2010 Research Report

Much of Ivanov's research in 2010 focused on MARSIS (Mars Advanced Radar for Subsurface and Ionosphere Sounding), a low frequency,pulse-limited radar sounder and altimeter used on the ESA Mars Express mission. It features groundpenetratingradar capabilities, which uses synthetic aperture techniques and a secondary receivingantenna to isolate subsurface reflections. The radar has been in operation since 2003 and has made many discoveries on on subsurface structure of Mars. This project is funded through NASA Mission of Opportunity.

MARSIS activity at PSI has been focused mostly on operational and data processing activities. Processing pipeline starts with level 1a data (instruments packets) and produces all the following products: raw formatted MARSIS data and observational geometry (level1b data), processed data (a.k.a compressed data, level 2 data) and corresponding browse products; geographically referenced browse products ( level 3 data, typical product is shown on Figure 1). Similar pipeline is also applied to the “flash data”, which is high resolution MARSIS data taken together with the main data stream. MARSIS obtained data on 761 orbits in 2009, which constitutes approximately 172Gb of processed data and browse products. The following tasks have been completed in 2009:

1. Upgrade of the level 2 processing algorithm. We now utilize topographic clutter model based on the MOLA DEM to improve quality of the data.

2. Phobos observations. Experiment planning, preparation and data analysis. [A B Ivanov and Thomas, 2010; Thomas et al., 2010]

3. Website maintenance. The website that serves data to the science team has been operational and is updated regularly with recent products.

The main source of noise in MARSIS subsurface data is ionospheric distortion, which is due to the fact that MARSIS frequencies are close to the ionospheric plasma frequency. One method to remove distortion is is a “contrast” method, which is a loop to estimate the value of the quadratic phase correction term of the signal Fourier spectrum. This algorithm optimizes the range compressed signal in terms of side lobes level, waveform shape, signal to noise ratio and range resolution, regardless any environmental behaviors. Another method is to modeling the total electron function (TEC) in the ionosphere. This method was described by [Safaeinili et al., 2003; Safaeinili et al., 2007] and [Mouginot et al., 2008] and it is currently employed for MARSIS data processing. In this paper we present an improvement to modeling TEC function. Current compensation scheme only uses Mars DEM [Smith et al., 1999] to compute distance between spacecraft and sub-spacecraft point, but it is not used as a constraint to compute TEC. In this work, we introduce a clutter model of Mars surface to be used as a constraint for TEC evaluation. Clutter model used in this work was developed by A. Safaeinili [Safaeinili et al., 2006]. Similar models were developed by [Nunes and Phillips, 2006], [Plettemeier et al., 2009].

The updated has now been implemented we are evaluating it on large sets of data. We have found  that it performs very well in places of interest. For example, echoes that come from polar layered deposits (PLD) terrains are now much better correlated with MOLA DEM. We also have a better results in other areas where observe subsurface signatures (e.g. Medussae Fossae Formation). Daytime observations, where signal-to-noise ratio is not big have improved considerably.

In February and March of 2010 Mars Express spacecraft has flown closer than 500km to Phobos, which is within MARSIS operational range. In total there were 6 fly-bys executed where MARSIS will be operational. This activity supported MARSIS operations near Phobos. It required planning of the upcoming fly-by, generation and validation of instrument specific timing coefficients. Currently data are being analyzed using topographic clutter model (Figure NN). Data come from different parts of

Phobos and can tell about surface and subsurface properties. Research activities have been focused on the understanding Phobos data from previous opportunities, polar area research and improving quality of data processing. A paper has been submitted to Science on the results of Phobos observations, but rejected by editors. A revisions is being prepared for JGR planets. Phobos observations suggest that there might be a secondary layer in some very restricted area on Phobos, but it is not ubiquitous. MARSIS has also allowed to improve on ephemeris of Phobos. 

MARSIS Radargram Radar echo simulation using Phobos topography

Reconstruction and analysis of Phobos surface model using MRO HIRISE Data There are two main hypothesis of Phobos formation: (1) this is a small body that was captured by Mars and (2) Phobos was ejected from Mars as a result of an impact. Both have their weak and strong sides. Surface of Phobos also has features, such as grooves, whose origin has been linked to catastrophic impacts (Asphaug and Melosh, 1993; Thomas et al., 1979) but detailed mechanism of their formation and properties of near surface regolith are still poorly understood. Grooves have also been discovered on other planetary bodies, like Eros (Prockter et al., 2002) and others. In this work we will focus on analysis of a HIRISE stereo pair and derive a high resolution (~20 m /pixel) digital elevation model. This will allow a fine analysis of crater depths and characterization of grooves’ profiles. We will aim at characterization of regolith properties on Phobos. Previously stereo and limb figure observations were done using Viking imagery (Duxbury, 1989; Gaskell, 2002; Tolson et al., 1978) which provided first estimates for the shape. Recently HRSC camera on Mars Express (Shingareva et al., 2008) was able to map Phobos and produce DEMs at 240 m/pixel. They have also studied properties of crater distribution and grooves.

