Jeff Master's rips Gray's April Forecasts

[Downdraft]

Here's the link

http://www.wunderground.com/blog/JeffMa ... amp=200704

[Matt-hurricanewatcher]

I agree with Jeff master, there has only been a handfull of 15 plus seasons over the last 150 years. This one doe's not look to be super favorable, in fact looks more like 98,99 or maybe 03. The Tropical Atlatnic is NOT as warm as 03,04,05 or even 06. The big difference being that there is a la nina developing. Also the fact that the high is forming in a better place for less over all shear over the tropical and western Atlatnic. So yes this year doe's in fact have its favorable area's. I say for shocking events maybe 2004? Maybe not, but we will see. I'm starting to think that everything science these days are turning into nothing more then something to scare people, where is the facts? Where is the history and the data supporting 17 named storms out of the box, that would be one of the most active seasons in history. I think 14 maybe 15 named storms. This is just my option.

[HURRICANELONNY]

It will be busier then last year. Probably analog year would be close to 98. The number means nothing. It takes one to make it your worst nightmare. Nobody can guess the # with much accuracy but it's the data that counts. We're in a more active period where the average is alot more names then the last 30 or so years.

[wxman57]

I would caution anyone here about trying to equate warm Atlantic SSTs with an increasing number of named storms. When I began researching past development patterns with respect to the AMO, I thought that surely I would find that more named storms formed when the Atlantic was warmer than during the cool decades. But I could find no such correlation. In fact, my studies showed slightly fewer (0.2) named storms formed when the Atlantic was warmer (1926-1969) than when it was cooler (1970-1994). Dr. Gray touched on this briefly during his talk on Friday. We don't see more named storms when the Atlantic heats up, what we find is that more hurricanes that DO develop tend to reach Cat 3-4-5 intensity - about twice as many.

Therefore, a warmer Atlantic in 2007 won't lead to any more named storms. Low and mid-latitude wind shear is a much more significant variable as far as determining the number of named storms that form each year. The water could be boiling but if wind shear is too strong, nothing will develop. So the forecast of a moderate La Nina could have a significant impact on the number of named storms, in combination with the new Atlantic Meridional Mode favoring lower wind shear and a weaker Bermuda high.

One other factor as far as the yearly "normals" of TC activity. I've been following the work of Chris Landsea and Roger Pielke closely. They estimate that in the pre-satellite era that an average of 3.2 named storms per year were missed in the Atlantic Basin. If we factor these number in, then the average number of named storms each season in the Atlantic rises from 9.6 to nearly 13. That also means seasons with 17 named storms may not be as rare as one might think.

Finally, combine the above with the fact that the NHC is now naming subtropical storms and the new seasonal average may rise to between 13 and 14 named storms. That said, 17 named storms doesn't seem so high.

Until I see stronger evidence of that moderate La Nina, I'll stick with my 14/8/4 initial guess. But I reserve the right to modify that number upward if La Nina does, in fact, develop. Also, I still have this gut feeling (or hope?) that the first named storm won't form until early August.

[Ptarmigan]

I would caution anyone here about trying to equate warm Atlantic SSTs with an increasing number of named storms. When I began researching past development patterns with respect to the AMO, I thought that surely I would find that more named storms formed when the Atlantic was warmer than during the cool decades. But I could find no such correlation. In fact, my studies showed slightly fewer (0.2) named storms formed when the Atlantic was warmer (1926-1969) than when it was cooler (1970-1994). Dr. Gray touched on this briefly during his talk on Friday. We don't see more named storms when the Atlantic heats up, what we find is that more hurricanes that DO develop tend to reach Cat 3-4-5 intensity - about twice as many.

Therefore, a warmer Atlantic in 2007 won't lead to any more named storms. Low and mid-latitude wind shear is a much more significant variable as far as determining the number of named storms that form each year. The water could be boiling but if wind shear is too strong, nothing will develop. So the forecast of a moderate La Nina could have a significant impact on the number of named storms, in combination with the new Atlantic Meridional Mode favoring lower wind shear and a weaker Bermuda high.

One other factor as far as the yearly "normals" of TC activity. I've been following the work of Chris Landsea and Roger Pielke closely. They estimate that in the pre-satellite era that an average of 3.2 named storms per year were missed in the Atlantic Basin. If we factor these number in, then the average number of named storms each season in the Atlantic rises from 9.6 to nearly 13. That also means seasons with 17 named storms may not be as rare as one might think.

