Where does the Equilibrium Sensitivity lie for a Doubling of CO2?

[Snow_Miser]

I'm kind of curious with regard to what the general thought is on this forum about how much warming we can expect with a Doubling of CO2. Discuss.

The Blackbody Sensitivity for a doubling of CO2 is estimated to be around 0.8-1.2 Degrees C.

[Snow_Miser]

Question for the people who selected 2.5-4.5 Degrees C:

Assuming the mean sensitivity is close to 3.5 Degrees C per doubling, then we should have observed a 1.4 Degree C temperature increase over the 20th Century when CO2 levels rose sharply, assuming all of the warming in the 20th Century was anthropogenic (which is wrong). Instead, we only observe a 0.6 Degree C increase.

Are you saying that there is 0.8 Degrees C in the pipeline for warming?

[PottercountyWXobserver]

Question for the people who selected 2.5-4.5 Degrees C:

Assuming the mean sensitivity is close to 3.5 Degrees C per doubling, then we should have observed a 1.4 Degree C temperature increase over the 20th Century when CO2 levels rose sharply, assuming all of the warming in the 20th Century was anthropogenic (which is wrong). Instead, we only observe a 0.6 Degree C increase.

Are you saying that there is 0.8 Degrees C in the pipeline for warming?

Yes, the heat didn't disapear. Majority of the rest of the heat was absorbed primarily into the oceans and arctic ice. Once the ice has practically vanished, heating in the atmosphere will increase dramatically, especially in the late fall time frame like it has but on a stronger note IMO. Octobers-Novembers will set big temperature records I believe coming up within the next few years.

[Snow_Miser]

Yes, the heat didn't dissapear. Majority of the rest of the heat was absorbed primarily into the oceans and arctic ice. Once the ice has practically vanished, heating in the atmosphere will increase dramatically, especially in the late fall time frame like it has but on a stronger note IMO. Novembers will set big temperature records I believe coming up within the next few years.

No one is saying that the extra heat accumulated from carbon dioxide disappeared. What I'm questioning is how much heat is actually in the pipeline. I find 0.8 Degrees C in the pipeline to be really not that believable, considering we haven't warmed for many years now.

That would mean that there is a very potent forcing completely canceling out the warming impact from CO2 if there really is 0.8 Degrees C of warming in the pipeline. What do you propose is this negative forcing counteracting the CO2 forcing?

[PottercountyWXobserver]

No one is saying that the extra heat accumulated from carbon dioxide disappeared. What I'm questioning is how much heat is actually in the pipeline. I find 0.8 Degrees C in the pipeline to be really not that believable, considering we haven't warmed for many years now.

That would mean that there is a very potent forcing completely canceling out the warming impact from CO2 if there really is 0.8 Degrees C of warming in the pipeline. What do you propose is this negative forcing counteracting the CO2 forcing?

Overturning of SSTs. The Oceans will absorb much of the warming caused from CO2 in my opinion. A lot of the heat is mixing throughout the column of water, and the deeper one goes, the warmer the temperatures have become compared to normal. Why do you think bottom melt in the arctic is such a huge player. Bottom melt in the arctic is a very positive forcing in my opinion, but oceans that are already warm to begin with will mix the colder deeper water which acts as a negative forcing eventhough this "negative" forcing is only temporarily until the layers have been thoroughly mixed. If I'm wrong in my thinking Friv, Vergent, or anyone please chime in.

[Snow_Miser]

Overturning of SSTs. The Oceans will absorb much of the warming caused from CO2 in my opinion. A lot of the heat is mixing throughout the column of water, and the deeper one goes, the warmer the temperatures have become compared to normal. Why do you think bottom melt in the arctic is such a huge player. Bottom melt in the arctic is a very positive forcing in my opinion, but oceans that are already warm to begin with will mix the colder deeper water which acts as a negative forcing eventhough this "negative" forcing is only temporarily until the layers have been thoroughly mixed. If I'm wrong in my thinking Friv, Vergent, or anyone please chime in.

So in other words, you believe that natural internal and dynamical changes in the ocean are responsible for the surface temperature halt in recent years.

How do you know that the oceans did not play an important role on the late-20th century warming trend, if the oceans are potent enough to completely eliminate the positive temperature increase?

[PottercountyWXobserver]

So in other words, you believe that natural internal and dynamical changes in the ocean are responsible for the surface temperature halt in recent years.

How do you know that the oceans did not play an important role on the late-20th century warming trend, if the oceans are potent enough to completely eliminate the positive temperature increase?

There are natural variations of ocean cycles without a doubt, but even with these cycles, the extra heat is being "hidden" in the oceans. Granted some of the heat experienced is due to the +AMO we have seen within the recent years, but we are also seeing an overlapping signal with the added heat. Places are experience record temperatures at deeper depths than ever measured. Natural internal and dynamic changes are just that, natural. These cycles can't add energy into the system, they only redistribute what is added to it. Eventually a neutral and -AMO standard today is going to be yesteryears +AMO.

