2012-02-23 03:17:25Monckton Misrepresents Reality (Part 3)
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Final part from us, although we'll probably do a potholer Part 4.  In particular, the runaway warming section could probably use a bit of work.

Monckton Misrepresents Reality (Part 3)

2012-02-23 03:45:33
MarkR
Mark Richardson
m.t.richardson2@gmail...
192.171.166.133

I agree that the runaway section does need work. Got to rush off now, will try to comment later.

2012-02-23 03:49:16
Alex C

coultera@umich...
67.194.46.72

I need to rush off myself, but one change I did make (to all the posts in fact) was to alter the second link so that it led to a page that *non-authors* could see.  The general page containing both parts of the Monckton Myth 17 is only an author page, the ones that were actually published were parts 1 and 2 (I linked to 1).

2012-02-23 03:54:52
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Yeah thanks Alex, I guess I only published the version that was broken up into 2 parts.

2012-02-23 11:08:28
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

Dana, I thought you may have missed my post here.  The gist is that Monckton has used the same statistical technique that he condemns so vocally, and which the IPCC did not use.  It also evident that at seperate times he has run a logically distinct argument from the same graph, and that he is likely to pull a switch on us.  Anyway, full post below:

 

"For Part 3, I want to draw your attention to Monckton's  St Paul speach which is criticized by Abraham.

In particular, Monckton states that the IPCC concludes that sea level rise from the melting of the great ice sheets will by 6 cm by 2100.  The way he states it on his slide (at 25:40) is "IPCC: 6 cm sea level rise from the great ice sheets in 100 years."

 That is of course false.  The IPCC draws no such conclusion.  The relevance is that he adduces as evidence this table from the IPCC:

In his response to Abraham, Monckton said:

"53: Did you also notice the IPCC’s table of observed contributions to sea-level rise over the past 40 years, from which it is not difficult to calculate that on present trends the great ice-sheets will contribute around 6 cm to sea-level rise over the next 100 years?" 

Now, clearly the maximum rate determined in the table is 0.42 mm/ year from the Greenland and Anartic ice sheets, or 4.2 cm/century.  Evidently Monckton has taken the trend from 1961-2003, and the trend from 1993-2003, and from them deduced a trend for 2001-2100.  In other words, he has himself employed exactly the "bogus statistical technique" he falsely accuses the IPCC of using.

I think we should definitely point this out in the response.

As an aside,  the IPCC AR4 projection for the A2 scenario was a combined Greenland/Antarctic ice sheet contribution betwee -11 cm to 5 cm primarilly due to increased snowfall in Antarctica.  However, that assumes no acceleration of discharge due to the break up of the ice sheets.

 

In any event, it is worthwhile also checking Monckton's St Paul lecture on the IPCC's purported statistical blunder (43:40), which he casts as an endpoing fallacy.  He attempts to rebut the IPCC by showing his own endpoint fallacies, showing trends from 1994, 1998, 2001, and 2005, all terminating in 2009.  What he clearly fails to mention, is that all of the IPCC trends are statistically significant, whereas none of his are.  (Well, perhaps the one from 1994, which is borderline, but also strongly positive).  The issue of statistical significance completely devestates his nonsense.

That is not how he has presented the supposed fallacy in his latest presentation, so no need for ammendment on that account.  But we should be aware that he is trying to run two distinct arguments from the same graphic, and is dishonest enough to switch mid discussion to avoid a devestating rebutal.  Expect him to now represent himself as using the St Paul version of the argument rather than the one he actually did use, and to which we have responded."

2012-02-23 11:15:27
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

I also have some issues with the runaway section, and no time.  I'll try to clarrify my issues this afternoon.

2012-02-23 12:14:26
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Tom, how do you think he gets from 4.2 to 6 cm?  An assumption that since there was an acceleration from 1961 to 1993, there will be some further acceleration?  That seems like a very un-Monckton assumption, essentially admitting that global warming will accelerate.  My problem with this is that since he doesn't actually state how he gets the 6 cm figure, we're assuming he's using this so-called 'fraudulent statistical technique', and assumptions can get you in trouble.

2012-02-23 14:02:48
Chris Colose

colose@wisc...
169.226.41.99

I haven't checked everything here, but after a quick scan:

1) The discussion of his radiative forcing-climate sensitivity calculation will probably lose a lot of readers.  That can all probably be summed up much more nicely.  I would not cite a specific number like "1.4 W/m2" of anthropogenic forcing.  This does a severe injustice to the large degree of uncertainty in aerosol effects, which is just one reason why it's not very worthwhile to use the observational record to deduce sensitivity (which makes his factor of 0.4 that supplements the Myhre equation unsuitable).  Of course, we can still agree that the total forcing is well under that of a doubling of CO2. 

