Ceres’ surface features, interior structure, and thermal evolution suggest a subsurface regional sea rather than a global ocean

Introduction:

Extensive evidence from the Dawn mission shows that dwarf planet Ceres had liquid brines in the subsurface in the past and probably the present, a finding with potential astrobiological significance [1].  Especially compelling evidence is found at Occator Crater, where the young, bright faculae have been shown to be the products of endogenic, salty water interacting with the surface [e.g., 2–6].  It is unclear whether these brines today exist regionally underneath Occator or globally in a subsurface layer that one could call an “ocean”. Because traditional observations used to detect subsurface oceans (e.g., libration, magnetic induction) are not plausibly applicable to a body like Ceres without a giant planet or large moon, more indirect methods must be used to test the extent of subsurface liquid water.

Here, I argue that the combination of surface features, inferences on Ceres’ interior structure, and results from thermal evolution models are most parsimonious with subsurface liquid brines concentrated in a region under Occator Crater but not present in the form of a present-day global ocean.  The key observation is that Occator is at low latitude and the region of greatest crustal thickness, both factors that favor brines at the bottom of the icy crust.

Methods:

 I use the finite element method (FEM) software COMSOL to quantify heat transfer in 3D and calculate the present-day temperature distribution within Ceres. The goal of the models is to determine if and where temperatures that permit liquid brine stability are reached. A similar approach with analytical equations has been previously used in 1D to show the feasibility of brines [7], and a 3D FEM approach is a sensible next step given Ceres’ surface temperature and interior structure.  Because Ceres’ surface is so close to its melting point (within 100 K in some locations) and its crustal thickness can exceed 10% the planetary radius, effects like surface temperature variations, crustal thickness variations [8], lateral heat conduction, planetary curvature, and ice’s temperature-dependent conductivity are all non-negligible.  I quantify these effects in the models.

I model Ceres’ crust in COMSOL in 3D. The outer boundary is set at the annual-average surface temperature, which varies with latitude, and the inner boundary has a heat flux applied from the deep interior.  The crust is a mix of ice, rock, and possibly clathrates and porosity [9], where the thermal conductivity is volumetrically weighted by each component.  I run hundreds of simulations, varying the crust’s composition and the heat flow.  At the end of each simulation, I note where, if anywhere, the base of the crust exceeds 273 K (melting point of pure H₂O) or 220 K (the eutectic temperature of plausible chloride-ice mixtures [10]). Figure 1 shows the model conceptually.

Results and Discussion:

Sample results from a simulation along different longitudes are shown in Figure 2, and results from many simulations are summarized in Figure 3. Many models consistent with Dawn data [e.g., 8] and thermal evolution models [e.g., 10] lead to temperatures at the base of the crust that exceed 220 K beneath the region around Occator Crater, but not globally. For example, the model shown in Figure 2 yields a basal crustal temperature of 231 K underneath Occator’s region but only 119 K underneath the south pole. The models show that reaching >220 K globally is very challenging and would require extensive clathrates, which have been suggested [9] but may be implausible on the basis of evolution models [11], or unrealistically high heat flows from the deeper interior.

Brines existing regionally on Ceres but not everywhere provides a natural explanation for why Occator’s faculae are so extensive, but other large, young craters have smaller faculae [12] or no bright faculae at all. Occator’s faculae are sourced from both impact melt and pre-existing subsurface water [3], but perhaps other faculae, like the smaller and less bright faculae in young Haulani Crater [13], result from the impact process alone without requiring endogenic brines.  This process would not explain the presence of faculae at Ahuna Mons [14], which could instead be locally sourced from the deeper interior rather than the base of the crust [15].

Conclusions:

The combination of Ceres’ surface features (Occator’s bright faculae), interior structure (thick crust in Occator’s region), and thermal evolution (models imply only a few mW/m² today) suggest that liquid brines exist today regionally in the subsurface under Occator, but not globally. This interpretation is consistent with a global ocean in the past, as the regional brine sea would represent a late stage of an ancient, freezing ocean [16].  Therefore, facula materials in Occator are still interpreted to be the products of an ocean world in this framework, and represent an excellent and accessible target for future spacecraft exploration [17].

References: [1] Castillo-Rogez et al.(2020), Astrobiology. [2] De Sanctis et al.(2016), Nature 536. [3] Scully et al.(2020), Nature Comms. [4] Bowling et al.(2019), Icarus 320. [5] De Sanctis et al.(2020), Nature Astron. [6] Nathues et al.(2020), Nature Astron. [7] Raymond et al.(2020), Nature Astron. 4. [8] Ermakov et al.(2017), JGR Planets 122. [9] Fu et al.(2017), EPSL 476. [10] Castillo-Rogez et al.(2019), GRL 46. [11] Castillo-Rogez et al.(2018), Meteorit. Planet. Sci. 53. [12] O’Brien et al.(2024), PSJ 5. [13] Krohn et al.(2018), Icarus 316. [14] Zambon et al.(2017), GRL 44. [15] Ruesch et al.(2019), Nature Geosci. 12. [16] Pamerleau et al.(2024), Nature Astron. [17] Castillo-Rogez et al.(2022), PSJ 3.

Figure 1. Conceptual diagram of our simulations, with a latitudinally variable surface temperature and heat flow Q applied to the base of the crust.

Figure 2. Example model temperature outputs along 220ºE (A–A’) and 60ºE (B–B’) for a crust with 50% ice, 30% clathrate, 10% silicate rock, and 10% porosity.

Figure 3. Results from a suite of models showing the predicted temperature at the bottom of the crust underneath Occator (solid lines) and the south pole (dashed lines) for different crustal compositions and heat flows applied from the deeper interior.