[Snow_Miser]

There are natural variations of ocean cycles without a doubt, but even with these cycles, the extra heat is being "hidden" in the oceans. Granted some of the heat experienced is due to the +AMO we have seen within the recent years, but we are also seeing an overlapping signal with the added heat. Places are experience record temperatures at deeper depths than ever measured. Natural internal and dynamic changes are just that, natural. These cycles can't add energy into the system, they only redistribute what is added to it. Eventually a neutral and -AMO standard today is going to be yesteryears +AMO.

The AMO like the PDO is an oscillation overriding a long term temperature trend, but that doesn't mean that it doesn't have significant impacts on the climate.

A significant portion of this long term trend upward is due to solar effects.

[csnavywx]

No one is saying that the extra heat accumulated from carbon dioxide disappeared. What I'm questioning is how much heat is actually in the pipeline. I find 0.8 Degrees C in the pipeline to be really not that believable, considering we haven't warmed for many years now.

That would mean that there is a very potent forcing completely canceling out the warming impact from CO2 if there really is 0.8 Degrees C of warming in the pipeline. What do you propose is this negative forcing counteracting the CO2 forcing?

I strongly suspect a significant portion of the slowing (note that we are still gaining heat) is a large increase in tropospheric aerosol loading and SO2 forcing (via Chindia) from a doubling of world coal burning rates over the past 10-12 years (China alone has nearly tripled its burn rate). I think the combined effects of the direct and indirect aerosol forcing are powerful, having dealt with the effects of the "Asian Brown Cloud" phenomenon personally (and having to forecast it). Coal is an extremely dirty fuel. Be glad for the Clean Air Act here in the US; China and India really have no such regulation. As a result, their pollution problems are becoming a policy and public health nightmare.

Our real-time data on aerosols is limited to PARASOL, optical depth measurements from satellites, and ground observation trends. As a result, the uncertainty on the magnitude of negative forcing is large (ranges from ~-0.4 W/m2 to -2.7W/m2 (!)), and this is based on mid 2000s data. I haven't seen a good recent forcing estimate or analysis with data past 2005 or so though. We need a satellite targeted on aerosols in a bad way as this would help narrow down the sensitivity estimates considerably.

I think the 2.5 to 3.5 estimate is a good one, based on "fast feedbacks". The slower ones obviously skew the value to the right (higher).

[tacoman25]

I strongly suspect a significant portion of the slowing (note that we are still gaining heat) is a large increase in tropospheric aerosol loading and SO2 forcing (via Chindia) from a doubling of world coal burning rates over the past 10-12 years (China alone has nearly tripled its burn rate). I think the combined effects of the direct and indirect aerosol forcing are powerful, having dealt with the effects of the "Asian Brown Cloud" phenomenon personally (and having to forecast it). Coal is an extremely dirty fuel. Be glad for the Clean Air Act here in the US; China and India really have no such regulation. As a result, their pollution problems are becoming a policy and public health nightmare.

Our real-time data on aerosols is limited to PARASOL, optical depth measurements from satellites, and ground observation trends. As a result, the uncertainty on the magnitude of negative forcing is large (ranges from ~-0.4 W/m2 to -2.7W/m2 (!)), and this is based on mid 2000s data. I haven't seen a good recent forcing estimate or analysis with data past 2005 or so though. We need a satellite targeted on aerosols in a bad way as this would help narrow down the sensitivity estimates considerably.

I think the 2.5 to 3.5 estimate is a good one, based on "fast feedbacks". The slower ones obviously skew the value to the right (higher).

Given the strong historical correlation between global temp trends and the PDO (on top of the underlying warming trend), don't you think the recent switch to -PDO phase is playing a role?

[ORH_wxman]

The AMO like the PDO is an oscillation overriding a long term temperature trend, but that doesn't mean that it doesn't have significant impacts on the climate.

A significant portion of this long term trend upward is due to solar effects.

This is not scientifically backed..at least if you are talking about amounts more than about a tenth celsius. (a tenth itself can still be considered significant over 100 years, but I am assumming you mean more than this...correct me if I'm wrong)

ENSO was deifnitely a significant contributer of the late 20th century warming during that +PDO period.

[csnavywx]

I don't doubt that the PDO has at least a minor role to play. My main beef with it being a big player has always been the fact that the vast majority of the amplitude change (temp change) of the PDO during the last big warming phase stopped around 1990 and overall warming continued at the previous rate for a number of years past that date.

[ORH_wxman]

I don't doubt that the PDO has at least a minor role to play. My main beef with it being a big player has always been the fact that the vast majority of the amplitude change (temp change) of the PDO during the last big warming phase stopped around 1990 and overall warming continued at the previous rate for a number of years past that date.

