the center was approached. (There could be a still problem because I believe hurricane winds speeds are somewhat lopsided.) But hurricanes are not stable. They get bigger and stronger and they also shrink and get weaker. So that metric would constantly need remeasuring. And since it is a new metric it would take time to figure out what it is actually telling us, if anything.

        3) There has got to be a better method than “highest wind speed”. Hey, you’re the hurricane scientist so go do your thing and figure it out and maybe win yourself a Noble Prize. The pinko, socialist, commies of the left are just licking their chops desperately wanting to hand out a Noble for climate work.

        Well, now you know why almost everybody just ignores me. Did i mention to you that I am the world’s ONLY expert on how to translate Biblical Hebrew? They ignore me also.

        Eugene WR Gallun

        PS — I have spent the last three days passing, not one, not two but three gall stones. I claim the pain has driven me crazy.

      • Hi, Mr. Gallun,

        Hope you are feeling okay, today.

        Your neighbor to the distant north,

        Janice

        P.S. Being ignored is tough and I can empathize. To help yourself realize that it is the lot of 90% of the commenters (to be ignored most of the time), notice how many, MANY, interesting and or witty remarks go unresponded to day after day. And also, to help yourself realize that you are being READ, note how many comments you smile at or nod your head at, thinking, “That was a good comment (or the like)” but do not say anything. Sometimes, I think there should be more threads where conversation is encouraged. I have heard (C.S. Lewis believed this to be true) that men are much less “epistolary” than women on the average (he received FAR more letters from female readers), thus, they do not enjoy “correspondence” of ANY sort. There is a fairly high percentage of men, here. HANG IN THERE, Mr. Gallun — there are some who will write back (Marcus would be a fun pen pal for you — but, sadly, he is still on moderation).

        P.P.S. And some of us would write more in response to more people, but we are regularly castigated for it by the WUWT curmudgeons.

      • NHC provides much more information in their forecast advisories than just the max wind/minimum pressure. They provide radii of 34-knot, 50-knot and 64-knot winds in four quadrants (NE, NW, SW, SE). They also provide a wide variety of graphical tools of potential wind and surge impacts.

      • Janice wrote to Eugene: “To help yourself realize that it is the lot of 90% of the commenters (to be ignored most of the time), notice how many, MANY, interesting and or witty remarks go unresponded to day after day. And also, to help yourself realize that you are being READ,”

        That’s right, Eugene. You are being read. And there are SO many witty posts on this website. Many, many, times I feel like complimenting someone on something, or telling them how funny they are, but I refrain for one reason or another. I bet there are a lot of people like me.

        Lots of good comments on this website. I laugh out loud a LOT reading here. It’s a lot of fun. And educational.

      • I don’t want to pose as an expert, but I have had some education in statistics, and two things I learned apply here:
        (1) summarising data with just one number is always misleading. Real phenomena vary. Even a range is misleading, but at least it’s less misleading.
        (2) (estimates of) extremes are unreliable.

        In particular, while it’s true that “the average wind would represent a wind that was actually unlikely to occur in that region”, the maximum wind represents a wind that is ALMOST CERTAIN to fail to occur in most of the hurricane. It is, if you like, one of the *least* representative measures possible.

        I can imagine, but I hope the NHC have some real data we could be enlightened by, that the maximum wind speed of a hurricane might vary a lot over time even while its total energy remained fairly constant.

        Of course, if I’ve read the article correctly, while the maximum wind speed is hard to measure, measuring ANYTHING inside a hurricane is fraught with difficulty, so it’s not clear what could be done that’s better.

    • Safir designed his scale idea based on wind speed damage from an engineers perspective. Since then a team of folks at NHC have been updating this standard. This standard was designed in the 70’s and we have category estimates that discuss stoms in terms of categories back in before the standard also determined by damage. What I am not sure about is if Safir intended for a one minute average to be used. It would be interesting to see his original math.

      • David Riser
        No matter what the math says what is in question is the NOAA standard for classifying storms based on the Safir-Simpson scale. Here is the NOAA standard as stated here:
        http://www.nhc.noaa.gov/pdf/sshws.pdf

        ” The maximum sustained surface wind speed (peak 1-minute wind at the standard meteorological observation height of 10 m [33 ft] over unobstructed exposure) associated with the cyclone is the determining factor in the scale.”

        As we have been informed by multiple authorities now, it is most often impossible to obtain the wind speed data for this metric.

      • anemometers rarely fail due to wind speed per se. I have certainly watched my anemometer on the ship clock steady 100+ knot winds. The failure is almost always related to flying objects. In the case of buoys its spray or dunking that does it. NOAA buoys are built to hang tough in the roughest environment without dumping the package into the water. Failure for anemometers is of two types, catastrophic and age. age related they slow down relative to wind. but they have to be pretty old for that to happen.

    • Janice Moore —

      Thank you for your kind words of comfort. Have you ever heard the expression — “sympathy for the Devil”? Here is a poem i wrote a number of years ago.

      THE DEVOUT OF DELA

      As evening over Dela spreads
      We heed the Temple horns
      And bind the roses round our heads
      Still guarded by their thorns

      For in the holy songs we sing
      The prophets make it plain
      Whatever gifts the gods may bring
      They also bring us pain

      The gods create, the gods destroy
      We live beneath their sword
      The gods will do what brings them joy
      We pray to be ignored

      Here is another poem that deals in one part with a similar idea.

      THE BEST OF A BAD GENERATION
      or
      THE POETIC CAREER OF A DIRTY OLD MAN

      On stage butt-naked with his cock getting hard
      Berating old Moloch and burning the flag
      A closet pedophile — in public a fag
      Crazy like his mommy with her Commie card
      And then after forty years of avant-garde
      As con-artist, Zen-fakir and Beatnik scag
      The old humbug pulled a last trick from his bag
      And donned cap and gown — an academie bard?????
      Life, perhaps, does make hypocrites of us all
      Our best shot may be the shallowness of fame
      But when Time, the slow reader, speaks of his name
      The godhead of Ginsberg — will get very small
      He was one part a poet, nine parts a fake
      Vachel Lindsay in vitro — no William Blake

      Eugene WR Gallun

    • When I was working on https://wattsupwiththat.com/2016/06/05/summer-of-1816-in-new-hampshire-a-tale-of-two-freezes/ I was intrigued that the average temperature back then was typically the average of morning, afternoon, and evening temperatures, e.g. 6:00 AM, 1:00 PM, and 9:00 PM, with the evening temperature counted twice. Presumably that was because the evening temperature was more representative than the morning or afternoon temperatures.

      Once Max/Min thermometers were invented, I presume that helped encourage our continuing focus on extremes. For future metrics, it may be worth looking at less extreme measurements, but for things with a long historical record we pretty much need to stick with what we do now.

  12. Dr. Klotzbach is absolutely correct in his assessment, and we can be assured that he is no warmist “adjusting” data for nefarious purposes. So modern methodologies use a variety of airborne, satellite, radar, and surface observations to define a more precise peak wind within a hurricane. My concern is that modern measurements of a hurricane categories at landfall are being compared to hurricanes a century ago whose landfall strength would be underestimated by a category or two due to much less complete data (no satellites or aircraft, less coastal development to indicate the peak damage, and only scattered anemometers that often failed). Which leaves us with recent hurricanes being “accurately” categorizd, being compared climatologically with older hurricanes that are underestimate.
    Unwittingly, this is statistically comparable to lowering early temperatures relative to recent ones to up the trend. Unless, of course, the assessments of earlier hurricanes somehow take the sparse data into account – perhaps by informed interpretation of pressures?
    BTW, I’d like to thank Dr. Klotzbach for carrying on the good work of our late hero, Dr. Bill Gray.

    • This.
      Plus as I wrote up post, the current Grading/Rating System is based on those old ground based measurements.
      To be consistent we should have 2 grading systems one continuing the use of Ground based Instruments and another for Satellite “ESTIMATED” values.
      Or as mentioned by Mr Keen are you going to go back and increase all the old Values to get a proper single scale

      • As I mentioned before, in all other global tropical cyclone basins (except for the NW Pacific from the late 1940s to 1987), satellite is the only tool used to estimate hurricane intensity. The majority of North Atlantic hurricanes and tropical cyclones also have their intensity estimated from satellite imagery. It’s only when they threaten land (or encounter a ship) that these other techniques are employed. The reason that the Hurricane Hunter planes are employed is to measure winds both at flight level (10000 feet), but also to release dropsondes and utilize the stepped-frequency microwave radiometer which estimates surface winds based on the state of the sea. The utilization of the sea state for estimating surface winds is a practice that has been long-utilized by the National Hurricane Center (since the 1950s).

  13. For The Sake Of Argument:
    I accept Dr. Klotzbach’s arguments at face value. There are better ways of measuring wind speed than previously existed.
    Now:
    We have a long history of hurricane data upon which the intensity scale, damage assessments (both before and after), historical patterns, number of storms, numbers of intense storms, etc., etc. are based. This data is sustained ground speed and is formally defined as as such. Note that whatever under-registering of wind speed is happening now has always happened in the record, and the record is based on it. Fair enough.
    Now we introduce a new, improved method which produces results which are systematically higher. How much higher? Make up a number, perhaps 20%, perhaps 30% or more. In other words, a lot. Enough to be at odds with the whole of the historical record. Any scientist would recognize that you have introduced a systematic bias into the record. Without an attempt to create a cross calibration between the two measurement methods, this is about as bad an error as you can introduce into a data collection system.

    At Best:
    Use the traditional measurement for the traditional uses, particularly for likely damage. Use the new measurements for any new purposes.

