A new twist on predicting U.S. hurricanes – watch African thunderstorms two weeks in advance

Predicting which African storms will intensify into hurricanes

Tel Aviv University researcher finds hurricanes ravaging US and Canada originate as disturbances in western Africa’s atmosphere

The photograph above was taken by astronaut Alex Gerst on September 8, 2014, from the International Space Station.

The photograph above was taken by astronaut Alex Gerst on September 8, 2014, from the International Space Station. It shows thunderstorms in the western Sahara Desert towering over a raging dust storm.

Hurricanes require moisture, the rotation of the earth, and warm ocean temperatures to grow from a mere atmospheric disturbance into a tropical storm. But where do these storm cells originate, and exactly what makes an atmospheric disturbance amp up full throttle?

A new study published in Geophysical Research Letters by Tel Aviv University’s Prof. Colin Price and his graduate student Naama Reicher of the Department of Geosciences at TAU’s Faculty of Exact Sciences finds most hurricanes over the Atlantic that eventually make landfall in North America actually start as intense thunderstorms in Western Africa.

“85 percent of the most intense hurricanes affecting the U.S. and Canada start off as disturbances in the atmosphere over Western Africa,” says Prof. Price. “We found that the larger the area covered by the disturbances, the higher the chance they would develop into hurricanes only one to two weeks later.”

Watching the clouds gather

Using data covering 2005-2010, Prof. Price analyzed images of cloud cover taken by geostationary satellites, which orbit the Earth at the precise speed of the earth’s rotation and take pictures of cloud cover every 15 minutes. This enabled Prof. Price to track the variability in cloud cover blocking the earth’s surface in West Africa between the months of June and November — hurricane season.

The coverage of clouds acts as an indication of atmospheric disturbances. The more clouds in an area, the larger the disturbance. Using infrared cloud-top temperature data gathered from satellites, Prof. Price assessed the temperatures of the cloud tops, which grow colder the higher they rise. He then compared his cloud data with hurricane statistics — intensity, date of generation, location, and maximum winds –from the same period using the National Hurricane Center data base.

“We first showed that the areal coverage of the cold cloud tops in tropical Africa was a good indicator of the monthly number of atmospheric disturbances — or waves — leaving the west coast of tropical Africa,” said Prof. Price. “The disturbances that developed into tropical storms had a significantly larger area covered by cold cloud tops compared with non-developing waves.”

What makes them special

According to Prof. Price, only 10 percent of the 60 disturbances originating in Africa every year turn into hurricanes. And while there are around 90 hurricanes globally every year, only 10 develop in the Atlantic Ocean.

“We wanted to know what was so special about these 10% of disturbances that develop into hurricanes. Was there something different about these storms at their genesis?” said Prof. Price. “By looking at each of these storms individually, we found again that the larger the cloud coverage originally in West Africa, the higher the value of the accumulated cyclone energy in a future hurricane. The conclusion, then, is that the spatial coverage of thunderstorms in West Africa can foretell the intensity of a hurricane a week later.

“If we can predict a hurricane one or two weeks in advance — the entire lifespan of a hurricane — imagine how much better prepared cities and towns can be to meet these phenomena head on,” Prof. Price says. He is currently examining the thunderstorm clusters around the eyes of hurricanes to study the intensification process of those destructive phenomena.

###

Advertisements

69 thoughts on “A new twist on predicting U.S. hurricanes – watch African thunderstorms two weeks in advance

  1. say what??? It’s been LONG known that most Atlantic hurricanes, and even east Pacific hurricanes, can be traced to African easterly waves, which also produce squall lines as they travel across Africa. NOTHING new. I know a little about this since my PhD dissertation was on the subject.

    • I’m just an interested observer, but this is exactly what I have understood for many years. Every Gulf Coast weather person starts foaming at the mention of “a disturbance off the west coast of Africa”.

