Bangs and Smoke

… Harry had long since learned that bangs and smoke were more often the marks of ineptitude than expertise.

— J.K. Rowling, “Harry Potter and the Half-Blood Prince”

In my opinion, that quote describes Christopher Monckton to a “T”.

But dwelling on his latest in a long string of unworthy essays would be a waste of time. Instead, I’ll address some aspects of the attempt to account for exogenous factors in global temperature data by “Steve F” which spawned it. It’s not just all bangs and smoke, but there are some unnecessary and un-called-for jibes, and some odd statements about Foster & Rahmstorf (2011) such as “(strangely) limited their analysis to post 1979” when those who read the paper would know the limit was imposed by the data — it includes satellite temperature time series which are only available since that time. But those little bits of sound and fury don’t signify very much.

Let’s reproduce the same procedure, but somewhat differently. For volcanic forcing, I’ll do the same: take AOD (aerosol optical depth) from NASA, multiply by 23, then exponentially smooth it with the same time scale. For solar forcing, I’ll use the same procedure: approximate the forcing as a linear function of sunspot number, then exponentially smooth that too.

Exponentially smoothing the forcing data represents the discrete version of solving the “1-box energy balance” model. If you’re going to do that (and in fact I think it’s a good idea), I think you should use a 2-box model. Using just one box underestimates the rapidity of the response of the atmosphere and overestimates the rapidity of response of the oceans. But that won’t matter for what follows.

For el Nino I’ll use the MEI (multivariate el Nino index) instead of his choice, but I’ll exponentially smooth it on the same time scale he chose. I think el Nino should be lagged rather than exponentially smoothed — but that too won’t matter for what follows.

Let’s regress temperature against those three variables as he did, but instead of the HadCRUT4 time series, I’ll use an artificial temperature signal that looks like this:


It’s a “schematic” representation of temperature, one which clearly isn’t influenced by volcanos or el Nino or the solar cycle. Then I’ll linearly de-trend the temperature data. After all, as “Steve F” says,

Using the original temperature data (not detrended) distorts the regression fit by essentially forcing the regression to explain all the temperature change, including any slow secular trend, using the three short-influence variables, and so yields very poor (even physically nonsensical) results.

It’s a pity he didn’t pay more attention to his own statement. He de-trends the data linearly, which, when applied to our artificial “schematic” temperature data, leaves this:


When we regress this on the three variables, we get an interesting fit:


It’s not only quite interesting, it’s also quite statistically significant. But, just as though it were full of sound and fury, it signifies nothing but the fact that the unremoved trend pollutes his model. Rather a lot. The essential result, when applied to actual temperature data, is to overestimate the volcanic influence and underestimate the solar influence (for the schematic data, the solar influence is — nonphysically — negative).

Steve F realized that a trend could pollute the fit to exogenous variables. But he utterly failed to account for the fact that removing a linear trend doesn’t de-trend the data and doesn’t solve the essential problem. Although he seems to know how to run the programs, I have my doubts about his “eptitude.”

I’ll probably have a good bit more to say on this in the near future.

36 responses to “Bangs and Smoke

  1. I see Steve F. as having a bit of a problem. He certainly has no love of F&R2011 with his opening words surely intemperate; describing F&R2011 as “a rehash” and interpreting the concluding remarks of F&R2011 (that say “Its(ie the AGW signal’s) unabated increase is powerful evidence that we can expect further temperature increase in the next few decades, emphasizing the urgency of confronting the human influence on climate.“) as to mean “that it is critical for people stop burning fossil fuels immediately” while adding “though it is not immediately obvious how a multiple regression model on global temperatures leads to that conclusion.” Indeed SteveF.’s interpretation does take some explaining.

    If SteveF. is troubled by examples of papers that are in his view “little better than a mindless curve-fit exercise“, I am surprised he has not set forth on such notable examples as Scarfetta or Akasofu or Tung/Zhou or even SteveF..

  2. Steven Mosher

    The strangeness of limiting your analysis to 1979 and beyond is quite simple to understand

    “F&R considered the influence of the solar cycle, a change of about 0.1% in solar intensity from peak to trough of the cycle, separately from the effects of stratospheric volcanic aerosols, even though both are expected to change the intensity of solar radiation reaching the Earth’s troposphere and surface. (Why should solar intensity change and volcanic aerosol forcing not be fungible?) Like some earlier publications, F&R also (strangely) limited their analysis to post 1979, even though data on volcanoes, solar cycles, and ENSO over longer periods is available. ”

    Recall the post you did on honesty. Now, its clear that one can choose to only look at satellite data, but that really might limit your ability to understand volcanic response and solar variablility. Hey go figure, more data might be a good thing. An honest person would investigate that. So, one wants to ask what does the answer look like if we dont selectively pick the data source. ? So, it is strange to pick 1979 when you have data for all your variables of interest that extend beyond that time period. Hmm note, strange is not dishonest. Its just strange.