Finally, combine the above with the fact that the NHC is now naming subtropical storms and the new seasonal average may rise to between 13 and 14 named storms. That said, 17 named storms doesn't seem so high.

Until I see stronger evidence of that moderate La Nina, I'll stick with my 14/8/4 initial guess. But I reserve the right to modify that number upward if La Nina does, in fact, develop. Also, I still have this gut feeling (or hope?) that the first named storm won't form until early August.

SST isn't everything. It also depends on high and cold the clouds are. A hurricane with cold cloud tops can get really strong not as favorable SSTs. Hurricane Wilma had cold cloud tops, which would explain why it had 882 millibars and 185 mph winds, despite forming over SSTs that was starting to cool down. I agree that there were probably more storms that we realized prior to 1965. 1886, 1887, 1899, 1903, 1926, and 1933 had more storms. I think 1933 could of had just had many storms in 2005. 1886 and 1887 were quite active, like 2004 and 2005.

[HURRICANELONNY]

I agree with both of you wx & pt. I think the #'s will be around 14-16. Also I believe your right to with 1933. It could of been worse or equal to 2005. There wasn't ships everywhere and they didn't name subtropical storms. My guess this year is 16/9/4.

[MiamiensisWx]

I completely agree with those who believe 1933 featured more tropical storms than currently indicated in the HURDAT best track database (which has not been reanalyzed beyond 1914). In addition, I see numerous flaws in the current best track positions and intensities in HURDAT as late as the 1970s and possibly the 1980s. Although we had ship observations before the 1940s and reconnaissance and satellite observations in latter years, we have only featured truly accurate estimates and measurements for Atlantic (and eastern and central Pacific) tropical and subtropical cyclones from the middle and late 1980s (at the earliest range) through the present. Ship observations (wind, pressure, and other features) are subject to many variables and the observations can easily be inaccurate (false pressure or wind reports or errors or discrepancies when compared to surrounding observations or the atmospheric synoptics of a particular system).

I believe many storms as late as the 1980s have been overestimated (see 1950's Hurricane Dog and hurricanes in the 1920s, 1930s, 1940s, and through the 1970s). Estimates before 1985 should be taken with a grain of salt, notably when dealing with storm intensities and tracks before 1975. Due to many variables, many storms as late as the 1980s may have also been underestimated. Many tropical storms may have reached hurricane intensity while others may have been tropical depressions. Small errors in reconnaissance aircraft fixes (and observations) and incorrect storm positions can easily result in incorrect storm positions in the best track database. Our knowledge of the atmospheric variables and pressure gradient/RMW relationships with regards to storm intensities has only been explored extensively in the most recent years. While we weren't clueless prior to the 1980s, we were vastly different in terms of our technology and equipment (and techniques) compared to 2007; thus, I would take only our most recent best track files for the more recent storms seriously in terms of overall accuracy and lack of discrepancies (in terms of intensity and best track positions).

I'll offer some examples. Storms like Dog (1950) and Camille (1969) are officially listed in the best track database with 160KT (185 mph) and 165KT (190 mph) one-minute sustained winds respectively. These estimates (and other estimates for many other storms prior to 1975) can easily be slightly inaccurate or even grossly overestimated (or underestimated). Reduction factors in flight-level winds may have been inaccurate. Storm size and pressure gradient play a large role in storm intensity; thus, Camille may have featured winds slightly below the best track estimate (in my opinion Camille peaked at 155KT to 160KT or 180 mph to 185 mph). In addition, Dog (1950) may have been a high-end Category 3 at peak intensity. Similarly, many other storms likely feature the same overestimated intensities (see Hurricane Four from 1926 which was likely well below its Category 4 designation in the HURDAT database due to discrepancies in observations). Hurricane Isbell (1964) is listed in HURDAT as a 110KT Category 3 hurricane at peak intensity and at landfall in southwest Florida, but the minimum pressure of 964 mbar (and larger pressure gradient for an October Gulf of Mexico hurricane) indicate the storm was likely a low-end Category 2 at its highest intensity. The rising pressure to 968 mbar (lowest pressure in Florida was measured at 973 mbar at Everglades City) also indicate the storm likely was weakening prior to Florida landfall (likely after peaking in the southern Gulf of Mexico near Key West) and did not maintain intensity across the state as indicated in the current best track data. Many other storms were likely also grossly overestimated prior to 1975, and I see many small (and large) errors in the track positions prior to the late 1980s and notably prior to 1975. These problems are numerous and extend back through the entire unanalyzed database. This illustrates why intensity and track estimates prior to 1975 must not be taken very seriously. The HURDAT reanalysis project has yet to resolve these errors beyond 1914 (official reanalysis changes have only been officially added for the 1851 to 1914 period).