      • Anyone who has worked in Academia, especially in the research areas, knows that there is a continual demand for both Masters and Doctoral ‘research topics’. Unfortunately, the definition of ‘novel’ appears to be ‘a subject that the research adviser hasn’t heard of’ so there can be repeated revisiting of research especially when an area of research that was somewhat esoteric becomes the funding flavor of the decade. What it shows is the lack of depth and breadth of knowledge in current academia which may be a symptom of more political activism and chasing after contracts/grants rather than knowledge. So whole departments churn out cookie cutter papers all citing each other (to get the hits) and spend a lot less time reading into subjects and doing detailed library searches. Instead ‘reading in’ and literature search becomes an afternoon on Google Scholar.

    • It seems that major scientific discoveries have been increasing rapidly in the past 20 or 30 years. I guess the trend is being helped by rediscovery someone else’s discoveries. I recommend a literature search to make this trend even stronger. (sarc)

      • This is what happens when billions of Dollars are spent each year on global warming / ‘climate change’ ‘research’.

        Abstract1992
        Christopher W. Landsea and William M. Gray
        Long-Term Variations of Western Sahelian Monsoon Rainfall and Intense U.S. Landfalling Hurricanes
        Western Sahelian rainfall during the primary rainy season of June through September is shown to he significantly associated with concurrent intense U.S. landfalling hurricanes during the last 92 years. The meet intense hurricanes (i.e., Saffir–Simpson Scale Category 3, 4, or 5) have an especially strong relationship with Sahelian rainfall, whereas weaker hurricanes show little or no association. The hurricane-Sahelian rainfall association is most evident along the U.S. East Coast but is negligible in the U.S. Gulf Coast region.
        =============
        Abstract1990
        Strong Association Between West African Rainfall and U.S. Landfall of Intense Hurricanes
        Intense hurricanes occurred much more frequently during the period spanning the late 1940s through the late 1960s than during the 1970s and 1980s, except for 1988 and 1989. Seasonal and multidecadal variations of intense hurricane activity are closely linked to seasonal and multidecadal variations of summer rainfall amounts in the Western Sahel region of West Africa…….

    • Just an amatuer weather follower, but just from the article it doesn’t seem like he is adding much. Maybe his paper is better or maybe he just needed to boost his published paper count.

      • I am in the same boat, but what I already knew about this subject gets a little more defined . If this study is correct (from Tel Aviv i’d say it probably is) the fact that the size of the T-storms and the amount of low temp cloud tops gives us a more exact # of hurricanes that could happen a better forecasting tool. What I missed is HOW BIG in volume and diameter these storms have to be to trip a hurricane? Maybe Roy can provide a bit of info, thanks.

      • Let’s chalk it up at a partially successful attempt to quantify something that has been qualitatively known for decades. It’s one thing to know that some African storms cause Hurricanes. It’s another entirely to predict which one will grow

      • The book I read as Atlantic Hurricanes in the early 1960s and it made the same observations as this article.

    • Yes, this is well known by anyone who takes an interest in following weather than could potentially bring a hurricane to the U.S. Let me quote how most reports start: “We are watching a strong tropical wave that has moved of the coast of Africa…” Usually you will then get a following report of the conditions ahead of this wave as a first observance of the likelihood of this wave developing further, i.e. conditions favorable or unfavorable for development and what those conditions are. They always conclude with “But it bears watching”.
      I suppose somebody got grant money for this study.

    • Just another of recent studies ‘discovering’ such things as the Jetstream and AMO (IIRC some of the excuses the UK Met Office used for recent cold winters).
      The UK free daily paper the Metro used to have a science brief section called ‘No S*** Sherlock’ for ‘discoveries’ such as this.

    • And the conclusion, that greater spacial coverage of clouds within the disturbance increases the chance of it later developing into a cyclone once over the ocean, could probably be more scientifically quantified, don’t you think? Like, did these systems have lower pressures and higher moisture content than the other systems or is it merely the spacial extent of the clouds that matters?
      Next they will correlate ENSO with the chances of Atlantic hurricanes developing and “discover” that there is indeed a correlation.

