Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.


Future exploration of Ceres as an ocean world

Long believed to be a primitive body, Ceres is now an ocean world with deep brines at a regional and potentially global scale. Further studies at Ceres’s conditions and — above all — a follow-up mission are needed to study its evolution and potential habitability.

After two years in orbit around Ceres, the Dawn spacecraft, in the last phase of its extended mission (June–October 2018), reached closer to Ceres’s surface, down to 35 km, and performed imaging, infrared spectroscopy, elemental spectroscopy and gravity science at resolutions never before achieved at an icy body1. The focus was on Occator crater, a ~20-million-year-old crater revealed earlier in the mission to be a site of recent evaporite exposure (faculae), primarily as sodium carbonate spread over >100 km2 (ref. 2). This discovery raised a debate in the community as to whether that activity stems from a deep brine layer, from melt produced as a result of heat generated by the impact that created the crater, or both. Liquids from an impact melt chamber evolve on different timescales than those from deeper (>40 km) reservoirs: the former would start warm, up to 100 °C, as a result of impact heat, but freeze over a few million years because of their close proximity to the surface. On the other hand, a perennial brine layer could be maintained by the presence of salts at subzero temperature if it was deep enough for internal heat to slowly leak through a thick crust.

The Dawn extended mission results published this month in Nature Astronomy, Nature Geoscience and Nature Communications shed new light on this question. First, the high-resolution images confirmed exposure of brines less than two million years ago3. Furthermore, the visible and infrared mapping spectrometer found a new compound, hydrohalite (NaCl.2H2O) (ref. 4). This material is very common in marine ice and it is the first time it has been found beyond Earth. Hydrohalite tops Cerealia Facula, a prominent site at the centre of Occator crater, whose composition is dominated by sodium carbonate2. It is a smoking gun for ongoing activity: that material is unstable on Ceres’s surface, and hence must have been emplaced very recently. Since impact heat dissipated long ago, recently exposed material had to be sourced from the deeper brine. Furthermore, hydrohalite precipitates at ~–30 °C while sodium carbonate forms at ~–5 °C. Therefore, the contrast in composition between the base and the top of Cerealia Facula points to an evolution of the source of that pile of evaporite. Lastly, imaging at <10 m pixel–1 shows thick deposits in Cerealia Facula and thin deposits farther away from the centre, in the Vinalia Faculae5. Altogether, these observations indicate that the first material forming the faculae was supplied by the warm brine solution created by impact melt and directly accessible below the surface, but the more recent material is coming from deeper, colder brines percolating via a fracture network throughout Occator’s subsurface6.

Analysis of Ceres’s topography suggests the presence of a low-strength layer ~40 km deep on a global scale7 that was interpreted as evidence, at the very least, of pore fluid. This led Ceres to be classified as a ‘candidate’ ocean world in the roadmap for the exploration of these objects commissioned by NASA8. More recently, the emplacement of Ahuna Mons, a 4 km tall by 17 km wide feature located hundreds of kilometres away from Occator, whose flanks are also covered by sodium carbonate9, was explained as the result of brine effusion carrying rock particles10. Ongoing activity in Occator brings additional and independent evidence for a deep brine layer and upgrades Ceres to the realm of ocean worlds. This discovery goes against the standard view of Ceres’s evolution. Its heat budget is determined by long-lived radioisotopes and it does not benefit from tidal heating like the other ocean worlds. In addition, its surface is heavily cratered, while highly heated ocean worlds, like Europa, tend to have been subjected to regular and recent resurfacing. However, it shares commonality with another dwarf planet.

