The discovery of Recurring Slope Lineae (RSL) suggests that drainage is occurring seasonally today (McEwen, et al. 2014). Within the RSL, observed changes in hydrated perchlorates provide support for flowing brines as the RSL source (Ojha, et al. 2015). Called Planum Boreum in the north and Planum Australe in the south, the polar caps of Mars are vast domes composed mainly of water ice sheets (Grima, et al. 2009) that extend 3-‐4 km above their surrounding terrains.
SHARAD returns from the base of layered sediments in Planum Boreum have sufficient power to constrain the amount of lithic fraction through the stack to less than about 5% (Grima, et al. 2009). Over time, as the planet rotated below (the rotation period is 24 hours and 37 minutes), MRO's near-polar orbit resulted in relatively dense coverage (compared to the rest of the planet) of the polar regions by SHARAD. The result of these two projects were the first high-resolution 3D images of the interior of Mars.
SHARAD Pre-‐processing
After years of mapping and analysis using SHARAD datasets from individual orbit passes, it became clear that the structural complexity of the polar ice caps was sufficient to merit 3-D processing of the SHARAD data. This process is performed in the frequency domain, after which the redundant half of the spectrum is set to zero, and the result is inversely Fourier transformed to obtain. Doppler shift correction: frequency and location dependent along-track phase rotation to correct for Doppler shift caused by the relative motion of the spacecraft and the center of the synthetic aperture.
Corrections applied to each frame via a polynomial adjustment of the derived differential time delays relative to the solar zenith angle (SZA) for each frame. The aggregate analysis revealed such residual relative ionospheric time delays that had not previously been assessed in individual radargrams, providing valuable feedback that benefited the SHARAD preprocessing workflow. Two typical radargrams over Planum Boreum (top) and Planum Australe (bottom) resulting from the SHARAD preprocessing workflow are shown in Figure 2.
3-‐D Processing and Analysis of SHARAD Data
This is only a partial migration in a 3-D structural environment, and since all terrestrial tracks above the poles have a full range of. For the Planum Boreum data, we simply used the radar return power, while for the Planum Australe data, we took the square root of the power value samples to convert them to the so-called reflectivity (Taner, Koehler and Sheriff 1979). so the lower SNR is better accommodated by detrending each frame before 3-D post-processing. Mass offset: remove most of the travel time on first returns for storage efficiency by applying a constant negative time offset and trimming each frame proportionally.
But for the large size of the grids and bin spacings, collecting 3-D polar volumes of Mars was routine by terrestrial seismic processing standards. Because of the latitude-dependent sparsity and irregularity of the SHARAD coverage, it was necessary to interpolate and smooth the 3-D forward volumes. Because the vast majority of the travel time occurs above the surface, the RMS velocity is almost unchanged by propagation through the ice caps even though increasing the dielectric constant within the ice caps above that of free space decreases their range velocities significantly.
The downward continuation of the entire Planum Boreum volume was performed piecemeal (due to software licensing and computer resource constraints), with one 237.5 km x 237.5 km section overlapping the coverage gap centered around the pole shown in Figure 7, before and after the downward continuation. As expected, the downward continuation has focused the structural features in the volume in a manner consistent with moving the acquisition datum from Earth orbit to within a few kilometers of the polar surface. On the scale shown, there are no discernible differences between the two results, indicating that virtually all focusing occurs with downward continuation.
The structural focusing across the surface of the ice sheet is striking, the deep troughs in the ice clearly defined. These images also show the significant impact that this surface topography has on the deeper time picture, essentially mirroring the topography of the ice sheet's interior and basal units. As a final note, it is evident that the downward continuation operator edge effects are more pronounced within the signal regions of the Planum Australe volume.
With the ongoing 3-D drop volume completed, the final step in the 3-D processing workflow is the 3-D Stolt migration, pending as of this writing.
3-‐D Interpretation Results
In particular, the 3D volume (Smith, Putzig, et al. 2016) helped fully map the extent of a surface discordance that they associated with a recent climate change (see Introduction). Partially buried impact craters show a distinctive signature that is repeated elsewhere in the volume, but without surface expression. For Earth's moon, crater statistics have been linked to radiometrically dated samples returned by the Apollo program, and the derived crater rates have been extrapolated and used to date surfaces on Mars and other solar system bodies.
