C02 and temperature proxies

[trbixler]

Lagging indicator.

"CO2, Temperatures, and Ice Ages"



wattsupwiththat.com/2009/01/30/co2-temperatures-and-ice-ages/

[trbixler]

Micro view.

"Temperature changes come before changes in carbon dioxide levels. While annual average values of CO2 show a simple upward curve, in reality carbon dioxide levels are fluctuating up and down. Until recently many climate researchers have ignored this trend. You might ask why – because it’s pretty clear when you look closely that temperature influences carbon dioxide levels and not the other way around (soil moisture also influences CO2 levels). Here is a graph showing this:"




reasonabledoubtclimate.wordpress.com/2011/08/09/carbon-dioxide-lags-temperature/

[steve]

Unless you can see in detail all the things that are going on it is impossible to show a direct cause and effect relationship.

However, we do know that CO2 is a greenhouse gas, and we can calculate the impact caused by increasing its levels in a given atmospheric composition. So we know it has the *potential* to warm the planet.

We know that the planet has a pretty sensitive climate. We know this because we know the climate has varied a lot in the past, and particularly that the correlation between temperature and the Milankovitch cycles (which on the face of it are causes of very minor forcing) is pretty good.

We also know that there is a good correlation between CO2 and temperature, not only for the ice ages, but also going back for hundreds of millions of years. And it is hard to see how long timescale climate variations could really be managed by mysterious ocean cycles/etc. that people like to invent to explain recent warming.

We do however know that there are processes in which CO2 levels will rise due to temperature rises, so certainly the link is not one-way.

On the shorter ice age cycles it seems that rising temperatures can cause CO2 to rise which accentuates the temperature rise, along with feedbacks like less ice cover. Eventually, the increase in outgoing longwave radiation is higher than the increase in CO2/ice/water vapour feedbacks and things are stable till something triggers a cooling at which point things go the other way.

On the timescales of 10s of million years, warmer temperatures caused by CO2 speed up the (still very slow) removal of CO2 by the action of rain dissolving the CO2 and the resulting acid reacting with rocks and ending up in the bottom of oceans.

[steve]

Jan 26, 2012 15:08:08 GMT Andrew said:

Jan 26, 2012 13:36:42 GMT steve said:

Unless you can see in detail all the things that are going on it is impossible to show a direct cause and effect relationship.

However, we do know that CO2 is a greenhouse gas, and we can calculate the impact caused by increasing its levels in a given atmospheric composition. So we know it has the *potential* to warm the planet.

We know that the planet has a pretty sensitive climate. We know this because we know the climate has varied a lot in the past, and particularly that the correlation between temperature and the Milankovitch cycles (which on the face of it are causes of very minor forcing) is pretty good.

We also know that there is a good correlation between CO2 and temperature, not only for the ice ages, but also going back for hundreds of millions of years. And it is hard to see how long timescale climate variations could really be managed by mysterious ocean cycles/etc. that people like to invent to explain recent warming.

We do however know that there are processes in which CO2 levels will rise due to temperature rises, so certainly the link is not one-way.

On the shorter ice age cycles it seems that rising temperatures can cause CO2 to rise which accentuates the temperature rise, along with feedbacks like less ice cover. Eventually, the increase in outgoing longwave radiation is higher than the increase in CO2/ice/water vapour feedbacks and things are stable till something triggers a cooling at which point things go the other way.

On the timescales of 10s of million years, warmer temperatures caused by CO2 speed up the (still very slow) removal of CO2 by the action of rain dissolving the CO2 and the resulting acid reacting with rocks and ending up in the bottom of oceans.

Steve

There is a difference between calculating and estimating. Your calculations for C02 will tend to rely on a large number of assumptions. Where importantly the modellers assumptions and bias and opinions will tend to drive the result he gets. There is no way you can calculate the impact that change of C02 will have on a complex system like Earth.

You are reading too much into what I said. It is quite easy to get a good estimate of the amount of radiative forcing that CO2 will cause in a typical atmosphere, and you will be hard pressed to find even a sceptic scientist who disputes this. What happens to that atmosphere and to the planet *as* CO2 forces it is a second (and of course important) matter as it is for the Milankovitch cycles.

The Milankovitch exert no net forcing integrated across the globe and over one year. However, they exert forcing of different amounts at different times of year which results in wild swings of temperature. What you are trying to tell me is that an actual forcing equivalent to many times more than the likely variation in the sun can be dismissed. What I am saying is that that amount of forcing is a) a given b) calculable and c) significant.

You are also very dismissive of things that could be an alternative to warming, which then tends to reinforce the idea - that exists in my mind and might reflect realities - that you are assuming C02 is the cause and your assumptions then lead you to the result that supports what you think.

In a complex system like the Earth, warming and cooling mini cycles - like have been going on in the Holocene - could be entirely random events.

The only thing I've been "dismissive" of is hypotheses of long term cycles with very little empirical support. I don't believe that the changes during the holocene are as big as is imagined. I don't believe that even if they are so big that they can have any role in explaining much longer term (100 thousand to 100 million year order) changes in climate such as those that show good correlation with CO2 levels.

For example there have been about 30-30 such cycles found to exist from observing the data for the largest glacier in mainland europe where a typical warming cycle averages 160 years i think. So todays warming could just be one of those random cycles of warming. And then, either already or tomorrow a cooling cycles begins for around average of 140 years. These are averages. Warming or cooling in a random cycle might continue for many hundreds of years.

Additionally, cooling such as the little ice age is a cooling mini cycle that is itself made of many cycles, where ultimately every day is just a random cycle - within limits of course.

Anyway the first point is that you cannot calculate the warming influence of C02. All you can do is make assumptions/guesses and have a go at an estimate.

How do you know that you are not simply trend following?

You cannot know that. No way can you know it because the system is too complex and you lack sufficient knowledge.

You might know that storms warm the oceans because when water moves it creates friction and therefore heat. The oceans themselves are massively more able to store heat than the quite puny ability of the atmosphere and absolute surface of the Earth. The ground of the Earth can store heat.

All of the above paragraph could be zero sum and yet on the surface and in the atmosphere there could be considerable warming. It would mean nothing at all. On the other hand the surface warming could create some additional change and some other random change could be underway, which hundreds of years later just comes back to where you were before. Etc etc etc.

But already you have decided this is irrelevant. You say you can calculate things so you know things.

Therefore as a beginning observation of what you have said, you are sounding too confidant you can know what is essentially unknown to you at this point in time.

There is also a second basic reality here when humans theoretically model things that involves a complex system. And that is it is impossible to get a result that is worthwhile unless you test it.

Modelling the earth is a bit rediculous because you can never test it. The earth cannot be set up with a set of parameters to run an actual test as you would for example an icebreaker for which you have just produced a physical example after doing calculations based on modelling and so forth.

You cannot calculate a better icebreaker. All you can do is make guesses and assumptions and test them and then alter the physical model until you finally get better real results, from which you then build an actual icebreaker and begin testing and changing that.

I won't dismiss the possibility that recent warming (last 50 years) is, say, 50% "natural" and that the models may have missed some important changes in forcing, though I think it is unlikely.

I will dismiss the possibility that CO2 has no or little role in the warming for the reasons given above.

Will you dismiss the possibility that it is mostly anthropogenic?

[sigurdur]

Steve:
When you look at the long term temperature change since 1800, it shows a steady trend.

Within that trend, one can use a channel. This is a standard tool employed when doing commodity analysis. The channel roughly establishes the price variations within a trend.

Temperature would follow the same dynamics one would think?

What we have to analyze is the deviations that would fall out of the normal channels. The past 50 years do not do this, which would indicate that the effects of rising co2 have not yet occured.

[steve]

In what way is saying that perhaps 50% of the warming is not due to CO2 "dismissing other random factors".

You are stating your case very forcefully but so forcefully that you are forcing words into my mouth.

The recent warming will at the moment tell us little about climate sensitivity because our observations of other forcings (and other natural climate variability) are poorly constrained.

The evidence from past climates (ice age and before) is important in telling us that the climate *is* sensitive.

If the climate is *still* sensitive then the risk is that the CO2 forcing that we know we are applying will cause a significant amout of warming.

I don't see what can cause the atmosphere to decide to hold more water than it currently does without some form of influence. I don't see the evidence that an influence that could cause the atmosphere to hold more water has occurred. I don't believe in "cycles" unless someone can give me a good physical basis for the "cycle". Anything else is just numerology and tenuous hypothesis.

The best we can hope for is that the sensitivity of climate today is less than it has been in the past (eg. because of current amounts of ice, or particularities of current land cover). I don't rule that out. Limited evidence is suggesting that the models are over-sensitive. But the evidence is very limited and there are plenty of unknowns (unknown forcings such as dust and aerosols, cloud variations, and uncertainties about ocean heat uptake) that can swing things either way.

In other words, I don't know but really there is no case for dismissing concerns.

[steve]

I am wondering why you appear fairly keen to dismiss potential
Holocene variations as 'imaginations' while you tell me that pre ice
age evidence tells you that you *know* that climate *is*
sensitive.

