Monday, November 08, 2010

The Anti-Watts Effect


For various reasons Eli has been thinking about aerosols and how the mid-century cooling is attributed to them. Now the Bunny has also been consorting with a bunch of acid rain and regional forcasting types and the thought occurred that maybe we have a case here of the urban cooling effect. To make a long story short, and this is really a WAGNER (wild assed guess, no explanation required) what if the large amounts of SO2 injected into the northern hemisphere atmosphere by WWII and the unrestrained coal burning (see London, smog) produced huge amounts of sulfate aerosol which shadowed and cooled downwind rural measurement sites. Sulfate aerosol gets rained out pretty quick, so the range would not be global. This would mean that the dip between 1940 and 1970 was in a sense an artifact, the UCE.

The figure at the left from GISS shows that the cooling was a northern hemisphere thing. Warming in the tropics and the southern hemisphere has been quite steady, even though there are plenty of aerosols there (although they are different, the major sources in the SH being sea spray and in the tropics sand from the Sahara as well as agricultural burning in Brazil and Africa.

38 comments:

James Annan said...

This is something that various people have looked at, eg Andronova and Schlesinger 2001:

http://www.agu.org/journals/ABS/2001/2000JD000259.shtml

(though this analysis is obviously dated)

Anonymous said...

http://www.skepticalscience.com/global-warming-early-20th-century-intermediate.htm
Including a comment from Mizmi
"One point to consider: oil and gas as fuels did not come into widespread use until the mid 1950's and began to supplant coal from 1960 onwards.
Coal was the major fuel before 1940 and emission controls virtually non-existant, so there would also have been a cooling effect from aerosols to (partially) balance GG emissions"

Taimino's volcanic lul has dissappeared
http://tamino.wordpress.com/2008/10/19/volcanic-lull/

Skeptical Science has the following link
Estimation of natural and anthropogenic contributions to twentieth century temperature change. Simon F. B. Tett and others

http://www.agu.org/pubs/crossref/2002/2000JD000028.shtml

Here is where I should put my own comments in to show I understand the question.

Little Mouse

Roger Jones said...

Oh my. Snap. I've been working on S Hemisphere non-linear change but hadn't seen this pic 'cause I'd been concentrating on the local chang in SE Aust. Thanks Eli. I'll chase the data.

cRR Kampen said...

Could be. But there's evidence for an analogy with (parts of) the 'Little Ice Age', which also appears to have been primarily a phenomenon of the Northern hemisphere. Volcanoes might have done this job (the Maunder Minimum apparently hardly did).

EliRabett said...

Well, yes and no. Andronova and Schlesinger looked at the global situation. What Eli was getting at is that the effect should be regional, e.g. you should see, in US terms a large effect east of Pittsburgh, Chicago, etc., but not as much to the west by considering the prevailing winds. A quick look at the station data in GISS showed this is not a stupid idea. Tett, et al, is more like it, but again, Eli's idea is not to look at global, but small regional responses

Anonymous said...

"What Eli was getting at is that the effect should be regional, e.g. you should see, in US terms a large effect east of Pittsburgh, Chicago, etc., but not as much to the west by considering the prevailing winds."

Yes, and no. My understanding, garnered from my interactions with the black carbon community, is that while the emissions are highly localized, and the forcing is somewhat localized (emissions blurred by transport), the temperature effect is much less localized, because temperatures changed by forcing can also be blurred by transport. The end effect being that the temperature effect of a given set of emissions end up being fairly equal around almost the entire latitude band. Adding to this effect is the fact that continental interiors respond to forcing faster than coasts and oceans. Transport is slower north-south, and especially across the equator, so those signals are easier to pick up.

Mind you, a number of the studies that the black carbon people use to get this intuition reduce emissions globally, so that would make it even harder to see patterns: I kind of want to see the modeled effect of a large point emission source, which I don't know that I've seen.

Also, the dimming effect is definitely localized to where the forcing is, so the dimming signal (which is a component of the temperature signal) may be more localized. Hmm. My theorizing aside, it is certainly worth looking at the data...

-M

Rocco said...

Eli: But wouldn't regional changes also be swamped by natural variability?

Anonymous said...

