Aerosols from coal burning, would a sudden stop actually raise temperatures?
gerh...@aston.ac.uk
70% of doubled CO2, and the response that that doesn't mean a 2C
increase right now, because of aerosols and thermal lag.
What I am wondering about is what would the effect of sudden changes
in aerosol production from coal fired power plants be?
Would shutting down all coal fired power stations over night yield a
rapid jump in temperatures, as aerosols are washed out of the
atmosphere within months?
And,
if you believe Chinese statistics on coal burning, there's been a huge
jump recently,
mighn't that, the short term at least, actually depress temperatures?
Would anybody here know whether there's a study on the matter/
James Annan
> We all know Lindzen's line about greenhouse gas forcings already being
> 70% of doubled CO2, and the response that that doesn't mean a 2C
> increase right now, because of aerosols and thermal lag.
>
> What I am wondering about is what would the effect of sudden changes
> in aerosol production from coal fired power plants be?
>
> Would shutting down all coal fired power stations over night yield a
> rapid jump in temperatures, as aerosols are washed out of the
> atmosphere within months?
There would be a rapid increase in forcing, but in broad terms this
would only result in a change in the _rate_ of warming, not a step
change in the temperature itself.
(Local effects could be more rapidly obvious, I expect.)
>
> And,
>
> if you believe Chinese statistics on coal burning, there's been a huge
> jump recently,
>
> mighn't that, the short term at least, actually depress temperatures?
Well it would probably tend to offset the CO2 forcing and perhaps might
be negative overall.
>
> Would anybody here know whether there's a study on the matter/
There have certainly been studies simulating different components of the
forcing, eg omitting all aerosols over the 20th century. The effect is
clearly noticeable but not overwhelming on the global scale.
James
Michael Tobis
See p 3 of this powerpoint:
http://www-new.mcs.anl.gov/climate/cwg/slides/ANL_CCW_Streets.ppt
Aerosols are dominated by low-tech burning of fuels for heat, and by cheap, primitive factories and motors.
This talk also indicates that global aerosol emissions have a slightly downward trend while greenhouse gases march inexorably upward.
mt
gerh...@aston.ac.uk
> would only result in a change in the _rate_ of warming, not a step
> change in the temperature itself.
Presumably because of thermal lag of the oceans.
Is that sufficient though to prevent a step change (?), thermal
inertia of the atmosphere is virtually zero, and I am not sure how
efficient heat transfer to the oceans would be.
Thermal lag gets me to an interesting point I've recently seen made
about cosmic rays: It's not necessary for (at least much or some of)
recent warming to be explainable by cosmic rays that there's a recent
trend in cosmic rays. It's entirely sufficient for them to have gone
up a few decades ago, and thermal (and other?) lags will mean
continued temperature increases.
Of course that in no way proves that the cosmic ray theory is anything
but hogwash, but it's still an interesting point.
gerh...@aston.ac.uk
>
> http://www-new.mcs.anl.gov/climate/cwg/slides/ANL_CCW_Streets.ppt
>
> Aerosols are dominated by low-tech burning of fuels for heat, and by cheap,
> primitive factories and motors.
Isn't that just for black carbon rather than sulphur, and isn't black
carbon a positive forcing agent, and sulphur related aerosols are
responsible for cooling?
Biomass doesn't contain much sulphur, coal does.
Raymond Arritt
>> See p 3 of this powerpoint:
>>
>> http://www-new.mcs.anl.gov/climate/cwg/slides/ANL_CCW_Streets.ppt
>>
>> Aerosols are dominated by low-tech burning of fuels for heat, and by cheap,
>> primitive factories and motors.
>
> Isn't that just for black carbon rather than sulphur, and isn't black
> carbon a positive forcing agent, and sulphur related aerosols are
> responsible for cooling?
The presentation is devoted specifically to carbonaceous aerosols, not
sulfates (sulphates) or aerosols in general.
Ray
James Annan
> Aerosols are dominated by low-tech burning of fuels for heat, and by
> cheap, primitive factories and motors.
Interestingly, I just saw a presentation yesterday indicating that
burning of agricultural waste (straw) is a significant source of ozone
and black carbon in China (at least, this source appeared to be
responsible for large peaks, although in the annual average it might not
be so important).
James
James Annan
>> There would be a rapid increase in forcing, but in broad terms this
>> would only result in a change in the _rate_ of warming, not a step
>> change in the temperature itself.
>
> Presumably because of thermal lag of the oceans.
>
> Is that sufficient though to prevent a step change (?), thermal
> inertia of the atmosphere is virtually zero, and I am not sure how
> efficient heat transfer to the oceans would be.
