# Open Mind

## Volcanic Lull

#### October 19, 2008 · 115 Comments

The question is often raised, what caused the global warming in the early 20th century? I’ve often said that one of the reasons is an unusual lull in climate-impacting volcanic activity at that time. But most find this explanation unpersuasive; how can a lull in climate-forcing volcanic activity lead to a steady rise in global temperature over several decades?

The answer lies in the fact that earth has several different system components which respond to the input of energy with different characteristic time scales; the atmosphere, land, upper ocean, deep ocean, and cryosphere all have different time scales for response to climate forcing. That atmosphere responds quickly, land takes longer, the ocean (especially the deep ocean) takes much longer.

The atmosphere responds with a time scale of about 1 year. That doesn’t mean that the full effect is felt in a single year! It means that if a forcing it sustained, then a fraction 1/e (where e=2.718… is the base of natural logarithms) is felt in the first year. But the huge thermal inertia of the oceans gives that component of the system a much longer time scale.

Computer model simulations indicate that the characteristic time for the climate system as a whole is about 30 years. Suppose we use a “two-box” model (a simple model of the climate system mentioned here) where one box has a characteristic time scale of 1 year and the other a characteristic time scale of 30 years. How would such a system respond to changes in climate forcing?

As mentioned in this post, the impact on temperature evolution of a time scale due to thermal inertia is to “exponentially smooth” the climate forcing. For a two-box system, the temperature due to a time-varying climate forcing F(t) is a superposition of two functions. One is the 1-year smooth of the climate forcing, given by

$S_1(t) = e^{-t} \int_0^t F(s) e^s ~ds$,

the other is the 30-year smooth given by

$S_{30}(t) = {1 \over 30} e^{-t/30} \int_0^t F(s) e^{s/30} ~ds$.

Hence the temperature will given by

$T(t) = c_1 S_1(t) + c_2 S_{30}(t)$,

where $c_1$ and $c_2$ are constants. The climate sensitivity in this model (sensitivity to unit climate forcing, not sensitivity to doubling CO2) will be $c_1 + c_2$.

To apply this model, we need to know the time-dependent climate forcing F(t). Fortunately, GISS provides climate forcing estimates from 1880 (the beginning of the GISS temperature estimates) through 2003. They even give a breakdown of the forcing by source.

If we consider volcanic forcing only, it looks like this:

The climate forcing impact of irregularly timed volcanic eruptions is evident; each downward spike in the volcanic forcing represents a major, climate-influencing volcanic eruption. Part of the impact of this will be due to its 1-yr smooth; thats looks a lot like the raw forcing data. The impact of the 1-yr smooth will be the “prompt response” to climate forcing, the global cooling noted for a few years subsequent to these eruptions. We can also plainly see that there’s a distinct lull in climate-impacting eruptions from just after 1910 to just before 1960.

The 30-yr smooth looks quite a bit different:

We note that there’s a large decrease early in the data. This is not necessarily reflective of an actual decrease, it’s because for the earliest few decades we’re estimating a 30-yr smooth based on considerably less than 30 yrs. of data. To approximate the 30-yr smooth with any accuracy, we should only use the data from about 1900 on, giving us 30-yr smoothed values based on at least 20 years of data. That looks like this:

Now we can plainly see that the lull in volcanic eruptions leads to a sustained increase in global temperature, due to the delayed response of that part of the climate system which exhibits a 30-yr time scale.

It’s also worth noting that the lull in volcanic activity isn’t the only climate-forcing factor operating during the early 20th century. Greenhouse gas forcing is also on the rise during this period:

To apply our two-box model, we should fit the observed temperature data to the 1-yr smoothed and 30-yr smoothed time series for net climate forcing:

Doing the multiple linear regression, we get this fit:

The fit is quite good, except for the few years preceding and during the 2nd world war. The estimated climate forcing $c_1 + c_2$ is 0.71 +/- 0.13 deg.C/(W/m^2), which includes within its error range the GISS modelE estimate of 0.68 deg.C/(W/m^2).

This is of course a simplified model, which can’t be expected to be as realisitic as general circulation models. But it illustrates excellently the point of this post: that the lull in volcanic activity in the early 20th century can indeed cause a sustained warming over multiple decades because of the long response time of certain components of the climate system. In fact, it can’t not cause such a sustained warming.

UPDATE UPDATE UPDATE

If climate forcing suddenly increases by a constant, and remains at that value, the two terms (a fast one due to $S_1$, a slow one due to $S_{30}$) look like this:

If we then compute a weighted sum of these terms, we’ll have a plausible time evolution of temperature due to the constant higher forcing:

Categories: Global Warming

### 115 responses so far ↓

• Raven

Tamino,

If the oceans can respond over decades then why can’t the respond over centuries? How do we know that a significant portion of the warming has not been caused by the release of heat stored in the ocean as previously submerged currents surface?

We simply do not know enough about the ocean to rule it out as a possibility.

[Response: I'll leave it to others to point out the many holes in your "theory." I will mention that it doesn't explain how additional greenhouse gases could fail to warm the climate system; no one has ever given even a pluasible, let alone demonstrable, explanation of that. Just as it's impossible for a lull in volcanic activity to fail to warm the climate, it's similarly impossible for greenhouse gases not to warm the climate.

It amazes me the lengths to which you will suggest implausible theories with zero evidence, rather than accept the reality of greenhouse gas warming.]

• Somewhat related, Lean and Rind (2008) create a simple empirical model using volcanic, solar, and anthropogenic forcings, plus ENSO. Their correlation coefficient is 0.87, and they calculate a contribution from solar that is roughly half of the Camp and Tung figure. What’s interesting (but not surpising given the trouble GCMs have), is that this model diverges the most during WWII, and a period of time centered around 1910 (the authors incorrectly identify this as during WWI, but not according to my history book). The 1910 period is “low” and the WWII period is “high” which leads to the large warming in the first part of the century. We know of the mid-century SST problems, which might shave down some of the mid century bump, but I wonder if there are systematic problems around 1910 as well.

The paper also has an interesting discussion of how ENSO and the various forcings effect regional climate, and how the models exaggerate warming at extreme high latitudes.

So Raven, How does heat radiated from the Oceans simultaneously warm the troposphere and cool the stratosphere? Why would the oceans continue to warm even as they radiate heat away? I can think of many things wrong with your theory. If you try, I’m sure you can, too. Or do you only apply skepticism to theories that are actually scientific?

• Raven

Tamino,

LPT is a well established phenomena. The only question is how much effect it has on he climate system. If it does have a large effect then the oceans would be the most likely vector. And yes, LTP is an extremely plausible hypothesis.

As for you other red herring: no one needs to explain why GHGs have a sensitivity less than that what the models suggest because the models don’t take LTP into account which means any sensitivity estimate by the models is wrong.

More importantly, we do know that the climate system has gone through cycles like the MWP and we don’t really understand why. Some have speculated that the sun causes it but that is highly disputed. LPT is one plausible explanation.

[Response: The gravitational redshift is also a well-established phenomenon. I suppose you'll be suggesting that next, as an alternate hypothesis to greenhouse-gas forcing.

You do not discuss; you babble.]

• dhogaza

Some have speculated that the sun causes it but that is highly disputed.

But wait! In a moment or two you’re going to remind us that the sun is causing *current* warming!

• Raven

dhogaza says
“But wait! In a moment or two you’re going to remind us that the sun is causing *current* warming!”

Direct changes in TSI cannot explain the warming. If there is a solar effect is must be indirect via clouds or the ocean. The short answer is we don’t know.

But we do know the models say the direct TSI effect on climate is 1/10th the effect of the sun when estimated empirically by papers such as Camp and Tung. The difference between the two suggests an indirect mechanism could be at work.