I'm neither keen nor not keen. The evidence as I see it is that the
climate was warmish during the MWP period, went coldish and has now
got warm and probably warmer. If you look at any sceptical analysis of
the MWP (for example the database held by the co2science.org lobby
group) you see a huge desire to emphasise the warm periods in
individual local "proxies" but ignore the cold periods.

The difference between the Holocene and the prehistorical periods I've been
talking about is that the Holocene has so far been quite short and the variation
in climate quite small. In the prehistoric periods we are talking about 100
thousand or 10s of millions of years with warmer or colder temperatures (on
average). Within those periods there will be massive amounts of shorter
term variation that doesn't show up in the record, but long term the climate has
been sustained at a particular warmer or cooler temperature based (in my belief)
on the long term stable forcings of that time. If "cycles" were more
important then
the evidence for links between climate and CO2 or Milankovitch would be weak.

If you can believe that holocene variations are small you have to
attribute more emphasis to forcing by some factor you have to provide
to account for the current warming, where if there were actually
significant holocene variations you would need to remove some of the
forcing factors.

If variations are not imaginations then your forcing factors are way
out.

That is terrible logic. If we were able to send our observing
systems back 1000 years who is to know whether we would easily
identify the likely causes of the climate variability. A 20th Century
anecdote that will go down in the annals of time is the US 1930s
dustbowl, but it is easy to see this huge historical event now as atemporary climate
variation (or just weather) coupled with bad farming practices.
Something like this happening a thousand years ago though would have
been attributed by co2science and the like as an impact of the MWP.

Maybe you also want to think that heat island impacts are imaginations
that can be lightly dismissed so that calculated warming has an extra
amount?

The suggestion that heat island impacts are lightly dismissed is
insulting to the scientists who have spent much of their life
assessing them and whose work has now been all but validated by the
failure of the surfacestations project to obtain its preferred result,
the work of the Muller BEST project and the analyses by the folk who
gather at Lucia's blog.

The impression i have is you want the result rather than you look at
the data and take what comes out of it.

Indeed!

Water in the atmosphere creates clouds, snow ice and water at the
surface as moisture or lakes. Waters influence is not just as a
greenhouse gas.

The Arctics ice acts to insulate the heat of the arctic ocean from
deep space. If you throw in a significant number of years of random
storms the ice breaks up and begins travelling south. Importantly the
ice has 80 times the cooling capacity of ice cold arctic water flowing
out of the arctic. Eventually the seas of the arctic are exposed to
deep space and eventually the seas returning to the arctic via the
gulf stream are cooler. Add in a significant period of no random
storms and the arctic can refreeze.

I just made this up now, however these kinds of cycles could take
thousands of years to play out. While all that is going on you have
changes in rainfall, different glacial amounts due to precipitation
and or cold, local coolings and heatings as ice comes and goes. Etc.

I find it very hard to believe that such random changes are not in
play all of the time.

You though have taken out this randomness from the holocene so that
you must provide a forcing factor for the observed warming

So you've made up a lot of things to invent a scenario. That's fine.
It would be relatively easy for someone to plug that scenario into some sort of
model and see the impact. You could even try to tweak it to see the
results if the effects of your changes are at the extreme end of
likelihood (in terms of the reality from the observations that you
input into the model and the reality of the physical assumptions).
Good luck with doing that *and* maintaining a stable and plausible
climate in the rest of your model. If you manage to do this then your
hypothesis may become theory.

But you are not the first to consider such things.

[trbixler]

So do the models use CO2 as a lagging indicator to the rise in temperature. Kind of a reality check to the models performance. If so what is the primary driver?

[steve]

You said my logic was terrible. What i am meaning is that if there is a period in for example the little ice age where there is a similar warming as that of today, then it suggests that your models are putting warming down to a factor you have had to insert in the model that then creates warming for you.

Where I think your logic is wrong is in assuming that if we did find a past period of warming like recently, that it validates the theory that current warming is part of some sort of cycle.

CO2 is not the only "forcing". Other forcings include volcanoes, solar variations, anthropogenic pollution (sulphur emissions etc.), methane. The attribution of current warming attempts to include *all* these factors and more.

If you find a similar past period of warming you should factor all these things in also. There is nothing that says a sustained rise or reduction in the number of sulphur emitting volcanoes could not cause similar temperature changes, for example.

I hope that is clear because it is an important point.

Assuming that it is clear, I do *not* dispute the existence of unknown forcings that can influence the amount of warming. But a) such forcings can cut both ways and b) existence of such forcings does not mean that you can simply deny the calculable radiative forcing that CO2 causes.