The atmosphere is pretty tightly coupled to the ocean, especially the
surface. I suppose it's possible that some altitudes would be affected
more significantly, depending on the details of aerosol behaviour (a
point that I'm not at all expert on).
> Thermal lag gets me to an interesting point I've recently seen made
> about cosmic rays: It's not necessary for (at least much or some of)
> recent warming to be explainable by cosmic rays that there's a recent
> trend in cosmic rays. It's entirely sufficient for them to have gone
> up a few decades ago, and thermal (and other?) lags will mean
> continued temperature increases.
Well unless one postulates an extremely high sensitivity to such forcing
(but not to other forcings), one would expect the warming to be
levelling off - and there's no sign of that except through some rather
specious cherry-picking vis a vis 1998...
James
Michael Tobis
Interestingly, the closest I have come is in another PowerPoint by David Streets:
http://www.giss.nasa.gov/meetings/pollution2002/present/1_streets.ppt
Relevant are slides 5 and 8. Slide 8 supports the impression that the trend in aerosol emissions from Chinese power plants is sharply downward, and that this applies to sulfur as well.
Slide 21 is also interesting in view of the discussion about possibly insufficient in situ measurement of important processes.
mt
gerh...@aston.ac.uk
> surface. I suppose it's possible that some altitudes would be affected
> more significantly, depending on the details of aerosol behaviour (a
> point that I'm not at all expert on).
I've been thinking quite a bit about thermal lag, it's an area that I
like, because of my background as a chemical engineer.
A good comparison ought to be seasonal changes. Over a day forcing
changes by about 1000 W/m2, over a season I don't know exactly, order
of magnitude wise it ought to be around 200 W/m2.
In Siberia, seasonal temperature changes I think approach 50C, in
England it's more like 20C.
Some of the forcing will go towards heating soil and air, some will be
compensated by more radiation.
Aerosols are currently around 1 W/m2 and with climate sensitivity of
3C that's equivalent to around 1C without thermal lag.
Currently, around 0.5 W/m2 or so go towards heating the oceans, 1 W/m2
is offset by aerosols (0-2.5) and 2.5 W/m2 is greenhouse gase
forcing.
I think this gives me some reasonable bounds for the response to a
sudden decrease in aerosol forcing. There ought be a jump of at least
0.1C over the first year, but more likely 0.5C.
The key here is looking at the ratio of the seasonal temperature
change and the seasonal forcing change. If there was no thermal
inertia (and no heat transfer from the tropics), 200 W/m2 should mean
roughly 200C with a climate sensitivity of 3C. When it is actually
50C, 3/4 of the forcing go toward heating air and soil.
The longer term bound seems to be about a third going to heating the
oceans, so 0.5 C after the first year, 0.7 C after 2-3 years and 1C
after decades to a century ought to be the response to a step change
in forcing by 1 W/m2 (roughly, roughly)?
---------------------
I find it quite helpful to look at the heat capacity of air and
oceans. 10 km of air weighs about as much as 10 m of water. The heat
capacity of water is 4.2 kJ/kgK, the heat capacity of air 1 kJ/kgK.
How long does it take to heat 10 km of air at 1 kW/m2 (full noon time
insolation)? 10 km times a metre is 10,000 m3 or about 10,000 kg. So
we need 10,000 kJ or a 1 kW, 10,000 second, or just under 3 hours.
Day/night time differences are 10C or higher and it doesn't take 30
hours to go from the daytime high to the daytime minimum, but that's
because only a fraction of the atmosphere actually goes through the
temperature change (the bottom most layer, temperature higher up is
much more constant).
How long does it take to heat 1 km of ocean at 0.5 W/m2 (current heat
take up by the oceans, roughly)? 1 km of ocean weighs 1,000,000 kg and
has a heat capacity of 4.2 billion J/K. At 0.5 W/m2, heating the ocean
by 1 K will therefore take 8.2 billion seconds, or 260 years.
> Well unless one postulates an extremely high sensitivity to such forcing
> (but not to other forcings), one would expect the warming to be
> levelling off - and there's no sign of that except through some rather
> specious cherry-picking vis a vis 1998...
I suppose, playing devil's advocate here, the recent lack of levelling
off might be due to say the declining aerosol forcing Michael Tobis
mentions ("the trend in aerosol emissions from Chinese power plants is
sharply downward").
Not being a climate scientist I am a little confused about what the
concept of forcing really means in the case of cosmic rays. My
understanding is that the forcing is measured in W/m2 at the ground,
assuming no feedbacks from water vapour, clouds etc.
