Introduction

Planetary radar observations provide a powerful tool for the post-discovery characterization of the physical and dynamical properties of asteroids, comets, the Moon, and terrestrial planets. Radar observations of near-Earth objects (NEOs) [e.g., 1] have increased our understanding of the diversity of NEO sizes, shapes, binarity, and composition. These characteristics are crucial also to planetary defense as they play a role in the selection of the optimal mitigation technique. Planetary radar systems can be used for range-Doppler imaging by mapping the reflected power as a function of the Doppler frequency and the range (based on the signal’s round-trip time), which allows imaging resolutions finer than 10 meters at best, and thus direct observations of morphologic features and possible moons. The data can be obtained at two orthogonal polarization states.

Due to the penetration depth of several wavelengths and the wide parameter space in scattering inversion problems, understanding the physical characteristics of NEOs based on their radar scattering profiles requires extensive numerical modeling. Traditionally, circular polarization ratio has been used as a first-order gauge to the surface roughness, but more recent advances in numerical modeling demonstrate that analyzing the reflectivity information in parallel with the polarization information is crucial (e.g., [2-3]). This allows distinguishing different scattering processes. For example, the disk function of (101955) Bennu shows little to no specular component, which indicates that wavelength-scale particles dominate the surface; a fact not available from the polarization ratio information alone. For contrast, the Moon has a strong quasi-specular spike, which is consistent with the fact that fine-grained regolith dominates the lunar surface.

Analytically derived scattering models typically assume that the surface is composed primarily of fine-grained regolith or a solid surface that forms a gently undulating interface with few or no wavelength-scale scatterers. This assumption has been reasonable for the surfaces of the terrestrial planets and moons but is not sufficient for asteroids that have often a “rubble-pile structure” and, as such, the asteroid surfaces have often a significantly greater coverage of centimeter-to-decimeter scale regolith than planets or moons. Empirical laws lack understanding of the meaning of the empirical fit parameters. In this presentation, I discuss the recent advances in scattering modeling methods and future requirements for improved interpretation of radar observations.

Aims

Here, we present recent advances in the modeling efforts of radar scattering for the characterization of planetary bodies. The goal is to improve planetary surface characterization by better interpretation of radar observations. As research has shown, examples of physical properties that can be derived include the near-surface density, regolith size-frequency distributions of centimeter-to-decimeter scale particles, and subsurface permittivity contrast that provides clues to the internal structure and composition.

Methods

The radar scattering processes in planetary bodies includes two components: Scattering by the undulating surface and scattering by the wavelength-scale particles. As the main part of this work, we conducted numerical computations of scattering properties of rough polyhedral particles 1) in touch with a surface to simulate surface particles [6], and 2) embedded in a host medium. In the first case, we investigate the roles of size parameter (x=2πr/λ, where r is the effective particle radius and λ is the wavelength) and the refractive properties. We selected two different refractive indices for comparative analysis: 2.17 + 0.004i and 2.79 + 0.0155i (particles) on a substrate with 1.55 + 0.004i, and two polyhedral morphologies with statistically distinct levels of roundness. In the second case, the refractive contrast relative to the host medium is compared for 1.4 and 1.8. Also, the effect of the particle packing density is investigated for the radiative transfer approximation with and without coherent backscattering included. The size-frequency distribution of regolith typically follows a power-law distribution with a power index of 2.5–3.5; a comparable size distribution range is used also in our numerical simulations. The range of sizes extends from sub-wavelength scale to several wavelengths.

For scattering by a fine-grained regolith substrate, we built synthetic rough surfaces and simulated radar scattering as a function of incidence angle using a geometric-optics approximation [4]. We used self-affine fractal surfaces, which are described using a horizontal-scale-dependent height standard deviation and Hurst exponent, because they have been shown to be more realistic for rocky surfaces than stationary surfaces. Research has shown that the Gaussian scattering law provides a good approximation for scattering by self-affine fractal surfaces [4,5].

Summary of the results

This work discusses and illustrates the different scattering processes taking place in planetary surfaces and what role they play in the observable parameters. The main part of the work discusses the role of particles on surfaces and below the surface. We find that for surface particles with a refractive index above 2.17, the refractive index plays an insignificant role in comparison to the particle abundance and shape, and that the surface-particle interaction is weak [6]. For particles embedded in the substrate, the contrast between the particles and host medium plays a noticeable role in the observed polarization and reflectivity. Coherent backscattering produces a significant enhancement, as expected. We discuss how to identify coherent backscattering – a signature of low-absorption substances such as water ice – when observations at a range of phase angles are not available.  

References

[1] Virkki, A. K. et al. (2022), Planetary Science Journal, 3, 222.

[2] Virkki, A. K. & Bhiravarasu, S. S. (2019), Journal of Geophysical Research: Planets, 124, 11.

[3] Hickson, D. C., et al. (2021), Planetary Science Journal 2, 30.

[4] Virkki, A. K. (2024), Remote Sensing 16, 890.

[5] Shepard M. K. et al. (1995), Journal of Geophysical Research 100, E6, 709.

[6] Virkki, A. K. & Yurkin, M. A. (2025), In revision. Pre-print available at https://arxiv.org/abs/arXiv:2501.10019.

How to cite: Virkki, A. and Leppälä, A.: Recent advances in planetary surface characterization using modeling of radar scattering, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-52, https://doi.org/10.5194/epsc-dps2025-52, 2025.

Planetary bodies can be affected by a number of geologic processes, including impacts, volcanism, volatile deposition, mass wasting, and weathering. Local stratigraphic sequences record the effects of these processes like a time capsule, revealing how geologic processes have shaped the site through time. Many geologic processes leave their fingerprints within the stratigraphy on many meters to decameters scales. While most remote sensing techniques are sensitive only to surface materials at the shallowest (µm to m) depth scales, radio frequencies (RF) are well poised for geologic characterization of planetary bodies at these deeper scales, providing insight on regolith thickness, subsurface deposits, and geologic chronology.

Ground penetrating radar (GPR) sounding is a non-invasive geophysical technique often employed to sense the subsurface that has been used on Earth, the Moon (the Chang’E 3 and 4 Yutu rovers’ Lunar Penetrating Radars, and the orbital Kaguya Lunar Radar Sounder), and Mars (the Mars Perseverance Rover’s RIMFAX, Mars Reconnaissance Orbiter’s SHARAD, and Mars Express’ MARSIS instruments), with Europa Clipper’s REASON instrument currently en route to Europa. Collecting GPR data, however, requires lateral translation of the antenna(s) to build up a 2D profile of the subsurface, adding risk and operational complexity. Additionally, a major challenge in interpreting traditional 2D GPR data is subsurface “clutter,” signals returned at the same time as subsurface targets of interest which disguise signals from the target.

Here, we present a surface-deployed RF instrument concept under development, which we call ARIA (the Askaryan Regolith Imaging Array) (Fig.1). ARIA features a dual-polarization, bistatic antenna array that measures full Stokes parameters and applies interferometric techniques that have never before been used for planetary radar, providing an unprecedented opportunity for 3D subsurface imaging at a landing site — all while stationary. ARIA utilizes a unique combination of traditional and cutting-edge RF techniques, including: active GPR, bistatic radar methods that capitalize on cosmic rays as a natural RF source, and passive radiometry for temperature profiling.

ARIA’s active radar system features a circularly-polarized transmit antenna that measures regolith properties, buried geologic units, and rocks embedded in the subsurface through the spectral, temporal, and polarimetric characteristics of radio signals received at 250–750 MHz using dual-polarized sinuous antennas, employed as a beamforming array to provide both lateral and depth resolution.

ARIA’s passive bistatic radar observes radio emission from cosmic ray cascades within the regolith, which illuminate an extended radius surrounding the antenna array. These relativistic particle cascades create highly impulsive, 100% linearly-polarized radio emission via the Askaryan effect (Saltzberg et al. 2001), a well-known and studied electrodynamic process in high energy physics, which has not previously been exploited for planetary science.

ARIA can also operate as a passive radiometer, measuring the RF spectra from 250 MHz out to 1500 MHz. Different frequencies are sensitive to temperatures at different depths, which can be exploited to constrain geothermal gradient (Siegler et al. 2023; Brown et al. 2023). ARIA would use its radiometric measurements in concert with independent constraints on dielectric properties from its other RF techniques to help pioneer a technique for measuring geothermal heat flux without the need to drill.

With ARIA, we can address many outstanding questions in lunar and planetary science, such as:

• Are there subsurface deposits of ice in the cold polar regions of the Moon?
• What is the thickness of the regolith at a given landing site, and what is the nature of the regolith-megaregolith contact?
• Are there buried impact ejecta or melt deposits present in the subsurface, illuminating the chronology of impact processes that have occurred in that region?
• How thick are individual volcanic units, what are the stratigraphic relationships between them, and how do these units vary laterally?
• What is the geothermal heat flux at a given landing site?

Figure 1: An implementation of ARIA depicted on the lunar surface.

Cosmic ray RF sounding was recently recognized in a report (CLOC-SAT, 2022) commissioned by the Lunar Exploration and Analysis Group. We can use the cosmic ray spectrum and radio emission properties to cross-calibrate the active GPR results. Using ARIA’s beamforming methods, we are able to map cosmic ray events to their reflection points, and ultimately back to the shower vertex. From the cosmic ray methodology, we can directly measure dielectric parameters of the regolith and subsurface interfaces, a capability not possible with traditional GPR, which must assume a dielectric constant to infer a given target’s depth. This methodology has been demonstrated by the NASA-funded Antarctic Impulsive Transient Antenna (ANITA) stratospheric balloon payloads (Gorham et al. 2009), and was used to measure Antarctic ice properties (e.g., Prohira et al. 2018). ARIA would be the first extension of this cosmic ray methodology beyond Earth. 

While here we focus on deployment of the ARIA instrument on the lunar surface, simulations by Costello et al. (2025) show that cosmic ray-induced RF showers could be detectable from a sensor deployed in orbit. Tai Udovicic et al. (2025) suggest hundreds of events should be observable from the Moon’s permanently shadowed regions during a 2-year mission with the Cosmic Ray Lunar Sounder (CoRaLS) detector in an LRO-like orbit. Simulations also show that utilizing RF pulses generated by the Askaryan Effect yield capabilities for sensing subsurface layers thinner than that detectable by more traditional radar sounding or synthetic aperture radar methods. Additionally, ARIA can be integrated with other instrumentation (e.g., a seismometer suite) for comprehensive and complementary investigations (Bramson et al. 2023). Lastly, while the Moon is a logical location to employ an instrument like ARIA, the utility of this RF instrumentation could be realized for many applications across the Solar System (Prechelt et al. 2022).

References:

Bramson et al. (2023) 54th LPSC, Abstract #1797,
https://www.hou.usra.edu/meetings/lpsc2023/pdf/1797.pdf.

Brown et al. (2023) JGR-Planets, https://doi.org/10.1029/2022JE007609.

Costello et al. (2025) GRL, https://doi.org/10.1029/2024GL113304.

Gorham et al. (2009) Astroparticle Physics, https://doi.org/10.1016/j.astropartphys.2009.05.003.

Greenhagen, Pieters, Glotch, and the CLOC-SAT Specific Action Team (2022). Continuous Lunar Orbital Capabilities Specific Action Team Report. Lunar Exploration Analysis Group.

Prechelt et al. (2022) arXiv:astro-ph.EP, https://doi.org/10.48550/arXiv.2212.07689.

Prohira et al. (2018) Phys. Rev. D, https://doi.org/10.1103/PhysRevD.98.042004.

Saltzberg et al. (2001) Phys. Rev. Lett., https://doi.org/10.1103/PhysRevLett.86.2802.

Siegler et al. (2023) Nature, https://doi.org/10.1038/s41586-023-06183-5.

Tai Udovicic et al. (2025) 56th LPSC, Abstract #2860,
https://www.hou.usra.edu/meetings/lpsc2025/pdf/2860.pdf.

