Memorandum 144
Submission from Stephen Salter, Emeritus
Professor of Engineering Design, Institute for Energy Systems,
University of Edinburgh
1. SUMMARY
- At a recent energy conference Simon Vasey,
trading manager of the major electricity provider Eon, said that
while profits of billions of Euros had been made from the first
round of the European carbon trading scheme not one kilogram of
carbon had been abated.
- The monthly addition of points to the Keeling
curve shows no reduction in the upward acceleration.
- Discussions of carbon emissions have used
per nation rather than per capita data. A judicious choice
of baseline date and the removal of shipping, aviation and the
proxy carbon associated with imported goods has allowed at least
one country to claim carbon reductions when in fact there has
been an increase.
- The track record of the IPCC with regard
to the timing of predicted events has been poor with several potential
positive feed backs, such as the loss of Arctic ice, happening
more rapidly than predicted in the earlier reports. People working
for the IPCC report privately that there is intense pressure to
modify wording from home governments.
- Ice core records show that have been many
abrupt rises in world temperatures of a size and rate that would
be catastrophic to a high world population. People who know a
great deal about the problem and who have been studying it from
the time when others thought it unimportant, now say that a sudden
rise, perhaps at the next el Niño event, is likely
and that, because the full effects of emissions lag their release,
we may already be too late.
- Even if there are strong reasons for not
deploying geo-engineering systems there is no case for not supporting
vigorous research into every possible technique and for taking
all feasible ones to the stage at which they could be rapidly
deployed. This view is not yet shared by DEFRA and UK funding
bodies.
- After 35 years work trying to develop renewable
energy systems I now believe that it may not be possible to deploy
enough of them quickly enough to prevent very serious consequences
of climate change. For the last four years I have been working
full time on the engineering design of one of the several possible
techniques. The idea, due to John Latham, former Professor of
Atmospheric Physics at the University of Manchester and now at
the Centre for Atmospheric Research at Boulder Colorado, is to
increase the reflection of solar energy from marine stratocumulus
clouds by exploiting the well-accepted Twomey effect. Engineering
drawings and design equations for a practical system are well
advanced and can be made available to your Committee.
- Like everyone working in geo-engineering
I do so with reluctance in the hope that it will not be needed
but fearful that it may be needed with the greatest urgency.
2. THE TWOMEY
EFFECT
1. Twomey says that, for the same liquid
water content, a large number of small drops will make a cloud
reflect more than a small number of large drops. We would expect
something like this from calculations of reflecting areas. We
can see it with jars of glass balls of different sizes. We talk
of dark storm clouds gathering when the drops become large enough
to fall.
2. Even if the relative humidity goes above
100% a cloud drop cannot form without some form of condensation
nucleus on which to grow. Over land there are plenty of suitable
nuclei, 1,000 to 5,000 per cubic centimetre of air. But in clean
mid ocean air the number is lower, often below 100 and some times
as low as 10. In 1990 Latham proposed that the number of condensation
nuclei could be increased by spraying sub-micron drops of sea
water into the turbulent marine boundary layer. Initially the
drops would evaporate quite quickly to leave a salty residue.
Turbulence would mix these residues evenly through the marine
boundary layer. Those that reached the clouds would provide ideal
condensation nuclei and would grow to increase the reflecting
area and so the cloud albedo.
3. The equations in Twomey's classic 1977
paper can be used to produce the graph below.
4. This follows the presentation used by
Schwarz and Slingo (1996) and shows cloud top reflectivity for
a typical liquid water content of 0.3 gm per cubic metre of air
for a range of cloud depths as a function of drop concentration.
The vertical bars show the range of drop concentrations suggested
by Bennartz (2007) based on satellite observations.
5. If we know the initial cloud conditions,
most especially the concentration of condensation nuclei, we can
calculate how much spray will produce how much cooling. The method
needs incoming sunshine, clean air, low cloud and the absence
of high level cloud. The position of the best places varies with
the seasons so sources should be mobile. Because the ratio of
solar energy reflected to the surface-tension energy needed to
generate drops is so large, it turns out that the spray quantities
are quite practical. In the right conditions a spray source with
a power rating of 150 kW can increase solar reflection by 2.3
TW, a ratio of 15 million. This is the sort of energy gain needed
if humans are to attempt to influence climate.
3. HARDWARE
1. The need to operate for long periods
in mid-ocean and to migrate with the seasons points to a fleet
of remotely operated wind-driven spray-vessels. These can obtain
the electrical energy needed to make spray by dragging turbines
like oversize propellers through the water. Thanks to satellite
communications and navigation remote operation is now much easier.
