The Curious Case of Record August Ocean Temperatures

September 19th, 2014 by Roy W. Spencer, Ph. D.

Several people have noted the apparent mis-match between the NCDC report of all-time record warmth of global average sea surface temperatures in August, and the satellite tropospheric temperatures which are nowhere near a record.

But, as I have cautioned, there tends to be a time lag between SST warming and atmospheric warming…typically 1 month during non-ENSO conditions, and 2-3 months during ENSO. Furthermore, tropospheric temperature variations are somewhat larger than the SST variations that drive them, making direct comparison of the numbers more difficult.

You can get around both of these problems by plotting one versus the other on a graph to see if the latest behavior departs from the normal relationship previously displayed by the two variables (ocean surface temperature and oceanic lower tropospheric temperature).

If you also “connect the dots”, you get what’s called a phase space diagram. If we make such a plot for the 1997-98 super-El Nino, the 2009-10 El Nino, and the current (still weak) El Nino, it looks like this:

Phase space plot of monthly sea surface temperature versus tropospheric temperature anomalies for three El Nino events.

Phase space plot of monthly sea surface temperature versus tropospheric temperature anomalies for three El Nino events (all begin in January, anomalies are relative to 1981-2010 averages).

The time lag of tropospheric temperature behind ocean surface temperature causes a curved trajectory in the data, as I’ve indicated with the light gray line.

What is interesting is that the “record warm” SST month of August, 2014 seems to be an outlier, with the SSTs being too warm (or the tropospheric temperatures too cool) compared to the usual behavior.

Barring some mistake in data processing, the only explanation I have for this is the possibility I blogged about yesterday, that near-record low ocean winds are allowing excessive surface warming while transferring less energy through convection to warm the troposphere. As I also mentioned yesterday, such an excursion would be due to natural variability…not due to “extra” carbon dioxide in the atmosphere, which basically remains unchanged from one month to the next.

31 Responses to “The Curious Case of Record August Ocean Temperatures”

Toggle Trackbacks

  1. ren says:

    Warm air is pushed off north by the waves in the stratosphere. Waves are formed as a result of ozone anomalies in the region of the southern magnetic pole. These anomalies are due to changes in solar activity.

  2. Frank says:

    Roy: To the extent (which may be small) that surface winds are linked to the Hadley circulation, slower winds might be associated with a lower lapse rate in convective regions (or with unusual warmth higher in the atmosphere.

    Why don’t we hear anything about “evaporation feedback”? If the ocean warms 1 degC, the increase in OLR is almost completely negated by the increase in DLR once the nearbly atmosphere has also warmed 1 degC. However, evaporative cooling of its surface increases about 7% – if wind speed and boundary layer humidity remain constant. If the increased evaporation results in more humidity in the boundary layer, more boundary layer clouds may form, turning “evaporative feedback” into cloud feedback. It seems as if the only way the ocean can warm with normal SWR is for wind speed to drop.

    • yes, the negative cloud feedback is a possibility. Cloud changes are difficult to predict, though, because for every additional cloud embedded in ascending air there has to be an equal amount of descending air elsewhere, which is usually cloud-free.

      • Frank says:

        Aren’t boundary layer clouds the result of local turbulent mixing at the interface between the boundary layer and the free atmosphere? If so, your comments about ascending and descending air masses may not apply to boundary layer clouds. If boundary layer clouds are found under the descending branch of the Hadley circulation, for example, then one can have negative cloud feedback even where the free atmosphere is descending.

      • RokShox says:

        “equal amount…elsewhere”

        Won’t the elsewhere be at a higher latitude? Would differences in insolation at the different latitudes tip the balance?

  3. ren says:

    On the animation visible ozone anomaly over Canada, which blocks the jet stream.
    It is associated with the magnetic field.

  4. dave says:

    “…barring some mistake in data processing…”

    In the film “Straw Dogs,” the bored wife of scientist Dustin Hoffmann carefully changes a minus sign to a plus sign in an equation on his blackboard.

  5. Walter Allensworth says:

    Dr. Spencer – I have an arrogant NOAA scientist friend who invalidates your work in satellite temperature measurement because you are a ‘denier.’

    In particular, he does not accept the UAH data record as valid.

    Could you provide some peer-reviewed fodder so I can refute his hand-waving dismissal of your work?

