News Note by Kip Hansen — 25 April 2025
The early predictions for this coming 2025 Atlantic Hurricane Season have come out.
From the group at CSU, we have this:
We anticipate that the 2025 Atlantic basin hurricane season will have above-normal activity. Current La Niña conditions are likely to transition to ENSO neutral conditions in the next couple of months; however, there remains considerable uncertainty as to what the phase of ENSO will be this summer and fall. Sea surface temperatures across the eastern and central Atlantic are generally warmer than normal, but not as warm as they were last year at this time. A warmer-than-normal tropical Atlantic combined with likely ENSO neutral (or potential La Niña) conditions typically provides a more conducive dynamic and thermodynamic environment for hurricane formation and intensification. We anticipate an above-average probability for major hurricanes making landfall along the continental United States coastline and in the Caribbean. As with all hurricane seasons, coastal residents are reminded that it only takes one hurricane making landfall to make it an active season. Thorough preparations should be made every season, regardless of predicted activity.”
As usual, the CSU Tropical Cyclones, Radar, Atmospheric Modeling, and Software Team (TC-RAMS), the team built by the late-great Bill Gray, provides their reasoning and basis of their predictions below the summary on the page linked above.
There is additional data, including deeper historic data, in a StoryMap description of their forecast here: StoryMap
[What’s a StoryMap? See here.]
Important for hurricane season prediction is the state of El Niño/La Niña – here are the model predictions, provided by Columbia University’s International Research Institute for Climate and Society (IRI):
[Author’s Note: Predicting ENSO requires models to use non-linear equations, multiple times, in models, which guarantees chaotic results (chaotic as in Chaos studies). What we see above are the results of using the more-or-less the same input data, observations, fed into non-linear models, producing a wide spread of essentially chaotic results, which are then interpreted (the heavy red and green lines) by averaging those results, which by the end of Oct-Nov-Dec 2025, are spread almost exactly evenly above and below zero. They call result “ENSO Neutral”]
From CSU’s StoryMap:
“PROBABILITIES FOR AT LEAST ONE MAJOR (CATEGORY 3-4-5) HURRICANE LANDFALL ON EACH OF THE FOLLOWING COASTAL AREAS:
- Entire continental U.S. coastline – 51% (average from 1880–2020 is 43%)
- U.S. East Coast Including Florida Peninsula (south and east of Cedar Key, Florida) – 26% (average from 1880–2020 is 21%)
- Gulf Coast from the Florida Panhandle (west and north of Cedar Key, Florida) westward to Brownsville – 33% (average from 1880–2020 is 27%)”
Other predictions:
Atmospheric G2 and The Weather Company have their own forecast:
But where do those landfalling hurricanes strike?
Here’s two interesting graphics I stumbled across, showing every U.S. landfalling hurricane ever recorded, produced by Michael Ferragamo [ story here with the original images ].
[ larger here ]
And just the US portion:
[ larger here ]
There are some interesting blank spots along the Mid-Atlantic Coast including almost all of Virginia, Maryland and Delaware.
And the mainstream media? What is their response to this rather mundane news?
Here’s the headlines folks:
“Forecasters predict another active 2025 Atlantic hurricane season — Colorado State University’s hurricane forecasting team is calling for yet another unusually active season with 17 named storms, nine hurricanes, and four major hurricanes.”
Forecasters Warn of Rough 2025 Hurricane Season: ‘Start Preparing Now’
Meteorologists share grim forecasts ahead of 2025 hurricane season: ‘Everyone needs to start planning and preparing’
Scientists sound the alarm over the looming 2025 hurricane season: ‘An above-average probability’
First Major 2025 Hurricane Season Outlook Released: Slightly Above Average Activity Expected The team at Colorado State University said the season will be somewhat more busy than average due to the lack of El Niño.
Only the Weather Channel makes an attempt to simply deliver the news. CSU expects the 2025 season to be a tad more busy than “average”, but less active than 2024.
And that’s the news….