The PDO+ peaked around 1990 but was still positive through the 1990s until a dip in the 1999-2001 La Nina before going mostly positive again 2002-2007, though to a lesser amplitude than the 1990s. Also, as the PDO was becoming "less positive" in the 1990s, the AMO increased out of the negative phase it had been in since the 1960s which likely helped offset the weakening positive PDO. In addition, most of the 1990s warming is because of the 1997-1998 Super El Nino...which was a temporary spike in the PDO back to stratospheric levels.

However, as long as the PDO was still predominately positive, then positive ENSO was being favored which of course increases global surface temperatures. Since the big PDO decline event in 2007, negative ENSO has been the favored phase and looks to to continue to be the favored phase in the near future.

The global temperature trend from the beginning of 2001 through August 2012 on GISS is 0.00C....dead flat. Most of the reason it is flat is because of the fall in global temps since the PDO shift in 2007. Temps were still rising before that, though at a slower rate than what they had been in the 1980s/1990s. Could this be due to the weakened +PDO state in the 2001-2007 time frame and then the flip to predominately negative in 2007 and beyond?

The link between ENSO and global temperatures is unmistakable on the data we have, and since the link between the PDO and preferred ENSO state is very significant, its likely the PDO is a significant contributor to global temperatures.

[The_Global_Warmer]

I don't doubt that the PDO has at least a minor role to play. My main beef with it being a big player has always been the fact that the vast majority of the amplitude change (temp change) of the PDO during the last big warming phase stopped around 1990 and overall warming continued at the previous rate for a number of years past that date.

When we take in solar variance( the longer term drop in TSI and the small rise since 2010, the slowing of OHC, the slowing of global SST rise, the higher frequency of Ninas GHG warming is still strong and certainly stronger than it has been up to this point.

I think those graphs that take out natural variance do a good job. There is a good chance 1998's peak might hold for another decade, but when we account for ENSO below and the PDO below

PDO:

2010** 0.83 0.82 0.44 0.78 0.62 -0.22 -1.05 -1.27 -1.61 -1.06 -0.82 -1.21

2011** -0.92 -0.83 -0.69 -0.42 -0.37 -0.69 -1.86 -1.74 -1.79 -1.34 -2.33 -1.79

2012** -1.38 -0.85 -1.05 -0.27 -1.26 -0.87 -1.52 -1.93

GLOBAL SST ANOMALIES FROM JAN 3RD 2010-SEPT 12TH 2012.

The Lower Troposphere is still warming and in the most recent two cool drops required strong NIna's during the NH winter to spend a couple months below normal, with the PDO running very cold.

2012 is going to be a neutral ENSO at best unless it warms up the next 3 months and we are not cooling at all. If 2013 starts out with a neutral positive ENSO and stays that way I think it will be close to 2010, with a mild NINO we pass 2010, with a strong one we might jump pretty high and beat 1998 because most of the year will sit at high anomalies instead of a huge peak.

If the PDO has a 3-4 year positive flip during this cold period we might jump pretty good upwards. When the PDO goes positive or is more neutral for a while we will jump up because GHG forcing is stronger and stronger.

The AMO doesn't fit to well in global temps on a seasonal scale.

2008 was cold during an ENSO that was very negative and 2009 was quite a bit warmer, but the 2008 AMO was higher. 2008 was higher than 2011, 2011 was WAY WARMER THAN 2008. 2010 was very warm but so was 2005 and 2012 the last few months but the results are not the same.

2002 0.206 0.194 0.176 0.059 -0.020 -0.086 -0.029 0.143 0.118 0.148 0.052 0.034

2003 0.080 0.013 0.135 0.104 0.181 0.239 0.319 0.457 0.487 0.460 0.256 0.255

2004 0.238 0.237 0.185 0.139 0.026 0.205 0.263 0.350 0.271 0.276 0.260 0.221

2005 0.143 0.153 0.312 0.324 0.319 0.357 0.485 0.475 0.454 0.270 0.166 0.245

2006 0.151 0.099 0.085 0.228 0.336 0.367 0.413 0.445 0.403 0.369 0.311 0.198

2007 0.198 0.243 0.156 0.190 0.140 0.122 0.168 0.093 0.134 0.195 0.210 0.139

2008 0.057 0.155 0.191 0.077 0.199 0.293 0.243 0.207 0.234 0.135 0.037 0.055

2009 -0.026 -0.131 -0.133 -0.097 -0.033 0.158 0.265 0.188 0.092 0.204 0.104 0.118

2010 0.074 0.207 0.319 0.463 0.492 0.482 0.488 0.565 0.487 0.363 0.275 0.246

2011 0.178 0.140 0.088 0.125 0.178 0.212 0.132 0.186 0.189 0.100 -0.039 -0.014

2012 -0.035 0.034 0.054 0.115 0.197 0.338 0.418 0.474 -99.990 -99.990 -99.990 -99.990

Outside of ENSO, it's rough. The AMO is sst's which is real time, so is ENSO. but ENSO variants much warmer of colder like the PDO.

