What is the correct value of Climate Sensitivity?

[-1=e^ipi]

Thought I would mention this because it has interesting implications. Bjorn Stevens has 2 papers recently, which help explain some of the discrepancy between climate model estimates, and observations.

The first is that Bjorn Stevens has reduced the uncertainty & median estimate on aerosol forcing and the result suggests that climate models (as well as simple radiative balance calculations) have been overestimating climate sensitivity (since the negative forcing of aerosols somewhat offsets the positive forcing of greenhouse gases in the instrumental record). Nick Lewis' instrumental best estimate of climate sensitivity drops from 1.64 to 1.45 C (a 12% drop).

http://climateaudit.org/2015/03/19/the-implications-for-climate-sensitivity-of-bjorn-stevens-new-aerosol-forcing-paper/

The other paper finds support for Lindzen's Iris hypothesis (which to my understanding means a stronger negative cloud feedback in the tropics). Since climate models aren't taking this effect into account, they are overestimating climate sensitivity. For example, this paper suggests that the climate sensitivity of ECHAM6 climate model decreases from ~2.8 to ~2.2 C.

http://judithcurry.com/2015/04/22/bjorn-stevens-in-the-cross-fire/

[jbg]

No, it was a result of the melting of the North American, Eurasian and other ice caps + glaciers at the end of the ice age, which was caused by Milankovitch cycles + the GHG & albedo feedbacks, which increased sea levels by 105-120 m. Is this supposed to be a rhetorical question?

No. I was pointing out that submersion and emergence of dry land can occur without man's help.

[-1=e^ipi]

Earlier I referenced a Lunt et al. paper that suggested that the ESS is approximately 1.4 times ECS (it compared changes between the Pliocene and the Holocene). However, this was using an ECS of ~ 3.05 C.

The ratio of ESS/ECS is unlikely to be constant. Rather, you can think of the ECS to be the 1.15 C no feedback CO2 sensitivity + fast feedbacks and the ESS to be the ECS + slow feedbacks. If we let a = fast feedbacks and b = slow feedbacks then:

3.05 C = ECS = 1.15 C / (1 - a)

1.4*3.05 C = ESS = 1.15 C / (1 - a - b )

This suggests that b = 0.108.

Now if ECS is actually lower (say 2 C), then that means that the fast feedbacks, a, are lower (0.425), which would give a lower ESS for the same slow feedbacks, b, (2.46 C). In this case, not only has the ECS dropped, but the ratio of ESS/ECS has dropped (from 1.4 to 1.23).

Edited April 23, 2015 by -1=e^ipi

[Michael Hardner]

Climate has always changed. With little or no help from humans we've had...

Restating this pointless argument once again makes no difference to the discussion.

Once we got a critical mass of this board to agree that the climate is changing and that (at a minimum) humans are impacting this, I bowed out of these discussions.

To me, the discussions we should now be having here (and at large) are about economic choices. TimG has been knowledgeable about these in the past.

[cybercoma]

Economic choices? You or someone dear to you is dying of kidney failure and you can be saved, but it's going to cost a lot of money. Do you make an "economic choice" here? Only a complete fool would claim that something so serious is only about dollars and cents.

[eyeball]

To me, the discussions we should now be having here (and at large) are about economic choices. TimG has been knowledgeable about these in the past.

Where, when, how and with what?

Cite?

[Michael Hardner]

I don't have time to comb through to site exactly what made my opinion on TimG's expertise, but I remember him quoting the economic consensus on the cost of climate change over time, and the costs of mitigation vs prevention in the same conversation.

It was awhile ago - TimG can you help me out here ?

[TimG]

but I remember him quoting the economic consensus on the cost of climate change over time, and the costs of mitigation vs prevention in the same conversation.

