An intriguing recent paper suggests bubbling the ocean to reflect sunlight back into space and reduce global temperatures (Russell Seitz, 2011, Bright water: Hydrosols, water conservation, and climate change, Climatic Change, 105:365-381). I like the idea: it is a stimulating and thought provoking argument, but I'm not yet convinced it can play a large enough role to make a global difference without much more experimental evidence rather than theoretical reasoning that might be, or might not be, complete. But it certainly is worth exploring in case he is right! He does not say so, but because of the density and viscosity differences it is energetically far cheaper to pump air to make bubbles than it is to pump water through sprays to make aerosols. That's why air pumping is used to circulate sewage ponds and fish ponds rather than pumping water around, but despite being packed full of all those really GREAT surfactants, even those bubbles don't last all that long! 

It is many years since I studied light scattering by particles in the ocean and by aerosols in the atmosphere (and I never was very expert in either) but I remember the theory as being far more complicated than is presented here, with very complex angular and wavelength dependences. If we take his argument that air bubbles in water are the mathematical equivalent of water drops in air, then like a London fog lit by lamps carried by armies of people searching for an honest politician, much of the light is internally scattered, giving an internal white glow with short path length to the next scattering event, which is why you can't see the man searching a few meters away even if you are aware of a diffuse glow in his direction, and why you'll die long before you find an honest politician, or even a common garden variety lying one! Most light does not escape the perimeter of the cloud, but still enough light leaves the outer boundary so that someone on a hill overlooking the thermal inversion sees a glowing fog beneath even while most light is internally trapped. The same will be true of reflections from dense bubbles. My point is that one cannot simply sum the light scattering of the individual bubbles, because most of the light will be reflected by the surrounding bubbles internally within the bubble layer, so only a fraction will be reflected back to the atmosphere and contribute to albedo increase. It seems to me that the path length of the light between reflection and scattering, strongly dependent on the size and density of the bubbles, greatly affects how much light actually leaves, and I'd assume that is not a linear function of the total number of bubbles though I'm not smart enough to calculate it.  

Having spent a bit of time inside breaking waves, I can confirm that they are very bright, you are surrounded by white and can't see any distance at all. But based on a lifetime on and under the water, I'm dubious about the optimistically large spatial and time scales he suggests can operate. I've spent a lot of time watching waves breaking on coral reef crests all over the world from both above and below, and I can tell you the vast bulk of bubbles in white caps dissipate very quickly, in seconds, not the months that are suggested possible due to surfactants (middle of page 372). Similarly, few ship wakes are kilometers long as stated, perhaps supertankers in flat calm condition, but certainly not most boats of any size I've ever been on. If the bubble spatiotemporal lifetimes are much shorter than hoped for, the energy needed to produce the effects needed will be correspondingly larger. 

One way to look at this to examine white caps as a function of wind speed. White caps do locally increase albdeo, during storms, but the sky is usually overcast and dark. On a global scale, averaged over time, this effect is fairly small. Stormy waters have lower albedo because flat water reflects much better than wavy water with a large range of slopes. Global wind speeds are increasing due to global warming induced increases in pressure gradients, and this is increasing white cap frequencies, providing a negative feedback on temperature, but I bet the effect is very small compared to the positive feedbacks in the system, and MUCH larger increases in global winds would be needed for the desired effects. I've been through Super-Typhoons while stranded on Pacific atolls or in small boats in the middle of the Indian Ocean, but those at most occupy a couple of days a year. Basically you would need energies equivalent to continuous super-Typhoons over most of the ocean, and anyway the clouds would greatly reduce surface light levels to be reflected, and most of that would be absorbed by clouds, not bounced out to space. I realize that use of pumps to make bubbles as he proposes is not at the same as big hurricanes (except insofar as bubble breaking spray aerosols are a major source of cloud condensation nuclei, especially if they have lots of surfactants), but the energies needed to make enough bubbles to cool the world enough to make up for what we are doing with fossil fuels might not be that different?

So the real question is whether practical increases in albedo that are large enough and long lasting enough can be induced at reasonable cost. The thing to do would be to take one of these new experimental ships using bubble production to reduce surface frictional drag of the hull and fly above in a helicopter to measure the increased albedo bubble plume from the air to see how large and long lasting they are, and what they cost to produce. A white plume stretching kilometers and lasting months might do the trick to reduce temperatures, but I'd be very surprised if that actually happened even if they are dribbling surfactants at the same time. But of course the bubble layer will reduce light levels beneath, with profoundly negative effects on the entire ocean biology. 

Having personally watched most of the corals in the world die from high temperatures, I'd love to believe that a ring of bright water at low latitudes could save them. While the idea is reasonable in principle, I'm skeptical that it can be done affordably at a magnitude that would make a difference, but I'd like to see it tested in case he is right!