By Patrick J. Michaels and Paul C. “Chip” Knappenberger
We have two new entries to the long (and growing) list of papers appearing the in recent scientific literature that argue that the earth’s climate sensitivity—the ultimate rise in the earth’s average surface temperature from a doubling of the atmospheric carbon dioxide content—is close to 2°C, or near the low end of the range of possible values presented by the U.N.’s Intergovernmental Panel on Climate Change (IPCC). With a low-end warming comes low-end impacts and an overall lack of urgency for federal rules and regulations (such as those outlined in the President’s Climate Action Plan) to limit carbon dioxide emissions and limit our energy choices.
The first is the result of a research effort conducted by Craig Loehle and published in the journal Ecological Modelling. The paper is a pretty straightforward determination of the climate sensitivity. Loehle first uses a model of natural modulations to remove the influence of natural variability (such as solar activity and ocean circulation cycles) from the observed temperature history since 1850. The linear trend in the post-1950 residuals from Loehle’s natural variability model was then assumed to be largely the result, in net, of human carbon dioxide emissions. By dividing the total temperature change (as indicated by the best-fit linear trend) by the observed rise in atmospheric carbon dioxide content, and then applying that relationship to a doubling of the carbon dioxide content, Loehle arrives at an estimate of the earth’s transient climate sensitivity—transient, in the sense that at the time of CO2 doubling, the earth has yet to reach a state of equilibrium and some warming is still to come.
Loehle estimated the equilibrium climate sensitivity from his transient calculation based on the average transient:equilibrium ratio projected by the collection of climate models used in the IPCC’s most recent Assessment Report. In doing so, he arrived at an equilibrium climate sensitivity estimate of 1.99°C with a 95% confidence range of it being between 1.75°C and 2.23°C.
Compare Loehle’s estimate to the IPCC’s latest assessment of the earth’s equilibrium climate sensitivity which assigns a 66 percent or greater likelihood that it lies somewhere in the range from 1.5°C to 4.5°C. Loehle’s determination is more precise and decidedly towards the low end of the range.
The second entry to our list of low climate sensitivity estimates comes from Roy Spencer and William Braswell and published in the Asia-Pacific Journal of Atmospheric Sciences. Spencer and Braswell used a very simple climate model to simulate the global temperature variations averaged over the top 2000 meters of the global ocean during the period 1955-2011. They first ran the simulation using only volcanic and anthropogenic influences on the climate. They ran the simulation again adding a simple take on the natural variability contributed by the El Niño/La Niña process. And they ran the simulation a final time adding in a more complex situation involving a feedback from El Niño/La Niña onto natural cloud characteristics. They then compared their model results with the set of real-world observations.
What the found, was the that the complex situation involving El Niño/La Niña feedbacks onto cloud properties produced the best match to the observations. And this situation also produced the lowest estimate for the earth’s climate sensitivity to carbon dioxide emissions—a value of 1.3°C.
Spencer and Braswell freely admit that using their simple model is just the first step in a complicated diagnosis, but also point out that the results from simple models provide insight that should help guide the development of more complex models, and ultimately could help unravel some of the mystery as to why full climate models produce high estimates of the earth’s equilibrium climate sensitivity, while estimates based in real-world observations are much lower.
Our Figure below helps to illustrate the discrepancy between climate model estimates and real-world estimates of the earth’s equilibrium climate sensitivity. It shows Loehle’s determination as well as that of Spencer and Braswell along with 16 other estimates reported in the scientific literature, beginning in 2011. Also included in our Figure is both the IPCC’s latest assessment of the literature as well as the characteristics of the equilibrium climate sensitivity from the collection of climate models that the IPCC uses to base its impacts assessment.
Figure 1. Climate sensitivity estimates from new research beginning in 2011 (colored), compared with the assessed range given in the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (AR5) and the collection of climate models used in the IPCC AR5. The “likely” (greater than a 66% likelihood of occurrence)range in the IPCC Assessment is indicated by the gray bar. The arrows indicate the 5 to 95 percent confidence bounds for each estimate along with the best estimate (median of each probability density function; or the mean of multiple estimates; colored vertical line). Ring et al. (2012) present four estimates of the climate sensitivity and the red box encompasses those estimates. The right-hand side of the IPCC AR5 range is actually the 90% upper bound (the IPCC does not actually state the value for the upper 95 percent confidence bound of their estimate). Spencer and Braswell (2013) produce a single ECS value best-matched to ocean heat content observations and internal radiative forcing.
