An outline of the paper: Climate instability on tidally locked exoplanets by Kite, Gaidos & Manga (2011).
The habitable zone around a star is the region where water may exist in liquid form on a rocky planet. The presence of a planet in this zone, however, does not ensure the presence of a stable atmosphere and a liquid ocean on that planet. Many climate instabilities have been studied, such as atmospheric collapse, photochemical collapse, runaway greenhouse, ice-albedo feedback and thermohaline circulation bistability in a large ocean. Kite et al. studied two more instabilities: enhanced substellar weathering instability (ESWI) and substellar dissolution feedback (SDF).
The rotation of a hypothetical rocky planet in the habitable zone of an M-dwarf is expected to be synchronised with the orbital period. Both instabilities require four conditions. Firstly, the atmosphere must be optically thin, and secondly, the substellar point of the planet must be much hotter than the planet-wide average. This latter condition is possible if the atmospheric or oceanic circulation is ineffective at redistributing the incident solar radiation. The third condition is that the substellar temperature increases with a decrease in pressure. Last, the two instabilities also require that the main atmospheric gas be a greenhouse gas.
Due to the poor heat redistribution assumption, there is a strong temperature contrast between the day and night side. The rate of weathering is strongly sensitive to temperature and consequently, the hot substellar point would host most of the weathering. This weathering would use up the greenhouse gas, but would usually be in balance with the volcanic outgassing.
A small decrease in pressure due to a dip in outgassing would usually reduce the substellar temperature. Counterintuitively, if the atmospheric recirculation of heat due to the now thinner atmosphere drops faster than this cooling effect, the substellar point can heat up. This would increase the rate of substellar weathering, pulling out more CO2 from the atmosphere and increasing the weathering rate even further. This enhanced substellar weathering can lead to atmospheric collapse (ESWI). The reverse can lead to a runaway greenhouse effect: a small increase in pressure causes the substellar point to cool by increasing the recirculation, reducing greenhouse gas consumption at the substellar point.
The authors note that Mars could have undergone ESWI in the past, given present day deposits of carbonates on Mars and the thin martian atmosphere. Strong greenhouse gasses can suppress ESWI, but the high UV flux from M-dwarfs could cause accumulation of CH4 and lead to anti-greenhouse haze effects. This would make ESWI possible at a larger range of pressures and stellar luminosities. In fact, the authors predict that we should observe a bimodal distribution of planets around M-dwarfs, with emission temperatures either close to isothermal or close to radiative equilibrium. The region in between is unstable to ESWI. On the other hand, substellar dissolution feedback should be a much rarer instability because it requires highly soluble gasses, small but deep and well-mixed oceans restricted to the substellar point, and precisely synchronous rotation – an unlikely mix of conditions.