Resolving the biases in century-scale sea surface temperature measurements reveals some interesting patterns

This new paper in GRL takes on the well-known buckets-vs-inlets issue (Steve McIntyre also visited the issue several times) related to ship based sea surface temperature measurements and as a result, produces an improved dataset. The results show a surprising period of warming, but not in the time period expected.  How would AGW advocates explain that most of the warming in the near surface layers of the ocean came well before Dr. Hansen’s supposed “safe” value of 350ppm of CO2 in Earth’s atmosphere. How would they account for the big rise before 1940?

Even more interesting, if you look at panel (b), (surface temperature in red) you see essentially no change from over 100 years ago. The shape and slope of Panel (a) looks much like the BEST and other surface data up until the mid twentieth century, then post WWII population growth set in. The differences between the two sets post 1980 (when we have the best measurements) is quite stark: 

Here’s the paper:

Consistent near-surface ocean warming since 1900 in two largely independent observing networks

Viktor Gouretski, John Kennedy, Tim Boyer, and Armin Köhl

We compare historical global temperature time series, based on bias-adjusted sea-surface temperatures with independent temperature time series, for the upper 20 meter layer
of the ocean based on the latest update of an historical hydrographic profile data set. Despite the two underlying data sets being different in number of data points, instrumentation and applied adjustments, both of the time series are consistent in showing an overall warming since 1900.

We also extend records of temperature change in the upper 400 m back to 1900. Noting that the geographic coverage is limited prior to 1950, the temperature change in the 0–400 m layer is characterized by two periods of temperature increase between 1900 and 1940–45 and between 1970 and 2003, separated by a period of little change. Citation: Gouretski, V., J. Kennedy, T. Boyer, and A. Köhl (2012), Consistent near-surface ocean
warming since 1900 in two largely independent observing networks,

Geophys. Res. Lett., 39, L19606, doi:10.1029/2012GL052975.

Introduction
Numerous studies have identified an overall rise of the surface temperature of the Earth since the nineteenth century [Smith et al., 2008; Hansen et al., 2010; Morice et al.,
2012]. The global-average surface temperature is estimated from a composite dataset that includes both land- and seasurface temperature (SST) observations. In addition to studies
analyzing surface temperature data, collections of historical hydrographic temperature profiles have been used to estimate the change in heat content of the global oceans
[Levitus et al., 2005, 2009, 2012; Gouretski and Koltermann, 2007].

Two main sources of uncertainty affect both the surface and subsurface time series based on in situ data. The first is related to insufficient data coverage both in space and time, with extremely irregular sampling in the earlier parts of the records. The second arises from instrumental biases which can be comparable in magnitude to real variability in the
climate. Jones and Wigley [2010] identified biases in SST measurements as the most important remaining uncertainty associated with estimating global average temperature change.

Prior to the 1980s, SST measurements were mostly made using buckets or in the engine rooms of ships. Folland and Parker [1995] described systematic errors in SST observations
associated with the use of uninsulated buckets for water sampling and developed adjustments. Uncompensated biases associated with a shift in the database from engine room measurements (relatively warm biased) to bucket measurements (relatively cold biased) occurred at the end of World War II and led to an apparent drop in observed SSTs in late 1945 [Thompson et al., 2008]. More recent studies [Kennedy et al., 2011a, 2011b] attempt to quantify SST biases and their associated uncertainties in the post war period.

However, Kennedy et al. [2011b] note that “Until multiple, independent estimates of SST biases exist, a significant contribution to the total uncertainty will remain unexplored. This remains a key weakness of historical SST analysis”.

Gouretski and Koltermann [2007] revealed significant biases both in the eXpendable BathyThermograph (XBT) and in the Mechanical BathyThermograph (MBT) data used
to measure subsurface ocean temperatures. The effect of this instrumentation problem appeared as an artificial pattern of ocean warming around 1975–1985 in the Levitus et al.
[2005] time series of ocean heat content within the upper 700 meters. Further studies have confirmed the general characteristics of the biases described by Gouretski and
Koltermann [2007] and correction schemes have been developed for both MBT and XBT data [Wijffels et al., 2008; Ishii and Kimoto, 2009; Levitus et al., 2009; Gouretski and
Reseghetti, 2010].

However, Lyman et al. [2010] showed that even in the recent record (1994–2008) the uncertainties of the bias adjustments applied to subsurface data were a major component of the total uncertainty in estimates of ocean heat content. It is often difficult to assess the effectiveness of bias adjustments in reducing the imprint of systematic errors in climate data because independent test data are rarely available. In this analysis an initial approach to resolve this uncertainty is made by comparing two independently derived estimates of near-surface ocean temperature. In addition, a time series of the mean temperature within the upper 400 meters of the world ocean is calculated back to 1900.

[...]

Conclusions
1. The time series of the temperature anomalies within the upper 20-meter and 400-meter layers were extended to the beginning of the twentieth century, although there are
gaps around the two world wars for the 0–400 m layer. Previous estimates started around 1950.
2. A good agreement is observed between the time series based on the sea surface and the near-surface data respectively, but differences suggest either residual uncertainty
of around 0.1C in the adjustments applied to minimize the effects of systematic errors, or actual differences between temperatures at the sea-surface and in the upper 20 meters.