• Richard Steckis

Dhogaza,

There is no current warming. The trend at the moment is either static or declining. With the PDO going into a cool phase, the assessment is that there will be a 20 to 30 year cooling phase ahead of us.

When will you people realize that GHG forcings are increasingly shown not to be the major climate forcers. CO2 continues to increase unabated whilst the climate system is starting to enter a cool phase.

• Gavin's Pussycat

Raven, LTP is a buzz word, not an explanation. If you do the math, you find that natural variability is of 1/f type; according to the data (over the instrumental period), according to the models (for as long as you care to run them) and according to the proxy reconstructions (no, there is no MWP to speak of). It all fits together.

The power spectral density is

(1/f)df = -(1/T)dT = d(ln f) = - d(ln T).

I.e., there is the same amount of power for every “octave” of frequency f, or characteristic time period T = 1/f. This expression diverges upon integration, but only logarithmically so.

If you look at the power density of the time derivative, you get

-(1/T) d (ln T) = -exp(-ln T) d (ln T)

which converges fast when taking the integral to T -> inf. Long term persistence thus has a negligible effect on the recent upswing in temperatures. The necessary power in the periods longer than the instrumental period just isn’t there.

See IPCC AR4 WG1 Figure 9.7 for the PSD, and Figure 9.5 for what it does over the 20th century. There’s much more literature on that, e.g., Manabe & Stouffer 1996, their Figure 17 IIRC; which is old but not paywalled.

BTW I would appreciate it if you changed your handle. There are a couple of ravens living near our summer home; beautiful animals. And smart ;-)

• Gavin's Pussycat

Tamino, what were c1 and c2 separately?

[Response: Response: c1 = 0.08, c2 = 0.63.]

• Murphy

Response to Steckis@4:22am:

it is elementary to show (using blackbody radiation calculations) that without the GHG effect, the earth would be about 30°C cooler than it currently is. Therefore, to say that the GHG climate forcing is insignificant is patently ridiculous. Can’t you come up with better denialism?

The only question left is to deduce the net contribution of each component of the GHG mix (at the current and anticipated future levels of CO2), in particular that of CO2. This isn’t straightforward since the climate system is a linked nonlinear and stochastic dynamical system, but a lower end estimate for CO2’s contribution is about 10% (see here: http://tinyurl.com/co2forcing ) of that 30°C GHG-forced warming, i.e. 3°C. Therefore, if we double the CO2 level, about a 3°C rise in temperature could potentially be expected.

[Resposne: RealClimate estimates the fraction of GH effect due to CO2 is about 25%.]

• Barton Paul Levenson

Richard Steckis writes:

There is no current warming. The trend at the moment is either static or declining.

Look again. Take out the hyphens:

http://www.geoci-ties.com/bpl1960/Ball.html

http://www.geoci-ties.com/bpl1960/Reber.html

• Murphy

Response to Tamino (RealClimate estimates the fraction of GH effect due to CO2 is about 25%). I was refering to the row ” CO2 91 (88, RC78)” in the table at the RC article which I linked to. According to that estimate, if all of the CO2 were to be removed from the atmosphere while leaving the rest intact, then atmospheric LW absorption drops to 91%, i.e. by 9% (drops by 12% using RC78 estimates). This can be considered to be the exclusive contribution of CO2 to the GH effect.

On the other hand, the same CO2 if all other LW absorbers were removed can absorb 25% (26% in RC78) of the outbound LW. The difference between 9% and 25% comes from the overlaps in the absorption spectra, as you know.

PS: the argument in my previously posted comment should be cast in terms of radiative forcings instead of temperatures in order to use the estimates from the table given in that RC article directly.

• It always amuses me when people like Raven say things like “We simply do not know enough…We don’t really understand…”, presuming to speak for science as a whole when what they mean is “I simply don’t know enough…”

Richard Steckis - you quote the assessment. Care to provide a link to wherever you got that assessment from?

• Lazar

Raven,

What GP said.
Warming since 1975 is not due to a linear trend. The linear trend is the statistical description. The causes are physical. Similarly, LTP is a statistical model of what has occured, not an ‘explanation’ of why what has occured has occured. The heat has to come from somewhere, by physical process. I’ve seen LTP thrown around like a buzzword, alone with no further explanation, posited as some ’cause’ quite frequently in the comments at Climate Audit.

• transplant

Re Richard Steckis, when will people like you realize they you haven’t the slightest idea what constitutes a climate trend?

• David B. Benson

LTP?

[Response: stands for "long-term persistence."]

• David B. Benson

Tamino — This is most interesting but also puzzeling and a bit troublesome in that it does not appear to agree with results by Reto Knutti et al. in their response to Schwartz:

Knutti, R., S.Krähenmann, D. J. Frame and M. R. Allen, 2008, Comment on ‘Heat capacity, time constant and sensitivity of Earth’s climate system’ by S. E. Schwartz, JGR, 113, D15103, doi:10.1029/2007JD009473

(A pdf of the paper is available from

http://www.iac.ethz.ch/people/knuttir/papers

but I don’t know how to provide a direct link. See Figure 1a on page 2 and the attached text. It is clear that the actual response is far from having an exponential form, and to my eye, far from your two box form.

[Response: But this post doesn't say the response to constant forcing (the case for the Knutti paper after year 200) is exponential. It says the response (in the 2-box model) is the sum of two exponential terms. If we take two exponential curves (like those plotted in Knutti et al.) and compute a weighted sum, we can produce a curve very similar to the evolution shown by their computer model. See the update to this post.]

• David B. Benson

Tamino — Thank you for the response and the update. It seems I couldn’t properly do the sum of the two exponentials in my head.

• Paul Middents

Tamino,

This is a lovely exposition. Just enough math for the geeks but still eminently comprehendable.

It has impelled me to kick in a little for the cause.

Paul

• Richard Steckis

to Transplant.

I understand what constitutes a climate trend. And it is not the 30 years that the IPCC trumpet. A climate trend is a change in the average weather over a decadal to mulit-decadal scale. Whithin this timescale trends can be recognised.

I am enough of a scientist to be able to read trends in data. even if you are not.

• dhogaza

A climate trend is a change in the average weather over a decadal to mulit-decadal scale. Whithin this timescale trends can be recognised.

I am enough of a scientist to be able to read trends in data. even if you are not.

This site is run by a professional statistician. Support your claim, and show your work. Otherwise, you’re just weenie-wiggling (and note that our host shows his work in each of his top posts).

I am enough of a scientist to be able to read trends in data. even if you are not.

You should also be - though apparently are not - enough of a scientist to understand that arguments to personal authority (”neener, neener, I’m smart and you’re not”) don’t hold water amongst scientists or interested people with a technical background.

• Richard Steckis

Dhogaza,

“Otherwise, you’re just weenie-wiggling (and note that our host shows his work in each of his top posts).”

Up to your usual insulting self Dhogaza. Tamino can afford to go into detail in his posts. After all he owns the site. He is a professional statistician. I am a professional biologist. That means we can both read science and interpret it. But from different viewpoints.

Richard Steckis,
If indeed you are a scientist (biologist, evidently), then you know that a “slope” is not necessarily a trend. To be a trend, it has to have meaning and be robust–i.e. not sensitive to starting and end points, methodology, etc. To call anything in climate a trend based on only 8 years of data is naive at best, likely delusional, and possibly mendacious.

• John C

[Resposne: RealClimate estimates the fraction of GH effect due to CO2 is about 25%.]

So if doubling the CO2 results in 25% of a 3 degree C rise, ie. only about 0.75 degree C, how come CO2 is stated as the key factor for changes from ice ages to interglacial periods when the CO2 content went from 180ppm to 280ppm …. not even doubled ….so might account for about max 0.5 deg C. The total T change was about 10 degrees C. What accounts for the other 9.5 degrees ?

[Response: So many things wrong, so little time...