I have referenced the social cost of carbon calculations on a number of times: http://en.wikipedia.org/wiki/Carbon_tax

These are estimates by economists to determine the harm caused by future emissions. These estimates depend on assumptions and vary widely but the generally accepted mean is around $40USD/tonne of CO2. This works out to a 3 cent/liter tax on gasoline. Since gas taxes already exceed this amount it is possible to argue that consumers are already paying for externalities caused by CO2 emissions from gasoline and no further taxes are required.

Even if we use the high end SCC of $200/tonne the required tax on gasoline is only 15 cents/liter which is lower than current taxes on gasoline in most industrialized economies.

These numbers tell me that the costs of adaptation may not be zero but they are hardly catastrophic. They also show that almost any CO2 reduction measures that cost more than the SCC are an inefficient use of resources and it is reasonable to say that such measures should not be adopted even if they do result in a reduction in CO2 emissions unless they have benefits other than the CO2 reductions.

Edited April 24, 2015 by TimG

[Michael Hardner]

This is where the debate should be happening, I think.

[TimG]

This is where the debate should be happening, I think.

I agree, but that position makes you a 'climate change denier' according to the typical alarmist.

I am not exaggerating, look at the what happen to Roger Pieke Jr when he got a column at Nate Silver's blog: http://www.dailykos.com/story/2014/07/28/1317347/-Climate-disinformer-Roger-Pielke-Jr-out-at-Nate-Silver-s-FiveThirtyEight-blog

Pieke Jr emphatically insists that CO2 is a serious problem. His only sin is he emphasizes a rational, evidence based approach to policy which is pretty close to my own position. The treatment of people like Pielke Jr is one of the reason why I think AGW alarmism is a religious cult that worse than Christian Fundamentalism in terms of the harm it can cause to our society.

Edited April 24, 2015 by TimG

[Michael Hardner]

The treatment of people like Pielke Jr is one of the reason why I think AGW alarmism is a religious cult that worse than Christian Fundamentalism in terms of the harm it can cause to our society.

That's just a finger pointing exercise. Others would argue that the conversation can't start in earnest until some groups stop fighting the fact that human-caused warming is happening.

[TimG]

That's just a finger pointing exercise. Others would argue that the conversation can't start in earnest until some groups stop fighting the fact that human-caused warming is happening.

You are making excuses. The people to blame for the lack of meaningful conversation on policy are the people who throw a fit because a moderate like Roger Pielke Jr got a column on a popular blog. The nasty treatment has forced him out of the field entirely which means the only people left arguing are the extremists. This is not so bad for the skeptics because a shouting match between unreasonable extremists ensures nothing gets done which suits them fine. It is only really bad for people who would like to see something done which means it is those people who have a responsibility to stand up and defend moderates like Pielke Jr or Lomborg. Edited April 24, 2015 by TimG

[Michael Hardner]

It is only really bad for people who would like to see something done...

Anyway, it's one group's bad behavior over another. Moving on...

[Moonlight Graham]

But we live in the real world and, IMO, eliminating CO2 emission is not remotely plausible given economic and political constraints. We could reduce the rate of CO2 emission growth but that is not going to make much of a difference. The only real options are adaptation.

I don't think eliminating CO2 emissions is realistic, and that's a strawman since not many are even seriously considering that option. But reducing would make a difference, and is an achievable option, the question is how much reduction if at all based on cost/benefit analysis. How can you say it "won't make a difference", espcially when you're not even defining the rate of reduction. If you reduce emissions by ie: 75% or 50% or 30% then yes of course it's going to make a difference.

What needs to be done, is to do our best to estimate the net longterm cost or benefit from rising temperatures (granted, very difficult task), and if there is a net cost, to calculate if the money it would take to significantly reduce CO2 using other power sources is less than any net cost to rising temps plus any costs for those who will need to adapt to climate change if we decide to do little/nothing to reduce CO2 (ie: flooded coastal cities). ie: If it costs the U.S. a net 1 trillion if we do little/nothing, but would cost 3 trillion to move to non-CO2 energy sources, then it should support doing little/nothing, and vice versa.