    [Response: What a slimy comment.

    You slip in a smear, calling my honesty into question with “Recall the post you did on honesty” but go for “plausible deniability” with “Hmm note, strange is not dishonest.” You, Steven Mosher, are the one who is not being honest.

    You say “it is strange to pick 1979 when you have data for all your variables of interest that extend beyond that time period” — please point us to RSS and UAH satellite data prior to 1979. You add “selectively pick the data source.?” when F&R used all 5 major temperature data sets. Slimier still.

    Steve F seem unaware of, and you seem to want to divert attention from, the salient point, that if you use data since 1979 (or 1975 which was my original intention, but I changed to 1979 specifically to have the same time frame for all 5 temperature data sets) then the long-term trend is at least *approximately* linear. It’s certainly not a perfect approximation, but it’s not completely poisoned by the tremendous nonlinearity in the trend since 1950. Which, by the way, isn’t the only ineptitude in Steve F’s analysis.

    Your sleaziness doesn’t go unnoticed.]

    • Gavin's Pussycat


      the analysis you call for has been done, Figures 1 and 2. The result is the same, considering of course that older data tends to be poorer, especially during World Wars.

      Choosing post-1979 makes sense from a data-quality viewpoint, and does address the bogus claim of global-warming stoppage / slowdown. Perhaps that irks you? Then here’s another one, Figure 2 :-)

    • Climate revisionist Steven ‘Sleazy’ Mosher didn’t go unnoticed.

    • The key to volcanic forcing are aerosol upper trop/lower strat aerosol measurements which were not too lovely before the satellite era either.

  3. steve fitzpatrick

    Humm… That guy Steve F may actually understand quite a lot more than you give him credit for. I am puzzled that you would not comment on the post where it appeared, rather than here. Is there something wrong with commenting at Lucia’s blog? We who have participated there have already discussed in some detail your objections to the linear detrending (before you objected). I invite you participate.

    • Steve F, are you vouching for “that guy” Steve F? Well, I guess he must be alright then.

      BTW, I vouch for me as well. I’m great.

      • stevefitzpatrick

        Apparently my humor was lost on you. I do hope that Grant visits Lucia’s blog…. You too.

      • Humor? If you say so.

        BTW, I’m puzzled why you care so much where comments appear. You can reply to Tamino on Lucia’s, he can reply to you on here. There are cross-links (how do you think I found your post on Lucia’s in the first place?) so following the discussion is no problem. The obvious benefit is that each can write an entire blog posting if they want. I’d much rather see a substantial reply to the issues raised regardless of the venue than whining about “censorship” and “selective editing”, as if censorship is even possible anyway.

        And I’m still great. Trust me.

    • Chris O'Neill

      “That guy Steve F may actually understand quite a lot more than you give him credit for.”

      That’s a hilarious piece of irony. The humor is not lost on me.

    • Straining not to go too far off topic but I’d suggest that having comments on blogs at all is an indication of the web being broken. It’d be much better if everybody could own their own comments and the web provided a way for the reader to find the ones they find interesting (without the owner of the original post having any part in the matter – either the ability to “censor” or providing any implicit endorsement by not doing so).

      This is one of the main distinctions between Ted Nelson’s Xanadu dreams and Tim Berners-Lee’s 404 pragmatism. The other distinction, of course, is that the web exists and, mostly, works; not an inconsiderable point in its favour.

      Anyway, replies of more than a few paragraphs seem better hosted on their author’s blog. In a web which is working well it should be easy for the reader to find them, e.g., by links from the article commented on.

  4. Susan Anderson

    Sorry guys, I can’t resist. Enjoy:

    And in case you don’t feel like going over there, some lovely bits (quote was one of Tobis’s – depending on how you feel – finer or more irrational moments. Greenfyre’s highlighting picks out “survival of the planet is at stake”, which is, after all, the point. It’s why we’re so angry, and it’s why we bother. Otherwise you guys could spout all the ill-willed garbage you wish:

    “Let me explain why. It is not because I am a pusillanimous chickenshit, Mosher. It is because the fucking survival of the fucking planet is at fucking stake.”
    What stood out for me was Tobis cutting to the core issue and expressing a decidedly less optimistic worldview than he generally does. Calling Mosher out that way for vainly playing self-indulgent games, suffering as he does from the misapprehension that he is some sort of rebel, was rather harsh, but then I wasn’t sure what Tobis’ real purpose was.
    Those, and the use of “pusillanimous“ which I have to admit I think is a lovely word and I always take note when someone uses it correctly.