First: When considering the entire greenhouse effect, CO2 is not "25% of 3 deg.C." It's 25% of 30 deg.C. And the entire greenhouse effect covers the range of CO2 concentration from 0 to the present value, so the effect is highly nonlinear (it also includes the non-logarithmic range) so a linear estimate doesn't apply.

Second: the global temperature change from glacial to interglacial times is not 10 deg.C, it's only about half that. The 10 deg.C applies only at the poles, which show a much greater change than the globe as a whole due to polar amplification.

Third: CO2 is not stated as the "key factor" for glacial-cycle temperature changes. It's one of the key factors. The "trigger" is changes in the distribution of incoming sunlight due to changes in earth's tilt and orbital configuration (Milankovitch cycles). And of course the disappearance of the vast ice sheets contributes a sizeable climate forcing due to the reduction of earth's albedo.

• John C

Raven wrote :
We simply do not know enough about the ocean to rule it out as a possibility.

http://www.weatherquestions.com/Global-warming-natural-PDO.htm

Yes, John C., We’re all looking forward to Spencer’s tapdancing around the difficulty of getting a steadily-rising trend out of an oscillatory driver. In case you never took differential equations, this would require imaginary coefficients, and indeed, Spencer’s work seems to increasingly have zero projection onto the real axis.

• John C

Thanks for the tips on the ice age to interglacial questions …… so if the average T change is 5 degrees and the CO2 changes from 180 to 280ppm, how much of the T change is due to the CO2 ?? Is it a degree … or half a degree ? Or 2 degrees ? Give that the CO2 lags the temperature by centuries (whereas other feedbacks are immediate), I’m just trying to understand how relevant it is in the glacial / interglacial cycle.

[Response: My best guess: about 1.5 - 2 deg.C. I suggest you ask at RealClimate, they know a lot more than I do.]

• dhogaza

He is a professional statistician. I am a professional biologist. That means we can both read science and interpret it. But from different viewpoints.

You mean statistics vary from one field of science to another? Does addition vary from one field of science to another, too? Is 2+2=4 only true in mathematics and physics, not biology?

• Bob North

Just so there is no confusion, here are the current WMO definitions of climate. Note that neither definition specifies a certain period of time (e.g., 30 years) as has been repeatedly claimed by some. While it seems clear (to me at least) that climate and climate trends should be multi-decadal in nature, there is nothing magical about 30 years and Richard Steckis’s definition is not that far off.

Climate - Synthesis of weather conditions in a given area, characterized by long-term statistics of the variables or state in the atmosphere in that area

Climate Trend - A climatic change characterized by a smooth, monotonic increase or decrease of average value in the period of record. Not restricted to a linear change with time, but characterized by only one maximum and one minimum at the end points of the record

From World Meteorological Organization — Guide to Climatological Practices, 2nd edition WMO Guide

• Lookumup. Any of these?

Every once in a while I try to put myself into the troll mind, and its baffling conjunction of business school, Sunday fund-raising megachurch rhetoric and crank science built on crank science .. LTP … “is it to Sigma 6?” etc.

… and then the spiders come.

• HankRoberts

> climate … 3o years

Global temperature trend is where the 30-year period seems relevant.

Local climate — depends on your use. If you want to plan a vacation, California’s climate is recommended.

If you’re a Spaniard seeing the Central Valley — at the time a rich marsh full of wildlife — you may think the climate ideal for longterm development — decades for sure.

But if you want to plan a civilization that will last centuries, look at the long droughts in the paleo record for California. Those make clear the need precautionary development of external water supplies .

Climate, in all examples. It depends.

• David B. Benson

John C // October 21, 2008 at 2:26 pm — Carl Wunsch has a wonderful, provockative paper (available from his website). He treats the Vostok ice core temperature record as an AR(2) process. He used the first half of the record to train the process, i.e., set the parameters. It then proceeds to do quite a good job of duplicating the second half.

His point, as I take it, is that the stade/interstade/interglacial oscillations of temperature are part of a nonlinear system; such are often able to entrain on a weak external forcing, in this case orbital forcing (which he didn’t need to put in his AR(2) process.

Note that CO2 isn’t even considered.

I certainly recommend the paper to all readers of Tamino’s Open Mind.

• Former Skeptic

Bob North:

The 2nd edition of WMO 100 that you cite is pretty archaic and the WMO is in the process of updating it. The more current definition of climate is this:

What is Climate?

Climate in a narrow sense is usually defined as the “average weather,” or more rigorously, as the statistical description in terms of the mean and variability of relevant quantities over a period of time ranging from months to thousands or millions of years. The classical period is 30 years, as defined by the World Meteorological Organization (WMO). These quantities are most often surface variables such as temperature, precipitation, and wind. Climate in a wider sense is the state, including a statistical description, of the climate system.

Further, the working draft of the 3rd edition of WMO pub. 100 also does specify the need for the 30 year period - As per their (WMO’s) instructions, I will not cite the exact quote as it is a working draft, but do read pg. 72:

http://www.wmo.ch/pages/prog/wcp/ccl/guide/documents/guide_third_edition_draft_may2007.pdf

• Bob North

Former skeptic - I won’t disagree that the 2nd edition is dated (1983 to exact), but is still the current edition. I was aware that there is a working draft but, like you, did not want to cite it since it has not been finalized and thus is not official. The discussion of “climate normals” (i.e., 30 year averages) on p. 72 is interesting, but climate normals (not “climate”) were defined as 30 year averages in the 2nd edition as well.

There were some other interesting points in the working draft about why the 30 year period was selected (also on p. 74) and concerning whether 30 years is the appropriate averaging period (2nd to last paragraph on p.74). Finally, the working draft gives its “definition” of climate at the bottom of p. 1. (Sorry I can’t link directly to these spots since the file is a pdf and I don’t want to copy and paste the working draft here).

My interpretation of these sections is that, while we should certainly think of climate trends as multi-decadal (which is what I said above), there is nothing magic about 30 years and that the definition of climate does not include a specific timeframe. Read the draft sections yourself and make your own judgement.

My only point in my first post was that many have repeatedly stated the definition of climate includes a specification of a 30 year period. From actually looking at the WMO’s definition, it does not. (Of course, one would have to accept the WMO as the arbiter for the “official” definition of climate)

• HankRoberts

Bob, you’re being silly. It’s a climate change blog. You’re reading discussions about global climate.

“… the statistical description in terms of the mean and variability of relevant quantities over a period of time …”

What’s relevant to global climate change statistics? That’s what people are talking about.

Here.

The meaning of words varies with context and changes depending on your time, location, and conversationalists. Who knew?

• Former Skeptic

Bob North:

…climate normals (not “climate”)…

Perhaps you see some significant difference in definition between “climate” and “climate normals”. I do not.

Climate normals are used to statistically describe the climate system in an area to allow for comparison with other stations (i.e. defined as a “benchmark” in the 3rd edition working draft of WMO 100). Arenn’t these standardized time series data a very important part of climate as well? Recall the definition of climate above as “…the state, including a statistical description, of the climate system.”

[As an aside, does anyone have access to WMO Pub. No. 49 (Technical Regulations)? Methinks there will be an explicit definition there that will answer all semantic doubts.]

Lastly, if you are insistent in claiming that climate has not been specified by the WMO as 30 years - how would you explain for the FAQ website link provided above? That’s official enough, IMO. :)

Leads me to, among many other links worth looking into, this list:
http://www.weather.gov/tg/wmodocs.html

Searching that for “49″ gives me

WMO Manual 49

TECHNICAL REGULATIONS

Volume I - General meteorological standards and recommended practices
… contains the regulations concerning the World
Weather Watch; climatology; …
ISBN: 92-63-15049-4

I hilight the ISBN. and toggle Book Burro
http://www.bookburro.org/
which tells me, hmmm
None of my local bookstores have it;
None of my preferred online bookstores have it.
None of my walking-distance libraries have it.
WorldCat starts with my ZIP code and searches outward and finds it available at:
- University of California, Davis, Shields Library
and at more distant libraries

Your libraries may vary. But you can find it by the ISBN and order it from Interlibrary Loan.