So I would attribute the AMO to 20-25% the impact on global temps compared to the PDO.

[PottercountyWXobserver]

[PhillipS]

Question for the people who selected 2.5-4.5 Degrees C:

Assuming the mean sensitivity is close to 3.5 Degrees C per doubling, then we should have observed a 1.4 Degree C temperature increase over the 20th Century when CO2 levels rose sharply, assuming all of the warming in the 20th Century was anthropogenic (which is wrong). Instead, we only observe a 0.6 Degree C increase.

Are you saying that there is 0.8 Degrees C in the pipeline for warming?

I'm not sure where you got the value of 0.8 C warming for the 20th century. Or why you would even limit the time period to just the 20th century. The increase in CO2 is generally referenced to the pre-industrial value of 280 ppm (from ice cores) so let's look at the temperature rise over as much of the period since the Industrial Revolution began as we have records for. Here is the long-term temperature record from the BEST project:

According to the BEST analysis the Earth has warmed about 1.4 C in the period that CO2 has risen from 280 ppm to today's 392 ppm (a 40% increase). A nice benefit of using longer periods is that quasi-periodic fluctuations, such as ENSO, tend to average out and have less effect on the underlying trend.

The poll was for the Equilibrium Sensitivity (ES) for a doubling of CO2 - which, of course, is the combination of fast-response and slow-response thermal processes. The fast response processes include temperature changes to air, land surface, and sea surface. Slow-response processes include ocean heat content, ice sheet mass changes, and permafrost melting - each of which can take centuries to reach equilibrium. That is what is meant by the phrase 'warming in the pipeline' - the slow-response processes haven't had time to fully react to the continuing changes in CO2 concentration. So most of the warming seen to date has been fast-response processes.

Just for the sake of determining an upper bound to ES I'll assume that natural variability has averaged out over the temperature record and all of the rise is anthropogenic. A 1.4 C increase in temperature for a 40% increase in CO2 works out to a fast-response sensitivity of 3.5 C for a doubling of CO2. When the slow-response processes are included the total ES would be higher. Depending on the ratio between fast and slow processes the upper bound for ES could be more than 5 C.

But, hopefully, not all of the measured 1.4 C temperature increase is anthropogenic.so the actual ES should be lower than 5 C. Assuming half of that 1.4 C is natural, the fast-response sensitivity would be about 1.7 C (0.7 / 0.4) and the overall ES some value higher. Say, around 3 C.

The difficulty of assuming a split like that between natural and anthropogenic warming is the need to identify a natural process that could have warmed the Earth that much since 1800. It can't be the 'usual culprits' like ENSO, PDO, or the Sun because none of those has shown trends of the necessary length and magnitude - so it would have to be something else. Let's call it Process X. However, if Process X has been warming the Earth for the past 200+ years - why would it stop now? What is to keep Process X from continuing to warm the Earth in synch with AGW? If it does then the warming we can expect to have to deal with is the Equilibrium Sensitivity to a doubling of CO2 plus the additional natural warming due to Process X. Alarming!

[ORH_wxman]

I'm not sure where you got the value of 0.8 C warming for the 20th century. Or why you would even limit the time period to just the 20th century. The increase in CO2 is generally referenced to the pre-industrial value of 280 ppm (from ice cores) so let's look at the temperature rise over as much of the period since the Industrial Revolution began as we have records for. Here is the long-term temperature record from the BEST project:

According to the BEST analysis the Earth has warmed about 1.4 C in the period that CO2 has risen from 280 ppm to today's 392 ppm (a 40% increase). A nice benefit of using longer periods is that quasi-periodic fluctuations, such as ENSO, tend to average out and have less effect on the underlying trend.

The poll was for the Equilibrium Sensitivity (ES) for a doubling of CO2 - which, of course, is the combination of fast-response and slow-response thermal processes. The fast response processes include temperature changes to air, land surface, and sea surface. Slow-response processes include ocean heat content, ice sheet mass changes, and permafrost melting - each of which can take centuries to reach equilibrium. That is what is meant by the phrase 'warming in the pipeline' - the slow-response processes haven't had time to fully react to the continuing changes in CO2 concentration. So most of the warming seen to date has been fast-response processes.

Just for the sake of determining an upper bound to ES I'll assume that natural variability has averaged out over the temperature record and all of the rise is anthropogenic. A 1.4 C increase in temperature for a 40% increase in CO2 works out to a fast-response sensitivity of 3.5 C for a doubling of CO2. When the slow-response processes are included the total ES would be higher. Depending on the ratio between fast and slow processes the upper bound for ES could be more than 5 C.

But, hopefully, not all of the measured 1.4 C temperature increase is anthropogenic.so the actual ES should be lower than 5 C. Assuming half of that 1.4 C is natural, the fast-response sensitivity would be about 1.7 C (0.7 / 0.4) and the overall ES some value higher. Say, around 3 C.

