Magma Ocean Evolution of the TRAPPIST-1 Planets

Abstract

Recent observations of the potentially habitable planets TRAPPIST-1 e, f, and g suggest that they possess large water mass fractions of possibly several tens of weight percent of water, even though the host star's activity should drive rapid atmospheric escape. These processes can photolyze water, generating free oxygen and possibly desiccating the planet. After the planets formed, their mantles were likely completely molten with volatiles dissolving and exsolving from the melt. To understand these planets and prepare for future observations, the magma ocean phase of these worlds must be understood. To simulate these planets, we have combined existing models of stellar evolution, atmospheric escape, tidal heating, radiogenic heating, magmaocean cooling, planetary radiation, and water-oxygen-iron geochemistry. We present MagmOc, a versatile magma-ocean evolution model, validated against the rocky super-Earth GJ 1132b and early Earth. We simulate the coupled magma-ocean atmospheric evolution of TRAPPIST-1 e, f, and g for a range of tidal and radiogenic heating rates, as well as initial water contents between 1 and 100 Earth oceans. We also reanalyze the structures of these planets and find they have water mass fractions of 0-0.23, 0.01-0.21, and 0.11-0.24 for planets e, f, and g, respectively. Our model does not make a strong prediction about the water and oxygen content of the atmosphere of TRAPPIST-1 e at the time of mantle solidification. In contrast, the model predicts that TRAPPIST-1 f and g would have a thick steam atmosphere with a small amount of oxygen at that stage. For all planets that we investigated, we find that only 3-5% of the initial water will be locked in the mantle after the magma ocean solidified.

Understanding planetary context to enable life detection on exoplanets and test the Copernican principle

“…The ways in which broader planetary context controls the long-term build-up of oxygen due to slight imbalances in sources and sinks is less understood. Oxygen build-up on planets around low mass stars due to extensive hydrogen escape during the pre-main sequence has been suggested 20 , and coupled models have explored whether such oxygen build-up could overwhelm magma ocean sinks 21,22,23 . The possibility of oxygen accumulation due to water loss in atmospheres with low non-condensable inventories has also been proposed 24,25 .…”

An upper limit on late accretion and water delivery in the Trappist-1 exoplanet system

“…to loss driven by ultraviolet radiation from the central star (especially for the inner planets). 35,36 Models of the early evolution of the Trappist-1 planets that also include the magma ocean phase 27,37 find that each planet may have lost a few to 20 Earth oceans of water. 27 Calculations that make assumptions designed to maximize water loss still find an upper limit of 25 oceans lost in the interval before planets e through h reached the habitable zone.…”

“…We used interior structure 23 and atmospheric 24 models constrained by the latest planetary mass and radius estimates. 3 Given that inferred water contents are subject to interior degeneracies, 3,[25][26][27][28] we estimated water mass fractions assuming five different rocky interior models (see Methods). Across all of our models we found that planets b, c and d are likely to be volatile-depleted whereas the outer planets (e through h) are more likely to have significant water contents (consistent with previous models 3,28 ).…”

Blue Marble, Stagnant Lid: Could Dynamic Topography Avert a Waterworld?

“…Planetary land/ ocean fractions emerge in a compromise between water's total budget and distribution between surface and interior reservoirs and the size of the basins carved out by topography (e.g., Simpson 2017). The resulting ocean mass from the former is largely stochastic: coded within it are the histories of volatile delivery during accretion (Raymond et al 2006;Morbidelli et al 2012), interior degassing from the magma ocean and succeeding mantle (Elkins-Tanton 2008; Schaefer & Fegley 2017;Katyal et al 2020;Ortenzi et al 2020;Barth et al 2021;Bower et al 2021;Guimond et al 2021;Lichtenberg et al 2021), atmospheric erosion by impacts (Zahnle & Catling 2017;Schlichting & Mukhopadhyay 2018;Howe et al 2020), and photodissociative atmospheric escape (Tian & Ida 2015;Zahnle et al 2019;Gronoff et al 2020), along with the surface temperature and pressure. In contrast, large-scale aspects of planetary topography may lend themselves to deterministic relationships with observable planetary bulk properties.…”