Uncertainties in sea surface temperatures

by Judith Curry

Two new papers have focused on the quality, uncertainties and  interpretation of global sea surface temperature data.

A call for new approaches to quantifying biases in observations of sea-surface temperature

Elizabeth Kent, John Kennedy, Thomas Smith, Shoji Hirahara et al. (long author list)

Abstract: Global surface-temperature is a fundamental measure of climate change. We discuss bias estimation for sea-surface temperature and recommend the improvements to data, observational metadata, and uncertainty modeling needed to make progress.

Global surface-temperature changes are a fundamental expression of climate change. Recent, much-debated, variations in the observed rate of surface-temperature change have highlighted the importance of uncertainty in adjustments applied to sea-surface temperature (SST) measurements. These adjustments are applied to compensate for systematic biases and changes in observing protocol. Better quantification of the adjustments and their uncertainties would increase confidence in estimated surface-temperature change and provide higher-quality gridded SST fields for use in many applications.

Bias adjustments have been based either on physical models of the observing processes or on the assumption of an unchanging relationship between SST and a reference data set such as night marine air temperature. These approaches produce similar estimates of SST bias on the largest space and timescales, but regional differences can exceed the estimated uncertainty. We describe challenges to improving our understanding of SST biases. Overcoming these will require clarification of past observational methods, improved modeling of biases associated with each observing method, and the development of statistical bias estimates that are less sensitive to the absence of metadata regarding the observing method.

New approaches are required that embed bias models, specific to each type of observation, within a robust statistical framework. Mobile platforms and rapid changes in observation type require biases to be assessed for individual historic and present-day platforms (i.e., ships or buoys) or groups of platforms. Lack of observational metadata and of high-quality observations for validation and bias model development are likely to remain major challenges.

Published in the Bulletin of the American Meteorological Society [link to full paper]

This paper provides a very good overview of the issues, and makes 9 recommendations for much needed improvements for the SST data set and the assessments of the uncertainties.

Assessing recent warming using instrumentally homogeneous sea surface temperature records

Zeke Hausfather, Kevin Cowtan, David C. Clarke, Peter Jacobs,  Mark Richardson, Robert Rohde

Sea surface temperature (SST) records are subject to potential biases due to changing instrumentation and measurement practices. Significant differences exist between commonly used composite SST reconstructions from the National Oceanic and Atmospheric Administration’s Extended Reconstruction Sea Surface Temperature (ERSST), the Hadley Centre SST data set (HadSST3), and the Japanese Meteorological Agency’s Centennial Observation-Based Estimates of SSTs (COBE-SST)  from 2003 to the present. The update from ERSST version 3b to version 4 resulted in an increase in the operational SST trend estimate during the last 18 years from 0.07° to 0.12°C per decade, indicating a higher rate of warming in recent years (Karl et al., 2015). We show that ERSST version 4 trends generally agree with largely independent, near-global, and instrumentally homogeneous SST measurements from floating buoys, Argo floats, and radiometer-based satellite measurements that have been developed and deployed during the past two decades. We find a large cooling bias in ERSST version 3b and smaller but significant cooling biases in HadSST3 and COBE-SST from 2003 to the present, with respect to most series examined. These results suggest that reported rates of SST warming in recent years have been underestimated in these three data sets.

Published in Science Advances [link]

Kevin Cowtan has posted a nice summary on his website [link], that includes a video explanation by Zeke Hausfather.

These are the main figures from the paper:




Plus this additional figure that Zeke sent me:


JC reflections

After a decade of exhaustive scrutiny of the land surface temperature record, we are now seeing a focus on sea surface temperatures.

For some context on this issue, see these previous posts:

I’m not sure how we will ever sort out the issues in the historical record prior to about 1980, and especially during WWII and prior to 1920.

But the more vexing issue is the discrepancies in the recent record — the last two decades and even the last 5 years.

Sorting these issues out, with comprehensive assessment of errors and uncertainties in the record, is very important.

Eventually we will be able to rely on coupled atmosphere-ocean data assimilation systems to produce global values of SST (and we will be able to go backwards in time also).  But coupled data assimilation systems are in their infancy.

Its good to see the international coordination reflected by the Kent, Kennedy et al. paper, as well as newcomers Hausfather, Cowtan et al. taking a fresh look.

We really need to get the data records for the last two decades sorted out.

With regards to Zeke’s paper, I haven’t had the time to dig into the paper.  I will say that this kind of analysis definitely needed doing.  And I give Zeke an ‘A’ (hah! still playing professor) for communicating the results of this research.  I understand that they have been giving a lot of interviews on this, lets see how this gets played in the media.

295 responses to “Uncertainties in sea surface temperatures

  1. Global surface-temperature changes are a fundamental expression of climate change.

    look at the slope of the temperature graphs. the temperature goes up a steeply and comes down a steeply. there are no changes in any external forcing or greenhouse gas that have enough watts per meter squared to do that.
    Ice always advances as it gets colder. Ice always retreats as it gets warmer. it always snows more when it is warmer. it always snows less when it is colder.

    ice advances because the ice on land has been replenished during a warmer period. ice retreats because the ice on land has been depleted during a colder period.

    The King has no clothes on, it really is this simple.

    • ice on land in the NH does regulate the NH temperature

      ice on land in the SH does regulate the SH temperature

      earth is colder than it was fifty million years ago because there is more circulation of warm water in polar regions and therefore, more snowfall to support more ice on land.
      Ice extent is a result of ice volume on land.
      ice volume on land is a result of snowfall.
      ice extent regulates temperature.
      the cooling by IR in the tropics accounts for most of the cooling but it does not have a thermostat with a fixed set point. The cooling in the polar regions has sea ice that freezes and thaws at a fixed set point and that does the fine tuning of temperatures in each hemisphere.

      • Think about that!

      • if you change the energy into earth, earth will adjust the ice to counter getting very far out of balance.

      • The IR cooling from the tropics, fifty million years ago, was much more than now, but we were still warmer. The IR cooling from the tropics, during the coldest parts of the major ice ages was much less than now, but we were still colder.
        This ice extent caused the difference, it was not a result of a difference.
        In all of the data, it snowed more in the warmer times and then it got colder, it snowed less in the colder times and then it got warmer.
        Study actual ice core data. They are the best proxies we can have because they reflect the temperatures of the oceans. Ice cores only go back 800 thousand years but the other proxies indicate the same profiles, going back fifty million years.

    • It is the old cause and effect question. The slopes show the issue: ” the temperature goes up a steeply and comes down a steeply”. The sea surface temperature is not a good indicator of global temperature, because it is also a major driver. If the ocean surface warms, the climate on land warms. When we talk about the difficulty of assessing natural climate variability versus human induced, we are mostly talking about the uncertainty in self induced ocean surface temperature variation from various effects. There are upwelling cycles such as La Nina/El Nino, but longer ones such as the 60 year cycle and probably much longer ones such as the 1,000 year cycle that caused the Roman and Medieval Warm Period as well as at least part of our own. What we have here is thus a record that is too short to draw conclusions.

      In fact, if we were to see a positive trend in sea temperature that matches the land temperature, our land based regression of Transient Climate Response (TCR) is in fact an Equilibrium Climate Response (ECR), because over the regression period at least, sea surface temperature did not lag land temperature, or not as much as would be expected in a CO2 effect only model. The IPC has published an ECR of 3.2 for a TCR of 2.0 deg Celcius. This indicates that a degree of ocean surface temperature increase causes 0.375 deg of land temperature increase. The measured trend of 0.12 deg/decade equals 0.84 deg in 70 years. So, if the ocean warmed this amount during the regression period, we should subtract 0.84*0.375 = 0.315 deg from the TCR estimate. It would lower the IPCC TCR to 1.7 deg/doubling. The more recent regression estimate of 1.3 by Lewis/Curry would lower to a TCR of 1.0 deg/doubling.

      It is amusing that the effort to find the “missing heat” in the ocean implies a lower climate sensitivity.

    • These minor adjustments of fractions of a degree are ridiculous. These time series are all red-noise because of the integrating effect of ocean heat storage on ocean temperature. The signal to noise ratio is zero. There is no signal. There is no trend. It’s just noise. See: http://blackjay.net/?p=386

    • There are three methods of heat transfer. They are conduction, convection, and radiant heat. Heat transfer to or from the earth can only be done by radiant. All material contains heat and is radiating it to cooler surfaces or absorbing it from warmer surfaces. The difference is the heat gain or loss of the material.
      The earth gains heat radiated from the sun and loses heat it radiates to outer space, called black sky radiation. Outer space is considered absolute zero.
      The amount of radiant heat hitting the earth from the sun daily is relatively constant. The radiant heat lost daily by the earth thru black sky radiation is constant since absolute zero is constant. The amount of heat gained by the earth’s surface depends on the surface area of the earth covered by water relative to that covered by land. Land area absorbs a larger percent of the radiant heat relative to the water area since the surface of the water reflects a percentage of the radiant heat back to outer space. The daily access heat, or loss of heat, is transferred to the oceans thru conduction and convection where it works its way to the poles and it freezes water adding to the polar ice caps or melts the polar ice caps thus keeping the temperature of the oceans, thus the earth, relatively constant. As the polar ice caps grow or melt, the surface area of the earth covered by land relative to that covered by water changes. This is the definition of global warming.
      That radiant heat absorbed by oceans and land masses is transferred to the atmosphere thru conduction and convection. When it is winter in one hemisphere it is summer in the other and the same with spring and fall. I would think the average temperature of the lower 5,000 feet of the atmosphere changes about 10’F to20’F each day. This probably takes more heat than man has added to the earth in the last 50 years. That heat man adds to the atmosphere each day is radiated to the black sky and the infinitesimal amount left helps melt the ice during global warming, should be called Global Defrosting, . The scientists have taken core samples of the polar ice caps and know how close we are to the ice left at the beginning of the last ice age. That is how close we are to the end of global warming or we have already begun global cooling, should be called global ice making.
      Absolute Zero is -459.68’F and the surface temperature of the sun is between 7,300’F and 10,000’F or an average of 8,650’F. If we could go back in time 12,000 years, the end of the last ice age, we would probable see that the average daily temperature of the earth was in the mid 60’F as it is today. You must understand the amount of heat gained every 24 hours is almost equal to that lost during the same 24 hours. Angle of the earth’s axis is 23.5’.
      The average surface temperature of the earth is 63.5’f. The heat loss to black sky radiation every 24 hours is constant. The average radiant heat striking the surface of the earth is relatively constant. Because the sun is an active star the average temperature will change over centuries. As the surface area of the earth covered by water increases, the radiant heat reflected back to the black sky increases. When the daily radiant heat gained by the earth from the sun in 24 hours is less than that lost by the earth’s surface Global ice making will end, or has already ended.
      The chart is CO2 levels of ice core in the antarctic

      The above chart is of an ice core sample from the Antarctic Ice. The red line, I believe shows, the CO2 level in the air trapped in the ice when it froze. The high point in the red line shows the ice formed at the beginning of the Ice Age. This line shows only ice formed in the first couple of millenium. The low point is the ice which was formed near the beginning but is the top layer when the new Ice Age began. As you can see there are 5 high points, thus 4 Ice Ages. I believe that as the new Ice forms some of the CO2 in the new ice is absorbed into the old ice, thus the angle on the vertical line. I also believe the small ups and downs of the red line are due to the change of the average temperature of the surface of the sun between the 7,300’F and 10,000’F. There is also an ice core sample from Greenland almost identical to the above.

  2. It’s as if she never left. And that’s a good and much appreciated thing.

  3. The difference between satellite atmospheric temps and surface/shallow water temps sure look like less clouds to me. Any shoreline swimmer knows how much one small cloud across the sun changes his experience, while in the shade of an umbrella he doesn’t notice a thing, even though we are well equipped to notice a 0.5C change.

    Karl et al doesn’t get rid of the problem. Neither does Zeke. Both only firm up the fact there is such a problem.

  4. I am reminded of Samuel Johnson’s remark about a poorly dancing bear. The oceans are vast. SST varies with wind/wave churning of the mixed layer, and current eddies. The surprising thing is that the ‘new’ metrics since 2000 agree as much as they do. The earlier ship data is hopelessly uncertain as well as trade route biased. Karlization did not change that uncertainty. What it did do was corrupt the here shown better post 2000 data for an agenda–disappearing a 21st century pause in delta T most others had already agreed existed. In the end, Karlization will contribute to busting CAGW.

    • The earlier ship data is hopelessly uncertain as well as trade route biased.

      Trade routes followed ocean currents. That is likely the best biased data from mixed ocean waters that we could possibly get from those times. It is the warm tropical oceans currents that warm the polar oceans that promote more snowfall.