    • Yes this is all common knowledge but did they look at how many waves… Do not become hurricanes…. and what the differences are between ones that do and do not. This actually would be helpful in determining future hurricanes. Prior to the desperation phase of AGW we could predict annual hurricanes relatively well. Now though, due to the need for headlines we have not hit the head on the nail in a few years. I actually hope this guy is onto something, hurricanes are to unpredictable but it might help.

    • Yes, this has been long known as the authors state in the abstract, “85% of all major Atlantic hurricanes originate as thunderstorm clusters in equatorial Africa,”. That was not the point of the study as I understand it. The point was, as the authors state, “only 10 percent of the 60 disturbances originating in Africa every year turn into hurricanes.” “…what was so special about these 10% of disturbances that develop into hurricanes.” They come up with a correlation between the spatial coverage of cold-top clouds associated with African Easterly Waves and the subsequent intensity of cyclones. It might seem logical that the stronger the wave the greater the cloud extent but has this in fact been quantified before? If so, a reference(s) would be handy.

    • I wouldn’t disparage it too much, there’s never been such temporally accurate data to observe such phenomena before. A cloud cover observation every 15 minutes, along with cloud top temperature measurements (not models), correlated with actual developing hurricanes. It doesn’t seem devoid of value to me.

    • Yeah, I would say I barely even qualify as an interested observer, and even I was aware of this. I didn’t actively seek out this knowledge or anything. It just came to me off-hand in general reading.
      Anyway, good that the science is being advanced, ironic that there is a claim of novelty, neat that Dr. Spencer did his dissertation on something related to this.

  2. Weird, yet another study that stops in 2010, what is it that they do not like about 2011 to 2014?

  3. I honestly thought that the article was sarcasm. I was looking for the /sarc tag. I’m an engineer but I grew up on the Gulf Coast and have known this for years.

  4. Perhaps this is a well known climate scientist with the IPCC or something. Obviously anything published by anyone else would be bad science…

  5. Next up, the impact of SAL on thunderstorm development and hurricane development during the Cape Verde season. sarc/

    • The Cape Verde Season has been known for over two generations.
      Additionally, few Atlantic hurricanes are Cape Verdes. Wikipedia says the average is only two per year. The claim “85 percent of the most intense hurricanes affecting the U.S. and Canada” may be true, but it’s likely due to the construction “most intense” and “affecting the U.S.”

  6. “If we can predict a hurricane one or two weeks in advance… imagine how much better prepared cities and towns can be to meet these phenomena head on,” Prof. Price says.”
    Do they really expect to be able to predict exactly where the hurricane is headed simply by observing atmospheric disturbances in Africa?

    • I can just imagine the goofies on the weather channel warning about a hurricane two weeks in advance, pointing to a 5,000 mile stretch along the Atlantic coast saying there is a chance a hurricane could impact somewhere in here.

  7. How is this going to help cities prepare? Even when they are fully developed in the Atlantic, they can’t say where, or if, it is going to make landfall.

  8. I’m going to start a reseaqrch programme to see if hurricanes produce strong winds and lots of rain.
    Can anyone suggest where I can apply for a grant?

  9. Leaders in cities know if they’re in a high probability hurricane path.
    They are always prepared.
    It’s the stupid people who choose to stay behind.
    They get hurt, die and bitch about how the government didn’t save them.
    Recommendation:
    Remove all warning labels and eliminate safety caps.
    In two generations…problem fixed.

    • New York, New Jersey with Sandy, New Orleans with Katrina had been warned for decades that they were likely to get a hurricane and should prepare. The mayors and governors disregarded these continual warnings and spent their time on such problems as gambling and sizes of soda cups, while approving building in flood plains and beaches to get more property taxes. When the inevitable happened it was ‘unprecedented’ and the destruction of buildings, lives and livelihoods was blamed on the new god Global Warming/Climate Change, as if all of the previous storms in history and the prior warnings had not happened. Politicians are only ever prepared after the event and are experts in blame avoidance.