Quite obviously, the IPCC is rapidly losing is credibility.
As a result, the Obama Administration would do better to come to grips with this fact and stop deferring to the IPCC findings when trying to justify increasingly burdensome federal regulation of carbon dioxide emissions, with the combined effects of manipulating markets and restricting energy choices.
References:
Loehle, C., 2014. A minimal model for estimating climate sensitivity. Ecological Modelling, 276, 80-84.
Spencer, R.W., and W. D. Braswell, 2013. The role of ENSO in global ocean temperature changes during 1955-2011 simulated with a 1D climate model. Asia-Pacific Journal of Atmospheric Sciences, doi:10.1007/s13143-014-0011-z.
=========================================================
Global Science Report is a feature from the Center for the Study of Science, where we highlight one or two important new items in the scientific literature or the popular media. For broader and more technical perspectives, consult our monthly “Current Wisdom.”
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
All this talk about climate sensitivity is playing their game. We need to stop and make them explain why for the past seventeen years the sensitivity has been zero.
Reprint is the top one listed here:
http://www.ncasi.org/Search-Results.aspx?q=loehle%20sensitivity
Did I read this correctly. They try to remove natural variation including solar activity to leave a modelled projection!! I’m sorry, but that seems to be an absurd approach to counter the CAGW crowd, and as JA says, playing into their hands.
The basic question is, why is there any belief, i.e. where is the evidence, that climate sensitivity to man’s CO2 is anything other than zero. And we should be perfectly clear about this; it is the ~3% man’s CO2 additions that is the subject of the debate, not the 97% of natural CO2.
I support government intervention to make the vehicle fleet more efficient and encourage the use of natural gas. These steps reduce dependence on foreign sources of energy and at the same time satisfy the apparent need to reduce greenhouse gas emissions. We should factor in the fact that if indeed the climate sensitivity is about 2 degrees C to doubling, the atmosphere is likely to have twice the CO2 content, and temperature will rise. And if it does, it´s evident sea level will increase and cause trouble. So the best solution is to cut emissions doing so in a prudent fashion and reducing the balance of trade déficit.
“Loehle first uses a model of natural modulations to remove the influence of natural variability (such as solar activity and ocean circulation cycles) from the observed temperature history since 1850.”
It is good to see this issue of low climate sensitivity getting another airing, but do we really know what the values of all the natural variability is, in order to remove it from the calculations in the first place. I suspect we don’t.
Keith Gordon
I’d like to see some of the older IPCC claims to be included on the chart as well. I think that’s important.
What is the current estimate of sensitivity based on data alone? In other words, look at the temperatures from 1900 to 2014, and fit to CO2. The math would end up using 1900 to about 1960 as baseline, then any increase above the expected rise would be attributed to CO2. (It might not be CO2, but it might be).
Such an estimate will be available with smallish error bars in 2100, but what is the estimate now?
It looks like we may be able to start ruling out climate sensitivities above 4C even today, just from data. As the decades run on, we get better data.
Settled to a scientist should mean predictions match actual data. There is basically no data yet.
But it’s hard to tax (cough) extort (cough) trillions of dollars out of the public if you can’t panic them….
It is somewhat disconcerting that the several low-end estimates have error bars that do not embrace the estimates. This may simply show that the models do not ‘know’ how to get the error bars.
Loehle’s estimate is something like what I am talking about, but he does a lot of data munging that could be hard to follow. I’m talking about a straight fit with only a few parameters – for instance not assuming that the lines have anything to do with climate.
Craig,
Wouldn’t using a straight transient to equilibrium sensitivity factor overestimate the equilibrium sensitivity? After all, only recent CO2 rises give truly transient warming, while older CO2 rises should include most of the equilibrium effect. I think takiing this into account would cause a transient sensitivity of 1.1C (which you found) to give an equilibrium sensitivity closer to 1.6 – 1.7 C. Is this correct or am I missing something?
