2018ApJ...864...75K

Many habitable zone (HZ) exoplanets are expected to form with water mass fractions higher than that of the Earth. For rocky exoplanets with 10-1000x Earth's H₂O but without H₂, we model the multi-Gyr evolution of ocean temperature and chemistry, taking into account C partitioning, high-pressure ice phases, and atmosphere-lithosphere exchange. Within our model, for Sun-like stars, we find that: (1) the duration of habitable surface water is strongly affected by ocean chemistry; (2) possible ocean pH spans a wide range; (3) surprisingly, many waterworlds retain habitable surface water for >1 Gyr, and (contrary to previous claims) this longevity does not necessarily involve geochemical cycling. The key to this cycle-independent planetary habitability is that C exchange between the convecting mantle and the water ocean is curtailed by seafloor pressure on waterworlds, so the planet is stuck with the ocean mass and ocean cations that it acquires during the first 1% of its history. In our model, the sum of positive charges leached from the planetary crust by early water-rock interactions is-coincidentally-often within an order of magnitude of the early-acquired atmosphere+ocean inorganic C inventory overlaps. As a result, p_CO2 is frequently in the "sweet spot" (0.2-20 bar) for which the range of semimajor axis that permits surface liquid water is about as wide as it can be. Because the width of the HZ in the semimajor axis defines (for Sun-like stars) the maximum possible time span of surface habitability, this effect allows for Gyr of habitability as the star brightens. We illustrate our findings by using the output of an ensemble of N-body simulations as input to our waterworld evolution code. Thus (for the first time in an end-to-end calculation) we show that chance variation of initial conditions, with no need for geochemical cycling, can yield multi-Gyr surface habitability on waterworlds.