Do the climate sensitivity estimates (est. warming response) incorporate the albedo effect? What’s “Earth System Sensitivity”?
Summary of Comments
Dr. Bart Verheggen: “The albedo effect due to the response from the big ice sheets is excluded (or held constant) in estimating [equilibrium climate sensitivity (ECS)]… That’s why e.g. Hansen and others call this the “fast feedback sensitivity”, since the really slow feedbacks (ice sheets and biosphere) are excluded.
Another way of saying this is to estimate ECS from the last glacial-interglacial transition, the albedo change due to ice sheet melting is taken as a forcing rather than as a feedback. See e.g. Fig S1 in the “target CO2″ paper by Jim Hansen. If instead you were to interpret the ice-albedo change and the biosphere change as feedbacks and thus exclude them from the forcing term, you’d get what’s commonly called the earth system sensitivity (ESS), and based on that same Fig S1 it would be approximately double the ECS value.
[ESS] is actually the more relevant climate sensitivity for long timescales (~milennia). The ECS sensitivity is in a way a transitional value between when the “fast” feedbacks have kicked in and “slow” feedbacks have not. ([Note:] “fast” is still fairly slow when compared to typical socio-political timescales). See also this post on SkepticalScience.com, especially their fig 1 (also taken from Hansen), where the dotted line is ECS and the solid line ESS.
Michael Quirke: “This is the first time “earth system sensitivity” has been mentioned in the Forum, and the first time I’ve heard the term. Just to be clear, are you saying that the ultimate result from a doubling of atmospheric CO2 in the VERY long-term – which takes into account the additional albedo-effect forcing and its subsequent response – is a warming that’s around TWICE the amount in the ECS estimates?!?!
This seems more than a little important in terms of weighing the costs and benefits of climate policies.”
Dr. Bart Verheggen: Yes indeed, on millennial timescales the temperature response is expected to be larger than the more common equilibrium climate sensitivity (ECS) would imply. How much larger is very much an open question.
From AR5: “The resulting equilibrium temperature response to a doubling of CO2 on millennial time scales or Earth system sensitivity is less well constrained but likely to be larger than ECS (…)” and “(…) medium confidence that Earth-system sensitivity may be up to two times the model equilibrium climate sensitivity (ECS)”
As a short-cut, one and a half times ECS may be a better ballpark number than twice ECS. For the long term commitment of the climate system, as a result of current emissions, this is indeed very important. During the last Interglacial, global temperatures may have been a degree or so warmer than now, but sea level was 5 to 9 metres higher (these estimates may be slightly outdated, didn’t check the latest references, but the point remains the same).
Different people may perceive the relevance of these long time scales very differently however: Some people care deeply about the long-term legacy of our actions and in-actions, whereas others might say something along the lines of “What has the future ever done for me?”.
One thing is for sure: Future generations can’t hold us accountable. That’s a key aspect of the climate issue: http://ourchangingclimate.wordpress.com/2010/09/06/future-generations-global-warming-is-not-our-problem/.
Dr. Scott Denning: “Bart and others have done a really great job of explaining the difference between “fast” feedbacks (on time scales of years to decades) and “slow” feedbacks (on time scales hundreds to tens of thousands of years). The “fast” ones control Transient Climate Sensitivity (TCS) or Equilibrium Climate Sensitivity (ECS) as estimated in models that hold ice sheets constant.
But think about Ice Age cycles for a minute, the “global wobbling” of another post on here by Sean Bryan. The changes in planetary radiation balance resulting from Milaknovitch cycles that control the timing of the Ice Ages are really tiny, only a few tenths of a Watt per square meter (more than 10 times weaker than the change due to doubling CO2), yet they produce monstrous changes to global climate, with giant ice sheets stretching from Hudson’s Bay to Chicago, then melting. This shows the strength of the ice-albedo and carbon cycle feedbacks over tens of thousands of years…”
Dr. John Nielsen-Gammon: “One other thing to keep in mind is that, although the Earth System Sensitivity is large, it operates on time scales comparable to the long-term removal of excess CO2 from the atmosphere. If we double CO2, we’d have to keep adding CO2 to the atmosphere for thousands of years to realize the full Earth System Sensitivity for CO2 doubling.
So it’s not something to realistically worry about, if the climate system is linear. If, on the other hand, Earth System Sensitivity is strongly affected by bifurcations (such as, for example, the long-term collapse of the West Antarctic Ice Sheet), we may reach something close to the full ESS in thousands of years even though CO2 would have largely recovered by then.”
Where does the albedo effect fit into the estimates of climate sensitivity (warming response)?
By Michael Quirke
Unlike other aspects of climate change science, albedo seems to be a fairly simple concept. In Latin, “Albedo” translates to “whiteness,” and in science, it means the measure of a surface’s reflectance. Thus, a dark surface has a low albedo (low amount of reflectance) and therefore absorbs more solar energy, which makes it hotter. A bright surface, in contrast, has a high albedo (a high amount of reflactance). Consequently, the brighter surface is much cooler because it reflects a lot of solar radiation.
