Gulf Summers Aren’t Global Averages: A Reply to Farron Cousins

Alan Reynolds, Senior Fellow, Cato Institute

DeSmogBlog contributor Farron Cousins writes, “Newsweek Gives Cato Institute Climate Denier A Platform.” That means my piece, “Hurricanes Harvey and Irma Can’t Be Blamed on Global Warming.” But how could I possibly be a “Climate Denier” when I openly accepted NASA’s estimate that “Globally-averaged temperatures in 2016 were 1.78 degrees Fahrenheit . . . warmer than the mid-20th century mean.”

The points I raised are about (1) hurricanes and floods depending on local weather not global climate, and about (2) the 1998-2013 hiatus in ocean temperatures, and about (3) “global temperatures in 2016 [being] majorly influenced by strong El Niño conditions,” to quote NOAA. None of that has anything to do with “denying” that mean global temperatures are higher than they once were. It just questions supposed connections to recent hurricanes.

After taking Climate Change into account, NASA’s estimate of average annual global temperatures is 56-58 degrees Fahrenheit (since it includes winters and polar regions). But a 1 or 2 degree rise in global temperatures around 57.3 degrees seem a doubtful explanation for August’s Gulf of Mexico water temperatures closer to 87 degrees (even though the overall national temperature in August was just 72.0°F).

Once a hurricane is in motion, moving across warm water can help it pick up speed – which doesn’t cause the hurricane but could make it more intense. Roy Jensen finds no historical connection between major hurricane landfalls and Gulf temperature. And the NAOO

In fact, the U.S. had more Category 3, 4 and 5 hurricanes from the 1900s to the 1960s than there were since 1970.


Flooding is a different issue, since heavy rain (unlike storm surge) doesn’t require hurricane winds. Flooding from Hurricane Harvey might have been “a few inches” lower if the Gulf water had been two degrees cooler, as Washington Post writer Jason Samenow estimated. But a few inches are too few to matter.

Cousins includes Sea Surface Temperature (SST), not just surface warming. But the graph he uses from the EPA is only about very long-term global anomalies from 1880 to 2015 –far removed from the actual time and place of recent hurricanes.

The issue is not about whether the Gulf of Mexico in August 2017 was a degree or two warmer than usual (it was), but whether or not that local and seasonal “anomaly” can be plausibly be attributed any long-term warming trend in the yearly averages of global ocean or surface temperatures.

“Like many deniers,” writes Mr. Cousins, [Reynolds] attempts to link the warming waters in the Gulf of Mexico to everything except climate change, including El Nino. While other weather factors do contribute to temperature variations, he ignores the fact that overall temperatures in both the Gulf and the Atlantic Ocean are rising at a steady rate” [time period unspecified].

If Cousins hopes to blame a recent spike in Gulf temperatures on global warming, he can’t change the subject by making cryptic claims about temperatures in just one ocean out of five. The Atlantic doesn’t cover the globe.

I wrote about all oceans, not one, and reported a long hiatus when temperatures did not rise. I cited a 2017 survey of four data sources for ocean temperatures which found that “within 1998–2012. . . the upper 100-m experienced a weak warming (IAP and Ishii) or cooling (EN4-GR10), coincident with the global surface temperature slowdown discussed in recent literature (Xie).”

The prolonged 1998-2013 period of stable or cooling ocean temperatures ended when the world’s oceans and surface were warmed by a weak El Niño in 2014 and very strong one in 2015-2016 pulled unusually warm water up from equatorial Latin America through the Tropical North Atlantic – which includes the Gulf of Mexico. El Niño peaked at 4.3 °F at the end of 2015, persisted through May 2016, and left behind a lot of warm water in 2017.

A study this April, “The extreme El Niño of 2015–2016 and the end of global warming hiatus” by Yale Geophysicists Shineng Hu and Alexey V. Fedorov says, “Our results confirm that weak El Niño activity. . . was the cause of the hiatus, while the rapid [recent] temperature rise is due to atmospheric heat release during 2014–2016 El Niño conditions…” Shang-Ping Xie at Scripps came to similar conclusions in, “What Caused the Global Surface Warming Hiatus of 1998-2013.”