How about this one:

http://www.nasa.gov/pdf/327055main_Fig3_Gisstemp_timeseries.pdf

From: http://www.nasa.gov/topics/earth/features/warming_aerosols.html

Anonymous said...

Sorry, forgot this one:

http://earthobservatory.nasa.gov/Features/GISSTemperature/Images/greenland_sulfur.gif

EliRabett said...

Wonder how much of the Greenland sulfur came from the deuterium separation plants in Nova Scotia? That must have been a huge source. The point about regional sources is that they should be very localized and thus would, at least on the regional scale, swamp natural variability.

Pbo said...

I'm quite sure this sulfur has traveled long - you can easily see the depression in the 30's and perhaps el chinon eruption '82 and pinatubo eruption 91'?

(the two last anonymous from me)

Pbo said...

Here's the source article for the sulphur graph from Greenland:

http://earthobservatory.nasa.gov/Features/GISSTemperature/giss_temperature4.php

Anonymous said...

Didn't Tamino do a home-grown analysis of the temperature trend in which he built a neat piece of software? Couldn't you take that and dump out all the data recording sites which are east of major industrial sites?

Lurker lawyer mouse

Anonymous said...

Are you aware of the National Atmospheric Deposition Program? This is the link to SO4 changes in precip from 1985 to 2005

http://nadp.sws.uiuc.edu/data/amaps/so4/amaps.html

EliRabett said...

Pinatubo and el Chichon were explosive volcanoes that pushed SOx up into the lower stratosphere where the fall out time is a year or so (very dry up there so there is no rainout. Normally SOx converts to sulfuric acid aerosol fairly fast and rains out locally.

One of the nasty things that happened is when the US Clean Air Act specified local emission limits, the utilities built huge stacks to move the acid rain deposition pattern further away. Even then the deposition pattern is only regional.

Eli was somewhat aware of the NADP but had not put it together with this, which is the point of this sort of speculation with others. You get a better picture.

Same Ordinary Fool said...

Would pan evaporation data be helpful in choosing your region? Wikipedia's Global Dimming article has some references.

Since "sulfate aerosol gets rained out pretty quick," are you going to be looking for wetter (or drier) months?

This is a variant: "Wild assed guess, no EDUCATION required." Just watch NOVA.

cRR Kampen said...

The acid rain laboratorium of the world is China. Data are unfortunately less prolific, I guess. But the problem might be termed one of the worlds largest natural disasters.

John Mason said...

Interesting thoughts, Eli. Well worthy of further investigation!

Cheers - John

Unknown said...

There is increasing cooling during the period - does this correlate with varying levels of sulfates/pollution in associated regions?

NH is still much more (visibly) hazy than SH. Historical accounts suggest the same was true in 19th century. Presumably there are some relevant observation series (cloudiness, visibility index?) at least since WW2. Another WAG.

David B. Benson said...

Tamino's volcanic lull was svaed in
http://webcache.googleusercontent.com/search?q=cache:cTkJPW3J9UoJ:https://tamino.wordpress.com/2008/10/19/volcanic-lull/

I suggest also looking at indicies of internal variability.

David B. Benson said...

DelSole, T., M. K. Tippett, and J. Shukla, 2010: A Significant Component of Unforced Multidecadal Variability in the Recent Acceleration of Global Warming. J. Climate, submitted.
ftp://www.iges.org/pub/delsole/dir_ipcc/dts_science_2010_main.pdf

Anonymous said...

China must surely be producing more global dimming with her unfiltered coal burning today than the industrialised countries did mid century.

Andrew W

Robert Way said...

Yeah i've never really bought the global mid-century cooling as being due to aerosols actually. I'm very much in agreement that it is linked to IMP or the AMO essentially just as David Benson shows above. I have done quite a bit of work on this subject and find that you don't NEED to include aerosols to get mid century cooling if the AMO is considered...

J Bowers said...

Some trivia about the AMO:

"Yeah, I came up with the term: Atlantic Multidecadal Oscillation. I coined the term in an interview with Richard Kerr [a writer for Science] in 2000 over a paper with Tom Delworth of the Geophysical Fluid Dynamics Laboratory and the NOAA Laboratory in Princeton, where we actually were the ones to articulate the existence of this oscillation. And you know what? It was celebrated by contrarians. My work has been celebrated by climate skeptics. It’s an interesting footnote." -- Michael Mann in an interview with Discover Magazine

EliRabett said...