How to cite: Bramson, A., Gorham, P., and Costello, E. and the ARIA Proposal Teams: ARIA (Askaryan Regolith Imaging Array): An Instrument Concept for Novel Radio Frequency Characterization of Planetary Subsurfaces, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1208, https://doi.org/10.5194/epsc-dps2025-1208, 2025.

Drone-based ground penetrating radar (DGPR) is a novel tool to retrieve internal properties of debris-covered glaciers, including supraglacial debris thickness, bulk thickness, and detection of englacial debris layers, demonstrating a unique potential of drones to study buried ice reservoirs on Mars [1, 2]. We present the results of surface clutter analysis as part of a project involving a GPR Geodrone 80, centered at 80 MHz, mounted on a DJI Matrice 600 drone.

Since the DGPR platform is above the surface, identification of subsurface interfaces requires the simulation of off-nadir surface reflections (“clutter”) in order to rule out false positive detections from other sources. There is an inherent uncertainty in confirming internal reflectors in our study sites due to sloped surfaces, the presence of boulders, and the proximity to valley walls and trees (Figure 1). Radar surface clutter simulations are employed to model radar returns from off-nadir surface topography and have been successfully used to validate internal reflectors in glaciers in Antarctica [3] using airborne sounding radar, as well as over lobate debris aprons on Mars from orbital sounding radar [4, 5, 6].

Figure 1. Diagram of the DGPR and the detection of subsurface and off-nadir targets. Clutter occurs when the reflection from the off-nadir object returns at a similar travel time to the englacial reflections. (a) A tree A has a return with similar delays to the bottom of the ice B. (b) An off-nadir surface boulder D has a similar return as a buried boulder C. (c) A headwall F has a return with a similar delay as an internal debris layer F.

We used digital elevation models (DEMs) derived from drone photogrammetry (~5 cm/px) available for Galena Creek and Sourdough [7] to generate clutter simulations (“cluttergrams”) using state-of-the-art radar surface return simulation software [8. 9]. For each trace in the GPR profile, the clutter simulation software generates a faceted representation of the topography surrounding the drone when the trace was acquired (Figure 2a). Then, the expected reflection power is estimated for each facet using a modified version of the Friis transmission formula (Choudhary et al., 2016) and the two way travel time is calculated from the distance between the drone and the facet center. The reflected power estimates and travel times are assembled into a simulated radar trace for comparison to the GPR data. Given that the antennas of the MALA Geodrone 80 are dipoles (Figure 2b), we use a normalized dipole radiation pattern by multiplying to a factor of the azimuth angle (φ) as shown in equation 1 [10].

Figure 2. (a) Setup of the faceted representation of the topography to generate the clutter simulations. The step size of the facets in both the along and across track is 0.5 m. The total length of the along and across track is based on the time window for each acquisition, in case of a window of 800 ns, the maximum distance to a facet is 100 m. (a) Normalized field patterns of a half-wave dipole antenna. (Visser, H.J., 2012). The red center arrow represents the angle .

We validated three scenarios with the clutter simulator

• Although there are off-nadir reflectors coming from the forest, the glacier base is still visible in some sections of the profile (Figure 3).

• Dipping reflectors interpreted as englacial debris bands are not associated with clutter coming from the headwall (Figure 4).

• The reflector corresponding to a boulder is not obscuring the englacial reflectors (Figure 5). This reflector could have also been identified as clutter with hyperbola fitting, given that its value is close to the speed of light.

Figure 3. DGPR radar profile (a) and its clutter simulation (b) in the lower section of Sourdough Rock Glacier. The base is a feature observed in the radar profile but not in the clutter simulation. (c) Drone-based orthomosaic with the nadir of the clutter simulation in red (d) Echomap with the first return of the clutter simulation in black. (e) Oblique photo of the DGPR takeoff with the forest in the background, location marked as E in panel c.

Figure 4. Radar profile and clutter simulation at the cirque of Galena Creek. (a) Radar profile with multiple dipping reflectors interpreted as internal debris layers and basal reflectors. (b) Clutter simulation, no clutter was associated with the dipping reflectors or basal reflectors shown in panel a. (c) Hillshade with the nadir return. (d) Power map of the clutter simulation with the first return. (e) Orthomosaic. (f) Ground view with a section of the profile.

Figure 5. Radar profile (a) and clutter simulation (b) of lunch rock in Galena Creek. (c) Hillshade, white arrow indicates the glacier flow direction. (d) Power map of the clutter simulation (e) Orthomosaic indicating the location of the weather station (magenta star) from where the photo in panel f was taken. (f) Ground view. The profile F-F’ in panels c, d, and f shows the first return of the clutter simulation. The yellow arrow indicates the boulder location in all panels.

References

[1] Aguilar et al. (2024) EPSC 2024, https://doi.org/10.5194/epsc2024‐1271. [2] Aguilar et al. (2025) LPSC 2025, https://www.hou.usra.edu/meetings/lpsc2025/pdf/1693.pdf. [3] Holt et al. (2006) JGR: Planets, https://doi.org/10.1029/2005JE002525. [4] Holt et al. (2008) Science, https://doi.org/10.1126/science.1164246. [5] Plaut et al. (2009) GRL, Https://doi.org/10.1029/2008GL036379. [6] Baker et al. (2019) Icarus, Https://doi.org/10.1016/j.icarus.2018.09.001. [7] Meng et al. (2023b). Journal of Glaciology, https://doi.org/10.1017/jog.2022.90. [8] Choudhary et al. (2016) IEEE GRSL, https://doi.org/10.1109/LGRS.2016.2581799. [9] Christoffersen et al. (2024) Zenodo, https://doi.org/10.5281/zenodo.10595007. [10] Visser (2012) Antenna Theory and Applications, https://doi.org/10.1002/9781119944751.ch5.

How to cite: Aguilar, R., Christoffersen, M., Nerozzi, S., Meng, T., and Holt, J.: Validating internal reflectors from drone-based GPR in debris-covered glaciers with clutter simulations, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1936, https://doi.org/10.5194/epsc-dps2025-1936, 2025.

Introduction

The investigation of the subsurface of Mars provides key insights into its geological history and reveal signs of extinct life. This is why the ExoMars/Rosalind Franklin rover mission has been designed to search for biosignature in the subsurface of Oxia Planum [1], especially thanks to a drill [2] able to collect samples down to 2 m below the surface. The WISDOM GPR [3] aboard the rover will guide the the selection of suitable and safe underground targets. For this purpose, WISDOM will probe the Martian subsurface at frequencies in the 0.5-3 GHz range, guaranteeing a centimetric resolution and a penetration depth of a few meters.

Field test campaigns in partially controlled or natural environments are of the outmost importance to assess WISDOM performance and validate the data processing chain. This paper presents interpretation tools developed and applied to experimental data acquired with the flight spare model of WISDOM during a field test campaign in Svalbard in March 2022. Svalbard already hosted the AMASE campaigns [4] between 2007 and 2011 to prepare the ExoMars mission. Svalbard’s permafrost is very favourable to the penetration of waves and the main goal of this campaign was to collect data in a well-documented geological and glaciological environment [5;6]. Besides, the regions includes polygonal terrains, caves and outcrops that can be regarded as relevant Martian analogues.

Svalbard field test location

The WISDOM data presented in this paper have been acquired in March 2022 in the Adventdalen Valley, near Longyearbyen. The map Fig. 1 shows 3 investigated terrains namely

(i) Ice cave meanders 6 to 10 m deep inside the glacier.

(ii) A heavily fractured sandstone outcrop (25-m long profile). Outcrops will be amongst the ExoMars rover's priority targets to drill and seek for biosignature.

(iii) A polygonal terrain (four parallel lines crossing two polygon troughs).

WISDOM data interpretation methods

The up-to-date WISDOM processing chain was applied to the data [7]. In addition,  electromagnetic simulations were conducted to characterize the shape, dimensions and permittivity value of detected buried structures. Simulations are performed with TEMSI-FD, a 3D Finite Difference Time domain (FDTD) code [8] which can account for the actual radiation pattern of WISDOM antennas.

We also exploit the different polarization configuration measurements of WISDOM to constrain the geometry and the shape of detected targets. For instance, spherical targets have a low depolarisation efficiency whereas angular and corner reflectors are strongly depolarizing. Tilted interfaces change the direction of propagation of the waves as well as its polarization. To display the intensity in the different WISDOM polarimetric configurations, we produce additive RGB synthesis radargrams [7] by combining three of the four polarization configurations presented on Fig. 2. The different radiation patterns in polarization 0 and 1 (Fig. 2) allow to constrain the position of the red (only 00) and green (only 11) reflections to the right rear and left rear of the rover traverse respectively.

We use in a complementary way the Bandwidth Extrapolation (BWE) technique, a super-resolution method [9;10] which can improve the range resolution of WISDOM radargrams by a factor 3, allowing to meet the requirements of the ExoMars mission in materials of relatively low permittivity value.

Results

This section illustrates the use of simulations and polarimetric analysis on Svalbard data.

Contribution of 3D simulations to quantified interpretation

Fig. 3b and 3c present a vertical slice of a 3D geo-electrical model representative of the trough (with and without ice wedge underneath) separating two polygons in the polygonal terrain investigated with WISDOM (profile in yellow on Fig. 3a). Fig. 3e-f shows the corresponding simulated radargrams, which are compared to the experimental radargrams (Fig. 3d). 

In both experimental and simulated data, snow/trough interface reflection weakens and disappears when the interface slope becomes too steep. The main contribution of these simulated radargrams is to remove ambiguity about the 3 intense reflections around 1.5 m deep. The middle reflection cannot be explained without the presence of an ice wedge, and the deepest signals are actually due to double reflections on the faces of the trough. The phase shift reported on Fig. 3d-3e supports the hypothesis that WISDOM waves travelled through a lower permittivity material, consistent with the presence of an ice body (ε=3.2) beneath the active layer material (ε≥7).

Polarimetric analysis of the data

The RGB display of the WISDOM radargram acquired above the ice cave of the glacier (Fig. 4a) reveals a variety of polarimetric signatures. Snow layers and snow/ice interface appear yellow (Fig. 4b) ; they are almost horizontal and relatively smooth, thus not depolarizing. From 1.5 to 2.5 m deep on Fig 4b reflections are either red or green; these structures are therefore off-track (not directly underneath the antennas). These features are nevertheless not depolarizing. In contrast, the received waves associated with the ice cave (Fig. 4c) are rather depolarized, meaning they may have experienced multiple reflections in the cavity.

Conclusion

The Svalbard data set allows to validate the WISDOM performances and the interpretation tools. The fine range resolution meets the mission’s requirements and the polarimetry constrains the position and shape or nature of buried reflectors, which is essential for the selection of the drill site. Simulations and polarimetry provide together quantified analysis of the subsurface structures, which will significantly contributes to the geological understanding of the site in synergy with the instruments of the payload. WISDOM will investigate the Martian subsurface in a complementary way to the lower frequency and non-polarimetric GPR RIMFAX and to RoPeR.

References

[1] C. Quantin-Nataf et al., Astrobiology, 2021

[2] F. Altieri et al., Advances in Space Research, 2023

[3] V. Ciarletti et al., Astrobiology, 2017

[4] A. Steele et al., AGU Fall Meeting Abstracts, 2010

[5] M. Ulrich et al., Geomorphology, 2011

[6] N. Ross et al., Norwegian Journal of Geography, 2005

[7] Y. Herve, PhD Thesis, 2018

[8] C. Guiffaut et al., IEEE, 2012

[9] K.M. Cuomo, NASA STI/Recon Technical Report N, 1992

[10] N. Oudart et al., Planetary and Space Science, 2021

[11] W-S. Benedix et al., Planetary and Space Science, 2024

How to cite: Brighi, É., Ciarletti, V., Le Gall, A., Oudart, N., Hervé, Y., Plettemeier, D., Benedix, W.-S., Mas i Sanz, E., Shestov, A., Mercier, L., and Harrar, L.: Performance validation of WISDOM, the GPR of the ExoMars 2028 mission, on a Martian analogue dataset acquired in Svalbard, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-710, https://doi.org/10.5194/epsc-dps2025-710, 2025.