If one has ever worn a black t-shirt or walked from a black asphalt parking lot to a white sandy beach on a bright summer day, then he or she understands how albedo works.
In climate science, the earth’s albedo is the measure of how much the sun’s energy is reflected back into space on a global scale. As explained by multiple scientists in the Forum, the large-scale melting of glaciers and sea ice results in a decrease in the earth’s albedo, and this “albedo effect” contributed to the warming that brought us out of the last ice age and into the current interglacial period:
Dr. Sean Bryan: “It is important to note that the orbital changes are not enough to drive the entire glacial-interglacial transitions. They got the ball rolling, but feedbacks such as albedo and greenhouse gases are necessary to explain the entire change.”
– Climate Change and Timescales (Part 1): Why global wobbling doesn’t matter by Dr. S. Bryan.
Dr. Will Howard: Some cite the Quaternary climate cycles as an example of “this has all happened before.” The logical contradiction is that these cycles are a source of our understanding of the role of greenhouse gases as powerful feedbacks, and of other feedback processes (e.g. albedo, among other short- and long-time scale processes) that mediate subtle changes in the distribution of incoming solar radiation into large-scale climate changes.
– Scientists’ Comment Thread under Fears of Freezing: The 1970’s are calling. They want their climate policies back by Dr. W. Howard.
Dr. Scott Denning: “Starting 20,000 years ago, the world warmed up by about 5 degrees Celsuis over 10,000 years. Why? Well, two things happened. First, the Ice sheets melted… The ice is bright and white and reflects a lot of sun, so when the ice melted, the earth’s surface got darker, which soaked up a lot more sun and it got warmer… “
So where does the albedo effect fit into all the recent climate sensitivity estimates shared in the Forum? Based on the excerpts above, the albedo effect seems to be a virtually certain positive-feedback, but one that has only begun to manifest itself recently.
– Teaching Climate Change Through Six Questions by Dr. J. Shakun.
In this recent NASA video, Dr. Nathan Kurtz of NASA Goddard Space Flight Center describes the high albedo of sea ice in the arctic as an “air conditioner for the planet”. He explains that since satellites observations began in 1979, there’s been a loss of sea-ice area equivalent to a third of the size of the United States and this is due to the rapid warming of the arctic. The narrator notes that the current volume of sea ice is only around 2/3rds of what it was in 1979. This makes the sea ice less “resilient” says Dr. Kurtz.
Am I correct to assume that one big reason why the climate sensitivity estimates from the paleoclimate analyses & climate models are always higher than the estimates from the observational/instrumental record (such as the one recently shared by Dr. Judith Curry), is because this particular positive feedback has only begun to manifest itself in the observations/instrumental record? I mean, with the exception of arctic sea ice loss, most of the ice sheets and glaciers of the world are going to take a long time to melt right?
Dr. Bart Verheggen: No, I don’t think that’s necessarily the case, at least no to the extent that you’re implying. The reason is that the response of the big ice sheets are excluded (or held constant) in estimating ECS (for a good explanation of different sensitivity definitions, see this post on RealClimate.org). That’s why e.g. Hansen and others call this the “fast feedback sensitivity”, since the really slow feedbacks (ice sheets and biosphere) are excluded.
Another way of saying that is that to estimate ECS from the last glacial-interglacial transition, the albedo change due to ice sheet melting is taken as a forcing rather than as a feedback. See e.g. Fig S1 in the “target CO2″ paper by Jim Hansen. If instead you were to interpret the ice-albedo change and the biosphere change as feedbacks and thus exclude them from the forcing term, you’d get what’s commonly called the earth system sensitivity (ESS), and based on that same Fig S1 it would be approximately double the ECS value. That’s actually the more relevant climate sensitivity for long timescales (~milennia). The ECS sensitivity is in a way a transitional value between when the “fast” feedbacks have kicked in and “slow” feedbacks have not. (where “fast” is still fairly slow when compared to typical socio-political timescales).
See also this post on SkepticalScience.com, especially their fig 1 (also taken from Hansen) where the dotted line is ECS and the solid line ESS.
Of course, a globe with lots of ice is going to respond differently to a change in planetary temperature than globe with comparatively little ice. You can see that in Fig 1 in the latest mentioned blogpost as the smaller difference between ESS and ECS under current conditions when compared to glacial conditions: ESS is smaller now than it was in glacial times, due to the smaller potential ice sheet response now. According to Hansen these values would be ~6 (glacial) and ~4.5 (interglacial) degrees. I’m not sure to what extent that would affect ECS estimates though.
[See follow-on comments below, in the Forum Comment Thread. This follow-on discussion is summarized at the top of this post.]
Featured image: Screenshot of NASA’s graphical rendition of Arctic Sea Ice on September,17 2013 from the video by NASA Goddard Institute of Space Studies.