The Mars Orbiter for Resources, Ices, and Environments (MORIE) Science Goals and Instrument Trades in Radar, Imaging, and Spectroscopy

Several major discoveries have provided new motivation for a next-generation Mars orbiter (Figure 1). Shallow Radar (SHARAD) and Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS) radar sounder detections have provided evidence for regional-scale buried ice sheets in the midlatitudes (Bramson et al. 2015; Stuurman et al. 2016). Steep scarps expose ice in midlatitudes (Figure 1), and ice exposures in hundreds of new impact craters and erosional scarps have been detected at latitudes as close to the equator as 39°N (Byrne et al. 2009; Daubar et al. 2013; Dundas et al. 2014, 2018; Figure 1). For landforms known as lobate debris aprons, recent radar and geomorphic analysis have shown them to be composed predominantly of water ice and thus constitute debris-covered glaciers (Holt et al. 2008; Plaut et al. 2009; Petersen et al. 2018). For both glacial and nonglacial midlatitude pieces of ice, the depths of burial remains poorly constrained, as the existing radar sounders have a relatively coarse vertical resolution (∼8–80 m), and thermal inertia and impacts only identify ice in the upper few meters (e.g., Piqueux et al. 2019). Also unconstrained is the global nature of subsurface ice, whether as pore filling, massive laterally contiguous deposits, or small, disconnected ice lenses (e.g., Mellon & Jakosky 1995; Mellon et al. 2009; Sizemore et al. 2015; Bramson et al. 2015). Recent work suggests that the presence of massive ice, rather than simply pore-filling ice, challenges current thermal models that predict any near-surface ice should be geologically young and actively retreating in the current climate (Bramson et al. 2017). The composition and origin of the Medusae Fossae Formation (MFF) are also contested, with different approaches offering contradictory interpretations as to whether the unit is a friable ash deposit or if it may offer an equatorial deposit rich in water ice, or possibly both (e.g., Bradley et al. 2002; Watters et al. 2007; Mandt et al. 2008; Carter et al. 2009; Kerber et al. 2012; Campbell & Morgan 2018; Wilson et al. 2018; Ojha & Lewis 2018; Campbell et al. 2021). In the north polar layered deposits (PLD), a sequence of recent layers has been identified and suggested to be related to obliquity variations over the last 370 ka (Smith et al. 2016). These new results have clarified the next questions to be studied in Martian ice science as noted by ICE-SAG and the Keck workshop (ICE-SAG 2019; Smith et al. 2020; Table 1), and significant advances are expected from a follow-on orbiter such as MORIE. Understanding the climate record retained in polar and ground-ice reservoirs requires better observational constraints on the lateral extent and volume of ice, as well as the depth to the top of the ice table at low latitudes. Quantifying the depth and extent of ground-ice deposits can constrain formation mechanisms and address fundamental questions such as what is the total water-ice reservoir? Is the ice mostly pure or in regolith pore space, massive sheets, or distributed lenses? What is the depth and thickness as a function of latitude? Is there water underneath or melt within the ice sheet or at the poles? What is the history of climatic variations recorded in the PLD, and how do those climatic variations relate to obliquity variations?

Zoom In Zoom Out Reset image size

The success of orbital and landed missions has led to a new understanding of the variety of aqueous environments on Mars as evidenced through the diversity of minerals detected at the surface (e.g., Squyres et al. 2004; Bibring et al. 2006; Squyres et al. 2008; Murchie et al. 2009; Boynton et al. 2009; Ehlmann & Edwards 2014). Orbit-based mineral detections were initially interpreted as indicating persistent wetter climates with greater water availability during the oldest epochs (Noachian), which produced phyllosilicates, followed by episodically wet conditions in later periods (Hesperian) conducive to the production of sulfate salts (e.g., Bibring et al. 2006). This broad outline is correct; however, higher-resolution orbital data and recent in situ missions have shown that the history of major environmental transitions on Mars was much more complex, with over a dozen types of aqueous environments varying in space and time (e.g., Murchie et al. 2009; McLennan et al. 2019).

An enduring mystery is the nature of the ancient Martian climate, what the environmental conditions were, how stable, and for how long, and the atmospheric conditions that lead to available liquid water early in Mars' history. In Noachian terrains, mineralogy and geomorphology indicate surface and subsurface aqueous activity with varying proposals for their relative prevalence (Hynek et al. 2010; Ehlmann et al. 2011; Carter et al. 2015; Bishop et al. 2018). Noachian paleoclimate models range from cold with uncommon liquid water (e.g., Fairen 2010; Wordsworth et al. 2013; Kite et al. 2013; Cassanelli & Head 2015; Palumbo & Head 2018), to warm periods ∼10⁷ yr or longer (e.g., Ramirez & Craddock 2018; Ramirez et al. 2020), depending on factors that are poorly constrained such as atmospheric composition and water inventory, and the response of the climate system to obliquity change. Confounding the simple story of a transition from a clay-forming to a sulfate-forming epoch is the appearance of sulfate units within the clay stratigraphy of Mawrth Vallis and in a complex set of paleolakes in Terra Sirenum (Bishop et al. 2020; Wray et al. 2011). This suggests periodic wetting and drying rather than exclusively wet then dry conditions.

On the ground, results from the Mars Science Laboratory (MSL) have demonstrated that a lake may have existed perennially or episodically in Gale Crater for up to 10⁵–10⁷ yr during the Hesperian period, with even longer-lasting groundwaters (e.g., Grotzinger et al. 2014, 2015). Clay, sulfate, and hydroxide minerals in the sedimentary rock are attributed to in situ weathering, authigenic precipitation, and diagenesis (e.g., McLennan et al. 2014; Bristow et al. 2018; Mangold et al. 2019; Rampe et al. 2020). A progression from Fe(II) to Fe(III) oxides may signal changing atmospheric conditions, lake depth, or diagenetic conditions (Hurowitz et al. 2017; Rampe et al. 2017). Higher in the stratigraphy, mineral phases may indicate lakes alternating with drying episodes where a lake was absent or reduced in extent (Bristow et al. 2018; Rapin et al. 2019; Thomas et al. 2019; Rampe et al. 2020). However, these mineral records of surface conditions are also complicated by the long-lived and varied history of diagenetic alteration, which is difficult to discern from orbital data alone (e.g., Fraeman et al. 2020). Thus, the Gale crater and other sedimentary records on Mars preserve a complex history of interactions between deposition, alteration, and erosion extending across billions of years (McLennan et al. 2019). Understanding whether these transitions reflect local conditions unique to the Gale crater region or global climate change requires investigation in situ at other landing sites as well as improved orbital infrared spectroscopy of the Martian rock record with sufficient spatial resolution for meter-scale analysis of stratigraphic relationships that record change.

The nature of Mars' earliest aqueous environments and when the planet transitions to drier conditions on regional and global scales are still unclear. How these transitions may have affected the habitability of surface and subsurface environments remains to be unraveled. Unlocking the geologic sequence stratigraphy and these transitions through time is crucial to interpreting the changing record at locations not yet visited by rovers, such as large water-related deposits in Valles Marineris (Figure 2), ancient stratigraphies of clay and sulfate in Noachian terrains, deep lake and inland basins at Terra Sirenum and Eridania, and the large connected system spanning the Uzboi, Holden, and Ladon basins. Filling these knowledge gaps demands more advanced mineral detection techniques and better stratigraphic reconstructions of the sedimentary and geochemical record on Mars, both of which are achievable with a next-generation orbiter. Key measurements include the ability to identify, spatially map, and quantify at a few-meters scale the water-related minerals: oxides, phyllosilicates, and salts that constrain water chemistry (salinity, alkalinity, pH, redox state, persistence), and the ability to characterize igneous mineral and glass assemblages at a few to tens of meters scale to understand the prevalence of alteration and how changing volcanism and impacts influenced climate. Such measurements would enable us to resolve environmental transitions occurring over more than a billion years at a time resolution of millions of years, and to correlate these changes across the planet.

