MEX MARSIS Derived Data

Derived Data - MEX Optimized MARSIS Radargram Data

Instrument:Mars Advanced Radar for Subsurface and Ionosphere Sounding

PDS4 Bundle: urn:nasa:pds:mex_marsis_optim DOI: 10.17189/wrxv-m053

Data Provider's User Guide: Optimized MARSIS Radargram Data (U.S.) PDS Archive User’s Guide

The Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS; Jordan et al., 2009) is one of the instruments on board the ESA's Mars Express (MEX) spacecraft. The "Optimized MARSIS Radargram Data" archive bundle are the result of higher-level processing of MARSIS instrument data performed by U.S. members of the MARSIS Science Team at the Jet Propulsion Laboratory (JPL). The main feature of these higher-level products is the application of an optimization technique to remove the distorting effects of the ionosphere of Mars (McMichael et al., 2017), resulting in a geometrically consistent representation of the MARSIS radar sounding profiles, commonly referred to as "radargrams." The products in this bundle are intended to facilitate use of the data set both quantitatively and qualitatively; data are presented in binary format, rescaled to byte format, in png image format and as annotated browse pdfs. Each MARSIS observation also includes maps of topography along the spacecraft groundtrack and simulated radargrams that portray the expected position of topographic surface "clutter." The ionospheric correction results in a set of parameters that are also provided, including an estimate of the total electron content (TEC) of the column of ionosphere between the spacecraft and the neutral atmosphere. The archive bundle has been assembled to comply with PDS version 4 standards (PDS4). The collections searchable through ODE are listed in the table below. For more information about MARSIS optimized radargram data products, see the User’s Guide.

Example Listing of Archive Products

Type	Product_Type in ODE	Description	PDS4 Collection
radargram_data	RADAR** (RADARUN - Uncorrected, RADARCH - Chapman, RADAROP - Optimized)	Binary radargrams data and images. In ODE, binary radargrams data are categorized into three subsets: Uncorrected, Chapman, and Optimized, according to their ionospheric correction schemes, with the goal of improving search performance.	urn:nasa:pds:mex_marsis_optim:radargram_data
topography_data	TOPO	Native sampled and resampled topography data from the MOLA instrument for the MARSIS radargrams	urn:nasa:pds:mex_marsis_optim:topography_data
cluttergram	CLUTTER	Surface clutter simulations in IMG and PNG Format	urn:nasa:pds:mex_marsis_optim:cluttergram
ionosphere_data	IONOSPH	Derived ionospheric parameters for each radargram frame	urn:nasa:pds:mex_marsis_optim:ionosphere_data
geometry	GEOM	Geographic and geometric properties for each radargram frame	urn:nasa:pds:mex_marsis_optim:geometry

Radargram

Binary radargram is the most basic level product in the "radargram_data" collection of this bundle. Three different ionospheric correction scheme versions of the processed binary data are provided. The first, "uncorrected," applies only a standard radar processing algorithm, ignoring any possible distorting effects of the ionosphere. The second, "Chapman," uses a priori ionosphere parameters from the model of Chapman (1931) to correct for dispersion and delay. The third, "optimized," applies the optimization algorithm of McMichael et al. (2017). For each of the ionospheric correction schemes, two versions of the binary data are provided, "wind" and "no-wind." These refer to the presence or absence of a Hann windowing function (Harris, 1978) applied during processing ("wind" has windowing applied, "no-wind" does not). The effect of the Hann windowing is to suppress the range sidelobes, resulting in a less-noisy image, at the expense of vertical resolution.

To facilitate viewing and manipulation of data by the user, collection of radargram images are provided in simplified 8-bit (1 byte) per pixel image formats. The pixel values reported as real numbers (IEEE754LSBSingle) in the binary radargram are converted to the range of 0-255 (DN, digital number). Radargram images with the scaling are provided in byte array format and as png images. The PDS4 "img" file and its corresponding png file are described by a single xml label file. The png versions contain the same data as the img versions, but rather than appearing as co-registered layers, multiple images are stacked vertically in the png products. Single look images contain the 3 Doppler filters at each of the 2 channels, for a total of 6 radargram images per product. Multilook images contain 1 multilook image for each of the 2 channels, for a total of 2 radargram images per product.

MARSIS radargram images in their original ("native") format have a frame spacing that varies depending on the altitude and velocity of the spacecraft relative to the surface under observation. This bundle includes all radargrams in the native format, and versions that have been resampled in the along-track direction to an equal lateral pixel spacing of 500 m/pixel. A standard linear interpolation is performed to populate sample columns in the resampled radargram products. Pixel spacing in the vertical dimension (time delay or range/depth) is unchanged from the native radargrams in the resampled products. All multilook radargram image products are provided in both native and resampled formats; single look products are in native format only.

