Sea-Going Hardware for the Implementation of the Cloud Albedo Control Method for the Reduction of Global Warming. Stephen Salter 1, Graham Sortino 2 1. School of Engineering and Electronics, University of Edinburgh.
[email protected]. 2. School of Informatics, University of Edinburgh.
[email protected]
ABSTRACT. The ideal solution to the problem of global warming would be for everybody to change immediately from fossil fuel to cheap, reliable and plentiful renewable sources of energy. But China will be building 1000 MW of coal-fired plant every four or five days for the next eight years and India the same every two weeks. World fossil fuel consumption rose by 4% in 2004. Coal consumption rose by 6%. Our leaders are betting the planet on the hope that carbon sequestration, renewable energy and improved efficiency will be able to compensate. The 2004 increase in the rate of rise of CO2 to 2.5 ppm suggests that the goal has not yet been achieved. If the leaders are wrong and the Siberian permafrost melts to release its stored methane, the global temperature rise will accelerate even more than now, even if we were to succeed in reducing carbon emissions. Back-up plans are urgently needed. This paper describes the hardware of a plan based on John Latham’s 1990 proposal to exploit the Twomey effect. Wind-driven spray-vessels will release micron-sized drops of sea water beneath marine strato-cumulus clouds. This will increase their albedo and so reflect enough solar energy back out to space to allow double present CO2 levels with no change of mean global temperature. Keywords: Global warming, Carbon dioxide, Twomey effect, albedo control, cloud condensation nuclei, spray generation, spray vessel, atomizer, Flettner rotor, Thom disk.
Albedo basics
Marine stratocumulus clouds are not permanent bodies but are more like banks of indicator lights giving information about local temperature and humidity. The clouds are there because of the random updrafts, sometimes with velocities up to one metre per second, which have produced a temperature drop. The clouds will vanish when turbulence moves the air down. We can therefore be confident that dispersion through the marine boundary layer will be reasonably fast. The lifetimes of sea water aerosol are set by coagulation and coalescence and are of the order of a few days, so that the control scheme has a rapid response, is locally variable, readily reversible and ecologically benign. Reversibility has a particular advantage.
The albedo of a body is the fraction of light that it reflects. The albedo of marine stratocumulus clouds, which cover about one quarter of the ocean surface, is between 0.3 and 0.7. Sean Twomey (1977) wanted to understand the variations in cloud reflectivity . He found that the albedo was set by the number and size of the cloud droplets. For a given amount of water, a large number of small droplets make clouds reflect more than a small number of large drops. Increasing global cloud albedo by only 13% would produce a cooling sufficient to offset the warming due to a doubling of current CO2. The front cover of this paper is a NASA satellite image which shows the inadvertent demonstration of the Twomey effect. The long streaks are Good places to do the spraying must have plenty of incoming caused by sulphates in the trails of exhaust from ship engines. sunshine to give something to reflect for much of the time. They must have the right kind of low-level marine John Latham (1990) suggested that the Twomey effect could stratocumulus clouds the kind with flat bases and fluffy tops. halt global warming if we deliberately introduced sub-micron They should have few high clouds because we cannot get drops of sea water below low-level marine stratocumulus condensation nuclei through the calm air above the marine clouds. Clean marine air masses normally have a deficit of boundary layer, and the bases of high cloud can reduce cloud condensation nuclei, often below 100 per cm3 and incoming energy and reflect energy back again. There should sometimes as low as 10 per cm3. The salty residue left after the be steady, reliable, brisk but not extreme winds to give plenty evaporation of a small drop of sea water is an ideal cloud of drive power. There should be a low density of shipping and condensation nucleus. For the middle of the albedo range the icebergs. There should be a low initial density of cloud fractional change in albedo is about one-twelfth of the natural condensation nuclei because it is the fractional change that logarithm of the change in nuclei concentration. A doubling of counts. This suggests sea areas distant from dirty or dusty the concentration increases albedo by just over 5.5%. Less land upwind. Because of a possible anxiety over the effect of spray will be needed if the initial concentration of nuclei