    Going Forward:

    a device called a stepped-frequency microwave radiometer which measures radiation transmitted from the ocean surface

    Sounds like a surface measurement to me. Take these readings where there are deep water buoys, shallow water buoys, and close by land stations. Compare data sets.
    In other words:
    Calibrate, Calibrate, Calibrate
    Having two measurement systems which routinely differ by one full category on the S-S scale, and frequently differ by 2 full categories, is just not acceptable.
    By relying on the new, higher values without any attempt to calibrate to historical data on the S-S scale, you trash the value of the historical hurricane record.
    This is why we work so hard to eliminate systematic errors in long running data sets which so commonly occur when there is a change in instrumentation or methodology.

  14. There must be some reason why the winds, once they reach land, are only 50% of that projected from the satellite measurements.

    • A most recent example being Hermine? I still have serious doubts it was a hurricane when it made land fall.

      • hermine went right over top of me.
        it rained.
        some spanish moss blew off the oaks.
        the weather gangsters shifted to a tale of how lucky we were and skulked off to look for another tyfun of doom so they could party on paranoia again asap.

      • hermine went right over top of me.
        it rained.
        some spanish moss blew off the oaks.
        the doom mongers got precious about how it might would could may should next time.

    • NHC has conducted a reanalysis of Atlantic hurricanes going back as far as 1851. Obviously there are large underestimates, especially as one goes back in time. There were also some over-estimates of winds in the late 1950s-early 1960s, when the MSLP/max wind relationship they were using was poorly chosen. Gabe Vecchi at GFDL has done some good work looking at potential underestimates in named storms and hurricanes prior to the satellite era using ship track measurements. When you include his estimates of missed hurricanes with the observed hurricanes, you see no trend in Atlantic hurricane activity since the late 19th century.

  15. We can only estimate the winds in a hurricane. There is no way to consistently measure the peak winds in a hurricane, and even if we could, the peak wind probably fluctuates fairly rapidly in such dynamic storms.

    There is another way to estimate the winds in a storm: by surveying the damage. If I remember correctly, that is how Hurricane Andrew in 1992 was raised from an estimated Category 4 storm at the time it hit southern Florida to a Category 5 storm after post analysis.

    Since the hurricane winds are estimated, that means there is larger uncertainty than if we had direct measurements. My guess is that the uncertainty is somewhere around 10 to 20 percent most of the time, and sometimes larger, depending on the information available for making the estimates. Uncertainty of the peak wind estimates for historical hurricanes before the satellite era is probably much larger.

    For hurricanes striking land, the peak winds are most likely to occur at the immediate coast where wind is blowing onshore. Even if there is a weather station right on the coast, like NOAA’s CMAN stations, if that measurement is at 10 meters above the surface in normal conditions, hurricane high tides may decrease the measurement height above the surface and wind driven waves may create surface roughness that decreases the winds substantially from what would be measured at 10 meters above the wave crests. As the wind moves inland from the coast, surface roughness increases rapidly in most populated areas, because of buildings and trees, and thus winds at 10 meters are almost always much lower even a few miles inland than on the immediate coast. There are always exceptions, and one I remember is Hurricane Celia which hit the Corpus Christi area in 1970. In that storm the Corpus Christi National Weather Service Office, about 22 miles inland from the Gulf of Mexico, measured a peak sustained “fastest mile” wind of 125 mph and peak gust to 161 mph according to the annual hurricane summary for 1970 reported in Monthly Weather Review. In those days the “fastest mile” wind was estimated from a chart recording of the anemometer output.

    The effect of surface roughness is to greatly increase turbulence, which drags down wind speed averages much more than peak gusts. However, it is the gusts that cause most of the wind damage and probably provide the best indication of what damage might occur. The exposure of a 10 meter anemometer also plays a roll. Poorly exposed sites with nearby buildings and trees usually measure lower winds than well exposed sites in wide open spaces, like airport runways. Nearby buildings can sometimes cause very localized higher winds because of the Venturi effect. I remember an extreme wind gust measurement of 236 mph made in Guam during Typhoon Paka in 1997 was suspected of being enhanced by the position of the instrument relative to nearby buildings, and thus not representative of the larger area around the measurement location. Many anemometers are blown away or damaged before peak winds occur in a hurricane landfall and often power outages precede the peak winds so that wind measurements are lost from lack of backup power. So it’s tough to get good wind measurements during a hurricane landfall, even in a populated area.

    • At least Andrew had the decency to rip the anemometer off the NHC with a final reading of 200mph.

    • Well said.
      Andrew damage was beyond anything in my 60 years. I went there a month after the storm, and saw what happened with my own eyes. In the damage path, almost everything above the surface was scraped clean. All trees were gone, some palm stumps. Mountains of debris were piled at every street corner. Over a 10 mile path, I saw one structure made of steel reinforced concrete that survived, it looked like someone had built a NASA blockhouse. Heavy steel girders and beams, such as those holding metal roofs over service stations and commercial buildings were twisted and mangled. Light steel structures were gone. All wood structures were gone. Temporary street lights were rigged, some of the heavy steel poles survived. And the structures in that area were mostly new enough to have been constructed under rules that are supposed to anticipate a direct hurricane strike.
      It was a very grim scene, because it reminded me of photos of Hiroshima after the bomb.
      It does not matter what talking heads say on TV, or politicians blather, you need to stay far away from any strong hurricane.

      • bw, it’s a miracle there was not a greater loss of life with Andrew. If the eyewall had hit the center of Miami instead of well to the south, damage and deaths would have been much higher. Thankfully the highest winds in the eyewalls of the most intense hurricanes usually impact an area not more than about 20 to 50 miles across and most commonly hit areas that are not as heavily populated as city centers. But like many natural disasters looming in the future, sooner or later a Category 5 hurricane will hit the center of cities like Miami, Tampa, New Orleans, and Houston. It’s just a matter of time. We don’t have a long enough record to estimate the frequency with much confidence, but my guesstimate would be about once every 200 to 500 years for such an event somewhere in the southeastern US. As coastal cities grow in population, the average recurrence interval will decrease just because there will be more and larger targets.

    • Please be careful about advocating a damage-based measurement system, First, for any given storm, it has no predictive value, as the damage has not yet occurred. But more importantly, damage will vary, for a given storm intensity, based upon the standards to which a building was built, how the area was affected by any storm surge and for such things as stop signs and trees, how well they were installed (the stop sign) and how saturated the ground was at the time (for trees). Damage assessment can be rather subjective.

      Some of you may remember the great hurricane of 1987 that hit England – it uprooted thousands (maybe tens of thousands) of old-growth trees. I don’t think it did much other damage – can anyone comment on that? Does anyone have any assessment of the strength of that storm?

      • The Saffir-Simpson scale has been consistently confirmed for decades. Wind damage is proportional to wind speed.

      • Jim, I certainly don’t advocate a damage only assessment of hurricane wind strength at landfall. It’s just one more piece of the puzzle and occasionally a decisive piece as in the case of Andrew. It’s also all there is for estimating hurricane wind strength back before the days of wind instruments.

      • I was living in Broadstairs, Kent at the time and experienced the Great Storm of 1987. Though easily of hurricane force, technically it was not actually a hurricane but a very deep mid-latitudes depression. The anemometers at Dover Docks were rated to 110 knots, and were constantly off-scale at the height of the storm. Had it not struck in the small hours, hundreds of people would have died.

  16. In the example above where an anemometer is hit by the eye wall, why would it measure low? Would we also conclude that all anemometers in every situation read low? If not, why just in huricanes?

    I’m not trying to poke a hole in the explanation, I’m just trying to understand.

    • An anemometer is a physical device, and it may read low during high winds for many reasons:

      – it may break if wind speeds exceed some threshold
      – it may be unable to report wind speeds above some threshold (e.g., limitations on electronic signalling)
      – it may be inaccurate for wind speeds outside a specific range (e.g., too much friction for turning paddles)

      • Anemometers are pretty sound. if they get the wind they will be accurate. Some silly model has no way to prove that t wind anemometer is not accurate. That being said the sources of error for a anemometer relates to a few things. bearing failure caused by age and wind not hitting the thing square. When you realize that wind is in three dimensions if it has a bit of updraft to it or downdraft your still just reading the forward part of the vector. Buoys have a vain that keeps them roughly into the wind but a confused sea can spin them a bit and they can get some blockage sometimes. overall though they are accurate up to failure unless they are already failing but you can usually spot this problem since at low wind speeds they wont spin at all. NOAA (NDBC) buoys manage this through paired sets of Anemometers which reduces error significantly.

  17. The relevance of a hurricane varies with intensity, dwell time and location.

    “If the eyewall of Hurricane Matthew had been 50 miles further west”

    If I had some ham I could make a ham & cheese sandwich if I had some cheese.

    Any discussion about hurricanes and their impact must consider the practical effect on populations. There are hurricanes that manifest off-shore that never hit land fall… and it is as if they never happened.

    Matt Drudge recognized that the democrat party activists were aching for a devastating climate related event. Many people were hyping it, and Hillary was planning a trip to the projected landfall location.

  18. > The adjustment factor would be approximately 0.90, which means if a buoy measures 80-mph eight-minute average winds, the one-minute average wind would be about 89 mph.

    Wouldn’t an adjustment factor of 0.90 with a measurement of 80 mph be 87 mph? (0.9 * 80 = 7.2)

    Then Phil also wrote:

    > The World Meteorological Organization recommends a 0.88 correction for a 10-minute to 1-minute average, so for 8-minutes, we approximated 0.90.

    If you just rounded it to 0.9, that’s fine, but if you “made it a little bigger” because of the 10-8 minute difference, the adjustment-to-the-adjustment-factor went the wrong way. An 8-minute average should be closer (smaller adjustment) to a 1-minute average than adjusting a 10-minute to 1-minute.

  19. When did satellite-measured wind speeds become part of the standard, and how did the agency ensure continuity with pre-satellite measurements?

    That is, once satellite measurement is added to the mix, the greater precision appears to bias wind speed measurements up by removing some of the negative bias (AMSU measurements are outliers significantly above the line from 10/6-10/8), so if a storm would have previously been measured at wind speed X, would it now be reported as wind speed X + something, because of these more precise measurements?

