Thursday, December 11, 2014

RealClimate's opinion on the WUWT widget

Two months ago, it has been ten years since this weblog was founded. Two months later, a group of fraudulent proponents of the climate hysteria founded RealClimate.ORG, a domain designed to spread misinformation about the climate issue.

I never planned to celebrate the 10th birthday because I find such celebrations stupid and I am shy – but if you want to drink some whiskey at home, be my guest! After these ten years, this blog run by one person (but made so inspiring and kind by many of you, thank you!) has welcomed the same number of visitors as RealClimate.ORG which is run by a dozen of folks, about 50% of the "global community" that wants to force the mankind to pay trillions of dollars. Not bad.

Congratulations to the 10th birthday of RealClimate.ORG.

Lacking the Lumoesque shyness, modesty, and focus on the beef, the RealClimate.ORG website has published not just one but three Happy 10th Birthday blog posts.

Meanwhile, their friends celebrated, too. All of them flew to Peru, Lima and set a new world record in the money wasted for a hysterical climatic conference. Their Greenpeace comrades, also in Peru, damaged and desecrated the Nazca lines by an ancient civilization that have been carefully protected for 1,500 years. (Ironically enough, the most irreversible damage has been done by the Greenpeace officials' footprints, too. The Peruvian government normally demands special shoes etc. for all the visitors.) The similarity of this vandalism to the liquidation of heritage by the Islamic State is way too obvious.

The Peruvian government is suing Greenpeace and because of the pricelessness of these geoglyphs, the Latin American nation undoubtedly has the moral right to liquidate the disrespectful terrorist organization. But my realism prevents me from believing that this outcome will actually materialize. But I want to discuss the previous RealClimate.ORG blog post – one about a... widget.

Stefan Rahmstorf wrote a rant for RealClimate.ORG whose title was
The most popular deceptive climate graph
It's pretty comical that a group of people who demand trillions of dollars to be wasted for the pseudoscientific fantasies and who claim to define the mainstream pick the 137th most important piece of software on a blog of a climate skeptic as their target.

But it's simply the case that Anthony Watts' climate blog has traffic that exceeds that of RealClimate.ORG – or my blog – by a factor of fifteen. Among many other things, Anthony Watts also maintains a "climate widget"

that you may find in the right sidebar of The Reference Frame, too. The RealClimate.ORG men who are getting tons of money are dreaming that they will earn a footnote in Anthony Watts' memoirs, or something like that, so they decided to analyze the widget. Here is the summary of Rahmstorf's criticism:

Well, I happen to strongly disagree with all these criticisms. In other words, if someone created a widget that would respect these criticisms, it would be much more misleading than Anthony Watts' widget. Let me say a few details about the individual points. WUWT published its own reply to the RealClimate.ORG criticism of the widget.

Ploy #1: monthly data

Watts is criticized for using the monthly data. Why doesn't Rahmstorf like the monthly data? Well, the answer is obvious and he openly admits it. He doesn't like it because the monthly temperature anomalies oscillate and the "global warming trend" that he would like to amplify and emphasize largely gets lost in the noise. So it's bad, isn't it?

The problem with this criticism is that the monthly oscillations are completely real and different temperature datasets agree about them more precisely than they agree about the trend. In this sense, these wiggles are more real than the "global warming" itself! The monthly noise is completely verifiable. Indeed, in combination, the data imply that the trend visible in recent decades doesn't really exceed the normal noise. And you know what? It just doesn't.

Any averaging of the data designed to show a simpler, smoother, perhaps nearly linear curve would be fraudulent. Such a manipulation with the data would convey the message that there is a clear underlying trend in the temperature data. But there is no clear underlying trend in the temperature data. The previous sentence has a very clear meaning: the amount of trend-like warming e.g. in 35 years is comparable to the noise – or at least, it isn't substantially greater than that.