Introduction: Polarimetric analysis of SAR imagery can provide extensive information about the scattering properties of an imaged surface. When looking at the Lunar South Pole, where the presence and abundance of ice has been debated for decades [1][2], having in-depth information about the present scattering mechanisms is invaluable. The Dual Frequency SAR (DFSAR) onboard ISRO’s Chandrayaan-2 provides fully polarimetric data in both L- and S-band, with the former expected to be able to penetrate up to 5m in dry, low loss soils [3]. With most of the subsurface ice expected to be located within the Permanently Shadowed Regions (PSR) of craters, the high-resolution coverage of these regions provided by Chandrayaan-2 allows the areas to be studied in greater detail.

Study Region: Two craters, Kocher and Wiechert E, were selected for this study due to their anomalous nature – this meaning that the Circular Polarisation Ratio (CPR) of the crater is higher in the interior [4]. They both exhibit high backscatter returns around their rims, contain a PSR and have average temperatures lower than 100K. With Kocher having a diameter of >20km and Wiechert E having a diameter of <20km, their floor texture is an unusual characteristic, as often craters with a diameter greater than 20km are more complex and have features such as central peaks [5]. The radar-bright regions of the crater walls are a possible indicator of unique surface or subsurface materials.

Methodology: Utilising bespoke code developed in python to detect and then radiometrically calibrate each polarisation to beta0 (HH, HV, VH, VV) as a baseline visualization, further code is used to extract the Coherency (T) matrix [6] and Pauli vectors to begin polarimetric analysis. Where H represents horizontal and V represents vertical, the Pauli vector k is defined by

This base-level decomposition can be utilised to differentiate between surface scattering from the first matrix component S_HH + S_VV, dihedral structure scattering from the matrix component S_HH – S_VV, and volumetric scattering with the final matrix component 2S_VH.

Another technique used for in-depth analysis of the surface scattering mechanisms is the Claude-Pottier decomposition: a second-order optimisation method for distributed targets that provides information on the randomness of the scattering mechanism (entropy) and the potential dominant component. The scattering components of the Cloud-Pottier decomposition are derived from the 3x3 [T] matrix.

Initial Results: Kocher crater can be seen in Fig. 2, with the hummocky textured floor and large PSR represented in blue. The bright radar return from the western wall could be due to wavelength sized scatterers within the crater wall, but equally it is common for sloped terrain perpendicular to the look angle of the radar to appear bright. Since the Pauli Decomposition has high-level links to the physical scattering mechanism, we can map its components in colours (see Fig. 1b): the volumetric scattering is represented in green, while the surface scattering seen across most of the swath is represented in blue. The volumetric scattering component from natural mediascould be a representation of either subsurface ice suspended within the regolith or crustal material, or other wavelength sized scatterers such as larger rocks mixed into the Lunar regolith. Further analysis utilising the Claude-Pottier code to constrain the origin of the scattering mechanisms is being undertaken – including the possibility of distributed subsurface ice.

Fig. 1. (a) 5x5 Lee filtered HH and VH swath surrounding Kocher Crater (b) Pauli Decomposition of swath surrounding Kocher Crater.

Fig. 2. West-looking S-band image from LRO Mini-RF of Kocher. The outlined blue areas are the PSR. Image taken using LROC Quickmap.

Future Plans: A deeper investigation integrating the results from different decompositions and the backscatter of all polarisation channels is aimed to be completed for the surrounding areas of each crater. The plan is to then mosaic the swaths together once projected onto the ground and create full images of each crater for a complete study.

When completed, the code used for this project is planned to be released.

With NASA’s Artemis III mission landing areas planned to be around the South Pole, and many other space agencies and commercial companies sending missions to this region, understanding the geological context of the area is key to success, in particular the presence and accessibility of water ice.

Acknowledgments: We acknowledge the use of data from the Chandrayaan-II, second lunar mission of the Indian Space Research Organisation (ISRO), archived at the Indian Space Science Data Centre (ISSDC). UK Space Agency for funding this project; grant no. ST/Y005384/1 as part of the UK Government’s Science Bilateral Programme. LROC Quickmap team for hosting datasets and the map.

References: [1] Watson K. et al. (1961) J. Geophysical Research, 66, 9. [2] Campbell D. B. et al. (2006) Nature, 443, 835-837. [3] Bhiravarasu S. S. et al. (2021) Planet. Sci. J., 2, 134. [4] Spudis P. D. et al. (2013) JGR Planets, 118, 10. [5] French, B. M. (1998). LPI Cont. No 954, pp. 27. [6] Lee J. S. and Pottier E. (2017) CRC press.

How to cite: McVann, P., Ghail, R., Gallardo i Peres, G., Mason, P., Marino, A., and Knight, C.: Polarimetric Analysis of Anomalous Lunar South Pole Craters from Chandrayaan-2 DFSAR Data, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1701, https://doi.org/10.5194/epsc-dps2025-1701, 2025.

Introduction:  Understanding the subsurface structure is important for elucidating a planet’s geological evolution and thermal environment from its formation to the present. The Lunar Radar Sounder (LRS) onboard SELENE has detected buried regolith layers formed between subsurface lava flows [1]. While most radar surveys have focused on large-scale horizontal features (tens to hundreds of kilometers), (hundreds of meters to several kilometers) is gaining importance. Theoretically, lunar magma ascends due to buoyancy and volatile exsolution, and if it fails to erupt, gas voids can form at the magma tip [2]. Detecting such cavities would provide evidence of volatiles in lunar magma. In addition, lava tubes, which are potential habitats due to their thermal stability and radiation shielding, are key targets for future lunar exploration and base construction [3].
To detect small-scale subsurface echoes from LRS data, it is essential to accurately simulate the surface scattering from the lunar surface and to subtract this component from the LRS data. Kobayashi et al. (2020) simulated surface scattering using the Stratton–Chu integral method[4], but the resulting echo intensities were often higher than LRS observations at depths greater than 500 m, possibly due to artifacts in the SLDEM elevation model used.
In this study, we develop the detection method of small-scale subsurface echoes using LRS data and surface scattering simulation based on high-precision DEM generated by a Generative Adversarial Network (GAN). Furthermore, we analyze the spatial distribution of subsurface echoes and evaluate the subsurface structure indicated by the detected subsurface echoes.

Methods: The analysis in this study consists of three steps: (1) generation of a high-resolution DEM using a GAN, (2) surface scattering simulation, and (3) detection of subsurface echoes. In the DEM generation phase, we construct a GAN model using high-resolution images (LRONAC images), a low-resolution DEM (SLDEM), and a high-resolution DEM (NACDEM). Then this trained model is then used to generate high-resolution DEMs (hereinafter GANDEM) by inputting SLDEM2015 and LRONAC images. In the surface scattering simulation phase, we calculate the electromagnetic field scattered from the lunar surface using GANDEM. The simulation is conducted based on the method proposed by Kobayashi et al. (2020) [4]. Finally, by comparing the simulated B-scans with LRS observations, we identify small-scale subsurface echoes. Figure 1 shows the overall process of the proposed approach.

Results: We generated GANDEM for the region of the Mare Tranquillitatis (latitude: 7°N-11°N , longitude: 31°E-34°E). Figure 2 shows the GAN-generated DEM (GANDEM) (Fig. 2(d)) for a portion of that region, as well as the corresponding  LRONAC image (Fig. 2(a)) and SLDEM2013 (Fig. 2(b))  NACDEM (Fig. 2(c)). SLDEM2013  cannot reproduce crater shapes with diameters of less than approximately 100 m, compared to NACDEM (Figs 2(b) and 2(c)). On the other hand, GANDEM succeeds in expressing small craters at the same level as NACDEM (Figs 2(c) and 2(d)).
Using the GANDEM, we performed surface scattering simulations for LRS observation lines in the Mare Tranquillitatis. Figure 3 shows results from one representative track: (a) LRS observation radargram, (b) simulation result, (c) and (d) are their respective track-averaged radargrams, and (e) shows the pixel-wise intensity difference between (c) and (d). Figure 4 presents the histogram of intensity differences aggregated from 15 ground tracks in the region. A Gaussian Mixture Model (GMM) was used to fit two distributions. The subsurface echo detection threshold was set at 3σ of the distribution centered near 0, corresponding to 4.50.
Figure 5 shows the distribution map of the extracted SECs. The color represents the apparent depth. The three dotted lines represent grabens, which are linear depression structures on the surface. SECs exist continuously on the northwest extension of graben A, shown by the white solid ellipse. In areas that were not extensional regions of the graben, some SECs were detected continually along multiple LRS ground tracks shown as white dotted ellipses.

Discussion: SECs have been observed along the northwestward extension of Graben A, which is likely formed by subsurface magma intrusion. These SECs may represent subsurface structures associated with the same intrusion. Additionally, SECs have been continuously detected along multiple LRS ground tracks in areas unrelated to grabens. This suggests the presence of other horizontally continuous subsurface features, such as lava tubes or magma-related voids.

We next consider alternative explanations for the observed SEC intensities and distributions. While echoes from regolith layers between lava flows have been reported in previous LRS studies [1], such layers would extend horizontally for hundreds of kilometers. This wide distribution cannot explain the continuous and linear distribution of SECs detected on the multiple LRS ground tracks in this study. Subsurface faults may return echoes stronger than –25 dB if their planes are oriented nearly perpendicular to the radar beam. However, with average lunar fault angles of ~60° [5], such reflections would occur in radargrams at depths of ~200 km—far deeper than the 350–800 m apparent depths observed for the SECs.  Consequently, no fault could be the reflection source for the SECs. Layers of ice and liquid water returning strong radar echoes have been discovered under the surface of Mars [6], but those would not exist in the lunar equatorial region.

Therefore, more plausible subsurface structures indicating SEC detected along multiple LRS ground tracks are considered to be the exsistence of lava tubes  or gas voids at magma tips.

References: [1] Ono  et al. (2009) Science, 323.5916, 909-912. [2] Head and Wilson, (2017) Icarus, 283, 176–223.   [3] Kaku et al., (2017) GRL, 44.20, 10-155. [4] Kobayashi et al., (2020) IEEE Transactions on Geoscience and Remote Sensing, 59.9, 7395-7418. [5]  Golombek  et al., (1979) Journal of Geophysical Research: Solid Earth, 84.B9, 4657-4666 [6] Orosei et al., (2018) Science, 361, 490–493.

How to cite: Nozawa, H., Haruyama, J., Kumamoto, A., Toyokawa, K., Iwata, T., Head, J., and Orosei, R.: Small-Scale Subsurface Structure Detection using Lunar Radar Sounder and Surface Scattering Simulations Based on GAN-Generated DEM, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1128, https://doi.org/10.5194/epsc-dps2025-1128, 2025.

1. Introduction

Secondary craters result from the impact of primary ejecta falling back to the surface. Closer to the primary crater, secondaries form at low velocities and show asymmetric shapes (McEwen et al., 2005; Pike & Wilhelms, 1978; Oberbeck & Morrison, 1973; Melosh, 1989), while farther away, higher-velocity ejecta produce secondaries resembling small primaries (Bart & Melosh, 2007). This range-induced morphologic diversity creates a challenge in distinguishing simple primaries from distant secondaries, affecting crater-chronology accuracy (McEwen & Bierhaus, 2006).

On the Moon, radar studies have found that some small craters exhibit radar-bright, asymmetric ejecta that is enhanced in cm- to dm-scale debris and may be attributed to secondary formation. Specifically, potential secondaries with ejecta extending downrange from Tycho crater have been identified within Tycho’s rays (Wells et al., 2010; Watters et al., 2017), including over the South Pole and intersecting Haworth, a potential Artemis landing site (Rivera-Valentín et al., 2024). Here, we used multiple datasets from the Lunar Reconnaissance Orbiter (LRO), including Mini-RF radar observations, to catalog and characterize Tycho secondaries along its rays. Our goal is to investigate the causes of these radar-bright, asymmetric ejecta deposits (termed “radar tails”) and explore their implications for secondary and ray formation processes.

2. Methods

Potential secondary craters along several of Tycho’s rays (Fig. 1) were first identified in radar imagery via the presence of a radar-bright, asymmetric ejecta signature oriented away from Tycho. We used S-band (12.6 cm) monostatic radar data from LRO Mini-RF, a hybrid-polarimetric SAR. Data products analyzed included circular polarization ratio (CPR) and m-chi decomposition images (Raney et al., 2012). The m-chi RGB composite displays even-bounce (red), random (green), and odd-bounce (blue) scattering.

Figure 1: Distribution of Tycho secondary craters with radar-bright extended asymmetric ejecta deposits (“radar tails”) identified in this study.  

In addition to our radar characterization, we measured the depth and diameter for each potential secondary crater using LOLA elevation data (Barker et al., 2016) and calculated depth-to-diameter (d/D) ratios. The same method was applied to nearby primary craters (>300 m in diameter, without radar-bright ejecta tails) for context.

To study the morphology of secondary craters and understand the cause of the ejecta tails observed in the radar products, we then used both LROC NAC images and ShadowCam images to describe the surface texture of the radar tails and document the presence of boulders and assessed the maturity of the craters.

3. Results

We identified 473 craters in Tycho’s South Polar ray, including 24 secondaries at Haworth (Fig. 2, candidate landing site for the Artemis missions), and 543 more secondaries in the other bright rays. Tycho secondaries at Haworth could offer a chance to sample Tycho ejecta and compare it to Apollo 17 material (Jolliff et al., 2020; Rivera-Valentín et al., 2024).

Supporting our interpretation that the craters with radar tails are secondary craters is the observation that radar tail craters have lower d/D ratios (~0.04) than primaries (~0.10) identified in the same area. Preliminary crater size-frequency distribution analysis also supports this interpretation.

Figure 2: Tycho secondary craters with extended asymmetric ejecta (magenta arrows) in Haworth.

We find that Tycho secondary craters exhibit three possible classes of extended asymmetric ejecta: (1) isolated secondaries (only one radar-bright tail extends from the crater), (2) clusters (multiple tails merge into one emanating from a cluster of craters), and (3) linear crater chains (closely spaced craters arranged roughly straight line). We observed that the enhanced depolarized backscatter signals associated with the radar tails can be attributed to three main factors: (1) subsurface scattering likely caused by buried structures, such as subsurface blocky material or debris (based on the lack of observed surface features on visual images, (2) scattering from surface boulders, and (3) presence of craters (with D>200m) along the radar tail.

Figure 3: We found 3 main tail textures that could be causing the depolarized scattering (a) buried structures, (b) surface boulders, (c) additional craters.

4. Conclusions

We confirm that radar data can be used to identify fresh secondary craters on the Moon via the presence of extended, asymmetric radar-bright ejecta "tails" that trace surface roughness and subsurface scatterers. These tails, aligned with topography and pointing away from the primary impact site, may reflect ejecta debris flow processes, such as a debris surge (Campbell et al., 1992; Wells et al., 2010; Watters et al., 2017). Radar features often extend beyond visible surface textures, suggesting that much of the structure lies beneath the surface. Radar datasets are therefore uniquely placed to examine secondary processes.

References:

• Barker et al. (2016). Icarus, 273, 346–355.
• Bart and Melosh. (2007). Geophysical Research Letters, 34(7).
• Jolliff et al. (2020). 54th LPSC 2023 (LPI Contrib. No. 2806).
• McEwen & Bierhaus (2006). Annual Review of Earth and Planetary Sciences, 34, 535–567.
• Melosh (1989). Impact cratering: a geologic process. New York: Oxford University Press; Oxford: Clarendon Press
• Oberbeck et al. (1973). The Secondary Crater Herringbone Pattern, 4, 570.
• Pike & Wilhelms (1978). Secondary-Impact Craters on the Moon: Topographic Form and Geologic Process, 907–909.
• Raney et al. (2012). JGR: Planets, 117(E12).
• Rivera-Valentín et al. (2024). PSJ 5(4), 94
• Watters et al. (2017). JGR: Planets, 122(8), 1773–1800
• Wells et al. (2010), JGR: Planets, 115, E06008
• Martin-Wells et al. (2017). Icarus, 291, 176-191

How to cite: Perez-Cortes, S., Rivera-Valentin, E., Ahrens, C., Bramson, A., Fassett, C., Nypaver, C., Morgan, G., and Patterson, W.: Characterization of Tycho Secondary Craters on the Moon Using LRO Mini-RF Radar Data: Implications for Formation Mechanisms, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-861, https://doi.org/10.5194/epsc-dps2025-861, 2025.

Introduction:
The Martian North Polar Layered Deposits (NPLD) are a kilometers-thick stack of icy layers, with a bulk composition of 95% water ice and the rest being silicate dust and lithics [1–2]. Individual layers within the NPLD vary in dust content, ranging from negligible amounts to potentially ~60% dust [3]. These differences are believed to be controlled by orbital forcing in Martian history, where each layer contains a snapshot of the paleoclimate at the time of deposition [4]. Specifically, it is believed that the large variations in Martian obliquity, ranging from ~15°– 60°, create alternating periods of ice accumulation and ice loss [5]. Our best tool to understand the layering within the entire vertical extent of the NPLD is orbital radar sounding, such as that from the Shallow Radar (SHARAD) onboard the Mars Reconnaissance Orbiter (MRO). However, visible imaging of exposures of the upper NPLD have shown that layers can have thicknesses down to decimeter to meter scales [6], smaller than SHARAD’s vertical resolution of ~10m. The impact that these sub-resolution layers have on radar observations is not completely understood [7–8], and it is hypothesized that a radar reflector within the NPLD may be the result of either a single relatively thick layer, or “packets” of thinly spaced layers [9]. Therefore, to understand the Martian paleoclimate recorded in the NPLD, we must have a better understanding of how different scales of layering influence the radar observations. To address this, we combine results from laboratory analog experiments and wave propagation simulations with Mars’ remote sensing observations.

Laboratory Setup:
We create a series of laboratory analogs to test how radar observations are influenced by various sub-resolution layer thicknesses and spacings. Our radar measurements are made using a 2.7 GHz GSSI StructureScan Mini XT system, which is higher frequency than SHARAD but allows for experiments of a tractable size. Our experiments consist of different configurations (Fig. 1) of 100% water ice layers and dust-rich layers composed of ~55% Martian regolith simulant MGS-1 [10] and ~45% ice, within a 72cm × 72cm × 45cm polyethylene container housed in a -20°C freezer. These experiments focus on dusty layer thicknesses and spacings ranging from λ/2 to λ/16 (where λ is the wavelength of the transmitted radar wave in free space, 11 cm for 2.7 GHz).

Fig. 1 Example of various dusty layer configurations to be tested. Experiments are conducted in a 72cm × 72cm × 45cm polyethylene container and radar data is collected using a commercial radar system with a center frequency of 2.7GHz (free space λ ≈ 11cm).

Preliminary Results:
Current results indicate that the critical thickness for a single dusty layer to have distinct upper and lower reflectors in our NPLD analog experiments is > λ/8, which matches theoretical expectations [7]. Fig. 2 shows radargrams for a single λ/4 thick dusty layer, where upper and lower reflectors are distinct, and a single λ/8 dusty layer, where the upper and lower reflectors are combined into one reflector. Based on early results, coarsely spaced layer packets may be differentiated from singular thick dusty layers based on total power returned from the lowest reflector. In Fig. 3, a singular λ/2 thick dusty layer has reduced power returned from the reflector corresponding to the bottom of the layer, compared to that from a layer packet configured with two λ/8 thick dusty layers, spaced λ/2 apart. Combining these initial results with more experiments and simulations to extrapolate results to larger scales and lower frequencies will establish a guideline to interpreting SHARAD data of the NPLD.

Fig. 2 A) Radargram of a single λ/4 (2.75 cm) thick dusty layer. Upper and lower contacts with the ice show distinct separate reflectors. B) Radargram of a single λ/8 (1.375 cm) thick dusty layer. Upper and lower contacts do not show distinct reflectors and combine into one reflector.

Fig. 3 Comparison of radar power returned for two experimental configurations: a single, thick (λ/2; 5.5cm) dusty layer and two thinner (λ/8; 1.375 cm) dusty layers spaced λ/2 (5.5cm) apart. Left side of each subpanel illustrates the stratigraphic configuration (centered on the depth of the middle of the experimental column), with dusty layers in orange and water ice in light blue. Right side of subpanels shows radar power (relative to transmitted power) returned for detected interfaces associated with the layer(s).

Future Work:
More experiments will be run, as shown in Fig. 1, to better understand how the power return of reflectors changes with thickness and spacings. Additionally, experiments focused on packets of λ/8 thick layers at increasingly thinner spacings will be conducted, to understand when layer packets appear as a singular layer in radar data. We are also running simulations of these experiments using two Finite-Difference Time-Domain based electromagnetic wave propagation softwares, gprMax [11] and Remcom’s XFdtd to allow us to relate our empirical laboratory results to theoretical results. For example, in Fig. 4, we show our measured experimental radargrams compared to preliminary simulations of waveform propagation for the same configuration. We will apply these laboratory results and simulations to apply our findings to SHARAD data and place new constraints on layer properties of the NPLD.

Fig. 4 Comparison of experimentally measured radargrams (left) and gprMax simulations (right). A) Results for the configuration of a single, thick (λ/2; 5.5cm) dusty layer. B) Results for the configuration of two thinner (λ/8; 1.375 cm) dusty layers, spaced λ/2 (5.5 cm) apart.

References:
[1] Malin (1986) Geophys. Res. Lett., 13(5), 444–447. [2] Sinha & Horgan (2022) Geophys. Res. Lett., 49(8). [3] Lalich et al. (2019) JGR Planets, 124(7), 1690–1703. [4] Byrne (2009) Ann. Rev. of Earth and Planet Sciences, 37, 535–560. [5] Montmessin (2006) Space Sci. Rev., 125 (1–4), 457–472. [6] Herkenhoff et al. (2007) Science, 317, 1711–1715. [7] Widess (1973) Geophysics. 38(6), 1176-1180. [8] Zeng (2009) Leading Edge. 28(10), 1192-1197. [9] Putzig et al. (2009) Icarus, 204, 443–457. [10] Cannon et al. (2019) Icarus, 317, 470–478. [11] Warren et al. (2016) Computer Phys. Comms., 209, 163–170.

How to cite: Harris, S., Bramson, A., and McGlasson, R.: Effects of thin layers on radar observations of the Martian polar layered deposits: An integrated approach using experiments, simulations, and spacecraft observations, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1086, https://doi.org/10.5194/epsc-dps2025-1086, 2025.

Introduction:  The Radio Echo Sounding (RES) technique is commonly used on Earth to study the internal structure of ice sheets and understand ice flow (e.g., [1], [2]). Although large ice caps also exist at the Martian poles, their internal deformation and flow dynamics remain largely unknown. This study presents radar data from the Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS; [3], [4]) at Ultimi Scopuli on Mars (Fig. 1A, B), revealing the radar first evidence of large-scale englacial folds within the South Polar Layered Deposits (SPLD). These features are similar to those found in Earth's ice sheets, indicating that the Martian ice sheet has experienced movement under either dry or wet basal conditions.