“Loehle first uses a model of natural modulations to remove the influence of natural variability (such as solar activity and ocean circulation cycles) from the observed temperature history since 1850. The linear trend in the post-1950 residuals from Loehle’s natural variability model was then assumed to be largely the result, in net, of human carbon dioxide emissions. ”
They should also remove or identify land use changes to further refine residuals. The focus is usually restricted to UHI and for some reason dismissed. But the change in land use should be extended to include agriculture, because open fields that are well drained are warmer or hotter even than what was there before. If anyone is familiar with London then this IR image clearly shows a large warm rectangle with a cool blob within. The cool blob is a lake in the Fairlop Waters Country park and the large yellow block on the right hand side is farmland. It is not what I expected. Notice that trees and water are ‘cool’.
http://wattsupwiththat.files.wordpress.com/2011/04/3_left_london_uhi1.jpg
Therefore the effect of CO2 is even less.
Gents, if there is in fact a thermostat negative forcing that counters warming (or cooling in the case of volcanic aerosols), what then can we say about CO2 doubling climate sensitivity and if it matters. In such a case it would have only theoretical interest as an effect of CO2 doubling, all other things remaining the same.
Personal speaking, I’m with the groundhog 😉
Dr Torch
Yes add in James Hansen’s 6 degrees centigrade estimates
Brian: in the discussion I point out exactly what you suggest, that conditions suggest a lower equilibrium response that I calculated.
My approach is very simple. No-one has measured a CO2 signal in any modern temperature/time graph, so there is a strong indication that the climate sensitivity for CO2 added to the atmosphere from recent levels is 0.0 C to one place of decimals or to significant figures.
When we really know, 15 or 20 yrs out, I bet it will be between 1.1 and .8.
I expect it to be about 1.1 now, and it will fall as temperatures rise.
“All this talk about climate sensitivity is playing their game. We need to stop and make them explain why for the past seventeen years the sensitivity has been zero.”
****
You’re wrong here. Sensitivity is the whole ball game. Moreover, sensitivity to Co2 is theoretically a constant. A period of no warming doesn’t mean that atmospheric sensitivity is zero, although it certainly argues for lower sensitivity given the ongoing rise in anthro Co2 during that time.
that the earth’s climate sensitivity is close to 2°C…
Then we are already there…so pack up, go home, and stop playing their game
Fernando Leanme says:
“…We should factor in the fact that if indeed the climate sensitivity is about 2 degrees C to doubling, the atmosphere is likely to have twice the CO2 content, and temperature will rise. And if it does, it´s evident sea level will increase and cause trouble….”
Even ignoring the fact that there is no data supporting a direct link between global temperature and CO2 (along with no explanation for the last 17 years), in mho this does not make much sense. According to NOAA, since 1960 atmospheric CO2 has been increasing by a factor of 1.00425 per year (from 317 to 396). At this rate it will take 163 years for CO2 to double. At least do a risk/reward analysis rather than arbitrarily picking a particular effect. Or pick data that supports doubling of CO2 resulting in a 30% increase in crop yield. A 2 degree increase in CO2 can arguable do more good than harm. Make decision on current issue such as energy sources and cost as well as known pollutants (CO2 is a plant food not a pollutant) and balance of trade not something that may or may not happened in 163 years or that may or may not be beneficial. Some of these arguments remind me of the philosophy of Pangloss in Voltaire’s Condide; “we live in the best of all possible worlds.” Condide didn’t find that to be true and I suspect that we also will not find that to be true.
I’m currently working on a study to try and constrain sensitivity using somewhat more complicated “simple” model, but I’m running into the problem that I haven’t got much experience with partial differential equations. Ah well.