    • Susan Anderson

      btw, Mosher and his pals seem to regard me as a pincushion. It’s rather a compliment, but not nice, not nice at all (ref. Smeagol).

  5. Steve F asked : ” Is there something wrong with commenting at Lucia’s blog? “

    The funniest thing about this comment by Steve F is that Lucia doesn’t allow rhetorical questions at her place.

    • steve fitzpatrick

      The question was not rhetorical.

      • steve fitzpatrick.
        So are you genuinely countenancing the possibility that ‘commenting at Lucia’s blog’, something you yourself practice, is wrong?! Or is your understanding of what is and is not ‘rhetorical’ less than perfect?

  6. “Calling Mosher out that way for vainly playing self-indulgent games, suffering as he does from the misapprehension that he is some sort of rebel, was rather harsh”

    Nothing is harsher than Steven “Piltdown Mann” Mosher deserves.

  7. Timothy (likes zebras)

    So, if I follow your argument correctly, the reason for choosing a post-1975/1979 data set is so that you can ignore the uncertainties with the impact of anthropogenic sulphate aerosols on climate?

    These uncertainties being the main reason why papers such as Gregory’s have such difficulty with using the observational record to constrain estimates for climate sensitivity.

  8. > zebras
    Have I followed your herring correctly?

    You seem to be accusing our host of trying to “ignore the uncertainties” of the anthropogenic sulphate peak of the pre-1970ish decades.

    Pointer welcome to a comparable satellite data set.

  9. With regard to the comment of June 16, 2013 at 11:44 pm, the moniker is the message. Wallowing in and glorifying stolen correspondence leaves a durable, essentially permanent stench lingering around one’s reputation and in the long run is a poor choice for advancing one’s own interests.

  10. Steve Mosher, (and Steve F)

    Regarding choice of 1975 as start point – Tamino explained his reasons here – he may be wrong, but I believe he was being genuine

  11. There’s not often a “correct choice” for trend removal, but the emphasis on the need to include data pre-1979 is ridiculous, and a distraction. It would be good for anyone doing such an analysis to test the robustness of their results to different time periods and statistical choices. Drawn out arguments about these choices, particularly without testing the sensitivity, is not particularly useful (note I agree with Tamino’s general argument here though I’m not sure how much it matters, I haven’t done it myself). The shape of the global temperature time-series does call for something other than mere linear regression if using a rather long time period.

    It would be interesting to do this at a grid box by grid box level in gridded datasets. Linear trends could be removed at each grid cell (I think this is fine via linear regression for some appropriate time period) and ENSO could be removed by estimating the “maximum lag” between the Nino index of interest and the variable of interest (e.g., T anomalies at the grid cell), which could be done by finding the maximum correlation at, say, 0-6 months lag at each grid cell and using that lag. This would remove contemporaneous relationships in the temperature field, but highlight the “maximum ENSO influence” which is realized at different lag times spatially via atmospheric teleconnections.

    • “It would be good for anyone doing such an analysis to test the robustness of their results to different time periods and statistical choices.”

      I tried here with multiple regression with either linear or quadratic in the mix, and starting at 1950 or 1979. I tried GISS and NOAA as well. I used nlm() to optimise the delay coefficient and peform the regression. I got generally much higher post-1997 trends than SteveF, although his case, HADCRUT 4 linear starting 1950, was an exception in several ways, and there my results were similar to his.

      • Gavin's Pussycat

        As I pointed out, all this has been done — using the real forcings. There’s nothing new to be discovered, but hey, it’s a harmless hobby…

      • Well, Lean and Rind did not lag the forcings – F&R added that, and this extends to some different cases, It uses an optimisation routine to determine the lag, which is implemented via exponential smoothing. A few things are different. Also the hubbub about a pause has grown, so the analysis needs to be kept up to date. I’ve posted the code.

      • I’ve posted the code

        More importantly, Nick has GISS updating each month at moyhu AND wood for trees!

  12. Steven Mosher, have you tried bathing in tomato juice?

  13. > got generally much higher post-1997 trends than SteveF, although
    > his case, HADCRUT 4 linear starting 1950, was an exception

    What makes it exceptional; or, what change would make it not exceptional? Is it exceptional only assuming that specific starting date? (if you can simplify for the non-scientist reader)

    I see Wikipedia says: “Overall, the net effect of HadCRUT4 versus HadCRUT3 is an increase in the average temperature anomaly, especially around 1950 and 1855 ….”

  14. “What makes it exceptional; or, what change would make it not exceptional? Is it exceptional only assuming that specific starting date?”

    I think it’s related to Tamino’s issue about the dip in the middle after detrending. All three datasets give lower post-97 trends if you start from 1950 with only a linear plus exogenous. HAD4 is just exceptionally low, and also has a very long lag.