is low. extra cloud condensation nuclei on rainfall we should avoid Spraying of the drops should be done where there is a frequent being close upwind of land with a drought problem. occurrence of the right kind of cloud as shown in figure 1. It will also be more effective where there is a high solar input. This Regions in which marine currents are flowing towards the Arctic would be of special interest because cooling this water will varies through the seasons as shown in figure 2. preserve Arctic ice cover, which is itself a powerful reflector of Latham calculated that the quantities of spray needed in solar energy and because we might achieve a reduction in the suitable regions are surprisingly small. An annual increase of release rate of methane from the melting of Siberian and the spray rate by 10 E18 drops per second would allow the Canadian permafrost. present rate of rise of CO2 to continue with no temperature rise. If the spray were done from 50 new sources each year, spraying There is a vast amount of existing information on most of these one micron drops, the water mass would be only ten kilograms parameters but it is distributed between many computers around the world and has been saved in different formats with of sea water per second from each source. different spatial resolutions and sampling rates using different Meteorologists call the lowest few hundred metres of the recovery software and different access protocols. This has atmosphere over the sea the marine boundary layer. The been collected, decoded, interpolated, unified and stored in average depth is about 850 metres. Friction between wind and database which allows selective interrogation by an efficient water generates quite high levels of turbulence and so mixing parsing routine, Sortino (2006). Examples will now be discussed. is rapid but stops sharply at the top of the boundary layer.
Spray places Figure 1 from ISCCP shows the distribution of cloud drop radius for the extremes of January and July together with a whole-year mean. We can see that drop diameters over the sea are always above 20 microns and sometimes above 30 microns. This is far above the target spray diameter.
Present solar reflection.
The reflections are spread round the full 360 degrees of longitude with values down by a factor of about three relative to the peak noon value. There is a large area of low reflection over Africa and Asia and lots more reflection in December from the southern hemisphere than in June from the north. Much more from http://www.cdc.noaa.gov/PublicData/.
Figure 2 shows data taken from the NOAA-CIRES Climate Diagnostic Centre database and gives the amount of solar energy at present being reflected out to space from cloud tops. Observations are for ten days either side of the winter (top map)and summer solstices averaged over the whole 24 Clearly spray sources must have seasonal mobility. hour period.
We then use the wind speed data for each cell and the fact that spray rate will go with the cube of wind speed subject to an upper If we are to have double present CO2 with no temperature limit of about 10 kg per second to get spray production from one change then the world needs to increase solar reflection by vessel for that place and time. about 4 watts per square metre, or 2000 Terawatts if it was spread evenly over the whole earth. We want to know how The present design is aimed at a maximum spray release rate many spray vessels with how much spray equipment placed of 10 kg per second but it looks as if this would often saturate a where at which season will be needed. Some of the existing concentration of 65 condensation nuclei per cm3 if drop lifetime data have now been assembled Calculations can be done turned out to be as long as one week. separately for each of the 6596 equal-area (7.72E10 m2) cells in the Sortino maps. The greatest uncertainties concern the most We place the cells in rank order to see how many we need to important of the parameters. These are the estimate of the treat to achieve any target cooling and either how many vessels present number of cloud condensation nuclei at various times we should put in each cell or how many cells should be treated and places and the drop lifetimes. This is because it is the by one vessel. By looking at the best cell list for the next month fractional change in drop numbers in clouds that drives the we can plan vessel movements
A sample calculation
change in albedo. In the absence of better data we can multiply the depth of the marine boundary layer (say 850 metres) in each cell of the map by the figures for the present concentration of condensation nuclei. This gives the number of drops in the treatment volume of each cell. For the average 65 condensation nuclei per cubic cm suggested by Bennartz 2007, the present number per cell is 4.27E21.