And go back to that first Google search, I didn’t read much of it and maybe there’s a fulltext source online.

• Richard Steckis

Dhogaza.

You really are silly. By different viewpoints I mean this. Tamino is a statistician and looks at the problem with the mind of a mathematician. I as a biologist look at the problem as a systems problem with. Biologists tend to be a little more tuned to the real world and its nuances rather than the more esoteric world of mathematics.

Both viewpoints are valuable and complementary.

• Richard Steckis

Oh. And Ray Ladbury. I do know what a trend is. I also know that in climate studies, 8 years of data does not necessarily mean a trend. But it can mean one is occurring and that it is wise to track it. It should not be summarily dismissed as “noise”. The eyes can deceive and so can statistics.

• Barton Paul Levenson

Richard Steckis writes:

Oh. And Ray Ladbury. I do know what a trend is. I also know that in climate studies, 8 years of data does not necessarily mean a trend. But it can mean one is occurring and that it is wise to track it. It should not be summarily dismissed as “noise”. The eyes can deceive and so can statistics.

So you’re depending neither on the visual impression nor on the statistics to say the recent cooling is a trend. How did you figure that out — with a crystal ball? A seance? Messages from extraterrestrials? I’m trying to think of a method that might use neither visual impression nor statistics to define a “trend.” Can’t seem to come up with one.

• John C

David Benson wrote :
Note that CO2 isn’t even considered.

Thanks for the tip on the paper - I’ll have a look. This opposes Hansesn view that CO2 is the major GHG contributor to the T increase when exiting an ice age (Kingnorth trial testimony). I just can’t get the maths to work. It just doesn’t jive with the commonly quoted number that a doubling of CO2 will lead to a 1 degree F rise (ignoring other feedbacks). And when exiting an ice age, the CO2 is a late feedback itself and only goes up 100ppm … so its contribution must be negligible.

• Richard Steckis

Barton Paul Levenson:

“So you’re depending neither on the visual impression nor on the statistics to say the recent cooling is a trend. How did you figure that out — with a crystal ball? A seance? Messages from extraterrestrials?”

What are you on about? Just because I said that both eyes and statistics can deceive does not mean that you don’t use them. It just means that one must be CAREFUL and circumspect when using them.

Do I have to explain the obvious because you can’t interpret language?

• Lazar

John C

This opposes Hansesn

It doesn’t.

First hit. Not everything that one reads on CA is true.

From David B. Benson’s description of the paper and what others have written, positive CO2 feedback is part of that “nonlinear system; such are often able to entrain on a weak external forcing”.

Hansen was making a point about exceptionally high CO2 levels effecting the potential for future glaciation, not the involvement of CO2 in previous terminations.

But shouldn’t Earth now, or at some point, be headed into the next ice age? No. Another ice age will not occur, unless humans go extinct. Orbital conditions now are, indeed, conducive (albeit weakly) to initiation of ice sheet growth in the Northern Hemisphere But only a small amount of human-made GHGs are needed to overwhelm any natural tendency toward cooling. The long lifetime of human-made CO2 perturbations assures that no human generation that we can imagine will need to be concerned about global cooling. Even after fossil fuel use ceases and its effect is drained from the system an ice age could be averted by chlorofluorocarbons (CFCs) produced in a single CFC factory. It is a trivial task for humanity to avert an ice age.

• dhogaza

You really are silly. By different viewpoints I mean this. Tamino is a statistician and looks at the problem with the mind of a mathematician. I as a biologist look at the problem as a systems problem with. Biologists tend to be a little more tuned to the real world and its nuances rather than the more esoteric world of mathematics.

Both viewpoints are valuable and complementary.

Not really. Replace “biologist” with “physicist” and voilá! You’ve described the real world of climate science, where physical scientists, mathematicians specializing in modeling, and statisticians work together to tell us WTF is going on regarding trends, physical forcings, etc.

Now, biologists play a role, too, in particular what AIBS now likes to call “whole organism biologists”. They’re contributing a lot of data regarding the in-the-ecosystem signal of global warming.

• HankRoberts

Richard S, are any of those papers Scholar found yours?

• David B. Benson

John C // October 22, 2008 at 2:25 pm — I should of said something like ‘the role CO2 is not explicitly considered’. The AR(2) process, representing the Vostok ice core proxy for the Southern Ocean temperature, was tained on half the data; this means that whatever influence CO2, methane, and even the external orbital forcing all ended up influencing the trained parameters. No contradiction to anything Hansen has written or said.

It turns out that Carl Wunsch, an oceanographer who writes extremely lucidly, used (at least for the final run in the paper), a time step of 400 years. More recent work suggests that this is a good compromise dwell time for water sent deep elsewhere and upwelling in the Southern Ocean. A better one would be 250 years, but then one would need an AR(3) process to capture the longer dwell time for Pacific waters.

Wunsch’s AR(2) process, it now seems to me, indicates the great importance of deep ocean mixing upon the millennial scale aspects of the pre-anthropogenic climate.

Now it might be often quoted that 2xCO2, absent other feedbakcs, warms in planet by 5/9 K, but you ought to find a more authoritative statement for this; looks to small to me.

[Response: Although I disagree with Wunsch's hypothesis about glacial cycles, I wholeheartedly agree that he is one of the best at lucid writing.]

• David B. Benson

Tamino — I believe you are refering to his hypothesis that orbital forcing plays no role in glacial cycles. He is essentially alone in this; I take his paper as indicating,
as I stated, the great importance of deep ocean mixing.

I admit to being puzzeled about how an AR(2) process, with a time step of 400 years, so a memeory of only the last 800 years is able to adequately resemble the intergalcial –> stade –> interstade –> stade –> interstade –> deep stade –> interglacial even once, mush less the twice his paper demonstrates; that’s about 200,000+ years so at least 500 time steps!

[Response: I don't know the whole scoop, but I think Wunsch's hypothesis involves more than just AR(2).]

• David B. Benson

Carl Wunsch’s web site:

http://puddle.mit.edu/~cwunsch/

The paper,

Wunsch, C., 2004, Quantitative estimate of the Milankovitch-Forced Contribution to Climate Change (pdf)

is linked about half way down the page and is named milankovitchqsr2004.pdf

I just reread the relevant sections and I see nothing but an AR(2) process trained on the first half of the data.

• David B. Benson

It seems that later Carl Wunsch came to understand the situation quite a bit better, further down his page find

E. Tziperman, M. E. Raymo, P. Huybers, C. Wunsch, 2006. Consequences of pacing the Pleistocene 100 kyr ice ages by non linear phase locking to Milankovitch forcing (pdf), Paleoceanography

the file being named tzipermanetal2006.pdf

I also recommend this paper to all readers here.

• Dave A

I’d like to ask a question of Hank, Lazar, GP, David Benson and dhogaza.

It is, are you all climate scientists?

If, as I suspect, you are not why do you always weigh in heavily against other scientists, for example Richard Steckis but also many others in numerous posts on this blog, who admit they are not climate scientists but put forward perfectly valid points of view?

Further, are you all saying that no one but a climate scientist can articulate a point of view about a scientific matter that may affect them considerably? Have you not yet joined the modern world ?

• HankRoberts

• David B. Benson

Dave A // October 23, 2008 at 10:27 pm — I am an amateur student of climatology, after being an amateur geologist since 1960. I am a retired scientist whose specialty is rather remote from most of climatology. However, like most scientists, I am a quick studier and usually can easily tell the difference between something which (might be/is) solid and something which is wrong/confused/ill-informed.