3. The upper 400 meters of the ocean warmed by about 0.3–0.7C since 1910, with a central estimate around 0.5 to 0.6C. The temperature change is characterized by
two periods of stronger temperature increase between 1900 and 1940–45 and between 1970 and 2003, separated by a period of little change in the global average.
4. 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. Whereas cooling in the tropical Eastern Pacific ocean is associated with frequent La
Nina events in the past decade, the cause of the cooling within the Southern Ocean remains unknown.

h/t to Dr. Leif Svalgaard, who has a copy of the paper online here.

About these ads
This entry was posted in ENSO, Sea Surface Temperature and tagged , , , , , . Bookmark the permalink.

59 Responses to Resolving the biases in century-scale sea surface temperature measurements reveals some interesting patterns

  1. Harold Ambler says:

    It needs to be repeated for those staring at these graphs that the zero-anomaly line is essentially meaningless.

    Of course, there are people who that think their being alive on planet Earth confers upon the climate system of the past century a magically deep significance.

  2. beesaman says:

    Maybe the BBC are slowly seeing the light?

    http://www.bbc.co.uk/news/science-environment-19848112

  3. George says:

    From about 1910 to 1940, we saw a significant global warming. This was then followed by a hiatus and then it warmed again to temperatures we saw in the the period around the 1930’s. The extent to which modern land temperatures exceed those of the 1930’s is simply an artifact of “adjustments”. I am not convinced we have seen any global temperatures exceed those of 1934.

  4. George P R says:

    Re George says. Born on 07/05/1926, I can well remember the mid thirties in Southern England and Southern Ireland being hotter than any period since then.

  5. Bob Tisdale says:

    Thanks for the link to the paper, Anthony and Leif.

    Hopefully, the new datasets will be updated to the KNMI Climate Explorer soon.

  6. Bob Tisdale says:

    The conclusion of the paper includes: “However, large regions of the oceans have experienced cooling since the 1990s…”

    And we showed that the sea surface temperatures for 75% of the surface area of the global oceans haven’t warmed in agreement with climate models since 1993 here:
    http://bobtisdale.wordpress.com/2012/09/20/a-blog-memo-to-kevin-trenberth-ncar/
    And in the YouTube video included here:
    http://bobtisdale.wordpress.com/2012/10/03/we-now-control-weather-extreme-heat-events-dirty-weather-climate-disasters/

  7. With all due respect….(love that phrase!)
    With all due respect, I as a matter of course, in the course of my work, measure the temperatures of air, gases and liquids.
    I use highly accurate, professional, calibrated equipment both digital and analog.
    You know what?
    Every time I read someone’s solemn assertion that the ‘Global’ SST in 1820 was…something POINT something degrees, or 1900 ‘Global’ Land temperature was something POINT something degrees…. I am torn between laughing at them and wanting to smack them across the head!

  8. vukcevic says:

    Global Temperature Natural Variability
    An image easy to remember
    http://www.vukcevic.talktalk.net/EarthNV.htm
    Earth has its ‘zing’ originating in the core and the sun has its cycles.
    When two are in phase the oceans absorb more energy, when two are out of phase the oceans cool.
    Jean Dickey of NASA’s Jet Propulsion Laboratory, Pasadena:
    One possibility is the movements of Earth’s core (where Earth’s magnetic field originates) might disturb Earth’s magnetic shielding of charged-particle (i.e., cosmic ray) fluxes that have been hypothesized to affect the formation of clouds. This could affect how much of the sun’s energy is reflected back to space and how much is absorbed by our planet. Other possibilities are that some other core process could be having a more indirect effect on climate, or that an external (e.g. solar) process affects the core and climate simultaneously.

  9. Henry Clark says:

    vukcevic says:
    October 7, 2012 at 12:59 pm
    Jean Dickey of NASA’s Jet Propulsion Laboratory, Pasadena: [...] Other possibilities are that [...] an external (e.g. solar) process affects the core and climate simultaneously.

    I have not gotten around to reading and looking at it in detail yet, nor done quantitative calculations to check plausibility for the timeframes and inertia involved, but, in case you might be interested, the authors of a paper including the following excerpt may be pursuing a partially related path of investigation:

    When the changes of the interplanetary magnetic field induced by solar wind exceed a certain threshold, by virtue of the Faraday law of electromagnetic induction they can become the reason for origin of inductive currents in the Earth liquid core. In other words, when the eccentricity verges toward its local maximum the magnetic flux variations of interplanetary field can cause the considerable variations of the Earth magnetic field, which, in its turn, are capable to induce electric current in the Earth liquid core. Such an additional current, which by virtue of the Le Chatelier law is oppositely directed relative to direction of convection current in the Earth liquid core, can partly disable the convection responsible for the Earth magnetic field generation.

    Such and further discussion is within a paper at

    http://arxiv.org/abs/0803.2766

    where the full-text pdf link is

    http://arxiv.org/pdf/0803.2766v3

    And part 1 of the paper is:

    http://arxiv.org/abs/0803.2765
    http://arxiv.org/pdf/0803.2765v3

  10. Stephen Wilde says:

    It is important to see that the temperature variations are most pronounced in the near surface region and not at the surface.

    That feature suggests that the culprit is the amount of solar energy able to pentrate past the evaporative layer at the ocean surface.

    A warmer atmosphere cannot do that nor can increased so called back radiation.

    So, clearly to me at least, we must look to the amount of solar energy getting into the oceans and that is intimately connected to global cloudiness and albedo.

    The pre 1940 increase on near surface warmth is coincident with solar cycles 17, 18 and 19.