It seems astonishing that one wind-driven spray vessel with a power rating of a small wind turbine - a few hundred kilowatts could trigger an increase in reflected solar energy approaching the electrical output of nearly 300 nuclear power stations. But we can get a cross-check by working back from the performance of the maximum theoretical energy gain.
The amount of energy that the drop will reflect depends on its projected surface area times the incoming power density times We then look up the incoming radiation in watts per square a geometric reflection coefficient times the drop lifetime. The metre averaged over 24 hours and multiply by the present reflection coefficient can be obtained from a complicated albedo for each cell and month to get the amount of energy scattering equation which depends on drop-diameter, wavelength and polarization. It is highest when diameter is being reflected now. comparable with wavelength, see Laven 2006.
For the very best cells in the NOAA maps of figure 2 at the December solstice in the South Atlantic, South Pacific and Indian Ocean this reflection can reach 180 watts per square metre. It is over 130 watts per square metre from latitude 30 to latitude 80 south for all longitudes. In the northern summer it reaches these values for about half the angles of longitude. The effect of injecting ten litres of sea water per second as one micron drops, evenly distributed through the boundary layer will be to increase the number of drops per cell by 1.91 E16 per second. It will take some time for turbulence to move the spray drops through the boundary layer and some further time before they wash out or coalesce with large drops. The lowest estimate for drop life is about one day giving an increase of 1.65E21 in each cell. For these guesses, the ratio of (old plus new) divided by old is 1.387.
The energy needed to create the original seeding drop by a perfectly-efficient spray-generator depends on its total surface area times the surface tension of sea water, which is 0.078 Newtons per metre. We divide the two quantities of energy. The drop diameter and cancel out, leaving the energy ratio as growth-factor squared times incoming power-density times the geometrical reflection coefficient times the drop-life, all divided by 4 times the surface tension of sea water. Consider, for example, estimates for a place where the 24-hour incoming solar power density at the cloud top is 340 watts, assume a drop growth to 15 microns, a reflection coefficient of 0.1 and a drop life of one day. (We would be most grateful for all suggestions of more accurate values.)
To use the Twomey equation we take the natural log of this nuclei gain ratio, which is 0.327, and multiply by 0.075 which The resulting ratio of reflected solar energy to drop creation energy is 2.12 E9: yes, more than two thousand million. If the gives the fractional increase in reflection of 0.025. efficiency of the spray mechanism was only 1% and only one We multiply this by the present reflected energy, say 150 watts of every thousand drops was fortunate enough to reach the per square metre over 24 hours for a good cell at a good time of reflecting level at the cloud top, the energy gain would still be the year, to get a reflected power of 3.679 watts per square over 200,000. metre. Engineers often have to make this sort of calculation based on We multiply this by the cell area of 7.72 E10 square metre to get inadequate information at the start of a project and then refine the reflected power of 284 GW for that cell for the present their estimates by the collection of existing data and guesses. It does not matter if the spray was spread outside the measurement of those that are unknown. If it is not possible to reduce CO2 emissions then there is the most urgent need to cell are because the lower reflection will be from a wider area. understand fully the present concentrations of potential We perform the calculation for every cell which has not been condensation nuclei, the concentration, growth rate and excluded on grounds of being downwind of land with dirty air, lifetime of small drops and the depth of the reflecting layer at upwind of drought-stricken regions or too close to busy shipping cloud tops. The community of meteorologists and cloudphysicists and their funding bodies could be criticised for our routes. present ignorance on such important matters.
The combination of data In figure 3 we see maps for four seasons showing suitability of different sea areas based on the combination of one possible set of the selection criteria in the table below. The software will redraw a map for any other selection choice by the interrogation of 140 Gbytes of data.
We can see that while the southern oceans are better in the southern summer there are useful regions in the approaches to the poles where the extra ice benefit may be obtained . The very best of all-year places are off the coasts of California, Peru and Namibia.