  10. I was waiting for new information but I didn’t get it in the summation of the paper. This just in, ‘When birds start chirping in the morning it means a sun rise will be accuring shortly.’ I was waiting for the summation of the paper to say something along the lines of ‘so if we can break up these clouds early on we can lessen the intensity of the hurricane.’ My response was going to be, but what about the unintended consequence of breaking up a Hurricane, does that make he next mid size hurricane into a CAT 5? But the author didn’t go there so my question remains moot until some other scientist travels down that road.

  11. I think the significance here is a possible correlation between disturbance size and the overall power and life length of any storm forming from it. I think the whole business of Cape Verde hurricanes forming from African Easterly Wave was probably news only to the University press person who wrote the release. It is necessary for scientists to review any press release prepared on their work to avoid errors and misunderstandings.

  12. I realize what most here have commented, that is the obviousness of these from previous many decades of observations of Atlantic Basin hurricanes, especially since the advent of the satellite era.
    But…..
    We have to keep in mind that researchers grow old, retire, and then die. And in doing so they will take what is in their head with them. What good would have it been for Einstein to work out his theories on paper, fold that paper up, put it into a drawer and die? He had to publish. He had to pass on the knowledge to graduate-level trainees at Princeton, at the German Science academy, etc. Writing and publishing trains critical thinking as anyone who has ever published peer-reviewed science knows. Physicists have their pre-print servers specifically for this purpose (http://arxiv.org/) for new and recent findings and gathering critical comment and learning within the community. Publishing is just a more terminal step in the training process.
    The advancement of human knowledge requires passage of knowledge, insights, and (most importantly) skills (such as mathematics, computational and analytic thought frameworks, etc) from one generation to the next. Publishing by doctoral trainees, post-docs, and young professors is especially critical in maintaining any serious field of science.

    • Thanks Joel I was getting a pretty sad feeling about all the negative comments my self. But now I wonder if this just a troll invasion to make the site look silly. All the years I have followed WUWT i have never seen such low brow comments. No matter the prof went over old territory, gold panners do it all the time and still keep finding nuggets!

      • I must be mistaken but I was lead to understand that funding / publishing of scientific work is for original research, not doing duplication. I apologize to commenters here for my mistake.

        Abstract – 1990
        Strong Association Between West African Rainfall and U.S. Landfall of Intense Hurricanes
        ========
        Abstract – 1992
        Christopher W. Landsea and William M. Gray
        Long-Term Variations of Western Sahelian Monsoon Rainfall and Intense U.S. Landfalling Hurricanes

    • The press release deserves the ridicule.
      “A new study published in Geophysical Research Letters by Tel Aviv University’s Prof. Colin Price and his graduate student Naama Reicher of the Department of Geosciences at TAU’s Faculty of Exact Sciences finds most hurricanes over the Atlantic that eventually make landfall in North America actually start as intense thunderstorms in Western Africa.”
      I suspect that the original researchers were not that thick.
      “85 percent of the most intense hurricanes affecting the U.S. and Canada start off as disturbances in the atmosphere over Western Africa,” says Prof. Price. “We found that the larger the area covered by the disturbances, the higher the chance they would develop into hurricanes only one to two weeks later.”

  13. I think it was Mark Twain who said “Those who can, do, those who can’t, teach, those who can’t teach, administrate”.
    For “administrate” read “write press releases”

    • I applied for such a job but only got to one phone interview. It was all over when I answered that I didn’t play a musical instrument.
      I hope that explains everything.

  14. “If we can predict a hurricane one or two weeks in advance — the entire lifespan of a hurricane — imagine how much better prepared cities and towns can be to meet these phenomena head on,”
    Predicting the formation of an hurricane is one thing, predicting where it will land something else entirely. Human nature and economics mean that precautionary steps will only be taken at the last possible moment, unless of course someone thinks there’s votes in it.