To “contradict” such points (when I made them), Cousins reprints an EPA graph of SST anomalies from 1880 to 2015 which uses one of the four data series I discussed. Too many years strain credulity and eyesight, making it difficult to see what the IPCC called a “global warming hiatus” between 1998 and 2013 (though another hiatus from 1937 to 1976 is visible).

The NOAA graph focuses on year-to-date anomalies from 1998 to 2017, for January to August. The baseline of 1901-2000 is much older and cooler than the EPA baseline of 1971-2000, making deviations appear larger.


Scholars from Yale and Scripps, among others, blame post-2013 warming on El Niño. But Cousins is an El Niño denier. He cites another denier Dana Nuccitelli who says, “The argument is easily debunked. While there was a strong El Niño event in 2015–2016, there was an equally strong event in 1997–1998. . . But according to Nasa, 2016 will be about 0.35°C hotter than 1998.” But drawing a line between two extreme years does not constitute a trend.

To indicate “steadily rising” temperatures the EPA/NOAA anomalies would have to have been steadily rising from 1998 to 2013. But they weren’t. The anomalies peak during El Niños in 1997-1998 and 2014-2016, but otherwise fluctuate up and down around the 1998 level. Warm (relative to 1901-2000), but not warming.

Nuccitelli advises sticking to “global temperatures on the surface where we live, and where temperatures are easiest to measure accurately.” But strained exertions to blame Hurricanes and floods on global warming depend entirely on water temperatures at the time and place of the hurricanes.

Another graph shows Gulf anomalies in red over the past two years. The anomalies were fairly high in August, but significantly higher earlier this year and twice in 2015-16.


Gulf of Mexico Anomalies

To reiterate, my previous blog/article argues that (1) annual ocean temps were not steadily rising from 1998 to 2013; that (2) volatile summer temperatures in the Gulf of Mexico have no apparent connection to annual averages of global temperatures; and that (3) El Niño is a likely suspect for much of the post-2013 warming of both global and Gulf waters.


55 thoughts on “Gulf Summers Aren’t Global Averages: A Reply to Farron Cousins

  1. If it has warmed, no matter what it is. It was caused by global warming(tm).
    If it has cooled, it’s just weather.

  2. A fine response, Alan.

    As to rainfall associated with Hurricanes, Harvey was not unprecedented, except in the location of the rainfall. For comparison, Hurricane Lydia (1963) stalled over Cuba, and dumped more than 100″ in 4 days, and that was after raining 57″ on Haiti.

    • In the year 1935, in the month of November, within the memory of some citizens and recorded in the historical records…

      …It rained so much into the lower San Jacinto, Brazos and Trinity river watersheds that the three rivers burst out of their banks and formed a giant lake, dozens of miles upstream of Galveston Bay. There were no automatic rain gauges, and no one knows with certainty how many inches of water fell on South East Texas. And no satellites to take photos from the heavens

      The Nils and Mellie Esperson building in downtown Houston had water in the first story. never before or since.

      Most of the town’s houses were washed into the ship channel (it took 8 months to dredge it clear….)

      And let us not forget the great Texas floods of July 1938 that devastated Austin, San Marcos, New Braunfels and all of central Texas.

      Or the fantastical rains of 46 and 48 inches in 1978 and 1979… And all the other rain events in between and since. And the punctuating droughts… Weather changes, and changes and changes again. Some larger patterns are evident, like hurricanes in August and September…

      This is Houston, we are on the Gulf, our blessings have a price attached, we’ve been paying that price for 181 years, willingly.

      In the Caribbean and the gulf, and the southeastern Atlantic coast, we know we have to dance with the devil every August and September.

      My family home in Galveston was destroyed in 1961, in Carla… My great grandfather’s brother, Otto, his wife and their 8 children were lost in the 1900 storm. My great grandma and her children, including my grandfather, survived the storm at the Ursuline convent. and we endure, because the blessings are well worth the price.