Even the part about transient vs. equilibrium (while important to point out) seems very drawn out to me, and I think readers will have a tough time distinguishing the main points.  Maybe it's just me.

 

2) On the issue of runaway warming, it's unclear how you define "f" in this post, which is inconsistent with a lot of usage in the literature. 

The whole point of "When you add CO2 to this model system, there is indeed positive feedback, so even more CO2 will end up in the atmosphere than you dumped in from the burning of fossil fuels" is extremely confused in the context of this topic.  When discussing climate sensitivity, carbon cycle 'feedbacks' aren't really feedbacks at all, since the resultant CO2 concentration is just meshed into the definition of the radiative forcing.  If you prescribe the final forcing  to be that of a doubling of CO2 (i.e., ~4 W/m2), then you can't talk about CO2 feedbacks.   You're stuck at a doubling of CO2!! This is one error Arthur Smith called Monckton out in his APS "paper" on climate sensitivity (which would be good to point to by the way, since it came up in the debate with Denniss).

But as a recommendation, rather than talking about feedback loops and such, a more intuitive way to get this across may be to simply mention that the planet radiates to space more efficiently in a warmer world.  This is ultimately what allows for equilibrium, and the net effect of this is still to win out over the positive feedbacks (like water vapor, albedo, etc).  However, those positive feedbacks make the radiative feedback response less efficient at coming back to equilibrium.  If you were to do a calculation that just paramterizes this, rather than the outgoing flux depending on T4, you might model it as T3.9 when including a water vapor feedback (over a limited range of temperatures in a back-of-envelope calculation).  The climate still can come to equilibrium, but it's a bit more sluggish at doing so.

3) Minor points- In references to the transient sensitivity, write yr-1 (as opposed to yr-1).  NASA GISSA is a typo. 

2012-02-23 16:35:56
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

If somebody else wants to take a stab at revising the runaway section, that would be helpful.  It's a tricky one.

2012-02-23 17:25:49
Chris Colose

colose@wisc...
96.249.11.201

Dana,

I can try to revise that section tomorrow if you wish.  SkS already has a post on this (the one linked in the Monckton article), though to be frank it is rather confusing and misguided in certain aspects, especially the 'Advanced' section.  If you don't mind, I can also write up a side tutorial post on how the outgoing longwave radiation behaves as a function of surface temperature in the context of a positive feedback system. 

2012-02-24 00:01:59
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

With regard to feedbacks, Monckton writes:

 

"I did not say that the climate will behave “exactly” as it has in the past. We are capable of exerting some influence over it, but not very much. The notion that we can exercise a large influence is based on the mistaken idea that the initial warming from a doubling of CO2 concentration (which might be about 1 K) will be tripled by net-positive temperature feedbacks. This unfortunate assumption is what truly separates the IPCC from scientific reality. The IPCC makes the mistake of assuming that the feedback mathematics that apply to an electronic circuit (Bode, 1945) are also applicable to the climate. In two very important respects that the models are tuned to overlook, this is not so. First, precisely because the climate has proven temperature-stable, we may legitimately infer that major amplifications or attenuations caused by feedbacks have simply not been occurring.

Secondly, the Bode equation for mutual amplification of feedbacks in an electronic circuit has a singularity (just above the maximum temperature predicted by the Stern report, for instance, or by Murphy et al., 2009) at which the very strongly net-positive feedbacks that reinforce warming suddenly become just as strongly net-negative, dampening it. I have not yet heard of a convincing physical explanation for any such proposed behaviour as applied to the climate. But if we must use the Bode equation then it necessarily follows from the climate’s formidable temperature-stability that the feedback loop gain in the climate system is either zero or somewhat net-negative. A climate subject to the very strongly net-positive feedbacks imagined by the IPCC simply would not have remained as stable as it has."

(My emphasis)

I believe he is refering the the equation:

 

f = 1/(1-g)  (7)

as detailed in Hansen et al 1984 (who cites Bode 1945).

 

f = the equilibrium temperature response/ the initial temperature change.  For f = 3, the aprox central estimate of climate sensitivity from the IPCC,  g = 2/3.

Clearly, as g approaches 1, f approaches infinity, ie, the singularity to which Monckton refers.  Further, for 2  > g > 1, as g gets larger, f approaches 1.  That is, it becomes the strongly net negative feedback to which Monckton refers.  Therefore I feel confident this is the equation that Monckton is refering to.