The difficulty of assuming a split like that between natural and anthropogenic warming is the need to identify a natural process that could have warmed the Earth that much since 1800. It can't be the 'usual culprits' like ENSO, PDO, or the Sun because none of those has shown trends of the necessary length and magnitude - so it would have to be something else. Let's call it Process X. However, if Process X has been warming the Earth for the past 200+ years - why would it stop now? What is to keep Process X from continuing to warm the Earth in synch with AGW? If it does then the warming we can expect to have to deal with is the Equilibrium Sensitivity to a doubling of CO2 plus the additional natural warming due to Process X. Alarming!

Your temp graph is only for land. Snowlover's 0.8C increase in temps since the industrial revolution began is accurate using both GISS and HadCRUT4. Both are in agreement that we have risen at 0.06C per decade since 1880 at the beginning of their records.

[Vergent]

Your temp graph is only for land. Snowlover's 0.8C increase in temps since the industrial revolution began is accurate using both GISS and HadCRUT4. Both are in agreement that we have risen at 0.06C per decade since 1880 at the beginning of their records.

Land temps are the valid metric. The oceans are a big heat sink with kilo-year time constant. You might as well argue that the last km^3 of ice in glacier national park has not changed temperature in the last 200 years.;>)

[Entropy]

The climate sensitivity of a doubling of atmospheric CO2 is from 2 to 4.5K, with a best guess of around 3K. However, this takes into account only fast feedbacks, such as changes in water vapor, sea ice, clouds, and aerosols. At these levels, the earth's climate system would still be in flux to some extent due to changes in ice sheet coverage, ecosystem types, and release of additional GHGs. Based on the work of Pagani et al. (2009) and Hansen et al. (2008), I'd estimate the actual, equilibrium sensitivity to be around 6K.

[PhillipS]

Your temp graph is only for land. Snowlover's 0.8C increase in temps since the industrial revolution began is accurate using both GISS and HadCRUT4. Both are in agreement that we have risen at 0.06C per decade since 1880 at the beginning of their records.

I should have been clearer in my post but it seemed to be getting long-winded and I was trying to be concise. Sorry about that.

I understand that the BEST long-term temperature record is only for land - but it is the most relevant metric for assessing the sensitivity of the fast-response processes to a doubling of CO2 since land temps respond quickly to changes in energy balance and the BEST record covers most of the period in question. Assessing fast responses, the observed record, was the point of my post. If you prefer to use a global temperature metric, such as GISS, that's fine,of course, just understand that you're mixing fast, medium and slow response processes since sea surface temps respond slower than land and are coupled to the massive thermal inertia of the deep oceans.

** Edited to correct my error **

But now that I understand where the 0.6 C temperature trend for the 20th century came from I'll run the assessment again. The CO2 concentration in 1900 was 296 ppm [source] and in 1999 was 368 ppm [source] so the 20th century increase in CO2 was 24% (368 / 296). As before, assuming the entire warming was anthropogenic to establish an upper bound for equilibrium sensitivity we get a mostly fast response of 2.5 C (0.6 / 0.24) for a doubling of CO2. And since the Earth hasn't reached equilibrium the total ES would be higher. This value is in good agreement with my earlier assessment of an upper bound around 5 C per doubling.

Now, assuming that half of the observed 20th century increase is natural and half anthropogenic, we get a mostly fast response sensitivity of 1.3 C (0.3 / 0.24) per doubling of CO2. Since the ES will be higher, even if we assume half the warming observed to date is natural then we are still looking at an equilibrium sensitivity of around 2.5 - 3.5 C per doubling of CO2.

As I pointed out above, if the claim is made that much of the observed warming is natural then the questions of what caused the natural warming, how long will the natural waming continue, and what will be the magnitude of the natural warming need to be answered. Trying to brush the questions aside by invoking 'natural variability' or a hypothetical solar amplification is rhetorical handwaving with no explanatory value. One might as well claim the non-anthropogenic portion of the observed warming was due to leprechauns and unicorns. If skeptics want to shift much of the observed warming away of anthropogenic causes (and remember that CO2 is only one of several ways we're changing the climate) then they have the burden of proof for that claim.

[ORH_wxman]

Now, assuming that half of the observed 20th century increase is natural and half anthropogenic, we get a mostly fast response sensitivity of 1.7 C (0.4 / 0.24) per doubling of CO2. Since the ES will be higher, even if we assume half the warming observed to date is natural then we are still looking at an equilibrium sensitivity of around 2.5 - 3.5 C per doubling of CO2.

This is relatively close with some of the more recent papers on climate sensitivity versus the higher estimate of 4C and beyond.

If you are going to be specific on the 1900 to 2000 time frame rather than 1880-present, then the temperature rise is closer to 0.6C...its more on the order of 0.77C (or rounded to 0.8C) if you use the entire period.

This is of course using GISS and HadCRUT4.