[-1=e^ipi]

To me, the discussions we should now be having here (and at large) are about economic choices. TimG has been knowledgeable about these in the past.

But you can't really have meaningful discussions about the economic impacts without first understanding the magnitude of change that is to be expected due to climate change (which is why establishing climate sensitivity is important). After that there are many other things that need to be established (such as how much GHG emissions can we expect given different policies, how do we quantify and compare all the environmental changes, what is the response time of the climate to absorbing excess atmospheric CO2, etc.) before getting to what policy makes sense. To me, what to do about climate change is a priori indeterminant so the question of what to do about climate change cannot be answered without going through empirical evidence.

If you want to discuss the economic choices, then perhaps it would be better to create a different thread because it might be better to have a thread that is dedicated to the science and discussing expected impacts of climate change (this one). I was thinking of making a thread to discuss emission scenarios and why I think some of the IPCC's emission scenarios such as RCP 8.5 are nonsense; perhaps trying to answer the emission scenario question next makes sense.

but I remember him quoting the economic consensus on the cost of climate change over time

But is there an economic 'consensus' given that there are so many unanswered questions such as climate sensitivity?

I agree, but that position makes you a 'climate change denier' according to the typical alarmist.

Oh yes. Welcome to the 'denier' club Michael!

Edited April 25, 2015 by -1=e^ipi

[-1=e^ipi]

Only a complete fool would claim that something so serious is only about dollars and cents.

Economics is about more than dollars and cents.

[On Guard for Thee]

Economics is about more than dollars and cents.

I suggest heading back to you and Tims little playpen and throw around some more little formulas. The grownups need to deal with the problem at hand. Even Harper is getting on board, although we know that is likely just another sideshow for his voter base.

[TimG]

I don't think eliminating CO2 emissions is realistic, and that's a strawman since not many are even seriously considering that option.

You obviously are not listening to what they are talking about at these annual UNFCCC conferences. The end goal requires setting targets to get close to zero emissions (after meeting intermediate targets). If you say that getting to zero is impossible you are a "denier". See http://en.wikipedia.org/wiki/United_Nations_Framework_Convention_on_Climate_Change

To stabilize atmospheric GHG concentrations, global anthropogenic GHG emissions would need to peak then decline (see climate change mitigation).[36] Lower stabilization levels would require emissions to peak and decline earlier compared to higher stabilization levels

The key point is at some point in the future CO2 emissions must be less than natural absorption rate which implies a reduction close to zero.

But reducing would make a difference, and is an achievable option, the question is how much reduction if at all based on cost/benefit analysis. How can you say it "won't make a difference", espcially when you're not even defining the rate of reduction. If you reduce emissions by ie: 75% or 50% or 30% then yes of course it's going to make a difference.

My position is any policy based on reaching reduction targets will fail and and is a waste of money. If you want to take absolute reduction targets off the table and instead focus on incremental improvements that increase GDP per unit of CO2 emitted then I will agree that some reduction is better than no reduction. The problem now is any discussion about CO2 mitigation policy is fixated on achieving absolute reductions and I cannot support such policies because they will do more harm than good.

The problem with policies based on targets is two fold:

1) Population is increasing so even of you reduce the CO2 emissions per person the total emissions will continue to rise. For that reason alone targets based on absolute reductions are simply unachieveable in high growth countries.

2) When policy is fixated on meeting unachieveable targets then loop holes are created to allow people to emit while getting credit for spending money on other projects that claim to reduce emissions (e.g. carbon credits). The trouble is these projects range from the ineffective to outright scams that do nothing about the total CO2 being emitted. i.e. emissions continue to rise but they are hidden by accounting tricks.