# # # # #
Author’s Comment:
It is not unexpected to see even these very early hurricane season predictions leading to a bit of overblown alarm in the press. I suspect it would be the same regardless of the prediction – even “below average”.
It is de rigueur for all the media reports about hurricanes to point out that no matter what the prediction, your area being hit by a hurricane is a big deal, dangerous, and all should be prepared. When we were living on our boat in hurricane country, we were always ready to run to a nearby hurricane hole where we could tough it out in relative safety.
The maps supplied by Michael Ferragamo are fascinating.
Thanks for reading.
# # # # #
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Why is the US surface temperature data on the right side of the page still showing February when it is already almost May? The March data should have been in no later than 10 days ago.
ID ==> No idea, but the original data and graph are available at
https://www.ncei.noaa.gov/access/monitoring/national-temperature-index/time-series/anom-tavg/1/0
It took a minute or two to load properly for me. You need to click on “ClimDiv” on the right above the graph to turn off the blue ClimDiv trace. It is current through March 2025.
Thank you, Kip. I appreciate the help. I don’t know why anyone is junking my totally innocuous comment.
I never knew there was a gap in hurricane landfalls from Virginia to New Jersey.
Tom ==> Fascinating, isn’t it? We could say “never hit New Jersey except for “4”in 1903. Of course, those passing close moving north played havoc with the NJ barrier islands and those off of the Delmarva Peninsula.
I had known hurricanes Long Island fairly regularly. I just never knew there was a gap to the south.
Donna in 1960 must have had an interesting track.
It shows her making landfall in North Carolina and also Connecticut!
She must have turned back out to sea.
Question: We know they call it “landfall” where one crosses into land. Is there a name for where one crosses back out to sea?
Here in Virginia, the hurricanes may not landfall, but they sweep through from other states and cause havoc. So you are not escaping hurricanes by moving here.
Virginia sticks out quite aways into the Atlantic so that it shelters coastlines to its immediate north.
Colorado State University hurricane forecasting team? Some kind of Rocky Mountain High deal?
Ron ==> The CSU Hurricane forecasting unit was established and run by the fabulous William “Bill” Gray. It produces the best and most accurate forecasts each year, by far.
DeSmog Blog boosts his reputation here, saying:
“Gray was a regular speaker at the Heartland Institute‘s International Conferences on Climate Change (ICCCs), a venue which gives climate change skeptics from around the world a platform to voice their views. Gray has also described global warming as a “hoax,” and something that “they’ve been brainwashing us [about] for 20 years.”
He appeared in a 2006 video news release sponsored by Tech Central Station titled, “Global Warming and Hurricanes: All Hot Air?” where he denied any connection between the severity of hurricanes and global warming.”
Don’t know why the unit was established in Colorado.
Big EPA and energy labs there. I think the weather satellites are run from there.
They pioneered seasonal hurricane forecasting. I was pretty skeptical about them for a while until I spent the time to read one of their forecasts and even more impressed when reading their post-mortem analyses, especially when the forecast didn’t verify. 1995 was one of those, as that’s when the AMO flipped and last ended the limits brought by the negative AMO. (Other big limiters are El Nino and Saharan dust.)
Bill Gray was very skeptical of CO2’s impact on tropical storms. Please take the 15 minutes to watch http://climateconferences.heartland.org/bill-gray-iccc9-panel-1/ from 2014.
Ric ==> Thanks for the reminder of, and link to, the 2014 Bill Gray presentation.
OK, OK, Bill Gray is the real deal. However, if there was a DOGE style review of how this is in Colorado, instead of SE US somewhere (a lot to be said for looking out your window at weather/climate phenomena), there is probably politics involved.
Ron,
I agree. Hurricane forecasting should best be done by universities in Houston, Florida, Louisiana rather than Colorado or Pennsylvania (no disrespect to Bill Gray’s historical fine work).