Results:  The radar sections collected at US (Fig. 1C) exhibit a general stratified ice structure, typically ~1500-2500 m thick, overlying a flat/slightly rough sub-horizontal bedrock (at MARSIS resolution). The ice stratigraphy is generally characterized by sub-horizontal and continuous or gently undulated and disrupted reflection horizons that are roughly equally divided in two principals radar units by a semi-continuous and sub-horizontal sharp and thick radar reflector (labelled “H1” in figures). In some cases, the radar horizons are locally slightly inclined and truncated by the H1, suggesting an unconformable discontinuity surface. The lower unit (labelled “U1” in figures) is ~700-1200 m thick and is mostly characterized by low radar reflectivity with few relatively thick and fuzzy internal reflection horizons. On the contrary, the upper unit (labelled “U2” in figures) is ~800-1200 m thick and mainly composed by a thinly stratified radar sequence with bright reflectors.

Observing the MARSIS radargrams (i.e., Fig. 2), in its interior the ice sheet layering is locally interrupted and disrupted by clear radar dark-areas of non-reflective ice (from now on “Non-Reflective Areas” or “NRA”). However, in their inner core, the NRAs sometimes show bright-toned and rounded/elliptical layering (“eye-folds” like structures”) and “anti-form” like folding. Some other folding can partially envelop the dark NRAs or are located near such structures. In general, the NRAs have an irregular rounded to elliptical shape and usually develop from the interface between the bedrock and the base of the icy mass. On average, surveyed NRAs are ~16 km long and several hundreds of meters thick (up to ~1500 m), thus they are markedly elongated along the horizontal x-axis. All the above-described radar features are mainly present in the basal unit (U1), below the depth of ~700-1000 meters (i.e., below the H1 horizon); nevertheless, in some cases (e.g., Figs. 3, S1 and S3) deformation structures appear to interrupt or to be upon the H1 horizon, partially involving the upper sequence (U2), however always lying below ~500 meters in depth from surface.

Discussion and Conclusions:  In terrestrial ice sheets, highly deformed layers linked to areas of high dynamic ice flow driven by variation in basal shear stress typically cause strong power loss in internal reflectors of ice-penetrating radars (e.g., [5], [6]). In RES images, these zones tend to be dark (i.e., “layer free”) and interrupt the lateral continuity of the horizons from the bedrock up to some depth since internal layering slopes are most extreme near the ice bed (it is exhibited strong deformation) and thus power loss intensify at depth [5]. In such circumstances, the lossy/diminished radar power areas can be bordered by bright folded reflectors/undefined areas [1], [7], [8] and can include folded bright reflectors in their core as well. All these radar structures are the signature of broad asymmetric/complex geological shear folds, characterized by one limb steeper than the other [5] and/or completely tilted and classified as overturned/recumbent and sheath folds (e.g., [1], [7, [8]).

Net of the unavoidable approximations due to the different frequency and resolution of the radar instruments, we compared the main NRAs radar features detected by MARSIS at US with some examples of the above described englacial structures observed by RES in Greenland and Antarctica polar ice sheets (e.g., Fig. 3 after [1], [7], [8]). The data matching points out their consistency in terms of radar morphology and morphometry, and the most plausible interpretation is that large-scale geologic folds are affecting the SPLD stratigraphic architecture of US ice-sheet. In particular, considering the overall setting of the main surveyed structures, the MARSIS dark NRAs and the deformed radar patterns in U1 unit seem to be consistent with the englacial sheath folds known on Earth. The existence of sheath folds states that large displacements have occurred in the US region due to ice-flow/basal sliding, similarly to what observed in the SPLD in Promethei Lingula by previous studies [12]. On Earth, the origin of large englacial folds is still debated, with various dry or wet mechanisms proposed. Several mechanisms (dry or wet) have been proposed since now, often depending on case to case (e.g., [1], [9], [13]). Understanding similar Martian structures is harder due to limited data, environmental knowledge, and structural analysis. Ongoing studies aim to shed light on all these aspects.

References: doi: https://10.1002/2014JF003215. [2] doi: https://doi.org/10.1017/aog.2020.11. [3] doi: https://10.1016/j.pss.2009.09.016. [4] Picardi G. et al. (2004) in Mars Express: The Scientific Payload, ESA SP-1240. [5] doi: https://10.3189/2014AoG67A005. [6] doi: https://10.5194/essd-14-763-2022. [7] doi: https://10.1002/2014GL062248. [8] doi: https://10.1002/2015JF003698. [9] doi: https://10.1016/j.jsg.2015.09.003 [10] doi: https://10.1016/j.jsg.2006.05.005. [11] doi: https://10.1111/j.1365-3121.2012.01081.x. [12] doi: https://10.1016/j.icarus.2012.06.023. [13] doi: https://10.1016/j.epsl.2015.10.024.

Acknowledgements: This research was supported by the Next Generation EU program, Mission 4, Component 1, through project “Combining mAchine Learning and optImization for Planetary remote Sensing missiOns” (CALIPSO), Unique Project Code C53D23010010001.

Fig. 1. MOLA location and MARSIS radargrams

Fig. 2. Examples of “NRAs” and englacial folding (“eye-folds” -EF- and “anti-form” -SF- folding like structures) in MARSIS radargrams.

Fig. 3. Example of folding in RES (Greenland) compared with MARSIS structures. Image after Figure 13 in [1] © AGU-Wiley.

How to cite: Guallini, L., Orosei, R., and Pettinelli, E.: First Evidence of Large-Scale Englacial Folding in the South Polar Layered Deposits (Ultimi Scopuli, Mars) Unveiled by MARSIS, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-72, https://doi.org/10.5194/epsc-dps2025-72, 2025.

Introduction: One of the main goals of Mars science and exploration is to explore the shallow subsurface (i.e., 10s-100s meters) and reveal its geologic nature and composition [1]. Although many remote sensing techniques can probe the subsurface, there is currently a sensing gap at depths between ~1 and 100s of meters. Radar surface reflectivity analyses can fill this gap thanks to sensitivity to changes in composition and geologic structures at ~5-15 m depth with the Shallow Radar (SHARAD, [2, 3]) and ~50-150 m depth with the Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS, [4, 5]). Historically, MARSIS surface reflectivity analyses were limited by insufficient coverage and ionospheric effects, which caused attenuation and distortion of the radar echoes. Thanks to new processing techniques and nearly global coverage after continuing MARSIS observations for nearly 20 years [6], it is now possible to probe the Martian near-subsurface at four distinct frequencies ranging from 1.3 MHz to 5.5 MHz.

Methods: We employ a recently released PDS MARSIS dataset that corrects ionospheric phase distortions and provides three empirical parameters associated with ionospheric total electron content (TEC), in turn related to radar echo delay and attenuation [7, 8]. MARSIS operates at four distinct frequency bands, centered at 1.8, 3, 4, and 5 MHz, each with 1 MHz bandwidth, resulting in a free space vertical resolution of 150 m. This allows us to construct four reflectivity maps, each corresponding to a MARSIS band and probing to larger depths as frequency decreases. We process the surface echo power for each frequency through the following steps:

• Automatic picking of surface echoes based on global topography and “clutter” simulations [9].

• Path loss correction to account for surface echo power changes due to spacecraft altitude.

• Ionospheric loss correction via normalization of surface power with respect to solar zenith angle (SZA) and the three TEC parameters (example in Fig. 1).

• Removal of surface geometry effects via normalization of surface power with respect to slope measured from the Mars Orbiter Laser Altimeter (MOLA, [10]) 463 m/pixel global DEM and surface roughness calculated at 0.6 km, 2.4 km, and 19.2 km baselines from [11] (example in Fig. 2).

Finally, we generate surface reflectivity maps for each MARSIS band by gridding corrected power values at 5 km/pixel resolution and accounting for the variable footprint of each echo, measured as the radius of the 1^st Fresnel zone.

Figure 1: Example of ionospheric effect corrections based on SZA (top) and TEC1 parameter (bottom) applied to the 3 MHz surface echo power data. 

Figure 2: Example of surface geometry corrections based on slope (463 m/pixel) and roughness (at 0.6 km baseline, [11]) applied to the 3 MHz surface echo power data. 

Results: We find a strong surface echo power dependence on the local SZA (Fig. 1), with lower frequencies experiencing loss at higher SZA values, an expected behavior for MARSIS [5, 8]. After applying the SZA correction, we still find a non-linear dependence on all TEC parameters (Fig. 1), showing that up to 3-4 dB of loss occurs even at low TEC values. There is also a significant dependence on slope and 0.6 km baseline roughness (Fig. 2), suggesting that MARSIS echo power is most susceptible to short wavelength roughness.

Figure 3: Preliminary MARSIS surface reflectivity maps at the 3 and 4 MHz bands, plotted on MOLA shaded relief. 

After empirical correction, the preliminary surface reflectivity maps (examples in Fig. 3) reveal several prominent features, such as high surface reflectivity associated with lava flows in Tharsis and the Elysium/Amazonis Planitiae, and low reflectivity zones corresponding to the friable Medusae Fossae Formation. The Martian northern plains reveal a wide range of reflectivity likely associated with variable terrain types including ice-rich deposits [12, 13] and the latitude-dependent mantle [14].

Discussion: The consistency of surface power measured across adjacent profiles in our preliminary maps indicates that the correction of path and ionospheric losses is effective at all SZA values, thus eliminating the need for filtering. Despite the low number of observations and significant ionospheric losses, we also obtained the first Mars surface reflectivity map at 1.3-2.3 MHz.

There is a weak residual surface power dependence on roughness, likely due to simplifications in our preliminary approach. To address this, we are currently running a detailed analysis of several roughness and slope parameters within the 1st Fresnel zone of each MARSIS echo, leveraging the entire MOLA shot point dataset [10]. This will allow us to capture the effects of short wavelength roughness highlighted by the preliminary results at 0.6 km baseline (Fig. 2).

All reflectivity maps show enigmatic “sinuous stripes” across the southern highlands with low power values compared to surrounding areas. These features are consistent across adjacent profiles and their location and shape resemble ionospheric anomalies associated with remnant crustal magnetization observed by SHARAD [15]. A simple empirical correction based on the total crustal magnetic field [17] appears to remove most of these features, suggesting that MARSIS is particularly sensitive to ionospheric phenomena linked to crustal magnetization. We will further test this hypothesis via analyses of TEC parameters and magnetic anomaly data.

Overall, these results offer new insights into volcanic, sedimentary, and ice-related processes. They also lay the groundwork for future missions targeting subsurface ice or landing sites.

Acknowledgements: This work was supported by NASA MDAP grant 80NSSC22K1079.

References: [1] MEPAG Science Goals document. [2] Campbell et al. (2013)JGR: Planets. [3] Seu et al. (2007)JGR: Planets. [4] Mouginot et al. (2010)Icarus. [5] Jordan et al. (2009)PSS. [6] Orosei et al. (2015)PSS. [7] McMichael et al. (2017)2017 IEEE RadarConf. [8] Plaut (2024)Optimized MARSIS PDS4 release. [9] Choudary et al.(2016)IEEE GRSL. [10] Smith et al. (2001)JGR: Planets. [11] Kreslavsky and Head (2000)JGR: Planets. [12] Bramson et al. (2015)GRL. [13] Stuurman et al. (2016)GRL. [14] Kreslavsky and Head (2002)GRL. [15] Campbell et al. (2024)GRL. [16] Campbell and Morgan (2018)GRL. [17] Langlais et al. (2019)JGR: Planets.

How to cite: Nerozzi, S., Christoffersen, M., Morgan, G., Grima, C., and Plaut, J.: Measuring the Mars Global Surface Reflectivity at 1.3-5.5 MHz with MARSIS, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1857, https://doi.org/10.5194/epsc-dps2025-1857, 2025.

MARSIS radar onboard MEX did numerous observations of Phobos the past 20 years, trying to highlight geological structures in the Phobos subsurface. Up to now, these measurements never allowed us to identify subsurface features while the onboard software upgrade in 2023 promises a significant sensitivity improvement. On the other hand, a deeper analysis demonstrated a deviation between the orbit prediction and the ranging measurements done by MARSIS and then allowed us to derive a new constraint for Phobos ephemeris calculation.

Phobos’ orbit is currently known down to a precision of 300m, mostly directed along its track. It has mainly been determined with imagery, and more recently with the Super Resolution Channel of the HRSC camera onboard Mars Express (MEX). This method is associated with an error mainly normal to the plane of imagery. By dynamical constraints, Phobos’ trajectory determination error is mainly spread along its orbit.

In order to refine the orbitography and reduce the range error of the measurements, we propose to use data from the MARSIS (Desage, 2024). To do so, we perform a SAR synthesis on the MARSIS data in order to locate the radar echoes in a range/along-track plane. For every one of the 35 datasets at our disposal measured between 2008 and 2021, we also perform a coherent simulation using a Phobos shape model by Willner et al. (2014), and apply the SAR synthesis the same way we did for the MARSIS datasets. Given the geometry of our simulations and the SAR synthesis, the simulated radargrams are not sensitive to a range error of a few km in MEX’s trajectory, they can therefore be taken as reference points. We measure range errors between simulations and MARSIS data, distributed around +1km, with a standard deviation of 350m. The measurements being spread all around Phobos, the most probable cause for the non-zero average of the offsets measurements is an instrumental delay. After subtracting this average from the measurements, we estimate the offset of Phobos along its track that would create this standard deviation.

We find that this offset is of about 100m before 2017, and that the estimated value is rising linearly after this date to reach about 1.3km in 2021, date of our last observation. Since 2017 is the date of the last control point of the NOE-4-2020 ephemeris used for this study, our measurements exhibit a significant drift after this time. The systematic analysis of MARSIS data set from 2007 to 2021 is available in Desage, 2024 and was taken into account in the last Phobos ephemerid (V. Lainey, 2024, NOE-4-2024-MMX-D3). The possibility of a joint Radio-Science and MARSIS observation is also under evaluation to better constrain Phobos distances during Radio Science measurement and then improve gravity field measurement.

How to cite: Herique, A., Desage, L., Lainey, V., Kofman, W., Cicchetti, A., Orosei, R., and Andert, T.: MARSIS data as a New Constraint for Phobos’ orbit , EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-286, https://doi.org/10.5194/epsc-dps2025-286, 2025.

Mechanisms hypothesized for the formation of notable geologic features on Europa (e.g., chaos terrains) can alter the surrounding icy regolith to depths of a few meters to tens of meters (i.e., the near-surface) [1]. On Ganymede, variations in bulk composition and porosity may exist across the surface indicative of impact erosion and mass wasting processes [2]. The ability to characterize these compositional and physical properties (e.g., porosity), particularly in the context of its surface geology, provides a window into processes governing the evolution of their icy regoliths, both spatially and with depth.

The application of radar reflectometry to study the ‘surface’ return from sounders has been demonstrated to be a promising technique for characterizing near-surface ice on Earth and Mars [3, 4]. Such an approach can also be applied to future observations from the Europa Clipper and JUICE missions, both currently en route to the Jovian system. Both missions host nadir-pointing ice-penetrating radars: the Radar for Europa Assessment and Sounding: Ocean to Near-surface (REASON) on Europa Clipper [5] transmitting simultaneously at center frequencies of 60 MHz and 9 MHz, with bandwidths of 10 MHz and 1 MHz, respectively, and the Radar for Icy Moons Exploration (RIME) on JUICE [6] at 9 MHz center frequency with a bandwidth of 1 or 2.8 MHz.

Prior work used to investigate bulk near-surface properties and surface roughness relied on characterizing the coherent and incoherent content encompassed in the total surface return [7]. They utilized radar observations collected at near constant altitude (h). However, Europa Clipper and JUICE will perform flybys throughout their nominal tours, where altitude rapidly changes across the observation window. Thus, understanding how changes in altitude can affect the balance between observed coherent and incoherent backscattered energy is needed to confidently apply reflectometry for these missions.

Here, we simulate the radar surface echo from synthetic terrains, using a version of the multilayer Stratton-Chu coherent simulator with rough facets [8]. We assess the coherent content of the total surface power to changes in altitude and surface roughness, by comparing the power derived from simulated surface echoes at 9 MHz center frequency (Fig.1). Coherent and incoherent power falls off at different rates (h² versus h³, respectively) with increasing altitude [9]. The coherent content of the total return at a particular altitude over the target of interest could be utilized to invert for near-surface properties. Therefore, increasing platform altitude could be leveraged as a natural filter to establish stable coherent returns.

We further validate this approach for reflectometry with ice-penetrating radar data collected over Beardmore Glacier, Antarctica. These data were collected with the Multifrequency Airborne Radar-sounder for Full-phase Assessment (MARFA) instrument, a 60 MHz airborne radar sounder operated by the University of Texas Institute for Geophysics [10]. This particular groundtrack was flown to incorporate altitude variations that scale with the center frequency of the HF channel of RIME and REASON. We find that power fall-off is neither entirely proportional to h² nor h³ (Fig. 2). This suggests that the surface return consists of both coherent and incoherent energy, perhaps characteristic of cracks and crevasses as well as changes in firn porosity (i.e., icy regolith) across the glacier’s surface. In the case of RIME and REASON, this work implies that different terrain types (e.g., chaos terrains versus ridged plains on Europa) may be better observed at certain altitudes from the perspective of reflectometry. In addition, our results can provide insight into favorable targets and altitudes suitable for cross calibrating RIME and REASON at their shared center frequency and bandwidth to enable comparative radar studies across the Jovian icy moons. This science objective has been recommended in the latest JUICE-Clipper Steering Committee (JCSC) orbital report.

Fig 1. Simulated surface power (P) for synthetic terrains with stationary surface roughness, for altitudes ranging from 20 to 1000 km. Simulations were conducted with correlation length l_c = 0.25λ.

Fig 2. Observed surface power (P) over Beardmore Glacier for altitudes ranging from 600 to 5000 m.

References

[1] Daubar et al. (2024) SSR, 220(1), 18. [2] Moore et al. (1999) Icarus, 140(2), 294-312. [3] Grima et al. (2012) Icarus, 220(1), 84-99. [4] Chan et al. (2023) The Cryosphere, 17, 1839–1852. [5] Blankenship et al. (2024) SSR, 220(5), 51. [6] Bruzzone et al. (2015) IEEE IGARSS, 1257-1260. [7] Grima et al. (2014) P&SS, 103, 191-204. [8] Gerekos et al. (2023) Radio Science, 58(6), 1-30. [9] Haynes et al. (2018) IEEE TGRS, 56(11), 6571-6585 [10] Young et al. (2016) Phil Trans of the Royal Society A: MPES, 374(2059), 20140297

How to cite: Chan, K., Grima, C., Gerekos, C., Desage, L., Young, D., Blankenship, D., and Patterson, W.: Altitude-dependent radar reflectometry to characterize the near-surface of Jovian icy moons: perspectives from an Antarctic terrestrial analog, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-865, https://doi.org/10.5194/epsc-dps2025-865, 2025.

Figure 1: ε’ and σ of two doped ice samples in the frequency range 1MHz – 100MHz.

Figure2: The ratio of the power of the emitted and received signal corresponding to a discontinuity solid ice-brine at different depth (0.5km - 6km) for two possible temperature profiles (fig.3). The simulation propagates the signals through a crust with constant salinity (20mM). The expected noise for RIME is represented by the dashed line. 

Figure3: Two possible temperature profiles hypothesizing an icy crust of 5km with a surface temperature of 100K a salty mushy solution at the temperature of 260K. The blue line is the linear profile proposed by [8] and the red line is a power low proposed by [16]. 

How to cite: Turchetti, G., Brin, A., Lauro, S., Cosciotti, B., Mattei, E., and Pettinelli, E.: The radar investigation of Jovian moons: dielectric measurements and signals propagation, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-75, https://doi.org/10.5194/epsc-dps2025-75, 2025.

Space Rocks, such as asteroids and comets, constitute primordial remnants of the formation of our Solar System.  During the last decades these objects have become increasingly available to study, particularly those that enter near-Earth space, due to the advancement of detection and follow-up techniques. With the discovery and cataloging of most Potentially Hazardous Objects (PHO), our ability to anticipate possible impacts has significantly improved. 

Impact cratering is an ubiquitous process in the solar system, and Earth is not exempt from this phenomena. However, detection alone is not enough:  without accurate orbit prediction and knowledge of an object's physical properties, we cannot develop effective deflection strategies. 

In this context, the physical and dynamical characterization of near-Earth objects (NEO) has grown into a rapidly advancing field—evolving from purely cosmological interest to a cornerstone of planetary defense.

For over five decades the Arecibo Observatory in Puerto Rico, served as a key facility for these studies, via its planetary radar system instrumentation. Even with the closure of the facilities, the data endures, and are available to the scientific community and continue to be of great scientific value.  This work includes an effort to catalog all the objects detected at the Arecibo Observatory with its Planetary Radar System, which includes over 800 NEAs. As well as the use of some of these observations to enhance our knowledge of physical characteristics of these objects. 

The thesis is presented as a compilation of peer reviewed articles pertaining to the physical and dynamical characterization of a subset of NEAs observed with the AO Planetary Radar System. 

The first article: “Radar and Optical Characterization of Near-Earth Asteroid

2019~OK”,~demonstrates the adaptability in observation scheduling and the impressive precision of the radar technique. With just a few minutes of data, the reported astrometry secured its orbit for the next two centuries. This object was found to be a very fast rotator, having a rotation period of 3-5 minutes.

The next publication: “The Fastest Rotators: Near-Earth Asteroids Observed with the Arecibo Planetary Radar System", studies a subset of NEAs with rotation periods less than 10~minutes. In this work we compare optical and radar derived parameters, and find, in general,  both methodologies to be in good agreement. 

The last publication included is: “2020 BX12-The Last Binary Asteroid Discovered at Arecibo", which presents the analysis and shape modeling of this PHA. A rotation period typical of the population, close to the spin barrier, is found for the primary. Data acquired during the 2024 apparition using the Gran Telescopio Canarias (10.4~m) provided a spectrum which is in agreement with the expected taxonomy based on the radar derived circular polarization. 

This research integrates radar and optical data (Figure 1) from selected NEAs to contribute to our understanding of their physical and dynamical properties. By combining both optical and radar data sets, we can cross-validate and refine our findings. Understanding their composition is essential for developing potential planetary defense strategies, should they be needed in the future.  

Beyond enhancing our knowledge of rapidly rotating asteroids and binary systems, this study preserves a detailed record of all observations made by the Arecibo Planetary Radar system. This catalog offers the scientific community a unique opportunity to explore and make use of the previously under-explored Arecibo data archive, unlocking new possibilities for future discoveries.

Figure 1. Comparison of Measured values (Red box) for specific physical and dynamical parameters of NEA’s obtained via optical and radar observations and relationships with derived/estimated parameters (green box). Colored lines to mark parameters relationship. Arrow end direct to calculated values, open circles are derived. Solid black circles are physical or dynamical parameters.  

How to cite: Zambrano Marin, L. F.: Constraining physical characteristics of near-Earth Asteroids using Arecibo legacy radar data and optical observations, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-845, https://doi.org/10.5194/epsc-dps2025-845, 2025.

Introduction

Bilobed objects (contact binaries) appear in both asteroid and comet populations with a bimodal mass distribution and bifurcated shape. They are common in both near-Earth asteroid populations, estimated from radar imaging to be near 30% contact binaries [1], but also further out in the Kuiper belt, where optical observations suggest that upwards of 40-50% of Plutinos could be bilobate or elongated in shape [2,3]. Additionally, spacecraft imaging and radar observations of comet nuclei, while few in number, suggest that the majority are contact binaries.