    • Roy Spencer says:

      OMG. Does he accept the RSS satellite record, which came after ours (we invented the AMS and NASA award-winning methodology), which basically gives the same answer as ours? In fact, the RSS record is cited MORE by skeptics because it shows slight cooling since ~1998, while ours has very slight warming.

      We have many peer-reviewed publications, including the best agreement with radiosonde data.

      Sounds to me like the NOAA person is “denying” more than I am. Sheesh. You sure this isn’t a NOAA social scientist? 😉

      • dave says:


        You have a jewel here – someone with perfect bad judgement.
        Flatter him, make him your financial advisor (he will do it for free). Just remember to always do the exact opposite.

        • bernie says:

          “…NOAA scientist…”

          Sounds like a reincarnation of the Florentine astronomer Francesco Sizzi:

          “Moreover, the satellites are invisible to the naked eye, and therefore can have no influence on the earth, and therefore would be useless, and therefore do not exist.”

      • Walter Allensworth says:

        Dr. Spencer – no, RSS is also lumped into the same ‘worthless’ category. When I pointed him to a plot of UAH his comment was:

        “Those aren’t the satellite data. They are the satellite data analysis from a single source who is an acknowledged climate change denier. It is also a data set that has been shown in a National Research Council report to be be unrepresentative of the overall climate system. Care to read the report?”

        I then showed him GISS, which also does not show an August spike, and pause, and he again ignores the last 18 years and fixates on the rise from 1079-1998.

        He only appears to accept the NOAA “blessed” temperature record based on thermometers. I did eventually get him to admit the “similarity” between HADCRUT4, RSS, UAH, GISS, and this was only to support his point that temperatures have been increasing. He ignores/deflects the existence of “the pause,” calling it cherry picking when I suggest that the temperature trend in the last 18 years has not been statistically different from zero.

        So to me, the big question is this jma data he cites (from really the be-all, end-all data set and no other data set is to be trusted? How do I combat this? Is there shortcoming of the jma data set that I should be aware of (that is peer reviewed so I can point him at the review)?

        Thanks so much.

  6. Jim Curtis says:

    I donno. It seems that they all get odd when you get up to a 0.2 to 0.25 deg C SST anomaly. 2009-10 stutter steps and even reverse loops. Over-analysis?

  7. Werner Brozek says:

    Why is there this huge difference between 1998 and 2014? Before I give my answer, I would like to refer you to a post by Bob Tisdale here:
    In this post, Bob Tisdale talks about how warm the north east of the Pacific ocean is at this point in time. And during the 1998 El Nino, it was the tropics that were very warm. Would it make a difference to UAH if warm tropics water rose by 3 C rather than if an equal volume of a much cooler north Pacific rose by 3 C? Due to the law of conservation of energy, if more water evaporates, then more water vapor condenses in the lower troposphere. And it becomes warmer. However if a colder part of the ocean warms up, there should be less evaporation and condensation.
    I believe this explains why the satellite anomalies are not as high this time.

  8. Thanks in favor of sharing such a fastidious opinion, paragraph is fastidious, thats why i have read it

  9. I am regular reader, how are you everybody? This post posted at this web page
    is really pleasant.

  10. ren says:

    Please compare with my previous entries. This also applies in the lower stratosphere.
    When solar activity is low ozone anomalies increases over polar circle, which affects the circulation across the atmosphere.
    Through the wind are formed the ocean surface temperature anomalies. N shift results in a reduction in the number of hurricanes.

  11. ren says:

    Let’s see the current temperature of the oceans.

  12. ren says:

    Here you can see the impact of wind on the temperature of the ocean, and why has not developed El Nio.,-5.92,635

  13. ren says:

    Dr. Spencer please see. This solar cycle 24 will not only be extremely low, but also extremely long.

  14. Bill_W says:

    CO2 is tricksy, Roy. It caused hidey-heat in the ocean, now it is causing hidey-heat in the air!

  15. In periods of extended calm normal wave mixing ceases and the upper 1-2 meters of the ocean can warm by several degrees C.
    At the same time evaporation is reduced by the lack of wind. As soon as the wind resumes the warm surface layer is dispersed by mixing. During such events the absorbed solar energy does not suddenly increase. It is only temporarily confined to the surface and when mixing resumes the ocean temperature quickly returns to close to normal for the season.