      And to my friends suffering in Florida, I remind you that 595 years ago, the entire 28 galleons, caravelles and brigantines of the Spanish treasure fleet sank in the keys, in a hurricane, of unknown magnitude. Not new, not then, not now. It is summer in the tropics, and we have to dance with the devil, to pay our dues for our blessings…


  3. Preaching to the population, trying to scare them, has had little or no effect on voters’ and investors’ perception of risk and what truly worries them. The average American looks askance at the wild claims about extreme weather and “climate”, and anyone older than 40 years has been around long enough to remembers “worst evah!” storms from the past (Katrina, Rita, Andrew, …). You can lead a horse to water …

  4. “How could I possibly be a “Climate Denier”? You work for Cato. Don’t try to make it even worse.

  5. I don’t understand this. A rise of .35 of one degree Celsius has these people in a tizzy? A rise of barely a third of a degree occurs and everything is on fire and hurricanes pound the Caribbean because of that? I’m more and more persuaded that these people are living on some plane of existence that denies reality. If you raise the temperature of a room one-third of a degree, Celsius of Fahrenheit, I doubt seriously that they’d notice it.

    Well, okay, this “rise” does not explain the massive July flooding in north central states like Illinois and Indiana, does it? No, it doesn’t. There’s a little video at that link which clearly shows the direction those storms came from and it was NOT the Atlantic or the Caribbean, but rather from the northwest, which is where a lot of our weather originates.

    The fact that the meteorologists have stated that our current hot, humid weather is due to Irma’s influence and Jose’s support of it, blocking the cooler air now providing a chance for snow to fall in Montana, Idaho and Wyoming, completely escapes these people, doesn’t it? Not to worry: the cool air of Autumn will return by the end of next week.

    What are they going to do about the problem with a massive volanic eruption now threatening Bali? It’s forced over 11,000 people to evacuate the vicinity of that volcano. Mount Agung on Bali is part of the Ring of Fire, you see, and while I doubt that most of these Warmians could find Bali anywhere, it’s southeast of Sumatra. It’s a big volcano, it has a long history, and there is nothing that says it can’t spew out enough particulate matter and gases to cloud the atmosphere and drop the global mean temperature by more than a couple of degrees, just like Pinatubo did. But they’ll blame this eruption on global warming, won’t they?

    The more I see of their responses to real data, the more I truly wonder what is wrong with them. There is always an excuse. I guess it’s up to us to keep putting the real data in front of the public and try to deflate the hysterics and emotional hyperbolic drivel as much as we can.

    • That volcano is large enough to impress, but it is sitting in the middle of a gigantic caldera. It could rock more than the neighborhood.

  6. ” Flooding from Hurricane Harvey might have been “a few inches” lower if the Gulf water had been two degrees cooler, as Washington Post writer Jason Samenow estimated.”
    jason just lazy. phil did it right.
    and flooding is something to be dealt with by dam, aqueduct and moving off the flood plain – not stopping rain.

  7. “But a 1 or 2 degree rise in global temperatures around 57.3 degrees seem a doubtful explanation for August’s Gulf of Mexico water temperatures closer to 87 degrees”

    talk about a category mistake.

    AGW theory states that if you increasing forcing ( solar or GHG) the planet will warm ON AVERAGE
    some places will warm more, some will warm less.

    The global temperature rise doesnt EXPLAIN or try to explain a LOCAL warming. The global average “causes” nothing. it is just a metric. the global average is 1-2 degrees higher because the LOCAL areas cause it to rise.

    The case for a warmer SST leading to SLIGHLY MORE intense storms, and storms that intensify more quickly is pretty staright forward.

    IF the sun were to warm the earth more than it does, if the sun were to warm the gulf more than it does now,
    that extra warmth would impact/alter/change hurricanes. FFS remember your butterfly wings.

    the critical point is this, the slight increase in stronger huricanes, may be uncertain. And cutting C02 might not be detectable in that metric. Still, we are ill prepared for the hurricanes of the past, much less the potentially stronger ones in the future. That argues for better adaptation regardless of the science of intensifying hurricanes. That argues we should not spend money subsidizing the continued development in hurricane prone and flood prone areas.