MARSIS radargrams are normally presented with the vertical dimension portraying time delay of the received echoes. In free space, the distance represented by an increment of one pixel in time delay is straightforward to calculate, as the wave speed is simply the speed of light. In a medium such as the subsurface of Mars, however, the wave speed will differ from that in free space, and thus the distance corresponding to the time increment will also differ. To compensate for a different wave speed in the subsurface, a slowing factor can be applied to pixels in the subsurface, based on the assumed refractive index of the medium. In this bundle, a simple (and strictly speaking, incorrect) depth correction assumption is made such that all pixels in the subsurface are placed as if the subsurface medium has the refractive index of water ice (real dielectric constant 3.1, refractive index 1.76). This assumption has been shown to be approximately correct for much of the polar terrains of Mars (Picardi et al., 2005; Plaut et al., 2007). The depth-corrected radargram filenames include the word "depth," while the nominal time delay radargram filenames do not. The vertical distance increment in the depth-corrected images is 26.767 m/pixel. The number of lines in the depth-corrected radargrams is 4196, twice the number in the standard delay radargrams, in order to retain the spatial resolution in the depth direction. If the vertical dimension of the delay radargram is expanded by a factor of 2, the delay and depth-corrected radargrams will co-register, with all pixels above the surface identical. Depth-corrected products are provided for multilook, optimized ionospheric correction radargrams only.

Topography Data

Analysis and interpretation of radar sounding data usually require knowledge of the surface topography of the observed area. Fortunately, a global data set of Martian Mars Orbiter Laser Altimeter topography (MOLA; Smith et al., 2001) is available to pair with MARSIS observations. The MOLA topography data associated with each MARSIS observation are provided in several forms in this bundle, in the collections "topography_data" and "topography_image."

The topography data are derived from the MOLA gridded radius products, as described in: https://pds-geosciences.wustl.edu/missions/mgs/megdr.html. The MOLA files used are Mars radius (megr*) at 128 pixels/degree, in cylindrical and polar projections. From these radius data, a standard Mars bi-axial ellipsoid is subtracted (ellipsoid radii: equatorial a=b=3396.19 km; polar c=3376.2 km). The pixel values are provided in the topography_data collection products as single precision floating point values (IEEE754LSBSingle; 4 bytes per pixel) representing the elevation relative to the ellipsoid. The MOLA topography swath along the groundtrack is provided in 2 projections, native and resampled. The native product provides one sample column per MARSIS frame, with 800 pixels at 500 m spacing across track (line direction). The nadir groundtrack lies along the center of this image, between lines 400 and 401. The resampled product is constructed at 500 m postings in both the along- and cross-track directions, reprojected from the original MOLA 128 pixels/degree gridded data. The resampled product thus aligns with the resampled radargrams in the radargram_image collection; the resampled radargram merged with the topography map is provided in the "resamp_map" image files in the radargram_image collection.

The topography_imagecollection includes strip maps and shaded relief. Both types of maps are provided in the native and resampled projections as described above for the binary topography data products. The strip maps report the MOLA elevation in meters relative to the ellipsoid as UnsignedLSB2, 16-bit integers. Conversion to MOLA ellipsoid-referenced elevation values can be accomplished using the scale and offset values provided in the xml label for the strip maps. The png file is also 16-bit format; both the img and png are described by a single xml label for that observation's strip map. The shaded relief maps portray the topography with a simulated illumination from the left of the image. They are provided in native and resampled projections, in 16-bit img and png formats. A horizontal black line indicates the position of the nadir groundtrack. The pixel values in the shaded relief maps should not be considered quantitatively meaningful. Some users may find the shaded relief maps' appearance can be improved with a simple contrast stretch.

Cluttergram Products

Cluttergrams derived from MOLA topography data along the ground track are provided for each MARSIS observation. The methodology for generating the cluttergrams is described in McMichael et al. (2017). The cluttergram portrays all echoes expected to come from surface topographic features, allowing the user to distinguish surface clutter echoes from candidate subsurface reflections. The cluttergrams are also used in the optimized ionospheric correction algorithm in the "optimized" products in this bundle (McMichael et al., 2017). Finally, the cluttergrams are used to identify the expected position of the first surface return echo, beyond which the correction factor is applied in the depth-corrected products. A single cluttergram is generated for each MARSIS observation; the clutter is considered to be independent of which MARSIS frequency band is used. The cluttergrams are provided in native and resampled projections, and in delay and depth-corrected vertical scaling. A depth-corrected cluttergram is useful for comparison to actual depth-corrected radargram data, as any clutter echo arriving later than the first return will be placed in a shallower position in both the cluttergram and radargram. The cluttergram images are provided in UnsignedByte (8-bit) img and png formats. They are designed to align precisely with the corresponding radargram image products processed using the optimized ionospheric correction scheme. The cluttergram products named "resamp_map" contain a merged product of the resampled MOLA strip map and the cluttergram, in 8-bit format img and png formats. The brightness of a cluttergram pixel is not intended to represent an actual calibrated MARSIS backscatter measurement, but simply a prediction of the position of the clutter echo and a simulation of its relative brightness.