  20. I doubt that Matt Drudge is arguing that Hurricane Matthew was not harmful. It clearly was. I think his point is that National Hurricane Center has modified their measurement systems in such a way as they report storms at a higher wind speed than they did several decades ago.

    The National Hurricane Center is constantly tweaking their methods. Their intent is to ensure more reliability. However, the result could be a bias in hurricane levels which would lead scientists to draw inappropriate conclusions from this biased data. The simple, obvious example is “Hurricanes are getting worse — therefore climate change is a contributor.

    Personally, I think Drudge is right. Air pressure at the eye is not a perfect determinate of hurricane strength, but it is pretty reliable. Of the 31 category five Atlantic storms which have occurred since the beginning of record keeping, only three have had pressure higher than 934 millibars — Edith in 1971 (943 millibars); New England Hurricane of 1938 (940 millibars) and Matthew (934 millibars.)

    Historically, those numbers are closer to the pressure in category 4 storms (these typically run 930-950 millibars). Category 4 storms can still cause plenty of damage — the only ones who really care about such differences are those trying to calculate ocean energy content for a research paper.

    • For those interested the wind map for the east coast is on page 37, other areas are on other pages,

  21. The problem is this… Now you have hundreds of thousands of people from Cape Canaveral to Mayport who think they rode out a “Major” hurricane within a few miles of the beach. My own parents carried out one of the most stupid evacuations, from their house on the beach to a motel ONE MILE from the beach, and they think that was just great. A real “major” hurricane would have destroyed that motel.

  22. What happens when alarmists keep crying wolf. Just like claim of 250mph winds off the Northwest coast about 2 weeks ago. Total B.S.

  23. Dr. Philip Klotzbach,
    As an Engineer, I am disappointed in your article, since I know for a fact that structures are not designed for the maximum velocities recently reported by NOAA. There are groups of engineers that look at the wind data and have specified the design wind velocity for every portion of the US and the structure design wind velocities are less than the NOAA reported 165 MPH velocities. The maximum design velocity is 150 mph in the Gulf which reduces to 120 mph in the NY Harbor region. See document below.
    https://law.resource.org/pub/us/cfr/ibr/003/asce.7.2002.pdf These are 3 second gusts at 33 feet above ground level and the design formulas clearly adjust the load for the elevation,

    I suspect the maximum velocities reported by NOAA are not the velocities at ground level and are likely misleading or exaggerated for ground level. Perhaps you can please clarify and explain why there is a difference.
    As an engineer I have designed and checked commercial tall and ground structures for wind loads and nothing like 165 mph is used for the east coast. For example the design wind velocity for structures close to NY harbor are designed for about 120 mph as I recall. Keeping in mind, that the force on the structure is proportional to the wind velocity squared, the load difference is huge if the NOAA reported loads are used rather than those in the referenced document, and there would be numerous failures if those loads were actually experienced.

    I do recall that several decades age the design wind velocities were increased after one of the Hurricanes in the Gulf, and some commercial structures were beefed up to handle the higher loads although I doubt that all commercial structures were upgraded. I recognize that many older homes especially the roof is not per the latest codes.
    Also I was disappointed at the casual dismissal of ground station data because I suspect many are designed to handle the loads and many gave reports without failures.
    Otherwise thanks for your article.

    • As I mentioned before, anemometers usually fail in high winds, either due to the winds themselves or due to the storm surge. The article describes why, in Hurricane Katrina, the maximum sustained surface wind reported was 86 mph (79 knots). You can go through the report here and see:

      http://www.nhc.noaa.gov/data/tcr/AL122005_Katrina.pdf

      Similarly in the case of Hurricane Camille (1969), no weather stations near the coast survived to report max winds. Colombia, MS (about 80 miles inland) reported max winds of ~115 mph:

      http://journals.ametsoc.org/doi/full/10.1175/BAMS-D-14-00137.1

      I spent a considerable amount of time doing post-storm reports for some wind engineers after Hurricane Ike (2008), and in that case, virtually all of the NWS stations close to the coast failed during the event. I ended up pulling Weather Underground stations to find higher max winds, but of course, the siting of Weather Underground stations could be suspect.

      In the case of Matthew, I wouldn’t expect to see Category 3/4 winds on land anyway, as the eyewall did not come ashore until Matthew was weaker. The best example of what we would have seen was an 107 mph wind gust reported at an elevated platform in Port Canaveral. The western part of the eyewall was within about ~10 miles of Port Canaveral as it scraped by the coast.

      If you want to see more what ground-based observations look like in recent hurricanes (e.g., Ivan, Wilma, etc.), I encourage you to check out the NHC post-storm reports:

      http://www.nhc.noaa.gov/data/tcr/

      • A gust is not a hurricane. Plenty of photos of Mathew damage show tropical storm force winds. The surface wind speeds recorded by surface anemometers are consistent over the entire path of the storm at below hurricane category 1. NDBC buoys recorded winds are consistently confirmed by nearby land based anemometers. Not one surface station shows any recorded sustained winds reaching category 1. You have no surface data anywhere from Florida to N. Carolina that show hurricane winds.

      • Sounds like instead of spending billions against “climate change” we should spend a few million beefing up NWS anemometers to survive 150mph winds. This has to include battery power and radio transmission. I was trying to follow weather stations during Matthew’s sidle along Florida, and was disappointed that several of them stopped reporting. Of course, it could be that certain authorities realized they weren’t going to report anything dramatic, so took them offline to avoid the emabarrassment. That’s a cynical view I know, but so much bad stuff has gone down in the CC arena that it’s hard not to be. I’ll be interested in reading a post-Matthew report on what happened to those stations.

        Rich.

      • Phil,
        Mathew did not In fact destroy any anemometers. buoys frequently survive cat one and two storms. Not so much major storms which tends to lend weight to the overhyping of the storm. one huge wind gust in one corner of the Hurricane does not a major storm make!

  24. My observations of wind speed came from weather stations on Wundermap and well as https://earth.nullschool.net.
    I make no claims regarding their accuracy.

    It appears from those sources’ indications that media reported max wind speeds that were occurring at and above 850 hPa.. whatever altitude that would be in a hurricane. Two or three thousand feet?

    At the surface, or on the ground… wind indications were much much lower in this and in previous featured hurricanes with those two sources. (Nullschool.net also reports wave heights.) It’s likely disagreement could happen if people are comparing observations devoid of altitude info.

  25. BW that as it may, sir, but the results speak for themselves: consistently over blown and inaccurate cries of disaster and doom. Stop beating up on those who point out that reality and start doing something about those who make you look less than credible: the ignorant media hacks, rent seeking activists, and opportunistic politicians. They are the ones doing you and those who claim to be responsible the damage.

  26. If the money spent on “renewables” and the “research” to prove that Human activity changes global climate had been spent on developing hurricane-resistant buildings and civil infrastructures, we might actually have benefited the economy and the coastal populations simultaneously.

  27. Hitlery would have been just as correct to state that the hurricane was made worse by “deplorables”.

  28. Oops, I triggered the spam disposal. meant to say
    Hitlary would have been just as correct to state that the hurricane was made worse by “deplorables”.

  29. The evidence is strong that coastal structures will be destroyed by storms long before being inundated by climatic sea level rise. We are on the wrong track, progressive dolts!

  30. I know you were well meaning with your article but my reading of it further cemented a part of Drudge’s point: we really have little idea what the winds at ground level will be when a hurricane comes ashore. Our estimates are bad and purposely intended to measure high.

    The other piece of Drudge’s point…intent…that one I feel much better about after reading the article. I always try to avoid attributing something to malice when “incompetence” will do. “incompetence” in quotes here because it’s clear how difficult managing the complexities of measuring hurricane wind speed can be.

    If you want to improve how you give data to the public report the expected ground level wind speeds with a confidence number. Every image of the hurricane published should include an overlaid map of these speed estimates and confidences as the storm moves along its path. This way we could see how small that 120MPH field is expected to be and how large that 50MPH field is.

    • This information is already available from NHC but only occasionally displayed from various news outlets. Regarding your other comment, I disagree. Devices such as stepped-frequency microwave radiometers are carefully calibrated to estimate surface winds. In addition, when storms get close to land, we also utilize radar imagery to estimate the surface wind speed. While admittedly these aren’t measured specifically at the surface, there are known adjustments that can be made to estimate surface winds. Of course, unless you are measuring the wind directly along the coastline or in an open field, you will likely get terrain sheltering effects and weaker winds.

      Also, realize that the metric that is predicted by the National Hurricane Center in terms of the maximum wind has been (and probably will always be), the maximum one-minute sustained wind at any point in a hurricane. This is a very elusive quantity to measure specifically with a weather station, which is why these other methods are also employed.

  31. “The National Hurricane Center provides the best analysis that science can offer.”

    Poor babies. They are finally admitting that it is not an easy task to measure wind speeds, let alone forecast the actual path a storm takes. After this entire article telling us how hard it is to get an accurate measurement of the wind speed after criticizing Drudge for throwing the BS flag.

    The National Weather Service, especially the National Hurricane Center, tries to do its job with the best tools it has. Unfortunately, the Television Weather Boffins are trying to look good to the home folks, and make a name for themselves, over-selling the crap out of hurricane models like they are some sort of religious experience.

    Science has its problems. It is becoming more and more apparent everyday that there is nothing special about scientists. They grub for money just like everyone else, and as a group, climate scientists are not adverse to fudging their data and faking the results, just like described in the Climate Gate emails.

  32. Most of us have heard of or even seen some spectacular weather events, but this is a new one to me:


    Video courtesy James Loveridge Photography (also on the BBC website)

    Unless you know otherwise lets call it : Dorset Fog-fall

    • I have woken up one fine morning and seen something similar in the Alps on a smaller scale. Just gorgeous! Imagine looking down to one side in a valley and seeing a mist flowing through and down a cliff like a slow motion water fall as it comes through a saddle like a river of mist between two peaks. . Then you look down the slope of the mountain your on and see a group of Red Deer peacefully grazing on high pasture. Then you look up slope on the mountain your on and there are smaller Roebuck working their way up through the rock. No camera can catch and record a thing like that.