So, say for CO2 we have say 2W/m2 out of 200 W/m2 (or actually 350 or
so, when taking into account greenhouse gases), simple radiation is
based on the fourth power of temperature, so from 291 to 293 K we are
talking an increase of just under 3% and that's not 2 W/m2, it's more
like 10. The difference is water vapour, clouds and other feedbacks
(that are operative on medium scales, or in other words, ice sheet
extent changes, or changes in surface albedo due to different plants
growing there are specifically excluded).
This is fine for a forcing like CO2, but when the forcing affects the
feedbacks it wouldn't be.
gerh...@aston.ac.uk
Thanks for that. One of the beauties of this group is that it's
virtually all high quality.
Having read the presentation now (it took ages with my dial-up
connection to download), I'd say two things, one it's a nice
presentation with lots of interesting data,
and two
it's heavily out of date.
Chinese statistics claimed a fall in coal production in the 1990's.
There are people who claim that this fall wasn't real, and due to bad
Chinese statistics. What I've heard specifically is that the Chinese
government was less than impressed with the safety record of small
mines and that local government officieals therefore hid the
production of many small mines.
I don't know what the truth of that is, but the fall does appear a bit
strange in the face of (according to official statistics) pretty good
economic growth over the period. Maybe Chinese economic growth wasn't
actually all that grand during the Asian crisis and the coal
statistics are real. Maybe there was a big efficiency improvement.
But, over the last five years Chinese CO2 emissions have supposedly
doubled due to doubled coal production, and energy consumption's
supposedly grown as fast or even fast than GDP.
I've got some healthy skepticism when looking at Chinese statistics,
but taking the official pronouncements of the Chinese government and
guesstimates from other sources, it very much looks like the fall
suggested by that powerpoint was a temporary aberration.
James Annan
>> The atmosphere is pretty tightly coupled to the ocean, especially the
>> surface. I suppose it's possible that some altitudes would be affected
>> more significantly, depending on the details of aerosol behaviour (a
>> point that I'm not at all expert on).
>
> I've been thinking quite a bit about thermal lag, it's an area that I
> like, because of my background as a chemical engineer.
>
> A good comparison ought to be seasonal changes. Over a day forcing
> changes by about 1000 W/m2, over a season I don't know exactly, order
> of magnitude wise it ought to be around 200 W/m2.
>
> In Siberia, seasonal temperature changes I think approach 50C, in
> England it's more like 20C.
I think the UK is more like 10C and a global average is roughly half
that. This would suggest that on the global scale the effect would be
very much at (below) the bottom end of your range. Of course I agree
that on a local basis the elimination of all aerosols could have rather
larger effects.
>> Well unless one postulates an extremely high sensitivity to such forcing
>> (but not to other forcings), one would expect the warming to be
>> levelling off - and there's no sign of that except through some rather
>> specious cherry-picking vis a vis 1998...
>
> I suppose, playing devil's advocate here, the recent lack of levelling
> off might be due to say the declining aerosol forcing Michael Tobis
> mentions ("the trend in aerosol emissions from Chinese power plants is
> sharply downward").
I don't think there is any suggestion that aerosols globally have
actually declined significantly (the emissions rate per power output may
have improved, but total power has increased). Crowley's forcing data
has aerosols monotonically increasing to 1998.
>
> Not being a climate scientist I am a little confused about what the
> concept of forcing really means in the case of cosmic rays. My
> understanding is that the forcing is measured in W/m2 at the ground,
> assuming no feedbacks from water vapour, clouds etc.
Roughly speaking it is at the top of the atmosphere assuming everything
else is held fixed (although there are a few variants of "top" and
"everything else").
James
gerh...@aston.ac.uk
> that. This would suggest that on the global scale the effect would be
> very much at (below) the bottom end of your range. Of course I agree
> that on a local basis the elimination of all aerosols could have rather
> larger effects.
I've looked up the monthly averages for Birmingham and Ekaterinburg
and the difference between January and July is 13C and 34C
respectively.
I don't know the hemispheric averages for seasonal changes or the
forcing changes for the whole hemisphere, which I suspect would be
smaller than 200 W/m2.
The beauty of using the whole hemisphere is that relatively little
heat crosses the equator, while a lot of heat makes its way from the
tropics to England (?).
The beauty of looking at individual locations is that it gives an idea
of how important the difference between continental and ocean
locations is, and I suppose faced with a step change in aerosol
forcings, there'd be a much more pronounced difference in warming
rates between ocean/land than is the case now.
> I don't think there is any suggestion that aerosols globally have
> actually declined significantly (the emissions rate per power output may
> have improved, but total power has increased). Crowley's forcing data
> has aerosols monotonically increasing to 1998.
Michael Tobis presented a link to a presentation and summarised it as
showing a recent decline in Chinese aerosol forcing. Without having
read the presentation, I thought that might be due to lower sulphur
coal or recent pollution abatement efforts. After reading the
presentation, I think it was only based on the well known dip in
Chinese coal use at the end of the 1990's recorded by official Chinese
statistics, which was followed by a near unbelievable doubling within
five years.
gerh...@aston.ac.uk
A little googling gave me the hemispheric values. For forcing it's
only a graphic, but it ought to be around 200 W/m2, not below as I
thought.
For temperature, it's 13C for the Northern hemisphere and 6C for the
Southern hemisphere. The graphic of temperature change is quite
impressive, showing how substantial thermal inertia over the oceans
is.
So, I suppose, a step change in forcing of 1 W/m2 might indeed only
give a global jump of 0.1C within six months to a year, maybe it would
be 0.5C within 10 years, and 1C within 100 years.
James Annan
> Michael Tobis presented a link to a presentation and summarised it as
> showing a recent decline in Chinese aerosol forcing. Without having
> read the presentation, I thought that might be due to lower sulphur
> coal or recent pollution abatement efforts. After reading the
> presentation, I think it was only based on the well known dip in
> Chinese coal use at the end of the 1990's recorded by official Chinese
> statistics, which was followed by a near unbelievable doubling within
> five years.
Coincidentally, I heard a presentation a few days ago from our
atmospheric chemistry group here on this subject. The speaker said there
were 3 sets of widely differing statistics (ranging from a large drop to
a very small one that is really more of a temporary levelling off), but
their opinion (based on modelling and observations of pollutants) seemed
to be that reality was at or beyond the upper end of this range, with no
dip at all.
I think this paper covers it:
Hajime Akimoto, Toshimasa Ohara, Jun-ichi Kurokawa, and Nobuhiro Horii,
Verification of energy consumption in China during 1996?2003 by using
satellite observational data, Atmospheric Environment, 40(40),
7663-7667, 2006.
James
Hank Roberts
> I think this paper covers it:
>
> Hajime Akimoto, Toshimasa Ohara, Jun-ichi Kurokawa, and Nobuhiro Horii,
> Verification of energy consumption in China during 1996?2003 by using
> satellite observational data, Atmospheric Environment, 40(40),
> 7663-7667, 2006.
>
> James
That fits everything I've read. A summary here:
http://www.technologyreview.com/printer_friendly_article.aspx?id=17963
"... In November [2006], the International Energy Agency projected
that China will become the world's largest source of carbon dioxide
emissions in 2009, overtaking the United States nearly a decade
earlier than previously anticipated. Coal is expected to be
responsible for three-quarters of that carbon dioxide...."
http://yaleglobal.yale.edu/display.article?id=5058
"Some scientists have estimated that 30% or more of the mercury
settling into America's ecosystems comes from abroad - China, in
particular.
gerh...@aston.ac.uk
> actually declined significantly (the emissions rate per power output may
> have improved, but total power has increased). Crowley's forcing data
> has aerosols monotonically increasing to 1998.
When reading the following paper by Crutzen for other reasons:
http://www.springerlink.com/content/t1vn75m458373h63/fulltext.pdf
I am across this snippet:
"In fact, after earlier
rises, global SO2 emissions and thus sulfate loading have been
declining at the
rate of 2.7% per year, potentially explaining the observed reverse
from dimming
to brightening in surface solar radiation at many stations worldwide
(Wild et al.,
2005). The corresponding increase in solar radiation by 0.10% per year
from 1983
to 2001 (Pinker et al., 2005) contributed to the observed
climatewarming during the
past decade."
Incidentally, there's also an estimate of the Pinatubo forcing and
related cooling in there (4.5 W/m2 and 0.5 C in the year following the
eruption, so another way to confirm your estimate of 0.1C for 1 W/m2
in the first year).
jdannan
paper, which seems relevant:
http://www.agu.org/pubs/crossref/2005/2005GL023902.shtml
they tried 2 experiments: a "commitment run", ie running on a
continuation of a 20th century simulation, with atmospheric constituents
held fixed at current (2000) levels, and a similar run in which all the
sulphates were removed (but everthing else the same).
"Calculations performed with the Hamburg Climate Model suggest that
climate change resulting from increasing greenhouse gas emissions would
become considerably more pronounced if air quality were dramatically
improved in the future. Specifically, the globally averaged surface air
temperature and amount of precipitation could increase in less than a
decade by 0.8 K and 3%, respectively, if the entire amount of
anthropogenic sulfate aerosols were removed from the atmosphere."
I think the temperature value is about where we ended up in our discussions.
James