Our research leverages state-of-the-art deep learning techniques to automate surface mapping and continuous monitoring on planetary bodies. We are also developing tools to analyze the model uncertainty and decision-making in AI models with evaluation in our surface mapping projects and beyond.

We focus on one of the solar system's most dynamic Earth-analog environment on terrestrial planets - Mars' northern polar region, a repository of the planet's climatic history within its extensive ice-layered dome. We detect small blocks [1] and their sources yielding a reliable method for monitoring mass wasting activity with valuable present-day erosion rate results [2].

In parallel, we investigate and map polygonal patterns on both Earth and Mars to assess the global distribution of polygons and their potential as indicator for climate conditions and changes. On Earth, polygons are indicators of the volume of ground ice and provide insights into permafrost vulnerability to climate change. On Mars, similar young landforms could be linked to geologically recent freeze-thaw cycles. This would be conflicting with the current environment and would have implications for the recent hydrologic past of the planet. The distribution of polygonal ground on Mars can provide valuable information on the role of liquid water in the recent past by shedding light on the formation mechanism.

We use AI models for automated surface mapping because they achieve highly complex decision-making. However, they are usually treated as Black-Box systems. To tackle this problem, we are developing software tools for analyzing model uncertainty and decision-making within an application-independent framework. Typical questions are why did the model produce exactly this response and how certain is it about the correctness of its results?

References: [1] Martynchuk O. et al., 2024. EGU 2024. [2] Su S. et al., AGU 2023.

How to cite: Fanara, L., Su, S., Martynchuk, O., Hauber, E., Schlegel, A., Ludwig, J., Melching, D., Hänsch, R., and Gwinner, K.: Unraveling the climate evolution on Mars and Earth with AI-driven surface mapping and explainable AI, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10083, https://doi.org/10.5194/egusphere-egu24-10083, 2024.

Mass wasting activity, in the form of ice block falls, has been observed as the main erosion process at steep scarps of the North Polar Layered Deposits (NPLD) [1,2]. Our study focuses on leveraging a state-of-the-art deep learning technique to map the sources of such events throughout the entire NPLD region. By quantifying water ice loss, we derive the current erosion and retreat rate for each active NPLD scarp. We notice that these scarps have varying degrees of erosion, from less than 0.01 up to 0.88 m³ per Mars Year per meter along the scarp. The current most active scarp shows a retreat rate of ~6 mm per Mars Year. We want to compare our results to the detected ice block falls at the underlying Basal Unit (BU) region [3], to understand the difference between the two units’ geological processes, and help to constitute important constraints to the present-day mass flux of the north polar region.

References

[1] Herkenhof et al., 2007. Science, 317(5845), pp.1711-1715.

[2] Dundas et al., 2021. J. Geophys. Res. Planets, 126(8), p.e2021JE006876.

[3] Martynchuk, et al., 2023. AGU23, 11-15 Dec.

How to cite: Su, S., Fanara, L., Xiao, H., Hauber, E., and Oberst, J.: Erosion rate of the north polar steep scarps on Mars, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19493, https://doi.org/10.5194/egusphere-egu24-19493, 2024.

The north polar region of Mars is one of the most active places of the planet with avalanches and ice block falls being observed every year on High Resolution Imaging Science Experiment (HiRISE) data. Both phenomena originate at the steep icy scarps, which exist on the interface between two adjacent geological units, the older and darker Planum Boreum Cavi unit, also called Basal Unit (BU) and the younger and brighter Planum Boreum 1 unit, which is a part of the so called North Polar Layered Deposits (NPLD). These exposed layers of ice and dust contain important information about the climate cycles of the planet. We are primarily interested in monitoring the current scarp erosion rate (quantified through analyzing ice debris) at the same time differentiating between the activity originating in the NPLD [1] from that originating in the BU

The large scale of the region of interest, combined with a growing amount of available satellite data makes automation key for this project. To achieve the latter we propose a computational pipeline consisting of three consecutive steps, namely: scarp segmentation, single image super-resolution and ice-block detection. For the final analysis Mean Average Precision (mAP.95) was used as a benchmark metric. The performance value of 93.6% was obtained on a test dataset, leading us to conclude that the network is able to perform even on small ice fragments (which comprise the majority of the debris). On a system running 4 RTX3090 GPUs the finished pipeline processes a single HiRISE product in just under 20 minutes, returning the scarp outline and precise ice boulder coordinates. Using this pipeline, we next plan to robustly monitor the mass wasting activity in the whole north polar region and throughout the entire Mars Reconnaissance Orbiter (MRO) mission.

[1] Su, S. et al., 2024. EGU 2024.

How to cite: Martynchuk, O., Fanara, L., Gwinner, K., and Oberst, J.: Computer vision model for monitoring block falls in the Martian north polar region, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-22233, https://doi.org/10.5194/egusphere-egu24-22233, 2024.