Variable
Jan-March
July-Sept
Weight /10
Percentage of Stratocumulus clouds
+10
Percentage of medium and high cloud
-5
Albedo of low stratocumulus
-3
Estimated low cloud drop concentration
-3
Incoming short wave energy at 60 mbar
+10
Boundary layer height
+3
Cloud base height
-6
April-June
Oct-Dec
Figure 3. Parameter combination based on one possible set of selection criteria. Red is best but yellow is fine.
Energy As well as having the right kind of cloud, we will need energy to make the spray. The proposed scheme would draw all the energy from the wind. Numbers of remotely controlled spray vessels would sail back and forth, perpendicular to the local prevailing wind. The motion through the water would drive underwater ‘propellers’ acting in reverse as turbines to generate electrical energy needed for spray production. These would need to be much larger than the propellers needed for propulsion of a similar-sized vessel. Batteries would store energy for intricate maneuvering or for several hours of propulsion in windless conditions. Each yacht would have a global positioning system, a list of required positions and satellite communications to allow the list to be modified from time to time, allowing them to follow suitable cloud fields, migrate with the seasons and return to land for maintenance. The problems of the remote operation of ropes and reefing gear would be avoided if we used Flettner rotors instead of sails. These are vertical spinning cylinders which can produce forces perpendicular to the apparent wind direction which are very much larger than forces from textile sails of the same area. They were used successfully for ship propulsion in the 1920s, Seufert and Seufert (1983). They allow a sailing vessel to turn about its own axis, apply ‘brakes ’ and go directly into reverse. They even allow self-reefing at a chosen wind speed. Figure 4 shows an artist ’s impression of Flettner rotors driving a flotilla of spray vessels.
Anton Flettner built two sea-going ships. The first crossed the Atlantic in 1926. He obtained orders for six more, only to have them cancelled as a result of the 1929 depression. The rotor system weighed only one quarter of the conventional sailing rig which it replaced. The power to spin the rotors was about 10% of the power of the ship ’s engines. Flettner used drums of steel and later aluminium. Today much lighter ones could be built with glass or carbon-reinforced epoxy materials. It is well known that part of the drag of an aircraft wing is the continuous vortex generated by air moving from the high pressure below the wing to the low pressure above it around the wing tip. The effect can be reduced by high aspect-ratio wings with narrow tips as used by the albatross and Burt Rutan’s wonderful aircraft designs. Flettner put end caps at the the of his rotors for the same reason. In the ‘thirties Alexander Thom (1934) experimented with series of fences along the entire length of the rotor. Hew found that fences three times the rotor diameter at intervals of 0.75 of the diameter could increase the lift coefficient from 9 in Flettner’s ships to 25 but would also reduce the drag coefficient to give a lift-drag ratio of 35 at a lift coefficient of 18. Unfortunately he greatly over-estimated the power needed to drive larger versions of the modified rotors and abandoned work. The mistake was spotted in the ‘nineties by Joseph Norwood who used the correct scaling equations and confirmed them with experiments. He used Thom fences on Flettner rotors to design very exciting sailing vessels.
Figure 4. Albedo spray vessels. They would sail back and forth square to the local prevailing wind. Flettner rotors with Thom fences can give lift coefficients up to 20 and lift drag ratios of 35, much higher than cloth sails. Artwork by John MacNeill.