2. Rather than solve the robotic problems
of handling ropes and textile sails we propose to use Flettner
rotors. Flettner rotors offer much higher lift coefficients and
lift drag ratios than sails or aircraft wings but their main attraction
is that a computer can control the rotation speed of a cylinder
far more easily that it can tie a reef knot. Anton Flettner built
a ship, the Baden-Baden, which crossed the Atlantic in 1926. She
won a race against a sister ship with a conventional rig and could
sail 20 degrees closer to the wind. The weight of rotors was one
quarter of the weight of the rig that they replaced. Flettner
won orders for six ships and built one, only to have the orders
cancelled because of the 1929 depression. Modern bearings with
spherical freedom and materials like Kevlar and carbon-fibre would
make rotors even more attractive. Enercon, the major German wind
turbine maker launched a 10,000 tonne rotor assisted ship on 2
August 2008. The television company Discovery Channel has funded
successful trials of a 34 foot yacht conversion. They also carried
out an experiment at sea which confirmed expectations of the very
high energy gain offered by the Twomey effect.
3. Design calculations and general arrangement
drawing of the first spray vessel are well advanced. It has a
waterline length of 45 metres and a displacement of 300 tonnes.
Early vessels have space for a crew as well as the option to transfer
control to an auto pilot and from land. Future ones may be a little
smaller. All sensitive equipment is in hermetically sealed cylindrical
canisters which can be individually and thoroughly tested on land
and quickly exchanged. With three spray systems it will be possible
to spray 30 kg a second as 0.8 micron drops. A fleet of 50 vessels
in well-chosen places could cancel the thermal effects of the
present annual increase of greenhouse gases. Work packages and
costings for a five-year development programme which would provide
a reliable tested design for the ocean going hardware are available.
4. The change of cloud reflectivity necessary
to stabilize global temperature despite a doubling of pre-industrial
CO2 is about 1.1% globally or 6% if evenly spread in cloudy areas.
The contrast-detection threshold for fuzzy irregular patterns
is much higher, about 20%. It will be necessary to develop a method
to convince non-technical decision makers that anything has changed.
The spray generation modules have been designed so that one of
them can be fastened to the hull of a conventional ship and can
produce spray at 10 kg a second, drawing electrical power from
the ship system. The ship would sail to a selected mid-ocean site
and then drift to a sea anchor so as to minimize its own exhaust
emissions.
5. The MODIS AQUA satellite system
crosses most of the world at the same local time each day. We
would download photographs of the shortwave radiation signals
(channels 1, 3 and 4). These would be translated to align the
ship positions and then rotated to bring the mean wind directions
to be coincident. Multiple images of the cloud system would be
added over a period of a few weeks. The random clouds should average
to a medium grey with contrast of the wake improving with the
square root of the number of photographs. Photographic superposition
will allow the measurement of the result of a very small spray
release.
4. POTENTIAL
SIDE EFFECTS
1. Our understanding of the world's climate
system is far from complete because it is so difficult to carry
out controlled experiments over the size range from condensation
nuclei to continental weather systems. All geo-engineers are anxious
about unintended consequences. Early models show that very large
spray injections can have effects in either direction at long
distances from the injection site in the same way that el Nino
events can influence climate far from Chile and Peru. We also
know that release from different sites can have quite different
results. We therefore must regard the world climate system as
having a large number of possible controls set by when and where
we choose to release spray. So far, we have no idea about which
control does what. However it should be possible to learn by a
series of very small experiments using release patterns modulated
on and off at the right periods in a known sequence followed by
the measurement of the long-term correlation of climate parameters
with the known input. This pseudo random binary sequence technique
works well with analysis of communication networks without being
noticed by users.
2. Modern computers do allow increasingly
sophisticated analysis and prediction. Recently there has been
a great deal of progress on computer simulation of all the effects
of albedo control. The leading team is at the National Centre
for Atmospheric Research at Boulder Colorado and is led by Philip
Rasch using the most advanced fully-coupled air/ocean model. This
produces results for nearly 60 atmospheric parameters presented
as maps, zonal graphs and mean values. Evenly spread releases
are less damaging than large point injections.
3. The amount of salt that cloud albedo
control will inject into the atmosphere is orders of magnitude
below the amount from breaking waves, some of which falls on land.
The difference is that albedo control uses a carefully chosen,
narrow spread of drop diameters.
4. The immediate effect of cloud albedo
control will be a reduction of solar energy reaching the sea.
The ocean temperatures are the primary driver of world climate
but oceans are a very large thermal store so the effect will be
slow. Currents and winds are efficient ways of distributing energy
and sharing it with the land so the eventual effects will be well
distributed. A short term engineering approach to choosing a cooling
strategy would be to look at historic data on sea temperatures
and attempt to replicate a pattern thought to be good with regard
to sea levels, harvests, hurricane frequency, floods and droughts.
Rather than thinking of the side-effects of we should really be
studying the side effects of NOT doing albedo control and letting
sea temperatures rise. We would then decide which of the outcomes
was the least damaging.
5. A first effect of warmer seas is greater
evaporation. Even though it is left out of many diagrams showing
the effects of greenhouse gases, water vapour contributes at least
an order of magnitude more global warming than carbon dioxide.
6. The second effect of warmer water flowing
north is the loss of summer Arctic ice.
7. A third effect is that surface water
temperatures above 26.5 C increase the probability and severity
of tropical cyclones, hurricanes and typhoons.
8. Warmer surface water increases the density
difference between it and the nutrient-rich cold water below it.