    • Historical records of when people went to live in Greenland because it was warm enough and of when the Chinese mapped the Arctic because it was warm enough and records that it got colder after the massive more snowfall during this Medieval Warm Period caused ice to advance that caused the Little Ice Age.
      We have wonderful proxy and historical data that is there to help us understand what caused what. It does always get colder after a warm period with more snowfall. It does always get warmer after a cold period with less snowfall. That is not random luck, that is always.

    • Steven Mosher

      “Does anyone believe Karl Cooked the books when independent satellite data shows he did not?”

      • SM, actually the ‘independent sat data’ shows that he did. Hence Lamar Smith’s subpoena and the NOAA refusal to comply therewith, in contempt of congress. Do try to keep up, albeit hard in Berkeley.

      • Cooking the books is tantamount to fraud.

        Using statistical techniques, however valid, to support an admistration’s objectives, is creeping up to the line where science gives way to politics.

        Not illegal. Not even unprofessional. Just a dilution of scientific independence.

      • Didn’t anyone want to investigate why the UAH trend since 1997 dropped by 60% between Versions 5.6 and 6? This is supposedly their best data, and it was a seismic shift using all the same data.

      • Steven Mosher

        Rud. Are you accusing Karl of misconduct?

      • Sounds like he’s saying satellite land and ocean overrules satellite ocean, despite satellite ocean being affirmed by ARGO and ocean buoys, and proves that Karl did something wrong… like being a good scientist who was attacked viciously by prevaricators in congress and on blogs and opinion pages.

      • Steven Mosher:

        Could you link to this fraud allegation please (or one of them if there are many). I have no idea what you are talking about.

        How does the paper rebut this fraud allegation – could you actually say something substantive about this issue.

        Thanks in advance.

      • Richard, Anthony Watts infamously wrote a quite intemperate letter regarding the Karl analysis. It’s archived by HotWhopper ( a blog I don’t particularly respect, but it does have a fun name ).

        That doesn’t mean Watts is wrong about everything or that the Karl analysis is the flawless truth.

        But it’s a reminder to reign in emotion,
        and to “Fall in love with the problem, not the solution.”

      • The data here shows that bouys, satellites, and Karlized NOAA agree since 2000. The objection to Karl is that his method of preferring ship to buoy data far earlier than presented here had the effect of reducing the cooling between 1945 and 76. Back then the bouys ran cooler than the ships, and he picked the ships, the equivalent of the “old” NOAA.

        Is cherry picking data fraud? No. Is it agenda based science?

  5. Just as long as the biosphere adapts as fast as the ocean changes we will be OK. First order changes like sea levels and OHC are only significant if it reduces the ability of the ocean to support the food chain. The average American consumed 15.5 pounds of seafood in 2015 (up 1lb. since 2014, the biggest leap in seafood consumption in 20 years). For reference 91 percent of U.S. commercial seafood is imported. World wide fish stocks are in decline according to a recent report by the Global Atlas of Marine Fisheries, paywall @ islandpress.org/book/global-atlas-of-marine-fisheries).

    • Are you suggesting that species adapt to climate change? I would argue that past climate change is the biggest killer of species and cause of extinctions, because most species don’t adapt that quickly. Some of these climate driven die-offs are still occurring today under our watchful eye, yet it goes completely unnoticed.

      • My interest in climate change has moved beyond temperatures, sea levels and polar ice. So called ‘first order’ effects. Plant and animals can’t read thermometers but if the environment they live in can’t support them then they either move or die. That’s the way it’s always been so the only important variable is how fast things change. With the exception of maybe the Great Barrier Reef it’s not a problem yet and I’ll be worm food before it maters to me.

      • Jack,

        I’m reminded of a paper that studied Andean bird populations and their reaction to a changing climate.

        Observational data for temps was good and was not significantly different to modeled expectations. What did not match the models was the behavior of the bird populations. Models predicted them moving up in elevation as temps increased were wrong. Populations were still strong, but also still occupied the same areas. The conclusion of the researchers? The birds were threatened because they were too dumb to move to higher elevations.

        For me the conclusions are the researchers should be flipping burgers.

      • tim,
        That antidote proves my point, bird’s can’t read thermometers (or models) so climate change is not a problem for most species. Being on the menu is probably their biggest problem (I hear they taste like chicken). Coral can’t read thermometers either and they are having a hard time lately but that’s OK because I don’t eat them.

      • Jacksmith: “With the exception of maybe the Great Barrier Reef it’s not a problem yet and I’ll be worm food before it maters to me.”

        Hasn’t Great Barrier Reef been shown to be highly influenced by ENSO? As in ENSO is most responsible for die off events or “bleaching?”

        It should matter to you, because some of us are dealing with climate climate change impacts right now. The problem is that it is climate change that occurred over the past 1,600 years, so no one cares about it. Everyone is more concerned with climate change that hasn’t happened yet.

      • JS: “so climate change is not a problem for most species. ”

        Oh really? Tell that to Coleogyne ramosissima, Populus tremuloides, Yucca brevifolia, Pinus ponderosa, and their obligate species.

        I am watching what could be an extinction event, or perhaps just a 99% reduction in a species as I type this, and it is very likely due to climate change, and it is nearly certain to be mostly due to climate change over the past 1,600 years. I am pretty sure climate change is a problem for all species.

        I am also wondering what it is you have been reading since you moved beyond temperatures and ice.

      • Oh really? Tell that to Coleogyne ramosissima, Populus tremuloides, Yucca brevifolia, Pinus ponderosa, and their obligate species.

        I am watching what could be an extinction event, or perhaps just a 99% reduction in a species as I type this, and it is very likely due to climate change, and it is nearly certain to be mostly due to climate change over the past 1,600 years. I am pretty sure climate change is a problem for all species.

        Can you substantiate your suspicions?

        Pinus ponderosa has a very wide range and with it, a very wide range of climates – it does not appear to me that climate is a problem:

        I had to look up Yucca brevifolia but I recently drove from Phoenix to Las Vegas, and it’s a drive I would recommend because the Joshua trees are thriving and spectacular.

        The problem with any extinction theory and AGW is that the change has been small compared to the inter-glacials of the Pleistocene which lasted thousands of years:

        And in fact, extinctions don’t appear to correlate with paleo temperature estimates:

      • Eddy. “Can you substantiate your suspicions?

        Pinus ponderosa has …”

        Is it growing where it was 2000 years ago?

        Have you looked at where the oldest pondo in the world is growing? Any thoughts on why that population is now isolated from its genetic cohorts to the southeast?

        Yucca is currently undergoing a shift in latitude, and it’s been going on for a while.

        Blackbrush isn’t so lucky, there was a great reduction in reproduction somewhere around 700-1000 years ago. A shift in elevations occured as recently as 250 years ago. It no longer establishes today in areas that were almost monocultures 5 years ago.

        Sorry, no citations at the moment.

      • Of course populations are dynamic and yes, climate can influence evolution, but the observed and even forseeable extent of change is small compared to known variations of the evolutionary recent past. I do not see a case for AGW extinction.

      • We are arguing for the same idea. Some of the forecasts I have seen for vegetation is calling for less of change in 60 years than we have experienced in the last 10, which isn’t much compared to the last 700.

    • Is this information useful?

      “A Japanese sushi chain boss has bid a winning 74.2 million yen ($632,000) for a 212 kilogram (466 pound) bluefin tuna.
      The winning bid Thursday for the prized but imperiled species was the second highest ever after a record 155.4 million yen bid in 2013.”

      Once something becomes scarce it gets pretty expensive. Wonder what they will cost by 2050? I guess being one of the top predators in the ocean last thing you are worried about is the water temperature…

      • People with too much money buy interesting things. Some buy tuna to resell to other folks with too much money. Others buy private jets to fly to global warming conferences so they can have tuna for supper. Neither do any good for anything. Most critters and plants have a fairly wide range of tolerance to temperature, or behavioural ability to avoid extremes (migration, hibernation, range extension etc.). Only species with a narrow tolerance, low reproductive capability or sought by man will be endangered. Such is the cycle of life on earth. One would find it difficult to find extensive “habitat” loss (and thus endangered species) owing to climate change. Examples are everywhere when land use (habitat loss) and over-harvesting are involved.

      • “People with too much money buy interesting things.” – Richard
        “No such thing as too much money” – Mammon
        “Everyone has their price” – jacksmith4tx

      • Keeping with my notion that the most important element of climate change will be spotting 2nd order effects in the biosphere To that end I found this interesting:

        “Ocean surface warming creates a layer of low-nutrient water separate from cooler, nutrient-rich water below. This process occurs annually in large regions of the open ocean where punctuated winter mixing allows for a short “bloom” of phytoplankton life that is followed by a summer season of warm, low-nutrient surface waters. Most of the bloom species disappear during summer because they are not effective competitors for nutrients at low concentrations. Ocean surface warming is leading to an expansion of these low-nutrient environments in a process known as ocean desertification…
        The first phytoplankton lineage appears to be an entirely new group of species of phytoplankton. Researchers believe its ancestor may be a single-celled protistan group that took a separate evolutionary path from the haptophyte algae, which arose between 1 billion and 637 million years ago. The second lineage appears to be closely related to haptophytes. However, it is a new group that doesn’t belong to any known species or class.
        How these groups contribute to the food chain and carbon cycle is currently unknown. Worden believes one possibility is that they get their nutrients though a combination of photosynthesis and feeding on other cells.”

        So if I read this right these new phytoplankton are some sort of predator that feed on other single cell organisms? At least one specie’s ancestors were last seen over 637 million years ago so that must be a good thing that they find today’s climate more hospitable.

    • About 1/8th of seafood consumed in the US is ‘farm raised’.(naturally raised is 7 times farm raised)
      US per capita consumption of fish has varied between 14.6 lbs and 16.6 pound since 1997.

      Seafood Facts

  6. I’m not sure how we will ever sort out the issues in the historical record prior to about 1980, and especially during WWII and prior to 1920.

    I really don’t care how that they fudge the data. We already records of what it was before they fudged it and we have ice core and other proxy data that we can use to better understand what caused temperature changes in the past and we will then know what is still working the same way. Their fudging does damage their credibility and that helps the skeptic case. Now, with a more skeptic head of government, it will help determine who to get rid of. We need honest science and discussion and debate where every theory is considered.

    • If the consensus climate people fail to show up at a skeptic climate conference to present and defend their theory and data, that is cause to get them off the public payroll.

      If the consensus climate people fail to invite skeptic climate people to their climate conferences to present and defend their theory and data, that is cause to get them off the public payroll.

  7. David Wojick

    It is the nature of uncertainty that sorting out may not be possible.

    • Steven Mosher

      Does anyone believe Karl Cooked the books when independent satellite data shows he did not?

      • You said that already upthread. Repetition does not equal truth. You do know that satellite T equivalent data does not measure SST, right? It ‘sees’ infrared emissions from atmospheric zones. Darned atmosphere hanging over the sea surface. Whodathunk.

      • If we knew that one of the measurement schemes was correct everyone would use it. Because there is debate, apparently none of the schemes is accepted as definitive. All “new” adjustments, therefore, are made against an existing data base that obviously has flaws or adjustments wouldn’t be necessary. New attempts may change a pattern that is closer to reality than the new results. The whole thing is a game, which would be harmless academic gymnastics if politics wasn’t involved. The price, like that of tuna, is way too steep. Getting excited about new annual highs based on 0.02 degrees shows how ridiculous the whole game has become. What’s next? Maybe weekly temps? then daily? then hourly? I can see it now – the high today for 2pm on January 5 sets a new record for this hour in history.

  8. A problem with finding and adjusting for systematic biases when measuring sea surface temperatures is that it is impossible to not find actual heat on the surface of the ocean than but it is impossible to find Trenberth’s missing heat at the bottom of the ocean and yet climate clowns believe it’s there anyway– that is the bias!

  9. it is impossible to find Trenberth’s missing heat

    No matter, the same heat was missing during the Roman and Medieval warm periods and earth always snows itself out of a warm period.

  10. Judith, typo in opening sentence of post:

    “Two new papers have focused on the quality, uncertainties and interpretation of global sea level data.”

    “sea level” should be “sea surface temperature”.

  11. Being diagnosed at an early age as a right brained person and thus discouraged from the pursuit of math and science, I am interested in the graphic representation of these data.

    Giving equal volume to x and y (years and decades against .0 temp changes) makes the changes look dramatic.

    Where is the drama?
    It goes up .3 or goes down .3.
    From where I sit, if it were flat I’d reassess my lack of church attendance.

    What does the ‘healthy’ test result look like?