Spray technology. It is the NUMBER of condensation nuclei disseminated, not the mass of spray, which matters. Many spray generation methods are available. The commonest is a Venturi passage sucking liquid into a high velocity air stream as used in carburetors but this produces a very wide range of drop sizes. A very narrow range of drop diameters can be achieved if a jet of liquid is directed to the centre of a spinning disk. The liquid is centrifuged out to a sharp-edged rim from where it will be thrown off when centrifugal force exceeds surface-tension force. Walton and Prewett (1949) give the equation for drop diameter as a function of density, disk size, surface-tension and spin speed. The technique is excellent for drop diameters down to about 30 microns but the speed required for the drop diameters for albedo control is too high for practical wheels. Ultrasonic atomisers are used for dispensing chest medications and even for mist effects in stage performances. They consist of a piezo-electric actuator which induces shortwavelength capillary waves known as Faraday waves at the liquid surface. If these are steep enough, drops are thrown out from wave crests. Viscous losses in such short waves are rather high but we are trying to reduce them. The behaviour of drops colliding with one another depends on a parameter known as the Weber number. This is drop diameter times density times the square of velocity divided by the surface tension and is the ratio of kinetic energy to the surface-tension energy. For values below one, the impact is not enough to break the air film between drops and so they bounce apart. For values between one and ten, the drops will probably coalesce. For values much above ten, drops will break up into smaller ones with larger numbers of smaller drops being created by higher Weber numbers. It may be possible to use electrostatic forces to make drops collide with one another or to bounce between parallel walls kept at opposite voltages. This would be much easier if we could produce a material with hydrophobic but electrically conducting properties which will not be contaminated by prolonged impact of seawater. The preferred technique will pre-charge filtered sea water with compressed air. At 400 bar one litre of water can dissolve about 7 litres of air. This will be released to atmospheric pressure through two nozzles directed at each other along a vertical axis. Air will come out of solution to blow the jets apart just as they meet an identical stream coming the other way at a relative approach velocity of over 500 metres per second. The Weber number for a 1 mm drop is over 4 million. The colliding flow should produce a thin horizontal disk spreading radially outwards being slowed by air drag. The disk will impact the inside of a 45 degree cone which will direct it up through the bore of the rotor which would be open to air at top and bottom. A block diagram is shown in figure 6. While ultrasonic spray generation might need a fan to move the air and spray mixture up through the rotor there is more than enough upward momentum in the water drops from the drop collision spray generator to entrain a vigorous flow of air. Indeed fitting the rotors with impellers could recover some useful kinetic energy and reduce the need for rotor drive power.
Demonstration Before anyone will pay to build one spray vessel, let alone one hundred, there must be clear proof that the system will work as claimed. The proof should be in visual form, easily understood by non-technical decision-makers. It could be a satellite photograph of an unnatural pattern of cloud brightness. However, image contrast from the first experimental spray plant will have to be much higher than the increase needed to offset the effects of double present atmospheric CO2 concentration. Under good laboratory conditions with separated grey patches the human contrast detection threshold is about 10%. See figure 5 below. Figure 7 shows mathematical shapes intended to represent the wakes of two point sources at sea level with a standard deviation divergence angle of 3 degrees and three rings representing the wakes from the flight path of a circling aircraft. The cross sections are the familiar bell-shaped Gaussian curve. Figure 8 shows a satellite photograph of marine stratocumulus clouds near Madeira with mean albedo set to 0.5. The brightness of the cloud image has been multiplied by 0.1, 0. 2 and 0.3 of the flight path images and 0.3 of the two wake images in figure 7. This shows that the demonstration must achieve albedo increases of at least 1.2 to convince a nontechnical decision maker. This will require an increase of the concentration of cloud condensation nuclei of about 11. This increase is far above what any final equipment would have to do, but possible if the experiment is done with very clean air masses. One plan is to take multiple photographs of the wake of a spray system over a long period under random cloud patterns. The images would be digitized and rotated by computer to align wind directions and added together. The random clouds should average to a medium grey with contrast improving with the square root of photograph number.
Design choices and parameters The spray sites should have frequent partial cover of marine stratocumulus clouds, steady wind speeds but not hurricanes, low initial levels of cloud condensation nuclei and not too much other marine traffic. If we can achieve long service intervals, the distance to base does not seem important. This confirms the selection of the trade wind belts especially in the southern oceans. However there may also be a special need for operations round the Arctic ocean to cancel loss of ice cover and the large amount of water in Greenland ice. There has been a long-term trend in naval architecture for vessel size to increase. This is because of the relationship between water line length and wave drag. Larger vessels can carry more sails or Flettner rotors and develop more power. However too much spray at one point will saturate the local clouds and so be ineffective, especially because of the logarithmic term in the Twomey equation. The vessel number should be chosen to achieve an even distribution of a low concentration over a wide area. With a 1.5 degree divergence of the standard deviation of the spray concentration we get a merging of wakes after a down-wind. range of 20 times the source separation distance.