When somebody comments that they ‘believe’, they need to have first established that their point of view is worthy of attention. Climatology is quite challenging; the hardest part for me has been the atmospheric physics; even learning quantum mechanics (in graduate school, just through Klein-Gordon) was easier.

Unfortunately, I have no choice but to exist in the modern world; I vastly preferred the previous century. :-)

Dave A., Of course on needn’t be a climate scientist to articulate a meaningful point of view on climate. All one need have done is publish a peer-reviewed article that advances understanding of climate sufficiently to be cited by several subsequent papers in climate science.

Just curious: Do you ask your plumber for financial advice? How about if you have chest pains–do you call your electrician? Do you really attach zero value to intensive study and expertise?

• Gavin's Pussycat

Dave A, I hold a doctorate in geophysics and published a few peer reviewed papers on post-glacial rebound and GPS sounding of the atmosphere. But then, on a blog like this D. Duck could claim the same with as little proof. I suggest you try to judge my statements on the science in this light. I’m not making things up.

• Gavin's Pussycat

David, impressive paper, that. Wunsch states in the caption of Figure 5 that “Fit was only to the first half of the record shown.” Presumably meaning that a1, a2 and \theta(t) (?) were estimated by fitting to t = -450..-225 ka, and then the red dotted curve was generated from them for t = -225..0 ka.

I just cannot believe this. If this were really true, the agreement on the right half of the picture would be pure magic. I could imagine the a1, a2 fit to be to the first half, but \theta(t) must come from the whole record. There must be a misunderstanding here. Why not ask him?

Note also that if you read carefully, he doesn’t actually rule out astronomical (or any other) forcing, but points out that it is overlaid by a huge random-looking variability. We know of course that during ice ages, small variations are amplified mainly through albedo feedback. So we might here be looking at amplified natural variability. As Figure 9 shows, even a small systematic effect, while drowning in the noise on the PDF plot, may lead to very visible effects on the time domain plot. Fig.10 explains how long a memory this particular AR(2) process has: it seems to actually be a rather special one: a1 + a2 = 0.9826, very close to 1, so if it starts out constant, it will take a few times 50 times delta-t, i.e., a few times 20 ka, to decay.

• Gavin's Pussycat

David, yes that 2006 paper ties all the loose ends together… I think we can say that there is no real mystery left ;-)
I do wonder though about the physical mechanism proposed. What is the role of sea level change and of isostatic adjustment to ice load, if any?

• dhogaza

If, as I suspect, you are not why do you always weigh in heavily against other scientists, for example Richard Steckis

I don’t deny Steckis his right to make his opinion known. What I do deny is his argument of personal authority simply because he is a biologist and because biology is a science.

I’d also ignore Tamino if he claimed to understand climatology because he’s a statistician.

But Tamino does no such thing. He argues about time series analysis, which is his speciality. When he talks about denialist or climate science claims about trends, reliability of data, etc he’s simply blogging about the same kind of stuff he’s paid for in his professional life.

• David B. Benson

Gavin’s Pussycat — “Presumably meaning that a1, a2 and \theta(t) (?) were estimated by fitting to t = -450..-225 ka, and then the red dotted curve was generated from them for t = -225..0 ka.” That’s my reading of what he did.

“… a few times 20 ka, to decay.” What is ‘a few’? 5? 10? If 10 then I guess now I don’t see any miracle.

Re: 2006 paper. The ‘physical mechanism’ is clearly just a simplified model to demonstrate non-linear entrainment on the externally imposed signal (orbital forcing).

Anyway, I suspect that Carl Wunsch no longer doubts the role of Milankovitch cycles on the climate.

• Gavin's Pussycat

> “… a few times 20 ka, to decay.” What is ‘a few’? 5? 10?

I meant that every 20 ka corresponds to decay by a factor of e. Very roughly of course.
Fig. 10 actually shows the decay behaviour.
> Anyway, I suspect that Carl Wunsch no longer doubts the role of Milankovitch cycles on the climate.
I’m quite sure of that… or more precisely, the role of Milankovitch forcing.

• Dave A

Ray,

I often ask my plumber for financial advice - until recently he was a senior investment banker :-)

And I do not attach zero value to intensive study and expertise. I do note, however, that there are many well qualified scientists, perhaps not in climatology, who comment here and elsewhere who are dumped upon if they do not support the ‘consensus’ .

I will turn the question back to you. As a well educated person do you meekly accept the word/ peer reviewed papers of all the many highly qualified experts in your own field or related areas, or do you have your own opinion about them ?

• David B. Benson

Gavin’s Pussycat // October 24, 2008 at 7:51 pm — Thanks for the reminder of what to look for in the 2004 paper. Theta, said to be the innovation, does indeed appear to be taken (somehow) from the entire Vostok record. While statistically ‘almost white noise’, it is no longer strictly random, but a random-appearing additive factor in the AR(2) process. So as I know see it, obtaining such a good fit for 550 time steps isn’t actually very impressive at all. Tamino was right when he wrote that there is something here other than just AR(2).

• Dave A

GP,

If you had claimed to be D Duck then I would have had serious reasons to doubt you:-)

Dave A., Don’t be daft. Of course I form my own opinion of research in my field. I read the paper. I do the math. I look for consistency and how it fits into the context of previous research. Likewise, I have done the same to the extent of my ability with climate research–as have many of the commenters here.
Now here is the thing, when I ask scientists who dissent from the consensus view what their objection is, I get no cogent evidence or argument. I get nothing that increases my understanding of climate. All I get are very weak arguments (this year is cooler than last) and appeals to ignorance (we don’t know anything…). And I certainly get no cogent arguments on why the known physics of greenhouse gasses should magically change at 280 ppmv.
What scientists look for is theory that increases my understanding of broad ranges of phenomena. Climate science does this. The denialists offer nothing in terms of understanding.

• Ian Forrester

Dave A said “I often ask my plumber for financial advice - until recently he was a senior investment banker” Now we know why the world financial sector is in such a disgraceful mess. I hope you are not hoping that the world’s climate will also go into such a mess.

Dave A said “I do note, however, that there are many well qualified scientists, perhaps not in climatology, who comment here and elsewhere who are dumped upon if they do not support the ‘consensus’” No, they are not dumped upon because they do not support the”consensus”. They get dumped upon when they deny the science, obfuscate and generally act in an unscientific manner.

• And there aren’t many. Show of hands please if you’re among those so put upon unfairly?

• Ian Forrester

Hank, you are correct. In fact, on many blogs it is quite the other way round and many knowledgeable scientists (not necessarily climate scientists) are “dumped upon” by the ignorant and vile mouthed deniers. I have stopped participating on a well known blog since the administrators allow these trolls complete and unmoderated access and are allowed to slander and smear both well known climate scientists and knowledgeable contributors.

Decent people can have lively debate but, unfortunately, many of the deniers can hardly be called “decent”.

• Philippe Chantreau

“The denialists offer nothing in terms of understanding.”

Ray you’ve summed it all in a few words; there is nothing to add to that, really.

• Gavin's Pussycat

> So as I know see it, obtaining such a good fit for 550 time
> steps isn’t actually very impressive at all.
Precisely. And note that Wunsch in the later (Tziperman) paper successfully performs the same AR(2) fit to model generated data… that settles it IMHO (and clearly Wunsch’s).

• Ray had another important point about how professionals evaluated ideas:”I look for consistency and how it fits into the context of previous research.”

Now the hooty thing about denialists is that they believe three impossible and mutually contradictory things before breakfast. Look at the nonsense that they simultaneously accept. They don’t care as long as it casts doubt on our understanding of climate. For example the Miscolczi paper and those by Chillingar and friends and so on.

• Gavin's Pussycat

Eli, yes, so true.

Many non-scientists have no clear idea of how highly rated redundancy is in science. You don’t really believe anything seriously before it has come from several independent sources. And those sources themselves are often internally redundant, like surface temperatures, monthly averaged, correlate over long distances.