In today’s extremely dry Mars, water vapor “adsorption” on regolith grains is thought to play crucial roles in subsurface water retention and water vapor exchange with the atmosphere (Fanale & Cannon, 1971; Zent et al., 1993, 1995, 2001; Böttger et al., 2005; Savijärvi et al., 2016, 2020). Global models that explicitly account for water diffusion in the shallow subsurface and calculate subsurface water distribution have assumed globally uniform regolith properties to simplify assumptions (Böttger et al., 2005; Schorghofer & Aharonson, 2005; Steele et al., 2017). However, Pommerol et al. (2009) examined the adsorption efficiency of six samples similar to the Martian regolith and found that the samples with smaller grain sizes store more adsorbed water due to their larger specific surface areas. Therefore, we have newly implemented a regolith scheme in a Mars Global Climate Model (MGCM), considering regolith properties like grain size, porosity, and the specific surface area. The grain size distribution was obtained from the empirical equation as a function of thermal conductivity (Presley & Christensen, 1997). The distributions of porosity and the specific surface area are also determined, referring to the laboratory experiments of Sizemore & Mellon (2008). Our results clarify that regolith grains with large specific surface areas in the northern low and mid-latitudes and the southern high latitudes, which have high adsorption coefficients, affect water storage. Subsurface water in the northern low and mid-latitudes exists up to 0.5–1wt% as adsorbed water. Regolith with high adsorption properties makes the depth of subsurface ice shallower in the southern high latitudes. Pore ice accumulates in regions poleward of 50°N and 50°S and the west of Elysium Mons and Olympus Mons, which is consistent with previous simulations. Also, with a homogeneous specific surface area, seasonal increases in pore ice were calculated at a depth of about 60 cm in mid-latitudes with low thermal inertia and high atmospheric water vapor content, but with the specific surface area map, the seasonal increases were not demonstrated. This study suggests that adsorption properties influence subsurface water dynamics, emphasizing the importance of considering inhomogeneous regolith properties in models of subsurface water distributions and the atmospheric water cycle including the regolith.

How to cite: Kobayashi, M., Kamada, A., Kuroda, T., Kurokawa, H., Aoki, S., Nakagawa, H., and Terada, N.: Effects of regolith properties on the Martian subsurface water distribution using a global climate model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7059, https://doi.org/10.5194/egusphere-egu24-7059, 2024.

Improving the data on the gravitational field of Mars can yield enhanced knowledge about Martian planetary dynamics and subsurface water reservoirs. In this study, we augment the VENQS software tool to perform simulations for a future dedicated satellite gravimetry mission at Mars following the archetype of GRACE-FO and as a result to study the challenges of such a mission.

The VENQS software tool consists of two parts: the VENQS App and the VENQS library. The VENQS App provides users with an easy access to a variety of simulation models, that can be combined to an individual VENQS library setup. These simulation models include amongst others orbit propagation of single satellites with embedded test masses, simulations of satellite constellations, and detailed disturbance analysis for satellites due to the space environment. Interaction with versioning systems allows the VENQS App to effectively track the software of the simulation models. In addition, a dedicated release management system enables the provision of different versions of the VENQS library.

Initially designed for satellites orbiting Earth, we are working on an augmentation of the VENQS library for interplanetary spacecraft or to be more precise for satellites orbiting arbitrary celestial bodies. In this context we want to propose the adaptation of VENQS for precise orbit propagation at Mars, which can assist the assessment of different mission influences on gravity field recovery (via dedicated software tools such as GRAVFIRE). We present the general simulation procedure including the modelling of perturbating forces along with gravitational acceleration for the orbit integration. Furthermore, we explain the differences to simulations of terrestrial spacecraft and outline occurring challenges with Martian atmosphere, time and reference frames, solid Mars tides as well as more complex satellite geometries inducing micro-vibrations and the non-availability of GNSS, that may deteriorate gravity field solutions.

How to cite: Bredlau, M., Bremer, S., Schilling, M., and Wassermann, N. K.: Simulation of a satellite gravimetry mission at Mars, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9136, https://doi.org/10.5194/egusphere-egu24-9136, 2024.

Previous studies have constrained the lithosphere at the north and south poles of Mars to be thick and cold, with elastic thicknesses of 330 to 450km [1], and >150km [2], respectively. The elastic thickness characterizes the stiffness of the lithosphere in response to loading and is directly linked to the thermal state of the lithosphere and the surface heat flow. Thus, elastic thickness estimates at the north and south poles provide crucial constraints on the present-day surface heat flow on Mars. Additional information on the present-day planetary thermal state comes from evidence of ongoing melting in the mantle, as indicated by the presence of both young lava flows in Tharsis and Elysium provinces and an active mantle plume beneath Elysium Planitia [3,4,5]. 

