Curiosities: Mixed Signals in Land and Ocean Temps

A guest post by David Smith (author of “Tiny Tim Storms” on Climate Audit)

The oceans have great capacity to store heat, much greater than the soil and rock of land. We see this often. For instance, daily sea breezes are the result of the different warming rates of the ocean and land. On a seasonal scale, oceans cool more slowly than land in winter and warm more slowly than land in summer – that’s one reason why beaches are popular getaways in July and August.

It seems reasonable that the same principles should apply in a world which is warming due to the greenhouse effect. Land should respond quickly to greenhouse warming while the seas should lag. It also seems reasonable that, on decadal scales, this difference in warming rates should be visible in the global temperature record by now if AGW is driving the climate.

If, on the other hand, the recent warming involves other significant factors, such as changes in ocean behavior in combination with greenhouse warming, then the differential warming pattern might be murky and mixed.

So, what does the record show? Below is RSS satellite-derived data for the lower troposphere, broken into three regions: the tropics, northern extratropical and southern extratropical. The lines represent the spread between land and ocean temperature anomaly in each region – the smaller the spread, the lower the point on the line. (The three lines are scaled so as to separate them on the graph.)

The small scale may make examination difficult so I’ll summarize. The tropics have shown a decrease (tropical land is warming more slowly than the tropical ocean). That surprises me and is not consistent with my understanding of AGW behavior. The southern extratropics is about breakeven and is, in any case, probably hard to measure due to the limited Southern Hemisphere land. 

What about the north? Well, over the three decades of the record the differential trend is upwards. Northern Hemisphere (20N-82.5N) land has warmed faster than the Northern hemisphere oceans. Here is a closeup:

What catches my eye is the odd pattern of flat/ quick increase/flat in the Northern extratropics.

Breaking the graph into two subsets gives this for 1979-1997:

This shows basically flat behavior for 19 years.

And here is 1998-to-today:

This suggests that the Northern extratropical differential is flat-to-declining over the last decade.

Before any howls begin let me state that I clearly am playing with start and end points. I make no claim (or suggestion) that greenhouse warming isn’t the driver behind global warming over the last three decades. I simply don’t know. (Also, I am a “lukewarmer” who thinks that the world is warmer than it would otherwise be due to anthropogenic gases (but doubts that the impact will be extreme)).

What I do suggest is that the patterns are curious. They are not what I expected to find. The patterns open the door wider for the idea that warming since the 1970s has included both natural and anthropogenic factors.

What I’d like is to learn how these patterns fit into the AGW hypothesis. Also, do the GCMs (which reportedly do a good job of replicating recent climate change) show these patterns. The ideal outcome would be to learn what happened in the Northern atmosphere to create these patterns.

As always, I may be making fundamental errors in my simple analysis. If you see such mistakes then please post.


24 thoughts on “Curiosities: Mixed Signals in Land and Ocean Temps

  1. One point is that your recent period of decline in the differential approximately corresponds with the decline in Arctic sea ice. How does RSS differentiate between land and ocean, what does the sea ice count as?

  2. The above graphs are consistent with a hotter sun (via an unconfirmed mechanism that affects the earth’s surface temperature) and less man-made aerosols in the atmosphere. In other words, from the viewpoint from someone standing on the earth in the NH (Asia, for example), the sun is brighter and hotter. However, here’s the kicker. The sun isn’t likely to get hotter (i.e., more active) and man-made pollution isn’t likely to get that much better. Thus, I expect NH land temperatures to start falling in conjuction with SH ocean temps. By the way, it’s my slightly confirmed opinion that air quality has improved substantially since the fall of the Soviet Union and the modernization of China (i.e., less burning of ag waste for heat in the winter). Did you notice that most of last month’s temp anomaly was over central Asia?

  3. “Lukewarmer”: Do we have a new term?

    REPLY: “Luke Skywarmer”, one of Gore’s graduates from the School of AGW that he’s training, i.e. the “army of thousands”?

  4. Pingback: Lukewarmer: New word? | The Blackboard

  5. HADCRUT3/HADSST Jan/Feb/Mar Anomalies

    NH Sea 0.204 0.212 0.209
    SH Sea 0.028 0.095 0.123

    NH Land 0.058 0.304 0.735
    SH Land 0.053 0.071 0.124

    Where the heck did that .735C come from?

  6. Hmmm… When exactly did high energy CRF start declining? Where are clouds most likely to be affected by CRF? And how are ocean surfaces primarily warmed, isn’t by direct sunlight or something?

  7. David,

    It is possible that if you analyze UAH data some of these curious signatures seen may not exist or be a bit different than the RSS data. These same types of differences between each group’s MSU lower tropospheric land data have been shown to be from differences in diurnal corrections to the time series (which you would see if you differenced land and ocean). Not saying it would change the overall outcome, but it would be interesting to see. It is also important to calculate the CI in these types of trends to ensure the uncertainty in trends, especially shorter time periods, is well understood. It is possible (for example) that the slight decrease in the Tropics has a + or – that would indicate a possibility of no trend or a positive trend.


  8. NOAA is the last to produce a monthly climate summary (March 2008 not out yet), but theirs is the most informative.

    The weekly US data shows a strong cooling trend over the last 3 months – 10 out of twelve weeks show a negative anomaly, including all of the last 7 weeks, with the latest week showing the largest negative anomaly.