[Vergent]

I should have been clearer in my post but it seemed to be getting long-winded and I was trying to be concise. Sorry about that.

I understand that the BEST long-term temperature record is only for land - but it is the most relevant metric for assessing the sensitivity of the fast-response processes to a doubling of CO2 since land temps respond quickly to changes in energy balance and the BEST record covers most of the period in question. Assessing fast responses, the observed record, was the point of my post. If you prefer to use a global temperature metric, such as GISS, that's fine,of course, just understand that you're mixing fast, medium and slow response processes since sea surface temps respond slower than land and are coupled to the massive thermal inertia of the deep oceans.

But now that I understand where the 0.8 C temperature trend for the 20th century came from I'll run the assessment again. The CO2 concentration in 1900 was 296 ppm [source] and in 1999 was 368 ppm [source] so the 20th century increase in CO2 was 24% (368 / 296). As before, assuming the entire warming was anthropogenic to establish an upper bound for equilibrium sensitivity we get a mostly fast response of 3.3 C (0.8 / 0.24) for a doubling of CO2. And since the Earth hasn't reached equilibrium the total ES would be higher. This value is in good agreement with my earlier assessment of an upper bound around 5 C per doubling.

Now, assuming that half of the observed 20th century increase is natural and half anthropogenic, we get a mostly fast response sensitivity of 1.7 C (0.4 / 0.24) per doubling of CO2. Since the ES will be higher, even if we assume half the warming observed to date is natural then we are still looking at an equilibrium sensitivity of around 2.5 - 3.5 C per doubling of CO2.

As I pointed out above, if the claim is made that much of the observed warming is natural then the questions of what caused the natural warming, how long will the natural waming continue, and what will be the magnitude of the natural warming need to be answered. Trying to brush the questions aside by invoking 'natural variability' or a hypothetical solar amplification is rhetorical handwaving with no explanatory value. One might as well claim the non-anthropogenic portion of the observed warming was due to leprechauns and unicorns. If skeptics want to shift much of the observed warming away of anthropogenic causes (and remember that CO2 is only one of several ways we're changing the climate) then they have the burden of proof for that claim.

Land temperatures are even deceptively low for sensitivity. They include permafrost, snow cover, and glacier areas where summer temperatures will not change until the ice is gone. The melting point of water is not an issue.

[PhillipS]

This is relatively close with some of the more recent papers on climate sensitivity versus the higher estimate of 4C and beyond.

If you are going to be specific on the 1900 to 2000 time frame rather than 1880-present, then the temperature rise is closer to 0.6C...its more on the order of 0.77C (or rounded to 0.8C) if you use the entire period.

This is of course using GISS and HadCRUT4.

I understand what you're saying but I used the value that Snowlover used in his earlier post - 0.8 C for the 20th century. It will be interesting to see if he wants to move the goalposts and change his value now. **Edit - my mistake - I took the wrong value from Snowlovers' post above. He did use 0.6 C for the observed 20th century warming. The error is entirely mine and I'll change the assessment to use his figure. **

Ultimately it doesn't make much difference to the value of ES. The observed warming over the instrumental period is too great to support ES values below about 2 C.

And I am interested in your take on the questions I raised about the non-anthropogenic component to the observed warming. The paleoclimate trend for the last 8K years or so has been a gradual natural cooling of about 0.25 C per millennium (0.0025 C / decade) IIRC . Agree? Any claim that for the most recent 200 years there has been a natural warming trend raises questions have have to be rigorously answered. Among them:

• When did natural climate processes change from cooling to warming? Even if one believes the LIA was a global rather than a NH phenomenon, the recovery from that cool period was finished long ago. Why would the Earth be naturally warming today?
  
• What triggered the change from natural cooling to natural warming? The answer to this could involve multiple factors but they need to be identified and their durations and magnitudes need supporting data.
  
• What is the magnitude of the natural warming? How would you split the observed warming into anthropogenic and natural (non-anthropogenic) components?
  
• How long will the natural warming continue? The long-term temperature record shows that the Earth has been warming at least since 1800 - more than two centuries. If we are facing another two centuries of continued natural warming on top of AGW then the problems we face are probably serious. If, on the other hand, the claim is that natural warming is over, or will be soon, that just raises additional questions of why did (will) natural warming stop and how can we predict whether it will begin again at some point in the future.
  

Because the answers to these questions affect any assessment of equilibrium sensitivity they are on topic, but I don't mean to clog up this discussion so I may start a new thread on natural global warming.

[ORH_wxman]

• When did natural climate processes change from cooling to warming? Even if one believes the LIA was a global rather than a NH phenomenon, the recovery from that cool period was finished long ago. Why would the Earth be naturally warming today?
  
• What triggered the change from natural cooling to natural warming? The answer to this could involve multiple factors but they need to be identified and their durations and magnitudes need supporting data.
  
• What is the magnitude of the natural warming? How would you split the observed warming into anthropogenic and natural (non-anthropogenic) components?
  
