Spectroscopy on Mars: A Look at What’s Been Uncovered About the Red Planet

Raymond Arvidson

Spectroscopy has played a significant role in the Mars expeditions, including the confirmation of the former presence of water on the Red Planet. Raymond Arvidson, the James S. McDonnell Distinguished University Professor at Washington University in Saint Louis, Missouri, is involved with the various National Aeronautics and Space Administration (NASA) missions to Mars and the spectroscopy incorporated in the instruments sent there. Here, Arvidson discusses those techniques, including a hyperspectral imaging system, an emission spectrometer, and an X-ray spectrometer, and what the results of the missions indicate about Mars so far. Arvidson is the winner of the 2016 Lester W. Strock award presented by FACSS. This interview is part of the series with the winners of awards to be presented at the SciX 2016 conference.

What kind of spectroscopic instrument capability do the Mars Reconnaissance Orbiter and Opportunity rover have installed? Can you tell us more about the Compact Imaging Spectrometer for Mars (CRISM) system and how it was developed?

The Mars Reconnaissance Orbiter has several instruments that operate over a range of wavelengths, including multispectral imaging systems for surface and atmospheric monitoring and characterization. CRISM is a hyperspectral imaging system, operating from 3.62 to 3.92 µm with 6.55-nm/channel spacing.  It was developed at the Johns Hopkins University Applied Physics Laboratory. CRISM is used to map both the surface and atmosphere of Mars. The best spatial resolution mapped onto the surface is ~12 m/pixel using along-track oversampled observations. Analysis of data collected from the CRISM instrument has revolutionized our understanding of Mars, with, for example, identification and mapping of phyllosilicate, hydrated sulfate, carbonate, and opaline minerals formed in the presence of water.

Opportunity landed in January 2004 and has far exceeded its mission design lifetime. This is testimony to a well-made American vehicle. The mast-based science instruments include a multispectral stereo camera system called Pancam, operating from 0.4 to 1.0 µm with 13 bands. An emission spectrometer called Mini-TES is also on the mast, although this instrument ceased operating some time ago. The arm-based instruments include a Mössbauer spectrometer with a cobalt-57 radioactive source that has been through many half-lives and thus the instrument is no longer used. The key arm-based spectroscopic instrument is the alpha particle X-ray spectrometer (APXS), equipped with a curium-244 radioactive source. When placed onto the surface, alpha particles and X-rays from the source result in X-ray emission from the soil or rock target and thus determination of elemental composition.

In a recent paper (1), you discussed the data gathered from the Mars Exploration Rover Opportunity’s trip to the Noachian age Endeavour crater on Mars where ancient aqueous environments were discovered. What spectroscopic techniques were used to gather these data, how were they analyzed, and what did the results show? How long does it take for the data gathered by Opportunity to reach the scientists that analyze them?

After spending ~8 1/2 years investigating the sulfate-rich sandstones on Meridiani Planum, a plain near the equator on Mars, Opportunity was commanded to traverse onto the Cape York rim segment of the ancient ~22-km-wide Endeavour Crater.  Pancam and APXS were the workhouse instruments that produced data to allow us to infer the presence of phyllosilicates, sulfates, and hematite, and manganese oxides within the rocks exposed on Endeavour’s rim segments.  These minerals definitely formed in presence of water, likely hydrothermal systems generated by the heat of the impact that formed Endeavour.

Reduced data in the form of spectral reflectance image cubes for Pancam and oxide compositions for APXS data are produced within a few days for science team use. After a validation period the raw and reduced data are deposited in the NASA Planetary Data System for community use.  Raw images are released to the community via the web as soon as they are generated from telemetry data.

Have those results had an impact on future plans to explore different areas on Mars?

The more we look on Mars, combining orbital spectroscopy and imaging with what the Spirit, Opportunity, and Curiosity rovers have discovered, the more evidence we find that indicate early in geologic time the Red Planet had extensive surface water such as rivers and lakes, and extensive groundwater systems. The Endeavour Crater results fall into the category of groundwater systems that moved preferentially along fractures, altering rocks and depositing aqueous minerals. Results show that there is a large range of possible targets for future missions focused on defining, in detail, the habitability of a variety of aqueous environments, and whether or not life existed or exists on the Red Planet.

In other recent work (2), you present an analysis of data from three CRISM full resolution targeted (FRT) and two along-track oversampled (ATO) observations over the Cape Tribulation Endeavor rim segment that have been processed from spectral radiance to surface single scattering albedo (SSA) spectra using first-principles radiative transfer procedures, and regularized using a log maximum likelihood approach. Can you briefly describe the sampling techniques used in this study? How does this approach to analyzing the data differ from previous approaches? What were you able to uncover about this data?

Our work is the first end-to-end retrieval of surface reflectances in which we explicitly modeled the absorption and scattering of light due to gases and aerosols in the atmosphere, and retrieved single scattering albedo surface spectra. These spectra are independent of lighting and viewing conditions and only depend on the complex index of refraction, grain sizes, and packing of surface materials. Further, the work is the first use of maximum likelihood processing procedures for this type of hyperspectral imaging data, regularizing the data to allow mapping to 12 m/pixel, even though an individual CRISM pixel projected onto the surface is 18 m/pixel. The increased spatial and spectral fidelity of the products allowed identification and mapping of phyllosilicate minerals in Marathon Valley, and directing Opportunity to specific localities to investigate the composition of geologic settings of these phyllosilicate-bearing deposits.

What do the various discoveries on Mars indicate so far?

Mars was warm and wet early in geologic time, when it still had a magnetic field to shield the solar wind from stripping away the upper atmosphere, and volcanoes were active and releasing greenhouse gases into the atmosphere. The planet was likely a habitable place, although it remains to be seen if life started and evolved during that early time period.

Are there any other analytical techniques that you wish you could apply to the data on Mars, and are there are any new developments that would be needed to accomplish it?

It would be nice to have a mini-CRISM on a rover to complement what CRISM is able to detect and map from orbit. Perhaps on a future mission.

The 2020 Mars rover mission is designed to acquire rocks cores and store them on the surface for a later mission to gather them and return them to Earth. Spectral analysis of these returned cores will undoubtedly again revolutionize our understanding of the Red Planet and move us far along to an understanding of whether or not Mars supported life.

What are the next steps in your research?

Lots to do with Opportunity mission planning and data analysis, with an intent to leave Marathon Valley and head to the next rim segment, Cape Byron, where orbital imaging shows a large gully system that may have formed due to surface waters. We intend to drive down the gully to test various hypotheses about how it formed: dry avalanche, debris flow lubricated by water, or fluvial processes. And then there is the Curiosity rover and its discovery in Gale Crater of ancient fluvial-lake deposits. And let’s not forget plenty of CRISM coverage and associated data products not yet reduced and analyzed!

References

•  R.E. Arvidson et al., Science343, DOI: 10.1126/science.1248097 (2014).

• V.K. Fox, R.E. Arvidson, E.A. Guinness, S.M. McLennan, J.G. Catalano, S.L. Murchie, and K.E. Powell, Geophys.  Res. Lett. DOI: 10.1002/2016GL069108 (2016).