    Dr. Curry,
    After yesterday’s post, it was great to click on your blog (my fav) and see a new article.
    Oddly, your blog and the comments to me are a better source for the state of the outside world than the NYT or WAPO.
    Faint praise, I know.
    Keep it about science, but it’s actually much more.
    Thank you.

  12. Judith ==> I have serious doubts about the physical reality of this statement in the abstract of the Kent et al. paper:

    “Global surface-temperature changes are a fundamental expression of climate change. “

    It is my understanding that sea surface temperatures are more dependent on ocean currents, mixing, up- and down-welling of ocean waters, ENSO, etc etc. and are NOT a “fundamental expression of climate change.”

    Am I just horribly off-base on this issue?

    • Am I just horribly off-base on this issue? YES!

      Sea surface temperatures are dependent on ocean currents, mixing, up- and down-welling of ocean waters, ENSO, etc etc. and are a “fundamental expression of climate change.”

      • Long term ocean surface temperatures are recorded in the ice core data from the Antarctic and from Greenland and those are records of climate change.

      • pope ==> So are you saying that SST is BOTH dependent, caused by, “ocean currents, mixing, up- and down-welling of ocean waters, ENSO, etc etc.” AND thus are a “fundamental expression of climate change.”?

        Can’t quite figure out what you mean to say there.

      • pope ==> “Long term ocean surface temperatures are recorded in the ice core data from the Antarctic and from Greenland” — recorded to tenths of a degree C? for the whole planet? Who would have thought?

        And here Dr. Curry thinks that there is some uncertainty…foolish lady.

        Really now, the issue, my question, is if a single-number whole Earth average SST metric is a “fundamental expression of climate change.”, given that monthly/seasonal/yearly SSTs are the result of lots of known causes not directly associated by climate change.

    • That struck me the same. It seems to be a chicken vs egg issue. It doesn’t seem like a global value describing sst would give any indication of climate change. Values describing the changes in ocean currents or sst in specific regions would.

      Of course I am wanting to look at actual climate change, not global values that have not proven very effective at identifying or describing climate change.

    • Steven Mosher

      “It is my understanding that sea surface temperatures are more dependent on ocean currents, mixing, up- and down-welling of ocean waters, ENSO, etc etc. and are NOT a “fundamental expression of climate change.””

      Interesting… down below you claim we cant really know anything

      • Mosher ==> Better take a break — you have descended to the level of ‘tweenage trolling (and not even good ‘tweeage trolling).

      • catweazle666

        I think Mosher’s ‘Trump Derangement Syndrome’ is playing him up.

        That’s the charitable explanation, at any rate.

    • That’s well spotted, Kip, and no way off-base.

      Writing papers includes the early laying out of foundation assumptions, and a smoothly written sound-bite-like opening like that sentence affords the programming of the assumption into the reader’s mind/mind-set as the information follows. Without stating that it is an assumption.

      Written differently would cause pause in the reader’s attention while they weigh up the strength of the assumption (pun intended-boom boom). How about, ‘In this paper we assume climate change is revealed principally by sea-surface temperatures when assessed with the data combined from all the oceans, NH-tropical-SH.”

      Note how the same info is provided, but the assumption and caveat declared.
      Problems with the sentence used include the term ‘climate change’ which should be defined, so as to differentiate from the term ‘a warming climate’. Use of the term ‘fundamental expression’ sounds very rock solid, but the thesaurus goes nuts for both words and ambiguity exists. For good measure, the word ‘global’ is used – which has a reflection in global warming, which we have already been unconsciously thinking since reading the term climate change.

      cheers, John

      • I have seen the same tactic used elsewhere. Often times they don’t support their claims because doing so would lead the reader to information that downplays the claims.

    • Ocean heat content is a measure of climate change, but sea-surface temperature tends to follow it. As with anything climate, you have to look at long-term averages rather than year-to-year variations, and these have long-term trends showing a climate change.

  13. Again we see that what is really a difference in determining the actual numerical metric — simply saying the measurements disagree with one another and we are curious about that — is expressed as a difference in “linear trend” — which is not a measurement but an artifact of beginning and end points and, of course in this case, a direct result of the end point being at the height of a (super-?) El Niño.

    The trend is artificial. (as are all trends….) See Cowtan’s page with the graph of comparing ship and buoy records. When the last El Niño in included, the ship record shows a graph absolutely flat to 2014. Only the 2015 El Niño data create the upward trend. It is a shame the longer term data is not included — leaving us with a rather truncated data set to look at.

    In the Kent et al. paper, they included this plea:

    Lost datasets – can you help?
    Over the years there have been several studies either comparing SST measurements made by different methods or detailed wind-tunnel and ship-based assessments of temperature change from buckets. We have learnt a lot from the papers and reports describing these experiments,
    but much more could be done if we were able to track down the original measurements. We’ve tried, and failed, but still hope they are out there and someone knows where they are. And of course if you know the whereabouts of any similar measurements we’d be delighted to hear from you

    Judith is quite right — we can’t really know what is going on unless the data sets can be sorted out — and some simply will not be sorted, the data is missing, uncertainty in the original data is far greater than the differences being measured, etc.

    • Steven Mosher

      well we can know that karl did not commit fraud as some have accused.

      ‘Judith is quite right — we can’t really know what is going on unless the data sets can be sorted out — and some simply will not be sorted, the data is missing, uncertainty in the original data is far greater than the differences being measured, etc.”

      1. We really CAN KNOW that Karl did not commit fraud. What do you say? take a stand and separate yourself from the tribe who claim he is a fraud.

      2. Not sure how you can know that the uncertainty is FAR GREATER than the differences being measured.. point me at your code and data which demonstrates this. or are you just making stuff up.

      • Mosher ==> All early SST data — ship based — was by human read thermometers which to my best understanding were recorded as whole degrees C. That means that the recorded temperature is really a range of one degree surrounding the the number recorded. The 1 degree C range of the original measurements is greater than the ~ 0.5 degree differences in the current data — no number derived from those ranges can correctly be stated without indicating the the result itself represents a range, one degree wide.

        Only in the digital buoy age have SSTs been recorded to a tenth of a degree.

        Kent et al. explain much of this in their paper.

      • That means that the recorded temperature is really a range of one degree surrounding the the number recorded

        That’s not correct. Even if your individual measurements have an uncertainty of +/- 1 degree, you can take multiple measurements, and the average will to converge to the expected value. Here, the average we’re looking at is the spatial average of the SST measurements.

        The mathematical name for this is Bernoulli’s theorem, or the Law of Large Numbers.

        Say you’re rolling a die, over and over. Over time, the average of the rolls will converge to 3.5, even though the range for an individual roll is 1-6. Same principle. In short; the low and high fluctuations cancel out.

      • BW you are only correct when the measurements are of the same physical item. As the measurements were taken at different times and different locations, each of which is a different physical item, there is no precision to be gained from multiple measurements.

      • Benjamin Winchester ==> I disagree. If you have multiple measurements of the same point in time and space, then the average is towards the expected value. However, measurements of different places at different times is an entirely different matter.

        100 thermometers in my backyard, each with some unknown error range, all auto-magically checked at the same moment, then averaged would tend towards the correct value of “temperature in my backyard” at that moment.

        However, temperature measurements were NOT with random individual errors. They are only recorded as rangers. The weatherman looks at his thermometer and writes 72. what he means is that what he saw on the thermometer was between 72.499999… and 71.5….anywhere between. This means that the temperature could be any value in the 1 degree range. The natural world does not have a tendency (or any physical reason) to favor even whole number degrees, therefore, the noted temperature, in each case, is a range 1 degree wide, by convention, written as the whole number that is the center of that range.

        When averaging data that are each a range, one has to average all of the possibilities for each datum…in the end, one discovers that the possible results range from a particular number (the average) up to the average plus 0.4999999 and down to the average minus 0.4999999… . A range 1 degree wide.

        Since the original measurements are ranges, there is no “error” to average out..the range must be maintained. Whatever the mathematical mean is with the range of the original data added on, such as 72 +/- 0.49999999….a range, not a whole number.

        Similarly, if our average turns out to be 72.68946529857 you still must include the +/- 0.499999 so your result is a range of the same size as that of the original measurements.

        You can not average the range away with wishful thinking.

        If we can’t agree, I’ll write an entire post on this. There will be a tremendous fuss over it, but it remains true.

      • Benjamin ==> forgot the paragraph:

        SST measurements are not multiple measurements of the same thing at the same time. They are lots of individual measurements of different times and different places with different equipment, all with their own biases and original measurement errors and original measurement properties (some are recorded as ranges of 1 degree width, some recorded as ranges of 0.1 degree width and certain amount of uncertainty to be added to that, some recorded an hour late by a tired ship’s Quartermaster at 0300 hrs who only glances at the thermometer which hasn’t ever been calibrated since manufacture (and has a measurement error of a degree or two).

        Thus, any claim to have found long-term changes in sea surface water temperatures must show the real ranges of the early data and add to that the expected error bars . More recent data is coming from digital equipment which is expected to be more accurate. Krigged data has additional error bars to be added (see the original BEST supplemental paper for the land krigged data error range).

        All in all, findings of differences to tenths of a degree C for SST of the Earth’s oceans is dubious at best, especially with no indication of uncertainty or error.

      • Kip

        Your 10 .53 post.

        I wrote a series of articles on the temperature uncertainties. This one best sums up the one regarding land and the wide range of readings that could be obtained from a single thermometer.


        If you have even half a dozen thermometers in the same place it is quite reasonable to remove the outliers and average the rest in order to get a reasonably accurate result of that days reading. However, only one reading was taken beset by uncertainties and inaccuracies


      • Kip Hansen, take the dice example. The more times you roll a die, the closer the mean of all your rolls gets to 3.5. It’s like that. More is better than less when you do the statistics. Ten thermometers are always better than one, and even if you only read them to the nearest degree, the final accuracy can be less than a degree.

      • Don’t be silly Jimbo.

        A dice gives a discrete reading with no possible range of error.

        When it reads – say – three – that’s it. Three is is.

        When a standard thermometer such as is under discussion reads three, it actually means 3±0.49999.

        There is no comparison whatsoever.

      • If one thermometer reads 10 C, but half of ten thermometers read 10 C and the other half read 11 C, then the mean of 10.5 C is more likely near the true temperature than 10 C. More thermometers would just confirm it even more. More measurements always narrows the error.

  14. Steven Mosher

    Looks like Karl was right.

    • In what regard?

      I think the thing that Karl did that rankled so much was the choice to reduce the accuracy of the most accurate buoy data rather than to try and correct the least accurate ship data.

      Zeke says that doesn’t matter to trends, and that makes sense, but it still is philosophically egregious to reduce accuracy in absolute terms, rather than to improve it. And it points out the unknown conditions of much of the old ship measurements which is why making physical based corrections to ship data is so problematic.

      The recent analysis ( which is more and more dominated by buoy measurements ) doesn’t solve the uncertainty of past ship based measurements.

  15. Steven Mosher

    A lot of folks accused Karl of fraud…but when you look at data preferred by skeptics…like satellites…He looks to be vindicated.

    Witness the difference between skeptics who were certain he was wrong and scientists like zeke who looked at the data

    • You seemed to know this would be the result from the start.

      • Steven Mosher

        hehe.. Been waiting a long time for this one to get through all the reviews.

        And Another shoe may drop.

        Its kinda funny, Some skeptics on some occassions raise interesting points..

        BUT they never do the science.. they just point at a problem and move on thinking they have done science.

        So.. if you pay attention and then go actually dig.. you can find all sorts of interesting things.. and shoe or two.. at least

      • richardswarthout

        Mosher you were quite critical of satellite records up to now. Now you love them because they mirror Karl’s technique of adjusting bouy data to look like bucket data – sort of like putting lipstick on a pig.

        Do ya think the satellites got out of adjustment?

    • I’m looking forward to the day that Mosher gets past the BadSkeptic/GoodWarmer template for all his comments.

      It’s a little predictable.


      • Steven Mosher


        Will you accuse Karl of fraud?
        Do you prefer satellite data?
        Satellite data confirms Karl..
        Do you want to deny satellite data now?

        your move.

      • “Will you accuse Karl of fraud?
        Do you prefer satellite data?
        Satellite data confirms Karl..
        Do you want to deny satellite data now?”

        False dichotomy.
        You knew that, but did it anyway.
        Files to throw back at you the next time you make an outrageous claim (next post?)

      • Enh, Mosher’s BadSkeptic/GoodWarmer template is really just another way of saying “people who don’t do their DD” and “people who do”.

        It’s appropriate to challenge skeptics to back up their claims with evidence; same as for everyone else. But most people won’t actually do any real analysis. That’s true on both sides.