What needs to be done, is to do our best to estimate the net longterm cost or benefit from rising temperatures (granted, very difficult task)

This has been done. It is called the "Social Cost of Carbon". The SCC is between 20 and 200 USD per tonne of CO2. 40USD/tonne is the middle of the road estimate used by the EPA. If an CO2 reduction policy costs more than the SCC then it is more cost effective to adapt. Almost all CO2 reduction measures promoted today cost significantly more than 40 USD/tonne and this is the basis for my claim that adaption is a smarter way to deal with the problem. Edited April 25, 2015 by TimG

[TimG]

Economic choices? You or someone dear to you is dying of kidney failure and you can be saved, but it's going to cost a lot of money. Do you make an "economic choice" here? Only a complete fool would claim that something so serious is only about dollars and cents.

All of life comes down to economics. If the choice was between sending kids to college or leaving a spouse without any retirement savings you would find a lot of people would forego the kidney. The only thing that shields most people from such choices are insurance policies that cover high cost/low probability events. But insurance policies cost money and in a lot cases the cost of the insurance is too high and people go without and accept the consequences.

Climate change itself is primarily an economic problem. i.e. will change interfere with the smooth functioning of human societies. If it does not then we don't care about it. If it does we do care about it. Climate change is not a problem on its own merits since the climate has always changed and always will change.

Edited April 25, 2015 by TimG

[-1=e^ipi]

The key point is at some point in the future CO2 emissions must be less than natural absorption rate which implies a reduction close to zero.

It really doesn't. Remember that in the long run the oceans absorb 85% of emitted CO2, plus the biosphere absorbs more CO2 at higher levels of CO2 due to the CO2 fertilization effect. So if say emissions were stabilized at say 600 ppm, the ocean concentration of dissolved CO2 times Henry's constant at the point of stabilization is roughly 300 ppm, and the decay time of atmospheric CO2 is 100 years, and it takes 2.13 gigatonnes of carbon of fossil fuels to increase atmospheric CO2 by 1 ppm (all realistic numbers), then this would suggest that humanity could burn 6.39 gigatonnes of carbon per year without increasing atmospheric CO2. Current annual CO2 emissions is ~ 10 gigatonnes. Therefore, to stabilize CO2 levels at 600 ppm would only require a reduction in global emissions by ~36% from current levels.

In any case, if we use an ECS of 1.5 to 3 C, then this suggests that this stabilization at 600 ppm would lead to a long run global temperature change of (1.5 to 3)*ln(600/278)/ln(2)-0.8 = 0.86 to 2.53 C relative to current temperatures.

[TimG]

It really doesn't. Remember that in the long run the oceans absorb 85% of emitted CO2

The data I saw says 50% and that is a relative figure - not an absolute one so if emissions drop the absolute amount absorbed may drop too. Also, the UNFCCC talks about needing to reduce the levels of CO2 in the atmosphere (i.e. keep it below 350 or 450ppm). Doing this means human emissions must be below the natural re-absorption rate.

In any case, if we use an ECS of 1.5 to 3 C, then this suggests that this stabilization at 600 ppm would lead to a long run global temperature change of (1.5 to 3)*ln(600/278)/ln(2)-0.8 = 0.86 to 2.53 C relative to current temperatures.

My comment is in response to someone saying that no one is seriously expecting emissions to drop to zero so the relevant numbers are 350-450ppm which the UNFCCC seems to think is necessary. Reality may be quite different but that was not the point of the comment. Edited April 25, 2015 by TimG

[-1=e^ipi]

The data I saw says 50% and that is a relative figure - not an absolute one so if emissions drop the absolute amount absorbed may drop too.

The 50% is rate of natural uptake relative to current emissions, not the equilibrium response. The rate of natural uptake is proportional to the difference between atmospheric CO2 emissions and Henry's constant times oceanic concentration of dissolved CO2; so uptake depends directly on atmospheric CO2, not human CO2 emissions. If our emissions suddenly halved, the rate of natural uptake would initially stay the same.

Also, the UNFCCC talks about needing to reduce the levels of CO2 in the atmosphere (i.e. keep it below 350 or 450ppm).

There is no scientific basis for this claim by the UNFCCC.

Doing this means human emissions must be below the natural re-absorption rate.

