Skip to content

SST Trends Since 1900.

April 27, 2015

By Paul Homewood

While we’re on the subject of sea surface temperatures, let’s take a look at the long term changes, as measured by HADSST3.


We can see the step up in 1998, since when things have largely levelled off. What is also noticeable though is the rapid warming from 1920 to 1950, which was a a similar rate to the more recent warming between 1980 and 2000.

As the earlier warming could not have been caused by the small increase in CO2 emissions during those years, clearly some other factor is at play.

Kenneth Richards also points out that Gouretski’s paper, “Near-Surface Ocean Warming Since 1900”, published in 2012, finds that sub surface temperatures (to a depth of 20m) rose more quickly between 1920 and 1950 than in recent decades.


Gouretski also notes that

These maps demonstrate that the first decade of the 21stcentury (2001–2010) was not uniformly warmer than previous decades. Before about 1920, the global ocean was almost everywhere colder than the reference decade of 2001–2010. After 1920, several regions of the global ocean were warmer than in the reference decadeLeaving aside the effect of underwater volcanoes and other seismic activity, the dominant energy source for ocean warming is, of course, the sun. But the ocean’s heat capacity is so huge that the effect of any changes can take decades to appear….

Decadal mean SST and 0–20 m layer anomalies calculated relative to the reference decade 2001–2010 give evidence of the general warming of the global ocean since 1900. However, large regions of the oceans have experienced cooling since the 1990s.

This is hardly the uniformly warming effect that you would expect increased levels of CO2 to produce.

The chart below, from an Indiana University publication, gives a good overview of how ocean currents move warm and cold water around the planet. According to them, a complete run through this current system is estimated to take about 1000 years. 


NASA describe the process whereby the cold, bottom water from Antarctica rise to the surface in the Pacific:

The resulting Common Water, also called Antarctic Circumpolar water, flows northward at depth into the three ocean basins (primarily the Pacific and Indian Oceans). These bottom waters gradually warm and mix with overlying waters as they flow northward. They move to the surface at a rate of only a few meters per year.


In other words, the whole process is an extremely slow one. The idea that an increase in GHG over just a few years can make any measurable difference to these enormously powerful oceanic processes is frankly ridiculous.

  1. April 27, 2015 5:13 pm

    The enormous heat capacity (and cooling ability) of the oceans makes me suspicious of all sudden changes in trend, could these be splicing artifacts between eras of different measurement technologies, rather like what you get when you splice old proxy temperature data onto modern instrumental data, a la IPCC propaganda/marketing brochures?

    Alternatively, could it have been good old-fashioned pollution causing coastal waters (where most measurements are made) to become less transparent, raising surface temperatures because more photons get absorbed there?

  2. April 27, 2015 6:35 pm

    Read the Gouretski paper. He descibes major problems with SST. Inconsistent methods, inadequate coverage, trade route biased…the heroic efforts to fix left me unimpressed.
    The last chart of thermohaline circulation is correct about its slow multicentenial processes. But there are many other ocean surface/near surface porcesses (overturning, upwelling, variation in known currents) that are much faster, timescales of years (ENSO) to decades (PDO, AMOC) that affect aggregate SST. Way to much uncertainty to conclude much beyond natural variation exists, and is inadequately represented in climate models.

  3. April 27, 2015 8:13 pm

    Paul, thanks for these interesting links. The graphic is the best I’ve seen depicting the MOC.

    As ristivan says, it’s the small scale, highly dynamic processes that drive oceanic internal variability. I am trying find out what is the state of knowledge, or conceptual model for this.
    For example this paper discusses many of the challenges:
    Climate Fluctuations of Tropical Coupled Systems—The Role of Ocean Dynamics
    P. CHANG et al 2005

  4. John F. Hultquist permalink
    April 27, 2015 9:30 pm

    As the earlier warming could not have been caused by the small increase in CO2 emissions during those years, clearly some other factor is at play.

    Unrecognized factor or factors at play is a good bet; especially sunlight. That requires either more getting into the water or less energy leaving. Both?

    Now, to the “small increase in CO2 emissions”:
    It is claimed, with good physics support, that the early emissions have the most effect and subsequent emissions have less – recall saturation, log effect, doubling times or any other term that meets with approval. That seems to be related to the atmosphere.
    If there is a connection between CO2 and SST it must require a mechanism worthy of W. Heath Robinson (or ‘Rube’ Goldberg).

  5. Bill Illis permalink
    April 28, 2015 1:29 am

    All the maps that show global ocean circulation patterns, like the one from the University of Indiana, are simplistic illustrations that do not represent what is really happening.

    For example, the Arctic bottom water in the Atlantic actually comes from the central Arctic ocean under the sea ice. This is the coldest, densest ocean water on the planet and it continues to accumulate in the Arctic Ocean basin until it overflows various choke-points in these canyons.

    The Denmark Strait bottom water overflow, for example, flowing 0.0C ocean water at the bottom, becomes the largest waterfall on the planet once the flow gets past the choke-point, falling 3,000 metres.

    The Gulf Stream plays no part in this bottom water flow or formation and the Gulf Stream, at 10C at this location, just continues flowing north well into the Arctic Ocean before it cools off.

    • tom0mason permalink
      April 28, 2015 5:01 am

      Bill Illis,
      Remarkably fine graphics but from where and who?

  6. johnmarshall permalink
    April 28, 2015 10:22 am

    There has been many changes to data collection methods and data accuracy. The Argo data should be the most accurate but mixing data from different instruments of varying accuracy is not good at producing meaningful conclusions.

  7. ulriclyons permalink
    April 29, 2015 7:59 am

    It makes more sense to measure SST trends from cold to cold or warm to warm AMO modes:

  8. Brian H permalink
    May 6, 2015 7:02 pm

    “than quickly than” ???

Comments are closed.

%d bloggers like this: