Conference PaperPDF Available

Robotic exploration potential of Martian caves


Abstract and Figures

Martian caves are among the best locations to search for evidence of extinct/extant life, may provide access to stable water ice deposits, and offer a protected environment for human habitation. We examined the robotic exploration potential of a subset of these features. Poster URL:
Content may be subject to copyright.
Cushing3, N. M. Bardabelias4, and N. G. Barlow1,5. 1Department of Astronomy and Planetary Sciences, Northern
Arizona University, Flagstaff, AZ;; 2Department of Biological Sciences, Merriam-Powell
Center for Environmental Research, Northern Arizona University, Flagstaff, AZ;; 3USGS
Astrogeology Science Center, Flagstaff, AZ;; 4Lunar and Planetary Laboratory, University of
Arizona, Tucson, AZ;; 5deceased.
Introduction: Martian caves are among the best
locations to search for evidence of extinct or perhaps
extant life, may provide access to stable water ice
deposits, and can offer a protected environment for
human habitation [1]. In recognition of the emerging
importance of caves on Mars, NASEM’s 2019
Astrobiology Strategy recommended emphasis on
research and exploration of subsurface environments
[2]. Additionally, mission architectures [1] and concepts
[3] were recently developed to provide
recommendations for examining Martian caves for
astrobiological and habitability potential (Fig. 1).
To date, 1,036 cave-like features have been
documented on Mars [4]. Williams et al. [5] modeled
cave water-ice stability on Mars and identified the most
favorable regions for ice to accumulate and persist.
These datasets combined were used as a first order down
selection approach to identify high priority caves for
Overall, we aimed to identify the highest priority
targets for a future robotic mission to a Martian cave.
Specifically, we: (1) defined caves within regions of
high subterranean ice accumulation; (2) calculated cave
densities per region; and, (3) identified concentrations
of caves on the Martian landscape. Additionally, we
provide two examples where we applied the selection
criteria for targeting high priority caves for robotic
exploration [6].
Methods: This project examined known cave locations
[4] and a subterranean ice accumulation model [5] to
deduce cave entrances within regions of predicted high
ice accumulation. We selected a subset of these caves
based on existing HiRISE imagery, which became our
dataset. We then evaluated these features using a
candidate selection criteria for robotic exploration [6].
These criteria include the potential for subsurface ice,
access to subsurface geology, and robotic accessibility
including terrain navigability and entrance
configuration. Geologic and structural characteristics
were determined using available geologic maps [7,8]
and HiRISE images. Nearest neighbor analysis was
performed to identify cave clusters. All analyses were
conducted using JMARS 5.1.8 and ArcGIS 10.8.1.
Figure 1. Locations of possible Martian cave entrances (red dots) [2] overlaid on the cave ice accumulation model
(where predictions of annual ice mass amounts (in kg/yr) within hypothetical caves) [5] for global Mars. Areas of
interest occur within the [A] Tharsis and [B] Phoenicis Lacus quadrangles. MOLA base map, courtesy NASA.
Robotic exploration potential of Martian caves: M. L. Kearney et al.
Results: We found 97 caves occurring within regions of
high ice accumulation (0.1-0.7 kg/yr) and 847 within
regions of medium accumulation (0.001-0.1 kg/yr; Fig.
1). The highest density (620 caves) was located within
the Phoenicis Lacus quadrangle with a density of 1.4 x
10-4 caves/km2, while the Tharsis quadrangle was the
second most dense with 0.5 x 10-4 caves/km2 (223
HiRISE imagery is available for 28 caves, which
were all within the Tharsis volcanic province. We
summarized structural and geologic characteristics for
these features (Table 1). Angular and jagged outcrops
along pit rims suggestive of examinable stratigraphy
was observed for 58% of the vertical features
potentially permitting robotic examination and
sampling of “deep” subsurface geology. At least 23
features would permit dual-axle rover access to the
entrance; this would be required for subterranean
ingress of a secondary tethered rover. Most (82%) of the
features were located on the lava fans and surrounding
plains of Arsia Mons. Nearest neighbor analysis
revealed that three features had from three to six
additional features within a 25 km distance and up to 10
potential cave entrances within a 50 km radius; clusters
will provide greater flexibly as multiple features could
then be closely scrutinized if aerial drone technologies
are employed.
While 28 features are currently being evaluated for
their exploration potential, we provide preliminary
results for two features as examples. The first feature is
a possible skylight situated on the southern lava fan of
Arsia Mons (Fig. 2A). Angular protrusions suggestive
of viewable stratigraphy are situated along the eastern
rim. Access could be gained via a drone or tethered
rover. We cannot discern whether a lateral entrance is
associated with this feature, which is a key requirement
for robotic exploration.
Our second feature is probably not viable as it is an
isolated feature occurring at an elevation of 20,244 m
within Caldera V [sensu 9], the southernmost caldera of
Olympus Mons. However, this is presently the only
feature suggestive of laterally-trending passage for
which HiRISE imagery is available. Located within the
southeastern wall of the caldera, the feature is likely a
conduit related to the caldera, possibly a vent.
Conclusions: While our work will ultimately involve a