        Mosher and Muller actually got out and did the work necessary, and they added something to science, showing that the other surface temperature series were solid. They did far more work, and with a far more rigorous analysis, than those claiming the science is fabricated.

      • Danny Thomas

        Mr. Winchester,

        If I may interject. Yes, I appreciate Mosher’s/Mueller’s contribution. The area of concern is prior to the Karl work confidence was high that temperature representations (code and data) based in science (to date) was a reasonable proxy. After Karl the same can be said (and has been adamantly here). One is quite obviously less of a representation than the other.

        Based on that evidence it’s equally “appropriate to challenge scientists to back up their claims”. Agreed?

      • The area of concern is … was a reasonable proxy. After Karl the same can be said (and has been adamantly here). One is quite obviously less of a representation than the other.
        Based on that evidence it’s equally “appropriate to challenge scientists to back up their claims”. Agreed?

        I’m sorry, can you rephrase that? I’m having trouble following. Probably low blood sugar. Specifically:
        “a reasonable proxy” for… actual temperature?
        What do you mean by “less of a representation”? Representation of what?

        I do agree that it’s always appropriate to challenge people to back up their claims. But then you have to look at the evidence they presented, understand what they did, and be able to critically analyze it for yourself for its strengths and weaknesses.

      • Danny Thomas


        In other words, Karl was ‘skeptical’ of the earlier data sets. Those same data sets were significantly defended. Karl applied adjustments. ‘New’ data sets now significantly defended. Both cannot be accurate.

        To directly address your questions, in both cases they are ‘representative’ of ‘actual’ temps. Apologies for lack of clarity.

      • Even prior to Karl, there was a suggestion that the shift from ships to buoys would make a difference, so Karl just quantified what was already suspected.

      • Danny Thomas

        Okay. What was the conversation involving satellites?

      • I am sure there was one. Work like this usually builds on discussions over time. People would be highly skeptical if it came out of the blue.

      • Danny Thomas

        Jim D,

        It’s just entertaining to me that those labeled as ‘skeptics’ have not reached a level of comfort with the temperature ‘data sets’ yet when Karl makes an adjustment (obviously he was ‘a skeptic) it’s defended as if it’s some sort of expectation. Yet none of the temperature series are even defended by those such as Mosher as being comprehensive, measurement based, and anything more than (IIRC his words) an estimate. Yet here we are. Sounds to me like the ONLY reasonable position is skeptical. Obviously I’ve missed something.

      • No one sees it as a completed job, but each adjustment is smaller, so it is converging. The key for confidence levels is having more independent datasets agreeing with each other as we have here.

      • Danny Thomas

        Sure. The concept makes sense. However, at the end of the day we’re still discussing something which is an ‘estimate’ for something known as ‘global temperature’.

        And ‘each adjustment’ is still an adjustment of something which is considered ‘state of the art’. Even here, Zeke et al are leading us towards an indication of accuracy. So should that mean no further future adjustments?

        Art seems an appropriate term.

      • Any proper presentation should have error bars, and future estimates should reduce those and remain within them.

      • It’s been awhile so correct me if I misremember, but Karl first adjusted the buoy data is match earlier data (i.e. warming the buoy data). Then because the buoy data is considered more accurate, he weighted it more than the ship based records. The result was the pause disappeared, but perhaps unexpected second result was the long term warming rate dropped. So, no pause, but also no worrisome rate of warming.

        I never used the term fraud. I did say that Karl’s method seemed very counterintuitive. Why adjust the more accurate data set and then weight it because it is the most accurate? Whether it was the case or not, his paper gave the impression of being produced for a specific purpose. The following refusal to respond to inquiries’ added to that impression. As I said above, not fraud, maybe not even bad science, but if politically motivated, something to be concerned about.

    • ==> A lot of folks accused Karl of fraud…but when you look at data preferred by skeptics…like satellites…He looks to be vindicated. ==>



      Plausible deniability strikes again?

      • As I have said since the Fyfe paper came out, and all the multiple other papers that have discussed the hiatus/slowdown/pause, usually in terms of explaining its cause, these scientists have no substantive disagreements with one another about anything to do with the theory of global warming and its pace. They’re not going to sit at a table and call each other names over the almost completely dead pause in global warming.

        The science explaining the pause contains a chilling message, which is included in the very important paper about the possibility of prolonging the pause:

        …One extreme example shows a warming of almost 1 °C in 15 years—a much greater 15-year warming rate than has occurred in the observations to date (red curves). These spring-back warmings illustrate another important potential consequence of strong internal multidecadal variability as simulated in CM3, and reinforce the need to better understand whether such internal variability actually occurs in the real world. …

        In the old fashioned sense of cooling: as in, when the GMST actually goes progressively downward for a few decades: as in, you know, the cooling phase of a ~65-year cycle… that has not happened since the late 19th century. The AMO is stuck on up and cannot fall down… same as that sputtering old Klunker: Looming Lobal Kooling, as it exists only in dreams.

        The stadium wave has come and gone. We’re in a heatwave. It could be near its end; it could last for another decade. Nobody knows. Several papers point to the Pacific having changed regimes, and that points to more warming soon.

    • Which “folks” specifically accused Karl of fraud? Name names for a change.

      • First, does the word “fraud” have to be used in order to accuse a scientist of committing fraud?

        A definition:

        Fraud – A false representation of a matter of fact—whether by words or by conduct, by false or misleading allegations, or by concealment of what should have been disclosed—that deceives and is intended to deceive another so that the individual will act upon it to her or his legal injury. …

        In my words, Karl was accused of rushing, at the behest of the White House, a scientific result that employed bad science in order to influence public policy.

      • jim2

        good question. Who here accused Karl of fraud? I certainly didn’t. I have consistently said on many occasions that I don’t believe scientists are routinely making fraudulent statements, creating a hoax or indulging in a conspiracy

        Instead of wild accusations let Mosh or others name names of those denizens who thought Karl was being fraudulent in the sense that we all understand the meaning of that word and let them then provide their defence.


      • You can read the article linked by Joshua just above. This was typical of the questioning of integrity that was done in the wake of the Karl paper.

      • You can read the foxnews article linked just above. This was typical of the questioning of integrity that was done in the wake of the Karl paper.

      • Yes, I can read and the word “fraud” isn’t there. So, based on what you’ve pointed out here, I have to infer you believe scientists should be beyond question.

        You are wrong.

      • His integrity was questioned in articles like this for no other reason than they didn’t like his results. If you want to attack data, you need to come up with your own data to do that, not attack the integrity of the person who provided it. This was just plain old-fashioned mud-slinging. Nothing stuck, but no one expects any apologies.

      • catweazle666

        “I have to infer you believe scientists should be beyond question.”

        In Jimboworld that is only applicable to scientists who adhere absolutely to “the Cause”.

        The absolute opposite applies to sceptics.

      • People question the adjustments because ALMOST every time adjustments occur, the trend goes up. I believe there might be one exception, but it makes people wonder why the trend increases or the pause gets wiped out. I think this is understandable.

        But to your point, the criticism should be based on what he did wrong, and I believe there is some of that sort also.

      • They question the people, not the adjustments. That’s the point. If they questioned the adjustments, they would have a whole different set of questions from the ones we saw in this and other articles.

      • Danny Thomas

        “They question the people, not the adjustments.”

        Suggested previously. Remove the label (in this case, Karl) and publish the work. It would be interesting to see the results.

      • Danny Thomas

        “They question the people, not the adjustments.”

        Suggested previously that labels are of concern. Remove the labels (in this case Karl) and submit the work. It would be interesting to see the results.

        (Mods, if duplicate posts please remove. First attempt evaporated)

      • Karl became vilified because of this work. He did not have any kind of bad reputation among these people prior to that, although I guess some still retain a grudge for him quantifying the TOBS correction. If it was another author, they would quickly become the target (see Marcott for example) . It’s the work, not the name, that draws the fire to the author.

      • Danny Thomas

        Jim D,

        In some venues Karl was vilified. In others he was crowned. The point is that the producer should make no difference only the work should. Either the work is valid or it’s not. Karl was on his way to retirement so notoriety was unimportant.

        One of the topics recently relates to peer review. Removing the label (names) and put the work out and discussion should then be directed at the work. Continuing to do the same thing over and over expecting different results (responses) and all.

        Even the recent study about the narrative form of presentation impacts the level of citations should be a clue.

        Oh, and maybe Mosher is recalling this: https://stevengoddard.wordpress.com/2015/11/25/biggest-fraud-in-history-perpetrated-by-tom-karl-and-gavin-schmidt/

      • Karl didn’t need to be crowned. He already had a great reputation because of his career of careful data work. On the other point, it is very difficult to keep scientific authors anonymous to reviewers, particularly in a small field like climate science. Most people build on and cite their previous work making it often easy to guess at least who is associated with a paper, if not the lead author. That, and they may have presented it in a conference before publication.

      • Danny Thomas

        Jim D,
        “it is very difficult to keep scientific authors anonymous to reviewers”

        I’m not suggesting reviewers. That should be a given. I’m suggesting the work in it’s entirety should be anonymous. As you suggest, it’s not the authors that matter. Academia can keep track anonymously for the ‘publish or perish’ rule.

        Even Steig, w/r/t Antarctica stated +/- 50% wasn’t understood. That’s something a scientist should have the freedom to state. Peer review vs. peer pressure.

        Unless of course notoriety is the actual goal.

      • That is impossible because of conferences and seminars given by scientists to their peers. Most scientists would be quick to claim a high profile piece of work too. How would you even refer to such works anyway? By a code number?

      • Danny Thomas

        Sure. DOI number.

        And, why does the presenter of the work have to be the author? Do not many other cite works of others? Presentation should be doable if the science is sound. And defensible.

        Look. We’re discussing the problem of ‘playing the man’ and not discussing the work. Since there is some agreement that this is a concern (as adamantly expressed here by Mosher) why not try another route?

        Maybe after some period of time (2 years, or 5?) attribution becomes public. The work stands (or falls) on it’s own merits, and the science is the focus. Why not?

        If it’s as you suggest that ‘climate science’ is such a small world this seems like a great area to try the experiment. This would mean those more climate concerned and those less so would have to address the ‘code and data’. One of Mosher’s big sticking points!

      • It is playing the man that is wrong. There is nothing science can do about that because science is exercised in the open, not behind closed doors. What you are advocating would increase suspicion and have the opposite effect to more openness. It should be possible to call up a scientist directly and question their paper person to person. This type of open debate, especially around recent papers, is how science progresses.

      • Danny Thomas

        Jim D,

        “What you are advocating would increase suspicion and have the opposite effect to more openness.” How? Those called ‘skeptics’ are predisposed to ‘suspicion’ of government/academic money influencing results. Those called ‘alarmists’ are predisposed to fossil fuel money influencing the skeptical ‘frame’.

        Suspicion is already there. Openess is doable. In lieu of a lead author as contact a lead communicator can accept questions/requests and process thru to those incognito. Anonymous blog names (already being done) can be used for public formats. Same in direct responses to authors. Does it matter if the response comes from one named ‘unknown’?

        If Mann does work and publishes it’s automatically rejected by one side. If Willie Soon does work it’s rejected by the other. Many on the different blogs know this to be true. You might be an example yourownself.

        If the Karl work had been put forth anonymously would anyone have accused him of ‘fraud’?

        Just because it’s a different thought doesn’t make it bad. Think it through.

      • Scientists do actually talk to each other about their work, and whatever you do, you can’t stop that process. Why raise publications to a special level, when they can just discuss a conference paper or seminar on the same subject as a way around your roadblock/embargo? It just doesn’t work in the real world.

      • Danny Thomas

        Jim D,
        “Scientists do actually talk to each other about their work, and whatever you do, you can’t stop that process.” No desire to ‘stop’ that process. If you and I collaborated on a work this does not mean ‘the public’ is privy to that having taken place. Most work goes right on by without my knowledge until a paper is made public. I’m not informed of it being submitted, going to review, coming back from review with comments, modifications (if any), and resubmission. Peer review is traditionally anonymous. ” Traditionally, peer reviewers have been anonymous, but there is currently a significant amount of open peer review, where the comments are visible to readers, generally with the identities of the peer reviewers disclosed as well.” The open process involves ‘peer pressure’.

        Further: “In anonymous peer review, reviewers are known to the journal editor or conference organiser but their names are not given to the article’s author. In some cases, the author’s identity can also be anonymised for the review process, with identifying information is stripped from the document before review. The system is intended to reduce or eliminate bias.”

        “It just doesn’t work in the real world.” Show me, don’t tell me. If it’s been tried, I’m unaware. If it’s not been tried you are unaware.