    • Oh Mossshhher the once Great and Powerful, I have only two questions.
      1. What slight increase in stronger hurricanes?
      2. Who is this “we” you speak for who is ill-prepared for hurricanes of the past?

    • Agree with Mosh on this. All sounds reasonable and accurate based on my understanding. This all can be represented as pdf’s (or “fuzzy”, as it was popularly called a few decades ago), and it makes sense that warmer waters, in absence of hurricane killing shearing forces, might tend to shift the mode of the pdf to the more ‘powerful’ side. The primary point of the importance of adaptation seems obvious…in the previous world, the current world, or future worlds.

      • S, what do you say to the propositions that first hurricanes are largely the result of the temperature differential between tropic and poles and second that differential is decreasing as the poles are said to be warming faster than the tropics?

      • …yet as many meteorologists have pointed out, it is the difference between atmospheric T that drives intensity, more then the anomaly of just 1/2 the equation.

        CAGW theory is predicated on that difference being mitigated both latitude wise and surface vs troposphere wise.

        Also, for a time period in the 1990s and 2000s we had an increase in solar isolation at the surface due to cloud cover and a strong solar cycle.
        The disparate ocean residence time of different W/L. isolation is not known, yet this was a likely factor in any ocean warming. We also do not know what ocean T was in The late 1930s early 1940s when the US was hotter then now, when the arctic ice was very low like now, when Greenland was very warm, when we had an extreme lack of measuring T ability in most of the world.

        I agree with Mosher about stopping continued rebuilding with federal funds in areas that are very prone to natural disasters.

    • If the local temperature is slightly warmer because of global warming from CO2, then it is warmer all round as well, meaning the Delta T is no different from before, meaning there is no more energy potential than before. That is what ‘global warming’ means – warmer everywhere.

      • Crispin, see my comment above;
        CAGW theory required LESS T differential, reducing storm and tornado intensity.

    • That sounds sound from Mosher. No problem with his point of view. Especially the fact that average is just a metric. As people at this site emphasize otherwise.

    • Stephen Mosher; as you are an expert in these matters please explain to me the cause of an extratropical cyclone developing into Hurricane Bawbag off the western coast of Scotland in the December of 2011.
      I particular I would be interested in your description of how the ‘warmer’ waters of the northern Atlantic had anything to do with it. Indeed how CO2 had anything to do with it. I am sure you know Bawbag had recorded sustained wind speed over the Scottish central belt, including Scotland well above those recorded by Irma or Maria at any weather station along their routes.
      I do agree with your comments about throwing good money after bad. Lousy infrastructure and buildings in ‘hurricane alley’ will not withstand high winds. Bawbag caused little damage despite its ground speed of over 100 mph in a heavily urbanised area, all power supplies were reinstated within two days, it pays to build good sea defences, underground power networks, and use stone/brick rather than wood/metal. But then the Caribbean islands expect to have their begging bowl filled every 10 years or so.

      • Mr Mosher, I await your response please.
        If you can’t explain Bawbag, I suggest you refrain from posting about hurricanes.

      • I have only one argument with your comment. Your suggestion to build with stone and brick may not be appropriate in seismically active areas. Wood and metal can better withstand the forces of an earthquake and it can be used in ways to better withstand wind, as well. It all depends on wealth and willingness.

    • The Earth gets warmer by storing more energy. The interconnected nature of oceanic/atmospheric systems could explain global warming as being entirely bound by intrinsic factors that result in increased atmospheric CO2 being a bit player riding on coat tails, not a driver. Type, timing, and location of clouds along with water vapor are by far the major players and likely have long term cycles with points along the way being adjusted by oribital mechanics.

      Enjoy the warmth, camp on the beach, and live inland. These warm periods have been and will continue to be the best environment for a productive planet. The cold periods are the ones we will struggle to survive, no matter where we live.