    The subsequent pause coincides with weaker cycle 20.

    The subsequent resumption of warming coincides with stronger cycles 21, 22 and 23.

    Obviously (to me) solar activity affects global cloudiness and albedo and thus the amount of energy entering the oceans to fuel the climate system.

    I have already described the mechanism by which that is achieved.

  11. wayne says:

    Would someone explain the blue Near-Surface plot in 1b? I must be misinterpreting this in some manner that, in the 1920’s, the temperature at that depth was varying about 20°C (±10°C) year-by-year and I assume seasonal variance, yet in the post 2000 temperatures they now only varies about 2°C? Then you look at figure 1c and the 2-sigma uncertainty is only 2°C in the 1920’s? Really? WUWT? The analysis goes on digging deeper from that point on but I can’t without some further explanation of what we’re look at.

  12. Three peaks are visible in HADSST2

    1878, 1944 and 1998.

    66 years apart (or close enough).

    1878 0.000
    1944 0.095
    1998 0.451

    Keep in mind that 2011 = 0.273 barely higher than 1944 or 1878.

    http://www.cru.uea.ac.uk/cru/data/temperature/hadsst2gl.txt

  13. Bob Tisdale says:

    wayne says: “Would someone explain the blue Near-Surface plot in 1b? I must be misinterpreting this in some manner that, in the 1920′s, the temperature at that depth was varying about 20°C (±10°C) year-by-year and I assume seasonal variance, yet in the post 2000 temperatures they now only varies about 2°C?”

    With a global dataset that’s complete, different parts of the globe are responding at different times so they they will cancel or dampen one another. There are fewer observations as you go back in time, and most of those observations were confined to shipping lanes, so there is less and less data from other areas to dampen or oppose those observations.

  14. vukcevic says:

    Henry Clark says: October 7, 2012 at 1:41 pm
    ……
    Mr. Clark
    Thank you for the info, I shall follow it up. I have suspected that the solar-Earth magnetic fields link is very likely, as confirmed by the (inverse) correlation between two, despite vigorous opposition from some experts,
    http://www.vukcevic.talktalk.net/TMC.htm

  15. RACookPE1978 says:

    OK.

    So, do we not HAVE to factor one more “recent change” into the Land-Based Temperature record for the last 50 years?

    If, as CAGW theists maintain, the earth’s net (or “average” or “theoretical-single-worldwide”) temperature anomaly is ONLY a function of net albedo, net total solar radiation in, and net IR (earth-to-space) radiation out …

    We know that the recent increases in CO2 levels have caused every green plant on earth to grow 13% to 27% more rapidly, in more areas, and with wider, thicker branches, stems, and leaves.

    Why do we “assume” that the land (or sea) albedo has remained the same?

  16. stpaulchuck says:

    didn’t the ice cores indicate that the oceans warm up some 800 years (on average) BEFORE the CO2 rises? This would look like it tracks that observation – warm ocean burps up CO2 (and other dissolved gases).

  17. Nick Stokes says:

    “if you look at panel (b), you see essentially no change from over 100 years ago. “

    I think that’s an effect of the axis scale. For near-surface it’s an extraordinary change – from 5°C to 20°C. I agree with wayne – we’d need to know more about that. Bob T suggests that they are measuring in different places. That makes the plot pretty useless – another reason why anomalies are better.

    REPLY: I was referring the surface temperature in red in panel b, updated for clarity – Anthony

  18. Steven Mosher says:

    “Even more interesting, if you look at panel (b), you see essentially no change from over 100 years ago. The shape and slope of Panel (a) looks much like the BEST and other surface data up until the mid twentieth century, then post WWII population growth set in. The differences between the two sets post 1980 (when we have the best measurements) is quite stark: ”

    huh?

    Figure 3 is the chart you want to look at to see the differences between early periods and later periods. relative to the base period of 2000-2010 the first decades are 1.2 C cooler and .76C cooler ( 0-20m and SST respectively )

    Some notes.

    1. They use IDW. There are better methods
    2. More data is a good thing, especially from other sources.
    3. In ICOADS there are just as many MAT (marine air temperature ) records as SST records. There is no reason to combine STT and SAT for a global average. One could combine MAT and SAT for the global air average. But for some weird historical reason Hanson combine SST and SAT for the global average. Just a thought for folks who are interested.. That is to say, looking at the change in MAT from 1950 on will be instructive and may rule out various conjectures and speculations.

    REPLY: I was referring the surface temperature in red in panel b, updated for clarity – Anthony

  19. Steven Mosher says:

    Wilde:

    “Obviously (to me) solar activity affects global cloudiness and albedo and thus the amount of energy entering the oceans to fuel the climate system.”

    Unfortunately, the hypothesis that solar activity “affects” “global” cloudiness, has no well defined empirical data for its support and no working proven physics to support it.

    To support that you will need.

    1. A well defined and audited historical record of “solar activity” whatever that means.
    2. A well defined and audited historical record of “cloudiness” whatever that means.
    3. A physical model ( as in equations ) to test. Mere correlation isnt science.

    It possible, of course, but I’ve yet to see anything that rises above the level of “mannian” science.

  20. tallbloke says:

    vukcevic says:
    October 7, 2012 at 2:14 pm
    I have suspected that the solar-Earth magnetic fields link is very likely, as confirmed by the (inverse) correlation between two, despite vigorous opposition from some experts,
    http://www.vukcevic.talktalk.net/TMC.htm

    Yes, and this is why Leif’s magnetically derived solar series diverges from the sunspot count. The cancelling effect of the opposite phases irons out the apparent solar variation calculated from the deflections of the magic magnetic needle.

