Thermal Modelling and Morphological Analysis of PSR-Hosting Craters at Mercury’s North Pole

Introduction:

The polar regions of Mercury host Permanent Shadowed Regions (PSRs), which remain in constant shadow due to the planet’s low obliquity [1]. These conditions allow surface temperatures to drop to extremely low values [2], potentially enabling the accumulation and long-term stability of water ice. Radar-bright deposits were firstly observed in PSRs from Earth-based radar [3,4] and interpreted as water ice. This hypothesis was later supported by comparisons with icy bodies [5], hydrogen abundance measurements taken by the MESSENGER mission [6], and thermal models [7]. Morphological studies of PSR-hosting craters [8] have revealed features potentially linked to ice-related processes. On an airless body like Mercury, local topography significantly influences solar illumination and, in turn, the thermal behavior of surface volatiles. In this study, we investigate the thermal environment of selected PSR-bearing craters using shape-based thermophysical modelling [9], with the goals of: (1) characterizing their thermal conditions, (2) predicting the evolution of their surface features, and (3) assessing whether temperature regimes could influence the formation of specific landforms.

Methods:

We adopted a multi-disciplinary approach, combining geomorphological mapping with thermophysical modeling. Our study began with a geomorphological analysis of two north polar craters, Fuller and Ensor, focusing on four distinct landform types:

• Fractures: Classified as “Landforms of Uncertain Origin” [7], these features may be related to subsurface ice acting as permafrost (Fig. 1A). On Earth, permafrost forms when the ground remains below freezing for at least two consecutive years [9]. Prior studies [6,11] suggested that PSR floors may host a lag deposit composed of dark, carbon-rich material—interpreted as residue from sublimated ice. This deposit could behave like an active layer in terrestrial permafrost, potentially deforming and cracking due to thermal gradients induced by nearby sunlit terrains [12]. Alternatively, the fractures could be associated with cooling of impact melt or volcanic infilling, both common processes on Mercury [13]. Fuller and Jimenez craters host these kinds of features.
• Landslides: These post-impact features partially cover the crater floors (e.g. Fuller’s floor, Fig. 1B) and have been interpreted as rockslides [14]. Their formation may be affected to thermal weathering, which varies with temperature. Since PSRs receive minimal insolation, thermal breakdown of rock and subsequent mass wasting may be less efficient in these regions.
• Rough Unit: This unit, located near Fuller’s central peak (Fig. 1C), is darker and rougher than both the smooth crater floor and the landslide material. Its distinct morphology and thermal behaviour make it a key target for further investigation.
• Bright ejecta: Ensor crater contains bright material in the shadowed wall (Fig. 1D). These are ejecta of small craters, which may have brought underlying ice to the surface on impact [8, 17].

Fig. 1 – The morphologies analysed in the studied: A) fractures within the Fuller’s floor, B) one of the Fuller’s landslide, C) the hummocky unit inside the Fuller crater; D) the bright ejecta on the shadowed wall of Ensor crater.

To investigate the thermal environment, we performed a detailed thermophysical analysis of seven craters using a shape-based 3D thermal model [9]. This model calculates the surface and subsurface temperature of each facet in a 3D mesh over time, incorporating direct insolation, multiple scattering of visible and infrared light, thermal emission, and terrain shadowing. Ray-tracing is employed to simulate the Sun as a disk, taking into account its large angular size in Mercury’s sky. This approach allowed to accurate model both umbral (full shadow) and penumbral regions, which are critical in high-relief terrains such as crater interiors.

Preliminar observations:

Plots in Fig. 2 show the max temperatures measured in 176 days, reached by different portion of the two craters (the colored curves are the different thermal profiles). From the graph, it is possible to notice that the southern portion of Ensor reaches the lowest temperature (slightly less than 100 K), particularly in the lower part of the wall, where ice deposits have been detected on the surface. In general, the floor with the dark material does not exceed 250 K.

In contrast, Fuller shows:

- the part of the floor with fractures experiences a maximum temperature of 200 K;

-  the blue unit (Rough unit) reached the lowest temperature, around 150 K, notably in the same region where radar-bright material has been observed;

- the surface affected by landslide experiences significant thermal fluctuations, ranging from 300 K to 550 K within a distance of less than 10 km.

Fig.2 – Arrows indicate the three analysed features located in the Ensor and Fuller craters (maps is taken from [7] and graphs is extrapolated by [8] simulations.

Conclusion:

We analysed craters with PSRs in Mercury’s north polar region, focusing on landforms (fractures, rockslides) potentially linked to thermal regimes. Our preliminary results suggest that temperature conditions within PSRs may influence the formation and evolution of surface features. Ongoing and future work will focus on the analysis of five additional craters.

Acknowledgements:

This work has been developed under the ASI-INAF agreement n. 2024-40-HH.0

References:

[1] Margot J.-L. et al. (2012), JGR:Planets, V. 117. [2] Susorney H. C. M. et al. (2021), The Planet. Sci. J., V. 2. [3] Harmon and Slade (1992), Science, V. 258(5082), pp. 640–643 [4] Harmon et al. (2011), Icarus, V. 211, pp. 37-50. [5] Butler et al. (1993), JGR, V. 98, pp. 15003 – 15023. [6] Wilson et al. (2019), JGR:Planets, V. 124, pp. 721 – 733. [7] Paige et al. (2013), Science, V. 339, pp. 300 – 303. [8] Bertoli et al. (2024), Jourtnal of Maps [9] Cambianica P. et al (2024), PSS [10] Dobinski, W. (2011), Earth-Science Rev., V. 108, pp. 158 – 169. [11] Syal M. B. et al. (2015), Nature Geoscience, V. 8. [12] Filacchione G. et al. (2022), EPSC Abstracts V. 16, EPSC2022-191, 2022 [13] Xiao Z. et al. (2014b), JGR:Planets, V. 119, pp. 1496 – 1515. [14] Crudes D. M. and Varnes D. J. (1996), Landslides Eng. Pract, V. 24, pp. 20–47. [15] Brunetti M. T. et al. (2015), Icarus, V. 260, pp. 289 – 300. [16] Molaro J. L. et al (2015), JGR:Planets, V. 120. [17] Deutsch A. N. et al. (2019), EPSL, V. 250, pp. 26 – 33