Same with replication: I can replicate with the best of them someone else’s coding errors by running their code. Independent replication, by different people, using their own code and methods, on different data (if you can get it) proves something.

• A lot of people who are non-specialists know a little bit about what they are talking about. I’m a student and I managed to scrap together a halfway decent blog which I hope teaches a few laymen-level people a few things. I don’t claim to think that I am going to “advance” the science at my level of understanding, or win a nobel…it’s just silly. Lots of bloggers apparently think that their youtube enligtenment on how CO2 lags temperature makes them experts on climate change. The arrogance! I have, however, expressed disagreement with people in the field such as Roger Pielke (pick one…) simply because I’ve read a lot of things to the contrary. Not because I know more than he does or because I don’t like him…just an informed judgment. If Dr. Bob, Jesus, Manabe, Gavin, or the Ghost of Christmas Past told me that orbital variations played no role in glaciations I would still be skeptical of them…not because I think I know more than they, just because it goes against a lot of other research. Ray says good when he says “I look for consistency and how it fits into the context of previous research.”

What I don’t do is take a concept such as “water vapor feedback” which has grounding in primary articles for many decades, is a mainstream view, and then go on my blog and say it’s all wrong or try to twise it to fit some crazy agenda. It would be silly and ruin any credibility that I really don’t have. When people like John Christy get in front of a camera and say things like “we can’t say CO2 causes temperature change: it certainly never did in the past” then as a laymen, why would I ever take his word over another experts again. The ‘boy who cried wolf’ story applies. Scientists have a right to be wrong or corrected, but not misrepresent.

What I’d rather see from non-specialists is “hey look, we got two opposing views, say Kerry Emanuel and Landsea on the future of global warming on hurricanes: Now I’m going to e-mail a few people, ask some questions, read some of their stuff, and make informed judgments on what may be more likely.” Denialists of climate change do NOT do these things…they simply say that all the scientists are wrong and that every other paper on urban heat islands, 1934 adjustments, “bucket corrections,” aerosols, and anything else debunks AGW. Quite silly actually.

• David B. Benson

Gavin’s Pussycat // October 25, 2008 at 7:48 am — Thanks for reeading the two papers and your questions and comments. I understand the two papers much better for this exchange.

• Dave A

Chris Colose,

“they simply say that all the scientists are wrong”

This is rubbish. Most sceptics do not say that at all, what they say is that climate scientists cannot be certain that they are right given the complexities of the Earth’s climate system and the fact that the models are, to put it simply, not very good ( maybe better than they were but still not very good).

You take Christy to task for “crying wolf” - do you do the same with Hansen?

• Dave A

GP

>You don’t really believe anything seriously before it has come from several independent sources

So how many ‘independent sources’ has the hockey stick come from?

• David B. Benson

Dave A // October 25, 2008 at 10:32 pm — Three that I know of. I gather there are over a dozen such studies.

• Dave A,

I haven’t seen Hansen do anything that would make me think that. In the actual scientific community, he’s regarded as one of the most prominent scientists in this field of our time. I only see people trying to discredit him through personal attacks and complaining (McIntyre?)

As for your post, you’re just confusing real skepticism and denialism. Take a glance at some of these papers (just don’t take em seriously) and tell me where the authors are trying to advance the science through honesty and sound analysis of the data and physics

Don’t kid yourself too much with the goody version of “objective balance.” It only sounds nice to the kiddies.

• dhogaza

This is rubbish. Most sceptics do not say that at all, what they say is that climate scientists cannot be certain that they are right given the complexities of the Earth’s climate system and the fact that the models are, to put it simply, not very good ( maybe better than they were but still not very good).

Yet, despite all this uncertainty, they repeatedly tell us that there’s nothing to worry about.

Umm … an honest acceptance of uncertainty would tell us the opposite.

• Gavin's Pussycat

You’re welcome David. Interacting with someone who actually wants to figure out something rather than intentionally remain ignorant (hint, hint) is refreshing and rewarding. That’s what this blog is for.

• P. Lewis

Dave A says:

…Most sceptics do not say that at all, what they say is that climate scientists cannot be certain that they are right given the complexities of the Earth’s climate system and the fact that the models are, to put it simply, not very good ( maybe better than they were but still not very good).

I’ve seen this aspect/meme (the plea to a sort of irreducible complexity argument) promulgated in the climate denial-o-sphere quite a few times. When I see/hear it I think Michael Behe, which rather says it all for me.

• Gavin's Pussycat

> So how many independent sources’ has the hockey stick come from?

Methodologically independent? Well, tree rings and boreholes come immediately to mind, as does glacier retreat. That’s three as David said.

They don’t all go back equally far — tree rings go back the farthest and have good dating.

Note that if you go further back in time, the number, global distribution and variety of available proxies diminishes, and eventually you lose redundancy. This “data horizon” was around 1400 when MBH98 was written; today it is more like 700-900.

Note also that lack of redundancy doesn’t prove that something is wrong; just that you wouldn’t be comfortable making hard statements on it.

• Barton Paul Levenson

Eli writes:

Now the hooty thing about denialists is that they believe three impossible and mutually contradictory things before breakfast. Look at the nonsense that they simultaneously accept. They don’t care as long as it casts doubt on our understanding of climate. For example the Miscolczi paper and those by Chillingar and friends and so on.

I can testify from personal experience that that’s true. I’m in a long running dialogue over on landshape.org where people are defending Miskolczi’s crackpot paper. One of them insists that the surface temperature of the Earth being higher than the radiative equilibrium temperature is caused by “stored energy,” not by the greenhouse effect. I pointed out that this would mean he’s disagreeing with Miskolczi, who at least believes there’s a greenhouse effect. Nothing doing. He says he didn’t see any conflict, and neither Miskolczi, who is also present on the thread, nor any of his supporters, was willing to correct this guy. “No enemies on the left.”

• Barton Paul Levenson

Dave A writes:

So how many ‘independent sources’ has the hockey stick come from?

At least 14, according to the National Academy of Sciences report.

Dhogaza raises an important point, which I would like to amplify:

Once a threat is established as credible–and by any reasonable standard, climate change meets this metric–uncertainty is no longer the friend of apologists for inaction. This is because you have to consider uncertainties on the high side as well as on the low side, and the high-side uncertainties result in much higher risk estimates. This is especially true for climate change, where the high-side uncertainty is much greater than the low side uncertainty. As such, Hansen’s warnings are still consistent with accepted confidence intervals, while the apologists for complacency aren’t even on the map. The key to respectability in science: Remain consistent with the evidence.

• Dave A

Ray,

The key to respectability in science: Remain consistent with the evidence.

So where is the actual empirical evidence that man made CO2 is doing what you say it is?

Why does global warming science have to rely on models that are not actually very good at replicating the climate system they are trying to model?

Why have numerous problems been reported with surface temperature measurements yet these are never acknowledged by GISS & Hadcrut?

Why do so many climate scientists feel the need to develop unique statistical methods to justify their work, often with little basis in the statistical mainstream?

• Dave A,

you are very confused. It comes from reading too many wingnut sites.

The radiative physics behind CO2 increase has been known (albeit not as much so) for over a century. Arrhenius’ 1896 paper didn’t have models or CO2/temperature data, etc. The physics speaks for itself. The “empirical” evidence is the many greenhouse fingerprints such as stratospheric cooling, changed LW fluxes, minimum temperatures warming faster than maximum, etc and other things which are inconsistent with other “natural mechanisms” We have satellites measuring changes in the sun, we see no changes in cosmic rays, we’re lookng at volcanism, we see heat going in the ocean and not out…so what else do you think is causing the warming, and also how is it cancelling out CO2 physics?