At 40 au from the Sun, Pluto is in an even more dire situation than Ceres in terms of heat budget. Its environment is very cold (35 K) and even though it has accreted antifreeze, that is, compounds like ammonia that can decrease water freezing temperature significantly, this is still not sufficient to explain the presence of abundant liquid suspected in that body. Independent studies for Ceres and Pluto concluded that long-lived preservation of liquid in these bodies requires the presence of gas hydrates (clathrates)11,12. Clathrates are molecules of water surrounding small gas molecules. Although they resemble and have about the same density as water ice, their thermal and mechanical properties are vastly different: their thermal conductivities are up to one order of magnitude lower than ice whereas they are up to three orders of magnitude stronger. The combination of these two effects results in significant slowing of heat loss through icy shells. Clathrates have commonly been assumed to be present in Titan’s crust13, and have been suggested for Europa14, but they have never been considered in a systematic manner in icy-body models even though they are expected to naturally occur in the hydrospheres of large icy bodies. Their probable role in preserving liquid in Ceres and Pluto carries a vast array of implications for maintaining liquids in other heat-starved bodies like other dwarf planets and Jupiter’s moon Callisto12. Still, the extra heat does not explain how activity occurs in Ceres, since its crust is strong and immobile. The new geological observations indicate that the crater opened fractures reaching >35 km deep6 and relieved the pressure in the gas-loaded brine reservoir15. No fractures are evident around Ahuna Mons; however, a subsurface fracture network offers the most probable pathway for mantle material to reach the surface10. Future in situ measurements are needed to confirm this hypothesis.

Ceres’s ocean is very different from Europa’s or Enceladus’s. Its high density indicates a high concentration of rock particles10,16. Furthermore, high-resolution gravity data obtained during Dawn’s extended mission revealed an abundance of dense material — that is, salts and rock particles — in the lower crust of Ceres17, suggesting that material is reaching saturation in the ocean and is being incorporated in the crust upon freezing. As such, these high-resolution gravity data provides insights into how ocean and crust interact as ocean worlds near the end of their lives. These findings emphasize the role of impurities (especially gas species and brines, a direct consequence of aqueous alteration) in driving the long-term evolution of volatile-rich bodies. The new observations also revealed hydrological and periglacial processes that resemble those on Earth and have also been observed on Mars18, but that have been influenced by the composition of Ceres’s crust19. These processes are also expected to act at icy satellites19. Hence, the Dawn observations can help prepare for the interpretation of future observations planned at icy moons, in particular Europa (Europa Clipper mission) and Ganymede (Jupiter Icy moons explorer (JUICE)), in the coming decade.

The next ten years of dwarf planet exploration requires focus to be brought onto habitability through time in these evolved oceans, which are likely to be rich in organic matter. Exposed evaporites in Occator provide direct sourcing of the deep brine below the crater and represent an obvious target for a future mission. The composition of the brines reflects >4 Gyr of evolution but a key question is whether the ocean environment (that is, its pH, temperature and redox conditions) remains static overall, in which case chemical reactions and organic processing would reach an equilibrium, or whether it was modified over time, and if so, how. In other words, was Ceres’s ocean a closed system or did it benefit from regular, or at least sporadic, input of fresh oxidizing material to create local chemical gradients? More generally, this question is of major interest to the ocean world community: without an influx of oxidants, the habitability potential of ocean worlds and their value as astrobiological targets decrease significantly. Three main processes have been identified to ‘open’ oceans, at least temporarily. Large impacts act in disrupting the icy crust, potentially reaching out to the ocean (in the case of basin-forming impacts), leading both to the loss of gas and to the supply of fresh material. An alternative source of volatile supply to the residual ocean comes from the thermal evolution of rocky mantles in the form of devolatilization that releases mostly oxidizing material to the ocean20. Lastly, carbon dioxide and other small molecules trapped in clathrates may escape to the impact melt chamber as impact breaks the clathrate ‘cages’. The extent of these processes on ocean worlds is poorly understood, and certainly their effects have not been fully considered. However, their implications on the evolution of organic matter are potentially significant. In situ exploration of Ceres’s evaporites, or even a sample return, can help to address these questions. However, data on the behaviour of these materials and their mixtures in conditions relevant to Ceres and other ocean worlds are lacking. For example, the relationships between organic compounds and salts or electrical conductivity of relevant brine mixtures are mostly uncharted territory to be explored by future experimental work. Convection in a brine–rock mixture and transfer of material in a fractured crust under low-gravity conditions are other critical processes for mid-sized ocean worlds for which theoretical frameworks are mostly missing. Hence, until the next mission to Ceres appears on the horizon, scientists need to continue exploring Ceres through the Dawn data and fundamental research, an endeavour that will benefit ocean world science as a whole (Fig. 1).