Figure 5. With a good grey-scale printer these patches should have a brightness increase of 10% of full white from one to the next. Only the brightest pair would be easily distinguished. Shapes with fuzzy edges would be even harder.
Figure 6. A block diagram of the spray generation system
Figure 7. Mathematical wake patterns representing stationary sources and circling aircraft.
Figure 8. The superposition of wake brightness from figure 7 on marine strato-cumulus clouds near Madeira. Contrast would have to be increased by about 1.2 for a convincing result.
The power into the propeller/turbine depends on the product of vessel speed and rotor thrust, which will itself depend on the square of vessel speed, so reasonably high speeds, just short of any sudden rise in wave drag, are attractive for energy and turbine cost. This makes for an interesting difference in design objectives compared with other sailing vessels. Vessel geometry could be mono-hull, catamaran, trimaran, proa or even a kite tied to a vestigial keel in the water. The mono-hull suffers more than other designs from water linelength effects on drag, provides more machinery space than we need, rolls a good deal in beam waves but has good recovery from extreme roll. The catamaran suffers less from wavemaking drag caused by short waterline length, rolls much less but recovers less well from extreme roll. However it seems less suited for the centrally placed machinery below Flettner rotors. This suggests that a trimaran configuration is attractive.
If the batteries are sized to give adequate maneuverability and endurance for the final approach to harbour at low speed, they will certainly be adequate for going about at the end of each leg of the spray path. The initial choice of hull material for the experimental vessels will be steel because of the ease of welding attachment points and drilling holes. Later small batches will be in glassreinforced plastic. However for large production numbers the low energy content and excellent endurance in sea water of ferro-cement is of interest.
Early prototypes will suffer frequent failures but, once the working environment and components are well understood, it does not cost very much to extend service intervals. One limiting factor may be fouling on the hull leading to high drag. However the start of fouling growth requires several days with water velocities below one metre per second and so vessels The beam of the vessel and the side-floats are chosen to give which spend little time in harbour should suffer less. We can adequate righting moment at maximum rotor drag force and hope for tours of duty of several years with condition monitoring wave slope. There is a sharp reduction in righting moment to give advance warning of problems. when one side-float becomes fully submerged or the other fully out of the water. The generator must operate over a wide range of speeds and power levels. A poly-phase permanent magnet machine using The need for remote control rules out traditional fabric sails moving neodymium boron magnets and static pancake coils whose only advantage is that they can be stowed by the crew. built into the rim of the rotor seems best. Switching-mode The first design ideas used solid wing sails rotating about a electronics using IGBT components, with series/parallel changevertical axis slightly forward of the centre of pressure with the over to cover a wide power range, will convert this to DC for angle of the wing to the apparent wind set by a torque rather than battery storage. a position command. Removing the torque lets the sail head directly into the wind with much less drag than a bare mast, so Motors for driving the several types of pump for water and air wing sails have very attractive survival features. However it can be standard induction machines running on a nominal 50 was not so easy to combine a wing sail with a spray generator. Hz supply from an inverter. However there is considerable Flettner himself had considered wing sails but abandoned competition for space at the bottom of the Flettner rotors for them for rotors. His circular section is a much better structural locating bearings, mechanical and electrical power beam than a wing sail. instrumentation, control and fluid passages. Although there are commercial designs of vertical axis induction motor with hollow The useful thrust from a Flettner rotor depends on the product shafts on the market, the vessel design looks less congested if of length and diameter but short fat ones will suffer losses we develop a special permanent-magnet rim-motor built into the analogous to the tip vortex loss of an aircraft wing. Flettner entry passage of each Flettner rotor and driving it directly with no used end plates and Thom added fences. It would be gears. It can run on direct current with local commutation interesting to look at tapered rotors. A length of 25 metres and depending on the angle of rotation. a diameter of 3 metres would be about the largest convenient for transport by road. We have to expect that rotors will interfere with one another if they are too close, just like the wings of a biplane. Flettner used two rotors on his first ship and three on the second. With two, the ship can turn about its own axis. With three some functionality would remain after the failure of one. Three or four seem right.