[steve]

Jan 27, 2012 14:08:35 GMT trbixler said:

So do the models use CO2 as a lagging indicator to the rise in temperature. Kind of a reality check to the models performance. If so what is the primary driver?

You'll have to explain yourself more. The observed relationship between CO2 and temperature during, say, the ice ages is not resolved to better than an order of 100s of years, and is not consistent. So not sure what it can tell you about a <100 year climate model run.

Also the CO2 variations we're talking about during the ice age cycles are small compared with current observed rise.

[steve]

How many times do i have to point out the obvious reality that you cannot calculate the radiative forcing that C02 causes?

It's late. So I'll only respond to this bit for now.

You haven't actually said that the radiative forcing *cannot* be calculated. You've said that there is a difference between calculating and estimating, which doesn't really say much if your estimate can confidently be said to be within, say, 10% of reality, does it?

I have said that the radiative forcing can be calculated if you increase the amount of CO2 in a typical atmosphere composition.

If you are worried that the calculation is an *estimate* because it depends on assumptions about the atmospheric composition or other reasons, then re-do the calculation with different assumtions. The answer won't be significantly different.

[numerouno]

The data of the last millions of years seems to show that C02 does not drive climate change but rather temperature changes drive the composition of the atmosphere.

Iceskaters, the current science (from the 1800s) holds that even the small portion of energy captured by CO2 and other GHC from the huge radiated energy of the Sun will have an warming effect on climate, regardless of what the Sun's current activity and the current state of the Milankovitch cycle might be. (Additionally, you might want to actually check your spelling of "CO2")



Where I think your logic is wrong is in assuming that if we did find a past period of warming like recently, that it validates the theory that current warming is part of some sort of cycle.

Yep, the silly denialist logic is like claiming we don't have to worry about or put off modern forest fires any longer, since bolts of lightning have ignited forests in the past.

[trbixler]

As usual Mr. Green is in charge.
Yes silly logic says manage the forests to limit the effect of forest fires. While Greens fight for no management to maximize the fire impact.

"“We’ve lost two people in my family because you thingyheads won’t cut trees down…”"



wattsupwiththat.com/2009/02/11/weve-lost-two-people-in-my-family-because-you-thingyheads-wont-cut-trees-down/

[icefisher]

You haven't actually said that the radiative forcing *cannot* be calculated. You've said that there is a difference between calculating and estimating, which doesn't really say much if your estimate can confidently be said to be within, say, 10% of reality, does it?I have said that the radiative forcing can be calculated if you increase the amount of CO2 in a typical atmosphere composition.



Usually the problem when dealing with this sort of stuff is in how words are used. I would suggest you have not calculated a forcing but instead an attenuation of certain frequencies of IR which is exactly what one would expect if a gas absorbed at certain frequencies and retransmitted at least some amount of that absorbed energy at normal frequencies of thermal radiation. In such an instance one cannot claim a forcing.


When one views radiation from space, the globe radiates the amount of thermal radiation that is expected.

Further I would suggest that if as you have calculated 99% of the gases in the atmosphere did not absorb any radiation what so ever the atmosphere/climate of the earth would be a lot warmer than it is today.

The basic argument why that would be true would be: Absent any IR absorbing gases in the atmosphere, the atmosphere would be much better insulated unless the amount of atmosphere was trivial.

GLC argued at length that greenhouse gases did not violate the 2nd law of thermodynamics because the gases instead effectively slowed (using an insulating blanket as an analogy) cooling of the surface. However, in truth it is well known that materials, all else being equal, which are the worst absorber/emitters are in fact the best insulators. Thus standing thermodynamics on its head.

The argument that the surface of the earth would then radiate all it received from the sun to space making the climate cooler is completely without merit in the presence of the ideal gas law and laws of thermodynamics.

I maintain that the earth's atmosphere is non-trivial and it would only be possible for the earth's climate to be as cool as the SB calculated temperature, in such an instance as above, if all of the following conditions were true. 1) the face of the surface of the earth radiated at 100% efficiency; and 2) that incoming radiation to any surface on the globe was uniform and invariable.

Care to take on any assertion above? I have been fine tuning my worldview and would love for it to be constructively criticized.

[numerouno]

Care to take on any assertion above?

Icefisher, I don't think your private (or actually on loan from other denialists, each with a slight personal variation) thermodynamics beats the one they teach in schools and universities.

I think you partly realize that GHSs work both ways, both on the rays' way in, and the way out. This is I suppose what your attempt at "insulating materials" means.