by Judith Curry
These results support the notion that the enhanced wintertime warming over high northern latitudes from 1965 to 2000 was mainly a reflection of unforced variability of the coupled climate system. Some of the simulations exhibit an enhancement of the warming along the Arctic coast, suggestive of exaggerated feedbacks. – Wallace et al.
Simulated versus observed patterns of warming over the extratropical Northern Hemisphere continents during the cold season
John M. Wallace, Qiang Fu, Brian V. Smoliak, Pu Lin, and Celeste M. Johanson
A suite of the historical simulations run with the Intergovernmental Panel on Climate Change Fourth Assessment Report (IPCC AR4) models forced by greenhouse gases, aerosols, stratospheric ozone depletion, and volcanic eruptions and a second suite of simulations forced by increasing CO2 concentrations alone are compared with observations for the reference interval 1965–2000. Surface air temperature trends are disaggregated by boreal cold (November-April) versus warm (May-October) seasons and by high latitude northern (N: 40°–90 °N) versus southern (S: 60 °S–40 °N) domains. A dynamical adjustment is applied to remove the component of the cold season surface air temperature trends (over land areas poleward of 40 °N) that are attributable to changing atmospheric circulation patterns. The model simulations do not simulate the full extent of the wintertime warming over the high-latitude Northern Hemisphere continents during the later 20th century, much of which was dynamically induced. Expressed as fractions of the concurrent trend in global-mean sea surface temperature, the relative magnitude of the dynamically induced wintertime warming over domain N in the observations, the simulations with multiple forcings, and the runs forced by the buildup of greenhouse gases only is 7∶2∶1, and roughly comparable to the relative magnitude of the concurrent sea-level pressure trends. These results support the notion that the enhanced wintertime warming over high northern latitudes from 1965 to 2000 was mainly a reflection of unforced variability of the coupled climate system. Some of the simulations exhibit an enhancement of the warming along the Arctic coast, suggestive of exaggerated feedbacks.
Recently published in PNAS; link to abstract [here].
In the latest issue of PNAS, Jerry North has a commentary on the Wallace et al. article listed on the PNAS early edition site [link]. Since all this is behind paywall, here are some excerpts:
Gerald R. North
One of the more intriguing mysteries of global warming is the fact that the high northern-latitude surfaces are warming faster than those averaged over the globe, a phenomenon known as [Arctic] amplification (AA).
Among the candidates for explaining AA are (i) unforced natural variability in the coupled ocean/atmosphere system; (ii) altered ocean heat transport into the region; (iii) local effects, such as thinning of sea ice and overall reduction of the area covered by sea ice ventilating heat from below into the atmosphere; and (iv) remote sea- surface temperature patterns that might correlate or “teleconnect” through long atmospheric wave patterns with polar conditions. It is possible that these and other effects overlap and combine in different ways in different seasons and that no one of them alone is dominant. An important fact that must be explained is that most of the AA is in boreal (Northern Hemisphere, NH) winter and over land. There is very little amplification in the SH.
The present study is about the trends in the mass and temperature distributions in the atmosphere and how one can parse out the natural fluctuations due to dy- namical instabilities in the atmospheric flows from the purely thermodynamic signals. These empirical dynamical patterns are to be distinguished from those mainly associated with heating due to radiative imbalances in the entire atmospheric column (this latter can be thought of as the pure “greenhouse gas warming” signal, although aerosols are also involved).
The starting point of the present study is the observation that variations in the pressure mode patterns excite the dynamical patterns in the thermal fields, in- cluding the surface temperature. Large pressure fluctuations mean more heat flow toward the poles due to storminess in the upper midlatitudes. Pure radiative imbalances do not induce much pressure variability. The pressure variability pat- terns are always stronger in the winter hemisphere because of strong latitudinal thermal gradients (equator to pole). Temperature responses to heat fluxes from the tropics induced by the pressure fluctuations are largest over large land masses because the effective heat capacity is much less than over oceanic areas. Be- cause there are virtually no large land masses in the high temperate latitudes in the SH, little surface temperature response to dynamical activity (pressure variability) is generated there. By contrast the NH contains the large Eurasian and the North American continents, with their extreme seasonal cycles in surface temperature (so-called extratropical continental climates).
After finding the dynamical component by least-squares regression with the pres- sure modes, the next step is to “remove” this component’s contribution to the trends, yielding the purely thermodynamic or radiative-imbalance signal, due mainly to human greenhouse gas and aerosol emissions.