      • I am saying trying to make published work anonymous in the real world is not workable. Often by the time it has been published, it has already been presented somewhere and discussed within the community. No scientist would want what you are asking for, which is probably why it hasn’t happened. I have not heard of a scientist wanting to be anonymous. If they did, perhaps they could get a colleague or Ph. D. student to publish for them, but I can only imagine that happening if someone feels their reputation is already shot.

      • Danny Thomas

        Jim D,

        ” I have not heard of a scientist wanting to be anonymous.” Is the science important or the scientist? Both? If it improved the discussion and reduced the declarations of ‘fraud’ is it worth consideration?

        “If they did, perhaps they could get a colleague or Ph. D. student to publish for them, but I can only imagine that happening if someone feels their reputation is already shot.” Again, not suggesting long term. Effectively suggesting an extension of the review process where names are not applied to the work until some time after it’s been publicized.

        Reviews have long been anonymous. Keeping the work within that framework for a period of time post publication should not be of issue if it’s really the science and not the scientist that’s of import. It’s a brave new world. New thinking might not be a bad idea.

        If it’s not been tried as suggested you cannot provide evidence the approach would not be beneficial in today’s politicized world.

      • There has to be a demand for a double-blind review, and there just isn’t, is the bottom line. I believe some minor journals within the field do have that, but probably none of the major ones. If someone wants to be anonymous at the review stage, they may have that option with some journals they could submit to. they can also suggest reviewers or people they don’t want to review their paper when sending it in.

      • Danny Thomas

        Jim D,
        “they can also suggest reviewers or people they don’t want to review their paper when sending it in.” Is that valid peer review? Part of the problem when considering ‘suspicions’ dontcha think?

        Extended review takes place much as what Henson (bravely) chose for his recent work. Put it out for all to see. Only difference would be had he left off his name for the near term.

        Nothing to fear from the ‘brave new world’ if the work stands. Where would we be in this discussion about Karl had he (being near retirement) just put the work out minus his name? He didn’t need the notoriety. Pretty certain the pension checks would still clear, and the conversation would not be about the politics.

        Not quite clear why you’re so adverse to the possibilities. It seems to me as the amateur that it’s worth discussion.

      • I am sure Karl did not mind that his name was on that work. He knew he could stand by it. Fringe views in the media do not impact the scientific community and putting your own name on your work is fundamental. What problem are you trying to solve? The scientists don’t care about it enough to remove their names from their own work. For journals it is up to the handling editor to pay attention to the suggested lists or not. It remains unknown to the author who they finally went with.

      • Danny Thomas

        ” What problem are you trying to solve?” The problem of accusations of fraud and or bias due to influence without consideration of the science. If it be Mann or Soon.

        Mosher’s comments about ‘code & data’ make sense. All that really matters is the code and data (the science IOW). The package is unimportant except to the packager. In the short term, the science is put forth for scrutiny. Notoriety can come over the longer term as academia needs.

        “It remains unknown to the author who they finally went with.” Not always.

      • Scientists always want their journal paper to bear their name and affiliation. You can’t deprive them of that. If someone starts a journal with no names, I would predict it gets almost no papers, except perhaps cranks who are not at all confident about their work, or are just doing it for political reasons. I would not trust a paper that someone did not sign their name to.

      • Danny Thomas

        So you’d consider any paper submitted anonymously as invalid if you were a reviewer therefore you’d reject it w/o comment? Strange, but that’s how I understand it’s done all the time. It’s only labelled at time of publication.

      • No, I would find it interesting that it was anonymously submitted and wonder why, so I would read it to find out mainly because that would be so unusual.

      • Danny Thomas

        Okay. Then apply that thought process over time, numerous papers, and as a norm.

        I recognize this is an off topic discussion but it’s interesting that no one else has chimed in with an opinion. I, for one, would appreciate others insights.

      • Are you talking about just review or actual anonymous journal articles? As I said, scientists have no reason to want their journal articles to be anonymous. They want their name on their work.

      • Danny Thomas

        Not sure how I can be more clear when I use the terminology of extended peer review. Name remains off until some time after publication. That time length could be a month. Long enough for publication and review in the ‘brave new world’.

        You keep saying ‘scientists don’t want’. I’m more concerned about the science than the tists. Had Karl who needed no notoriety published with no name, fraud would not have been politically suggested IMO. This would apply to all. You’ve indicated your own tendency towards bias due to authors name (they been right in the past). Sorry, but that’s not science.

        Just think it’s worth discussion.

      • I think your premise is wrong, and that Karl couldn’t care less about the effect of having his name on his work. If senators want to persecute him for his work, they will hunt him down anyway, so it doesn’t prevent that. This is why what you suggest is completely pointless, unless you can also figure out how to hide from Congress who wrote something, because that is where the s**t hit the fan as far as Karl would have been concerned. How would you prevent that?

      • Danny Thomas

        I don’t know how or intend to prevent (or assist) congress from doing anything. And if Karl was willing to take the heat (do you think he really expected to be called ‘a fraud’) is not important. Karl had(s) standing. Should the science be subject to open conversation post publication for being good science, or should the standing of the scientists affect acceptance?

        Did Karl personally take heat because of the quality of the work, because of the result, or because of his position within an administration? And did that heat benefit science? Your answer to this question will be of interest.

      • Scientists publish papers all the time. He probably would have expected pushback on this one, but he wasn’t intimidated by that. This is how science works. Publish what the data says and let the critics try to attack the science, and they did, but not really the science. Instead they attacked his motives for showing his results. Not his problem because the science was solid. Their problem. Could have predicted that Smith would do all kinds of subpoenas on him and NOAA? Probably not. Would he still have published knowing that? Yes, because he knew he was right and they were wrong, and politics should never stop science. This whole episode has been a net positive for Karl because in the end his has become the accepted view.

      • Also can you imagine the press release describing such a paper “Some scientists from somewhere and somewhere else are saying that sea levels are rising faster…” Is that what you want?

      • Danny Thomas

        Jim D,

        Press release? What does that have to do with the science? That’s only about the notoriety. If a work is put forth in a format such as this one, or RC or the like wouldn’t your expectation be that discussion be about the work? Would that be refreshing?

        There would be no need to be concerns about accusations of fraud by a person because of anonymity.

        Is that what I want? Sure. Why not? Do I really care if Mann puts out a new work or that he puts out work that withstands scrutiny? Is that what you want, or do you prefer name brands?

      • Papers often come with press releases and that is usually the first place the public would hear of them. Since those often come from the university to tout their important work, they would probably not be issued if the university could not be named because there is no longer any point. Again, no one is asking for this.

      • Danny Thomas

        Extended peer (and public) review. Notoriety comes later. Universities still able to tout their research ‘chops’ for recruiting. No one is asking for it? Has anyone asked in formal setting. Has it been bantied about? Got a link?

        Again, not suggesting a permanence? Just a delay before it’s labelled.

      • If any scientist wants this, you find them. It is not up to me. Journals would suffer too. Imagine a journal with no names and identifying information redacted for years.That would be a pretty sorry state to be reduced to by lay critics.

      • Danny Thomas

        I read stuff all the time from those who’s work I’m unaware. You’re more ‘experienced’ than I. Do you know all the authors you read? Do you have propensities for acceptance of work from those you know? Rejection? If you don’t then anonymity matters not. If you do, that’s exactly the bias for which I’d suggest this method might relate.

      • Knowing the reputations of people or groups behind the papers helps in gauging them for sure. It is like you may have certain journalists or editorials that interest you more than others because they have been right in the past, or at least have been somewhat sensible in their opinions. Similarly, the author’s name is a big part in deciding whether to read a paper first in a new journal issue, apart from the relevance to your own interests.

      • ==> People question the adjustments because ALMOST every time adjustments occur, the trend goes up. I believe there might be one exception, but it makes people wonder why the trend increases or the pause gets wiped out. I think this is understandable. ==>

        Bull. People question the adjustments because they have a predisposition to (1) think that the adjustments are being made by fraudsters and (2) reject any information that puts in doubt the existence of a “pause.”

      • ==> good question. Who here accused Karl of fraud? I certainly didn’t. ==>

        This is what I meant by “plausible deniability” “We think he should be investigated by the government for conspiratorial advocacy science, and he’s an” alarmist” just like others we have called frauds thousands of times, but we didn’t call him a ” fraud. “

      • Steven Mosher

        “They question the people, not the adjustments. That’s the point. If they questioned the adjustments, they would have a whole different set of questions from the ones we saw in this and other articles.”

        Yes precisely. Which is WHY they go after the emails rather than just looking at the data. Which is Why people like Rud, rant about whistleblowers..

        Blowing the whistle on WHAT?

        We have the data
        The method is clear.

        Do you guys really want all the details of who used the exact words fraud or misconduct?

        But you all, especially tony b, Miss the larger point

        If skeptics don’t POLICE THEMSELVES, then quite simply you will all be lumped in with the most extreme members. Sounds unfair I know.
        The same way Gore is seen to ‘represent” climate science, you most extreme members represent your tribe. Like it or not. fair or unfair. That’s just the reality of politics. Tough.

      • catweazle666

        “We have the data”

        We had.

        Until you dodgy lot mangled it with your AlGoreithms to match the output from your silly computer games…

        “The data doesn’t matter. We’re not basing our recommendations on the data. We’re basing them on the climate models.”
        – Prof. Chris Folland, Hadley Centre for Climate Prediction and Research.

        But hey, what fool would trust the temperature reading from a $10 thermometer when they’ve got a $100,000,000 supercomputer game making temperature data up, right Mosher?

      • Mosh

        What is this larger picture that all of us, especially me, miss?

        I consistently confirm my belief that scientists do not indulge in hoaxes, conspiracies etc, yet you try to tar us all with the same brush by this insistence we all shouted fraud.

        As you know, I object to incomplete, inadequate or unusable data being used to help promote an alternative viewpoint of history and I certainly think there is an element of group think. There is also a failure to look at the historical context, that of continual rises and falls throughout the Holocene, which allows us to put the current three hundred year long warming into its context

        . But you will have to spell out this larger picture, preferably at the same time naming those you say are calling fraud here.


      • Steven Mosher

        “The emails will at a minimum show how Karl and Huang hid that in their papers, three levels down. Lamar Smith’s committee knows about this because I posted the sleuthing in a comment to Judith’s post on Karl 2015 at CE, then wrote it up and sent it to his oversight committee. Karl and Peterson are toast, and they know it. Hence the subpoena stonewall by NOAA comprising clear contempt of congress, 2USC192, enacted in 1938.”

        too funny.

      • Questioning emails is fine. But as far as policing goes, skeptics aren’t an organization like, say, the AGU. There’s no central authority to police anything. And I know I don’t read everything, everywhere that self-identified skeptics write. And even if I did, it ain’t my job to “police” anyone.

      • Steven Mosher

        tony b

        you were one of the first offenders on the previous thread.
        Your preferred tactic is to
        A) say something about uncertainty
        B) Insinuate something about the authors in the form of a question
        ( wasnt this author etc etc)

        That way you can stand here and pretend to be clean
        but you are not.

        naming names?

        Bob Tisdale
        Anthony Watts

        Its pretty dang simple. Without exception the leading figures of skepticism accused Karl of lying (Tisdale), Misconduct and fudging(Rud)
        Changing data for political reasons (Watts) and virtually everyone on the prior thread accused them of conflict of interest.

        very few people made any substantive comments on the actual science

        Now.. the work is done.

        The karl data has been compared to New data never used before.

        he was correct

        Now contrast the reaction Zeke and I had ( on the first thread) with everyone else.

        1. I said I withhold judgement until I can study it. Rud abused me for taking my time.
        2. Zeke said he would have to look closely at it.

        What The hell do you think happened after that?

        basically Judith and other commenters ( not you) made some suggestions, gave some homework..

        So now folks have the answers to the questions SOME SKEPTICS asked

        What about Argo? ( karl is right)
        What about satellites (Karl is right)
        What about other series? The old NOAA series was the worst.

        Reminds me of another time when skeptics made suggestions to look at temperature series..

      • Steven Mosher

        ““They question the people, not the adjustments.”

        Suggested previously. Remove the label (in this case, Karl) and publish the work. It would be interesting to see the results.”

        Yes on the original thread even “tony b’ was casting aspersions on certain authors..

        Never the science, play the man

      • Steven Mosher


        It’s simple enough. When you see folks discussing the man and not the science.. Point out that they are not following the scientific method.

        you enjoy making that sort of point..

      • Mosh

        Too funny. So now you are referring to something I said in a previous thread and reading all sorts of things into it. Please link to the comment instead of making vague generalisations and accusations.

        You have been flailing round for days shouting fraud.. Bob tisdale and Anthony watts do not habitually post here so why go round accusing all the denizens of fraud and demanding we recant something we never said?

        If you have an issue with what Rud said please take it up with Him and stop tarring us all with the same brush. As you well know I consistently refute the notions of fraud etc and get it in the neck for my troubles.

        Now, please provide the link to a previous thread where I am supposed to have done something dreadful.


      • “They question the people, not the adjustments.”

        It was inevitable that a comment like that would generate a reply like this:

        “Which is WHY they go after the emails rather than just looking at the data.”

        Data requires interpretation. Interpretation, in this context, is done by humans and even more contextually, it was Thom Karl who was doing the interpretation that led to his (their) adjustments. Some are just being disingenuous by suggesting an either or scenario, i.e.; “they’re going after the emails instead of just looking at the data”, assuming, either by arrogance or some other malady, that Lamar Smith has not looked at the data, or at the very least listened to advisers who have looked at the data and offer a different interpretation than that of Huang and Karl.

        How does one simply “look at the data” to know how effectively those making the adjustments reigned in their own bias? It is not as if Karl’s dismissal of the hiatus was met with universal approval by scientists and even Micheal Mann, joining Fyfe et. al, found disagreement with Karl on that matter. Is it being suggested that Fyfe, Meehle, England, Mann, Santer, Flato, Hawkins, Gillett, Xie, Kosaka and Swart have failed to “look at the data” in order to arrive at disagreement with Karl?

      • My recollection is that Karl’s team came to visit Smith to answer all the data questions they had, but Smith just sent his underlings to that meeting. This makes it seem that they really weren’t interested in the scientific details. That was the facade for pure political sabre-rattling, and perhaps to intimidate any other scientists who had thoughts of putting real results on climate change out there. (Look out, Zeke).

      • “This makes it seem that they really weren’t interested in the scientific details.”

        That’s one interpretation of the data. The ‘data’ being, in this instance, your recollection.

      • The most likely scenario is that the skeptic tanks were telling Smith ‘this is killing us, do something’, so he did.

      • “The most likely scenario is that the skeptic tanks were telling Smith ‘this is killing us, do something’, so he did.”

        Did the skeptics go to Fyfe and Mann, et. al. as well?

      • As far as I know, Fyfe and Mann weren’t asking for his emails. Even Zeke was skeptical which is why they sought independent ways to verify Karl’s result, but it proved solid. This is how science operates. Not via subpoenas.

      • “As far as I know, Fyfe and Mann weren’t asking for his emails.”

        You’re better than this, Jim D. You don’t have to deflect and obfuscate if your cause is just. The Fyfe paper disputes Karl’s findings of no evidence of a hiatus. Presumably this is what you meant by the skeptics going to Smith and telling him “this is killing us”. You claim that such a scenario is “most likely” but in order to derive such a scale you necessarily have to ignore paper’s like ‘Making Sense of the Early 2000 Slowdown’.

        It is likely your “most likely” needs another review of the data and then some adjustments.

      • “You’re better than this, Jim D. You don’t have to deflect and obfuscate if your cause is just.”


        Best laugh I’ve had this week!

      • I don’t know about Fyfe, but I guess they were having a scientific debate which is the way it should be done. I am sure people will still question Zeke’s results too. At least with Fyfe and Karl you can compare their arguments side by side using actual data, which can’t be said for Smith. I found this for more information on Fyfe. He didn’t argue about the buoy correction, which was the main thing Karl added.

      • As you can see from that, whether it is a slowdown or a continuation, depends what base period you use for the previous warming. Fyfe refers to a slowdown, not a hiatus, but that is relative to a faster-warming base period. These are just details of definitions of periods and not fundamental disagreements.

      • Jim D.;

        Let me address the latter of your back to back comments.

        “Fyfe refers to a slowdown, not a hiatus…”

        Read harder! Fyfe uses the word at least 16 times and the first time they use it is in their first sentence.

        As to your former comment, your conflation of Smith with Fyfe and Zeke is what I mean by you being better than that. Lamar Smith is a Congressman and has different functions than scientists do. Rule 607 of the Federal Rules of Evidence should help you better understand that difference.

        If you cannot be bothered to read that for yourself, here’s a hint. Rule 607 acknowledges the right to attack the credibility of any witness.

      • Even Fyfe’s title is ‘slowdown’. He often refers to the ‘big hiatus’ as a term of reference for the long undisputed hiatus period up to the 1970s, and the other ‘hiatus’ as one that he doesn’t actually agree with as a hiatus. As for Smith, the motives of the person should not be attacked unless first you can prove they said something wrong in the first place, which he didn’t, so he jumped the gun on that one.

  16. Reminiscent of the Large Hadron Collider statisticians torturing the data to tease out a tiny little bump that they “identified” as the Higgs Boson. Nobel prize to follow (in math). Does anyone believe that a measured change of 0.1 degree in the average Earth temperature means anything?

    • Does anyone believe that a measured change of 0.1 degree in the average Earth temperature means anything?

      I do, it means the alarmists are desperate to promote fear to advance their alarmist agenda because that is money coming in to them from our tax dollars.

    • Steven Mosher

      Does anyone believe Karl Cooked the books when independent satellite data shows he did not?

      • That’s irrelevant if you read the previous comment.

      • Steven Mosher

        crypto do you believe Karl Cooked the books when independent satellite data shows he did not?

        Will you stand here and accuse him of fraud?

      • Hold on a moment – give people time to look into that paper.

      • Mosher, “Does anyone believe Karl Cooked the books when independent satellite data shows he did not?”

        I believe that ERSSTv3b dropped satellite data for interpolation because it showed a “cooling” bias and ERSSTv4 increased weight for buoys and satellites to eliminate a cooling bias associated with ships. This is a lot like all the adjustments required the land temperatures in order to make Karl et al.s methodology match others when no TOBS adjustment is required when one uses a better methodology to begin with. Making you work “fit” after the fact is very similar to cooking the books.

        Making your work fit after the fact – X numbers of dollars, Not screwing up to begin with – priceless.

      • Mosher,

        You can massage data to say anything you want. Karl published what he did. Not fraud. Quit flogging a dead unicorn.

      • Let’s just call it statistical sophistry shall we, Mosher”

        You know, as in “lies, damned lies and statistics”?

    • “Reminiscent of the Large Hadron Collider statisticians torturing the data to tease out a tiny little bump that they “identified” as the Higgs Boson”

      Idon’t know what you are talking about… the Higgs boson statistical prediction based on handful of events was perfectly correct, as confirmed by the thousands and thousands of more bosons generated at exactly the energy predicted by the announcement.

      • I think that’s the point.

        No matter how much data confirms Karl’s original work, it’ll always be wrong if it tells them what they don’t want to hear. Just like the Higgs Boson announcement is “wrong”, even though normal physicists consider it verified.

      • Says Benjamin who totally failed on his Bernoulli’s Law explanation earlier in this thread.

  17. Averaging the sea surface temperatures around the globe to create the global anomaly makes little sense, particularly if that is done over a short period of couple of decades.
    About 3 years ago (it needs updating) using the NOAA data, I compared the SST for two strips of the Atlantic Ocean at high latitudes (north and south hemispheres).
    Rise and fall of the SST in the two hemispheres periodically moves in and out of phase.

    • v ==> Yes, that’s why I say “I have serious doubts about the physical reality of this statement in the abstract of the Kent et al. paper: “Global surface-temperature changes are a fundamental expression of climate change. “

      More of the silliness of “single-number metrics” — One Number to Rule Them All.

      • Fundamental statements should only be made after most or all uncertainties have been resolved or lease known not to be settled. Now how does one accomplish this when the science is claimed to be settled and “settled” expects quietness? By mere definition, “settled” is never correct to claim.

      • Steven Mosher

        1. Address the following

        A) Skeptics claim Karl Cooked the books. With No analysis to back this up.

        B) Zeke looked at new data from satellities and Argo.

        C) The data show that Karl did not cook the books,

        Where does Karl go to get his good name back, and when will you and other stop making false accusations?

      • Yes, and doesn’t it not assume complete coverage of the ocean with sensors?

        As an analogy, look at terrestrial ecosystem assessment methods using different indicators to describe an attribute, such as erosional patterns to describe vegetation function. It is repeated ad nauseum: “A single indicator cannot be used to describe an attribute.” That is on scale ranging from a few to a few thousand acres.

        And here were are assessing the state of climate based on a single global value? That’s nuts.

      • Mosher ==> Again you are mistaking me for some imaginary enemy. I have never made a “false accusation” against Karl (or anyone else to my knowledge). I have not written about Karl or any of his papers.

        You really need to rein yourself in — you are the one who makes false accusations with your wild generalizations of “skeptics”.

        That kind of utter nonsense is what makes the comment sections of Climate Science blogs resemble elementary school playgrounds — with opposing teams of bullies shouting tribal-based insults at one another.

      • Steven Mosher

        So Kip

        Since the satellities back up Karl, What is you opinion of people
        who claim he was cooking the books.

        Tell us what you conclude.

        Do side step . Answer.

      • Mosher ==> Again, you mistake me for someone who would bother to answer your weird questions.

        You know very well that I have no interest in, and do no engage in, these silly Climate Warrior games.

      • Steven,

        Are you ok? You appear to be hunting windmills.

    • If it is assumed that the NOAA data is good enough (within the reason) then:
      – if two hemispheres are compared like for like (longitude x latitude is 10 x 10 degrees in both cases) these two parts of the same ocean do their own thing regardless of what CO2 was up to.
      – while global ‘warming’ was apparently surging ahead these two patches of the Atlantic refuse to comply.
      – ergo: sun provides the energy to keep oceans warm, but the oceans’ currents govern how the absorbed energy is distributed.

      • My central problem with NOAA is (example) they state that month X of a given year was the warmest ever by 0.04 degrees C. When in fact, they and we all know that really means that it is unknown which month (or other similar months) were the warmest. Then continually stating that global temperatures have risen in Y number of consecutive months.

        Also, when subtracting 2016 readings from 1890 or 1900 or 1920 etc. are apples and apples. Similar technologies? Just adding more and more measurement points across the planet, over time, might show increasing temperatures… and if it did, should not all prior readings be incremented knowing that the lack of more measurement positions caused the earlier readings to be lower than actual?

        NOAA joyfully proclaims rising temperatures per month while joyfully promote the rise over the last quarter when a month drops. It just does not ring well with me.

        Scientists (myself an engineer 46 years) can do great things in their studies … but promoters and soothsayers they shouldn’t.

        A president may determine that action should be taken per scientific information. But at the same time scientists of that input should correct that president when states that the science is settled and the debate must end.

      • Damn vuk,

        That should be obvious to any high school grad. That you have to tell people that doesn’t speak well of your audience.

        Seriously, one doesn’t need to have Vuk’s level of qualifications or intelligence to understand this.

  18. It would be useful if this analysis included some spatial representation of the differences of coverage for each data set and how that might effect the differences between them.

    It appears that COBE includes ice covered Arctic/Antarctic while buoys and ships do not.

  19. Steven Mosher

    • If “scientists” had a clue what was going on with sea temperature they wouldn’t need to be making constant ‘adjustments’ ‘fudge factors’ ‘bias adjustments’, etc…
      I mean, just saying….

      Its more like archaeology than science it seems.

  20. Global warming: our children may never know snow, sleet or ice again… bring it!

    • Steven Mosher

      do you believe Karl Cooked the books when independent satellite data shows he did not?

      Will you stand here and accuse him of fraud?

      • Danny Thomas

        I’ve read this question numerous times in this thread. I’ve seen no one here accuse Karl of fraud. Can it stand at that?

        It seems a more pertinent question is what were the errors which caused the prior data sets to be in need of correction and have methodologies been implemented in the numerous (4-5) major temperature data generation processes to insure those errors have not been embedded elsewhere in the records, nor will they be in the future?

        IIRC some stood by the prior temperature data sets as being representative of actual. Yet apparently, based on satellite confirmation, those records were rightly suspect.

        But according to some it’s poor form to debate via question.

      • I’ve read this question numerous times in this thread. I’ve seen no one here accuse Karl of fraud.

        Watts did, so perhaps Mosher should troll there not here.

      • The guy who frequently uses the term “karlized” on this blog hasn’t shown up yet.


        He doesn’t use it complementarily.

        He won’t admit to you Steven that he is wrong.

      • Are you NOW suggesting that sat data is better than surface?
        Aren’t you the one who said sat data has MASSIVE structural uncertainty, therefore making surface data (presumably BEST) better?

        Let’s see – was Gavin’s Hottest year eva!” claim fraudulent? (being only 38% sure and all)
        Is a 0.02C increase from previous record something to worry about?

        Perhaps it is better to call Mosher a fraud and be done with it. Yes, let’s do that… Mosher, you are a FRAUD! You provide short, sharp “Wrong!” answers, then refuse to expand, even when asked. When you ask, refusal to answer is often taken as proof of vexatious claims. Sauce for the goose and all that.

      • BD, wrong. I did, several times already. And stand behind my post thereon over at WUWT, despite the goof of actually assuming a NOAA graph had a properly metriced Y axis. ( The offical version didn’t, and I failed to catch. Commenters did. So I commented a correction. Did not change the conclusion.
        As a general recommendation, never refight past wars you have already lost once. Doesn’t go well.
        Jeez, at least know your enemy’s name. And Books. And pedigree. You need to reread Sun Tsu’s The Art of War. Because you are failing.

      • ZH says they used three independent records, not just satellites. All three… as SM often says… same answer.

      • Lamar Smith is one that needs to weigh in about whether this independent confirmation finally satisfies him. He won’t, of course. Now he’ll ask Berkeley for their emails in a more likely scenario.

      • Climate fraud is just a tenth of what it used to be…

      • Wagathon means that the more they look for fraud, the less they find.

      • Steven Mosher

        It’s easy guys. If you don’t believe he committed fraud say so. Then you are on record.
        Better yet.. call out skeptics who do make these claims

        As it stands the transition team is taking advice from ball and goddard… they represent your flock. .like it or do something.

      • Steven Mosher

        Are you NOW suggesting that sat data is better than surface?
        Aren’t you the one who said sat data has MASSIVE structural uncertainty, therefore making surface data (presumably BEST) better?

        See the difference between a MSU and the sensor used in this study.
        See the difference between a series of 9 satellite stiched together and one series

        You won’t

      • > And stand behind my post thereon over at WUWT

        Linkies would show even more courage, Sir Rud.

      • Can we compare Karl and UAH for the period 1979 to 2015?

        Therein lies the answer to Steven’s question.

        To me it looks like Karlization give the most accurate result, because the trends differ by 0.001 and Karl’s has 30% less uncertainty

      • Depends which version of UAH you use. He tweaked it recently. Just a tad.

      • “You won’t”

        So it’s less than 30 years then – so weather, not climate.
        Like all alarmists, you screech and complain when short-term trends are presented as climate-significant, and then crow about how the latest month is a record breaker; how the satellite record is “dodgy” until it suits your needs, then it becomes “proof”; etc ad nauseum.

        It’s one thing to play devil’s advocate, it’s quite another to be hypocritical to try and “score points” – you’re doing the latter more and more, and frankly you’ve gone from some-one whose comments are worth reading to some-one I’m more likely to skip over. All bluster, gotchas, “read harder”s and cryptic rubbish – sad, really, because you can be so much better than that.

        So, for the record: anyone who uses lower resolution, higher uncertainty data to adjust higher resolution, lower uncertainty data is a fool or a fraud. Anything that gets through peer review doing so, is published and cited as authoritative, is foolish or fraudulent in a field of fools and frauds. So which is Karl then – fool or fraud?

      • You keep walking further out on that limb. Several have already said they don’t think fraud. Now it is on you to identify who you think did. Or shut up and get your rest.

  21. As soon as the 2016 data is available, a revisit of the trends is in order.

    But it appears as if all the data sets still indicate a flat trend from 2001 through 2013. But that flat trend is still consistent with the longer term trends and ended with 2014.

    So, Karl didn’t bust the pause, the recent El Nino did.

    What lies on the other side? Probably something consistent with the trend:

    • You know the “pause” was never statistically significant nor adequately defined by any so called skeptic.

      • Danny Thomas

        “You know the “pause” was never statistically significant nor adequately defined by any so called skeptic.”

        That’s a bit of a framing concern.

        What definition of ‘statistically significant’ do you prefer? Prior to or after ‘normalized’ temperatures?

        Pause/hiatus is a plateau running over an ‘uncertain’ time frame. Seems to be the point of contention.

        So was it ‘statistically significant’ and defined then the statistics modified and definition too?

        The definition of skeptic seems a bit unclear.

      • Danny Thomas

        Just happened by this:
        “But Gerald Meehl, a climate scientist at the National Institute for Atmospheric Research who has published on the “hiatus” and attributed it to changes in the Pacific Ocean, said that notwithstanding this work, something really did happen, temporarily, to the planet’s climate in the 2000s that could be called “hiatus” and is worth understanding.

        “The early-21st century ‘slowdown’ (or ‘hiatus’) was mostly a product of internally-generated naturally occurring climate variability,” said Meehl by email. “It was notable compared to the previous 20 years when there was accelerated warming, also with a big contribution from internally generated variability.”


      • Danny –

        What is the difference between a “pause in global warming” and a “relatively short-term slowdown in a longer-term rise in surface temperatures (only)?”

      • Josh,

        Definition of terms and points-of-view of defenders of a particular position.

        It’s like asking the difference between Global warming and climate change. Changing participants and moving goalposts.

      • “The early-21st century ‘slowdown’ (or ‘hiatus’) was mostly a product of internally-generated naturally occurring climate variability,”

        naturally occurring climate variability is something that all the alarmists and many of the skeptics do not understand. Most do not even have a clue.

      • You know the “pause” was never statistically significant nor adequately defined by any so called skeptic.

        That’s probably true. There’s a flat trend between 2001 and 2013, but not particularly meaningful given natural variability.

        Are the trends above significant? It doesn’t appear so – natural variability is still greater than the trends.

      • My training in statistics is limited to linear regression, standard deviations and the like.

        For a linear regression of a temperature trend, I would define statistically significant as where the measurement of the trend exceeds the measurement of the uncertainty of the trend by a factor of 3 for a positive or negative trend. And for a pause, an uncertainty measurement of less than 0.05 C/decade.

        And of course I would expect the raw data to be run through some sort of approved quality assurance inspection and adjustment to minimize errors.

        I like Karl’s work because over the life of the metric, his work matches UAH the best over the period 1979 to 2017.

        A skeptic is one who comes to conclusions after examining the evidence.

      • Danny Thomas

        “A skeptic is one who comes to conclusions after examining the evidence.” That’s an interesting definition. Some might disagree and use the term in derogatory fashion. By appearance that’s not your choice. It’s not mine either.

    • Curious George

      The X axis is “NOAA Estimated ..” So what?

    • …and, similarly, 1998 created the pause. El Nino giveth and taketh away.

      • Danny Thomas

        That’s a pretty clear statement that there indeed is/was a pause (plateau) and runs counter to the Karl work and subsequent purported confirmation.

      • If you want to start your trend at a super El Nino, yes, but some might call that cherry-picking your end point, which is not a good statistical practice (not that that matters to them, of course). Maybe it makes some more sense to start and end at large El Ninos like 1998 and 2016, but there are better ways to get robust trends, like the trend in the thirty-year mean temperature which really is solidly 0.17 K/decade, and is more meaningful for climate change.

      • Karl and the scientists who have written papers on the causes of the pause have no real disagreement. They use his adjustments.

      • “If you want to start your trend at a super El Nino”

        Of course, that also means you can’t use 2015 as an ending year.

      • I wouldn’t. Like I say, I would at least average out the ENSO cycle before getting any trends, but probably also solar cycles, which means you need at least decadal averages to compare with each other.

      • Yeah, if your results change drastically based on the inclusion/exclusion of one or two data points, then it’s not very robust.

        The “pause” was so much hullabaloo for that reason; cherry-picked starting points. It doesn’t work if you start from 1996 or 2000. It’s not statistically robust, nor scientifically rigorous.

        In short, the “pause” was a con that deniers pulled on people without much statistics training.

      • I really don’t like removing ENSO, volcanoes, etc. They are part of the climate system. Just rinsing them out and saying the background warming is there is an unsatisfying argument. Sorry to Jim D, but saying 1998 created the pause is not good.

        The pause was caused by physics. Something happened. It was a real thing.

        The one thing the “pause” has done is spur a lot of science, and some of it looks to be really solid. What it is hinting at, rather than models being too hot, is that natural variability has been suppressing ACO2 warming since around 1985 to 1995.

        Before the Karl adjustments I believe, from memory, there were 30-year trends in the .15 ℃ per decade range. Now ~.175 ℃ is solid going back far past 30 years, but people can still argue models are running hot. So that is the next big lie that is going to go down the hole. Natural variability has been running cold, and it is looking like it has shifted to a hot regime, which means observations will soon, if anything, indicate models are running cold.

      • JCH, the definition of climate is a 30-year average. On those time scales solar cycles, ENSO and individual volcanic effects cancel out, but there may still be solar and volcanic trends on 30-year time scales. The 30-year temperature has a rather clean tendency, no ENSO, no volcanoes, no 11-year cycle, to distract from climate change itself.

      • They didn’t cancel out. Over the 30 years, 1984 through 2013, the PDO index was net negative and ONI was net negative. A vast area of the earth’s surface was persistently cool. So models ran hot, and people have fooled themselves into being lukewarmers and into thinking climate sensitivity is low and CO2 is just plain old plant food.

        Two years of ENSO and PDO working together:


      • Much variability occurs on 15-year time scales. When comparing models to observations, it depends which 15 years you choose, so I agree with what you are saying. Longer period trends have the models doing better, and should not be ignored.

      • The four humps of the negative slide of the PDO index

        The four negative phase ramp ups resulted in:

        1997 – warmest year
        1998 – warmest year
        2005 – warmest year
        2010 – warmest year
        2014 – warmest year
        2015 – warmest year
        2016 – warmest year

        and still managed to cause an 19-year pause in warming.

      • But in applications of the method to times between 100 million and 400 million years ago, Franks finds hints of a foreboding message. During documented episodes of global warmth, he says, the method reveals relatively low CO2 values, nothing like the levels of 2000 ppm or more suggested by other proxies. If these downward revisions hold, Earth may be even more sensitive to injections of CO2 than current models predict. “Temperatures are going to climb further for less carbon and we better be mindful of that,” Franks says.

        Climate sensitivity is not low.

      • Climate sensitivity is not low.

        Has someone falsified Steffan-Boltzman?
        Otherwise, can’t get too far from SB.

        AR4 sez low scenario is 1.8C per century.
        Actual rates are 1.6C per century, why?
        OHC can explain this, but OHC is part of the system.

      • jch

        your link regarding sensitivity raises several conundrum.

        During the Pliocene which concluded some 2 million years ago, temperatures and sea levels were higher than today at roughly todays co2 levels.

        Similarly temperatures were warmer than today during the early part of the Holocene although at a much lower co2 concentration and in amongst the bursts of warmth were bursts of extreme cold-Dark ages, LIA etc-at levels lower than today


        So temperatures can be higher than today or lower than today at co2 levels higher than or lower than todays.

        (PS Has Mosh found out who exactly called Karl a f5aud yet? I don’t recall the majority of sceptics here claiming that)


      • These records suggests that intermediate waters were 1.5–2 °C warmer during the Holocene Thermal Maximum than in the last century.

        Intermediate water masses cooled by 0.9 °C from the Medieval Climate Anomaly to the Little Ice Age.

        The records suggest that dynamic processes provide an efficient mechanism to amplify small changes in insolation into relatively large changes in OHC.

        The kimikamikaze… it’s all you got.

      • The records suggest that dynamic processes provide an efficient mechanism to amplify small changes in insolation into relatively large changes in OHC.

        Probably an error to conceive of the changes as small.
        The global annual mean changes may have been small, but the regional and seasonal changes were quite large:

        The Northern tropics experienced a much greater insolation energy during the Northern summer months – much greater than a 2xCO2. During winter, of course, the HCO exhibited a roughly commensurate decrease in insolation for most Northern latitudes.

        So, were the proxies inordinately influenced not by the annual average, but by the maximum insolation? It sure seems possible.

        “We suggest that even very small radiative perturbations can change the latitudinal temperature gradient and strongly affect prevailing atmospheric wind systems and hence air-sea heat exchange. “
        There’s no need to imagine some indirect effect on gradients during the HCO, because the gradient of insolation alone during the HCO was quite large and imposed a substantial weakening of the Northern gradient from November through June ( black: 2010, orange: Holocene Climatic Optimum )

  22. Hummm, so what if the SST’s go up and raise global Temps a few hundredths of a degree?

    What is the down side?

    We have more of a growing season for more crops and more areas to place and feed an ever growing population globally?

    Really, aside from modeled catastrophe, where is the threat?

    Droughts, tornadoes, hurricanes, flooding? Not from the data…..

    Perhaps with SLR we can get some of the people in Miami and New Jersey to move out of the landfills they live in. Is that a bad thing?

