Perhaps the most fraught ethical question in the whole sustainability universe is our obligation to the deep future. Specifically, the question of whether, if we could obtain substantial certainty that our present actions would have significant consequences in the deep future, whether we have any obligation to prevent such an outcome.

As David Archer often points out when he raises this question, people do weigh such questions in deciiding on the ethics of nuclear power, but they don't do so in considering the ethics of carbon releases.

Methane clathrate is a form of water ice that contains a significant fraction of methane molecules caged in the ice crytal mesh. It is unstable at standard pressure and evanesces gradually, but it has the eye catching property that it's flammable. It forms under pressure at low temperatures. At some locations on the sea floor, naturally occurring carbon rain from detritus of plankton and similar life forms decays, producing methane, which creates the conditions for formation of clathrate.

Some of the clathrate may be economically recoverable, and as such may be considered as part of the potential reserves of natural gas. Because methane has 20 times the global wamring potential (more or less) of carbon dioxide, a large release of methane could be a serious matter. A methane release from clathrates is still, as far as I know, a leading candidate for the cause of the Paleocene-Eocene Thermal Maximum of 65,000,000 years ago, the previous really drastic global warming event, in which humans obviously played no role.

How did the world suddenly contrive to release all its clathrates? It might have a consequence of some other warming trend, which propagated warm temperatures to the ocean depths, perhaps destabilizing the clathrates. Wikipedia quotes estimates of 1500 Gt carbon released, about triple anthropogenic emissions to date. The extent to which this forcing was multiplied by the extra global warming potential due to the carbon being in methane form depends on the time period over which the release took place.

Archer and Buffet 2004 looked at the global inventory of sea-floor methane and came up with a total on the order of 2000 Gt, about equivalent to economically viable fossil fuel deposits today.

Presuming that most of the clathrate is not economically recoverable, Archer & Buffet argue that the anthropogenic warm signal will propagate down to the depth (below the sea floor) of the clathrates, causing a significant release in the distant future creating a second global warming crisis, that one of longer duration than the one we are now so eager to precipitate.

I know that David was trying to get this work published for quite a while, with a baffling lack of success. I suppose that predictions on a time scale much longer thna a human lifespan are arguably not testable.

Therefore the most salient quote I found remains this abstract from an AGU presentation:
Mechanistic models for the distribution of methane clathrate in marine sediments predict a steady-state inventory of 5000 Gton C. These models also predict a strong sensitivity to changes in ocean temperature. Increasing the ocean temperature by 1.5° C is expected to decrease the steady-state inventory by roughly a factor of 2. The time scale for adjustment to a new steady state is not well known. However, we can parameterize the time-dependent behavior using first-order rate constants. Methane accumulation is expected to occur on time scales of several million years, whereas methane release could be comparatively fast. We calculate the rate of methane release by dividing the clathrate reservoir into fast and slow parts. The fast part responds through slumping of continental margins, whereas the slow part responds by diffusion and oxidation of methane below the seafloor. The evolution of the clathrate reservoir through geologic time is sensitive to the choice of rate constants. When the time scale for the fast part is too short, runaway melting is caused by the positive feedback from radiative forcing (assuming CH4 has oxidized to CO2). Thus the existence of the present-day inventory imposes a constraint on the rate of release. The time scale for accumulation must by greater 5 Myr to avoid unrealistic fluctuations in δ13C during glacial cycles, but shorter than 10 Myr to allow the clathrate reservoir to build up during the geologically recent cooling. The constrained model predicts a methane release of 200 GTon C or less on deglaciations. Future methane releases of 2000--4000 GTon C are expected in response to a 2000 GTon C anthropogenic carbon release. Anthropogenic climate change differs from deglaciations in that it warms the ocean to temperature not seen in millions of years.
Emphasis added.

So this raises an ethical question which economics will "discount" to zero, and which I have trouble accepting as having zero value. Presuming the Archer/Buffett work holds water, what are our ethical responsibilities to the distant future?

--
Burning ice image via PinkTentacle probably from Asahi.com