What’s your hottest take on the tropics?

[Hurricanehink]

Once the lid is off and there are proper advisories for Medicanes and South Atlantic storms, I suspect there will be lots more research and find other older storms in their respective basins.

[chaser1]

Wanna really separate the boys from the men (not intended to be sexist, ladies )? I'd like to see Seasonal Forecasts evolve to include a new additional metric. Instead of the current (eg. 13/6/2 ACE: 125), those with the cojones to REALLY put it on the line should be willing to also include a numerical number or score for that season's IRP, or Impact Risk Potential. Whether measured in terms of percentage of population, percentage of named storms projected to impact land, or some broader simple color code seasonal projection suggestive of low, medium, high, extreme numbers of named storms anticipated to threaten population (or landfall) regions.

Such a metric would be less focused on numbers, but perhaps better highlight anticipated risk of impact based on forecast location of tropical genesis (or "hot-spots") and anticipated track pattern forecasts for the upcoming season.

For that matter, I could envision an even more useful application for any given Invest, T.D, T.S. or Hurricane within NHC T.W.O.'s or Storm advisories. Doing so might result in the general public to be more alerted & aware of some new Invest forecast to develop, then some recurving Cat 2 hurricane out in the East or Central Atlantic.

I am skeptical it would be possible to come up with a seasonal forecast model for IRP with any skill at any lead time. The impact level of even an individual storm is dictated by effectively quasi-random events on small timescales, good luck coming up with a model that can show skill at the seasonal level.

For example, if conditions had been a little different, Irma could have hit Miami at close to cat 5 intensity, Andrew could have made a direct hit on Miami or passed over the keys, Dorian could have come ashore in Florida instead of turning north over the Bahamas, Floyd could have been a worse version of Andrew, and if it had moved a bit further north and started its rapid intensification 24 hours earlier, Rita could have slammed into south Florida as a cat 4.

I think you misunderstood my point, or more likely that I probably did not better explain it.

You site many fantastic examples where outcome scenarios could so easily have been different for any number of different circumstances, many of which are often tied to timing. Timing as a result of other weather systems, the speed of any given storms' forward motion, or even the minute impact of a storms short term deepening can result in a temporary stall or decrease in forward motion. Add David from 1979 to the several great examples that you sited. A slight overall deviation to David's storm track could have resulted in South Florida to have received the brunt of the wrath which the Dominican Republic ended up receiving. Simply take Bermuda as an even easier example to illustrate your point. So many prior "near-misses" could have easily ended up as direct landfalls had circumstances for that particular storm - in that particular year been ever so slightly tweaked.

What I understand you to be saying is that such a small margin for error or "game of inches" can in no way be so perfectly predicted weeks or months in advance. I couldn't agree more.

My idea of Impact Risk Potential would have no intention of guessing "who drew the short straw" and ended up most impacted by a tropical cyclone for any given week, month, or Hurricane season. My idea of the IRP for any given point on the map (Houston, Merida, Key West, Bermuda or Atlantic City etc) would simply be more akin to a tropical "traffic report". A good analogy would be like getting bit by a shark. If you and 10 friends go swimming in shark infested waters, the risk factor would be "high" regardless if no one got bit, everyone got bit, or the same one person keeps getting bit every time (THAT dude should just find a new hobby). Well, shark migration patterns can be predicted. Or, changes to shark migration patterns could occur due to water temp, habitat, or food sources. To put it simply, I believe that the science is advancing enough for us to somewhat anticipate where storm genesis is more apt to occur in an upcoming season. I also think that we're beginning to gain an aptitude for anticipating broad scale long wave or potential TUTT patterns, or roughly forecast where we believe greater mid level ridging (and orientation) might set up and how anomalous it may be for the months to come. I firmly believe that global and regional patterns evolve in a manner where one could potentially predict that Jamaica seemingly may have a Low Impact Risk Potential for 2025, however possibly be forecast to have a "high" IRP forecasted for next year just prior to the 2026 Hurricane Season.

Hit or miss, if a higher IRP were forecast for S. Florida for each of the years that Andrew, David, Irma, Floyd or Dorian occured then the IRP forecast for that area would have proved accurate. The difficulty here would be how narrowly or broad can "risk" be fairly accuratly forecast. Kinda like the question of:
At what point does an 80% chance of rain, become a busted forecast for any given area?

[al78]

Wanna really separate the boys from the men (not intended to be sexist, ladies )? I'd like to see Seasonal Forecasts evolve to include a new additional metric. Instead of the current (eg. 13/6/2 ACE: 125), those with the cojones to REALLY put it on the line should be willing to also include a numerical number or score for that season's IRP, or Impact Risk Potential. Whether measured in terms of percentage of population, percentage of named storms projected to impact land, or some broader simple color code seasonal projection suggestive of low, medium, high, extreme numbers of named storms anticipated to threaten population (or landfall) regions.

Such a metric would be less focused on numbers, but perhaps better highlight anticipated risk of impact based on forecast location of tropical genesis (or "hot-spots") and anticipated track pattern forecasts for the upcoming season.

For that matter, I could envision an even more useful application for any given Invest, T.D, T.S. or Hurricane within NHC T.W.O.'s or Storm advisories. Doing so might result in the general public to be more alerted & aware of some new Invest forecast to develop, then some recurving Cat 2 hurricane out in the East or Central Atlantic.

I am skeptical it would be possible to come up with a seasonal forecast model for IRP with any skill at any lead time. The impact level of even an individual storm is dictated by effectively quasi-random events on small timescales, good luck coming up with a model that can show skill at the seasonal level.

For example, if conditions had been a little different, Irma could have hit Miami at close to cat 5 intensity, Andrew could have made a direct hit on Miami or passed over the keys, Dorian could have come ashore in Florida instead of turning north over the Bahamas, Floyd could have been a worse version of Andrew, and if it had moved a bit further north and started its rapid intensification 24 hours earlier, Rita could have slammed into south Florida as a cat 4.

I think you misunderstood my point, or more likely that I probably did not better explain it.

You site many fantastic examples where outcome scenarios could so easily have been different for any number of different circumstances, many of which are often tied to timing. Timing as a result of other weather systems, the speed of any given storms' forward motion, or even the minute impact of a storms short term deepening can result in a temporary stall or decrease in forward motion. Add David from 1979 to the several great examples that you sited. A slight overall deviation to David's storm track could have resulted in South Florida to have received the brunt of the wrath which the Dominican Republic ended up receiving. Simply take Bermuda as an even easier example to illustrate your point. So many prior "near-misses" could have easily ended up as direct landfalls had circumstances for that particular storm - in that particular year been ever so slightly tweaked.

What I understand you to be saying is that such a small margin for error or "game of inches" can in no way be so perfectly predicted weeks or months in advance. I couldn't agree more.

My idea of Impact Risk Potential would have no intention of guessing "who drew the short straw" and ended up most impacted by a tropical cyclone for any given week, month, or Hurricane season. My idea of the IRP for any given point on the map (Houston, Merida, Key West, Bermuda or Atlantic City etc) would simply be more akin to a tropical "traffic report". A good analogy would be like getting bit by a shark. If you and 10 friends go swimming in shark infested waters, the risk factor would be "high" regardless if no one got bit, everyone got bit, or the same one person keeps getting bit every time (THAT dude should just find a new hobby). Well, shark migration patterns can be predicted. Or, changes to shark migration patterns could occur due to water temp, habitat, or food sources. To put it simply, I believe that the science is advancing enough for us to somewhat anticipate where storm genesis is more apt to occur in an upcoming season. I also think that we're beginning to gain an aptitude for anticipating broad scale long wave or potential TUTT patterns, or roughly forecast where we believe greater mid level ridging (and orientation) might set up and how anomalous it may be for the months to come. I firmly believe that global and regional patterns evolve in a manner where one could potentially predict that Jamaica seemingly may have a Low Impact Risk Potential for 2025, however possibly be forecast to have a "high" IRP forecasted for next year just prior to the 2026 Hurricane Season.