The authors choose the reference in- terval 1965–2000, a period when the global average surface air temperature was rising at a nearly uniform rate. They wanted to conduct their separation under these most optimal conditions and apply it to multidecadal trends. The dynamical contribution of the trend in temperature is virtually all in the NH winter during the period of study, and it is dominant over the large landmasses. The purely thermo- dynamic component is pretty boring. It does not have much dependence on sea- son or hemisphere or even land vs. sea surface—it just keeps on warming linearly essentially everywhere in all seasons. These qualitative findings hold for both the observed surface temperature fields and those derived from published climate model simulations that cover the same interval. The study includes two types of model runs: (i) a multimodel average that includes the conventional drivers of climate: changing aerosols, increasing greenhouse gas concentrations, volcanic dust veils, and solar variability; and ii) a second series of runs in which the aero- sol, volcanic, and solar forcings are omit- ted but a pure CO2 forcing is included, with concentration increasing at 1.0%/y. In each case, data or model simulation, the dynamical effects were removed in the same way.
The dynamical trends were similar in the observations and the model simu- lations, except that the dynamical effect was much larger in the observations than in both model experiments. In other words, the dynamical effects are clearly present in the state-of-the-art global climate models, but they are much too small to match the data.
There are multiple lessons to be learned here. First, there are unresolved issues in computing the general circulation of the atmosphere. This could be plain old errors in the models, or it could be that some key effect or component is being omitted.
The article shows that a large portion of the climatic change over the period being examined is attributable to the dynamical component rather than direct radiative forcing. This former is the part that is associated with a change in the atmospheric circulation system, such as widening the giant tropical cells known as the Hadley Circulation, and causing the midlatitude storm belts in both hemispheres to gradually extend their centers and boundaries polewards. Satellite observations and other data indicate that these pole-ward shifts of the storm belts are larger than expected according to current climate models that include the conventional forcings. It seems that these shifts have had a larger effect during the study period than has the direct radiative imbalance due to increasing concentrations of greenhouse gases, although the latter is substantial.
What causes the two primary decadal empirical pressure patterns to ramp up their amplitudes and line up their phases in such a way as to encourage lots of warm air polewards onto the big continents during the study period and not at other times? Wallace et al. do not address this issue but list several contenders.
One possibility is that the anthropogenic factors are actually inducing this change in circulation, perhaps even through stratospheric connections as mentioned earlier. They seem to lean toward unforced natural variability at the decadal scales. For example, if there is a long-term unforced oscillation pattern in the coupled system, it might be setting and evolving sea surface temperatures world- wide, and these are in turn causing the conditions favorable for the high-latitude pressure variations.
The central contribution of the present article is that it provides a framework for seeing how the different factors combine in a few simple indices explaining most of the AA. There seems to be a large multidecadal variability in the complex ocean–atmosphere system that can su- perimpose itself atop the global warming signal. It seems to be identifiable with a few large-scale patterns in the temper- ature fields; when phases of the pressure modes match up one with another they can enhance the rate of warming, especially on the large, wintertime continents. It is not yet known whether cooling periods can happen as well in this scheme. Nor do we know exactly the origin of the multidecadal long-term variability that has been identified. It could be related to anthropogenic causes, or it could just be part of the natural system’s internal variability.
JC comment: I like the Wallace et al. paper, and Jerry North did a nice job with his commentary. However, none of this is news to me, since I have published two papers previously that came to same conclusions:
- Recent Arctic sea ice variability: connections to the Arctic Oscillation and the ENSO
- Causes of the northern high-latitude land surface winter climate change
While there is a trend in the Arctic, the amplification is associated with natural internal variability. As per google scholar, the sea ice paper has 7 citations (miniscule) and the land paper has 41 citations (moderate). Neither paper was cited by Wallace et al., presumably they are unfamiliar with these papers.
This brings to my mind the issue of ‘unknown knowns’, to expand upon Donald Rumsfeld’s vernacular. An ‘unknown known’ is something that somebody knows, but isn’t generally known (in this case it was known by myself and my coauthors and a handful of people that read the papers, but not sufficiently known to make into say an IPCC assessment or to be referred to in more than a few scientific papers).
So how to make the knowns actually known? Sign up to be an IPCC author, so you can cite your own papers. Or issue a press release. Hmmm . . . There oughta be a better way.
I’m not being critical of colleagues like Wallace et al.; in fact I wouldn’t have known about the Wallace et al. and North articles if Ronald Hirsch hadn’t sent me an email. How to really mine the published literature for knowledge remains a challenge, although the internet is an enormous boon. Known unknowns and unknown unknowns are extremely challenging to deal with; it seems like dealing with the unknown knowns should be relatively tractable.
I am very appreciative of those of you who send me links to articles, this helps me deal with the unknown known issue in terms of my own understanding and in sharing these through Climate Etc.