But the natural re-absorption rate depends on atmospheric CO2 levels; a higher concentration of CO2 means a higher re-absorption rate.

[cybercoma]

Economics is about more than dollars and cents.

Not in the context that Michael Hardner was describing it.

[Michael Hardner]

Economic choices? You or someone dear to you is dying of kidney failure and you can be saved, but it's going to cost a lot of money. Do you make an "economic choice" here? Only a complete fool would claim that something so serious is only about dollars and cents.

Economy is about how we use limited resources, including the environment, people, and energy sources.

If course you make an economic choice to save your potted plant, your cat, or your grand-dad. I hate to be crass but on a certain level everything is about limited resources, abilities, and what we can/can't do.

[-1=e^ipi]

@ Michael - I've changed my mind; if you want to discuss the economic impacts of climate change in this thread, I am fine with that.

Anyway, I thought I would try to estimate a very crude model to predict CO2 emissions under a no-mitigation scenario.

I'll use the identity: Annual CO2 Emissions = Global Population * Real GDP per capita / (GDP per unit of CO2 emitted)

I'll use 3 data sets over the years 1950-2008 (the area of overlap):

Global Annual CO2 Emissions (in metric tonnes of carbon): http://cdiac.ornl.gov/trends/emis/tre_glob.html

Global Population: http://www.geohive.com/earth/his_history3.aspx

Global real GDP per capita: http://www.ggdc.net/maddison/maddison-project/data/mpd_2013-01.xlsx

It is well known that for centuries, the global population was increasing roughly exponentially. However, currently the global population growth rate is slowing down (people are having less children) and it is expected to peak or plateau mid century. A simple model that can approximate this behaviour is the logistic function (http://en.wikipedia.org/wiki/Logistic_function). The logistic function is used frequently in ecology to model the population of other species, so why not humans?

To estimate the logistic function of human population growth, I merely have to perform a linear regression of dN/dt = A*N + B*N^2 + error, where N is the population of the planet and dN/dt is the annual change in population of the planet.

Estimating this gives A = 2.76*10^-2 and B = -2.31*10^-12.

Using the above estimate of the differential equation, plus the fact that the global population in 2010 is 6916183482, I get that the 95% confidence interval for the global population in 2100 is 11.28 +/- 0.14 billion people. Note that this error only includes the error assuming the model is true; so is an underestimate of the true uncertainty.

For changes in real GDP per capita over time, this follows a roughly exponential trend (one can even look at real GDP per capita of the USA over the past 200 years to see that this trend can hold for centuries). Therefore, to estimate future real GDP per capita, I merely have to perform a linear regression of ln(y) = C + D*t + error, where y is real GDP per capita and t is time.

Estimating this gives C = -31.57 and D =0.0202.

Using the above estimate, I get that the 95% confidence interval for real GDP per capita in 2100 is ~(47630 +/- 4510) dollars (1990 US $).

For changes in economic output per unit of CO2 emitted, this appears to follow a roughly quadratic trend since 1950 (the ratio is both increasing and accelerating over time). To estimate future economic output per unit of CO2 emitted (under a no-mitigation scenario), I perform a linear regression of Ratio = E + F*t + G*t + error.

Estimating this gives E = 3.90*10^12, F = -3.99*10^9 and G = 1.02*10^6.

Using the above estimate, I get that the 95% confidence interval for economic output per unit of CO2 emitted under a no-mitigation scenario in 2100 is 24.56 +/- 1.32 trillion dollars (1990 US $) per gigaton of carbon.

Now, using the identity Annual CO2 Emissions = Global Population * Real GDP per capita / (GDP per unit of CO2 emitted), I get that in 2100 the 95% confidence interval for Annual CO2 Emissions under a no-mitigation scenario is 21.9 +/- 2.4 gigatons of carbon per year.

So under a no-mitigation scenario, we are looking at slightly over twice current emission levels (roughly 10 gigatons of carbon per year) by 2100.