Speaking of Pennsylvania, what is Michael Mann predicting for 2025, he sure blew the 2024 prediction…
These are predictions, not guarantees. So residents of hurricane-prone regions should be well-aware what precautions should be taken in the event that one or more materializes in his/her area. . At least we have the advantage of satellite and radar tracking weather technology. One-thing is certain though: residents of hurricane-prone zones aren’t going to make large-scale permanent evacuations to colder climates.
Edward ==> Everyone living in hurricane prone areas knows about the dangers — but many get complacent.
My experience is with the most vulnerable community: those living on boats on hurricane prone islands and coasts. Even some sailors get lazy and don’t take proper precautions — many throw out an extra anchor or two and pretend they’ll be ok….
I predict they’ll only do that once. We went through a TS a few years ago in the Chesapeake that made believers out of us 🙂
I remember that storm. I was crossing the Bay Bridge at mid span when it hit. White knuckle time the whole way.
Ho hum, another gloom prediction, or maybe not too gloomy.
Gotta admit we had a few “outliers” last year, and what happened with Helene was really off the charts, very unusual. Milton could have been much worse, but we came thru that sucker much better than we expected.
Main thing we folks in Hurricane alley do is have our go bags freshly packed, spare propane for the BBQ, and never let gas go below a half a tank.
This is where the climate realists excel as they don’t depend on “renewable” sources of electricity once the wind dies down and you still have a roof.
Hope El Nino take over, as when it does, the Gulf coast seems much less vulnerable. However, it’s about the time that the east coast from Brunswick , GA to New Jersey takes a few hits.
STORM2K website is a great place to learn
Gums sends…
Earth orbital changes from 2024 to 2025 will increase NH April-June solar EMR by 0.8W/m^2. The Oct to Dec solar EMR will be down by 1w?m^2. The area of the NH ocean reaching 30C will increase by about 0.5%. This means July to Sep advection will be higher than it was in 2023 to 2024.
My bet is on a slight increase over average.
Rick ==> And CSU and The Weather Company agree with your overall view.
Thanks Kip. I know folks that live in coastal areas and communicate with them when hurricanes approach. There are a lot of people newly moved into those regions that likely do not comprehend the seriousness. Good luck to all.
I accept how serious strong storms are having sat through a tropical cyclone. I do not try to diminish efforts to cope with them.
But what is the motivation for this pointless exercise?
Nature will do what Nature does.
Where are the real uncertainty estimates of these forecasts?
Can emergency services be better planned in advance because of these forecasts?
Is the math acceptable or not?
Geoff S
Geoff ==> The CSU has a pretty good record — in the mid-80%s. Besides producing news stories, some emergency planning can be done in advance — my Church positions tractor trailer loads of hurricanes recovery supplies — quantities based on season severity predictions. No area can be over-prepared.
In the end, it is a crap shoot.
There are numbers, but not dates. Without dates, planning is purely speculative.
Best to have an emergency kit and keep the car gassed up. Pay attention and skedaddle when the time is right.
For one example, how much supplies to stockpile in advance of hurricane season will depend on how bad you believe the season will be.
I found this interesting. Do they ever produce forecasts for the likelihood of a California tropical system for the year?
https://en.m.wikipedia.org/wiki/List_of_California_hurricanes
johnesm ==> I have never seen a prediction of hurricanes for the California coast — though I grew up there. They are exceedingly rare — only one actual hurricane hit LA in 1939.
Recently a hurricane went up the Gulf of California into Southern Cal and Arizona as a tropical storm.
All eight parameters are forecasted up from the average (-:
Why am I not surprised?
Steve ==> Because it looks like ENSO neutral or shading into La Nina – and general Atlantic SSTs are high-ish? It is physical conditions that determine the likelyhood of hurricanes.
Keep in mind the state of lawfare in this countries.
If the forecasts are too low, lawsuits will follow.
Two quick comments.
First, thinking that in a chaotic system model the average of multiple runs is more likely than any of the individual runs is a misunderstanding of the math. There is no underlying normal distribution as there is in error theory.