In March of 2008 HIRISE camera took two images of Phobos: PSP_007769_9010 and PSP_007769_9015 (these images have been released to PDS prior to January 2009). They provide a very detailed look at an area of Phobos around 0 longitude. We have created anaglyph (shown in Fig. 2) as first step in the proof-of-concept. It allows visual analysis and see how images can be registered to each other. The next step was to manually pick a set of control points (similar to approach taken by (Duxbury and Callahan, 1989)) and then run them through our geometry software to find XYZ locations of intersections in Phobos reference frames. Results are shown in Figures 5 and 6. XYZ points were triangulated to obtain a simple shape of surface. Comparison with the current Phobos figure (Duxbury, 1989; Gaskell, 2002) showed that our approach is valid and we can now proceed with automated tiepoint generation. This work has been continued from the effort in 2009. Instead of manual feature matching we were able to employ automatic tiepoint generation and considerably improve resolution of the generated Digital Elevation Model. This approach resulted in a much better results, which allowed analysis of individual features on Phobos. Work continues to improve matching on Phobos and better calibration of the geometry.

Figure 5. Part of the original image (PSP_007769_9010, left panel) and derived DEM (right panel). The original image was selected as it contained most interesting parts of Phobos - craters, depressions and parallel grooves. The goal of this study was determine whether automated image correlation methods will be able to resolve features on the surface of Phobos. DEM panel shows the disparity intensity as a shaded relief map(darker area). Original image is blended in together with the shaded relief. The DEM does not cover the whole image due to complicated geometry of observations. Current process is not yet optimized for Phobos. Profiles across a crater (A-A’) and grooved terrain (B-B’ ) are shown in Figure 3 Figure 6. Profiles across features on Phobos derived from the preliminary digital model of elevation. In both panels horizontal axis shows distance across the profile in meters, assuming per pixel resolution of 6.8 m/pixel. Vertical axis shows detrended disparity (or feature displacement distance). Once full geometry is processed it will be possible to calculate a proper XYZ positions of feature points and create a true digital elevation model. Vertical scale and horizontal scales are the same for both panels. Top panel is a profile across a crater (line A-A’ in Figure 2). Bottom panel is a profile across parallel grooves (line B-B’ in Figure 2). These profiles prove feasibilty of constructing a high resolution DEM. Final resolution reported is approximately 13.6 m/pixel (twice the original resolution). Due to compilcated geometry of observations resolution will vary in final DEM .

Publications:

40th Lunar and Planetary Science Conference, (Lunar and Planetary Science XL), held March 23-27, 2000

in The Woodlands, Texas, id.1588.

Safaeinili, A., W. Kofman, J. F. Nouvel, A. Herique, and R. L. Jordan (2003), Impact of Mars ionosphere on orbital radar

sounder operation and data processing, Planetary and Space Science, 51(7-8), 505-515.

Safaeinili, A., Y. Gim, A. Ivanov, D. Plettemeier, J. Plaut, and G. Picardi (2006), Interpretation of MARSIS Radar Signal over the

Mars South Polar Layered Deposit, Fourth International Conference on Mars Polar Science and Exploration, October

2-6, 2006, Davos, Switzerland. LPI Contribution No. 1323, 8077.

Safaeinili, A., W. Kofman, J. Mouginot, Y. G. Gim, A. Herique, A. B. Ivanov, J. J. Plaut, and G. Picardi (2007), Estimation of the total

electron content of the Martian ionosphere using radar sounder surface echoes, Geophysical Research Letters, 34

(23), -.

Smith, D. E., et al. (1999), The Global Topography of Mars and Implications for Surface Evolution, Science, Vol. 284, Iss. 5419,

1495 (1999).

Thomas, N., R. Stelter, A. Ivanov, N. T. Bridges, K. E. Herkenhoff, and A. S. McEwen (2010), Spectral Heterogeneity on Phobos

and Deimos: HiRISE Observations and Comparisons to Mars Pathfinder Results, in Lunar and Planetary Institute

Science Conference Abstracts, edited, p. 2595.