  33. The question of whether we’re still in a “major hurricane drought” is still important as a talking point to adress those who exaggerate. I understand that helping keep people safe is the primary function of the NHC, and I understand the reason for your complicated processes as you have described them.
    But there is still a difference between the East Coast being “brushed” by a category 4 hurricane, as mentioned in the article, and a category 4 hurricane brushing *by* the East Coast. Based on the analysis of the NHC, are we still in Major hurricane drought, or not?

    • Correct, we’re still technically in a major hurricane drought, but obviously there are some technicality issues given that Ike (2008) may have been classified as a major hurricane in the earlier part of the 20th century when damage was often utilized to assess storm intensity. Ike was a high-end category 2 hurricane (95 knots) that had a massive surge associated with it given its large size. Also, the drought only continued this year given a very small wobble prior to a potential landfall with Hurricane Matthew.

      I recently posted a figure though, relaxing the metric to Category 2-5 hurricanes, and showed that we’re currently at a 11-year running average low dating back ~160 years:

      • Ike was quite memorable. The damage ratings approach is interesting and full of promise, but the climate hype community rejected that because it requires adjusting for constant dollar values and increases in coastal development, and that once again shows a flat line storm change. Ask the Pielke’s who were hounded out of weather related research for pointing out that inconvenient truth.

      • “Major Hurricane Drought” I think of it more as a hiatus. A pause. Who could have predicted this great relief from these deadly storms.. Everyone should be thankful we in the lower 48 have been so fortunate. Even the Bozos that have claimed the opposite would happen or will happen. But we all know by their actions, deeds, and words, that there are some who are disappointed. Some who have minimized the rarity of such a pause. Some who almost seem to salivate every time a Major Atlantic hurricane forms that may come ashore here in the US. Now with the SOI indicating a major seasonal shift in the weather pattern coming old man winter looms large come the middle of November and thus the probability of a hit this year declines significantly. They will almost certainly have to wait for next year to await the storm they so much want to hold up as an example of what climate change can do and thus see the end of the “drought”. And if history and long range forecasting is any guide they are very likely to get it.

      • Phil, in your comment about “if Matthew had been further west” you are still clinging on to your belief that it would have been at least Category 3, when all evidence from damage and recorded speeds in the Bahamas implies that it would have been in the TS to Category 2 range, so not a major hurricane.

        Rich.

  34. There was Cat 5 cyclone in Australia 6 years ago where the highest wind gust recorded before the stations were damaged were about 110 mph or 175 kph. Somehow anemometers from the early 20thC survived 220kph gusts. The experts guesstimated that it was Cat 5 at landfall despite there not being the damage even consistent with a Cat4. Its hard to trust experts when things like that happen.

    Taking away any mention of the F word, there is still the issue that these cyclones/hurricanes are compared with others that only have what was probably a highest recorded windspeed significantly less than could have been measured in modern times. Not chance that cyclone Yasi would have been called a Cat 5 if it happened in the early 20thC but it is compared with earlier cyclones as if there is a trend of more severe storms due to climate change.

    • And the real worry is that people who came through a cyclone such as Marcia think that they survived a Cat 5 cyclone so they’ll be right next time- when the winds and damage e.g. here in Rockhampton were nothing like Cat 5. There’s a similar problem with Severe Storm Warnings. My suggestion is for well publicised estimates of cyclone or storm strength afterwards, when engineers can make estimates from building and vegetation damage.

  35. Quote: There are several reasons buoys are likely to underestimate a hurricane’s wind.

    As people have said above this is the key to the issue. Once upon a time buoys, anemometers and actual damage were the only data available. From that data an estimate of x mph was made. Newer and better data and calculations would have calculated an estimate of y mph.

    If y > x then there is no doubt. Guilty as charged.

    The new methods really do give higher results than the old. It is just the same when the people who fudge the temperature record do their thing and insist that the new way of calculating things is better than the old.

    The proper approach is not to shoot the drudge messenger but to explain why y > x. Then of course there is the question of communicating the results in a way that is useful to people in harms way.

  36. The hype and continuously “adjusted” data, both current and historical make a
    non metorologist like me distrustful of anything produced by “professionals.”

    My favorite quote from the Weather Channel is “this wave is not developing
    the way you would like to see it”.

    From my non professional point of view, why not use ground station barometric
    pressure to compare storms?

    • Pressure was used frequently when comparing Matthew with historical Florida storms. The lowest pressure in the eye is a much easier to measure quantity in a hurricane than is the maximum sustained wind speed at a single point.

      • And seems much more reasonable and objective- if your community will educate the media and public as to what it means. The over hype of storms is ultimately going to lead to “boy crying wolf” syndrome, where people will ignore all warnings, even the one where there really is a wolf.

    • I was working my way down to the bottom to say exactly that.

      Air pressure is essentially the same across the eye, that’s much larger area than some narrow strip in the eyewall buffeted vortexes created in the eyewall.

      I’m going to start talking up central air pressure in the future. Lessee, isn’t that a leading indicator anyway? I.e. a fall in air pressure precedes an increase in wind speed.

  37. Phil, With the limitations you mention, it seems to me unlikely to achieve a measurement that satisfies the formal definition for sustained hurricane wind speed, and which happens to be measured at the point of highest wind speed. The best that can be done is to say the highest measured formally, and estimate the max from the other, non-formal measurements, using the alternative techniques.

    That said, it seems to me that the winds occur in association with other physical facts that can be measured more reliably, that do not break the instruments or mis-measure if shaded by structures. Can we calculate the energy of the hurricane from eye latitude, eye air pressure, eyewall diameter and eyewall direction?

    Thanks for your excellent article.

  38. I guess I, like probably many others that have started learning this stuff relatively recently, have the feeling of having been misled for years. For years the NHC grading of storms has been presented in the general media as an absolute based on simply reading hard data. It is still presented that way today. In my ignorance I assumed that was the case only to learn over time that in fact their categorization is anything but absolute. The older I get, the more I learn, the more skeptical, I become. The trick is not becoming too cynical at the same time as one delves ever deeper into it.

  39. I was working with Prof. Rozencrantz and Prof. Staelin when Phil did his stuff on wind speed estimation from microwave sounders. It is better than nothing, but it really relies on a lot of highly non-linear things in the variation of sea surface reflectivity in the microwave spectrum with wind speed. It is not as accurate as a dropsonde but it is a valid estimation technique.

  40. So wind speed is measured, calculated, and estimated. How do you decide which to report for use by public?

    • Just to clarify, I don’t work for the National Hurricane Center, so I don’t make the final decision on what is reported to the public. When a storm is approaching land, it’s the best estimate of the National Hurricane Center forecaster on duty using a combination of satellite, aircraft, radar and in situ data.

  41. Dr. K, your discussion of wind speed measurement challenges and limitations is fascinating. However you apparently missed the point millions and millions of Americans, including Drudge, notice: The over hype and catastrophism that dominates and distorts weather reporting. Most on this forum, and most Americans, get it: it is tough to measure weather systems. That has nothing to do with the shallow, reactionary, panic mongering *based on your work* by media and political leaders that you and your community have so far declined to forcefully address. Matthew was no exceptional in track, strength, impact, risk, etc . Where is your moralistic Op-Ed condemning the hype and anti-science idiocy of those who falsely claim that this minor storm (in terms of impact) was unique, unusual, caused by “climate change”, etc. That would be a bold move. Blaming Drudge is low hanging fruit in a forum like WaPo. How about holding classes for media and politicals so they can learn how to understand your work product- its features, benefits, short comings, etc.?
    Respectfully,
    hunter

    • A couple of points here… Matthew was an incredibly devastating storm for portions of the Caribbean as well as portions of the eastern United States. With most hurricanes, the real killer isn’t the winds… it is the water. 45 people died in the United States from a storm that didn’t make landfall until it was a weakening Category 1.

      NHC doesn’t engage in fear-mongering tactics… had the storm come ashore in Florida, its results would have been devastating. In addition, I would hardly claim that Hurricane Matthew had minor impacts in the southeast United States. Insured damage estimates currently run from $3-$7 billion dollars. I would argue had Matthew tracked 50 miles further to the west, the insured damage estimates could have been 5 to 10 times as high. There is a large difference between a hurricane’s outer bands hitting a 100-200 mile stretch of coastline than if the western eyewall had actually come ashore.

      If the hurricane had come ashore and associated sustained winds in the 80-110 mph ballpark ensued, I expect Drudge would be singing a different tune.

      • (1) Given wind speed and pressure differentials, the value apportioned to damage is a consequence of development density and the sophistication of technologies that are damaged and destroyed, refer Pielke Jr for a fuller explanation.
        (2) Argue all that you like, but this storm did not track 50 miles further west and as such your argument is merely supposition unless you are an expert in construction technologies and cost … as am I.
        (3) Drudge then, is not singing a different tune.

      • Phil, at the time I was really pleased that Matthew missed Florida. But now I’m regretting it, as I really would like to have known how bad it could turn out. And I don’t buy your estimate of 80-110 mph.

        To justify my rude contradiction of you, please consider your own graph of Buoy 42058. Now, I agree that sustained winds are suspect, so let’s ignore that. But the maximum gusts were 80 knots, and the symmetry either side of the eye passing shows the buoy got a good look at two edges of the eyewall. 80 knots is 92mph. From that gust, what would you estimate the sustained speed to be? Presumably not 110mph :-).

        Do you think that Matthew would have been stronger at Florida landfall than it was when it passed over Buoy 42058? I will accept a “yes”, but only if accompanied by credible data.

        Thanks in anticipation,
        Rich.