    • Rather than the postulated catastrophe being a product of mathematical modeling I’d liken it more to writing a script, which makes it good theater but bad science.

      • “Rather than the postulated catastrophe being a product of mathematical modeling I’d liken it more to writing a script, which makes it good theater but bad science.”

        Even with plays and screenplays with their inherent logic, the story must remain consistent with its inherent logic. Failure to do attracts a tepid audience reaction at best and at worst it drives the audience away. A good writer can make the implausible seem plausible but only as long as the inherent logic remains consistent.

  23. DLyons123@aol.com

    Julia, As a metrologist I have been most concerned about the accuracy of the instrumentation used to take these temperature readings. Are they calibrated? Is there traceability to the NIST standards? Some of the measurements referenced for example here in the US were using instruments with accuracies of only +5%. That leaves for a lot of uncertainty. Doug Lyons

  24. I downloaded the 2005-2015 Argo SST data and compared against ERSST V4. Argo’s trend was about 0.007C/dec warmer than ERSST over that period. (

  25. Earthly sea levels have been relatively stable for centuries. We have learned from verifiable satellite information that we do not face an imminent crisis of disastrous sea level rise caused by humans going about the business of living and that such fears are superstitious ignorance in general and worse, are nothing more than ideologically-based fearmongering for political purposes…

  26. David Wojick

    Death Of Global Temperature ‘Pause’ Greatly Exaggerated

    • The North Atlantic Ocean Heat Content dropping below 1950s level is one of the more astonishing headlines I’ve read in quite sometime, especially given much of the discussion on this post. Duchez st al 2016.

  27. Three cheers for Zeke and his colleagues for clearing up part of this controversy – the relationship between the buoy trend and composite trends.

    Because of changing technology, the SST record has always been a mess. In theory, all technologies should be reporting the same trend with different y-intercepts. (If the trends are different, one technology or the other needs to be discarded.) When records are combined into a composite with the wrong intercept correct, one obtains a composite record with a biased trend. Unfortunately, differing methodologies are reporting on different fractions of the ocean at different times. The only way to find the correct intercepts is to compare the records from a common area over as long a period as possible. Thank you, Zeke et al.

    My only question are: 1) Why isn’t this done clearly and transparently every time every a new composite record is released? Didn’t peer reviews demand to see this evidence? 2) Why was Karl’s work published in Science, a journal which doesn’t provide enough space to adequately and transparently document the creation of such composite records?

  28. Figure 1 of Hausfather is a bit of a puzzle. According to this, ship measurements were fine until around 2009, when some gremlin started pulling them lower. This at a time when the proportion of buoys was increasing, so one might expect increasing convergence over regions of common coverage, not divergence?

  29. …. if however, during the time global temperature rise was claimed to be in a pause, this would had been proven the case if the temperature over the next 600 years remained the same. Also, since it currently seems to “still” be rising, that could change in 50 years, and really pause. So at this point it is still a prediction not a known scientifically determined fact.

    The social debate however, is not about warming or cooling or status quo, it is about what is the cause and is the result (more CO2 in the atmosphere) good or bad.

    Since temperatures have been rising, say since 1970 and the human population had doubled, it would seem more CO2 is pretty good for humans and all we really need to do is better position cities like New York rather than in on a island out in the Atlantic Ocean.

  30. Danny Thomas


    Link to Hostetler uncomplimentary comments about this blog, please.

    • Danny Via Facebook, I’ve asked Zeke if he would comment on things said here at CE numerous times in the past year. His response has been that he tends to avoid reading stuff on Curry’s Blog. This is our loss at CE. Dr. Curry should be encouraging a healthy environment (not of bullies and personal attacks) for people like Zeke to give their views.

    • JPZ,

      Speaking freely is one thing, but may not make for the most open environment. The impression is Stephen is lamenting a more welcoming environment. I don’t think we should expand that to a great fearmongering.

      • Danny,

        First, happy new year to you brother.

        Secondly, some people will find offense even with the most innocuous of welcome mats. You can only welcome those who want to be welcomed. If Hostetler does not want to be welcomed here in CE, there is no such welcome mat that will change his mind.

        My fear is that what will change his mind is if certain people were made to feel less welcomed in this site so that he may feel more welcomed. I don’t think I am fear mongering by expressing that concern, and I also think there is a Grand Canyon of a difference between speaking freely and freedom of expression.

        Our freedom of expression is better preserved when we the ones expressing ourselves endeavor to reign in how we express ourselves. Still, ain’t no reigning in on my own language that will satisfy some and sometimes my reigning in of language frustrates others who seem to expect a more full throated response.

      • Danny Thomas


        Happy New Year to you also. Highest regards.

        If you followed the link to the 2014 post, Zeke was raked over the coals a bit unnecessarily IMO, but it’s my belief that the welcome mat is ever present from Dr. Curry where the importance lies. Being a blog, and one loosely moderated, it’s my hope that Zeke would ‘turn the other cheek’ and return when appropriate.

        Frankly Stephen surprised me just a bit when he stated he’d requested Zeke respond to some sort of commentary here. When we feel a need for a response we grow by researching and responding for ourselves. But to each their own.

        No one should be made to feel unwelcome be it Zeke or other commenters, but civility is a hoped for goal. Stephen, I perceive, is frustrated that we don’t police our ourselves. Maybe in that he’s correct. There are unfortunately ‘sides’. When seemingly there should be a common side of seeking accuracy and ‘the truth’ (whatever that may be).

        2017 being a new year brings new opportunity.

    • JPZ For at least 2 years (at least) my main observation here at CE is that the coupling of two Dr. Curry points, presents a very positive message: (1) Lewis/Curry view on TCR; PLUS (2) Dr. Curry’s favorable view of fast mitigation of SLCPs of smog, methane, black carbon, HFCs.

      • Stephen,

        Thank you for your reply. Dr. Curry’s endeavors to create a place where positive messages can be made is more than a worthy goal. I would respectfully argue that both Ristvan’s arguments of Karl’s adjustments and your arguments of Rud’s stance, i.e., your post below, are aspects of that positive atmosphere. Sure, some can find the underlying negativity in the arguments, but that’s the deal. Positive has a flip side.

        In my estimation, more speech not less is an attribute of a positive atmosphere.

  31. Three problems. First, you spelled my name wrong. Second, you missed my previous posts and comments here and at WUWT on Karlization around the 2015 event just before he retired. Third, only warmunists would assert I have alleged fraud. I have explained Karl’s bad science, yes. Several different ways. But never alleged fraud in either the legal or colloquial sense. Never impute to malice that which is adequately explained by mere incompetence. Nor did Rep. Lamar Smith in his Karl subpoena. Because you and yours have on this thread, you are the self-designated frauds.
    Have a not so nice day.

    • Stay tuned for a post on this topic next week

      • Just watched you on Tucker Carlson’s show on Fox. You presented yourself well and your answers were persuasive. Your answer to the “98% consensus” claim was just right.

        Well done. Hope to see much more of you on the MSM.

    • Did this new paper just introduce two brand new reconstructions, buoy and satellite?

    • If I understand what you are asking, two long, buoys and satellite, and three short… all ARGO.

    • Steven Mosher

      ristvan | June 4, 2015 at 5:40 pm |
      You need to learn the difference between improvements and plain old fudging. I refer to my physics professor, Hamorski, and his Hamorski fudge factor: a multiplicative value derived by dividing the ‘right’ value by the experimental value, applied to the observed value, before submitting the experiment.
      Essay When Data Isn’t in ebook Blowing Smoke. Please return with rational explanations for what is indelibly shown there. Rutherglen and Reykjavik, for example. Or DRX 2013 to NClimDiv 2014 for Maine.

      Bad science is one thing
      Claiming they Fudged?
      Why that sounds actionable

    • Steven Mosher

      You want the links where you call it misconduct?

    • Steven Mosher

      ristvan | June 4, 2015 at 11:57 pm |
      Have just done the same. Concur with you.
      Looks like this was a planned multipaper setup. The fingerprint ‘evidence trail’ is clear. Legal types should be shouting “willful misconduct”

  32. What’s all the fuss about sea surface temperatures anyway? Seems to this non-expert that the steric component of sea level alterations is both more easily measured and a more sensitive measure of total ocean heat content. Arguing about hundredths of a degree changes that are beyond the precision of the instruments used with data that are very poorly sampled spatially and temporally seems a waste of time, energy and resources.

  33. Animations of SST show significant apparent randomness in anomalies:
    (long term) http://www.youtube.com/watch?v=1ir1w3OrR4U

    (4th qtr 2016) http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/gsstanim.shtml

    • Curious George

      Are these anomalies all taken to the same base? I believe NASA uses a 5-year-average sliding base. With a sliding base, a uniform cooling – or warming – would produce always the same anomaly.

  34. I find the intolerance of any view but the .”Humans are the cause of all climate change .The lack of open debate ,with the fear of reprisal for saying the earth will always change over 10,000 years periods or much less time .

  35. Geoff Sherrington

    Here are some truisms and comments about surface sea temperatures.
    Unlike some basic properties like mass or length, temperature is rather variable, so its accurate description will often require a specification of the time duration of its measurement.
    At any point in the seas, there will be a ‘true’ temperature for a fleeting time. However, it is practically impossible to measure that temperature, and to reproduce it.
    Therefore, all reports of sea temperatures are estimates that are known to be wrong unless, by unknown passing chance, one is right now and then.
    Apart from temperature changes with time, there are sea temperature changes with depth. e.g. https://books.google.com.au/books?id=tZary8a4HMwC&pg=PA71&lpg=PA71&dq=surface+temperature+of+ocean+in+micro+detail+of+layers&source=bl&ots=XjxoBPFHz6&sig=TMwCeOnS1ICZckt6eYVtE0SCa7A&hl=en&sa=X&ved=0ahUKEwiryeWJq6_RAhWIFZQKHXPfCccQ6AEIODAF#v=onepage&q=surface%20temperature%20of%20ocean%20in%20micro%20detail%20of%20layers&f=false
    Of course, the SST, at the very surface, is rather different over the brief time before and after the start of rain. The duration of differences depends on mixing, which is not able to be measured with present state of the art as, for example, an addendum to ARGO type instruments.
    What is loosely named a sea surface temperature is thus a rather variable entity, with its variability being significantly large from time to time and depth to depth at a point compared with the uses to which the measurement is being put.
    Any definition of SST will be susceptible to these time and place errors; in hard science, all errors demand explanation and estimation, especially errors of bias. Such error descriptions are rare in SST climate papers. Further, there is no way to trace measurements back to a primary standard, as is often demanded in the best of science.
    The ability of a set of sea surface temperature probes to agree with each other is not a valid measure of accuracy. It is necessary, but not sufficient.
    As with many matters, extraordinary claims demand extraordinary proof. The large socio-economic consequences flowing from errors in SSTs are extraordinary, so extraordinary accuracy has to be demanded. Accuracy is not presently within a couple of orders of magnitude of what might subjectively be called acceptable.
    It is wrong that a crazy optimism about the accuracy of SST is influencing global climate policies.

    Those who will respond that the errors are by now under control and manageable have missed the points entirely. (One being, accuracy is much more complex than statistical precision).

  36. Here is the BBC coverage of the Hausfather et al paper:

    The headline being “Fresh doubt over global warming ‘pause'”.

    But doesn’t this miss the big picture? If the heat is going into the ocean that may help to explain the pause but does not explain it away. There has still been a hiatus in land surface temperature rise (likely ongoing when the current el nino fades), and the models still failed to predict this. And it is LST that is relevant to human welfare and most of the ecosystems we care about.

    Because the heat capacity and density of water are much higher than of air, then (if these papers are correct) the predicted tenths of a degree of warming over land became a few hundredths of a degree of warming of the sea surface. And if the heat finds its way into the deep ocean (as has been suggested elsewhere) then it becomes a few thousandths of a degree and of entirely negligible impact.

    If the heat is finding its way into the oceans faster than the models predicted, then AGW is much less of a problem than we thought. But the story is that there was no ‘pause’, so by implication it is more of a problem than we thought. As always, whatever the science says the spin is alarmist.

  37. I have had issues with the Karl paper that I have discussed with the authors of that paper and including Karl just before Karl retired. I have discussed these issues at other blogs and this one. Tom Karl did not make the changes from ERSST v3b to v4 as that was the work of Huang. Karl took data from the newer version in an attempt to show that the recent warming slow down was not significant (and that it was with v 3b). I doubt that Karl even understood all that went into the changes, but he apparently got the blame from his critics for making them.

  38. Pingback: Weekly Climate and Energy News Roundup #254 | Watts Up With That?