      • Your philosophy is, in my opinion, spot on. I will likely often quote your final statement in the above post. Thank you.

  8. Don’t take the ‘climate denier’ charge too literally. What “they” mean is that anyone who disagrees with even the smallest part of their analysis is automatically defined as a denier.

  9. The thesis of this post, that the NA tropical cyclone basin does not represent globally averaged tropical cyclone activity is an extremely good one. What climate science says about the theoretical effect of AGW on TC formation and intensification by way of rising SST is derived from climate model experiments. What we learn in these model run results is that over the next century, AGW should gradually increase TC intensity and rainfall amounts and decrease the number of TCs that form and that these trends can be seen only in global averages of all seven basins and only at decadal or longer time scales and not in individual basins and not at an annual time scale because of wild variance in year to year tc activity and in individual basins. This means that a single tc season in a single tc basin tells us nothing about the effect of agw on tc activity. I would like to add that I checked the hadsst data to see if sst warming can be related to fossil fuel emissions and found no correlation to support the claim that the observed warming in sst can be attributed to emissions. Here is the link to that result.

    • Jamal, the month by month anomaly analysis is interesting, it is quite striking how different the various monthly anomalies are over time.

      However, I question the use of “sliding trends” and even the idea of “detrending” the data when looking for a correlation in two quantities which are both rising over the period of record and where the fundamental question is the presence of causation in this long term rise.

      A sliding trend is very similar to doing a running mean on the rate of change. Running means are very distorting and also fail to remove a lot of the short term variability as you can see in your panel of 30y trends which are still very noisy at the inter-annual level.

      I suggest you look at this article on the distortions caused by running means:

      The profile of such sliding trends graphs changes enormously as you change the length of the window. Your paper suggests you have looked at this so you must be aware. 30 year windows provide quite linear ramping periods due to the underlying 60y cycle. But this is quite subjective and the profile changes rapidly to other forms with windows only a little longer or shorter than 30y. So what you are doing here is subtracting a distorted, low-passed filtered version of both datasets and then comparing the correlation of the residuals.

      A running mean of 30y will produce maximum distortion at around 20y, that maybe a lot of the reason why you got no rejections of the null hypothesis at that timescale.

      There have been two things which I have railed against in climatology over the years: the use of distorting running averages and detrending of data. You manage to combine the two.

      I would not have any interest in looking at the correlation after such data mangling, since I would not have any faith that it represented anything meaningful about the data or the climate system.

      Apart from the use of the word “generational” , which refers to an approximate and irrelevant feature of the human reproductive cycle, I do not see any explanation of the logic for subtracting the 30y trend. It would seem more logical to low-pass filter the data ( using a non distorting filter ) and then look for correlations in the long term change, rather than trying to remove it.

      Sorry, you seem to have put a fair bit of work into this paper, but I think the underlying method is poorly conceived.

  10. A sports bar in Cambridge, Mass holds a Thursday-night contest modeled a bit on the popular “Jeopardy” program. Patrons are invited to form and name teams, the names the quirkier the better.

    One team last night called themselves “Global Warming Isn’t Real.”

    Remarkably, and in a young crowd comprising lots of university students, the phrase didn’t elicit a single negative response.

    A sign the tide is turning?

  11. The 1998-2013 period had the 1998 El Nino spike in it’s beginning, so the lack of warming in that period is not due to lack of global warming. The rise of 2016 over 1998 is due to global warming. I think that is less than .35 degree C though, and I can’t check HadCRUT4 on my phone without leaving here.

    • “The rise of 2016 over 1998 is due to global warming. ”

      I do not see why the relative heights of two weather events should be taken as a measure of “global warming” even if you mean that without the often implied AGW inference.

      We need a couple of years beyond the El Nino to see where it settles. Maybe they will be permanent offset as there was after the 1998 event. The best thing is to low-pass filter the data to remove inter-annaul changes that are ( questionably ) said to be “caused” by ENSO.

      I think Nino3.4 is a metric of global scale natural variability, not the “cause” of that variability.