    Which is why a number of solar-terrestrial experts don’t accept Leif’s attempted revision of sunspot numbers.

    http://tallbloke.wordpress.com/2011/07/16/does-sunspot-number-calibration-by-the-magnetic-needle-make-sense/

    K. Mursulaa, , , I. Usoskinb and O. Yakovchouka, 1

    aDepartment of Physical Sciences, University of Oulu, Finland
    bSodankylä Geophysical Observatory, University of Oulu, Finland

    Accepted 18 April 2008.
    Available online 10 May 2008.
    Abstract

    It has been suggested recently that early sunspot numbers should be re-calibrated and significantly corrected using the observed daily range of the geomagnetic declination (so-called rY values). The suggested “correction” method makes an a priori detrending of the rY series and then extends the linear regression between rY and sunspot numbers established for the last 25 years to earlier times. The suggested “correction” of sunspot numbers by roughly 30% goes far beyond the traditional estimates of observational uncertainties of sunspots. Concentrating here on Zürich sunspot numbers (Rz), we demonstrate that the rY values do not actually imply that the observed Rz values in the 19th century are systematically underestimated. Rather, we find that the Rz numbers are fairly uniform after mid-19th century. The suggested “correction” is largely induced by the detrending of the rY series, which enhances the rY-based sunspot activity in the 19th century relative to later times. We also show that while the annually averaged declinations have a rough relation between sunspots and other related solar parameters, this relation is strongly seasonally dependent and, therefore, not sufficiently accurate or uniform to allow annually averaged rY values to be used as a very reliable proxy of solar activity in early times.

    The paper can be purchased here: http://www.sciencedirect.com/science/article/pii/S136468260800117X

  21. Steve Short says:

    There is a very interesting fine structure to Southern Ocean (from 30S) atmospheric CO2 levels if one calculates residuals relative to the ‘official’ NOAA global average. This also applies to individual Northern Hemisphere (NH) and Southern Hemisphere (SH) monitoring stations such as Mauna Loa (MLO).

    Over the period 1982 – 2007 CO2 levels at MLO were always greater than the global average. For MLO 1998 was an obvious peak in exceedance of the global CO2 average but despite the fading of the large 1998 El Nino, at least until 2007 the trend for MLO seemed to be for an increasing margin above the global average.

    Conversely, over the period 1982 – 2007 CO2 levels over the Southern Ocean were always lower than the global average. For the Southern Ocean 1998 was not a special year with respect to the negative residuals for CO2 levels i.e. below the global average.

    Over the period 1982 – 2007 the (negative) residual of CO2 levels over the entire Southern Ocean relative to the global mean has trended towards greater (negative) values. In other words, over 1982 – 2007, CO2 levels over the surface of entire Southern Ocean (below 30S) have slowly lagged increasingly below the (rising) global average surface CO2, falling from about 0.35% below the global average to above 0.55% below it in recent years – a counter trend of about -0.1%/decade against the (always rising) global average CO2 level.

    Similarly, the CO2 residual below the global average over the Southern Ocean has now approached close to the long term average residual at the Easter Island Station (EIC), which has always typically lagged about 0.65% below the global average CO2 level since records (at EIC) commenced in 1994. The Easter Island Station is located near the center of the Southwestern Pacific Gyre – a region of high cyanobacterial productivity and a well-known (and studied) high level of persistent cloudiness.

    As I see it, the only logical explanation for this trend is that cyanobacterial productivity (which abstracts CO2 and bicarbonate from surface waters) over the entire Southern Ocean is steadily increasing (relative to the past and relative to the global oceanic productivity average). It is likely therefore that the degree, and persistence, of cloud cover over the entire Southern Ocean is increasing in response to increasing atmospheric CO2. Hence it may be inferred that global albedo is also increasing as a biogenic negative feedback to AGW.

    If someone could tell me how to post a graph I can show them the data upon which the above statements are based. All data is from NOAA. I would be happy to post it to e.g. YouSendiT for anyone who emails me. FYI, I have not seen any peer-reviewed paper identifying and discussing this long term, and continuing, trend despite careful and periodic seaching over the last 5 years.

  22. Mike Smith says:

    If the ocean near surface temps really increased by 15 degrees Celsius between 1900 and 1940 (as suggested by panel b) we do indeed have reason to be quite alarmed.

  23. Billy Liar says:

    beesaman says:
    October 7, 2012 at 11:44 am

    Maybe the BBC are slowly seeing the light?

    http://www.bbc.co.uk/news/science-environment-19848112

    Exactly the opposite of what the same outfit (the University of Reading Dept of We Haven’t a Clue) were saying 5 years ago:

    Dr Peter Stott, a climate scientist at the University Of Reading and co-author of the paper, said: “The paper is saying there is a significant human influence on global rainfall patterns and this includes an increase of precipitation north of 50 degrees northern latitude, an area that includes the UK.
    “In the UK wetter winters are expected which will lead to more extreme rainfall, whereas summers are expected to get drier.

    Source:
    http://www.telegraph.co.uk/science/science-news/3301412/Man-made-global-warming-increases-rainfall.html

  24. P. Solar says:

    Steve Short says: If someone could tell me how to post a graph I can show them the data upon which the above statements are based.