In this study we explore the thermal evolution of Mars using global 3D geodynamic models. These models improve upon our previous work [6] by including updated interior structure information from the InSight (Interior Exploration using Seismic Investigations, Geodesy and Heat Transport) mission [7,8] and by considering constraints on the present-day thermal state of the planet as noted above. Thermal evolution models using the most recent crustal thickness estimates [8,9], require that the crust contains more than half of the total amount of heat producing elements (HPEs) to explain localized recent volcanic activity on Mars [8]. 

We find that the crustal thickness variations control the surface heat flow and the elastic thickness pattern, as well as the location of melting zones in the present-day Martian mantle. The strongest constraint for the thermal history and present-day state of the interior is given by the elastic thickness at the north pole. While at the south pole, all models show values >150km, compatible with the latest estimate [2], only a few models present an elastic thickness >300km at the north pole, with values still lower than the recent estimate of [1].  A larger elastic thickness at the north pole could indicate: 1) a northern crust less enriched in HPEs, 2) a colder lithosphere due to a weaker blanketing effect caused by a thinner or higher-conductivity crust on the northern hemisphere, 3) ongoing viscoelastic relaxation, suggesting that the observed surface deflection beneath the north polar cap is not the final one [1], or a combination thereof. 

In contrast to the cold lithosphere inferred for the Martian polar regions, recent volcanic activity suggests a warmer interior beneath Tharsis and Elysium provinces [3,4]. This reveals an important spatial variability in the thermal state and thickness of the Martian lithosphere. Our work shows that only a narrow range of models can match elastic thickness estimates at the polar caps and explain Mars’ recent volcanic activity, thereby providing important insights into the structure and thermal evolution of the interior.

References:

[1] Broquet et al., 2020. [2] Broquet et al., 2021. [3] Voigt et al., 2023. [4] Hauber et al., 2011. [5] Broquet & Andrews-Hanna, 2023. [6] Plesa et al., 2018. [7] Stähler et al., 2021. [8] Knapmeyer-Endrun et al., 2021. [9] Wieczorek et al., 2023.

How to cite: Plesa, A.-C., Broquet, A., Voigt, J. R. C., Wieczorek, M. A., Hauber, E., and Breuer, D.: Thermal state of the Martian interior at present day as constrained by elastic lithosphere thickness estimates and recent volcanic activity, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12473, https://doi.org/10.5194/egusphere-egu24-12473, 2024.

Understanding the intricate thermal dynamics on Mars is crucial for accurate scientific measurements, particularly for seismological studies. The InSight Mission to study the interior structure and composition of Mars has recorded the Mars seismograms and in-situ data for the initial assessment of Mars' geothermal heat flow. Given that these measurements are obtained in close proximity to the lander at the surface, a primary concern is the presence of thermo-elastic noise, originating from fluctuations in solar radiation, within the collected data. Experiments such as those conducted by SEIS on Mars have specifically identified this phenomenon, detecting noise during the eclipse of Phobos (Stähler et al, 2020). While managing periodic temperature variations of instruments is feasible, challenges arise with other factors, such as those associated with moving shadows on the ground and solar radiation fluctuations. This implies that the presence of the lander will introduce thermal perturbations, causing alterations in both local surface and subsurface temperature measurements. These challenges necessitate numerical quantification due to difficulties in filtering them from the data. Hence, this study investigates first how the shadowing effect from the lander's structure and solar radiation variations impacts subsurface soil temperatures and consideration of this effect on the tilt recorded on the seismometers. We develop a 3D numerical model within Comsol Multiphysics 6.1 finite element package. The key element in adapting this model for use on Mars is accurately replicating the illumination conditions on the surface. Based on sub solar latitude and longitude derived using the JPL Horizons Ephemeris output, an illumination model is set at the instrument site for a desired duration. Unlike the Moon, where no atmospheric contribution affects temperature variations, Mars possesses a thin atmosphere that contributes to convective heat transfer. First, an analytical model is employed to find the transient solution of temperature at any given depth and time instances. The solution to the energy balance analysis determines the boundary conditions at the ground surface, which are then applied in the heat conduction equations governing subsurface temperature distribution.  The numerical temperature distribution output at an unperturbed location, far away from the lander is then compared with the analytical solution.  Once the 3D model is calibrated, the resulting temperature profiles can be utilized to assess the tilt of the seismometer feet and the sensitivity to additional solar radiation fluctuations. The findings suggest that the presence of a lander can exert substantial effects on the surrounding temperature environment under Martian conditions. This can introduce noise into the data collected by the seismometer, emphasizing the importance of accounting for and mitigating such influences in both the design and data analysis.

How to cite: Kizhaekke Pakkathillam, S., Lognonne, P., De Raucourt, S., and Kawamura, T.: Lander Induced Thermo-Elastic Noise at InSight Location on Mars, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12497, https://doi.org/10.5194/egusphere-egu24-12497, 2024.