As far as these models that don’t do anything good, does this graph mean nothing? Or can models only not be credible if they are incapable of reproducing the modern warming trend w/o GHG’s? They predicted polar amplification, stratospheric cooling, increases in specific humidity, they showed good quantiative responses to the Pinatubo eruption. It seems models are incapable of ever being “relaible.”

• Sekerob

Start with doing some reading on AGGI e.g. The Sun fluctuates about 1.2 Watt Meter square in energy output between the maximum in a solar cycle and the minimum (TSI). The median thus about 0.6. The GHGs on their own, without feedback have added since 1990 alone about 0.5 Watts Meter Square, that’s nearly matching the cyclic mean variation. If the solar flux is enough to cause all the ups and downs in climate, then 0.5 Watt Meter Square is pretty darn significant or?

As for your questions: Occupational Therapy. All been answered… look them up.

• dhogaza

The radiative physics behind CO2 increase has been known (albeit not as much so) for over a century…

Look, DaveA has been told this stuff innumerable times in any number of places, and he comes back with the same.

If God lighted a bush and called to him, then struck him on the head with a stone tablet proclaiming that Physics 101 actually has some relevance to the Real World, DaveA would say - “I don’t believe you because of my political beliefs”.

Just give up with serious discourse. Poke fun at him, ridicule him, etc. Much easier and just as ineffective as treating him seriously.

• Barton Paul Levenson

Dave A writes:

So where is the actual empirical evidence that man made CO2 is doing what you say it is?

1. Laboratory work by John Tyndall showed that carbon dioxide was a greenhouse gas back in 1859.

2. Roger Keeling and others showed that carbon dioxide was steadily increasing in 1958, and every year since then.

3. The radioisotope signature of the new carbon dioxide shows that it comes from the combustion of fossil fuels. This was first pointed out by Hans Suess in 1955.

4. Warming is shown not just in land surface temperature readings, but in sea surface temperature readings (are there urban heat islands on the ocean?), balloon radiosonde readings, satellite observations, borehole temperatures, melting glaciers, movement of tree lines toward the poles, migration of insect and bird populations, earlier dates of blossoming of flowers, etc., etc., etc.

Put ‘em all together, they spell Anthropogenic Global Warming. Which of the points above do you dispute?

• Sekerob

A chart for fun, a spin-off from the one that showed AGGI on it’s own, but I’ve added TSI anomaly, from a random soource for perspective and Land temps of the Northern Hemisphere, BECAUSE, about 90% of the post Homo Erectus species live on this part of the planet.

http://i137.photobucket.com/albums/q210/Sekerob/AGGI2007-1.png

No, I did this as said for fun. Not going to discuss with the Climate Audit or WUWT types, sorry.

• Dave A

chriscolose,

Pretty graph. Try this

you spend a lot of time working on something, and you are really trying to do the best job you can of simulating what happens in the real world. It is easy to get caught up in it: you start to believe that what happens in your model must be what happens in the real world. And often that is not true…The danger is that you begin to lose some objectivity on the response of the model and begin to believe that the model really works like the real world…

• Dave A

dhogaza,

There you go again. Now you apparently ‘know’ my political beliefs.

I could possibly intuit your political beliefs based on your posts, but wouldn’t because life is much more complicated than that and surely a person could not just have a view of climate science based on their political view ? Well perhaps some might but I’m not one of them.

Besides climate science has sufficient contradictions within itself to cause one to question it.

• HankRoberts

Dave, you already told us what your political beliefs are and why you’re posting here.
You know how to find your own opinion, eh?

Dave A.. You will notice that that is a climate modeler who admonishes people not to confuse the model with reality. Here’s a hint, Dave. They already know this stuff. Like so many “sociology/anthropology of science” studies, this one misses the mark.

• Dave,

any serious discussion on climate models should involve their limitations, strengths, weaknesses, where we improved, where the yare still wrong, etc. Please cite a specific model for instance. There is no evidence that the current generation of models are so far off in their handling of the global mean temperature change as a response to external forcings, or that any flaws in models can remove the general conclusion that CO2 is a greenhouse gas and has been rising and efecting temperatures.

• Dave A

Ray,

Yes its a climate modeller responding honestly.
I agree they know this stuff already.

But why then do the university/institution press releases always stress the efficacy of the models and why when the MSM then report the results as ‘facts’ is there no corrective applied?

• Dave A

Hank Roberts,

Apologies, please tell me what my political beliefs are - in your own words of course.

• David B. Benson

Dave A // October 28, 2008 at 10:16 pm — The university/institution press releases are written by strictly speaking, PR types. MSM stuff is written by reporters.

There are some very good people writing about science for the public; Carl Zimmer comes to mind. But there are only a few who also know something about climatology; read carefully “The Discovery of Global Warming” by Spencer Weart:

http://www.aip.org/history/climate/index.html

Review of above:

http://query.nytimes.com/gst/fullpage.html?res=9F04E7DF153DF936A35753C1A9659C8B63

regarding the development of computerized climate models.

Also, Gavin Schmidt’s publications page has two papers regarding GISS ModelE; on one paper he is lead co-author, the other is a solo authorship paper. Both help to understand how the models are constructed.

But one thing is certain. No GCM offers short range predictions; many different runs are combined to form a statistical envelope. Oberservations within that envelope tend to confirm the models validity (statisticians have a fancier term for what I am attempting to express).

• HankRoberts

Nope, I trust your own statements here.

Dave A., Now, where exactly is there ever a “corrective” to what the media report–mainstream or otherwise? Do they correct them when they report lattice guage theory predictions of quark-gluon plasma as reality? No. How about models of Earth’ geodynamo? Nope.
The models do a very good job, actually at reproducing trends and elucidating the physics–which is their purpose. That is why they are the consensus choice–no model with low sensitivity can add understanding of the real world, so such models are relegated to the status of toy. Certainly modelers continue to tinker, as any model with low sensitivity that reproduced the physics and trends would be a very interesting beast, both scientifically and politically. Alas, they just aren’t realistic.

• Tamino, how hard is it to add a third component to the two you’ve charted?

[Response: It's pretty straightforward.]

• Bob North

Tamino - Back to the original topic of this post, I have several questions.

First, for a transient event such as volcanic eruption, the short term response is obvious. But is there good evidence for the longer term “slow” response being about 30 years or could the “effects” of the volcano wear off quicker.

[Response: The slow response is based on both model simulations and basic physics; the thermal inertia of the oceans is vast, and it takes longer for forcing to bring about equilibrium. The actual estimated time scale (30 yr) is based on computer model simulations.]

Second, is the point of the post that during any volcanic lull, there should be a warming trend since you are having less and less a negative forcing (I think this is the point) or is it that the volcanic lull allowed the early GHG signal to start showing up?

[Response: The former; the lack of intermittent negative forcing raises the long-term average forcing and leads to warming (with or without GHG changes).]

Finally, on the graph showing GISS and FIT, what exactly constitutes the FIT? Is it the combination of the short and long volcanic forcings along with the GHG forcing?

[Response: The FIT is the multiple regression best-fit for both short-term and long-term net forcing. The net forcing includes volcanic and GHG, as well as other forcings (sulfates, ozone, etc.) as estimated by GISS for the 20th century.]

Bob

• John C

[Response: My best guess: about 1.5 - 2 deg.C. I suggest you ask at RealClimate, they know a lot more than I do.]

I asked around …. the official answer is that a doubling of CO2 leads to a 1.2 degrees C rise in T (without any additional feedbacks).

http://www.grida.no/publications/other/ipcc_tar/?src=/climate/ipcc_tar/wg1/044.htm

So for the change from an ice age to an interglacial period (180ppm to 280ppm) the rise due to CO2 is about 0.75 degrees C (out of the total 5 degrees C).

You’re describing the hypothetical — doubling the number of carbon dioxide molecules in the atmosphere instantly, with no other change.