    Satellite measurement of SST only measures the actual temperature at the sea surface. Normally this temperature is close to that throughout the upper mixed layer of the ocean above the thermocline which is usually about 100m in depth.

    It is important to recognize that the temporary surface warming associated with periods of calm do not indicate a similar warming of the upper mixed layer.

    A hot sidewalk on a sunny still day does not mean the whole planet is heating up.

  16. D ou g says:

    Actually, Roy, over the long-term it is the temperature at the base of the troposphere (determined by the proven gravito-thermal effect) which determines the temperature of the sea surface, not the other way around. I know you will be surprised to learn this, but it is now a proven fact.

    Sea surface temperatures are a local maximum, it being significantly colder 500m below and 500m above the surface layer. Yet that surface layer is almost completely transparent to solar radiation and completely opaque to radiation from cooler regions of the atmosphere, that radiation being pseudo scattered back into the atmosphere. So it is not radiation which accounts for the temperature of the surface layer of the oceans.

  17. D ou g says:

    You see, Roy, you prove my point when you say in this article “Furthermore, tropospheric temperature variations are somewhat larger than the SST variations that drive them.” In general, large variations in matter with low specific heat drive smaller variations in matter with higher specific heat. That, surely, is intuitively obvious. Look more closely at your time lags. Could it not be the oceans lagging previous variations in the atmosphere? In any event, weather conditions, currents and other factors may lead to a false impression of what lags what.

  18. D ou g says:

    So, Roy, do you still think the Second Law of Thermodynamics is incorrect, and does not tell us that there will be a propensity for the troposphere of any planet to tend towards the state of thermodynamic equilibrium with no unbalanced energy potentials?

    Do you still think, Roy, that you could have more mean gravitational potential energy at the top of the troposphere, and yet the same mean kinetic energy, and thus have unbalanced energy potentials with your own special version of equilibrium, not actually observed on any planet with a significant atmosphere?

    And do you still think the Sun can shine through your window (like the thin surface layer of the oceans) warm yourself and, presumably using the same energy twice, warm the window glass right through?

    And do you still think water vapour warms, rather than causing the mean surface temperature of Earth to be about 10 to 12 degrees cooler than it would be in dry air with a steeper temperature gradient – even though empirical evidence proves you wrong, Roy?

    I’m glad to see the Australian Prime Minister is not attending that climate conference, even though he will be in NY at the time. He’ll just send our female foreign minister, probably because he’s had so many emails from me pointing out the errors in the “fissics” promulgated by the hoaxers like yourself, Roy.

    Now we learn that even the inaudible sound from wind turbines causes serious health hazards, in addition to all the death hazards that will result from the greatest hoax of all time.

  19. D ou g says:

    Perhaps the most compelling evidence supporting what can be deduced from the Second Law of Thermodynamics that downward convection occurs in a planet’s troposphere (in order to restore thermodynamic equilibrium) is the temperature of the ocean surface.

    It is obvious that there must be a transfer of significant amounts of thermal energy from the surface (at, say 18C) in the tropics to the colder floor of the ocean. It is also obvious that this is a one-way downward process, because the temperature differential is always there and the temperature gradient quite steep in the thermocline region.

    So the energy has to escape from the lower tropical oceans and it can only do so in the polar regions where such temperatures (around 4C) are found near the surface.

    But how does the thin transparent surface layer of the ocean receive sufficient energy input in order to both send some energy back to the atmosphere and space by way of radiation, and some at night by non-radiative processes, and also to send some down into its depths in non-polar regions? It cannot possibly acquire all the necessary energy by radiation because it is nothing at all like a black or grey body (being transparent) and most of the solar energy is being absorbed below the surface layer. The back radiation doesn’t even enter the surface layer, and it’s no use saying that it slows radiative cooling, as it does, because how would the surface get to the observed temperature in the first place? Well it does receive the required energy, because it doesn’t cool right down, but the energy into the surface comes from the base of the troposphere by non-radiative convection, diffusion and conduction at the interface.

  20. Nate says:


    Interesting plot. The more recent circles on your plot are shifting up and to the right a significant amount. How do you account for this, other than by a significant warming trend over this period (the last 17 years)?