    • “IF” something had happened other than the way it did, then you would look better.

      But it didn’t. So right now, I’d say you don’t look very good.

      Andrew

  42. ‘I’m certainly not a fan of over-hyping climate change’

    You certainly wouldn’t want to over-hype hype.

  43. As with any story, hurricane or otherwise, always utilize at least two sources. I use the NHC and earth.nullschool.net. The NHC/Media never wants to get caught understating a weather event, so they end up overstating. Consistently overstating ones case, puts one in the same dilemma as the boy who cried wolf.

    • NHC’s estimates of max wind and minimum pressure are their best estimates. They aren’t overstating the case in their advisories. In the case of Matthew, you had an incredibly difficult forecast of a major hurricane tracking parallel to a coastline. Had the storm tracked 50 miles further east, it could have been a virtual non-event for Florida. Of course, if the storm had tracked 50 miles further west, the results would have been absolute devastation.

    • Nullschool is not a data source. If I understand correctly, they display the results of the GFS models, not physical data.

      They do make great eye candy and do a nice job of showing motion of large air masses.

      • I am not a weather scientist. Neither am I blessed with an over abundance of I.Q., but I do know what i trust. Earth.nullschool.net does have beautiful images. Nullschool does let me zoom into large weather events, like Hurricane Matthew, and check average wind speeds. It allows me to see how the hurricane wind speed vary from the leading edge to the trailing edge. The color coding is just stunning.

        It seems that we have a version of the Wizard of Oz playing out before us. Pay no attention to the man behind the curtain. I do not want to lose faith in the NHS and the media, but when I suspect they are not being straight with me, I cannot help it.

  44. NOAA has proven it’s willingness to alter data to better match global warming theory. Moreover, NOAA actively supports the theory and the associated government grab for more power and money. Thus, those who understand science have recognized NOAA’s willingness to put political policy positions ahead of science. The NHC is part of NOAA and now suffers the results of being associated with an untrustworthy institution.

    Think of this way, if you knew someone was a con man, would you trust their best friend’s investment advice?

    Sorry NHC, but it’s perfectly reasonable to assume any reports you issue are questionable.

  45. A good excuse + a bad result is not equal to a good result.

    Unlike meteorology, in my business the numbers actually have to add up.

    There was a signiicant delta between what we heard on the news and what was actually measured, which is a valid reson for complaint for any number of reasons.

    And this: “But the hurricane’s deadly encounter with Hati and multibillion-dollar brush with the East Coast wasn’t enough for some people” does not lend credence to your article, nor does it increase my respect for your opinion. Here in New Jersey we were evacuating the shore, and didn’t even receive the equivalent of a good thunderstorm. Many businesses suffered as a result, as the tourists that make up their customer base stayed home.

    This may come as a surprise to NHC, but accurracy is important.

    • NHC never posted tropical storm watches/warnings for the East Coast of the United States north of North Carolina. Model forecast skill has improved significantly over the past 50 years, and NHC’s forecasts have improved as well:

    • While the early tracks showed Matthew coming up the coast, I don’t think the timing was short enough to warrant NJ evacuation orders. Wasn’t some of the out-to-sea effects from weather systems entering the US west coast while Matthew was in Haiti or the Bahamas?

  46. I live on a boat in Florida. The problem with over stating the hurricane wind speeds is that the general public does not appreciate that the pressure from the wind goes up exponentially with the wind speed. Once someone rides out a published 2,3 or 4 hurricane they think they can do it again.

    We are on part of the coast that got hit by Mathew. Afterwards everyone that did not evacuate was saying how easy it was. My wind instrument recorded the highest gust at 59kts. That is a long way from what was stated for the storm.

    Also, the problem with surface winds is that a storm that hits the coast will often see the bottom of the storm slow down and the top will tend to slide off. The upper level winds will become the ground winds very quickly with some power provided by the sheer. This can be bad!

    As for the hurricane hunters – not sure what the sampling interval is for their data (just don’t remember). It appears from the raw data that a single hit will set the wind speed. It could be that their sampling interval is set long enough where one hit equals their “sustained” wind speed period. They should have to record several data points to rate sustained winds.

    Having a very large vested interest in getting the hurricane predictions correct, I track the hurricanes almost all summer (Mike’s Spaghetti Weather page is probably the best amalgamation of the available data). There have been what I call ‘pretend’ hurricanes predicted. These are storms that the data does not seem to support the sustained winds that are advertised.

    This over inflating of the hurricane wind speeds gives boaters and homeowners a false sense of security. More of them are staying put when a hurricane is forecast because ‘the last cat X storm was not that bad.’ This over-forecasting is as dangerous (perhaps more) than under-forecasting. Either way is bad!

    The goal of any forecaster should be to provide reliable and understandable forecasts. Right now that does not seem to be happening.

    • The eyewall of Matthew did not come onshore, so I wouldn’t expect that winds would be that strong along the coast. As I’ve stated multiple times before, when major hurricanes do come onshore, anemometers very rarely record the strongest winds. They either fail or are destroyed by storm surge. The strongest sustained wind that I could find any U.S. hurricane was a sustained wind of 126 mph in Hurricane Andrew, which equates to a Category 3. In storms like Camille and Katrina, storm surge wiped out pretty much everything standing along the coastline.

      In the case of Matthew, virtually all stations in the Bahamas stopped reporting at some point during the storm. The one station that managed to report impressive winds was the Exuma International Airport, which reported a sustained wind of 119 mph and a gust to 144 mph.

      Details on hurricane hunter measurements are here:

      http://www.hurricanehunters.com/plane.html

      Flight-level winds are not what is used for the surface wind. These winds are averaged in 30-second intervals. These winds are then adjusted downwards to account for the weakening of winds from flight level (~10000 feet) to the surface. Dropsondes assist in determining how much the winds are weakening with height. In addition, the SFMR device measures surface emissivity from waves to assess surface winds. When nearing land, radar can also significantly assist in assessing just how strong the winds.

  47. Dr Klotzbach,
    I don’t disagree with what you are saying, my concern is the the wind velocities constantly reported are not what one expects at ground level. NHC should also report and predict the ground level velocities which are more important from the wind destruction and safety viewpoint. Survival at the reported high velocities is near impossible as one News caster reported “children will die” For emphasis, below an earlier comment is repeated.
    “There are industry recognized groups of engineers that look at the wind data and have specified the design wind velocity for every portion of the US and the structure design wind velocities are less than the NOAA reported 165 MPH velocities. The maximum design velocity is 150 mph in the Gulf which reduces to 120 mph in the NY Harbor region. See document below.
    https://law.resource.org/pub/us/cfr/ibr/003/asce.7.2002.pdf
    These are 3 second gusts at 33 feet above ground level and the design formulas clearly adjust the load for the elevation. Look at page 37 for a map of the East Coast. I am sure these are conservative and represent the best estimate for establishing safe structural design.

    Since the Max wind velocities (at possibly 10,000 ft.) are reported and people see the ground velocities from other sources, one will become suspect that the much higher velocities are hyped. NHC and NOAA should recognize the misrepresentation and make a correction to the information they present rather than trashing the messenger.

    As an engineer involved in the design of structures, I have known for years, that the reported velocities do not occur at ground level since essentially no commercial structure would survive as the associated forces on the structure are proportional to the wind Velocity squared.

    While there is criticism directed at Drudge, would you also expect a correction from the same organizations to a presidential candidate claiming that Matthew was worse because of Global warming although data indicates otherwise? Do you agree with that candidate’s statement statement?

    • Flight level winds are not what is utilized for NHC’s intensity estimates. These flight level winds are just one of several tools that NHC utilizes in its assessments. Also, remember that NHC is reporting the maximum one-minute sustained wind at 10-meter height anywhere in the hurricane. Thus, unless your building is well-removed from any trees or other terrain impacts, you aren’t going to see the strongest winds.

      While NWS stations are few and far between, networks of weather stations like Weather Underground can help to fill the gaps. However, most of these stations aren’t likely to have anemometers that are 10 meters high. For example, in my neighborhood, my HOA would not be happy if I decided to put a 10-meter high anemometer!

      Regarding TCs and climate change, here’s my most recent peer-reviewed publication on the topic:

      http://tropical.atmos.colostate.edu/Includes/Documents/Publications/klotzbachlandsea2015.pdf

  48. “The Information Age”

    We got these satellites now! Didn’t use to have them. When libs are in charge, they like to take advantage of opportunities such as these to exaggerate their lib claims. Lib media loves it.

    No story here, anyone able to think their way out of a wet paper bag already knows this…

  49. words do means things…”maximum sustained winds” are NOT the peak found during the one minute of observation that reading is a “GUST” the MSW is the LOWEST reading during that minute, the lowest reading is the maximum SUSTAINED wind…….

  50. Hours before Drudge’s comments on over-hyping appeared, I was online tracking the available buoy and land-based wind measurements in real time, as the hurricane moved through the Bahamas. On one of my screens I had the radar picture. Having spent many months of my life sailing through these waters, I am familiar with the water depths and land configurations in all of these islands. All of the instrumentation wind-speed maxima (available in real-time online) were far, far below the Category 4 levels being forecast during those moments, as I compared the radar eye-wall positions with the instrument readings on Andros, Berry Islands, Bimini and Grand Bahama, as well as from buoys. (Some of the latter were at locations in the lee of the land, or otherwise were not likely to be sheltered by extremely high waves). Drudge’s comments confirmed exactly what I had seen.

    A few days after Hurricane Matthew I drove down the U.S. East Coast (from the Chesapeake to the Florida Keys) and saw storm damage, at the worst, consistent with 35-45 knot peak gusts.

    I have a background in geophysics and when I first read (in the New York Times) a version of the above article, I was hoping to see an expert retrospective analysis of these same specific instrumentation data points, with a comparison of the forecasts and some explanation of the huge actual discrepancies in eyewall maxima (typically 45-55 knots rather than 120-130 knots). Instead, what I see in the article is apologetics, hand-waving and models, with little actual data. My worst suspicions are confirmed.

    • Your observations are correct. Typical plot of winds show Mathew was below hurricane force.
      Eg. http://tinyurl.com/hbmobbr
      should shows the Charleston, SC winds during Mathew much less than claimed by the NHC.

      All of the NDBC buoys off the US coast recorded consistent sustained winds in the 40 knot range. Two exceptions, the St. Augustine station SAUF1 recorded 55 knot sustained winds maximum.
      Also, Diamond Shoals, North Carolina (NDBC Station 41025) recorded 52 knot sustained winds.
      Land stations showed consistently lower winds compated to offshore buoys.

      The bahamas station SPGF1 recorded 68 knot sustained winds before it failed, but the plot of the winds from that station showed it was likely that any later winds were not greater much greater. Also, SPGF1 is not a buoy. It’s a land tower with anemometer height of 9.8 meters. Mathew was Category 1 over the Bahamas.

      There are peer reviewed papers that analyze and compare buoy data to other data. For example, Gilhousen 1986 tests the quality and accuracy of the offshore buoy observations.
      http://journals.ametsoc.org/doi/abs/10.1175/1520-0426(1987)004%3C0094%3AAFEONM%3E2.0.CO%3B2
      The bottom line is that the buoy based anemometers are reliable, thats why they have been out there for decades. The photos of damage confirm the winds measured by the buoys and surface stations.

      Another example of NHC claimed winds above reality is when Mathew passed directly over NDBC 42058 located south of Haiti that showed peak sustained winds of 65 knots at the same time that the NHC was claiming 125 knots. A time plot of sustained winds recorded by that buoy clearly show the eyewall on approach, the drop as the eye passed over, and eyewall winds on the trailing side. There is no possible explanation for that buoy being 60 knots too low. The NHC claims are not confirmed by reliable surface measurements or surface damage.

      • OK, so a Category 1 hurricane overturned cars in Freeport?

        AMS studies show that winds typically have to be at least 115 mph to overturn a mid-sized sedan. Also, if you look at the background of the same picture, palm trees are virtually completely toppled. While it’s hard to tell for certain what type of palm tree those are, sabal palms are typically only completely felled by Category 4/5 hurricanes.

  51. When Matthew was reported to be a major hurricane, its maximum winds were not onshore in Florida. Given your background in geophysics you would understand that assuming the pressure was 940 mb and that Matthew was not an extraordinarily large storm, you would have to have stronger winds than 45-55 knots to equal the pressure gradient force.

    Here’s max wind gusts from TWC.

    Cape Canaveral, Florida: 107 mph (on an elevated tower at 54 feet above the ground)
    Tybee Island, Georgia: 96 mph
    Jennette’s Pier, North Carolina (Outer Banks): 91 mph
    Daytona Beach, Florida: 91 mph
    Hilton Head Island, South Carolina: 88 mph
    Jacksonville Area: 87 mph
    South Ponte Vedra Beach, Florida: 84 mph
    Duck, North Carolina: 83 mph
    Beaufort, South Carolina: 83 mph
    Fort Pulaski, Georgia: 79 mph
    Folly Beach, South Carolina: 76 mph
    Oceana NAS, Virginia: 75 mph
    Myrtle Beach, South Carolina: 74 mph
    Savannah, Georgia: 71 mph
    Melbourne, Florida: 70 mph
    Charleston, South Carolina: 69 mph
    Florence, South Carolina: 67 mph
    Lumbertgon, North Carolina: 66 mph
    Fayetteville, North Carolina: 62 mph
    Sumter, South Carolina: 61 mph

    As I’ve stated several times before, it is very hard for a single surface station to sample the maximum 10-meter wind in a hurricane.

    • Why show only gusts?? Show the sustained wind speed data for those same stations. I was watching the data in real time, no sustained winds reaching over 56 knots at any surface based anemometer. Not just the majority of stations, all the stations in the path of Mathew show the same sustained wind speeds from Florida to North Carolina.

  52. think of a race car going around an oval track….obviously it hits higher speeds on the straights and slower in the turns, so its maximum sustained speed would be the LOWEST reading on the curves, NOT the maximum on the straight(GUSTS).

  53. One real important factor here is the effect of over-hype, but also the fact that peoples own experiences with claimed wind-speeds vs real-wind speeds can lead to real deadly mistakes.

    Back in the day, when I was very young, we had to regularly look at the wind-chill numbers to dress appropriately. Lots of these numbers made little sense but over time I got to understand them. A couple of real close calls on frostbite and deip body chill was enough to respect the numbers and be happy I lived through my exposure. But the fact is, IT IS HARD to understand QUANTITATIVELY what they mean. When people hear difficult to understand numbers and then experience something far less, this will lead to deadly consequences and distrust of the reporting authority.

    Missing the predictions badly just makes NHC look bad. But the effect of the apparent low winds on land and the reporting of higher winds gusts by the NHC makes us think we should perhaps not trust them.

    An important consequence of them missing it. As the next time around LOTS of people may try to ride it out cuz they did the last one. Trouble. Maybe they should do a post anaylsis session after each hurricane and then admit when it appears they goofed or show how accurate they were.

  54. Dr. Klotzback… I have been a meteorologist nearly all my life and I am now 71. I was a professional meteorologist for over 40 years and am a Fellow of the AMS. I have come to appreciate your work and that of Dr Gray more so as the years have gone by. However, as mostly an observer of hurricanes (I have only personally experienced the effects of Hazel, Agnes and David) it has been my experience that rarely have observed wind speeds matched those of the NHC bulletins, especially so with Matthew.

    You keep mentioning how bad conditions would have been had Matthew been 50 miles farther to the west. Granted they would have been worse but would it have been the catastrophe we have been led to believe? After all, the eye wall did pass over Nassau (did it not?) and as at least a category 3, maybe 4. But look at the damage described in this article from a comment above by Frank N. http://eleutheranews.com/?p=6765

    The article states “NASSAU, THE BAHAMAS – The 47 marina and allied members of the Association of Bahamas Marinas are open for business and are anticipating a busy winter season, almost all members having experienced no significant damage during the recent hurricane Matthew.” How can this be if the winds were truly as high as NHC reported?

    Prior to the storm, Joe Bastardi frequently mentioned this was a worst case scenario for Nassau. Well, no, that’s not what happened. So next time a storm of this supposed intensity is forecast to strike, how many people will disregard the warnings based upon what happened this time?

    I don’t for a minute believe NHC is inflating wind speed numbers but I do believe the current state of warnings does not give residents a realistic picture of what may occur. I applaud your attempt to explain the uncertainties involved in determining wind speeds but I don’t believe it resolves the disconnect between what people see with their “lying eyes” and what the bulletins are stating. Good luck with that by the way.

    • There was some pretty significant storm damage in Freeport. While I can’t quantitatively say exactly what kinds of winds are required to overturn cars, it’s certainly well into the sustained hurricane-force wind category.

      Given that all other cars are in parking spots, this was certainly done by the wind and not by water. The Bahamas have some very strong building codes, especially for their businesses. They are no stranger to major hurricanes. This is evidenced by the fact that the airport on Exuma measured a sustained wind of 119 mph and a wind gust to 144 mph, but yet according to that report, was ready for business just a few days later.

      In the case of the United States, except for a few exceptions like Hurricane Andrew (1992), most damage in hurricanes is caused by water, not wind. Had Matthew tracked 50 miles further west, the storm surge would have been much higher, inundating many coastal communities. The difficult challenge with Matthew was that it was tracking parallel to the coast. A 50-mile track farther east, and it would have been a virtual non-event for the state of Florida. Several reliable models brought the storm onshore. In this situation, I don’t see how NHC could do anything differently than they did.

      Neil Frank used to say that you have to evacuate three times for every one time that you need to evacuate. While hurricane forecasts are getting better, these storms aren’t happening in a lab where they can be forecast perfectly.

      • Is anyone arguing whether or not Matthew was devastating?

        It doesn’t matter what category you call it, the damage in Haiti, the Bahamas, and the Carolinas was substantial. A category 3 hurricane has devastating wind speed. I’ve also seen Tropical Storms that have created substantial damage. Sandy was devastating even though it was no longer hurricane strength when it struck New Jersey.

        The only question that I see has to do with trend analysis. If there is an upward bias to the measurements, then people who study the relationship between hurricane strength and climate change will be affected by this upward bias. Already we had people crying “climate change” about Matthew before it left Florida even though it was one of the first major hurricanes in a decade.

  55. Thanks to everyone who has commented on this, very educational. However, I haven’t seen a comment about this:
    “… one 2014 study found that on average, “a single perfect anemometer experiencing a direct hit by the right side of the eyewall will underestimate the actual peak intensity by 10-20% percent. Even an unusually large number of anemometers (e.g., 3-5) experiencing direct hits by the storm together will underestimate the peak wind speeds by 5%-10%.””

    But from the abstract of the study:
    “This study uses an observing system simulation experiment (OSSE) approach to test the limitations of even nearly ideal observing systems to capture the peak wind speed occurring within a tropical storm or hurricane.”

    So are the anemometers really underestimating wind speeds, or has the data been adjusted to the model? Have the models been verified? Shouldn’t they actually be verified by the anemometers, or as others have sugggested, by the observed damage?

    • i appreciate your efforts, but it sure seems to me the term “sustained winds” is not even being discussed…..sustained means constant winds, NOT the highest winds in one quadrant… sustained winds SHOULD mean the highest wind that is constant in the entire storm center.

      • Averaging winds over the eyewall would be a meaningless quantity… In a sheared situation you could have 120 mph winds dominating on one side and 60 mph winds on the other side. To say that maximum sustained wind was 90 mph would not be helpful to anyone. That’s why NHC has the maximum sustained wind but also specifies winds in various quadrants.