    The USA temperature data has shown a divergence from the rest of the world (significantly less warming) and this divergence appears to be increasing. Make of it what you will.

  9. Bruce,

    To answer your question (in my opinion) for land temperatures:

    1) Bias in the land temperature record (changes in land use, UHI, etc.)
    2) Less man-made aerosols from Former Soviet Union and rural Asia.
    3) Less of volcanic ash.

    Most of rural Asia (approx. 90% of all Asia) burns all sorts of crap to keep warm during the winter. I’ve seen it myself in northern China. They have corn stover piled higher and wider than the house during winter (it looks like a 2nd addition to the farm house). In more western parts, people burn coal in their homes.

    Would it not have been more reasonable for the IPCC to study the most logical reasons for the land temperature rise before resorting to science-fiction inspired reasons? Is anyone else puzzled why GW is localized mainly over Asia (i.e. the biggest land mass in the NH)? According to March’s RSS imagery, the biggest anomalies occur over northern and western China.

  10. David Smith: I enjoyed your post. I extended your graphs back in time to 1900, using the monthly NCDC ERSSTv.3 temperature anomaly data. Each graph includes the raw monthly data, the same data smoothed with a 13-month running average filter, and a 3rd order polynomial trend.
    Northern hemisphere land minus ocean:

    Tropics land minus ocean:

    Southern hemisphere land minus ocean:

    Over the long term, there doesn’t appear to be anything unusual or anthropogenic.

  11. What do you mean where did the 0.735C come from?

    The satellites actually showed NH land being even warmer. UAH showed Northern Hemisphere land as being 1.01C above the 1979-1998 mean. In fact, it was the third warmest NH land anomaly on record behind only January 2007 and February 1998. The only reason the overall anomaly wasn’t as large is because sea temperatures were actually below the 1979-1998 mean.

  12. “The satellites actually showed NH land being even warmer”

    Not the USA part of the NH. It was -.55C.

    NoExt Land was 1.34C (30deg up)
    NoPol Land was .74C

    Pretty much all of North America was cold. Real cold.

    And Europe/Asia were hot.

  13. Just checked this and cryosphere (ARCTIC AND ANTARCTIC ice UP UP UP) will arctic go above anomaly?)
    and COLA (which has not changed much in the last 6 months except March asia (which was warmer). (climate outlooks).
    So It looks like GISS and HADCRUT may have one H– of a problem in “forcing temps up” for April 2008. Also check the 4.4 -7.5 km height UHA/RSS data here (seems to be diving)
    One consolation prize for the AGW’ers… NH Snow cover is slightly below normal LOL

  14. “This suggests that the Northern extratropical differential is flat-to-declining over the last decade.”

    The same statement can be made about the tropics too.
    But take a look at the Southern extratropical differential. It’s increasing over the last decade. Why are the North and South behaving oppositely?

  15. VG; re: the Cryosphere site; there seems to be a contradiction between the animation showing the Arctic ice-loss over the 2007 year and the comparaitive side-by-side images of 1980 and the present; especially given their quote from Gore and the “major tooth decay” comment.

  16. Thanks for the link to the Grid Map, Bob. It is interesting to see the parts of the world that do not have reliable records – most of Africa, much of Brazil etc.

    It is also very interesting to look at the GISS map page. I recommend setting the smoothing radius to 250km, rather than the default 1200km. Also changing the projection to polar gives a different perspective. Of course neither projection really reflects the fact that 50% of the surface of the globe is between 30S and 30N.

    At the 250km smoothing the maps are mainly coloured grey – showing just how much extrapolation GISS do to come up with their figures.

    I note that on the GISS maps the oceans are nearly all grey (i.e. missing data) yet HadCRU maps have a reading for almost the entire oceans. Does anyone know why that is?

  17. Bob, thanks for the grid too.

    Canada + Greenland 12 million sq km … about half of North America.

    For the most part they have no temperatures recorded.

  18. sfod223: No it does mean that GISS is not as accurate that it should be. Also the current low temperatures does not mean AGW is wrong. Temperatures raise and fall quite rapid in a few months or years. Thats natural variations.

    But things that could proove AGW to be one of mans biggest misstakes are
    – We have no temperature rise since 10 years allthough CO2 emissions are raising faster then ever.
    – For the last 6 years the temperature has been falling.
    – Oceans have not been heating, slight decrease have been measured.
    – The ammount of CO2 that is added to the atmosphere seems to follow global temperature much more than anthropogen emissions.

    Those are things that can not be explained very well within the current AGW theory, at least as far as I know.

  19. Patrick, you probably forgot to select Hadl/Reyn_v2 for the ocean category. It’s defaulted to “none”.It’s in the drop down labeled “ocean”.

    Using 250km, you can really see how much estimating goes on for the 1200km. The U.S and Europe are covered nicely.

  20. Does anyone know of any work done to calculate or measure the amount of energy transferred to the earth’s core by fluctuations in the Sun’s magnetic field? It seems to me the amount of heating in the core from such interactions might be significant.

  21. I note two comments here regarding regional temperature anomalies, a negative one for the US for the last seven weeks and an apparently positive one for Asia (China). So China in growth mode generates more heat and the US in recession generates less heat? Is global warming local? Can regional anomalies be used to estimate economic activity?

Comments are closed.