Second, average weather is not normal, it is actually rare. There is no such thing as normal weather, so talking about weather as different from “normal” is a misuse of that word.
David ==> Right on both points, though the second relies on narrow thinking of everyday words.
Brown at Duke University called averaging chaotic results “madness” if i recall correctly.
Climate models have generally chosen 1850 as the year of perfect weather.
Trillions have been spent building wind turbines, solar farms, storage batteries and interconnecting all to existing power grids in a vain effort to restore perfect weather on Earth.
When I look around me at 37S I only notice overwhelmingly positive change as my outer suburb has been converted from hard exposed rooftops to expansive green canopy. Our winters not as bitterly cold and our summers a little cooler but more humid.
Kip
I’m in no way, shape or form a meteorologist or weather enthusiast. But I kind of wonder if we’ll see a repeat of 2024 with normal or above normal tropical cyclones in the Western Atlantic, Caribbean, Gulf, but virtually no storms in the Eastern Atlantic coupled with considerable Summer rainfall in the Sahara.
The rainfall in the Sahara sounds like the (purported) climate of 5000-10000 years ago. Might even be an (unexpected) consequence of climate change. Probably not, but it’s not like climate scientists, geographers, or geologists seem to have much if any idea exactly how the distributions of landmasses, mountain ranges, and orbital parameters produced the climates of the past 500 million years.
Don ==> There was rain in 2024 in the Sahara — however, it would take hundreds of years to return the Sahara to a green Pampas climate and environment.
The southern edges of the Sahara are greening, as you know.
My prediction –
This year will probably be much the same as last year, or possibly the year before that.
On a more serious note, I wonder about the usefulness of “14.4 named storms”, or “3.2 major hurricanes”. Maybe someone was trying to appear “scientific” by including pointless numbers.
Either that, or trying to imply that the future can be derived by sufficiently rigorous dissection of the past.
Ferragamo’s graphics are very nice, and point out the need for appropriate preparation. Areas of “gaps” don’t mean that a hurricane won’t strike there in the future. Nature is often capricious.
I don’t consider the fractional numbers pointless. (Umm, they _do_ each have a decimal point. Sorry.)
“3.2 major hurricanes” suggests to me that it’s more likely that there will be 3 major storms than 4. However, the numbers you are griping about are the average for 1991-2020. If there were 94 major storms in that period, what would be more useful to you? “Three?” “Three or four? As few as “X” or as many as “Y”?
Note that CSU is providing integers in their forecast column. I assume that pleases you.
Ric, not griping, as far as I know.
When you say ” . . . what would be more useful to you . . . “, my response would be “neither”. The number of storms in the past has no quantifiable influence on the future, does it?
Predicting the future state of the atmosphere is impossible. If you accept that the atmosphere behaves chaotically, that is. If you don’t, Richard Feynman has demonstrated that future states of the atmosphere are impossible to predict – purely based on quantum physics.
The physicist Niels Bohr said “Prediction is very difficult, especially about the future..
I was just pointing out that averaging something which is counted in integers – major hurricanes, named storms, humans, etc., is even more pointless than averages in general. If you think that the averages you mention have some utility, I would be interested in your reasoning, because I can’t see any at all.
I’m not sure why you write “I assume that pleases you.”, but I thank you for your obvious concern, even though I am mystified about your reasons.
We know that the ENSO cycle influences how many hurricanes there are in a year.
We know that the AMO cycle influences how many hurricanes there are in a year.
Your belief that the future can never be known with infinite precision, so we shouldn’t bother trying to do forecasts is anti-scientific.
Mark,
Anti-scientific? What the heck is that?
Maybe you could try quoting what I sad, rather than putting words in my mouth. I’ll point out that ENSO and AMSO are merely statistics of weather observations, and control precisely nothing.
If you believe that you or anyone else can predict the future state of a chaotic system better than I (or a smart 12 year old), I will accept your challenge. I will bet that neither you nor anyone else can predict the future better than I. Read what I said, not what you think I should have said, then announce your wager. You’ll lose.