So the temperature data are pretty nicely compatible with their interpretation as "noise", especially if you choose some pink noise (noise of an appropriate "color"). This correct conclusion may only be obtained if we carefully look at the relative magnitude of the noise and the hypothetical underlying trend. It is very clear that any method to artificially suppress the noise is fraudulent because it makes the signal-to-noise ratio look greater than it actually is.

Ploy #2: surface data preferred

Rahmstorf says that the surface data, and not the lower troposphere temperature data, should be plotted because "the surface is where we live". The difference between the surface-based and troposphere-based datasets isn't too large but let's assume that we care about the difference.

Why does Rahmstorf prefer the surface data? Because they show a slightly higher underlying trend – still negligible for all practical purposes.

Why is it more relevant to draw the troposphere's temperatures? There are at least two reasons:
  1. the satellites are more reliable than the weather stations, especially because they don't depend on complicated and subtle procedures of homogenization (and things like the urban heat islands)
  2. the troposphere, and not the surface, is where the CO2 greenhouse global warming theory is actually predicting the most visible signal, so it's the place where we should look if we want to have the best chance of validating the hypothesis!
When we look in the troposphere, we see a trend that is compatible with the interpretation of noise and that is 2-4 times smaller than what is routinely predicted by the global warming hysterics. So the predictions are falsified, the would-be science behind the hysteria has been discredited, and it's arguably inconvenient for those who have been spreading this hysteria for years.

But this disagreement is not the fault of Anthony Watts' widget.

I must make one more comment about Rahmstorf's assertion that "no one lives in the troposphere". His alternative is to plot the global mean temperature near the surface but no one lives at the "average of the whole globe", either! Everyone lives at some particular place of the surface at every moment and if we plot the local weather, we will get much higher variations – even relatively to the monthly oscillations that Rahmstorf disliked – and indeed, the "global warming" would become almost entirely invisible. After all, 30% of weather stations saw a cooling in the last 100 years or so.

Ploy #3: the Sun should be overlooked

You know, I don't consider the hypotheses that the fluctuations in the solar activity significantly influence the climate on Earth to be settled. Clear evidence doesn't exist in one way or another. The Sun surely has some impact – but whether it was producing a temperature change in the last 40 years that is comparable to the apparent "trend-like" temperature change during that period is as open a question as a similar question for CO2.

But the point is that many users of the widget are interested in the Sun and its possible role in climate change. And the widget correctly conveys the idea that both CO2 and the solar activity may influence the climate and we may want to study both of them. None of the two graphs is accurately (and not even approximately) mimicking the shape of the global mean temperatures.

At the end, it is rather likely that neither CO2 nor the solar activity are extremely important. That's surely what e.g. Richard Lindzen would tell you. It's a fallacy to try to attribute the climate change and look for an "external" and deterministic explanation. The wiggles in the temperature graph are likely to be random numbers produced by the internal complex processes in the Earth's atmosphere (and oceans) that are largely unpredictable.

But given the apparent correlation of the solar activity with some cold periods in the history and because of the known candidate theories that allow the solar activity to modulate the climate, it is surely fair for the widget to show some of the "solar weather" – after all, this phrase contains the word "weather", too.

Ploy #4: the CO2 graph looks too steep

The graph on the widget shows both the global mean temperature (in the troposphere) and the (nearly linearly increasing) CO2 concentration. They have different units so of course the relative scale on the y-axis is a matter of conventions.

However, the widget uses a simple convention that maximally uses the area in the whole rectangle reserved for the graphs.

And there's one more reason why the relative scale in Watts' widget is nearly perfect: the line representing the CO2 concentration may actually be read as the changing expected temperature predicted to be a function of the CO2 concentration by IPCC-like models. And the CO2-induced temperature may be compared with the actual temperature. They may be placed in the same graph, with the same scale on the y-axis, and when we do so with the IPCC-like values of the sensitivity, we indeed get what Watts' graph shows!