Standard global positioning systems are fine for navigation, with changes in the required mean position communicated by satellite. However most of the communications will be meteorological data on atmospheric pressure, air and sea temperature, solar input, the speeds and directions of current and winds, perhaps even plankton count, being sent from vessels back to meteorologists and biologists ashore who will The rotor thrust rises with the first power of spin speed times greatly appreciate hundreds or thousands more observation the apparent wind velocity rather than its square and self-limits sources. above a certain wind speed. For Flettner’s ship Buchau this limit was 12 metres per second, a value often chosen by wind The dispersion of plumes can be very complicated and is turbine designers for the maximum rating of their machines. discussed in (Anon 1999). For engineering purposes we can However the power needed to drive the rotor is likely to rise simplify some complicated relationships to just the angle by with the cube of the spin speed and so there is an incentive which the standard deviation of the concentration of nuclei towards rather large but slow rotors. Numerical work by Mittal diverges from the mean wind direction. In the stable conditions and Kumar (2003) at low Reynolds numbers shows that rotor due to steady temperatures of the sea the horizontal angle of surface speeds below two or above four times the wind can dispersion is between 1.5 and 3 degrees while the vertical one induce quite large oscillatory forces analogous to von Karman is between 0.3 and 1 degree. The former will affect cross-wind vortex shedding on a non-rotating cylinder. We are trying to dispersion and the latter the rate at which drops fall back into the sea. However there is a further complication because of predict effects at very much higher Reynolds numbers. movements below and within clouds and by effects of charge.
Required investment
Conclusions
A development programme has been planned to reduce technical uncertainties as rapidly as possible. These are mainly connected with the design of an efficient spray generator, drop life and dispersion, the present distribution of cloud condensation nuclei and meteorological modeling of effects. Very few will remain after the expenditure of the first £5 million. It will need perhaps £25 million to complete research and development and £30 million for tooling before the returns begin. Depending on spray rates and distribution effectiveness it is possible that 50 spray vessels costing a few million pounds each with a life of 20 years could cancel the thermal effects of a one-year increase in world CO2.
Present efforts to prevent global warming are like moving the deck chairs to the side away from the iceberg hole in the hope of sinking more slowly.
Several groups specializing in the provision of capital for activities linked to global warming have reported that at present there is no commercial value in global temperature control. The crucial political requirement is that albedo cooling should be included in something like the carbon-trading market, which was set up before Lathams ’s proposal was known to carbontrading planners.
Leading atmospheric physicists agree that albedo control by exploitation of the Twomey effect could allow double CO2 with no temperature change provided that all the engineering problems of spray generation and distribution can be solved. To optimize the design and make an accurate calculations of the number of spray vessels will require better knowledge of aerosol concentration, drop growth-rate, diffusion rate,lifetime and the reflecting thickness of cloud tops in marine air masses. Spray distribution could be done from a fleet of remotelycontrolled wind-driven spray vessels. The Flettner rotor with Thom fences is an attractive alternative to sails because of its high lift coefficients, high lift-drag ratios, easy control by computer and the convenience for housing spray plant and ejecting spray.