Zoom In Zoom Out Reset image size

Both shallow ground-ice and hydrated mineral deposits are of interest to NASA as potential resources for human missions (e.g., NEX-SAG 2015; MEPAG 2020). Recent work by Piqueux et al. (2019) used data from the Mars Climate Sounder (MCS) and Thermal Emission Imaging System (THEMIS) to derive the depth to the water-ice table within 1 m of the surface. This result and those of the project "Subsurface Water Ice Mapping (SWIM) on Mars," (Putzig et al. 2020) indicate a greater extent of near-surface water ice at lower latitudes than previously mapped. Regional studies and the SWIM project have leveraged multiple data sets in an effort to span the gap between shallow and deep ice detections, but the 5–20 m depth range effectively remains a blind spot for current orbital assets.

Two projects are underway to create global maps of hydrated minerals with data from the Observatoire pour la Minéralogie, l'Eau, les Glaces et l'Activité (OMEGA) and the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM; Riu et al. 2019; Seelos et al. 2019). However, these global maps are at coarse spatial scales and only able to map two-dimensional surface exposures in regions where dust coverage is low. To understand the potential for hydrated minerals as a future resource, both higher spatial resolution spectroscopy and the ability to link subsurface deposits to surface exposures through radar will be important. New orbital data would significantly enhance future science investigations as well as the resource mapping that requires the same information.

Radar sounders have been operating at Mars for over a decade, with MARSIS on Mars Express (Jordan et al. 2009) and SHARAD on MRO (Seu et al. 2007). They have proven to be invaluable assets for probing the subsurface to depths of over 3 km in the polar deposits (Plaut et al. 2007; Phillips et al. 2008), and tens to hundreds of meters in midlatitude ice-rich deposits (Holt et al. 2008; Plaut et al. 2009; Bramson et al. 2015; Stuurman et al. 2016; Petersen et al. 2018) and volcanics (Watters et al. 2007; Carter et al. 2009; Morgan et al. 2015). However, due to the relatively long wavelengths required for such penetration, and the inherently small associated bandwidths, the shallow subsurface (uppermost 10–20 m) remains unresolved.

In addition to higher-resolution sounding, synthetic aperture radar (SAR) imaging is another tool that could be utilized to explore the shallow depths of Mars. Orbiting SAR imagers have mapped Earth (e.g., Elachi et al. 1982; Jordan et al. 1995), Venus (Saunders et al. 1992), our Moon (Raney et al. 2011), and Titan (Stofan et al. 2007), but Mars has not yet had an orbital SAR mission. While sounders are nadir-pointing instruments, generating a profile of surface and subsurface echoes as the radar moves along the orbital path, SAR imagers emit signals from a side-looking radar to generate a swath of images based on the signal reflected by the surface and near-surface. The returned signal imparts information on surface roughness and the composition and structure of the near-surface (Figure 3). In the case of polarimetric SAR (PSAR), the polarization of the returned signal is recorded to place further constraints on the nature of the surface and near-surface scatterers (van Zyl et al. 1987). The polarization ratio for ices is unique; thus, PSAR provides a means to search for ice that is otherwise ambiguous or undetectable by other instruments. This technique has been used to detect ice on Mercury (Slade et al. 1992) and to investigate ice deposits on the Moon (Stacy et al. 1997; Campbell et al. 2003; Spudis et al. 2010). On Mars, similar to the Moon, PSAR can also be expected to constrain the composition through varying dielectric constant and surface roughness, geologic mapping under thick dust mantles, and regolith thickness and distribution (e.g., Cahill et al. 2014; Patterson et al. 2017).

Zoom In Zoom Out Reset image size

The radar image resolution is determined by antenna size. Through synthetic aperture processing, each scatterer is sampled multiple times as the radar travels along the orbital path. Leveraging the Doppler shift (among other effects) associated with the variation in the relative velocity of a sampled point and that of the radar's orbit, SAR processing replicates the same azimuth resolution as would be achieved by a substantially larger physical antenna (Elachi 1987). SAR is therefore a useful approach to operating radar on planetary missions. For example, Mini-RF on the Lunar Reconnaissance Orbiter is a C-band (7.14 GHz, 4.2 cm wavelength) and S-band (2.38 GHz, 12.4 cm) imaging SAR that weighed only 15 kg, with a small 1 m² antenna that provided pixel scales of 7.5 × 15 m (Chin et al. 2007; Raney et al. 2011).

Interferometric SAR (InSAR) utilizes multiple viewing geometries to construct a phase difference map of the surface in order to create a topographic model and to enable the quantification of surface change with subwavelength resolution, typically centimeter-scale displacements for usual SAR frequencies (Gabriel et al. 1989; Burgmann et al. 2000). These multiple geometries can be achieved in a single pass with dual antennas, or in multiple passes with a single antenna. Change detection, afforded through this technique, could include the quantification of seasonal deposits at the poles, either through a series of phase correlations over time with repeat passes, or differencing elevation models produced at intervals. InSAR-derived topography would provide much broader coverage than current imagery-based digital terrain models (DTMs) and could perhaps provide an order of magnitude higher vertical resolution than MOLA (centimeters rather than meters). This would be a major advance for geomorphic studies of potential ice-related features and would also enable clutter simulations for a high-resolution sounder.

Radar sounding at a resolution an order of magnitude higher than SHARAD (i.e., ∼1 m) would be complementary to imaging and ice detection with PSAR/InSAR. Sounding data would provide better constraints on near-surface structure and stratigraphy than PSAR/InSAR, but without the coverage, imaging, or detection of ice through polarimetry. Such sounding could be accomplished at the same central wavelength as SAR imaging, but to enable greater depth of signal penetration, it might ideally be done at frequencies between those of SHARAD and those of an imaging SAR, thus requiring a separate antenna.

Our study examined:

• 1.  
  Wavelength, horizontal resolution, and penetration of PSAR versus InSAR.
• 2.  
  Vertical resolution and total penetration depth tradeoffs for radar sounding.
• 3.  
  Antenna complexity and size for PSAR.
• 4.  
  Antenna complexity, size, and data rates for InSAR, with dual-antenna, single-pass operations, or single-antenna, multipass operations.
• 5.  
  Wavelength, bandwidth, antenna type, and size for radar sounding, including the possibility of using the same frequency as PSAR/InSAR.
• 6.  
  Coverage for InSAR topography and change detection versus operational mode.
• 7.  
  Combined use of an antenna for communications and PSAR/InSAR.

Over the course of our evaluation, it became clear that InSAR at frequencies needed for subsurface penetration would not likely have the ability to vertically resolve seasonal ice and that dual stereo cameras (see Section 4.2.2) would provide higher spatial and vertical resolution topography. Hence, the team ultimately selected the combination of PSAR and sounding with stereo cameras.

As noted above, our ability to map both the global distribution of ice and its horizontal and vertical structure is limited by the coarse resolution of current radar sounders, and the sensing gap between ∼5 and 20 m (Figure 4). To optimize measurements of the ice table in the existing observational gap, our study concluded that both sounding and SAR imaging in the P band (200–450 MHz, 1.5–0.7 m wavelength) are required, ideally at distinct center frequencies. The P band has also been advocated in various past studies (Campbell et al. 2004, 2012; Paillou et al. 2006; Pettinelli et al. 2007; Rincon et al. 2019). By contrast, the higher frequencies of the L band (900–1400 MHz, 33–21 cm) do not penetrate sufficiently to advance the state of knowledge of the depth to subsurface ice reservoirs. As noted in Figure 4, the lower frequencies of the P band would sense directly in the current observation gap for ice and will determine the ice lateral extent, volume, and presence at latitudes nearer the equator where ice is expected to be buried under several meters of dry overburden.