That is unfortunate for Monckton, as Hansen defines g as change in temperature due to feedbacks / the equilibrium change in temperature.  The equilibrium change in temperature is further defined as the initial change in temperature plus the change in temperature due to feedbacks.  Therefore, g can only be greater than 1 and less than 2 if the initial change in temperature is of opposite sign to the equilibrium temperature, and the change in temperature due to feedback is greater than the equilibrium change in temperature.  Further, f is only undefined when g equals the equilibrium change in temperature and the initial change in temperature is 0.

In other words, the equation is perfectly well behaved.  It is just that Moncktion is giving it mathematically nonsesnical interpretations. 

I think this is worth bringing out in the response.  However, it will take somebody more mathematically adept than I am to make this clear, and to make clear how absurd are Monckton's objections.

Over to you Chris :)

 

Note, the claim that this unsettling behaviour occurs at only a slightly larger temperature increase than that assumed by Stern is complete nonsense if assumed to refer to f, rather than g, and it is of course f that gets all the common discussion, not g.

2012-02-24 00:08:16
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

Dana, with regard to the sea level rise, in his response to Abraham, Monckton states quite clearly that his method is an inference from the "observed contributions to sea level rise over the past 40 years", and that is the only data he presents in his St Paul performance in support of the claim.   That data consists of two trends with a common terminal year.  The shorter period of the two trends has the largest trend, but is smaller than the trend he postulates for the 21st century.  Therefore he is not simply extrapolating a trend, but infering a future trend from the progression of past trends, ie, exactly the argument he claims is invalid.  I do not think we need confirmation to make that point.

2012-02-24 01:40:47
Chris Colose

colose@wisc...
96.249.11.201

Tom

Hansen reversed the terminology compared to what is used in the modern literature (f vs. g, so most people would write g = 1/1-f, which only applies in a linear feedback system).  It is confusing but I'll try to convey it simply without getting into all those complexities.

2012-02-24 03:08:52
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Chris, thanks, look forward to seeing what you come up with for the runaway section.  Maybe we can use it to revise/replace the existing advanced rebuttal too.  Also let me know if you'd like to be added as a co-author to this Monckton Part 3.

Tom - I see your point.  How about we create a graphic to illustrate what Monckton appears to be doing with the sea level rise data, i.e. draw a linear trend of 0.19 mm/yr from 1961-2003, 0.42 mm/yr for 1993-2003, and then a dashed linear trend from 2003 to 2100 at about 0.6 mm/yr?  Basically mirroring the IPCC graphic to illustrate that he's doing the same thing, of course while noting that it's our interpretation of Monckton's argument, not his own graphic.

2012-02-24 11:32:19
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

Dana, the graphic is a good idea, but unfortunately currently beyond my graphic abilities.  (One of these days I'll learn how to make calc make decent graphs.)  In introducing the contradiction, we should probably say something along the lines of:

"Curiously, although the IPCC is not guilty of the bad statistical argument Monckton attributes to them, one frequent commentor on Climate Science is - Christopher Monckton.  In his St Paul adress (and other places) he has cited the IPCC as predicting a 6 cm contritubtion from the great ice sheets of Greenland and Antarctica to global rise in sea level by 2100.  The IPCC, however, quotes no such figure.  Indeed, Monckton bases his claim, not on any statement from the IPCC about the contribution of the great ice sheets by 2100, but on the trends from 1961 and 1993 to 2003, as shown on the IPCC WG1 table 5.3, and displayed on Monckton's slide.  Monckton clarrifies how he derived the figure in his response to Abraham, saying:

"[N]notice the IPCC’s table of observed contributions to sea-level rise over the past 40 years, from which it is not difficult to calculate that on present trends the great ice-sheets will contribute around 6 cm to sea-level rise over the next 100 years".

[graphic]

"Present trends", if extrapolated linearly only show a rise of 4 cm by 2100, so Monckton must have calculated a change in trend based on the greater trend between 1993 and 2003 than that between 1961 and 2003.  That is, Monckton has made the same type of inference which the IPCC refused to do, and which he describes as  a "fraudulent statistical technique"." 

IMO, the Graphic should show to solid line trends to 2003 labelled as the observed trend, and three dashed lines to 2100, one being (and labelled)  the linear projection of the 1963-2003 trend, one being (and labelled) the linear projetion of the 1993 trend, and the third being and labelled as "Monckton's inferred trend".