Ionosphere products

The optimization algorithm of McMichael et al. (2017) is primarily intended to correct geometric and smearing effects of the ionosphere on the MARSIS subsurface sounding mode data. A byproduct of the ionospheric correction is a set of parameters that can be used to characterize the ionosphere itself. The derived ionospheric parameters are reported in the collection "ionosphere_data." The data are reported in an ASCII table that contains 1 row per frame and 17 columns of data, as described in the xml label files. Key parameters reported for each frame are the MARSIS frequency band, the solar zenith angle (SZA), and the time delay and 3 phase coefficients for each MARSIS channel (see Table 2 in the User’s Guide). These tables are formatted as comma separated values (csv) and can be opened directly in spreadsheet applications or text readers.

The collection "ionosphere_browse" contains pdf format browse products of plots of several of the ionosphere parameters. The products with "coefficient" in the filename contain plots of the 3 phase coefficients as a function of the observation frame number, with a second horizontal axis showing the SZA. The products with "delay" in the filename contain plots of the derived ionospheric delay in microseconds, again as a function of frame number, with SZA also shown.

Geometry Products

Ancillary data are provided for every frame of every observation in the "geometry" collection. The fields reported in these ASCII table files are shown in Table 3 of Section 2.7 in the User’s Guide. The tables are formatted as comma separated values (csv) and can be opened directly in spreadsheet applications or text readers. The column headings are provided in an accompanying file in the top level of the geometry collection. Useful parameters reported in these tables include the latitude and longitude of each frame's footprint, the MARSIS frequency bands used, the spacecraft altitude, and the solar zenith angle at the surface footprint.

Note: Please refer to Section 3 and Section 4 in the User’s Guide for more information on how to utilize the data within this bundle.

More information can be found in:

The User’s Guide.for the MEX Optimized MARSIS Radargram Data (U.S.) PDS Archive
Chapman, S. (1931). The absorption and dissociative or ionizing effect of monochromatic radiation in an atmosphere on a rotating earth. Proceedings of the Physical Society, 43, 26, https://doi.org/10.1088/0959-5309/43/1/305.
Harris, F. (1978). On the use of windows for harmonic analysis with the discrete Fourier transform. Proceedings of the IEEE, 66, 51-83, https://doi.org/10.1109/PROC.1978.10837.
Jordan, R., Picardi, G. Plaut, J., Wheeler, K., Kirchner, D., Safaeinili, A., Johnson, W., Seu, R., Calabrese, D., Zampolini, E., Cicchetti, A., Huff, R., Gurnett, D., Ivanov, A., Kofman, W., Orosei, R., Thompson, T., Edenhofer, P., Bombaci, O. (2009). The Mars express MARSIS sounder instrument. Planetary and Space Science, 57(14-15), 1975-1986, https://doi.org/10.1016/j.pss.2009.09.016.
McMichael, J., Gim, Y., Arumugam, D., Plaut, J. (2017). Radar autofocus algorithm incorporating terrain knowledge for correction of Mars’ ionospheric distortion in MARSIS observations. 2017 IEEE Radar Conference (RadarConf), 0873–0878, https://doi.org/10.1109/RADAR.2017.7944326.
Picardi, G., Plaut, J., Biccari, D., Bombaci, O., Calabrese, D., Cartacci, M., Cicchetti, A., Clifford, S., Edenhofer, P., Farrell, W., Federico, C., Frigeri, A., Gurnett, D., Hagfors, T., Heggy, E., Herique, A., Huff, R., Ivanov, A., Johnson, W., et al. (2005). Radar Soundings of the Subsurface of Mars. Science, 310, 1925-1928, https://doi.org/10.1126/science.1122165.
Plaut, J., Picardi, G., Safaeinili, A., Ivanov, A., Milkovich, S., Cicchetti, A., Kofman, W., Mouginot, J., Farrell, W., Phillips, R., Clifford, S., Frigeri, A., Orosei, R., Federico, C., Williams, I., Gurnett, D., Nielsen, E., Hagfors, T., Heggy, E., et al. (2007). Subsurface radar sounding of the south polar layered deposits of Mars. Science, 316(5821), 92-95, https://doi.org/10.1126/science.1139672.
Smith, D., Zuber, M., Frey, H., Garvin, J., Head, J, Muhleman, D, Pettengill, G., Phillips, R., Solomon, S., Jay Zwally, H., Bruce Banerdt, W., Duxbury, T., Golombek, M., Lemoine, F., Neumann, G., Rowlands, D., Aharonson, O., Ford, P., Ivanov, A., et al. (2001). Mars Orbiter Laser Altimeter: Experiment summary after the first year of global mapping of Mars. Journal of Geophysical Research, 106(E10), 23689-23722, https://doi.org/10.1029/2000je001364.