It’s false to suggest that “the rise due to [a change in] CO2″ [or any other forcing] over time [such as from an ice age to an interglacial] is only that hypothetical number. On this living planet, with this atmosphere, climate factors are connected.

In the past, in the absence of human activity, between ice age and interglacial, CO2 increased after warming, and warming increased further due to that rise in CO2.

You know this, of course; but you could mislead some innocent reader by failing to be clear.

Pull or push on one and others change as well.

• David B. Benson

The opposite of a vulcanic lull, the super-eruption of Mt. Toba.

This event has been dated, by two different well-established radiological techniques to either around 74,000 years ago or else aboout 71,000 years ago; the error bars do not overlap.

The geologists do not seem the slightest upset by this, but the 3,000 years discrepancy is actually of some concern to those studying the peopling of the planet. This super-eruption caused a population bottleneck but it is not clear (and may never be) whether or not modern humans reached Southeast Asia, possibly even South China, before the super-eruption or only afterwards. Knowing the date of the super-eruption might help settle the matter.

What does interest the vucanologist is estimating the great quantity of sulfates released:

“In addition it has been calculated that 10^(10) metric tons of sulphuric acid was ejected into the atmosphere by the event, causing acid rain fallout.”

from

http://en.wikipedia.org/wiki/Lake_Toba

According to the Scientific American article about the event, about 3 summers ago, the global temperatures plumented 4–8 K for 3–6 years. Despite this and the tremendous ashfall in South Asia, some humans (not necessarily modern ones) susvived in even that location.

Regarding the dating, NGISP ice core shows a tremendous, abrupt temperature drop right around 74,000 years ago, with a cold regime lasting for 1000+ years. While it is tempting to blaim Mt Toba for this, note that neither Vostok (centennial resolution) nor the LR04 stack (millennial resolution) give any supporting evidence whatsoever; there is no remarkable change in either dating between 80 and 67 kypb.

This suggests that the globe as a whole experienced a severe but short cold spell after the super-eruption but quickly returned to what was normal for the time; except possibly in Greenland.

• HankRoberts

Mt. Toba

Hmmm, anyone know how well the hemispherical air circulation is separated? I recall seeing suggestions it takes several years before particulates released in the Northern hemisphere average out across the Equator, but that was very old (nuclear war scenario) stuff, before electrons were invented, back when I was a child.

If there’s really a significant separation of the air circulations — whileIndonesia is on the equator, favoring mixing into both hemispheres, Mt. Toba was an explosion, not a longterm event, if I recall. Anything to suggest that particular huge debris cloud could have gone mostly north but not much south?

• David B. Benson

This well-done book chapter

http://www.andaman.org/BOOK/originals/Weber-Toba/ch4_climate/textr4.htm

shows that the majority of the evidence points to a 73,000 year ago data for the super-eruption; this is within the error bar for the older date.

Assuming an older date, the one site in South China I know about with defnitely modern human remains is sufficiently later (70 +- 1) kybp that modern human migration to that region could well have followed the Mt Toba event, but need not have done so. Taking Alan Templeton’s genetic flow work into account, it seems more likely to have been earlier; this is a subjective Bayesian assessment of the data I know about so far.

• David B. Benson

HankRoberts // October 29, 2008 at 10:19 pm — Yes, see George Weber’s chapter in my just prior post. Essentially all of the ash is in the northern hemisphere. This tells us that the event occurred during the summer monsoon, when the winds would indeed blow the ash towards and into South Asia (about 6–15 cm deep over all of India, up to 6 m in some places, up to 9 m on the Maylasian Penninsula).

• David B. Benson

Whilel George Webster’s chapter 5 is of interest, I’ll close this bit on the Mt Toba anti-volcanic-lull by referring interested reades to

which includes Stephen Oppenheimer’s dating on 74,000 years ago (good agreement with the start of the Wurm stade in the NGRIP ice core temperatures), but more, his careful working out of genetic data. That is from his book, “Out of Eden”, where the matter of the Mt Toba super-eruption’s effect of dividing human genetics in two is more fully treated.

I’ll mention that along the way, I found papers showing that Mt Toba ash fell into the South China Sea and also in considerable depth on Java, where the event seems to have extirpated the Homo erectus (Jave Man) population there.

• John C

Hank Roberts wrote :
In the past, in the absence of human activity, between ice age and interglacial, CO2 increased after warming, and warming increased further due to that rise in CO2.

Indeed, this is the point I was trying to make. The CO2 itself is a feedback. As is the albedo and the water vapour increase. Its all interconnected. What it looks like is there is a warming kick-off due to a beneficial (for warming) change in irradiance due to the orbital cycles and then all these feedbacks occur with the net effect of about 5 degree rise in T. But how much is due to Co2 and how much to other the other (also self-perpetuating) feedbacks. I guess the way to work that out is to model it with no CO2 change (ie. stays at 180ppm), and see how much warming occurs out of a glacial period. Has this been done ?? Or is the answer yes - its the 3 degrees C for doubling effect ?

• Sekerob

“warming increased further due to that rise in CO2.” MEANING it started acting as a forcing as soon as enough was released! So is my reading of it all.

This past century plus it’s a different ballgame. First CO2, then the temps in tow and has been from the get go a forcing. And now the headache of CH4 coming along with a sudden hike, with all the [false] global cooling fanfare.

Anyway, looks like the answer is in the question: What do you think whether it was modelled or not? The percent 95% plus confidence level I think applies here. :D

• Hank Roberts

John, one of the online models might help:
http://forecast.uchicago.edu/models.html

• Hugh

Re Toba: Here are the conclusions from a 2003 paper on the subject of the bottleneck (that probably informed the SciAm article):

1. Mounting geological evidence strongly suggests the eruption was significantly larger than previously estimated, and caused a millennium of the coldest temperatures of the Upper Pleistocene.

2. Numerous genetic studies suggest that the population bottleneck was real rather than “putative”, and that it occurred during the first half of the last glacial period. Estimating the duration of this bottleneck is complex, but it was unlikely to have been shorter than 20 generations, and may have been longer than 500. Mass extinctions were not a feature of this event, nor of other explosive volcanic eruptions of comparable magnitude throughout Cenozoic history.

3. Capacities for modern human behavior were undoubtedly present during the last interglacial, but the stable environments of this period did not foster widespread adoption of the strategic cooperative skills necessary for survival in the last glacial era. Modern humans may have eventually developed such strategies during the last ice age, but they were crucial for survival when volcanic winter arrived. We are the descendants of the few small groups of tropical Africans who united in the face of adversity.

AMBROSE, S. H. (2003) Did the super-eruption of Toba cause a human population bottleneck? Reply to Gathorne-Hardy and Harcourt-Smith. Journal of Human Evolution, 45, 3 231-237.

• David B. Benson

Hugh // November 3, 2008 at 5:55 pm — Thanks, but the Scientific American article was solely about the volcanology. Furthermore, I prefer Stephen Openheimer’s analysis to that of Ambrose; early departure from Africa, early enough to be in Southeast Asis and possibly southern China at the time of the super-eruption. This interpretation is supported by Alan Templeton’s more recent work on human genetics; there is a thread about it on Carl Zimmer’s blog, The Loom.

• John C

John, one of the online models might help:

Interesting website. I ran one of the models on orbital forcing …. but I think three is an error in it. I think it misses out the impact of precession completely. If I’m not mistaken, precession means that in mid June now the northern hemispshere faces the sun, but in 10,000 years in mid June northern hemisphere will face away fron the sun in mid June (a 20,000 year cycle). So why, when you run a time series over several hundred thousand years does this not show up on day 180.

[Response: Keep in mind that "precession" as defined in climate science isn't the same as precession as defined in astronomy. For climate purposes, it's the *difference* between the direction of the vernal equinox, and the direction of the perihelion of earth's orbit. The astronomical precession cycle is about 26,000 years, while the climatological precession cycle varies from about 19,000 to 23,000.]