In addition, many storms consequentially may have been underestimated in intensity prior to 1975. One notable example is Hurricane Two of 1935 (the infamous Labor Day Hurricane that struck the Florida Keys, paralleled the Florida west coast, and made a second landfall in the northwest portion of the state west of Cedar Key). Despite the storm's small RMW and windfield (nearly as small as Charley of 2004) and extremely rapid intensification prior to landfall at Craig Key (the 892 mbar central pressure was measured at this location in the eye) the current best track database lists this storm's peak intensity at 140KT (161 mph). It also indicates the storm weakened rapidly after landfall in the Keys and made landfall at Cedar Key as a 80KT (92 mph) Category 1 hurricane. In my opinion the lowest central pressure observed in the Florida Keys at the landfall point (and small storm size) and the rapid intensification prior to its first Florida landfall indicates the storm was much stronger. In my opinion peak one-minute sustained winds shortly prior to the Florida Keys landfall reached 165KT (190 mph) and winds at landfall were at least 160KT (185 mph) with gusts in excess of 200 mph. The small eye diameter (approximately 8.5 miles wide in my opinion) at landfall also demonstrates a small radius of maximum winds in the Keys and illustrates the small size of the core. In fact I believe the storm's pressure may have been slightly lower just prior to Keys landfall and peaked at approximately 890 mbar just before landfall. This storm was likely a low-end tropical storm as it crossed the Bahamas and likely did not attain hurricane status until it passed Andros Island early on September 1 (observations do not support winds higher than 50 mph on its path through the islands). From that point the storm likely began its rapid intensification before it even entered the Gulf Stream (in my opinion the storm reached upper-end Category 4 status with 135KT winds several hours before it is indicated in HURDAT and likely reached 155KT a few hours before it is even first listed in best track as a Category 5 hurricane).

I also believe this intense hurricane did not weaken as rapidly as indicated in HURDAT after it made landfall in the Keys on September 2. Personally I believe it was still a Category 4 hurricane with 125KT winds west of Sarasota (it retained a small diameter with only a gradual increase in its windfield) and I believe it was still a 105KT Category 3 hurricane as it made landfall west of Cedar Key near Perry, Florida (further west than indicated in HURDAT due to no lull in winds reported at Cedar Key and minimum pressure only at 984 mbar at this location). At its second Florida landfall it struck a region with a very sparse population and few observations available. Due to its landfall location a considerable distance west of Cedar Key it is likely far more intense at its second landfall than indicated in the best track data. In my opinion the minimum pressure at its second Florida landfall was still within the 958 to 960 mbar range at the highest pressure range (a ship reported a peripheral pressure of at least 941 mbar when the hurricane was west of Tampa Bay and nearing its second landfall). In my opinion the storm did not weaken rapidly to tropical storm intensity after moving inland near Perry, Florida; hurricane-force winds were reported well inland through Georgia and considerable damage was reported through the state and into the Carolinas (it likely maintained intensity due to baroclinic interaction with a potent mid-level trough and only weakened to a tropical storm when it entered South Carolina). Unlike its HURDAT designation as a restrengthening hurricane after exiting the United States near Cape Henry in Virginia, I believe it was transitioning to an extratropical system and became an extratropical low quickly after exiting southeastern Virginia (pressures reported as low as 950 mbar in the shipping lanes as it passed south of New England and off Canada were likely associated with another extratropical low which absorbed the system). Some great information on this hurricane can be found here in the September 1935 Monthly Weather Review. The MWR has a lot of information and observations that are often not mentioned or are poorly known by most other references. It is an excellent source.

In a nutshell, there are many variables to atmospheric systems and the current data is often inaccurate in the best track information prior to 1975. There are many errors in the current Atlantic best track databases that can skew climatology and our knowledge of tropical cyclones. These issues need to be resolved. In short terms I would take current official data before 1975 with a large grain of salt and put the data in huge perspective. This is why I don't trust Unisys best track data, official NHC information and the deadliest and costliest hurricane data, and climatology.

[Downdraft]

How about the point of what he says? Looking at the validation of the forecasts the April numbers are worse than climitology? If your April forecasts have a negative skill level what good is putting them out at all? I say right or wrong they can only increase awareness. Those that will not prepare won't care one way or the other. Those that will don't need the forecast to get ready in the first place. So your left trying to sway the undecided or uninformed. Unfortunately here you have two very skilled experts at odds over the product. If that's the case what's the public to believe?

[MiamiensisWx]

How about the point of what he says? Looking at the validation of the forecasts the April numbers are worse than climitology? If your April forecasts have a negative skill level what good is putting them out at all? I say right or wrong they can only increase awareness. Those that will not prepare won't care one way or the other. Those that will don't need the forecast to get ready in the first place. So your left trying to sway the undecided or uninformed. Unfortunately here you have two very skilled experts at odds over the product. If that's the case what's the public to believe?

I disagree with you. After the 2006 bust and other past forecasting errors people will view this outlook by Gray and Klotzbach as hype. Look at the comments posted on every real estate and storm-related article on the Miami Herald and South Florida Sun-Sentinel websites. You are in emergency management and the public safety department and I have enormous respect for your posts (I love your points) but this post is puzzling. There is a lot at stake. I agree with your point on preparedness but last year's bust and the combined media hype and downplaying and lack of general public awareness and knowledge on emergency situations (and forecasts and weather in general) is a recipe for huge problems. In addition, we have an uninformed press who knows nothing about meteorology. Take global warming and the downplaying of storms as good examples of the media's incompetence. The way the press handled Katrina (and other storms) was inexcusable (via declaring New Orleans off the hook and downplaying Wilma's impact on Florida without even listening to the NHC). I can name countless other previous examples going back many years. The insurance industry makes a dismal situation absolutely horrid.

Many people believe any attempt to stop rampant coastal development in vulnerable locations is only feeding the hungry business in the insurance industries. When you add the aforementioned problems into the mix you have a potential disaster lurking around the corner. Don't misread my post. I think you make great points and I'm getting prepared before June 1 (I have four gas cans and numerous batteries and other supplies) but the current overall situation leaves little room for optimism.

[Ptarmigan]

I disagree with you. After the 2006 bust and other past forecasting errors people will view this outlook by Gray and Klotzbach as hype. Look at the comments posted on every real estate and storm-related article on the Miami Herald and South Florida Sun-Sentinel websites. You are in emergency management and the public safety department and I have enormous respect for your posts (I love your points) but this post is puzzling. There is a lot at stake. I agree with your point on preparedness but last year's bust and the combined media hype and downplaying and lack of general public awareness and knowledge on emergency situations (and forecasts and weather in general) is a recipe for huge problems. In addition, we have an uninformed press who knows nothing about meteorology. Take global warming and the downplaying of storms as good examples of the media's incompetence. The way the press handled Katrina (and other storms) was inexcusable (via declaring New Orleans off the hook and downplaying Wilma's impact on Florida without even listening to the NHC). I can name countless other previous examples going back many years. The insurance industry makes a dismal situation absolutely horrid.

Many people believe any attempt to stop rampant coastal development in vulnerable locations is only feeding the hungry business in the insurance industries. When you add the aforementioned problems into the mix you have a potential disaster lurking around the corner. Don't misread my post. I think you make great points and I'm getting prepared before June 1 (I have four gas cans and numerous batteries and other supplies) but the current overall situation leaves little room for optimism.

The press screwed up with Andrew, Katrina, Rita, and Wilma. The press was mostly in the French Quarters or downtown when Katrina hit, not Lakeside or East New Orleans.

[Jam151]

The press screwed up with Andrew, Katrina, Rita, and Wilma. The press was mostly in the French Quarters or downtown when Katrina hit, not Lakeside or East New Orleans.

The national media still hasn't made it to New Orleans East yet. The 9th ward is often confused with this area. OK back to regulary scheduled programming.