    • “If we can predict a hurricane one or two weeks in advance — the entire lifespan of a hurricane — imagine how much better prepared cities and towns can be to meet these phenomena head on,”
      If the Professor actually said this, and I can only assume he approved the press release, he deserves all the ridicule heaped on him in these comments. Such a comment suggests he has no real knowledge of the behavior of hurricanes – that even a couple of hours before landfall they frequently change course significantly – meaning predictions a week in advance are pretty useless.

  15. It’ll give the people in the cities more time to flap around and cluck about the coming catastrophe.
    You KNOW the politicians will use this advance knowledge to aid themselves. Somehow.
    Probably force everyone into underground caves for the duration. “In an abundance of caution.”

  16. The tag line about how their research is so vitally important to the survival of citizens downwind is trite, self-serving, and little more than patting themselves on the back. Where’s the news story about the hurricane victim that is grateful to be alive thanks to the advance warning from the upwind storm? I suppose it is good to get a week’s notice to put up the plywood shutters and stock up on batteries and water.

  17. I think some of you may be overreacting to the understanding of the unamed journalist who wrote this article about the Price-Reicher paper. (abstract: http://meetingorganizer.copernicus.org/EGU2015/EGU2015-15140.pdf).

    Do West African Thunderstorms Predict the Intensity of Atlantic
    Hurricanes?

    Colin Price (1), Naama Reicher (2), and Yoav Yair (3)
    (1) Tel Aviv University, Israel, (2) Weizmann Institute of Science, Israel, (3) Interdisciplinary Center, Israel
    Since 85% of all major Atlantic hurricanes originate as thunderstorm clusters in equatorial Africa, we have investigated the connection between these African thunderstorms and the consequent development of these disturbances into tropical storms. We have analyzed METEOSAT infrared cloud-top temperature data to determine the areal coverage of cold cloud tops over a six year period from 2005-2010. In addition, hurricane statistics from the same period (intensity, date of generation, location, maximum winds) were obtained from the National Hurricane Center (NHC) data base. We first show that the areal coverage of cold clouds (with brightness temperatures Tb<-50oC)in tropical Africa is a good indicator of the monthly number of African Easterly Waves (AEWs) leaving the west coast of tropical Africa. Furthermore, the AEWs that develop into tropical storms have a significantly larger area covered by cold cloud tops compared with non-developing waves. Finally, we show that on a storm-by-storm basis, the cold cloud coverage in West Africa is positively correlated (r=0.57) with the accumulated cyclone energy (ACE) of the future tropical cyclones that develop out of these waves.

    So Price and Reicher are not claiming priority on discovering that African tropical waves eventually turn into Caribbean storms (as the first sentence of the abstract makes clear). Rather, the paper describes their research in using cloud-top temperatures extracted from satellite imagery (METEOSAT) of these waves to predict the future coverage and intensity of said storms, for which they claim to have found some moderate correlations.

  18. But … but … but, what about all those aerosols surrounding the storm clouds.
    Which is most important to climate?

    • “…what about all those aerosols…?”
      Those are dust storms over the Sahara, but as can be seen above, they are very noticeable even on a planetary scale. Reminds me of Martian dust storms, which are truly a planetary-scale phenomenon:
      http://i2.wp.com/www.universetoday.com/wp-content/uploads/2008/06/duststorms.jpg
      Aerosols are downplayed by the IPCC reports, because they say the particles exhibit highly variable density and individually have relatively short residence time in the atmosphere.
      But aerosols are always present and can absorb and reflect downwelling shortwave (solar) and upwelling longwave (terrestrial) IR radiation and so have complex behavior, cooling and/or warming depending on the height in the atmosphere.
      So aerosols (and water vapor) have a much stronger effect on climate than the IPCC care to admit.

      • @ Johanus, re dust storms traces of Sahara storms and those over China have been both found on the East and the West coast of the USA and Canada respectively. If we had a slightly less dense atmosphere ( IE H20) we might see lot larger effects.

      • The dust storms coming off the Sahara do have an impact on hurricane development. It is now included in the models for hurricane development. After Katrina, the predictions were for more and stronger storms, remember. I told colleagues that that wasn’t going to happen because of the dust being blown into the Atlantic basin, either cooling the water or by blocking the sunlight. This isn’t just blah, blah, blah, it’s historical record. Go back and look. By the way, the iron oxide in the dust gets blown all the way across and kills coral in the Caribbean.

    • Don’t we already know at least one week to ten days or more ahead of time that a potential hurricane is headed in our direction?
      Yes, we have known about these easterly waves from African for decades and can predict their tracks fairly reliably a week or so in advance.
      But this paper claims to have made progress in this research area by finding correlations between the cloud-top temperatures of storms in Africa with the eventual intensity and surface coverage of those storms which later developed into hurricanes.
      Remember this is a PhD candidate, developing and validating a theory for his thesis, so some “hand-waving” can be expected in these “preliminary findings” papers. That’s fine. That’s the nature of scientific research.
      But if this research does pan out successfully, then some day we might be able to predict that a particular Saharan storm, with probability X, will develop into a Category 5 hurricane with a diameter of 1000 km etc. I.e. we can add that capability to the existing capability of predicting the hurricane’s track.

  19. Marq de Villiers had it all described in the history of Hurricane Ivan in his book Windswept in 2006 – this is not news…..
    Windswept: The Story of Wind and Weather was published by Walker in New York and McClelland and Stewart in Toronto in January 2006.

  20. My understanding of the article is that it centers in storms IN AFRICA, not in the Cape Verde Islands, not off the African coast, but actually still in the interior of the continent. Measuring their area of coverage or other details of the storms while still over land may furnish details farther in advance of storm development than would be possible by waiting for the hurricanes to form hundreds or thousands of miles out in the Atlantic. Obviously, I don’t have all the previous research or articles, but some of the comments above did seem to me to misstate or omit the central message of the article, as I read it.

    • How much farther in advance do we need? IF the conditions are right, and IF a hurricane forms near the Cape Verde Islands we in the U.S. have more than a week to watch it. Even at this time and stage there is no telling what will happen as it makes its way across the Atlantic. No one is going to shutter up, evacuate or take any other action this far out. It may not even stay a hurricane or it may not even come close to the U.S. even if it does no one knows at this time where it could make landfall. Usually at this time of year there are tropical systems closer to home to be watched first.

      • No duh.
        Ya know, I recently read that there are a million PhDs in the world now. And those in academia are under the “publish or perish” dictum. Meaning they have to scrape the barrel to find ANYTHING to do a paper on, that no one else has already done. No joke.

  21. I’m a florida lifer. I half remember seeing something about how the Cape Verde islands help kickstart African waves into hurricainable waves during August. September season. I wish I could find that again. This article broke one of two of my Cardinal rules. Don’t waste your money with Israelis studying hurricanes. And never never let Florida state people study arctic anything

  22. Yup another case of scientists announcing the incredible genius of discerning the obvious insights of Farley Mowatt and other astute individuals who write about hurricanes or try like hell to avoid them in this case “What is upstream will end up down stream and the misery all depends on what happens along stream……”

  23. Wait a minute… Tropical storms in CANADA???? Is that a misprint, or are they daffy?
    Also, their math doesn’t add up. “According to Prof. Price, only 10 percent of the 60 disturbances originating in Africa every year turn into hurricanes. And while there are around 90 hurricanes globally every year, only 10 develop in the Atlantic Ocean.
    Am I the only person who calculates 10% of 60 as 6, yet they claim that 10 hurricanes “develop in the Atlantic” every year.
    Since when does 6 equal 10?
    I don’t get it.

  24. Since most hurricanes that develop from Africa are quickly and easily tracked already, And we on the east coast don’t really do anything until it gets much closer and the path is known with more precision. I guess I’m not seeing the need or even potential for more advanced warning based on this study.

Comments are closed.