Fig. 1: Comparative ocean world planetology.

NASA/JPL-Caltech/UCLA/MPS/DLR/IDA (top left); NASA/JPL-Caltech (top right); NASA/JPL (bottom)

Ceres shares commonalities with other ocean worlds: a low-density, high-porosity mantle, as also suggested for Enceladus, and a thick shell, and hence a residual ocean, as also suggested for Callisto. Synergistic studies of these various bodies can help unravel evolutionary processes, such as material transfer in a porous rocky mantle and a thick but fractured crust.


  1. 1.

    Castillo-Rogez, J. C. & Rayman, M. D. Nat. Astron. (2020).

  2. 2.

    De Sanctis, M. C. et al. Nature 536, 54–57 (2016).

    ADS  Article  Google Scholar 

  3. 3.

    Nathues, A. et al. Nat. Astron. (2020).

  4. 4.

    De Sanctis, M. C. et al. Nat. Astron. (2020).

  5. 5.

    Scully, J. E. C. et al. Nat. Commun. (2020).

  6. 6.

    Raymond, C. A. et al. Nat. Astron. (2020).

  7. 7.

    Fu, R. R. et al. Earth Planet. Sci. Lett. 476, 153–164 (2017).

    ADS  Article  Google Scholar 

  8. 8.

    Hendrix, A. R. et al. Astrobiology 19, 1–27 (2019).

    ADS  Article  Google Scholar 

  9. 9.

    Zambon, F. et al. Geophys. Res. Lett. 44, 97–104 (2017).

    ADS  Article  Google Scholar 

  10. 10.

    Ruesch, O. et al. Nat. Geosci. 12, 505–509 (2019).

    ADS  Article  Google Scholar 

  11. 11.

    Castillo, J. C. et al. Geophys. Res. Lett. 46, 1963–1972 (2019).

    ADS  Article  Google Scholar 

  12. 12.

    Kamata, S. et al. Nat. Geosci. 12, 407–410 (2019).

    ADS  Article  Google Scholar 

  13. 13.

    Lunine, J. I. & Stevenson, D. J. Icarus 70, 61–77 (1987).

    ADS  Article  Google Scholar 

  14. 14.

    Prieto-Ballesteros, O. & Kargel, J. S. Icarus 173, 212–221 (2005).

    ADS  Article  Google Scholar 

  15. 15.

    Quick, L. C. et al. Icarus 320, 119–135 (2019).

    ADS  Article  Google Scholar 

  16. 16.

    Ermakov, A. I. et al. J. Geophys. Res. 122, 2267–2293 (2017).

    Article  Google Scholar 

  17. 17.

    Park, R. S. et al. Nat. Astron. (2020).

  18. 18.

    Schmidt, B. E. et al. Nat. Geosci. (2020).

  19. 19.

    Schenk, P. et al. Nat. Commun. (2020).

  20. 20.

    Melwani Daswani, M. & Vance, S. E. Goldschmidt Abstr. 2020, 1777 (2020).

Download references


The research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). US Government sponsorship is acknowledged. The author acknowledges discussions with many colleagues on the topic of Ceres research and implications for other ocean worlds.

Author information



Corresponding author

Correspondence to Julie Castillo-Rogez.

Ethics declarations

Competing interests

The author declares no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Castillo-Rogez, J. Future exploration of Ceres as an ocean world. Nat Astron 4, 732–734 (2020).

Download citation


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing