Poster's note : this background article is interesting as it touches on
issues which apply to all the SRM techniques.

NB I'm going to be posting some more 'back to basics' content soon,
including classic papers, basic science guides, etc., clearly marked with
<b2b> in the subject line. If anyone doesn't like this idea, please reply
with "shut up Andrew" in the subject line.

http://www.newscientist.com/article/mg22329850.800-burning-blue-sky-earths-cloud-shield-is-failing.html?full=true#.VL-rRyUYbFo

Burning blue sky: Earth's cloud shield is failing - environment - 08
September 2014 - New Scientist
Stephen Battersby
Magazine issue 2985.

It's the clouds that stop the oceans boiling. But as the planet warms, our
main defence against the sun's fierce heat is weakening

(clipped)

Clouds have a vital role that few people appreciate: their overall effect
is as a global heat shield, reflecting sunlight that would otherwise bake
the Earth and obliterate life.

Much depends on what happens to this heat shield as the planet warms. It
might grow a little stronger, slowing the warming somewhat. Or it could
weaken, meaning the world will warm even faster. This is a crucial question
because it could mean the difference between a planet that is 3 °C hotter
next century – very bad but probably survivable – or 6 °C – which would be
catastrophic. To narrow this range of uncertainty we need to understand
clouds much better.

In recent years, we have started to make progress. It is now clear, for
instance, where climate scientists should be focusing their attention. "We
are hot on the trail, in a way that we haven't been before," says Bjorn
Stevens at the Max Planck Institute for Meteorology in Hamburg, Germany.
That trail leads to Earth's tropical seas, where great expanses of low
cloud exert a powerful influence over the climate of the entire planet.

Like all clouds, they trap heat below them in the form of long-wave
infrared radiation. This is why temperatures fall less on cloudy nights
than on clear ones. But clouds also reflect some sunlight straight back
into space and, less obviously, act as radiators, emitting infrared to
space from their tops. So a cloud is a parasol, blanket and cooling fin all
at once.

The overall result depends on the height and type of clouds. Low clouds
cool the planet: although they trap some heat, they also reflect a lot, and
their fairly warm cloud tops emit a lot of heat to space. High clouds emit
much less from their colder cloud tops, and often reflect little too, so
they help warm the planet.Low cloud is more widespread than high, which is
why clouds cool the planet overall. In fact if you were to strip away all
clouds, it might lead to a runaway greenhouse effect that would eventually
boil away the oceans, according to calculations published in 2013 by Colin
Goldblatt at the University of Victoria in British Columbia, Canada. That's
not going to happen, but we do need to know how clouds will change in a
warmer world.The best way to find out, you might think, would be look at
how clouds have changed over the past century as the planet warmed by 1 °C.
This turns out to be extremely tricky. If you have ever been mesmerised by
writhing wisps of cloud, you will appreciate that they are rather hard to
pin down.

Every approach to cloud observation has some shortcomings. Weather stations
on land are no use for the more widespread and important ocean clouds.
Observations from ships are patchy and subjective. Instrument-laden planes
are scarce. Weather satellites give some insights, but drift and decaying
orbits plague their data. And the dedicated climate satellites of NASA's
Earth Observing System have only been watching clouds for a decade or so,
not long enough to catch long-term trends.Even if we did have a good global
record of cloud behaviour, it might not be a reliable guide to what happens
when the planet gets even warmer. As the temperature soars we might pass
some threshold that produces big changes in cloud behaviour.

If we understood exactly how clouds work, we could predict future behaviour
in a climate model. But cloud computing isn't easy. The inner workings of a
cloud involve turbulent flows of air on scales ranging from a few
kilometres to a few metres. This is invisibly small to global climate
models, which slice the atmosphere into cubes a hundred kilometres wide.
Specialised small-scale models can now capture eddies down to a hundred
metres or so, but these cannot encompass large weather systems.On even
finer scales inside clouds, droplets of water and crystals of ice are
colliding, coalescing, condensing and evaporating. Many of these processes
– collectively known as "microphysics" – are well understood, but not all
of them. Zoom in even more, and you see that clouds cannot form without a
fine mist of aerosols: airborne particles less than a micrometre across
that act as nuclei around which water can condense or freeze. With more
particles you may get a whiter, longer-lived cloud, making a better
parasol.Models cannot capture all of these processes, so they have to rely
on approximations, such as the observed relationship between cloudiness and
humidity, say, or temperature. These relationships can then be plugged in
to the models. But as we have seen, observations are not perfect, so we
have no universal relationship between all the properties of the atmosphere
and the amount and type of cloud that you should get.

That gives modellers too many options. For example, cloud cover correlates
well with the temperature difference between ground level and 3 kilometres
up. But it correlates equally well with another measure that includes both
temperature and humidity. Model results vary depending which option is
chosen. "They give completely different predictions for what happens when
Earth warms up," says Steven Sherwood at the University of New South Wales
in Sydney, Australia.Despite these difficulties, there has been progress
with some types of cloud. Models and observations agree that high clouds
will, on average, be pushed higher still as temperatures rise. That makes
their cloud tops even colder, so they become less effective at radiating
heat. Meanwhile, storm tracks will probably shift towards the poles, where
clouds reflect less solar heat. Both of these factors will act to amplify
warming.

A much more important part of the global heat shield is found in the
tropics and subtropics, where great blankets of low stratocumulus cloud
stretch over much of the oceans on most days. Here the models clash. Some
predict almost no change in these low clouds, others a sharp decline that
amplifies global warming.