    I have never been able to get WP not to screw up any attempt to get an image included in a post. Maybe that is intentional. All I can suggest is putting the graph somewhere else, like tinypic.com , for example, and then you just add the URL and WP makes it into a link automatically.

    What you describe looks interesting, please post links to data too.

  25. murrayv says:

    By stretching the charts verticall one can make several interesting onservations. Trend lines eyeballed in Chart a). Coolng from 18xx to about 1909/10, then warming to ca 1945, cooling to 1976, warming to 2005. Curiously there seems to be a warm plateau from about 1940 to 1955 and from 1997 to 2012. The real post peak cooling should be about to start now. Total warming from 1910 to 2005 (1.5 cycles) is about 1 degree C. Warming from 1900 to 2000 is 0.6 degrees C. 2005 peak is about 0.3 degrees C warmer than 1945 peak. (I suspect there is still a little warming bias). Eyeballed warming trend 1910 to 1945 is almost exactly parallel with 1976 to 2005.
    One century is about 10 years longer than 1.5 cycles, so does not give the total 1.5 cycle warming of 1 degree C. 100 year warming is between 0.6 and 0.9 degrees C/century depending on start and end dates chosen.
    Chart b) seems to show warming of 4 degrees C valley to peak, but does not have tenths of degree scale. These 2 charts need to be reconciled.

    [Consider copying this and re-editing it into separate lines: It will be easier to follow. Mod]

  26. Paul Vaughan says:

    Aggressive suggestion for Mosher, Tallbloke, Wilde, Vukcevic, Clark, & others:

    Pay more respect & attention to the Russian luminaries:

    Sidorenkov, N.S.; Lutsenko, O.V.; Bryazgin, N.N.; Aleksandrov, E.I.; & Zakharov, V.G. (2005). Variation of the mass of the ice sheet of Antarctica and instability of the Earth’s rotation. Russian Meteorology and Hydrology 8, 1-8.

    The Russians are well ahead of NASA JPL on this particular file:

    Solar Driven Geomagnetic Jerks:
    http://i48.tinypic.com/4na4n.png

    Glaciologists:
    Important: See Sidorenkov’s conclusions. This is the opportunity of a lifetime to make a major contribution to human understanding of multidecadal climate.

    ——–
    Vukcevic, do they [ http://www.vukcevic.talktalk.net/TMC.htm ] use TSI as an input to their field models? (I wrote to ETH Zurich requesting basic info. The response: No reply despite the passage of many weeks …and no wonder, given the tremendously counterproductive chill climategate put on private climate-related correspondence with academia. Suggestion for everyone: Do cost/benefit, not just benefit when accounting.)

  27. pat says:

    Oh oh. There is no loss heat. There is no heat storage level in the sea. There obviously is no atmospheric heat retention (it would be apparent at the surface). If there is any global increase in temperature it is entirely transitory.
    Time for a study of earth temperatures. Literally. A pattern of thermometers in virgin soil about 18″ deep. Remotely sensed for about a decade or so.

  28. RACookPE1978 says:

    pat says:
    October 7, 2012 at 4:58 pm

    Oh oh. There is no loss heat. There is no heat storage level in the sea. There obviously is no atmospheric heat retention (it would be apparent at the surface). If there is any global increase in temperature it is entirely transitory.
    Time for a study of earth temperatures. Literally. A pattern of thermometers in virgin soil about 18″ deep. Remotely sensed for about a decade or so.

    Weeelllllllllll ….

    Lettuce sea.

    How about a long-term 4 inch soil temperature?

    When I last lived in Idaho, near the Idaho National Engineering Lab (nuke testing and reactor facility in the high lava deserts ..) the radio reported “the daily 4 inch soil temperature” every hour for the potato farmers.

    Too cold? Can’t plow for planting.
    Too hot? Need irrigation water.
    Not cold enough? Can’t dig ‘em up and harvest yet.

    Those guys ought to have some 100+ odd years of continuous daily temperatures in what can ONLY be described as “thoroughly non-urban” non-hotspots.

  29. P. Solar says:

    It was argued here: judithcurry.com/2012/03/15/on-the-adjustments-to-the-hadsst3-data-set-2
    that hadSST3 was removing most of the variability from most of the SST record on the basis of speculative bias corrections.

    This new assessment of near surface temps seems to leave that variation in. This probably accounts for the significant difference in pre-1940 sections of their graphs.

    It is unfortunate that they did not go back to 1880. There is a significant downward trend of which we just see the end here during 1900-1910. This was also played down in hadSST3, it would be interesting to see how the ‘near surface’ data shows this period.

    When you see two downward sections and two upward sections, it’s hard not to realise that there is a (majoritarily) strong cyclically varying component in the record.

    One thing I cannot understand in the graphs presented here is that absolute temps shown in figure 1b both seem to show a clear downward trend since 2000, however, this is not apparent in 1a showing the ‘anomalies’.

    Now, as I understand anomalies to be calculated , that should not happen. Can anyone explain why we do not see the same downward trend in the anomaly plot?

  30. pat says:

    RACookPE1978 says:
    Re 4″ continuous temperature record of upper soil.
    Wow. I was totally unaware such a body of knowledge existed. Even if the soil itself were degraded (i envisioned short horizontal tunneling) the body of data would be entirely consistent for an analysis of true atmospheric temperature change in the region.

  31. silver account says:

    These sea surface temperature maps are based on observations by the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Aqua satellite. The satellite measures the temperature of the top millimeter of the ocean surface. In this map, the coolest waters appear in blue (approximately -2 degrees Celsius), and the warmest temperatures appear in pink-yellow (45 degrees Celsius). Landmasses and the large area of sea ice around Antarctica appear in shades of gray, indicating no data were collected.

  32. LazyTeenager says:

    Charles Gerard Nelson on October 7, 2012 at 12:55 pm
    I use highly accurate, professional, calibrated equipment both digital and analog. I am torn between laughing at them and wanting to smack them across the head!
    ———-
    Yeah, I have the same reaction when someone demonstrates they don’t understand averages.

  33. Gunga Din says:

    Charles Gerard Nelson says:
    October 7, 2012 at 12:55 pm
    With all due respect….(love that phrase!)
    With all due respect, I as a matter of course, in the course of my work, measure the temperatures of air, gases and liquids.
    I use highly accurate, professional, calibrated equipment both digital and analog.
    You know what?
    Every time I read someone’s solemn assertion that the ‘Global’ SST in 1820 was…something POINT something degrees, or 1900 ‘Global’ Land temperature was something POINT something degrees…. I am torn between laughing at them and wanting to smack them across the head!
    ==================================================================
    “Significant digits” is something I wonder about when global temperature is talked about. Especially average (Hi, LT!) surface temperatures in light of Watts et al. Just how significant are any of those “digits”? Significant enough to base our economy on them?

  34. David Ball says:

    murrayv says:
    October 7, 2012 at 4:50 pm
    Good post. Thank you.

  35. David Ball says:

    LazyTeenager says:
    October 7, 2012 at 6:04 pm
    “Yeah, I have the same reaction when someone demonstrates they don’t understand averages.”

    Ok LT, now I am going to ask you to demonstrate your understanding,……

  36. Willis Eschenbach says:

    Nick Stokes says:
    October 7, 2012 at 2:59 pm

    “if you look at panel (b), you see essentially no change from over 100 years ago. “

    I think that’s an effect of the axis scale. For near-surface it’s an extraordinary change – from 5°C to 20°C. I agree with wayne – we’d need to know more about that. Bob T suggests that they are measuring in different places. That makes the plot pretty useless – another reason why anomalies are better.

    REPLY: I was referring the surface temperature in red in panel b, updated for clarity – Anthony

    I understand what you mean, Anthony. In Figure 1 a), which shows anomalies, there is a rise in the anomaly in fits and starts from beginning to end. The most recent period is the warmest.

    But in Figure 1 b), which shows the actual average temperatures, the surface temperature peaks in the 1940s and decreases since then … I’ve put a ruler over Figure 1 b) to show what I mean.

    Let me suggest an explanation. This is an artifact of the varying locations of the samples as you go back in time. Which brings up its own set of problems, but that’s another story.

    w.

  37. phlogiston says:

    Paul Vaughan says:
    October 7, 2012 at 4:55 pm
    Aggressive suggestion for Mosher, Tallbloke, Wilde, Vukcevic, Clark, & others:

    Who are you calling a Solar Driven Geomagnetic Jerk?

  38. phlogiston says:

    Slightly OT but the BoM Monthly Subsurface Pacific Ocean Equatorial Temperature Anomalies down to 400 Meters shows significant cooling from west to east across the Pacific.

    http://www.bom.gov.au/climate/enso/sub_surf_mon.gif

    The absence of east-west asymmetry that normally drives ENSO does suggest a period of neutrality, although some major new upwelling could possibly push the system toward yet another La Nina.

  39. tallbloke says:

    Willis Eschenbach says:
    October 7, 2012 at 11:26 pm
    Let me suggest an explanation. This is an artifact of the varying locations of the samples as you go back in time. Which brings up its own set of problems, but that’s another story.

    Hi Willis. The legend to the graphs states that:

    (b) Time series of
    the sea surface temperature (red) and of the 0–20 m layer temperature (blue). Both time series are obtained by weighted
    averaging of 5  5-degree box absolute temperatures.

    So the weighted averaging of the boxes should take out a lot of the effect of varying location. I would agree with you that there will still be some regional biases, though this should cancel out.

  40. markx says:

    Paul Vaughan says:October 7, 2012 at 4:55 pm

    “…..Pay more respect & attention to the Russian luminaries: Sidorenkov, ……”

    Paul, I think you are on to something there.

    Record (satellite era) low sea ice in the north, record high sea ice in the south, sure sounds orbit and tilt related …..

    Sidorenkov etal may be the only ones looking in the right direction….. some of his charts are impressive: Figure 7. Temporal variations in the specific mass of ice in Antarctica: curve 1, the theoretical value curve 2, the empirical value.

    Astronomical and Astrophysical Transactions Vol. 24, No. 5, October 2005, 425–439
    Physics of the Earth’s rotation instabilities N. S. SIDORENKOV*

    Abstract This paper generalizes the results of investigations on the instabilities of the Earth’s rotation and related geophysical processes. Long series of observations of the Earth’s orientation parameters are demonstrated. The tidal variations in the length of the day are described. The temporal variations in the atmospheric angular momentum and their contribution to the instabilities of the Earth’s rotation are studied. The mechanisms of seasonal variations in the length of the day and polar motion are discussed.
    The probable geophysical processes responsible for the decades-long (2–100 years) instability of the Earth’s rotation are discussed.