The excellent performances of quantum accelerometers, due to their very good behaviour in the low frequency measurement bandwidth and to their intrinsic stability, which does not call for periodic calibration of the sensors, foster their application to extra-terrestrial investigations. In particular, the study of Mars and of its planetary composition, evolution, density and surface properties is going to be of great importance in the next decades for many reasons, both for the enhancement of the scientific knowledge and for applications in future missions.

So far, the gravity models of Mars have been derived from tracking data of different missions. Preliminary simulations performed at POLIMI considering a one-arm gradiometer pointing in the radial direction, flying on a polar orbit and acquiring data for a time span of two months show that a significant improvement in the knowledge of the gravity field of Mars could be achieved by launching a dedicated mission collecting gravity gradiometry observations by means of a quantum sensor. Even taking into account a degradation of the solution due to more realistic conditions, allowing for a possible mission lifetime of a few years (which is feasible under Mars conditions) would mean that the already available CAI technology could lead to very high benefits in terms of the scientific knowledge of the Martian gravity field.

How to cite: Reguzzoni, M., Rossi, L., and Migliaccio, F.: A quantum gradiometry mission concept for the improvement of Mars gravity field models, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12635, https://doi.org/10.5194/egusphere-egu24-12635, 2024.

Due to its axial tilt of ~25°, Mars has seasons. During its fall and winter, when temperature drops, there exist two depositional mechanisms of atmospheric CO₂, that is, precipitation as snowfall and direct surface condensation in the form of frost (Hayne et al., 2012). Up to one third of the atmospheric CO₂ exchanges with the polar surface through the seasonal deposition/sublimation process. Therefore, accurate measurements of the evolution of the seasonal polar caps can place crucial constraints on the Martian climate and volatile cycles. 

Recently, by reprocessing and co-registering the MOLA profiles, Xiao et al. (2022a, 2022b) derived both spatial and temporal thickness variations of the seasonal polar caps at grid elements of 0.5° in latitude and 10° in longitude. However, the MOLA-derived results can suffer from biases related to various processes, for example, pulse saturation due to high albedo of the seasonal deposits, non-Gaussian return pulses due to rough terrain and dynamic seasonal features, incomplete correction for the global temporal bias, and penetration of the laser pulses into the translucent slab ice. Furthermore, MOLA altimetric observations are limited to Mars Year 24 and 25 which prevents the detection of possible interannual variations in the CO₂ seasonal transport. 

In this contribution, we will show how the shadow variations of fallen ice blocks at the bottom of steep scarps of the North Polar Layered Deposits (NPLDs) allow us to infer the thickness evolution of the seasonal deposits (Xiao et al., 2024). For this, we utilize the High Resolution Imaging Science Experiment (HiRISE/MRO) images with a spatial resolution of up to 0.25 m/pixel (McEwen et al., 2007). We successfully conduct an experiment at a steep scarp centered at (85.0°N, 151.5°E). We assume that no, or negligible, snowfall remains on top of the selected ice blocks, the frost ice layer is homogeneous around the ice blocks and their surroundings, and no significant moating is present. These assumptions enable us to separately determine the thickness of the snowfall and frost. We find that maximum thickness of the seasonal deposits at the study scarp in MY31 is 1.63±0.22 m to which snowfall contributes 0.97±0.13 m. Interestingly, our thickness values in the northern spring are up to 0.8 m lower than the existing MOLA results (Smith et al., 2001; Aharonson et al., 2004; Xiao et al., 2022a, 2022b). We attribute these differences mainly to the remaining biases in the MOLA heights. Furthermore, we demonstrate how the long time span of the HiRISE images (2008—2021; Mars Year 29—36) allows us to measure the interannual variations of the deposited CO₂. Specifically, we observe that snowfall in the very early spring of Mars Year 36 is 0.36±0.13 m thicker than that in Mars Year 31. 

Hayne et al. (2012). JGR: Planets, 117(E8).

Xiao et al. (2022a). JGR: Planets, 127(7), e2022JE007196.

Xiao et al. (2022b). JGR: Planets, 127(10), e2021JE007158.

Xiao et al. (2024). JGR: Planets (In Revision).

McEwen et al. (2007). JGR: Planets, 112(E5).

Smith et al. (2001). Science, 294(5549), 2141-2146.

Aharonson et al. (2004). JGR: Planets, 109(E5).

How to cite: Xiao, H., Xiao, Y., Su, S., Schmidt, F., Lara, L. M., and Gutierrez, P. J.: Thickness of the seasonal deposits by examining the shadow variations of the fallen ice blocks at Martian North Pole, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14925, https://doi.org/10.5194/egusphere-egu24-14925, 2024.