Hm, from the forcing dataset I have made up (which I can provide to those curious and explain the derivation, and has the added benefit that you can adjust the aerosol forcing) over the period focused on by this study as when the anthropogenic trend begins (1959-2013), the “known” forcings (Greenhouse Gases (Methane, N2O, CFCs, in addition to CO2), volcanic eruptions, total solar irradiance) trended at ~0.44 W/m^2/Decade, and if I use a factor of -0.9 (I think this is about the right magnitude? Can anyone comment on this?) to multiply my aerosol index (which is basically just sulfur dioxide emissions scaled from 0 to 1 from 1850-2013) I get a “total” trend of ~0.436 W/m^2/Decade (which I had to bring out another decimal place because otherwise it would have rounded to an identical value). So aerosols should play relatively little role over this period-this is because world SO2 emissions peaked decades ago. They have relatively short atmospheric life times, so their loading in the atmosphere should be about proportional to emissions-also, several papers suggest recent brightening, which is consistent with declining emissions leading to declining aerosol loading. Anyway, if I divide the decadal rate quoted in Loehle’s paper over that period (.066 K/Decade) by the decadal rate of forcing increase, and multiply that by the forcing from a doubling of CO2 (3.7 W/m^2), I get a transient value of .56 K. Assuming as he does that models have the right ratio of TCR/ECS, I’ get ~1.02 K per doubling.
In other words, considering the other known forcings could cut his estimate of the transient response in half.
Of course, that’s just working under the assumption his method is correct. I can’t speak to that. I can however speak to the importance of considering other anthropogenic factors beyond just the CO2 forcing. On balance it looks like neglecting them leads one to over estimate the sensitivity.
Gary Pearse says:
February 28, 2014 at 10:28 am
Gents, if there is in fact a thermostat negative forcing that counters warming (or cooling in the case of volcanic aerosols), what then can we say about CO2 doubling climate sensitivity and if it matters. In such a case it would have only theoretical interest as an effect of CO2 doubling, all other things remaining the same.
Gary,
Well stated.
After reading Craig Loehle’s post, I was ruminating about this aspect when I read your comment.
Thanks,
Mac
Here is how I did it…
We have 17 years of no surface warming, so I will simply say that the energy going into the atmosphere / land is zero for 17 years (it has no appreciable heat capacity anyway since over a year you will overwhelm any local heat capacity by seasonal variation).
Over the same 17 years, Levitus tells us (indirectly) via the change in ocean heat content that the total accumulation in the ocean is 0.5W/m^2 over the period using the 0-2000m dataset (which also agrees with 4 Hiroshimas/second, a number that struck me as totally insignificant, which is why I converted to W/m^2). I went ahead and gave full credit for 0.52 doublings of CO2 (log2(400ppm/280ppm)) that would be active right now compared to preindustrial. 0.52*3.7W/m^2 per doubling = 1.94W/m^2 direct forcing from CO2 right now. Since we know that only 0.5W/m^2 is accumulating, and we know that 1.94W/m^2 direct forcing is active from CO2, we know that the feedback is 0.5-1.94 = -1.44W/m^2.
We know that earth is warmer since preindustrial, for whatever reason. In fact, it’s warmer enough that it should be emitting about 5W/m^2 more than it used to, just using SB equations and surface temp. It must be doing that already because there is no getting around the fact that the only increase happening now is 0.5W/m^2 and it’s going into the oceans. (the surface is cooling too by plenty of measures, but I’m ignoring that for now since it has no heat capacity to speak of compared to oceans)
So. We have 0.5W/m^w accumulating, and 1.94W/m^2 direct from CO2, all of which is fairly well accepted by enough people it’s worth talking about. 0.5W/m^2 imbalance means that given 1°C warming for the direct forcing of 3.7W/m^2 per doubling CO2, we should be expecting 0.5/3.7*1=0.135°C further warming to establish equilibrium. It’s nothing. So for all practical purposes, the earth is at equilibrium NOW. This means we don’t have to worry about transient vs ECS, we can calculate it directly. The effect is 0.5W/m^2 / the 1.94W/m^2 direct forcing, or 26% of the direct effect. If the direct effect per doubling is agreed to be 1.0°C to 1.2°C, then the sensitivity is 0.26°C to 0.31°C per doubling after accounting for feedbacks…
OK? Or did I run it off the rails somewhere… Thanks.
BTW, the same analysis using ocean heat since 1957 gives a similar result, but more sensitivity, since I actually integrated the direct effect of CO2 over the time period. That one resulted in 0.66°C per doubling.
All this talk about climate sensitivity is playing their game.
It’s the relevant concern.