The NASA Lucy mission discovered and imaged Selam, moon of (152830) Dinkinesh and the first known contact-binary satellite orbiting another asteroid in 2024 [4]. In April Lucy performed a flyby of its second target, asteroid (52246) DonaldJohanson, confirming the presence of another contact binary in the main asteroid belt. Now, 24 known contact binary objects have been imaged by spacecraft or modelled using radar to confirm their shapes [5]. Notable examples of these objects include (25143) Itokawa (NEA), visited by the Hayabusa mission; (486958) Arrokoth (KBO), visited by the New Horizons mission; and 67P/Churyumov–Gerasimenko, visited by the Rosetta mission.

The current sample of known contact binaries can be visually categorised into different types of objects, with the level of bifurcation varying dramatically between objects such as (4179) Toutatis and (388188) 2006 DP14. By modelling more of these objects, we can investigate whether particular morphologies are more prevalent, giving more insight into the methods behind contact binary formation.

Targets

We are modelling two targets with radar and optical data to investigate the differing shapes and morphologies within the sample of modelled contact binaries.

We are refining the shape model of asteroid (275677) 2000 RS11, henceforth RS11, which was preliminarily modelled in 2014 with only radar data [6], with the inclusion of the optical lightcurve data, for which the current model does not strongly agree. RS11 appears to have a ‘rubber duck’ shape similar to 67P, with the smaller lobe nearer to the other's shortest principal axis. This contrasts with most of the other modelled contact binaries, which tend to have their lobes aligned with the longest principal axis of the larger lobe. The data used to model RS11 are the same archival radar observations from the Goldstone DSN antenna and Arecibo collected during the asteroid's close approach in 2014 and used in the previous modelling. We combine these with optical lightcurves collected in 2014 and one lightcurve collected in 2023 with the Danish telescope at La Silla. The addition of the 2023 lightcurve allows for a better constraint on the target's rotational period and provides a different viewing geometry to help define the rotational pole solution.

We are also shape modelling asteroid 2024 ON using optical and radar data collected over a period of 3 months in late 2024, when it was discovered. This object is therefore a good analogue for possible planetary defence efforts, such as the recent concern over 2024 YR4, which demonstrated that potential impactors may only be known to us one or two apparitions before impact. The capability to quickly model the shape of an asteroid and determine estimates of additional properties like composition and bulk density is relevant to planning any planetary defence missions, such as the NASA DART mission. Despite the data for this target concentrated in one close approach of the target, the object was discovered before its closest approach, so a wide variety of viewing geometries could be collected in a short period. Some example radar images of this target are displayed in Fig. 1.

Fig 1. Radar observations collected on the 16th of September 2024 at the Goldstone DSN antenna of asteroid 2024 ON during its close approach.

Modelling

To create a radar shape model, we first use the optical data to constrain the spin state of the object. Using convex inversion [7,8], we can estimate the periods of 2000 RS11 and 2024 ON to be 4.4456 +/- 0.0005 h and 6.014 +/- 0.001 h, respectively (estimates of RS11’s period in 2014 were 4.444 +/- 0.001 h). With a strong initial period, we use the SHAPE modelling software to combine the optical and radar observations [9]. We created simple bi-ellipsoid models and performed fits at different fixed rotational pole solutions to investigate the rough location of the rotation pole before refining this with higher-resolution pole scans. Once the rotational pole is well defined, we create a vertex shape model for the best solutions to allow the model to replicate more complex features in the data. Preliminary results suggest that both objects have a pole solution in the southern hemisphere, as is common among near-Earth asteroids. With the addition of the optical lightcurves for 2000 RS11, the pole solution is likely closer to the southern pole of the ecliptic than previously thought. This emphasises the need for optical observations to be used to support radar modelling.

Conclusion

We are refining the shape model of RS11 and creating a new shape model for the recently discovered 2024 ON using a combination of optical and radar data. Using the SHAPE modelling software, we will model the spin state and morphology of both objects to investigate how they compare to previously modelled contact binaries and the implications they hold for their formation histories and the population as a whole.

Bibliography

[1] Virkki A. K., et al., 2022, http://dx.doi.org/10.3847/PSJ/ac8b72 The Planetary Science Journal, 3, 222

[2] Thirouin A., Sheppard S. S., 2018, http://dx.doi.org/10.3847/1538-3881/aac0ff The Astronomical Journal, 155, 248

[3] Brunini A., 2023, http://dx.doi.org/10.1093/mnrasl/slad076 Monthly Notices of the Royal Astronomical Society, 524, L45

[4] Levison H. F., et al., 2024, http://dx.doi.org/10.1038/s41586-024-07378-0 Nature,

629, 1015

[5] Cannon R. E., et al., 2025, http://dx.doi.org/10.1093/mnras/staf371 Monthly Notices of the Royal Astronomical Society

[6] Brauer K., et al., 2015. p. 213.03, https://ui.adsabs.harvard.edu/abs/2015DPS....4721303B

[7] Kaasalainen M., Torppa J., 2001, http://dx.doi.org/10.1006/icar.2001.6673 Icarus, 153, 24

[8] Kaasalainen M., Torppa J., Muinonen K., 2001, http://dx.doi.org/10.1006/icar.2001.6674 Icarus, 153, 37

[9] Magri C., et al., 2007, http://dx.doi.org/10.1016/j.icarus.2006.08.004 Icarus, 186, 152

How to cite: Cannon, R., Rożek, A., Snodgrass, C., Brauer, K., Busch, M. W., Pravec, P., Brozović, M., Benner, L. A. M., Hornoch, K., Kušnirák, P., Kučáková, H., and Fatka, P.: Shape modelling contact binaries 2024 ON and 2000 RS11 with ground-based radar, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1068, https://doi.org/10.5194/epsc-dps2025-1068, 2025.

We use multiple observations of NEAs to constrain thermophysical models, probe the thermal inertia, and determine sizes from albedos. We examine a variety of NEAs observed at the NASA Infrared Telescope Facility (IRTF) using SpeX [1] in the 0.8-5 micron range to better understand their structure and surface properties. When radar observations contribute to shape models, the thermophysical modeling can be more sophisticated and better utilized. However, even basic shape information such as obliquity constraints and rotation rate (i.e.fast or very slow) can improve the modeling results. We choose targets that are planned to be observed with radar, or that already have radar observations to serve as an independent measure of the size. The radar range depth is a direct measure of the asteroid’s extent in the line-of-sight at the time of observation. If NEOWISE observations are available, these can also be very helpful for individual objects, provided the images are carefully checked [2]. 

Shape models of NEAs show a remarkable variety of shapes. At diameters of 100m and larger, rubble-pile structures with spheroidal shapes are common (about 25%). However, simple thermal models like NEATM [3] that assume a spherical shape can give albedo best-fit values that do not match the asteroid’s diameter. There are many reasons that a simple model may miss important realistic characteristics, and the shape is not always the most important factor. Large solar phase angles, non-zero obliquity, and rough and irregular surface texture can also be important. At the same time, thermal observations of very slowly rotating asteroids, including those in non-principal axis (NPA) rotation states, often fit remarkably well using a spherical model even with very elongated or irregular shapes. Interpretation of the model results in cases where the size and shape are known independently from radar observations can be instructive. Figure 1 (top) shows thermal observations from the NASA IRTF in 2018 of (163899) 2003 SD220. This NEA was observed at both Arecibo Observatory and Goldstone [4]. A radar image is shown in Figure 1 (bottom). This object is in an NPA rotation state with apparent rotation of about 285 days. The solar phase angle was 80-84 degrees at the time of observation. The changing beaming parameter indicates that other effects such as self-shadowing or surface roughness are important to match the observations. We will present additional examples of NEA thermal modeling and explore the reasons for the model fit to be either better or worse than one might expect.

Figure 1. (top) Spectra and model fits for 2003 SD220 taken in 2018 on four different dates. Thermal inertia is assumed to be zero due to the very slow rotation period. A simple model fits well despite the very irregular shape. The radar image on the bpttpm was taken on 17 Dec 2018 in bistatic mode [4] transmitting from Goldstone and receiving at the Green Bank Telescope. This asteroid has an apparent size of at least 2.5km and shows ridges and surface concavities.

[1] Rayner, J. T., Toomey, D. W. Onaka, P. M. Denault, A. J. ; Stahlberger, W. E. ; Vacca, W. D. ; Cushing, M. C. ; Wang, S.(2003) PASP 115, Issue 805, pp. 362-382.

[2] Myers, S. A., Howell, E. S. Fernández, Yanga R. Marshall, Sean E. Magri, Christopher Vervack, Ronald J. Hinkle, Mary L., (2025) NEOWISE Data Processing and Color Corrections for Near-Earth Asteroid Observations, PSJ 6, Issue 4 id 80 12pp.

[3] Harris, A. W. (1998), A thermal model for near-Earth asteroids, Icarus 131, 291-30

[4] Rivera-Valentin, E. G., Taylor, P. A., Reddy, V.; Jao, J. S.; Benner, L. A. M.; Brozovic, M.; Naidu, S. P.; Virkki, A. K.; Marshall, S. E.; Sanchez, J. A.; Bonsall, A.; Seymour, A.; Ghigo, F. D.; Busch, M. W., (2019) “Radar and Near-Infrared Characterization of Near-Earth Asteroid (163899) 2003 SD220”, 50th LPSC LPI contrib. 2132 id3016.

How to cite: Howell, E., Vervack, R., Fernandez, Y., Myers, S., Hinkle, M., Magri, C., Marshall, S., and Rivera-Valentin, E.: Combining thermal and radar observations of near-Earth asteroids, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-462, https://doi.org/10.5194/epsc-dps2025-462, 2025.

The shape modeling of near-Earth asteroids (NEAs) has gained significant attention in recent years due to its critical role in understanding the physical properties and dynamical evolution of potentially hazardous objects. NEA (154244) 2002 KL6 is particularly intriguing because of its reported negative YORP effect [1], implying that its rotation is decelerating contrary to other asteroids which were observed to be under the YORP effect. In this study, we present an updated and more detailed shape model of 2002 KL6, derived from a combination of delay-Doppler radar imaging and optical light curves. Our optical dataset includes dense photometric observations spanning 16 years, primarily from TRAPPIST [2], supplemented by additional observatories. Optical observations are used with radar data by providing extended temporal coverage and a wider range of phase angles, both essential for accurate spin-axis determination and shape modeling. Using the DAMIT inversion code [3], we performed a period search in the interval 4.2 to 5.2 hours, yielding a best-fitting period of 4.6102 +/- 0.0003 hours. With this period, a pole search revealed a unique solution at (λ= 148.3°, β = -88.8°  ecliptic coordinates), in agreement to recently published results. Our extensive and dense dataset allowed us to reject the mirror solution for the pole. The derived pole orientation also implies a retrograde rotation. With the derived spin-state parameters, we performed a convex shape inversion[5], resulting in an elongated shape (Figure 1), consistent with the radar images obtained by the Goldstone Observatory (Figure 2).

Our analysis is supported by archival radar data from both the Arecibo Observatory (AO) and Goldstone Observatory (GO). AO observations were conducted between July 7 and 16, 2016, and GO observations occurred from July 17 to 25, 2016, and again from July 30 to August 18, 2023. The closest approach during these intervals was approximately 0.06 au. Using the SHAPE modeling software[4] and the convex inversion model as initial input, we will derive a refined 3D shape model and spin-state solution for the asteroid. The radar-based model will reveal surface features visible in the radar images, offering greater detail than achievable with light curves alone. This work emphasizes also the importance of integrating multiple observational techniques in advancing our understanding of NEA evolution and contributes to ongoing efforts in planetary defense.

[1] Rodríguez Rodríguez, Javier, et al. "Light-curve analysis and shape models of NEAs 7335, 7822, 154244, and 159402." Monthly Notices of the Royal Astronomical Society 533.4 (2024): 4160-4187.

[2] Jehin, Emmanuël, et al. "TRAPPIST: TRAnsiting planets and PlanetesImals small telescope." The Messenger 145.2 (2011).

[3] Ďurech, J., V. Sidorin, and Mikko Kaasalainen. "DAMIT: a database of asteroid models." Astronomy & Astrophysics 513 (2010): A46.