• How long will the natural warming continue? The long-term temperature record shows that the Earth has been warming at least since 1800 - more than two centuries.
  

1. Likely in the 18th century sometime after the Maunder Minimum. This however, does not mean that there isn't a longer term natural cooling trend on the scale of a millenium, but on the shorter century scale, the natural variation likely stopped favoring significant cooling at that point. We did see another more brief period of cooling with the Dalton min along with smaller temporary (but severe) coolings with both the Tambora and the Laki eruptions.

2. The changes were likely caused by a combination of solar, volcanic, and ocean variability. The cooling wasn't continuous since the Holocene Max. The climate likely warmed significantly from the period around 300 AD to the peak ofthe medieval warm period near 1,000 AD, and one would assume that other smaller scale warmings occurred in the previous 7,000 years within the gradual cooling period.

3. I think #3 is probably the most difficult to answer right now. The IPCC currently thinks that roughly half of the warming in our modern temperature record (about 1880-present) is natural. Some of it due to solar and some of it due to lowered volcanic activity. There are likely some other internal oscillations that contribute to some level, but we just aren't able to accurately identify and quantify them due to our limited period of modern weather data.

4. I don't think we know this either. Solar is down which favors cooling. Volcanic activity is still low which favors warming. The shorter term PDO cycle favors cooling on a decadal scale for the time being, but we still are not certain about longer term ocean cycles or whether solar variability would be more enhanced or not if there were a period of low maximums (thus far I don't think the evidence has shown to support this). Underlying all this, we should still be on a slow longer term cooling trend on our way to the next glacial maximum due to orbital variation.

Obviously answering these questions accurately would increase our ability to calculate climate sensitivity more precisely than the current ranges.

[Entropy]

I think what some are considering is radiative forcing (measured in W/m²) and not temperature sensitivity (measured in K/(W/m²)). Temperature sensitivity is the change in temperature caused by a change in radiative forcing. A temperature sensitivity of zero is physically impossible, as it would imply the earth's temperature is static.

I consider the term climate sensitivity, which is widely used in the literature, to be a bit of a misnomer. The property described is actually temperature sensitivity, and not climate sensitivity. It only tells us how an esoteric value, such as a globally averaged temperature, is likely to respond to a change in forcing. Its tells us nothing about how the entirety of the earth's climate system will respond to that change in temperature. Schmittner et al. (2011), for instance, found a lower sensitivity than other analyses (about 2.3 K/(W/m²)). However, their model also assumed global temperatures during the last glacial maximum were only 2.6K lower than during the Holocene interglacial. So, if you think of the implications of this, its not really a good thing. Yes, the temperature rise might be less, but the effects are still the same. If a 2.6K change in global temperature was enough to mean the difference between the northern US being buried under thousands of feet of snow and ice and the present climate, a low sensitivity is hardly comforting.

[Snow_Miser]

This is not scientifically backed..at least if you are talking about amounts more than about a tenth celsius. (a tenth itself can still be considered significant over 100 years, but I am assumming you mean more than this...correct me if I'm wrong)

ENSO was deifnitely a significant contributer of the late 20th century warming during that +PDO period.

Statistically, the sun can account for 80% of the temperature increase, though the author of this paper neglects to acknowledge the amplifying forcing that has been observed over the course of a solar cycle, and concludes that even though statistically the sun can explain most of the temperature change, the effects from TSI are too small, so the sun only accounts for 20% of the increase. This is still a significant contribution, but the author neglected to consider the amplifying factors from the sun, and got lower numbers than he should have gotten for the contribution of the sun to the warming trend over the last 100 years.

[Snow_Miser]

I'm not sure where you got the value of 0.8 C warming for the 20th century. Or why you would even limit the time period to just the 20th century. The increase in CO2 is generally referenced to the pre-industrial value of 280 ppm (from ice cores) so let's look at the temperature rise over as much of the period since the Industrial Revolution began as we have records for. Here is the long-term temperature record from the BEST project:

According to the BEST analysis the Earth has warmed about 1.4 C in the period that CO2 has risen from 280 ppm to today's 392 ppm (a 40% increase). A nice benefit of using longer periods is that quasi-periodic fluctuations, such as ENSO, tend to average out and have less effect on the underlying trend.

The poll was for the Equilibrium Sensitivity (ES) for a doubling of CO2 - which, of course, is the combination of fast-response and slow-response thermal processes. The fast response processes include temperature changes to air, land surface, and sea surface. Slow-response processes include ocean heat content, ice sheet mass changes, and permafrost melting - each of which can take centuries to reach equilibrium. That is what is meant by the phrase 'warming in the pipeline' - the slow-response processes haven't had time to fully react to the continuing changes in CO2 concentration. So most of the warming seen to date has been fast-response processes.