To work out which point to the real future, we need a better understanding
of what might be going on above those warm tropical waters. "In general I
find physical mechanisms to be more compelling than 'my model predicts X so
it must be true'," says Peter Caldwell at the Lawrence Livermore National
Laboratory in California. The past few years have seen a flurry of new
mechanisms explored.

Some look like good news: they are "negative feedbacks" that act to slow
warming, the warmer things get. For example, where warm, dry air descends
towards tropical oceans as part of a global circulation pattern it can trap
sheets of low, cooling stratocumulus cloud. With warm air above and cooler
air below – a temperature inversion – the air cannot rise and lose its
moisture by raining. And as global temperatures rise, the warm downdrafts
should get warmer, strengthening the inversion effect and increasing cloud
cover on average.

Trapping more heat

At least, that is what observations and small-scale models suggest,
Caldwell and his colleagues reported last year. But it is only one
mechanism. "I think this negative feedback will be offset by a variety of
positive feedbacks," says Caldwell. "I'm on the fence about whether
stratocumulus will increase or decrease, though most of my colleagues seem
to think it will decrease."

That is because they have realised several positive feedbacks could be at
work. For one thing, the clouds could be starved of moisture. Low clouds
get their moisture by a roundabout process: as heat radiates from the cloud
tops, cold parcels of air form and sink down. This pushes up damp air from
near the sea surface, which forms more cloud as it cools and condenses. In
2009, two teams that included Caldwell and Stevens pointed out that rising
greenhouse gas levels will trap more heat, reducing heat loss from the
cloud tops. That means less cooling, less sinking air, less moisture
dragged up and less cloud cover on average.

Or clouds could lose moisture to the dry air above. Even where a
temperature inversion traps the clouds, there is some mixing between damp,
cool air below and dry, warm air above. As Stevens and colleagues suggested
in 2012, warming could drive stronger updrafts from below and increase this
mixing, dissipating the vital water. The result would be reduced cloud
cover and amplified warming.

Even if mixing doesn't get stronger, there could still be more moisture
loss to the dry air above. Warmer air can hold much more water vapour, so
in a warmer world a given air current will carry more moisture away.

To find out how big this feedback could be, Sherwood's team recently looked
at data from weather balloons to see how much mixing there is today
(Nature, vol 505, p 37). It turned out to be pretty vigorous – more than in
many models. "Models that have more mixing are closer to the truth," says
Sherwood.Different models suggest that a doubling of CO2 could lead to
warming of anywhere between 1.5 °C and 4.5 °C in the short term – a figure
known as climate sensitivity. But the models with realistic mixing are the
ones with greater sensitivity, Sherwood found. If they are to be trusted,
then Earth's short-term sensitivity will be 3 °C to 4.5 °C.

"This work is a great step in the right direction, but I don't think it is
definitive," says John Fasullo at the National Center for Atmospheric
Research in Boulder, Colorado. One problem with Sherwood's approach, he
says, is that observations of mixing are limited – relying on a scattering
of weather balloons – so it may be difficult to confirm the theory.Fasullo
prefers to compare cloudiness directly with humidity, which can be measured
globally by satellites. In 2012 he showed that models often overestimate
the humidity in the subtropics. His finding was also bad news: the models
with more realistic low humidity tended to predict greater warming
(Science, vol 338, p 792).

These findings are casting some light on the great cloud conundrum, but it
is still rather a dingy grey light, just hinting at which models might be
most trustworthy. "Are the more 'successful' models getting the right
answer for the right reasons?" asks Fasullo.

As computer power grows we can build models with finer resolution, but we
won't reach some paradise of perfect modelling. Even ignoring the
microphysics, important air movements are happening on scales as small as 5
or 10 metres. It will be several decades at least before global models can
include such fine detail. So models must keep using approximations for this
small-scale stuff, making it all the more important to test them against
direct observations.

Feeding clouds

One answer may be to make the best of weather satellites. "For climate you
need a stable observing network, but the weathersats that show clouds on
the evening news were never intended to be stable in that way – if a sensor
degrades or the orbit drifts a bit you can still see where a hurricane is,"
says Joel Norris at the Scripps Institution of Oceanography in San Diego,
California. With a thin cloud layer, whether you see it all depends on the
angle you look at it, so as satellites spiral closer to Earth, their record
of cloudiness can be distorted. They can also move geographically so they
are seeing a given spot later in the day when there is typically more or
less cloud.

To some extent these distortions can be corrected, and Norris is now
working to do that with two of the main weathersat databases.

We need to watch not only the visible clouds, but also their invisible
vaporous foodstuff. "Water vapour is the single most important variable,"
says Stevens. "If you ask how good are our global measurements – well, it's
a crime, we are off by tens of per cent. But the great thing is we have
some instruments now that can measure water vapour accurately." Raman
lidars can fire a laser into the air and measure the spectrum of light
scattered back by water molecules. "We need more of those – and also in
space," says Stevens.

So we haven't mastered the science of clouds yet. But both observations and
models suggest that far from coming to our rescue, clouds are going to
suffer along with us. And many independent lines of evidence point to the
same conclusion. Looking at past climates, for instance, cannot tell us how
clouds behaved but does reveal strong warming in response to rising
greenhouse gas levels. A slew of studies published this year all conclude
that the climate's sensitivity to CO2 is at the higher end of the range.
The forecast, then, is disturbingly clear and uncloudy.

This article appeared in print under the headline "Clearing skies"

Stephen Battersby is a freelance writer based in London
>From issue 2985 of New Scientist magazine, page 42-45.

-- 
You received this message because you are subscribed to the Google Groups 
"geoengineering" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/geoengineering.
For more options, visit https://groups.google.com/d/optout.

Reply via email to