Zoom In Zoom Out Reset image size

The ability to sound and image surfaces covered by dust or regolith will be highly valuable to geologic applications, providing information on buried features, including valley networks and fluvial channels, volcanic flows and ash, and basin fill and other sedimentary deposits. Support for an orbital SAR instrument is demonstrated by Earth-based radar maps of Mars that show tremendous detail in surfaces that are dust covered, illuminating blocky lava flows in the Tharsis region, and deposits from the large outflow channels (e.g., Campbell et al. 2012; Harmon et al. 2012; Harmon et & Nolan 2017)

4.1.1. PSAR and Sounding

For MORIE's mission goals, an imager would enable the additional study of ice, particularly as exposed by craters, and data for key hydrous mineral stratigraphic sequences by increasing coverage and spatial resolution. The NEX-SAG report identified high-resolution (HiRISE-class; ∼30 cm/pix or better) optical imaging as a critical capability, which was included as a baseline instrument. However, HiRISE has covered <3% of the Martian surface, and a large imager is both a cost and mass driver. We chose radar as the large instrument with precedence for this mission concept, given the critical nature of the ice-mapping science goal. Additionally, particularly for studies of active processes, there is great value in coordinated targeting and simultaneous data collection at multiple scales from one spacecraft, as shown by MRO with coordinated imaging between CTX and HiRISE. Lower-resolution imaging, as envisioned by ICE-SAG, can expand spatial coverage, which is valuable for tasks such as identifying new ice-exposing craters. NEX-SAG briefly considered ultra-high-resolution (5 cm/pix) optical imaging but noted accommodation challenges and did not examine what could be achieved with a coarser resolution than that of HiRISE.

We considered the following imager trade spaces to determine which instrument parameters best addressed the science objectives and complemented the other instruments under consideration:

• 1.  
  Spatial resolution and coverage
• 2.  
  Stereo capability
• 3.  
  Signal-to-noise ratio
• 4.  
  Wavelength coverage
• 5.  
  Cost and mass

The MORIE study team considered a range of possible instrument pixel scales. After detailed study and image resolution modeling using HiRISE, the 1 m per pixel (1 m/pix) scale was found to be ideal for both observing new ice-exposing craters and doing follow-up analysis without requiring a very large telescope (both a mass and cost driver). In particular, the MORIE imaging strategy combines modest-resolution stereo imagers (CTX scale) for increased spatial coverage, and a 1 m/pix imager (with color) to achieve all imaging goals (Sections 4.2.1 and 4.2.2 below). This combination of instruments at different spatial pixel scales is sufficient to identify ice-exposing impacts as small as a few meters across. This imager complement is also capable of detecting many other forms of surface activity such as recurring slope linea (RSL; e.g., McEwen et al. 2011) or gully flows (e.g., Dundas et al. 2019). A low-cost and low-mass wide-angle imager allows for continued monitoring of weather and seasonal frost.

4.2.1. High-resolution Imaging

4.2.2. Dual Stereo Cameras

4.2.3. Wide-angle Imaging

The surface of Mars has been measured at a variety of wavelengths and spatial resolutions (Figure 5). The visible, near-infrared, and short-wave infrared (0.4 to ∼5 μm, VNIR/SWIR) and the long-wave (or thermal) infrared (∼5–50 μm, LWIR or TIR) provide complementary information on surface composition. The VNIR/SWIR is most sensitive to iron oxides, oxyhydroxides, as well as anions resulting from high- or low-temperature water-driven alteration and has been widely used on Earth and Mars to identify and discriminate among phyllosilicates, carbonates, sulfates, opal, etc. On Mars, this wavelength range was also crucial to identifying primary minerals such as high- and low-Ca pyroxenes and olivines. The long-wave infrared, in addition to sensing the alteration minerals noted above, is sensitive to bonds of the SiO molecule in varying crystallographic coordinations. All silicates have diagnostic absorption features at LWIR wavelengths and this range is useful for feldspars, quartz, and other common rock-forming minerals that lack strong features at shorter wavelengths. Both wavelength ranges have been used to map the composition and grain size of the Martian ices: water and CO₂.