      • where was averaging suggested? for certain claiming there is a sustained wind based on the maximum of a tiny portion of the storm is also meaningless…….bottom line for me is the public needs accurate info not hype and matthews winds were HYPE = nothing even close verified on the ground

  56. The problem is not that modern techniques for assessing hurricane strength are inaccurate. Clearly they are more accurate. And when it comes to giving warning to people in the path of a Hurricane more accurate data can only be to the better. However these more accurate techniques were not available in the past. Would Matthew have been recorded as a 5 using the tools available in the 1940’s? Doubtful I’d say.

    Better measurement techniques means that Hurricanes are probably assessed as being stronger today than the same Hurricane would have been in the past. And this biases the long term record.

    When anemometers were the main way of recording wind strength they did sometimes record sustained (8 minute average) wind speeds of Cat 5 strength as hurricanes passed over them. If for all the valid reasons mentioned, these anemometers were underestimating then those Cat 5 Hurricanes of the past must have been absolute monsters. If ever such a monster hits us again I suspect the usual suspects will want to invent a Cat 6 to describe it, and claim that global warming has made Hurricanes worse.

    • Most older measurements from hurricanes are not from anemometers over the open ocean but from the Beaufort Scale used by mariners. I have not yet found a single measurement of sustained Category 5 winds from a hurricane in the United States. Anemometers were few and far between in 1935, Camille took out all anemometers within a ~50 mile radius of the coast, and Andrew’s top winds from a surface station were 146 mph (Category 4).

      But yes, there are likely underestimates as you go back further in time with the Atlantic hurricane database. That’s why most climate studies for the Atlantic start in the 1960s when geostationary satellite should have captured most systems. Of course, the Atlantic also has very large natural multi-decadal cycles in hurricane activity that would make any trend detection hard to quantify.

      • beaufort measurements from ships are based on the ships anemometer. We have had those on ships since the 15th century. They work!

      • Phil I doubt any sailor going back to the 1500’s would sail without an anemometer. This is a reporting issue vice an equipment issue. Even the paper discusses the form issue and I would say that many periods are devoid of military weather reports due to the inclusion of location which is frequently a piece of information military vessels just do not report without some prodding. Prior to 1950 your dealing with a variety of issues that have nothing to do with the weather or instrumentation on ships.

  57. The point is not windspeed per se for climate change, but wind speed RELATIVE to previous storms. If new methods are used to upgrade hurricanes, a “fix” must be applied to the historical data.

    Apples and oranges. The msm and alarmists call hurricane Matthew a Cat 4 or 5, though it made landfall at Cat 1. Anecdotal bad weather now equals historical measurements when dending AGW.

    Global warming is a witch; a lack of proof is not proof, once again, of a lack of witches.

  58. However, while we should never underestimate the impact of a storm nor should they be hyped, in 2013 the BOM in Australia hyped a storm near Rockhampton from Cat 2 up to cat 5 that’s a hype of 8 TIMES – damage was relatively minor because surface wind speed was only category 1 – 2. The result is that next time a cat 5 storm is bearing down on them they won’t take it seriously enough. The BOM was crying wolf.
    So it is with Matthew, Drudge is right to call it out, it was misclassified by 10m surface wind velocity. People need accurate information and not be desensitized to cat3+ storms by the weather people hyping low category storms. What the satellites show was not accurate and Matthew was probably never a category 5 in terms of surface wind.

    That doesn’t mean it wasn’t dangerous, just that we were being lied to about how dangerous

    • Yep. Trees still standing, and they show a picture of an old house without a roof to talk about the “devastation.” Ignoring that in the background was a new house that not only didn’t lose its roof, it didn’t even lose its SOLAR PANELS!! That was no Cat 5.

      And what’s worse is that despite having a weather station in the path of an eye, the BoM disappeared 3 days of data because they only showed Cat 2 readings (I had a screenshot before all the data went away). It was physically impossible to escalate from Cat 2 to Cat 5 in a couple of hours – it didn’t happen, and the data shows is didn’t happen as does the distinct lack of damage.

      Then they did it again in the Peoples Republic of South Australia. “Unprecedented” “50 year storm event.” One that didn’t show a gust over 115km/h (just 60kt) on the mainland, when Woomera had recorded 167km/h in 1979 which was a lot less than 50 years ago.

      Next they’re going to delete the records from Woomera to make their claim true.

      See also Typhoon Haiyan. I saw people filmed on their roofs, hanging on for dear life while keeping out of the certainly fatal storm surge that sent waves of debris-laden seawater through their streets. Tragically, they probably died. But the point is they were out there with a fighting chance – there is no way a guy hangs onto his roof in the “strongest storm in history!”

      The whole hoax came out of the BBC reporting a gust as “mph” instead of “km/h” and someone claiming a world record on the strength of a typo. 4 years later, the typo has become the official record and anyone saying it was a Cat 4, weakening to a Cat 3 at landfall. 99% of the damage filmed was from storm surge – actual wind damage was limited.

      I’m not accusing the NHC of the same thing – Mathew was a very big hurricane and had it turned left Floridians were in heaps of trouble.

  59. I have no problem with the view that current data is accurate, thanks to better recording methods that can assess the entire storm rather than relying on a direct hit at peak speed on an anemometer that doesn’t break in half. Fine, the wind speed recorded was a Cat 4, so it was a Cat 4.

    My problem is not comparing like for like, and deducing that modern events are unprecedented / increasing / weirded.

    If the 1920’s had a 10 year period between Cat 3 landfalls hitting a weather station, then that’s don’t take a radar reading 100km between stations and claim it’s accelerating. Either scale up the history (like they adjust the past temp data), or record “weather station only” datasets to allow inter-generational comparisons and just compare old to old data.

    • Well, your mileage may vary. Because I do have a problem with current data. Dr. Klotzbach says that the NHC can estimate different wind speeds in each quadrant. Well, that’s great, so when a quadrant passes over a buoy, like 42058 in the article, the buoy’s gusts (being generous and discounting sustained wind because of swells) should at least match up to those speeds. Did they? No.

      But then, it was only one buoy. Perhaps the winds saw it and, taking pity, slowed down. Just like, before the recent El Nino, the global warming heat was apparently going into the deep sea without touching the surface… It’s all alarmingly bad, but we just don’t have any witnesses to prove it.

      Rich.

  60. Well, I think people being unreasonable is as much at fault as any external cause for alarm fatigue. Human beings are famously bad at understanding and managing probabilities and risks, and our biases often hinder us in recognizing good information when we see it.

  61. I am sorry, but I am with Matt Drudge on this one. It wasn’t that the ground stations measured 10-20% less. The largest gusts were 1/2 of what was claimed and some weather stations got pretty close to the center of action.

    I think the issue might be with the estimate rather than the ground stations. But never doubt the simulations/computer models.

    • I’ve stated in my response several times that these are surface wind estimates, not winds recorded at any particular level. I certainly have not “ducked” the question as you posit.

      You can track all aircraft reconnaissance through a nice viewable format at Tropical Tidbits:

      http://www.tropicaltidbits.com.

      Aircraft recon reports winds measured at flight level (10,000 feet) along with estimates of surface winds from the SFMR instrument. You can also get the data reported by individual dropsondes.

  62. Does anyone here remember how the “science community” responded during the wild accusations and claims that were made at the time of Hurricane Katrina? To the weather-people: you reap what you sow.

  63. marque2,
    I agree with you. I personally think NOAA/NHC report the maximum recorded velocity at any elevation rather than the velocity at ground level. I have raise this question with the writer several times and he seems to duck the question.
    I have offered as evidence the design winds that engineers use for the east coast are lower than the reported NOAA hurricane wind velocities and has various factors for elevation adjustments and exposures, and the author seems to duck the question. These numbers are not different because of adjacent structures as suggested since many structures are not protected by other buildings, and rise to rather high elevations. At the risk of repeating again:

    See document below.
    https://law.resource.org/pub/us/cfr/ibr/003/asce.7.2002.pdf
    These are 3 second gusts at 33 feet above ground level and the design formulas clearly adjust the load for the elevation. Look at page 37 for a map of the East Coast. I am sure these are conservative and represent the best estimate for establishing safe structural design.

    Note that these are 3 second gusts which is a shorter timeframe than the NOAA reported timeframe. These folks .who establish this have a lot of responsibility to keep us safe and don’t take the responsibility lightly.

    • NHC’s reported winds are at the surface, not at any level in the storm. I suggest following their aircraft reconnaissance reports in real-time. They report winds both at flight level (10,000 feet) as well as at the surface as estimated by the SFMR instrument.

      The eyewall of Matthew did not come ashore in Florida when it was a Category 3/4, so I wouldn’t have expected to see 100+ mph winds on land. As I’ve posted in several other replies, there were wind gusts reported in the 80-90 mph range with Matthew’s landfall in South Carolina, which is consistent with a Category 1.

      As far as its strength through the Bahamas, Exuma airport reported a maximum sustained wind of 119 mph and gusts to 144 mph. Freeport’s station failed before the center of the storm came ashore, but as I’ve shown earlier in the discussion, cars were flipped over by the wind. Since you’re the engineer, you can tell me what kind of wind it requires to flip a car, but it’s certainly in the Category 3/4 range.

      • Dr Klotzbach
        Thanks for your reply
        FYI I have never judged the accuracy or motive of the information provided by the NHC or NOAA, I was skeptical on the media and NOAA storm descriptions reported wind velocities based on experience and the link I provided from the asce as to how they relate to surface winds used for design of structures. If I understand you correctly the velocities are “ground Level”?

        “NHC’s reported winds are at the surface, not at any level in the storm.”
        So are you saying that when I hear the Hurricane is packing 145 mph winds these are at the surface somewhere, recognizing they reduce further away from the center?
        Or is the surface wind velocity reported elsewhere? If so please advise where I should look to find it. As a boat owner it is important to me to make decisions to try to protect property.