thorough examination of all 28 potential cave entrances,
we have provided examples of how the selection criteria
will be applied. Ultimately, we will examine and
quantify the robotic exploration (as well as human use)
potential for all features using geospatial techniques.
Caves occurring within clusters will be further
examined and recommendations will likely be made for
additional HiRISE imagery acquisition.
Figure 2. [A] Skylight on southern lava fan on Arsia
Mons, diameter ~30 m (HiRISE ESP_042942_1680).
[B] Possible laterally trending feature on Olympus
Mons, length of apparent entrance, north to south, is ~10
m (HiRISE ESP_007669_1980).
References: [1] Titus, T.N. et al. (2020) Planetary
Caves Decadal Survey white paper. [2] NASEM (2019)
doi:10.177226/25252. [3] Phillips-Lander et al. (2020)
MACIE Decadal Survey white paper. [4] Cushing, G.E.
(2017) Mars Global Cave Candidate Catalog (MGC3),
PDS Archive. [5] Williams, K. et al. (2010) Icarus 209:
358368. [6] Wynne, J.J. et al. (2014) NASA JPL
Planetary Cave Workshop white paper. [7] Masursky,
H. et al. (1978) USGS Mars Geologic Map of the
Phoenicis Lacus Quadrangle. [8] Scott, D. & J.
Zimbelman. (1996) USGS Geologic Map of Arsia Mons.
[9] Neukum, G. et al. (2004) Nature 432: 971-979.
Lava Fans and Surrounding Plains of
Arsia Mons
Caldera and Flanks of Shield -
Cave category
Number of caves/
Diameter (m)
Elevation (m)
6,296 - 12,330
14,477 - 20,244
Entrance type
All vertical
3 vertical, 1 lateral
Table 1. Geologic and structural characteristics of the 28 caves with HiRISE imagery. Cave types: atypical pit crater
(APC), crack (Crk), skylight (Sky), irregularly shaped feature (Irr), entrance to rim of caldera (Rim), small rimless pit
(SRP), laterally trending entrance (Lat), and pit in the end of a trough (End) [4].
... Within the caves, temperatures are likely cold (<−25°C or 248 K) and thus could harbor metastable water ice. 74 Cave locations and predicted ice accumulation rates 75 ...
Since the 1980s, national and international planetary protection policies have sought to avoid contamination by terrestrial organisms that could compromise future investigations regarding the origin or presence of Martian life. Over the last decade, the number of national space agencies planning, participating in, and undertaking missions to Mars has increased, and private-sector enterprises are engaged in activities designed to enable commercial missions to Mars. The nature of missions to Mars is also evolving to feature more diversity in purposes and technologies. As missions to Mars increase and diversify, national and international processes for developing planetary protection measures recognize the need to consider the interests of scientific discovery, commercial activity, and human exploration. The implications of these changes for planetary protection should be considered in the context of how much science has learned about Mars, and about terrestrial life, in recent years.
... This feature would likely become a high priority target for groundtruthing. For Mars, a small number of aggregated pixels representing a potential cave may be a low priority target for further examination as a human shelter or storage depot, but potentially a higher priority target for robotic exploration-as it may be more buffered from the surface environment (e.g., [96,97]). Thus, the grouping of pixels with cave-like thermal signatures into clusters will enable workers to prioritize which potential cave localities require further examination (i.e., physical groundtruthing of terrestrial caves or additional imagery interpretation for planetary caves). ...
Full-text available
Since the initial experiments nearly 50 years ago, techniques for detecting caves using airborne and spacecraft acquired thermal imagery have improved markedly. These advances are largely due to a combination of higher instrument sensitivity, modern computing systems, and processor intensive analytical techniques. Through applying these advancements, our goals were to: (1) Determine the efficacy of methods designed for terrain analysis and applied to thermal imagery; (2) evaluate the usefulness of predawn and midday imagery for detecting caves; and (3) ascertain which imagery type (predawn, midday, or the difference between those two times) was most informative. Using forward stepwise logistic (FSL) and Least Absolute Shrinkage and Selection Operator (LASSO) regression analyses for model selection, and a thermal imagery dataset acquired from the Mojave Desert, California, we examined the efficacy of three well-known terrain descriptors (i.e., slope, topographic position index (TPI), and curvature) on thermal imagery for cave detection. We also included the actual, untransformed thermal DN values (hereafter "unenhanced thermal") as a fourth dataset. Thereafter, we compared the thermal signatures of known cave entrances to all non-cave surface locations. We determined these terrain-based analytical methods, which described the "shape" of the thermal landscape hold significant promise for cave detection. All imagery types produced similar results. Down-selected covariates per imagery type, based upon the FSL models, were: Predawn-slope, TPI, curvature at 0 m from cave entrance, as well as slope at 1 m from cave entrance; midday-slope, TPI, and unenhanced thermal at 0 m from cave entrance; and difference-TPI and slope at 0 m from cave entrance, as well as unenhanced thermal and TPI at 3.5 m from cave entrance. Finally, we provide recommendations for future research directions in terrestrial and planetary cave detection using thermal imagery.
ResearchGate has not been able to resolve any references for this publication.