[4]  Magri, C., Ostro, S., Scheeres, D., et al.: Radar observations and a physical model of Asteroid 1580 Betulia, Icarus, vol. 186, issue 1, pp. 152-177, 2007.

[5] Kaasalainen, Mikko, and Johanna Torppa. "Optimization methods for asteroid lightcurve inversion: I. shape determination." Icarus 153.1 (2001): 24-36.

How to cite: Miftah, M. A., Ferrais, M., Jehin, E., Marshall, S. E., Jabiri, A., and Benkhaldoun, Z.: Combined shape modeling of Near-Earth asteroid  (154244) 2002 KL6 from Radar and optical observations, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1712, https://doi.org/10.5194/epsc-dps2025-1712, 2025.

Asteroids preserve critical evidence about early solar system processes, including accretion, thermal metamorphism, and impact history. This project focuses on asteroid (1036) Ganymed, a large S-type near-Earth asteroid (NEA), with the goal of generating a 3D shape model using radar observations and optical lightcurves, analyzed via the SHAPE software (Magri et al., 2007). As part of the 2025 Asteroid Radar Modeling Workshop, this work will serve as a year-long scientific and technical study of asteroid surface evolution. The specific science goal is to search for large-scale remnant impact structures preserved on Ganymed’s surface, which may record ancient collisional events and offer insight into the fragmentation history of S-type asteroid parent bodies.

Asteroid (1036) Ganymed is one of the largest known NEAs, with an effective diameter of ~35 kilometers and a rotation period of 10.31 hours. It follows an Amor-type orbit with perihelion near 1.24 au. Despite its current NEA classification, its size and dynamical history suggest that it likely spent most of its lifetime in the main asteroid belt, only more recently transitioning into a near-Earth orbit. Its spectral classification as an S-type (Bus-DeMeo taxonomy) indicates a silicate-rich composition dominated by olivine and pyroxene. S-types are widely believed to be the parent bodies of ordinary chondrite meteorites—the most common meteorites found on Earth (Dunn et al., 2010; McGraw et al., 2025). Studying surface morphology on such bodies helps constrain the geological context for meteorite samples and their shock histories.

This project uses radar data collected from Arecibo Observatory and the Goldstone Deep Space Communications Complex during Ganymed's 1998 and 2011 apparitions, as well as optical lightcurves from eight apparitions between 1985 and 2024. These data provide strong constraints on the asteroid’s shape, spin state, and surface roughness. Using the SHAPE software (Magri et al., 2007), we will iteratively refine the model through inversion of radar echoes and photometric data.

While radar imaging has revealed concavities on Ganymed’s surface at scales of ~10 km (Figure 1), it is difficult to resolve smaller features with confidence. One specific goal of this project is to test the resolution limits of our shape models by quantifying the smallest concavity sizes that appear uniquely and robustly across modeling iterations. By exploring how model features vary with input assumptions and data subsets, we aim to establish a practical threshold for interpreting large-scale impact structures.

Large impact structures can persist for billions of years on low-gravity bodies, although modified over time by regolith movement and secondary cratering (Bierhaus et al., 2005; Marchi et al., 2012). If Ganymed preserves such ancient features, it could provide a unique record of early main-belt collisional history. In this context, Ganymed complements broader research efforts aimed at linking asteroid surfaces to meteoritic samples and understanding the evolution of ordinary chondrite source bodies (Vernazza et al., 2014).

The modeling workflow includes processing radar data; investigating shape, spin, and scattering properties; and iteratively adjusting the model to minimize residuals. Surface concavities and basin-like structures will be examined both visually and via automated shape analysis tools, as used in prior studies (Benner et al., 2015). While we will assess candidate features, we emphasize that this model’s resolution limits what can be robustly interpreted as impact structures.

The result will be the best shape model of Ganymed produced to date, incorporating more radar and lightcurve data than any prior study. This project builds on previous shape modeling by Medina et al. (2023) and includes new observations and refinements. The resulting model will be used to examine large-scale surface morphology and investigate whether preserved craters can be linked to specific collisional epochs.

This abstract reflects ongoing collaborative work supported by the Asteroid Radar Modeling Workshop. The final model and crater analysis will inform future studies of asteroid surface evolution and meteorite analogs, while contributing to the broader planetary defense community’s understanding of large NEA structure.

References

Benner, L. A. M., et al. (2015). Radar observations and physical modeling of near-Earth asteroid (162421) 2000 ET70. Icarus, 245, 362–378.

Bierhaus, E. B., et al. (2005). Cratering on asteroids: Reconciling regolith processes and crater populations. Icarus, 175(2), 486–500.

Dunn, T. L., McCoy, T. J., Sunshine, J. M., & McSween, H. Y. (2010). A coordinated spectral, mineralogical, and compositional study of ordinary chondrites. Icarus, 208(2), 789–797.

Magri, C., et al. (2007). Radar observations and a physical model of asteroid 1580 Betulia. Icarus, 186(1), 152–177.

Marchi, S., et al. (2012). The violent collisional history of asteroid 4 Vesta. Science, 336(6082), 690–694.

McGraw, A. C., Reddy, V., & Sanchez, J. A. (2025). The Gefion Asteroid Family: Parent body puzzles and ordinary chondrite pieces. Monthly Notices of the Royal Astronomical Society, 537(4), 3145–3159.

Medina, V. A., Marshall, S. E., Devogèle, M., Taylor, P. A., Brozović, M., Ferrais, M., & Jehin, E. (2023). Shape modeling of 1036 Ganymed from radar and lightcurve data. EPSC-DPS Joint Meeting 2023.

Vernazza, P., et al. (2014). Multiple and fast: The accretion of ordinary chondrite parent bodies. The Astrophysical Journal, 791(2), 120.

Figure 1: Views of the current best-fit shape model of asteroid (1036) Ganymed along each principal axis, based on radar and lightcurve data. The model reveals large-scale surface concavities and asymmetries that may be remnant impact features. Scale bar = 50 km.

How to cite: McGraw, A., Marshall, S., Medina, V., Lipitz, M., Devogèle, M., Taylor, P., Brozović, M., Ferrais, M., and Jehin, E.: Shape Modeling of Asteroid (1036) Ganymed: Searching for Large-Scale Remnant Impact Structures on a Large S-Type NEA, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1154, https://doi.org/10.5194/epsc-dps2025-1154, 2025.

This work presents results from radar observations of the binary asteroid 2005 LW3, conducted in the wake of the ESA project “NEO Observation Concepts for Radar Systems”, which aimed at a future development of a European radar system for near-Earth objects (NEOs).

Thanks to collaboration with JPL/DSN, a continuous-wave (CW) radar signal at 7167 MHz was transmitted using the 70-m DSS-63 antenna in Madrid (Spain) as part of a multistatic radar experiment. The echo was received by the 32-m “Grueff” radio telescope in Medicina (Italy), operated by INAF – Istituto di Radioastronomia, and by the 100-m Effelsberg radio telescope (Germany) of the Max Planck Institute for Radio Astronomy.

The observations were performed on November 23, 2022, when the ~400-meter potentially hazardous asteroid (PHA) 2005 LW3 made a close approach at approximately 3.1 lunar distances (LD) from Earth. Both receiving antennas successfully detected the echo with a high signal-to-noise ratio and resolved it well in the frequency domain. Time-domain data were processed using a phase-stopping technique to correct for Doppler drift due to the target’s radial motion [1]. The resulting high-resolution power spectra allowed us to estimate the asteroid’s rotation period (~4 hours, assuming an equatorial aspect) and revealed a frequency offset of 1.0 ± 0.1 Hz from the predicted ephemeris, information that can be used to improve the orbital knowledge.

Fig 1:  Full-track integrated power spectra of radar echoes produced with Effelsberg (left) and Medicina (right) data, at resolution 0.1 Hz and 0.25 Hz, respectively. Zero frequency is the expected center of mass (COM) frequency of the asteroid. The spike at ~4 Hz is the echo from the asteroid’s satellite.

Delay-Doppler imaging independently acquired at Goldstone (NASA/JPL) revealed that 2005 LW3 is a binary system, identifying a companion approximately 50-100 m in diameter orbiting at ~4000 m from the primary body [2].

Fig 2: Delay-Doppler image of 2005 LW3 and its satellite obtained at Goldstone (courtesy of NASA/JPL).

Our CW Doppler spectra provide an independent confirmation of this binarity: the satellite was detected as a distinct secondary peak in the high-resolution spectra, demonstrating that fine structures can be revealed even in CW observations where delay information is unavailable. 

The Medicina telescope also detected echoes in both same-sense (SC) and opposite-sense (OC) circular polarizations, allowing us to estimate a circular polarization ratio between 0.1 and 0.2. This value, which is sensitive to surface and subsurface roughness at radar wavelengths, suggests a relatively smooth surface. Such polarization ratios are particularly important as they provide insight into the scattering properties and surface microstructure of the target [3].

Fig 3: Echo power spectrum of 2005 LW3 in the OC polarization (solid blue) and in the SC polarization (dotted red) derived from Medicina data. 

Beyond classical observables, we explored the feasibility of extracting shape and physical properties from Doppler spectra alone. We first implemented a method to estimate the convex hull of the asteroid, interpreted as the pole-on projection of its convex envelope, based on the technique described in [4]. After subtracting the satellite’s spectral contribution, the echo edges were identified as the frequencies at which the spectral profile crosses the zero-sigma level, corresponding to the statistical noise floor (zero-crossing criterion). 

The asteroid’s radial profile was then modeled with a truncated Fourier series under convexity constraints and fitted using weighted least squares to account for measurement uncertainty. This method assumes adequate rotational coverage, which in our case was about 88% of the asteroid’s spin period. The resulting shape dimensions are in good agreement with those reported in the literature, supporting the reliability of our estimation.

Fig 4: Pole-on projection of the 2005 LW3 convex envelope. The red cross marks the rotation center. 

The current convex hull estimation relies on the variation of edge frequencies with rotation, but disregards the full spectral shape, leaving much of the spectral information unused. To advance beyond edge-only analyses, we implemented an algorithm to estimate the third semiaxis of the asteroid - assuming a triaxial ellipsoid geometry - by fitting the full sequence of observed radar power spectra. The algorithm simulates Doppler spectra from a Lambertian scattering model over a rotating ellipsoid, given known equatorial semiaxes from convex hull analysis. We jointly estimate the polar radius and sub-radar latitude at observation by minimizing residuals between observed and simulated spectra via nonlinear least squares. This approach leverages the complete spectral echo profile, allowing for improved 3D shape constraints even with limited viewing geometries.

We extended the method by perturbing the reference ellipsoid using small-scale deformations and harmonic distortions. We plan to test various perturbation schemes and compare the results with available delay-Doppler reconstructions, which serve as ground truth in this context. If successful, the comparison would demonstrate good agreement in both spectral features and estimated dimensions, validating the approach as a useful tool for preliminary shape reconstruction.

These findings demonstrate that Doppler-only shape reconstruction is feasible and robust, especially when supported by convex hull constraints and physical scattering models. Further improvements could be achieved through better knowledge of sub-radar latitude and spin axis orientation. This could be accomplished, for example, by multi-apparition observations or by the inversion of joint radar-optical data.

Bibliography

[1]: Molera Calvés G. et al. (2014) A&A, 564, 1-7. 

[2]: Green D.W.E. (2022), IAU Circular No. 5198, 2022 Dec. 10.

[3]: Virkki A. and Muinonen K. (2016) Icarus, 269, 38-49.

[4]: Ostro S. et al. (1988) Icarus, 73, 15-24.

How to cite: Margheri, A., Pupillo, G., Righini, S., Orosei, R., Bach, U., Bortolotti, C., Kraus, A., Maccaferri, G., and Roma, M.: Advanced Doppler-Only Characterization of Binary Asteroid 2005 LW3: New Insights from the European Multistatic Radar Campaign, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1542, https://doi.org/10.5194/epsc-dps2025-1542, 2025.