Hit or miss, if a higher IRP were forecast for S. Florida for each of the years that Andrew, David, Irma, Floyd or Dorian occured then the IRP forecast for that area would have proved accurate. The difficulty here would be how narrowly or broad can "risk" be fairly accuratly forecast. Kinda like the question of:
At what point does an 80% chance of rain, become a busted forecast for any given area?

Ok I think I see what you are getting at. I am thinking too deterministically. If I have understood what you are suggesting, you are thinking more in line with probabilistic forecasts. The idea is there might be some skill pre-season as to where storms are more likely to form which translates into a higher landfall risk for certain landmasses. If a forecast gives an elevated risk for S Florida, that does not mean a forecast that Miami will get hammered, it means that S Florida has an elevated risk of a landfall, the consequence being the areas of high value real estate around there also have a higher than normal risk of being impacted.

When it comes to probabilistic forecast, they cannot be validated on one outcome alone. The standard way to validate probabilistic forecasts is to acquire a large number of them, categorise them into bins where each bin (for example) is 0-10%, 10-20% etc, then for the observations corresponding to the forecasts, how frequently did the event happen. If a probabilistic forecast is going for an 80% chance of rain, then if you have 100 such forecasts archived, you want around 80 of the corresponding observations recording rain. If the forecast probabilities and the observation frequencies are way off, the forecasts are unreliable.

[kevin]

Mediterranean tropical(-like) cyclones are a lot more common than most people think and should be tracked/researched/archived as TCs. Papers researching them often have way stricter requirements for them than TCs in any other basin. The well-known paper "A long-term climatology of medicanes" found 99 medicanes, but used the following requirements:
*At least 20 mb below background pressure
*At least half of the lifetime of the TC must be over water
*The TC must have at least traversed 200 km during its lifetime
*The storm can not change direction by more than 60 deg during more than half of its time steps
*The storm must have a warm core (Hart 2003 is used) for more than 6 hours or 10% of its lifetime, whichever number is greater
*The storm must have a wind speed of 35+ kt for more than 6 hours or 10% of its lifetime, whichever number is greater
*The wind speed at 850mb must be higher than at 300mb
*The storm must have a max. wind speed of 55+ kt for more than 4 hours
*All requirements (with the exception of the 55+ kt one) must be met for 6+ hours

As I said, with all these requirements they found 99 medicanes for the time period 1948 - 2011. They also tested it keeping all requirements the same except the max. wind speed one which they lowered to 50+ kt. Then they already found 202 storms. With 60+ kt they found 70 medicanes. It also isn't clear what type of wind speed they used, but considering European-based forecasts usually use 10-minute sustained wind speed, I think this is also the case here. In that case, the number of TCs will be even higher when using 1-minute sustained wind speeds. Discarding this last note for now, I can image the number of TCs will increase dramatically if the max. wind speed is adjusted to 35+ kt as is the case in practically any other basin. A season with 5 - 10 TCs of which 1 - 2 reach 60+ kt might be realistic.

Besides the argument of curiosity and completeness of the global database (it's very weird that the European basin of all places is the only one in the world without a consistent historical archive), these Mediterranean tropical storms can also be extremely damaging. Daniel in 2023 killed more than 5000 people and other medicanes such as Numa, Apollo and Ianos resulted in multiple deaths and hundreds of millions or even 1+ billion euros in damages. And those are all storms from the last 10 years alone. I think a lot of deaths and damages in the past were simply undocumented due to the lack of a well-preserved database. In conclusion, I think all of this is more than sufficient data and arguments to indicate that a medicanes should be tracked in a similar manner as TCs are tracked in other basins and that doing so will likely lead to the conclusion that the European backyard basin is a lot more tropically active than most people think.

[chaser1]

I am skeptical it would be possible to come up with a seasonal forecast model for IRP with any skill at any lead time. The impact level of even an individual storm is dictated by effectively quasi-random events on small timescales, good luck coming up with a model that can show skill at the seasonal level.

For example, if conditions had been a little different, Irma could have hit Miami at close to cat 5 intensity, Andrew could have made a direct hit on Miami or passed over the keys, Dorian could have come ashore in Florida instead of turning north over the Bahamas, Floyd could have been a worse version of Andrew, and if it had moved a bit further north and started its rapid intensification 24 hours earlier, Rita could have slammed into south Florida as a cat 4.

I think you misunderstood my point, or more likely that I probably did not better explain it.

You site many fantastic examples where outcome scenarios could so easily have been different for any number of different circumstances, many of which are often tied to timing. Timing as a result of other weather systems, the speed of any given storms' forward motion, or even the minute impact of a storms short term deepening can result in a temporary stall or decrease in forward motion. Add David from 1979 to the several great examples that you sited. A slight overall deviation to David's storm track could have resulted in South Florida to have received the brunt of the wrath which the Dominican Republic ended up receiving. Simply take Bermuda as an even easier example to illustrate your point. So many prior "near-misses" could have easily ended up as direct landfalls had circumstances for that particular storm - in that particular year been ever so slightly tweaked.

What I understand you to be saying is that such a small margin for error or "game of inches" can in no way be so perfectly predicted weeks or months in advance. I couldn't agree more.

My idea of Impact Risk Potential would have no intention of guessing "who drew the short straw" and ended up most impacted by a tropical cyclone for any given week, month, or Hurricane season. My idea of the IRP for any given point on the map (Houston, Merida, Key West, Bermuda or Atlantic City etc) would simply be more akin to a tropical "traffic report". A good analogy would be like getting bit by a shark. If you and 10 friends go swimming in shark infested waters, the risk factor would be "high" regardless if no one got bit, everyone got bit, or the same one person keeps getting bit every time (THAT dude should just find a new hobby). Well, shark migration patterns can be predicted. Or, changes to shark migration patterns could occur due to water temp, habitat, or food sources. To put it simply, I believe that the science is advancing enough for us to somewhat anticipate where storm genesis is more apt to occur in an upcoming season. I also think that we're beginning to gain an aptitude for anticipating broad scale long wave or potential TUTT patterns, or roughly forecast where we believe greater mid level ridging (and orientation) might set up and how anomalous it may be for the months to come. I firmly believe that global and regional patterns evolve in a manner where one could potentially predict that Jamaica seemingly may have a Low Impact Risk Potential for 2025, however possibly be forecast to have a "high" IRP forecasted for next year just prior to the 2026 Hurricane Season.

Hit or miss, if a higher IRP were forecast for S. Florida for each of the years that Andrew, David, Irma, Floyd or Dorian occured then the IRP forecast for that area would have proved accurate. The difficulty here would be how narrowly or broad can "risk" be fairly accuratly forecast. Kinda like the question of:
At what point does an 80% chance of rain, become a busted forecast for any given area?

Ok I think I see what you are getting at. I am thinking too deterministically. If I have understood what you are suggesting, you are thinking more in line with probabilistic forecasts. The idea is there might be some skill pre-season as to where storms are more likely to form which translates into a higher landfall risk for certain landmasses. If a forecast gives an elevated risk for S Florida, that does not mean a forecast that Miami will get hammered, it means that S Florida has an elevated risk of a landfall, the consequence being the areas of high value real estate around there also have a higher than normal risk of being impacted.

When it comes to probabilistic forecast, they cannot be validated on one outcome alone. The standard way to validate probabilistic forecasts is to acquire a large number of them, categorise them into bins where each bin (for example) is 0-10%, 10-20% etc, then for the observations corresponding to the forecasts, how frequently did the event happen. If a probabilistic forecast is going for an 80% chance of rain, then if you have 100 such forecasts archived, you want around 80 of the corresponding observations recording rain. If the forecast probabilities and the observation frequencies are way off, the forecasts are unreliable.

Exactly. In fact, I think the "80%" example used would need to be the minimum benchmark achieved for an overall consensus that some benefit could be garnered from a probabilistic type forecast. In terms of a newly developing tropical cyclone that may well be over 1,000 miles away from most land or population impact, this type of "color-coded" IRP forecast could easily be displayed over an entire basin that would give people everywhere a quick digestible awareness and understanding regarding the eventual risk of needing to make critical preparation or change of plans. Naturally, the IRP factor (or color-code) could be updated once or twice daily, or as often as new storm advisories are issued.

A similar concept applied to pre-season forecasts could be created as well but would be far less beneficial to any one person other then perhaps major event planning, or a tool that some businesses or government agencies might consider for readiness, long term or contingency planning purposes.

[zal0phus]

My other hot hurricane take is people should be making more personified hurricane characters that look like hurricanes (it seems like everyone either makes them into people or gives them two eyes for some reason) but this is more of an art than science gripe

Guilty as charged. Mine are Paleozoic sea-creature humanoids that enter and control hurricanes. The protagonist of my project in the works originated as a personified Hurricane Wilma, though.

[storm_in_a_teacup]

My other hot hurricane take is people should be making more personified hurricane characters that look like hurricanes (it seems like everyone either makes them into people or gives them two eyes for some reason) but this is more of an art than science gripe

Guilty as charged. Mine are Paleozoic sea-creature humanoids that enter and control hurricanes. The protagonist of my project in the works originated as a personified Hurricane Wilma, though.

Not gonna lie, growing up I only really saw personified planets, storms, etc in children’s stuff, where they usually just put a face on the planet or thundercloud etc. So I assumed this was how everyone personified things.

Then I use social media for the first time in 2015 and almost all the personified celestial objects, storms, etc I see are drawn like people.

I never was expecting that. I guess I just don’t like humans enough to want to make everything look like them lol

(I will admit in middle school there was one comic book I sketched where 1900 turned human for a day. But like he hated it lol)

[Fancy1002]

Before I say my obvious hottake, Will the moderators/admin’s remove the post or delete my account if they disagree with it?

[chaser1]

Before I say my obvious hottake, Will the moderators/admin’s remove the post or delete my account if they disagree with it?

Delete your acct?? Doubtful as long as its not a personal attack on other S2K Members or an NHC official, or anything political (however I can't speak for Moderators or Admins). Btw, your "obvious hot-take" isn't obvious lol.

[Fancy1002]

Before I say my obvious hottake, Will the moderators/admin’s remove the post or delete my account if they disagree with it?

Delete your acct?? Doubtful as long as its not a personal attack on other S2K Members or an NHC official, or anything political (however I can't speak for Moderators or Admins). Btw, your "obvious hot-take" isn't obvious lol.

Also, I might be getting this wrong, but the definition of hot take is an opinion that you hold that you think other people are likely to disagree with right, or am I thinking of another term.

[tolakram]

Before I say my obvious hottake, Will the moderators/admin’s remove the post or delete my account if they disagree with it?

If you violate S2K rules the post will be removed and you will receive a PM or warning depending on the severity. It's not too hard to understand, but if you don't then just don't post your hot take.

[mitchell]

Also, I might be getting this wrong, but the definition of hot take is an opinion that you hold that you think other people are likely to disagree with right, or am I thinking of another term.

"Hot take" means an opinion based on a quick analysis and conclusion, as opposed to doing thorough research and reserving your conclusion until after you have gathered all of the facts. Whether you suspect others will disagree or agree is irrelevant.

[storm_in_a_teacup]

Also, I might be getting this wrong, but the definition of hot take is an opinion that you hold that you think other people are likely to disagree with right, or am I thinking of another term.

"Hot take" means an opinion based on a quick analysis and conclusion, as opposed to doing thorough research and reserving your conclusion until after you have gathered all of the facts. Whether you suspect others will disagree or agree is irrelevant.

Really? I always assumed a hot take by definition had to be controversial

[Category5Kaiju]

I find powerhouse hurricanes with bigger eyes more picturesque than hurricanes with pinhole eyes. I’ll take looking at a Melissa, Humberto, or Isabel over an Erin, Milton, or Wilma!!

[AnnularCane]

I find powerhouse hurricanes with bigger eyes more picturesque than hurricanes with pinhole eyes. I’ll take looking at a Melissa, Humberto, or Isabel over an Erin, Milton, or Wilma!!

In Wilma's defense, she got herself a big eye eventually.