Potential returns
Some way must be found to include albedo cooling in the carbon trading market even if planetary survival has no The carbon-trading market works very much like the Papal perceived commercial value. However the albedo control indulgences which so annoyed Martin Luther. It is intended to option should never be used to reduce efforts to develop keep emissions at 5.2% below 1990 levels by requiring that renewable energy. people wanting to exceed that level should pay others to counteract the increase. Recent high and low spot values were €28.6 and €18.6 per tonne of CO2. The total market for 2005 was €10 billion but distortions have been introduced by false claims References for the reference base. The potential market for albedo control would be this amount minus removals by carbon sequestration Anon. European Process Safety Centre. Atmospheric times a factor for any increased perception of the dangers of Dispersion. Institution of Chemical Engineers. Rugby 1999. global warming. (Translation from French of a report from 1995.) However the carbon-trading markets were set up by people who were unaware of the potential of albedo control. Some renegotiation will be necessary if commercial investment is to fund it. The prediction of atmospheric CO2 for 2020 is for an increase of by a factor of 1.7. At a cost of €20 per tonne this would represent an annual market of €3500 billion. The calculation is complicated because opinions about the cost of losing low-lying countries are subjective and the dangers of global warming are non-linear. For example, the amount of land flooded by rising sea levels will rise very sharply as levels increase. The frequency and severity of hurricanes rises abruptly for sea temperatures above 26.5 C. There are events such as release of methane from permafrost, loss of ice cover or changes in the thermo-haline circulation which will, in turn, trigger other rises. None of the non-linearities so far identified suggest any reduction in ‘market value ’ of reduced temperatures. Albedo control will produce only a thermal effect and does nothing about the chemical effects of CO2. However some extra CO2 could be beneficial for crop production and this might to some degree compensate for the problems caused by increased acidity of the oceans. Albedo control would also be able to regulate temperature rises which are not connected with CO2, such as variation in solar inputs.
Acknowledgements We are grateful to William Rossow, Yuanchong Zhang, Chris Best, Ely Duenas, Keith Fielding, Shoaib Sufi, Brian Lawrence, Ted Lungu, Paul Carter and Natalie Mahowal for supplying information for the data base and to Philip Laven for advice on drop reflection.
Anon http://www.cdc.noaa.gov/PublicData/ Bennartz R. Global assssment of marine boundary layer cloud droplet number concentration from satellite. Journal of Geophysical Research, vol 112, 12, D02201, doi:10.1029/2006JD007547, 2007 From http://www.agu.org/pubs/crossref/2007/2006JD007547.shtml Bower K, Choularton T, Latham J, Sahraei J, Salter S. Computational assessment of a proposed technique for global warming mitigation via albedo-enhancement of marine stratocumulus clouds. Atmospheric Research vol. 82 pp 328-336 2006. Latham J. Control of global warming. Nature 347 pp 339-340 1990. Latham, J. Amelioration of global warming by controlled enhancement of the albedo and longevity of low-level maritime clouds. Atmos. Sci.Letters. 2002doi:10.1006/Asle.2002.0048. Laven P. http://www.philiplaven.com/mieplot.htm Mittal S. Kumar B. Flow past a rotating cylinder. J. Fluid Mech. vol. 476, pp. 303-334. 2003. Norwood, J. Performance prediction for 21st century multihull sailing yachts. Amateur Yacht Research Association. London 1991. Seufert W and Seufert S. Critics in a spin over Flettner ’s ship. New Scientist 10 March 1983 pp 656-659. Sortino G. A data resource for cloud cover simulation. MSc thesis, School of Informatics, University of Edinburgh, 2006. Thom A. Effects of Disks on the air forces on a rotating cylinder. Aeronautical Research Committee Reports and Memoranda 1623 1934. (from Cranfield University). Twomey, S., 1977: Influence of pollution on the short-wave albedo of clouds. J. Atmos. Science., 34, 1149-1152. Walton W H and Prewett WC. The Production of Sprays and Mists of Uniform Drop Size by Means of Spinning Disc Type Sprayers ”.Proc. Phys. Soc. (London) B , 62, p. 341 (1949).