Obviously, the violet CO2 line is much steeper but that's because the climate sensitivity preferred by the climate hysterics indeed overestimates the trends actually observed in the real world by the factor of 2-4!

So Watts' widget does seem to convey the message that the CO2-based predictions for the warming are falsified by the observed data – and this message is entirely correct. Mr Rahmstorf may find this truth inconvenient but this discomfort doesn't make the truth any less true.

For these reasons and hundreds of others, I consider Herr Rahmstorf to be an immoral lying crook. He is a splodge on the European science.

And that's the memo.


  1. Just looking at solar influences on 14C content, as variations in solar wind modify this quantity on top of geological influences (old carbon introduced into atmosphere through dissolution of marine carbonates, e.g. through the thermo-haiine circulation in North Atlantic, I would say that there is very good evidence for the fluxes in past solar activity/ Czech physicists Vaclav Buch and son also look at this in terms of geomagnteic influences in more recent past, at decade timescales. In as much as these fluxes align in time sequence with past secular or century-timescale climate changes, as well as the decadal oscillations in teh ENSO and NAO or synoptic systems assessed more recently, the solar factor really should be considered as significant.

  2. The key question in climate forecasting – and the one to which the establishment climate scientists should really turn their attention i e Where is the earth with regard to the natural solar millennial cycle ? Several posts at my site deal with this question. In summary I would make a few observations.
    1 The period fluctuates generally between 960 and 1020 years –
    2.Look at the ties between solar “activity” and temperature in figures 4,5,6.7,8,9,10,11,12, in e.g. post

    3 . This post also says
    “NOTE!! The connection between solar “activity” and climate is poorly understood and highly controversial. Solar “activity” encompasses changes in solar magnetic field strength, IMF, CRF, TSI, EUV, solar wind density and velocity, CMEs, proton events etc. The idea of using the neutron count and the 10Be record as the most useful proxy for changing solar activity and temperature forecasting is agnostic as to the physical mechanisms involved.
    Having said that, however, it is reasonable to suggest that the three main solar activity related climate drivers are:
    a) the changing GCR flux – via the changes in cloud cover and natural aerosols (optical depth)
    b) the changing EUV radiation – top down effects via the Ozone layer
    c) the changing TSI – especially on millennial and centennial scales.
    The effect on climate of the combination of these solar drivers will vary non-linearly depending on the particular phases of the eccentricity, obliquity and precession orbital cycles at any particular time.”

    4.It is obvious by inspection of Figs 5 and 9 that we are just approaching, right at or just past the millennial temperature peak.
    5.We must distinguish between the peak in the solar activity driver which is behind us see Figs 13 and 14 and the temperature peaks which lag the driver peak by variable amounts (maybe 20 +/- 8 years) according to the metric used – i.e..the various measures of land and sea temperatures , OHC and the area measured e.g. NH SH Global or on the various climate plates.
    6.My cooling forecasts are included in the posts.

  3. I was watching the BBC news last night, and it was announced that Greenpeace apologized for desecrating the NAZCA lines.
    I guess that's OK, then.....
    I support some of their aims, but they do so many "Bull-in-a-Chinashop" pig- ignorant things that make it impossible to physically or fiscally support them. They also tend to have messianic leadership at times.
    RealClimate, however has few redeeming features---it is an exclusive club of sock puppets who tolerate only one viewpoint.
    The redneck version of RealClimate is DeSmogBlog, which, however, has NO redeeming features---it is a smug,gloating, self-righteous scurrilous,
    libelous blog that raises tabloid style smears to a new level.

  4. "The redneck version of RealClimate is DeSmogBlog" LOL, excelent metaphore :-)

  5. LOL, is it so pretty? I always imagine the top of the Pilsner Tower - the St Bartholomew Cathedral - with its green color.