    Keywords: Earth; Rotation; Instabilities

  41. Nick Stokes says:

    Willis,
    “Let me suggest an explanation. This is an artifact of the varying locations of the samples as you go back in time.”

    Yes, that’s what the authors say:
    “As illustrated by Figure 1b the mean monthly absolute temperature time series for the sea surface and for the 0–20 m level diverge before 1940 with nearsurface sampling being biased to higher latitude resulting in a colder mean temperature.”

    You can see this in the little maps in Fig 1a. Although SST is better measured than near-surface, there is much variation. And I don’t think weighting is helping (per TB); elsewhere they refer to area weighting (converging longitudes), and I think that is what is meant here. It won’t help with missing data.

    Another giveaway is the stated average temp – about 20°C. Now I don’t swim much in the ocean nowadays, but that’s not my memory. In fact, since average global temp is about 14°C and 3/4 of that is SST, something is wrong. I think it is that they are (in 1b) trying to average absolute temperatures rather than anomalies.

  42. P. Solar says:

    Willis Eschenbach … I’ve put a ruler over Figure 1 b) to show what I mean.

    That’s exactly what I pointed out here:
    http://wattsupwiththat.com/2012/10/07/resolving-the-biases-in-century-scale-sea-surface-temperature-measurments-reveals-some-interesting-patterns/#comment-1103693

    but your ruler makes it very clear and indisputable. How is it possible that this downward trend in the data is not present in the ‘anomalies’?

  43. P. Solar says:

    Gourestski et al : ” The higher trend in the subsurface analysis prior to 1945 is
    also seen in the SST analysis subsampled to have the same
    coverage as the subsurface analysis and is therefore likely to
    be due largely to poor geographical sampling. The subsurface
    data for the years before ca. 1920 have a strong geographical
    bias with the majority of the data coming from the (North)
    Atlantic Ocean (see Figure 1a). ”

    Yet if we look at papers covering hadSST we see that increase in that pre-1940 period is reduced on the basis of NMAT , deck height and bucket bias adjustments , not geographical bias. Smith and Reynolds found that if they did not use NMAT adjustments: “1854-1941 bias correction is about constant”.

    This is discussed in more detail with John Kennedy here:
    http://judithcurry.com/2012/03/15/on-the-adjustments-to-the-hadsst3-data-set-2/#comment-188254
    and here:
    http://judithcurry.com/2012/03/15/on-the-adjustments-to-the-hadsst3-data-set-2/#comment-187260

    Hadley remove about 2/3 of the variation from this part of the data. None of that on the basis of geographical bias.

    The 5×5 degree grid smoothing would not remove inter-regional bias between one ocean basin and the global average, however, the difference between ICOADS 2×2 grid processing and Hadley 5×5 has been shown to be the operation that removes much of 50y to 100y scale variation.

    In the extensive literature on HadSST processing they have never directly invoked this geographical bias, though there may be some geographical component to the NMAT “correction”.

    The current paper also says:

    ” The second [uncertainty] arises from instrumental biases.
    which can be comparable in magnitude to real variability in the
    climate.”

    Here “real” variability presumably means what they believe to be real as a result of their “corrections”. In fact it is simply a confirmation that the changes they are making in the name of bias correction are comparable to what is left. This is in agreement with Goodman’s statement that hadSST3 processing removes 2/3 of the long term variation from the majority of the record.

    A statement that John Kennedy was forced to agree with in that discussion. He also agreed that the bias adjustments were based on hypothesis not observation.

  44. P. Solar says:

    Anyone know whether this data is publicly archived are required by GRL publishing requirements?

    It seems like someone needs to plot this and calculate the anomalies to see what is going on.

  45. Peridot says:

    beesaman says:
    October 7, 2012 at 11:44 am

    Maybe the BBC are slowly seeing the light?
    http://www.bbc.co.uk/news/science-environment-19848112

    Or maybe not – This written report is very different from the television report on BBC News channel. There it was all doom and gloom due to ‘climate change’ linking the loss of ice at the Arctic to all the rain in the UK and drought in the USA. None of the doubt and detail in the written report made it into the recorded report (which was nearly 15 minutes long).

    Shukman also, with a straight face (deception or ignorance?) referred to the ice calving off a glacier as being due to the glacier RETREATING!

    I wish I could have asked him what he thought an ADVANCING glacier would look like.

  46. P. Solar says:

    Nick Stokes says: I think it is that they are (in 1b) trying to average absolute temperatures rather than anomalies.

    Why “trying to” and what is wrong with real temperatures (be careful with the the term “absolute temperature” that’s – kelvin scale – which is probalby not what you meant). Anomalies are just a way of removing most of the annual variation without loosing the end of the data as would happen when using a filter.

    The usual meaning for “anomaly” is the difference of a monthly (or dailly) value from the mean of the same months (or days) over the reference period.

    The two should be similar on the decadal scale though there may be differences in the detail in years that have more/less annual variation than the ref. period.

    If your understanding is different to that please explain.

    Clearly SST was higher in 1950 than today in figure 1b yet the anomaly plot tells a very different story.

  47. P. Solar says:

    Tallbloke: Both time series are obtained by weighted averaging of 5  5-degree box absolute temperatures.

    I think that just means that the gobal average plot is an area weighted average of the gridboxe values, this is nothing to do with the averaging that goes into making the gridded data.