[AussieMark]

I remember in 2004 the press was hyping up Tampa with Charley

[Derek Ortt]

press was hyping Tampa for Charley

Then Charley scored a direct hit on Orlando

[Downdraft]

Would Gray's forecasts mean more if expressed as percentages of probability rather than number of storms? Say a 70% chance of more than 15 storms, a 50% chance of more than 12 but less than 15 or a 30% chance of less than 12? I think when you put on a number on developed systems then the focus makes a target of the number. Wouldn't it be possible to extrapolate a percentage based upon observed or forecasted conditions rather than try to pick an exact number that seldom verifies? Just thoughts comments certainly welcome.

As for the merit of the forecast in the first place once again the educated consumer places a value on the product. For awareness I think they are great and to those who know more than the typical "Harry Homeowner" form your own more aware conclusions.

[wxmann_91]

Keep in mind that this is as much Mr. Klotzbach's forecast as it is Dr. Gray's.

[Ptarmigan]

I completely agree with those who believe 1933 featured more tropical storms than currently indicated in the HURDAT best track database (which has not been reanalyzed beyond 1914). In addition, I see numerous flaws in the current best track positions and intensities in HURDAT as late as the 1970s and possibly the 1980s. Although we had ship observations before the 1940s and reconnaissance and satellite observations in latter years, we have only featured truly accurate estimates and measurements for Atlantic (and eastern and central Pacific) tropical and subtropical cyclones from the middle and late 1980s (at the earliest range) through the present. Ship observations (wind, pressure, and other features) are subject to many variables and the observations can easily be inaccurate (false pressure or wind reports or errors or discrepancies when compared to surrounding observations or the atmospheric synoptics of a particular system).

I believe many storms as late as the 1980s have been overestimated (see 1950's Hurricane Dog and hurricanes in the 1920s, 1930s, 1940s, and through the 1970s). Estimates before 1985 should be taken with a grain of salt, notably when dealing with storm intensities and tracks before 1975. Due to many variables, many storms as late as the 1980s may have also been underestimated. Many tropical storms may have reached hurricane intensity while others may have been tropical depressions. Small errors in reconnaissance aircraft fixes (and observations) and incorrect storm positions can easily result in incorrect storm positions in the best track database. Our knowledge of the atmospheric variables and pressure gradient/RMW relationships with regards to storm intensities has only been explored extensively in the most recent years. While we weren't clueless prior to the 1980s, we were vastly different in terms of our technology and equipment (and techniques) compared to 2007; thus, I would take only our most recent best track files for the more recent storms seriously in terms of overall accuracy and lack of discrepancies (in terms of intensity and best track positions).

I'll offer some examples. Storms like Dog (1950) and Camille (1969) are officially listed in the best track database with 160KT (185 mph) and 165KT (190 mph) one-minute sustained winds respectively. These estimates (and other estimates for many other storms prior to 1975) can easily be slightly inaccurate or even grossly overestimated (or underestimated). Reduction factors in flight-level winds may have been inaccurate. Storm size and pressure gradient play a large role in storm intensity; thus, Camille may have featured winds slightly below the best track estimate (in my opinion Camille peaked at 155KT to 160KT or 180 mph to 185 mph). In addition, Dog (1950) may have been a high-end Category 3 at peak intensity. Similarly, many other storms likely feature the same overestimated intensities (see Hurricane Four from 1926 which was likely well below its Category 4 designation in the HURDAT database due to discrepancies in observations). Hurricane Isbell (1964) is listed in HURDAT as a 110KT Category 3 hurricane at peak intensity and at landfall in southwest Florida, but the minimum pressure of 964 mbar (and larger pressure gradient for an October Gulf of Mexico hurricane) indicate the storm was likely a low-end Category 2 at its highest intensity. The rising pressure to 968 mbar (lowest pressure in Florida was measured at 973 mbar at Everglades City) also indicate the storm likely was weakening prior to Florida landfall (likely after peaking in the southern Gulf of Mexico near Key West) and did not maintain intensity across the state as indicated in the current best track data. Many other storms were likely also grossly overestimated prior to 1975, and I see many small (and large) errors in the track positions prior to the late 1980s and notably prior to 1975. These problems are numerous and extend back through the entire unanalyzed database. This illustrates why intensity and track estimates prior to 1975 must not be taken very seriously. The HURDAT reanalysis project has yet to resolve these errors beyond 1914 (official reanalysis changes have only been officially added for the 1851 to 1914 period).

In addition, many storms consequentially may have been underestimated in intensity prior to 1975. One notable example is Hurricane Two of 1935 (the infamous Labor Day Hurricane that struck the Florida Keys, paralleled the Florida west coast, and made a second landfall in the northwest portion of the state west of Cedar Key). Despite the storm's small RMW and windfield (nearly as small as Charley of 2004) and extremely rapid intensification prior to landfall at Craig Key (the 892 mbar central pressure was measured at this location in the eye) the current best track database lists this storm's peak intensity at 140KT (161 mph). It also indicates the storm weakened rapidly after landfall in the Keys and made landfall at Cedar Key as a 80KT (92 mph) Category 1 hurricane. In my opinion the lowest central pressure observed in the Florida Keys at the landfall point (and small storm size) and the rapid intensification prior to its first Florida landfall indicates the storm was much stronger. In my opinion peak one-minute sustained winds shortly prior to the Florida Keys landfall reached 165KT (190 mph) and winds at landfall were at least 160KT (185 mph) with gusts in excess of 200 mph. The small eye diameter (approximately 8.5 miles wide in my opinion) at landfall also demonstrates a small radius of maximum winds in the Keys and illustrates the small size of the core. In fact I believe the storm's pressure may have been slightly lower just prior to Keys landfall and peaked at approximately 890 mbar just before landfall. This storm was likely a low-end tropical storm as it crossed the Bahamas and likely did not attain hurricane status until it passed Andros Island early on September 1 (observations do not support winds higher than 50 mph on its path through the islands). From that point the storm likely began its rapid intensification before it even entered the Gulf Stream (in my opinion the storm reached upper-end Category 4 status with 135KT winds several hours before it is indicated in HURDAT and likely reached 155KT a few hours before it is even first listed in best track as a Category 5 hurricane).

I also believe this intense hurricane did not weaken as rapidly as indicated in HURDAT after it made landfall in the Keys on September 2. Personally I believe it was still a Category 4 hurricane with 125KT winds west of Sarasota (it retained a small diameter with only a gradual increase in its windfield) and I believe it was still a 105KT Category 3 hurricane as it made landfall west of Cedar Key near Perry, Florida (further west than indicated in HURDAT due to no lull in winds reported at Cedar Key and minimum pressure only at 984 mbar at this location). At its second Florida landfall it struck a region with a very sparse population and few observations available. Due to its landfall location a considerable distance west of Cedar Key it is likely far more intense at its second landfall than indicated in the best track data. In my opinion the minimum pressure at its second Florida landfall was still within the 958 to 960 mbar range at the highest pressure range (a ship reported a peripheral pressure of at least 941 mbar when the hurricane was west of Tampa Bay and nearing its second landfall). In my opinion the storm did not weaken rapidly to tropical storm intensity after moving inland near Perry, Florida; hurricane-force winds were reported well inland through Georgia and considerable damage was reported through the state and into the Carolinas (it likely maintained intensity due to baroclinic interaction with a potent mid-level trough and only weakened to a tropical storm when it entered South Carolina). Unlike its HURDAT designation as a restrengthening hurricane after exiting the United States near Cape Henry in Virginia, I believe it was transitioning to an extratropical system and became an extratropical low quickly after exiting southeastern Virginia (pressures reported as low as 950 mbar in the shipping lanes as it passed south of New England and off Canada were likely associated with another extratropical low which absorbed the system). Some great information on this hurricane can be found here in the September 1935 Monthly Weather Review. The MWR has a lot of information and observations that are often not mentioned or are poorly known by most other references. It is an excellent source.

In a nutshell, there are many variables to atmospheric systems and the current data is often inaccurate in the best track information prior to 1975. There are many errors in the current Atlantic best track databases that can skew climatology and our knowledge of tropical cyclones. These issues need to be resolved. In short terms I would take current official data before 1975 with a large grain of salt and put the data in huge perspective. This is why I don't trust Unisys best track data, official NHC information and the deadliest and costliest hurricane data, and climatology.

Good points you make. I read that Hurricane Camille might of bottomed out at 901 millibars and had sustained winds of 205 mph. There was also Typhoon Marge which had sustained winds measured at 215 mph and central pressure of 888 mb. Gilbert had 888 mb and 185 mph winds and it was a monster. Lot of those hurricane datas I see are dubious.