Evidence of groundwater discharge and lake formation in Arabia Terra

Introduction

We examined an unnamed impact crater roughly centered at 27.8°N, 11.5°E, in the northeastern portion of Arabia Terra. Its uttermost relevant characteristic is the presence of multiple fan-shape deposits on its floor [1; 2; 3]  exhibiting an impressive preservation of water-related features (Figure 1). 

We carried out geomorphological analyses and mapping, which can provide valuable insights into the amount of water needed to generate the observed landforms, and is directly relevant for understanding the past climate and the associated geological processes. Additionally, we modeled the groundwater depth at the time of crater formation. The results of this numerical model can help us to evaluate if the observed landforms are consistent with the thermal conditions necessary for groundwater to exist at this location. These models are also used to compare the depth to the base of the cryosphere at this specific location to the planetary average at the time of crater formation. These measurements allowed us to reconstruct the different stages of aqueous activity within the crater.

Figure 1. CTX mosaic of the study area. Contour lines are displayed every 250 m using a color gradient from green (-4050 m) to orange (-2300 m). High-interest geomorphological features are mapped in blue (valleys) and brown (fan-shaped deposits). 

 

Methods

For our geomorphological observations, we relied on imagery from the Context Camera (CTX; ∼6 m/pixel) global mosaic [4]. For a regional topographic context we used the Mars Orbiter Laser Altimeter colorized elevation model (∼460 m/pixel) [5]. Observations, mapping, and measurements were made using the QGIS software, while for age determination we used the crater size-frequency distributions (CSFD) of impact craters [6].

The CTX Digital Elevation Models (DEMs) used in this work were first downloaded as raw images from the PDS Geosciences Node. The image processing was carried out through the USGS ISIS3 [7]. The resulting point clouds were then aligned to the MOLA global topography to improve the accuracy and finally converted to DEMs.

Volumetric estimations were carried out using the 3D point cloud and mesh processing CloudCompare (CC) software (cloudcompare.org) [8]. In the CC environment, through the 2.5D Volume Calculation Tool, it is possible to estimate volumes between a point cloud and an arbitrary plane, thus efficiently simulating the given case scenario of water volumes inside an impact crater.

We used global thermal evolution models [9] to estimate the groundwater depth at the time of the crater formation by comparing the local subsurface temperature with the melting temperature of water ice. Our models employ the most recent constraints on the interior structure of Mars from InSight seismic data [10]. The crustal thickness in our models is constant in time but varies locally according to crustal thickness maps derived from gravity and topography data and anchored by seismic measurements at InSight landing site [11]. We used several crustal thickness models [11] to test the effects of crustal thickness variations on the depth of groundwater. 

 

Results

The examined crater displays numerous water-related features suggesting that during its evolution it must have experienced at least a period of intense water activity despite its relatively young age. Nevertheless, the crater is missing a connection with large-scale fluvial networks capable of surface runoff. 

As the emergence of springs is connected to groundwater availability, it implies that the head of the valleys indicates the groundwater level that generates them. Accordingly, the elevation of valley head locations in the studied crater can provide us with information about groundwater supply during the period of formation. Specifically, valley heads are located close to -2900 m elevation, suggesting  that this may represent the top of the groundwater table level at the time when landforms developed after the crater formation at 3.5 Ga.

Overall, from the collected evidence, we can reconstruct a timeline of key events that shaped the examined area (Figure 2). The projectile collision with the Martian surface leads to the crater excavation, reaching the depth of the groundwater table. Groundwater flows into the basin and forms the observed water-related features observed today.  The exceptionally good match between geomorphological observations and the expected groundwater depth obtained through modeling suggests that the crater infill likely took place shortly after the impact when a warmer subsurface temperature and, hence, a shallower groundwater level were present. In this context, groundwater flowed from the subsurface toward the basin creating sapping valleys and filling the crater with water and fan-shaped sedimentary deposits. As the Martian subsurface cools with time,  i) the water input stopped; ii) the water already present in the basin got lost in the ground and atmosphere; iii) the groundwater table reached a deeper level given the thinner, less insulating crust at  the crater location, eventually arriving at a scenario comparable to that currently observable.

Figure 2.  Main events of the examined area. t-1 = unperturbed pre-impact crust; t0 = impact (3.5 Ga); t1 = water sapping, valleys and fan-shaped deposits formation, the basin fills with water; t2 = the basin dries out, geomorphological features connected to water circulation are left in a subaerial environment and the groundwater level adapts to the new surficial roughness

 

References

[1] Irwin, R.P., et al., (2005). J. Geophys. Res. 110. [2] Salese, F., et al., (2019). JGR: Planets, 124(2), 374–395. [3] De Toffoli, B., et al., (2021). Geophys. Res. Let., 1–10. [4] Malin, M. C., et al. (2007). J. Geophys. Res 112, E05S04. [5] Zuber, M. T., et al. (1992). J. Geophys. Res.  97, 7781–7797. [6] Hartmann, W. K., & Neukum, G. (2001). Sp. Sci. Rev, 96, 165–194. [7] Becker, K. J., et al., (2007). Lun. Plan. Sci. Conf.1338, 1779. [8] Penasa, L., et al. (2021). Wiley, 3D Geo. Mod. [9] Plesa, A. C., et al., (2022). Geophys. Exp. Sol. Sys. 63, 179-230. [10] Knapmeyer-Endrun, B., et al. (2021). Science, 373(6553), 438-443. [11] Wieczorek, M. A., et al. (2022). JGR: Planets, 127, e2022JE007298.

Acknowledgement

The authors thank the European Union – NextGenerationEU and the 2023 STARS Grants@Unipd programme “HECATE” support.