Data that includes the statistic “mean” without including “variance” and “standard deviation” is, or should be, suspect. It is not telling the whole story. From MS Copilot …
Together, variance and standard deviation add depth to our understanding of the dataset by describing how consistent or variable the data is around the mean. This is particularly useful when comparing datasets or evaluating the reliability of a mean value.
I agree, but I also accept that this post (and my comments) are subsets of the full paper. Please review https://tropical.colostate.edu/Forecast/2025-04.pdf . There are nine references to standard deviation.
Why do you say “variance” and “standard deviation”? The former is the square of the latter, I would have used “or” instead of “and.”
Personally, I’d recommend the detailed forecasts from CSU instead of MS copilot.
I would have preferred the median, not the mean, not the average.
So, the Doom Criers are out in force trying to terrify people. Typical.
Zero hurricanes here at 37 deg north latitude on the West Coast. Once in a while some dying Mexican storm trash makes it up here and we get lightning and showers in Sept -Oct, but not usually.
So we never get terrorized about hurricane predictions.
We don’t even get terrorized about earthquake predictions because they can’t be predicted. But we do get terrorized when they actually happen.
Remember, the largest risk to your health that you can ever take is to ride in an automobile. So stop doing that if you scare easily.
My driving is not the problem. 😉 Already seeing the TV weather twaffles talking about a bad hurricane season, cue up video of “journalist” being blown sideways by wind and rain as people walk past, drinking a Slurpee and smoking a blunt. Never forget Geraldo in the flood water to his knees and the sound guy standing next to him with water just above his ankles.
My favorite was the news bimbo in the canoe – when the two guys walked behind her in 4″ deep water.
Forgot about her! That was precious.
Kip, can you clarify please…
“Normal” would be in a given range, say “mean +/- 2 SDs”.
Is your forecast within this range ?
If it is, then it is “normal”, not “above normal”
bnice ==> 1) None of these are my forecasts. 2) Each of the forecasting groups uses a long-term average as if that were “Normal” — they should really use the better term “long-term average”. 3) the CSU group specifies “Average for 1991-2020” and does not use the term normal — that’s usually the journalists. 4) With number in ranges 20 and below, and variation narrow, SDs don’t help much, or, maybe better said, wouldn’t improve the understanding of the readers of newspapers.
Thanks Kip. The misuse of “average” which is a single number, and “normal”, which in weather is usually a range, is typical of climate messaging. 😉
Correct me if I’m wrong. I didn’t follow Hurricane forecasts as closely when I lived in the Midwest. But…in my experience I’ve never seen a Hurricane forecast that said that this year is likely to have fewer storms than average.
This seems to be a Lake Wobegon set in which all Hurricane seasons are above average. If I’m wrong, please direct me to the years and numbers.
DonK31: I get your point and don’t entirely disagree. But my frequently faulty memory tells me that Atlantic Hurricane numbers tend to vary with the ENSO index. There are more when the Eastern Pacific is relatively cold (La Nina) and fewer when it is warm (El Nino). Why? quien sabe. I’m pretty sure that “mild” Atlantic Hurricane season forecasts coincide with probable El Nino years.
“I’ve never seen a Hurricane forecast that said that this year is likely to have fewer storms than average.”
I have not seen one, either.
When was the last time the official forecasters forecast a “less than normal” hurricane season?
There were no major hurricane (Cat 3, 4, or 5) strikes on the United States for 12 years, from 2005 to 2017.
What were the official hurricane forecasts for 2006 through 2017?
I don’t know when “official” NHC seasonal forecasts started, but they started with folks like Chris Landsea and other PhD candidates spreading out from CSU and Bill Gray’s tutelage.
If you are willing to accept CSU’s forecasts, see their archive at https://tropical.colostate.edu/archive.html Keep in mind that both CSU and NHC’s forecasts cover the entire Atlantic basin, and that lull in US landfalls had several devastating hits elsewhere, e.g. Irma in 2017 that devastated Puerto Rico, a US territory.
I see that the NHC has forecast verifications at https://www.nhc.noaa.gov/verification/verify3.shtml going back to 1968. However, those old verifications are for forecasts of individual storms during their existence, not seasonal forecasts. Please take a look there and summarize what you find. I guess even the later, far more detailed reports still focus storm track and intensity accuracy.
If you want good seasonal data and analysis, rely on CSU. I don’t bother with NHC’s seasonal forecasts except to compare their numbers (about all that one can find easily) with CSU’s.
“We” made a big deal about the 1995 AMO flip from negative to positive. Negative AMO periods suppress Atlantic tropical storm development. They also correlate with weak periods of Bill Gray’s THC (Thermohaline Circulation).
One of my great annoyances is that interest in the effects of the AMO has steadily waned since then, leading to a remarkable difficulty in finding one of my all-time favorite maps that shows the difference in major hurricane tracks before and after the 1969 flip.
As we are in a period similar to 1945-1969, it takes merely a glance to understand why you may have never seen a below average hurricane forecast since 1995.
From https://tropical.atmos.colostate.edu/Includes/Documents/Publications/grayklotzbach2012.pdf here is the map.
By the way, I’m surprised that no one has pointed out that Michael Ferragamo’s marvelous maps don’t include Superstorm Sandy that hit northern NJ. I assume that’s because it was in the process of undergoing an extratropical transition at the time (it was complicated – I’m willing to call it Superstorm). Sandy went on to bring a lot of snow to West Virginia, another forgotten attribute.
My grandparents had a wonderful old summer house on Long Beach Island off the NJ coast. Dad’s stories of tropical storms there are responsible for my interest in them. While there are very few direct hits, there have been several close brushes.
Please keep in mind that hurricanes north of the Gulf Stream generally weaken quickly and that nor’easters often bring much greater damage. That old house was destroyed in March 1962, a very sad memory. See my https://wattsupwiththat.com/2012/03/06/50-years-ago-the-great-atlantic-storm-of-1962/
I’m sure you know this, but others may not.
Superstorm Sandy was actually the combination of two powerful storms. There was Sandy and there was also a very powerful storm coming in from the west which combined with Sandy over New Jersey and New York.
That’s what made Sandy a superstorm.
I suspect his has more to do with the fact that nobody talks about hurricane forecasts when they are predicting an “average” of below “average” year. So they don’t stick in your memory.
Last year was predicted by the wonderful Disgrace to the Profession Michael Mann to have 33 named storms (outcome 18?). It was an ‘above average’ prediction generally. The wonderment is how can we have ‘Climate Change’ when the most dangerous and measurable metric – cyclones – shows no change for a century? (Cato Institute et al et al ). I don’t think there’s been any climate change. Prove me wrong. Why so few in the SH? Because Mann lives in Pennsylvania
So-called climate science has some examples where various model predictions are made, then averaged to see if that works.
Thank you, Kip, for this illustration where the methodology is applied to storm forecasting.
About a year ago in Quadrant magazine, I wrote about similar questionable math for countries claimed in the scientific literature to be warming faster than the average. My analysis clearly showed that all countries examined, indeed all continents, claimed faster than average warming. That is clearly a mathematical nonsense.
People unaware of the reality of scientific chicanery in climate research need to understand failures as well as claimed successes. Chicanery is grim, it is expanding recently and it should be called out.
Geoff S
https://www.geoffstuff.com/faster.docx
Geoff ==> Thanks for the link to the doc.
It is telling that CSU does not mention the standard deviation in the 30-year averages. Here is their prediction with some context: named storms 17 (30 year average 14.4 ±5.5), ACE 155 (30 year average 123 ±65), hurricanes 9 (30 year average 7.2 ±3.3), and major hurricanes 4 (30 year average 3.2 ±1.9). So all predictions are well within one standard deviation of the 30-year average. Color me unimpressed. Does anyone have data on how their predictions pan out for the last 10-20 years?