    If you replace the metal by the same amount gold, we will probably tolerate that we will have sacrificed the priceless green color. ;-)

  6. Meanwhile the bladeless input data Marcott 2013 hockey stick remains unretracted after bizarrely issued a FAQ instead, verbally welding high time resolution thermometer data into low resolution proxy data.

    It's amusing how they give their own game away by revealing exactly how they would introduce bias into their version of such a widget! My favorite graphology ploy is the way warming is made to falsely look to the public and policymakers to have recently surged using mere visual trickery:

    My post-Climategate posting this type of infographic thousands of times a year to news and science sites and conservative blogs helped turn half of the West into skeptics. is scandalously still site registered to the notorious junk science law suit partner PR firm Fenton Communications via their Environmenta Medial Services (EMS) division.

  7. Dear Nik, I agree with the broader points but I don't really see how the simple barcharts or the incorporation of a horizontal axis into a graph "deceives" the viewers.

    Those things are among the most normal ways to visualize a function, aren't they? It's normal to include a horizontal axis in a graph, even if its height is arbitrary. And it's normal that barcharts show red color above zero and blue color below zero.

  8. You are a seasoned Harvard-level string theorist! My audience for that chart was everyday worldwide news site readers with zero background in mathematics or science who, after the Climategate event sent millions of conservatives searching for facts, were going to screw it all up by citing crappy arguments instead of understanding the basic historical facts of climate. Stuff like "volcanoes put out more CO2 than Mankind" sort of crap. Neither the public nor policymakers are scientists, as a rule, but after Climategate, conservatives who are naturally averse to scientific claims, were going to screw it up. With nearly *every* official scientific organization very loudly equating climate alarm skeptics to Moon landing deniers, it was crucial in my mind to present the very essence of basic climate data. Few skeptics even, when I started doing this is 2008/2009 were aware that both nearly all of the oldest tide gauge records and the oldest single site thermometer records showed strict linearity right up to the present day. So I made graphs of those too, and made them into single point infographics, and presented them far and wide, saturation style, for many years.

    I note that Rahmstorf is in essence also sociopathically aware that the real audience for output is common folk, via their activist base. He is giving the few remaining bitter ender activists marching orders to attack Tony Watts and yourself at the exact same level my own infographic addressed. So thank me, dear skeptic, for being one of the few to have actively addressed normal voters, already.

    I live on the Upper West Side of NYC, where I was very familiar with the entry level nature of activist acceptance of climate alarm. And *indeed* a simple graphing style was enough to convince them. I also needed content that the online climate activists couldn't mock via some link, some arguments not in common use on skeptical blogs. The very basic climate data of temperature and tides was what I found.

    As an intuitive empiricist I strongly desire a correct chart. The convention of using bar graphs hides completely the nature of the data since that horizontal line hides the middle era as if it was just a random lull. It's just stupid to plot things like that. But it's very smart for scammers.

    -=NikFromNYC=-, Ph.D. in chemistry (Columbia/Harvard)

  9. Dear Nik, I am grateful to your services etc.

    It's still not clear to me what's wrong with barcharts and horizontal lines.

    If the proximity of the graph to the horizontal line suggests that for some periods, the function may be interpreted as noise, it's probably because it's true, right?

    The function is more or less noise in the whole 35 years, and in almost every 20-year period, this becomes really obvious. There's not much non-noisy information in the graph. Interesting, your comments sound like alarmists' ones to me. ;-)

    The blue/red barcharts "highlight" the sign of the function relatively to a base, which is largely arbitrary, but I don't see anything wrong about it, either. We sometimes may want to quickly see which years were warmer or cooler than the average of the interval, for example.

    Interesting you never objected against barcharts. I would almost always draw the monthly or annual temperature as barcharts, by the BarChart command in Mathematica, see e.g. this blog post a week ago:

  10. well, big trend or small trend, it's been a trend in temperatures to justify some alarmism... but it is of interest how important this becomes to planned politics thereafter.
    btw when we talk about solar weather we mainly imply solar wind fluctuation? ;-)