Therefore, if a statistically significant number of putatively buried craters can be proven to be true impact craters, they will provide an important measure of the age of glacial stratified deposits that is independent of climate models. Despite the detection and removal of residual ionospheric time delays in the data, there remain some residual time delays in the data that cannot be identified as variations in ionospheric phase distortion as they are not correlated with the solar zenith angle or the calculated modeled ionospheric phase distortion parameter in the region. In particular, the finer shallow layering and structures evident in the input 2-D observations are not resolved in the 3-D volume, hampering the ability to (1) improve upon previous efforts to correlate radar layering with that seen in visible images, e.g. Christian, et al for mapping small troughs and waves at the surface associated with climate signals, e.g.
A renewed effort by members of this team will further reduce the remaining delays, and SHARAD data collection will continue in the fourth extended mission of the MRO, with a goal of completing coverage in areas targeted for 3D processing. As a final note, a comprehensive project report was written documenting the details of the entire Planum Boreum 3-D processing project, including discussion of tested methods for signal amplification and noise reduction and sources of error in the processed datasets.
Conclusions and Next Steps
In addition, coverage is still incomplete at the 475-‐m bin size, further limiting the resolution of finer features. The SHARAD 3-‐D volumes led to a better understanding of the effects of the ionosphere on the 3-‐D consistency of the data, thereby leading to improved modeling of the ionosphere. Moreover, through proper 3D imaging, the SHARAD 3D volumes provided structural clarity in the data, thereby improving its interpretability and usefulness for conducting scientific research.
As of this writing, the first properly imaged SHARAD 3D section of Planum Boreum has been completed, with the first section of Planum Australe nearing completion and to be followed by other sections imaging smaller targets at lower latitudes. Efforts are underway to archive the SHARAD 3D processing methodology and resulting 3D data volumes in NASA's Planetary Data System, making them available to the broader scientific community. Additionally, because of the diagnostic power of 3D data analysis, an effort is being made to detect and sift out persistent ionospheric time delays and other timing inconsistencies between the individual radargrams. If this is successful, it will improve the resolution and therefore the interpretability of the SHARAD 3D volumes.
Finally, through iterative sensitivity analysis and refinement of the 3-D image provided by the SHARAD volumes, we hope to extract velocity and relevant dielectric constant information for the polar ice caps directly from the SHARAD data, further exploiting its content of information about science. discovery.
Acknowledgments
34;SHARAD Signal Attenuation and Delay Shifts Due to Martian Ionosphere." IEEE Geoscience and Remote Sensing Letters, July. 34;HiRISE Observations of New Impact Craters Exposing Martian Ground Ice." Journal of Geophysical Research 119, no. 34; Origin and Characteristics of the Martian North Polar Basal Unit and Implications for Polar Geological History." Icarus 174, no.
34;Vertical distribution of dust in the Martian atmosphere during the northern spring and summer: High-altitude tropical dust maximum at northern summer solstice." Journal of Geophysical Research 116, no. 34;A revised surface age for the North Polar Layered Deposits of Mars: A REVISED SURFACE AGE OF MARS NPLD." Geophysical Research Letters 43, No. 34; Martian South Polar Layered Deposit Stratigraphy and Implications for Accumulation History.” Journal of Geophysical Research 113, no.
34;Massive CO2 Ice Deposits Sequestered in the South Polar Layered Deposits of Mars." Science, No. 34;MARSIS: Mars Advanced Radar for Subsurface and Ionospheric Sounding." Special Publication -‐ Mars Express: A European Mission to the Red Planet (European Space Agency), No. 34; An Inventory of Impact Craters on Mars' South Polar Layered Deposits." Lunar and Planetary Science XXXVI, Abs.
34;Radar evidence for ice in lobed debris aprons in the mid-northern latitudes of Mars." Geophysical Research Letters 36, no. 34;3-D radar images of structures and craters within the Martian north polar mantle." Earth and Space Science., in prep. 34; Mars Orbiter Laser Altimeter: Experiment Summary After the First Year of Global Mapping of Mars." Journal of Geophysical Research 106, no.
Note also that the faint "cross-hatching" in the time slices, made especially visible by the color scheme used for display (red is low intensity), is due to truncation artifacts from the adjacent downward continuous slices. Note also that the faint "cross-hatching" in the time slices, made especially visible by the color scheme used for display (reds are low power), is due to truncation artifacts from the adjacent down-direction. Note that the faint “cross-hatching” visible in the time slice is due to lingering truncation artifacts from the adjacent descending continuum pieces (see text for further discussion).