If you want to be a bit cautious about this, we can explicitly mention that we have inferred Monckton's technique from his public statements; that Monckton has declined to answer questions directed to him about sources etc, and finish with a challenge to Monckton to show the exact calculations he used to derive the 6 cm sea level rise from table 5.3.  That way when he denies he used that technique (which he will regardless of what he did), he will be compelled to show his transparently erronious workings, or by ignoring the challenge be shown to have had no basis for his projection.

2012-02-24 11:48:50
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

I can make the graphic easily enough.  I think I'll push this post back to Saturday so that we have ample time to make all these changes.  That way I can make the graphic tomorrow.  It would be nice if I could find some global sea level data to use prior to the satellite record.  I've always had a hard time finding that.

2012-02-24 14:22:39
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Hi Dana,

I just emailed you the GSL data from Church and White (2011).

2012-02-24 14:48:54
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Oh sweet, thanks Alby.

2012-02-24 15:40:08
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

dana, I'm not sure you will want actual sea level data in the graphic.  Actual data would be the "observed" in table 5.3, ie, 3.1 mm per year, where-as the discussion is focussed on the Greenland Ice Sheet and the Antarctic Ice Sheet contributions.  If you want observational data, you need data for the Greenland and Antarctic Ice sheets mass balance, using mm of sea level contribution as your units using the conversion factor mentioned in the caption (100 Gt/yr is loss = 0.28 mm/yr sea level rise)

2012-02-24 15:44:53
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Bah crap you're right, I forgot we were just talking about Greenland and Antarctic contributions.  Well, just plotting the trends will suffice.

2012-02-24 16:04:14
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Dana,

Not sure if this helps, but Church and White seperate out the contributions for different time periods from GIS, Antarctic Ice Sheet, as well as other terrestrial glaciers and ice caps.  Those data are in their Table 1 IIRC-- you should be able to access it here.  If not, try here.

Other poosible papers of interest here.  Note Jacob et al. (2012).

2012-02-24 16:10:51
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Thanks Alby, but Church and White only give long-term trends, which IPCC also give.  I was hoping to be able to plot annual data and then add the trend on top of it, as in the IPCC figure Monckton criticizes.  If Jacob has data, it's only since 2003.  Not a big deal though - it's mainly just for illustrative purposes, and the trends are what's important for that.

2012-02-24 16:50:32
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

How's this look?  Problem with not having the annual data is that I didn't know where to start the trends from, so it looks a little different from the IPCC due to the temperature variability in their figure.  I think this basically gets the point across though:

monckton slr

Suggestions?

2012-02-24 17:45:51
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

It's a shame the inflexion point between the 1993-2003 trend and Monckton's inferred trend is so slight.  It may be possible to make it more obvious either by having a large inset focussing on 1985-2015, or by extending the red dotted line to intersect with the x-axis.  Alternatively, it may be better to have the main chart be the 1985 - 2015 closeup, with an inset showing the full 140 years.

 

Also, to highlight the similarity between this graph and Monckton's, we need the the midpoint of the observed 1993 -2003 trend line intersect with the 1961- 2003 trend line, and the inferred trend line intersecting the middle of the observed 1993-2003 trend line.

 

If you want annual data as a backdrop, you need to use inverted mass balance data appropriately scaled.  I found this image at GRIDA:

Data is cited as being from Rignot et al 2008 (pay walled for me).  Unfortunately their trends may not agree with the IPCC trends given that the IPCC is based on multiple studies and methods.  Further, we would need equivalent Antartic data, and to combine them.

The graphs from the IPCC itself are useless for this purpose:

All in all, annual data is probably more trouble than it is worth.

2012-02-25 03:00:56
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Agreed on annual data.  I was also thinking that the solid lines should intersect around 1998 rather than 1993.  I could try extending them to the x-axis as well.

2012-02-25 03:20:47
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Also see and here.

2012-02-25 12:35:57
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Okay, I incorporated the sea level graphic and point, and also replaced the runaway section with a summary of and link to Chris' post.  I suggest tomorrow morning we publish Chris' post, then publish this one in the afternoon.

Chris, please have a look and make sure my summary of your runaway post is correct.

2012-02-25 14:01:23
Chris Colose

colose@wisc...
169.226.41.99

Dana, it's good, except this part is backwards

//"However, it would take a substantial OLR increase to create a situation in which the global system can't warm enough"//

Just the opposite, the issue is that the OLR can't possibly increase enough.