Just for the sake of determining an upper bound to ES I'll assume that natural variability has averaged out over the temperature record and all of the rise is anthropogenic. A 1.4 C increase in temperature for a 40% increase in CO2 works out to a fast-response sensitivity of 3.5 C for a doubling of CO2. When the slow-response processes are included the total ES would be higher. Depending on the ratio between fast and slow processes the upper bound for ES could be more than 5 C.

But, hopefully, not all of the measured 1.4 C temperature increase is anthropogenic.so the actual ES should be lower than 5 C. Assuming half of that 1.4 C is natural, the fast-response sensitivity would be about 1.7 C (0.7 / 0.4) and the overall ES some value higher. Say, around 3 C.

The difficulty of assuming a split like that between natural and anthropogenic warming is the need to identify a natural process that could have warmed the Earth that much since 1800. It can't be the 'usual culprits' like ENSO, PDO, or the Sun because none of those has shown trends of the necessary length and magnitude - so it would have to be something else. Let's call it Process X. However, if Process X has been warming the Earth for the past 200+ years - why would it stop now? What is to keep Process X from continuing to warm the Earth in synch with AGW? If it does then the warming we can expect to have to deal with is the Equilibrium Sensitivity to a doubling of CO2 plus the additional natural warming due to Process X. Alarming!

You are using a preliminary dataset for the land only, so you will naturally get larger warming rates for the land only.

[Snow_Miser]

I strongly suspect a significant portion of the slowing (note that we are still gaining heat) is a large increase in tropospheric aerosol loading and SO2 forcing (via Chindia) from a doubling of world coal burning rates over the past 10-12 years (China alone has nearly tripled its burn rate). I think the combined effects of the direct and indirect aerosol forcing are powerful, having dealt with the effects of the "Asian Brown Cloud" phenomenon personally (and having to forecast it). Coal is an extremely dirty fuel. Be glad for the Clean Air Act here in the US; China and India really have no such regulation. As a result, their pollution problems are becoming a policy and public health nightmare.

Our real-time data on aerosols is limited to PARASOL, optical depth measurements from satellites, and ground observation trends. As a result, the uncertainty on the magnitude of negative forcing is large (ranges from ~-0.4 W/m2 to -2.7W/m2 (!)), and this is based on mid 2000s data. I haven't seen a good recent forcing estimate or analysis with data past 2005 or so though. We need a satellite targeted on aerosols in a bad way as this would help narrow down the sensitivity estimates considerably.

I think the 2.5 to 3.5 estimate is a good one, based on "fast feedbacks". The slower ones obviously skew the value to the right (higher).

I highly doubt that aerosols are driving the recent hiatus in temperatures, considering that the net forcing from SO2 remained largely unchanged when the decadal temperature trends significantly changed. This can be seen in a figure from Kaufmann et al. 2011.

Note that there is no change in the SO2 forcing (purple line) during the hiatus.

[Snow_Miser]

I think what some are considering is radiative forcing (measured in W/m²) and not temperature sensitivity (measured in K/(W/m²)). Temperature sensitivity is the change in temperature caused by a change in radiative forcing. A temperature sensitivity of zero is physically impossible, as it would imply the earth's temperature is static.

I consider the term climate sensitivity, which is widely used in the literature, to be a bit of a misnomer. The property described is actually temperature sensitivity, and not climate sensitivity. It only tells us how an esoteric value, such as a globally averaged temperature, is likely to respond to a change in forcing. Its tells us nothing about how the entirety of the earth's climate system will respond to that change in temperature. Schmittner et al. (2011), for instance, found a lower sensitivity than other analyses (about 2.3 K/(W/m²)). However, their model also assumed global temperatures during the last glacial maximum were only 2.6K lower than during the Holocene interglacial. So, if you think of the implications of this, its not really a good thing. Yes, the temperature rise might be less, but the effects are still the same. If a 2.6K change in global temperature was enough to mean the difference between the northern US being buried under thousands of feet of snow and ice and the present climate, a low sensitivity is hardly comforting.

The IPCC defines Climate Sensitivity as the temperature increase that we can expect with a doubling of CO2. Schmitter et al. did not find a sensitivity of 2.3 K/w/m^2. They found a sensitivity of 2.3 K/3.7 w/m^2.

I think it is still too early to determine what the exact sensitivity is, but when you account for the indirect solar effects on the climate, the sensitivity is greatly reduced to that of a black body or less.

[csnavywx]

The IPCC defines Climate Sensitivity as the temperature increase that we can expect with a doubling of CO2. Schmitter et al. did not find a sensitivity of 2.3 K/w/m^2. They found a sensitivity of 2.3 K/3.7 w/m^2.

I think it is still too early to determine what the exact sensitivity is, but when you account for the indirect solar effects on the climate, the sensitivity is greatly reduced to that of a black body or less.

And you continue to post this nonsense. You declare "it's too early to determine what the exact sensitivity is", which is true, but then stretch out a time-tested minor forcing to claim this? Give me a break. There's a mountain of research pointing against this notion.