Zoom In Zoom Out Reset image size

LWIR data have also been used to model surface thermal inertia and detect near-surface ice deposits by their thermophysical properties (e.g., Bandfield & Feldman 2008; Piqueux et al. 2019; Sizemore et al. 2020). Comparison of thermal inertia and radar ice detections will allow us to trace the increase in the depth to the ice that is expected from higher to more equatorial latitudes.

Both NEX-SAG and ICE-SAG strawman payloads included a CRISM-like instrument and a THEMIS-like thermal mapper both at improved spatial resolution over current measurements. The best SWIR data are from CRISM at 18 m/pixel, but with sparse coverage (∼2%), and global maps in the SWIR and LWIR are at a few hundred meters/pixel. THEMIS, a multichannel LWIR imager, cannot provide detailed mineralogy and is usually supplemented with data from the Thermal Emission Spectrometer (TES). However, TES had a very coarse spatial resolution (∼3–6 km). TES enabled quantitative petrology of regional igneous compositional variability on Mars (e.g., Rogers & Christensen 2007), but it was insufficient for detecting outcrops of either primary or alteration minerals. Higher spatial resolution LWIR spectroscopy could lead to major advances in our understanding of mineral assemblages on Mars, as LWIR spectroscopy allows for quantification of mineral abundances of both alteration and primary minerals, facilitating the retrieval of more complete mineral assemblages (Feely & Christensen 1999). High spatial resolution LWIR data could also expand and enhance our understanding of the myriad alteration sites identified with CRISM.

Advances from CRISM imaging spectroscopy have demonstrated that high-resolution <20 m/pixel compositional information is required to resolve the details of transitions in mineral assemblages that signal changes in environmental conditions (e.g., Murchie et al. 2009). One way to achieve this increase in spatial resolution is to decrease spectral resolution, as demonstrated by the improved spatial resolution but limited spectral channels of THEMIS (Christensen et al. 2004). At shorter wavelengths, HiRISE, HRSC, and CaSSIS color data demonstrate that the inclusion of infrared multispectral color can reveal compositional variations unobservable at visible wavelengths (e.g., Delamere et al. 2010). This was a major success of the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) instrument observing Earth with narrow channels concentrated around wavelengths where alteration minerals have numerous absorption features (e.g., Mars 2013).

The most scientifically compelling instrument payload includes multiple spectral instruments, but we conducted a more detailed investigation of this trade space to identify the most effective combination of spectral range/resolution and spatial resolution across SWIR/LWIR wavelength ranges for both environmental transitions and polar ices and stratigraphy. Consequently, our study considered:

• 1.  
  Wavelength ranges and spectral resolution needed to advance information on mineral and ice compositions and abundances over current data sets (Figure 5), including emerging instrument capabilities.
• 2.  
  Instrument suites with different combinations of imaging spectroscopy versus multispectral imaging versus point spectroscopy to obtain high-resolution composition and morphology required for stratigraphic interpretation.
• 3.  
  Whether each instrument required a dedicated telescope.

The final MORIE spectral suite takes advantage of instruments that have flown and those that are being developed for near-term launch opportunities such as the Mapping Imaging Spectrometer for Europa (MISE), Earth Surface Mineral Dust Source Investigation (EMIT), and Polar Radiant Energy in the Far Infrared Experiment (PREFIRE). We gather spectral data from the visible through the thermal infrared using channels from the high-resolution imager (Section 4.2.1) and both short- and long-wavelength infrared spectrometers. A key resource driving factor is not the spectrometer or detectors but rather the telescope(s) to attain high resolution. The SWIR and LWIR spectrometers have similar telescope requirements in terms of diffraction and signal-to-noise ratio. Another result of our study was the ability to use a beam splitter to allow the instruments to share a telescope and reduce overall mass while still achieving the desired spatial resolution in both SWIR and LWIR.

4.3.1. Short-wave Infrared

4.3.2. Long-wave Infrared