        I don’t claim to be an expert to interpret wind velocities or as one who develops the ASCE design velocities used throughout the country, I just use them for structural calculations and would like to understand any differences. FYI the ASCE design velocities are for 3 second gusts and vary throughout the elevation of a structure and appear to be circa 10% less than velocities reported.

        FYI I looked at velocities to flip a car which is reported as “115 to 180 MPH to flip a car”.

        I apologize If anything I said appears to be derogatory, I was just trying to understand whether the reported velocities should be expected at ground level and why different than the ASCE in some cases.
        Appreciate your reply.

      • The official NHC forecast is for winds at 10 meters above ground level. There a ton of graphical products available from NHC which highlights the wind field for each storm:

        http://www.nhc.noaa.gov/archive/2016/NICOLE_graphics.shtml

        These products are excellent, as they can highlight significant storm asymmetries. For example, in sheared storms, hurricanes may have very little (if any) hurricane-force winds in specific quadrants of the storm. Levi Cowan at Tropical Tidbits (http://tropicaltidbits.com) has a great aircraft reconnaissance page which displays real-time flight level wind measurements by the Hurricane Hunter plane in real-time.

  64. Dr. Philip Klotzbach, thank you for your guest post here as well as your comments. But you seem to be trying to work a PR scheme in the interest of the NHC.

    The past 7 years or so hurricanes have been over-hyped to an unprecedented degree by both the NHC and of course the press. Only a fool would believe what the press reports but I would expect at least some degree of the truth from the NHC. I have followed in real time several reported hurricanes and tropical storms in the Atlantic/Gulf basins the past several years. Two in particular were Irene and Sandy. Had family in the Wilmington, NC landfall area and multiple tabs open on my computer which I monitored for hours. Irene was waaay over-hyped. Then along comes Sandy and it was the same business as usual (over-hyped) from the NHC. Feel free to look into the WUWT archives as Sandy traveled up the Atlantic seaboard as I (eyesonu) documented weather station reports/data from many locations that were in extreme conflict with the NHC. Just for fun I was also tossing in weather /data from some city on the great lakes that was not under any influence of Sandy for comparison.

    You have done well in your PR efforts here but I’m not buying it. Good luck with any grants you may be seeking as your efforts here will look good on your resume.

    • Sandy was a rain and surge producer, not a wind event. NHC never stated the winds were the main problem. I’m sure those living in New York/New Jersey would beg to differ that Sandy was over-hyped. $75 billion dollars in damage is hardly a small event.

      • I have always felt that if it hit somewhere further from where the press and government are ensconced it would have never been crowned with the made up nomenclature of “super storm” no matter how populated or how much damage it did at landfall.

        That is the type of overhype that I believe many of us in flyover country often perceive. Anything that ever happens is worse when it occurs along the I-95 corridor from the NYC area to the DC area. Just like the hype when they had a high in the mid 90s in Central Park this summer. That wasn’t even close to a record but you would have thought it was life threatening for the general masses the way it was covered.

    • Have you looked at earlier notorious hurricanes? Katrina’s maximum sustained wind on U.S. soil was 86 mph… was that also NHC hype?

      I suggest a thorough search of maximum sustained winds in other hurricanes. In Category 5 Camille, there were no anemometers left within 50 miles of the coastline after the storm.

      I am surprised that everyone is so fixated on anemometers and ignores constant wind reports from aircraft, as well as estimates from the SFMR and radar. There’s a reason that NHC uses these and just doesn’t depend on surface observations. For example, in Hurricane Ike, all NWS weather stations within close proximity to landfall failed.

      Are you saying that they are deliberating fudging the data, or just that there are massive wind reductions from 10,000 feet to the surface. If the former, that displays considerable paranoia and if the latter, you are ignoring many published studies as well as the dropsonde data that helps establish the appropriate reduction from flight level to the surface.

      But if you want to declare yourself the expert, so be it.

      • Phil Klotzbach said:

        “I am surprised that everyone is so fixated on anemometers and ignores constant wind reports from aircraft, as well as estimates from the SFMR and radar.”

        It’s partly because many people have painted themselves into a corner with uninformed criticism. To climb back down from their position is to acknowledge that they didn’t have sufficient understanding to take such a position in the first place. The belief that the NHC and other organizations are deliberately trying to mislead the public is a deep part of many peoples identities, and to acknowledge otherwise is a much bigger step for them than simply acknowledging they were wrong about some specific scientific detail.

      • Dr. Klotzback

        Do you deny that there is a natural tendency for inflation of both incidence and severity as compared to past records due to the march of technology in detecting and measuring TCs?

        NOAA certainly recognizes such inflation in the US tornado count record and is attempting to adjust for it.
        http://www.spc.noaa.gov/wcm/adj.html

        BTW I notice that their adjusted count for this year is very close to that of the historic minimum set in 2013.

      • Phil there is a huge difference between anemometers reporting low numbers and being destroyed. only a very small number of anemometers died to Mathew. where as Camille, Andrew and Katrina killed them all. You need to rethink what your saying.

      • David Riser said:

        “Phil there is a huge difference between anemometers reporting low numbers and being destroyed. only a very small number of anemometers died to Mathew. where as Camille, Andrew and Katrina killed them all. You need to rethink what your saying.”

        Camille, Andrew and Katrina made landfall. Mathew didn’t. Why would you expect similar damage to anemometers from a hurricane that didn’t make landfall?

    • Sure, I agree with you that there is a natural tendency for inflation in TC numbers, especially for weak named storms. Chris Landsea did an excellent job documenting this in a series of studies at NHC. When it comes to stronger hurricanes, they are probably reasonably consistently observed in the Atlantic since the mid-1960s, given that we’ve been using aircraft reconnaissance to fly most of the stronger storms. Underestimates likely plague most other global TC basins until ~1990. So, when looking at metrics like Accumulated Cyclone Energy, I would say that values since the mid 1960s probably aren’t too far off from reality in the Atlantic. Prior to that, there are likely underestimates. Gabe Vecchi and colleagues looked at ship traffic prior to 1966 to try and estimate “missed” hurricanes. When you apply these underestimates, you see no trend in hurricanes since the late 1800s. Similarly, when I assume that each missed hurricane generated ~10 ACE units (equivalent to the average observed Category 1-2 hurricane from 1966-2015), you get no trend in ACE since the late 19th century.

  65. With regard to failing anemometers, Dr. Klotzbach wrote:

    “I suggest a thorough search of maximum sustained winds in other hurricanes. In Category 5 Camille, there were no anemometers left within 50 miles of the coastline after the storm.

    “I am surprised that everyone is so fixated on anemometers and ignores constant wind reports from aircraft, as well as estimates from the SFMR and radar. There’s a reason that NHC uses these and just doesn’t depend on surface observations. For example, in Hurricane Ike, all NWS weather stations within close proximity to landfall failed.”

    Back on April 12, 1934, the weather station on Mt. Washington, NH, recorded a peak wind gust of 231 miles per hour. “The anemometer used to record the record wind was a heated anemometer designed special for Mount Washington. It was constructed in Cambridge MA, and tested in the wind tunnel at the Guggenheim Aeronautical Laboratory of the Massachusetts Institute of Technology in Boston.

    “After the wind measurement, the anemometer was run through a number of tests by the National Weather Bureau and the historic measurement of 231 mph was confirmed to be a valid reading.”

    Source: https://www.mountwashington.org/about-us/history/world-record-wind.aspx

    My comment and question for Dr. Klotzbach: For the general populace, surface winds are the only relevant measure, as they cause the damage. Why is there not in place in “Hurricane Alley” a comprehensive network of “special” anemometers, a la, Mt. Washington, designed to measure and withstand Hurricane Camille – force wind gusts?

    • I think the real problem with anemometer failure is that in most of the strongest storms, storm surge takes them out (as opposed to the wind). Of course, in the case of Hurricane Andrew (1992), winds did take out the anemometer mounted by NHC.

      While there isn’t a special anemometer network in Hurricane Alley is a bit curious to me as well. I will note in the case of the Mount Washington Observatory, it is continuously manned by a team of observers. These observers go to nearly heroic efforts during the winter months to keep the anemometer clear of rime ice. I encourage you to check out the videos that they post in the winter. It’s pretty crazy how intense those winds are in the winter.

      The Mount Washington Observatory is also a non-profit which has additional funding beyond that coming from NOAA.

  66. That was supposed to assure us ? Color me unimpressed … Basically the author showed that they have no way of telling a normal person not living at 10,000 feet in the path of a storm what the expected winds could be … So basically useless … Typical a** covering academic nonsense

    • Did you actually read the article? There have been many studies that have looked at adjustments between 10,000 feet and the surface. They also utilize dropsondes to measure winds descending from 10,000 feet to the surface. The SFMR utilizes the sea state to estimate winds at the surface. The use of the sea state to estimate surface wind speed has been a technique employed for over 50 years. Radars can also be used to quite accurately estimate surface winds close to land.

      On another note, they used to fly aircraft much lower than 10,000 feet into hurricanes, but a Hurricane Hunter plane in Hurricane Janet (1955) crashed into the sea when it attempted to penetrate Janet’s eyewall at 700 feet, which is when the aircraft decided to fly considerably higher.

      https://weather.com/storms/hurricane/news/hurricane-hunter-plane-crash-janet

      But if you want to knock 60+ years of research by hundreds of scientists as “academic nonsense”, that is your decision.

  67. As a boy, I witnessed the landfall effects of the 1962 Columbus Day storm in Washington state (Bellingham area, which is unprotected from weather coming through the Straits of Juan de Fuca). Very impressive. And the 1993 Inauguration Day storm, the aftermath of which made my cul-de-sac look like a logging camp.

    We recently were warned of a typhoon landfall that was expected to be of comparable strength. It came and went without much effect. So, the NHC estimated too high? I have no complaints! Anyone who wants to complain about being disappointed with lower-than-advertized winds can have my share.

Comments are closed.