How about replacing the paragraph section "In short...cause a runaway greenhouse effect" with:

In short, a system with positive feedbacks will not be as efficient at changing its outgoing longwave radiation (OLR) in response to forcings (like changing sunlight or CO2).  Therefore, a higher surface temperature change must result in order to accomodate the necessary change in OLR to re-establish planetary energy balance.  There are situations in which the OLR cannot increase enough to match the amount of sunlight that Earth absorbs (Figure 2) in which case a runaway greenhouse can be triggered; however, Earth is nowhere near this state of affairs. 

2012-02-25 15:05:44
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

Very minor point, if Monckton's inferred trend as shown in 0.6 mm/year, then the the intersect with the 1993-2003 trend should probably be in 2001.  Nor sure that is worht redoing the graphic for.  

 

Other than that, my only remaining concern is the runaway feedback section.  That adresses the principles he is ignoring, but not his explicit errors as discussed in my post of 24th Feb, 12:01 am.  I understand that including discussion of both the fundamental issues (as in Chris's post) and the technical details of Monckton's error (as I suggest) may make the section too long, and confuse some people.  Further, if only one should be retained it is Chris's discussion as being more educational to our readers.  Never-the-less, I want to flag the lack so that the decission to not explore the details of Monckton's error is deliberate rather than inadvertent. 

 

Which evey way Dana decides to go on that, the post is certainly ready for publication as is, IMO.

2012-02-25 17:11:28
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Chris - thanks, change made, except I put the reference to Figure 2 at the end, because I didn't use the graphic showing the runaway greenhouse effect, I used the previous one.

Tom - I updated the figure to intersect at 2001.  I had the same thought earlier, simple change to make.

I didn't incorporate your 12:01 comment because of Chris' subsequent comment that Hansen swapped f and g, plus I think Chris did a really nice job covering the subject in his post.  Part 3 as a whole has already gotten quite long too, so unless you think it's an important addition, I'm happy with it as-is.

2012-02-25 17:28:34
Tom Curtis

t.r.curtis@gmail...
112.213.180.137

dana, whether it is an important addition depends on our primary purpose for the post.  Do we want to educate and counter Monkcton misinformation?  In that case a clear exposition of the reason there is no risk of a runaway greenhouse effect despite climate sensitivities of of 3 or 4 degrees C per doubling of CO2?  In that case Chris Colose's explanation is clear and elegant, and further discussion only detracts from it.  But if instead we want to nail Monckton's ass to the wall and reveall to all and sundry that he is a fool?  Well, in that case showing that his argument is based his making mathematically absurd assumptions is definitely worthwhile.  

 

Now, I freely admit to having a prejudice in the later direction.  This is not only because I think making Monckton a laughing stock so ridiculous that not even Watts will touch him will do a great service to public education on climate science.  That service is twofold, for it elimnates Monckton as an effective propagandist against climate science, but it also tarnishes the reputation over every denier who has embraced him closely, which is all the corporate funded denier organizations. Not only that, but also because I can be a vindicitive son of a bitch, and the existence of people like Monckton who so abuse their public voice is odious to me.  

 

However, despite my personal preferences, I recognize the former is far more consistent with the aims of SkS.  I also firmly believe that clear exposition of the science in easilly understandible terms is the best weapon we have against the climate propagandists in the long term, and detracting from that strategic ambition to pursue the more tactical end (nailing Monckton to the wall) may not be desirable.  So, with those considerations in mind, your call (unless Alex C, as co-author, has a strong preference).

 

Edited to add:  the sea-level graphic is now perfect.   

2012-02-25 17:32:57
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.137.108.231

Well, I think we've pretty well nailed Monckton to the wall already Tom.  I mean, this is a 3 part post :-)  One more point won't make much difference, and personally I'd prefer clarity over an extra nail.  Though I'm open to adding the further point (assuming the f-g swap doesn't change the discussion) if others think it's worthwhile.

There's always the overkill backfire effect to worry about though, and we've already got a ton of material in this post.

2012-02-25 18:01:53
Chris Colose

colose@wisc...
72.226.121.32

If you guys do want to throw in some stuff about g's and f's and a more formal mathematical treatment, Gerard Roe has a good 2009 paper here that goes through all the equations and gives some good interpretation, as well as some diagrams that might be handy.

By the way Dana, I've slightly modified the quote in my article that you put in the Monckton article,:

//"There is in fact a negative feedback that always tends to win out in the modern climate.  This is the increase in planetary radiation emitted to space as temperature goes up.  Positive longwave radiation feedbacks only weaken the efficiency at which that restoring effect operates."//

Compared to the last quote, I just wanted to make sure it is understood that the statement is valid in the modern climate, and not necessarily in all situations (which is why we can talk about a runaway greenhouse or snowball Earth to begin with).