    • Donald, I took the .35 degree rise to regret to the difference in SST ONLY in the gulf, not global.
      The global difference between 98 and 2016 is likely within the error margins of measuring GMT methodology and changes and adjustments over the years.
      As I recall the UAH GMT difference was about .125 degrees warmer in 2016.

      Now if for some unscientific reason 100 percent of that change was applied to CAGW, and one further assumes this will continue for 84 more years, then this equates to a year 2100 warming of about 1.5 degrees.

  12. In preparing my “ACE in a hole” article on the relationships of SST and ACE, I looked at both tropical SST and basin wide N. Atlantic SST. The tropical data shows a much less clear correlation than the whole N. Atlantic.

    Both the WWII period ‘plateau’ and the recent one showed a marked drop in ACE , just when SST was at it’s highest. Apparently hurricanes are not as simple as some would like to think.

      • I suspect that wind shear varies too much locally and vertically to be able to do this. Plus, I don’t think we have constant measures of it in many places. And certainly not in the past.

  13. A 1.78°F rise over 70 years is so far within the margin of error for the massaged dataset and shit mathematricks they’re using as to be a meaningless number. Its SMALLER than the error introduced by suspicious UHI and IHI effect on poorly placed or encroached thermometer stations.

  14. Since we have records based on increases in the 20th century it is now time to have a second set of records based on increases in the 21st century to see whether these increases are bigger or smaller than those of the 20th century. The climate change panic was entirely based on the idea of progressive ever greater rates of increase with the stupid name of forcing. This is possibly because if they had honestly called it positive feedback the idea of an unstable climate that this means would have been openly ridiculed by so many other professions with a better handle on real world and less blind faith in poor computer models. This combined with a predicted normal climate as predicted by the climate fraternity which thanks to their primitive method takes no account of the cycles involved and readily detected cycles that coincidentally relate to earth and lunar year cycles interaction times that make a one degree rise normal makes this comparison essential.

    • David says…” The climate change panic was entirely based on the idea of progressive ever greater rates of increase with the stupid name of forcing”

      David, my understanding is forcing and feedbacks both apply, yet clearly the feedback forcing is not manifesting and the primary forcing DECREASES over time. Any direct forced warming caused by going from 280 PPM to 400 PPM will be considerably less over the next 120 PPM increase to 520 PPM.

      As the feedbacks appear to be very overstated, then the KNOWN benefits of CO2 increasing in a linear fashion, while the theoretical harms decrease exponentially, bodes well for humans.

  15. DeSmogBlog contributor Farron Cousins writes, “Newsweek Gives Cato Institute Climate Denier A Platform.” That means my piece, “Hurricanes Harvey and Irma Can’t Be Blamed on Global Warming.” But how could I possibly be a “Climate Denier” when I openly accepted NASA’s estimate that “Globally-averaged temperatures in 2016 were 1.78 degrees Fahrenheit . . . warmer than the mid-20th century mean.”

    Branding someone with the preposterous label “climate denier” when they patently are not one is malicious defamation. The issue is not whether you agree or disagree with NASA, but whether you deny the climate. I think you have an open-and-shut case for libel against DeSmogBlog if you want to pursue it, Alan.

  16. I looked up Cousin’s academic qualifications to see if he was qualified in Climate Science before I went to read his article. I found that the only one he has is in Political ”Science”. I immediately repaired to the smallest room in my house and imagined pictures hidden in the wallpaper whilst I attended to much more serious business.

  17. The question for Farron Cousins is: did global warming also cause the 12 year hiatus from 2005 – 2017 in category 3 or greater hurricanes making US landfall?

  18. I just have a problem with using anomalies especially when the data set being used is from 1850s to 2017. You can make anomalies graphs say a lot of different things just based on the years in the average. Correct me if I am wrong but over that time period SSTs 1850 t0 2017 were most probably NOT collected in the same quantity, quality or using the same methodology. Heck the thermometers in the late 19th century probably did not have the same precision as whatever was used early this century.

Comments are closed.