If nutrients cannot flow to where there is light there will be
no phytoplankton to act as the start of the marine food chain
or as the source of dimethyl sulphide and a sink for carbon dioxide.
At present dimethyl sulphide accounts for about 90% of the cloud
condensation nuclei, (Charlson 1987) and sea warming will reduce
the area producing it.
9. The sea has been soaking up much of the
anthropogenic CO2. Rising temperature will release it.
10. Very large amounts of methane are stored
in permafrost and even larger amounts as clathrates in the seabed
at depths of a few hundred metres. The release of either could
be regarded as an extreme side-effect of warmer seas and has been
linked to the Permian extinction.
11. So far the only suggested negative effect
of increasing cloud condensation nuclei is the possibility of
reduced rainfall, something that people in Britain and Bihar would
greatly welcome. The production of rain is a very complex process.
A gross engineering over-simplification is that rain needs quite
large drops to fall through deep clouds collecting smaller drops
in their path so that they get big enough not to evaporate in
the drier air below the cloud before they reach the ground. It
is known that too many small drops due to nucleation from smoke
from bush fires can reduce rain.
12. Clearly we must be cautious about doing
albedo control up-wind of a drought-stricken region. However the
driest regions are dry because subsiding air prevents winds blowing
in from the sea. Perhaps a larger temperature difference between
land and sea could produce a stronger monsoon effect to oppose
part of the subsiding flow.
13. The effects of the nuclei that we produce
will fade quickly. The marine stratocumulus clouds we will be
treating are usually not deep enough to produce rain. But we could
argue that if they were, the immediate effect would be to stop
the rain over the sea and coastal regions. This would leave more
water vapour in the air to give rain further inland where its
value will be greater.
14. If we do not yet know enough about the
side-effects of albedo control, at least we know more than about
those of uncontrolled temperature rise. But the strongest defence
is that we can start with small steps, move away from places where
problems occur and stop in a week if some natural event, such
as a volcanic eruption, should provide unwanted cooling.
5. POLITICS
1. Control of the UK climate is in the hands
of DEFRA. Official funding goes to many laboratories who tend
repeat the conclusions from the previous funding that the climate
problem is even more serious than previously thought and argue
that more funding is necessary to find out how much more serious.
There is a reluctance to fund any research into technology which
is "not yet soundly proven". The present DEFRA policy
is that carbon reductions are the best solution to the climate
problem and also that they should be the only solution on the
grounds that the possibility of alternatives might reduce pressure
to reduce emissions. This is strikingly close to the view of senior
officers in the RFC in world war I that issuing parachutes to
pilots "might impair their fighting spirit". They were
not even allowed to buy their own. The geo-engineering community
agrees with the rank order of desirability of emission reduction
to geo-engineering but asks "what progress in emissions reduction?"
2. People from the vigorous carbon trading
market are emphatic that there could be, even should be, no parallel
thermal trading equivalent and so it seems that, at present, there
is none of the commercial return needed to attract research funding.
Many geo-engineers agree that decisions about deployment should
not be based on commercial considerations.
References
Bennartz R 2007. Global assessment of marine boundary
layer cloud droplet number concentration from satellite. Journal
of Geophysical Research, 112, 12, D02201, doi:10.1029/2006JD007547,
From http://www.agu.org/pubs/crossref/2007/2006JD007547.shtml
Bower K, Choularton T, Latham J, Sahraei J and Salter
S 2006. Computational assessment of a proposed technique for global
warming mitigation via albedo-enhancement of marine stratocumulus
clouds. Atmospheric Research 82 pp 328-336.
Charlson RJ, Lovelock JE, Andreae MO and Warren,
SG April 1987. Oceanic phytoplankton, atmospheric sulphur and
climate. Nature 326 pp 655-661.
Latham J 1990. Control of global warming. Nature
347 pp 339-340.
Latham J 2002. Amelioration of global warming by
controlled enhancement of the albedo and longevity of low-level
maritime clouds. Atmos Sci Letters. 2002 doi:10.1006/Asle.2002.0048.
Latham J, Rasch P, Chen C-C, Kettles L, Gadian A,
Gettleman A, Morrison H, and Bower K, 2008 Global temperature
stabilization via controlled albedo enhancement of low-level maritime
clouds. Phil. Trans Roy Soc A Special issue October 2008.
Salter SH, Latham J, Sortino G, Seagoing hardware
for the cloud albedo control of reversing global warming. Phil
Trans Roy Soc A Special issue October 2008.
Schwartz SE and Slingo A 1996. Enhanced shortwave
radiative forcing due to anthropogenic aerosols In Clouds Chemistry
and Climate (Crutzen and Ramanathan eds.) pp 191-236 Springer
Heidelberg.
Websites
About parachutes: http://www.spartacus.schoolnet.co.uk/FWWparachutes.htm
Collected papers http://www.see.ed.ac.uk/~shs
INDOOR DEMONSTRATION OF THE TWOMEY EFFECT
The jar on the left is contains 4 mm clear glass
balls and has an albedo of about 0.6. The one on the right has
glass balls one hundredth of the size and an albedo over 0.9.
September 2008
| 