The problem with an AMO or IMP is that it is a post-hoc rationalization of observations, rather than something with physical causes. As to China, they are cleaning out a lot of the old coal burning plants, and replacing them with more efficient ones. Still coal, but a lot less black carbon.

Anonymous said...

Robert Way says: I have done quite a bit of work on this subject and find that you don't NEED to include aerosols to get mid century cooling if the AMO is considered...

But then how did the aerosols manage to not do much of anything?

Pete Dunkelberg

Anonymous said...

In the interest of perfection, it would help to have the WW2 sea surface temperatures cleaned up with respect to the change in method of measurement.

A large discontinuity in the mid-twentieth century in observed global-mean surface temperature. Thompson et al. 2008 Nature.

I suppose this work is under weigh somewhere. Meanwhile is there a good guess of the difference it will make?

Pete Dunkelberg

David B. Benson said...

Eli Rabett & Robert Way --- AMO is a proxy, or index, of internal variability. It certainly is not a perfect one as, for example Tamino's volcanic lull illustrates. Probably IMP is a better such index. However, the AMO does appear to be correlated with MOC rate changes, although that is quite a difficult subject due to insufficent data.

There is also the question of whether the midcentury change in the method of measuring SSTs makes a difference. I'm under the impression there is a paper recently on a combination of that factor and aerosols.

Tropospheric aerosol production ought to make a difference and the suggestion
to look just regionally for that signal seems a good one to this amateur.

As best as I can determine, using just zero dimensional models, even with
http://data.giss.nasa.gov/modelforce/RadF.txt
as the radiative forcing and accounting for ENSO and AMO as internal variability, the models don't manage the fully explain the mid-century (minor) cooling. So I suspect that the earlier SSTs need adjustment to agree with the modern measurement method. Of course I have no idea how to do this. I am under the impression that work on this matter contins at CRU.

David B. Benson said...

EliRabett & Pete Dunkelberg --- The AMO is an actual pre-existing quasi-periodic oscillation as determined by, amoung other proxies, northern european tree rings dating back hundreds of years before the beginning of the instrumental period in 1850 CE. Nonetheless, some aspects of volcanic and other aerosol variations are surely going to influence its exact form.

Anonymous said...

Didn't tamino put this one to bed?

http://tamino.wordpress.com/2010/08/23/antrhopogenic-global-cooling/

David B. Benson said...

drj11 --- Not entirely. There is still a small effect due to changes in the method of measuring SSTs and also a long-term internal variability approximately measured by the AMO. That just CO2 and AMO work well, here is a zero-dimensional, zero reservoir, decadal model for the instrumental period:
http://www.realclimate.org/index.php/archives/2010/10/unforced-variations-3-2/comment-page-5/#comment-189329

EliRabett said...

David, the diff is that Eli is hypothesizing a strong regional signature

David B. Benson said...

EliRabett --- That would certainly be of interest to determine. The entire problem is a sysmtem identification question which, as I understand it, is not completely resolved. This is partly because there is no sufficiently decent proxy for the long term climate variability.

J Bowers said...

More trivia.

A WUWT accolyte stumbles across a lamp and rubs it. !!!FLASH!!! a Genie appears and grants him one wish. The WUWT accolyte ponders for a while and says, "I know. To make sure I can keep knowing more about science than those damned CAGW Alarmists, make me a million times smarter!" !!!POOF!!! The Genie turned him into a climate scientist.

As you were.

Horatio Algeranon said...

Horatio was under the impression that the "Nega-Watts Effect" was the cooling bias in "poorly sited" surface weather stations

Anonymous said...

So are we closer to a paper?

Little Mouse

EliRabett said...

Closer than one might think. The mighty data banks are having a run - Eli

John E Pearson said...

Anonymous said "I kind of want to see the modeled effect of a large point emission source, which I don't know that I've seen."

I'm not sure how relevant this is but during Gulf War I there was a lot of smoke from oil well fires that was modeled reasonably well.