  48. vukcevic says:

    I found these references useful:
    1 – Corrections to Pre-1941 SST Measurements for Studies of Long-Term Changes in SSTs, Jones et al
    http://icoads.noaa.gov/Boulder/Boulder.Jones.pdf
    2 – Assessing bias corrections in historical sea surface temperature using a climate model, Folland
    ftp://ftp.wmo.int/Documents/PublicWeb/amp/mmop/documents/JCOMM-TR/J-TR-13-Marine-Climatology/REV1/joc1171.pdf
    3 – Reassessing biases and other uncertainties in sea surface temperature observations, Kennedy et al
    http://www.metoffice.gov.uk/hadobs/hadsst3/part_1_figinline.pdf

  49. JJ says:

    Steven Mosher says:

    Unfortunately, the hypothesis that solar activity “affects” “global” cloudiness, has no well defined empirical data for its support and no working proven physics to support it.

    Nor does it have any well defined empirical data for its refutation, and no working proven physics to deny it. And ‘global warming theory’ is ginned up using multiple ad ignorantiam fallacies that rest on such utter lack of information.

  50. Steve Keohane says:

    A little late to the discussion. Here is another version of plot (b), with the Y-axis exaggerated, and .5°C hash marks from 19.5-25°C.
    http://i48.tinypic.com/2rqdijs.jpg

  51. Stephen Wilde says:

    “Unfortunately, the hypothesis that solar activity “affects” “global” cloudiness, has no well defined empirical data for its support and no working proven physics to support it.”

    Some pretty supportive observations though, unlike AGW theory.

    It only needs a few more years to confirm or rebut but in the meantime I recommend it as an area of investigation.

  52. Paul Vaughan says:

    markx (October 8, 2012 at 2:07 am) suggested:
    “Sidorenkov etal may be the only ones looking in the right direction…..”

    In the long run, solar cycle frequency/length changes centrally-limit terrestrial Antarctic ice specific mass & geomagnetic field jerks:
    http://i49.tinypic.com/wwdwy8.png

    For many geophysical variables, Earth’s response to solar input is nonlinear due to the volume, distribution, and state of water at and near the surface.

    NASA JPL is reorienting to follow the Russian lead on this file.
    There’s also a very bright group in Paris.

  53. P. Solar says:

    Paul Vaughan says: In the long run, solar cycle frequency/length changes centrally-limit terrestrial Antarctic ice specific mass & geomagnetic field jerks:
    http://i49.tinypic.com/wwdwy8.png

    What is the origin of that graph? What is that “solar cycle deceleration”? Looks pretty similar to the variation in length of day.

  54. P. Solar says:

    vukcevic says: … links.

    most of those links and more are at the end of article I linked to at Climate etc. ;)

  55. vukcevic says:

    P. Solar says: October 8, 2012 at 11:31 am
    ……
    For some reason my pc is ‘allergic’ to Climate etc, it takes forever to load, even cooling fan goes to extra power. I avoid following links to (or on) the JC’s blog, since takes too long to return back where I started from.

  56. P. Solar says:

    “For some reason my pc is ‘allergic’ to Climate etc, ”

    That’s unfortunate, I think you’d find the article interesting. It looks at how the adjustments affect the frequency spectrum of the data.

    It’s very odd that you have problems there. It’s hosted on the same WP platform as WUWT , has a lot less links and peripheral content and no ads. As a quick test, I just reloaded the CE page in question and this page (which is much shorter in comments). CE took 6s this page took 8s.

    I usually open links in a new browser tab rather than going directly. That means “coming back” just involves changing tab or shutting down a tab. Much quicker and I don’t loose my place in the referring page.

    Trying to understand all the supposed bias corrections, and corrections to the corrections, applied to hadSST3 is quite a paper chase. How many of their guesses about the nature and timing of equipment and sampling changes hit the mark I don’t know. But a lot of fairly broad assumptions seem to be involved.

  57. P. Solar says:

    Nick Stokes says: “Another give-away is the stated average temp – about 20°C…. In fact, since average global temp is about 14°C and 3/4 of that is SST, something is wrong. I think it is that they are (in 1b) trying to average absolute temperatures rather than anomalies.”

    Good point. That means the whole data set has a fairly significant geographical bias even at the end. These are based on “available data” in each plot. There is some infilling and project involved in the creating the 5×5 gridded data but these graphs apparently don’t include projecting infilling of areas with no data.

    More coverage of Southern Ocean in recent period may explain why temperature time series curve down slightly in recent decades.

  58. Paul Vaughan says:

    P. Solar (October 8, 2012 at 11:29 am) asked about http://i49.tinypic.com/wwdwy8.png :
    “What is that “solar cycle deceleration”?”

    Solar Cycle Deceleration = Rate of Change of Solar Cycle Length
    Rate of Change of Solar Cycle Deceleration = Rate of Change of Rate of Change of Solar Cycle Length (i.e. 2nd derivative)

    In this case I measured from sunspot numbers, but you could use other solar variables.

    Use a complex wavelet (e.g. Paul or Morlet) to measure the cycle length as a function of time and then take derivatives. Tune the grain to something near Schwabe (~11 years) and the extent to something near 64 years. (You’ll find the results are robust even if you don’t tune carefully.) Cautionary Tip: You need to use a very, very, very wide support span to get good estimates of higher cycle length derivatives.

  59. phlogiston says:

    These data appear to confirm that recent warming of the oceans (i.e. of the planet) stopped around 2003.

Comments are closed.