Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Radar evidence of an accessible cave conduit on the Moon below the Mare Tranquillitatis pit

Abstract

Several potential subsurface openings have been observed on the surface of the Moon. These lunar pits are interesting in terms of science and for potential future habitation. However, it remains uncertain whether such pits provide access to cave conduits with extensive underground volumes. Here we analyse radar images of the Mare Tranquillitatis pit (MTP), an elliptical skylight with vertical or overhanging walls and a sloping pit floor that seems to extend further underground. The images were obtained by the Mini-RF instrument onboard the Lunar Reconnaissance Orbiter in 2010. We find that a portion of the radar reflections originating from the MTP can be attributed to a subsurface cave conduit tens of metres long, suggesting that the MTP leads to an accessible cave conduit beneath the Moon’s surface. This discovery suggests that the MTP is a promising site for a lunar base, as it offers shelter from the harsh surface environment and could support long-term human exploration of the Moon.

This is a preview of subscription content, access via your institution

Access options

Buy this article

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Experimental results for the MTP (8.3355° N, 33.222° E) imaging with Mini-RF.
Fig. 2: Results for MTP radar simulations (8.3355° N, 33.222° E).
Fig. 3: Reconstructed MTP cave conduit based on an inversion of the the Mini-RF radar data.

Similar content being viewed by others

Data availability

The Mini-RF data are available through NASA’s Planetary Data System Geoscience Node (https://pds-geosciences.wustl.edu/). Wagner and Robinson’s17 internal morphology point cloud of the MTP is available at https://zenodo.org/records/6622042. The LROC NAC images and DTMs used in this study are publicly available through the Planetary Data System LROC Node at https://wms.lroc.asu.edu/. The data supporting this study are openly available at Zenodo via https://doi.org/10.5281/zenodo.11005458 (ref. 28).

Code availability

All the relevant analyses on the experimental data were performed with MATLAB. RaySAR is open source and available at https://github.com/StefanJAuer/RaySAR.

References

  1. Greeley, R. Lava tubes and channels in the lunar Marius Hills. Moon 3, 289–314 (1971).

    Article  ADS  Google Scholar 

  2. Halliday & William, R. Terrestrial pseudokarst and the lunar topography. Bull. Natl Speleol. Soc. 28, 167–170 (1966).

    Google Scholar 

  3. Horz, F. in Lunar Bases and Space Activities of the 21st Century (ed. Mendell, W. W.) 405–412 (Lunar and Planetary Institute, 1985).

  4. Haruyama, J. et al. Possible lunar lava tube skylight observed by SELENE cameras. Geophys. Res. Lett. https://doi.org/10.1029/2009GL040635 (2009).

  5. Robinson, M. S. et al. Confirmation of sublunarean voids and thin layering in mare deposits. Planet. Space Sci. 69, 18–27 (2012).

    Article  ADS  Google Scholar 

  6. Wagner, R. V. & Robinson, M. S. Distribution, formation mechanisms, and significance of lunar pits. Icarus 237, 52–60 (2014).

    Article  ADS  Google Scholar 

  7. Kaku, T. et al. Detection of intact lava tubes at Marius Hills on the Moon by SELENE (Kaguya) Lunar Radar Sounder. Geophys. Res. Lett. 44, 10–155 (2017).

    Article  Google Scholar 

  8. Donini, E., Carrer, L., Gerekos, C., Bruzzone, L. & Bovolo, F. An unsupervised fuzzy system for the automatic detection of candidate lava tubes in radar sounder data. IEEE Trans. Geosci. Remote Sens. TGRS.2021.3062753 (2021).

  9. Chappaz, L. et al. Evidence of large empty lava tubes on the Moon using GRAIL gravity. Geophys. Res. Lett. 44, 105–112 (2017).

    Article  ADS  Google Scholar 

  10. Horvath, T., Hayne, P. O. & Paige, D. A. Thermal and illumination environments of lunar pits and caves: models and observations from the Diviner Lunar Radiometer experiment. Geophys. Res. Lett. 49, e2022GL099710 (2022).

    Article  ADS  Google Scholar 

  11. Kobayashi, T., Kim, J. H., Lee, S. R. & Song, K. Y. Nadir detection of lunar lava tube by Kaguya lunar radar sounder. IEEE Trans. Geosci. Remote Sens. 59, 7395–7418 (2020).

    Article  ADS  Google Scholar 

  12. Nesnas, I. A. et al. Moon diver: exploring a pit’s exposed strata to understand lunar volcanism. Acta Astronaut. 211, 163–176 (2023).

    Article  ADS  Google Scholar 

  13. Haruyama, J. et al. in Moon: Prospective Energy and Material Resources (ed. Badescu, V.) 139–163 (Springer, 2012).

  14. Carrer, L., Castelletti, D., Pozzobon, R., Sauro, F. & Bruzzone, L. A novel method for hidden natural caves characterization and accessibility assessment from spaceborne VHR SAR images. IEEE Trans. Geosci. Remote Sens. 61, 1–11 (2022).

    Google Scholar 

  15. Nozette, S. et al. The Lunar Reconnaissance Orbiter miniature radio frequency (Mini-RF) technology demonstration. Space Sci. Rev. 150, 285–302 (2010).

    Article  ADS  Google Scholar 

  16. Raney, R. K. et al. The lunar mini-RF radars: hybrid polarimetric architecture and initial results. Proc. IEEE 99, 808–823 (2010).

    Article  Google Scholar 

  17. Wagner, R. V. & Robinson, M. S. Lunar pit morphology: implications for exploration. J. Geophys. Res.: Planets 127, e2022JE007328 (2022).

    Article  ADS  Google Scholar 

  18. Staid, M. I., Pieters, C. M. & Head, J. W. III Mare Tranquillitatis: basalt emplacement history and relation to lunar samples. J. Geophys. Res.: Planets 101, 23213–23228 (1996).

    Article  ADS  Google Scholar 

  19. Wagner, R. V. & Robinson, M. S. Occurrence and origin of lunar pits: observations from a new catalog. In Proc. 52nd Lunar and Planetary Science Conference (Lunar and Planetary Institute, 2021).

  20. Wynne, J. J. et al. Planetary caves: a Solar System view of processes and products. J. Geophys. Res.: Planets 127, e2022JE007303 (2022).

    Article  ADS  Google Scholar 

  21. Cushing, G. E. Candidate cave entrances on Mars. J. Cave Karst Stud. 74, 33–47 (2012).

    Article  Google Scholar 

  22. Sharma, R. & Srivastava, N. Detection and classification of potential caves on the flank of Elysium Mons, Mars. Res. Astron. Astrophys. 22, 065008 (2022).

    Article  ADS  Google Scholar 

  23. Henriksen, M. R. et al. Extracting accurate and precise topography from LROC narrow angle camera stereo observations. Icarus 283, 122–137 (2017).

    Article  ADS  Google Scholar 

  24. Auer, S., Hinz, S. & Bamler, R. Ray-tracing simulation techniques for understanding high-resolution SAR images. IEEE Trans. Geosci. Remote Sens. 48, 1445–1456 (2009).

    Article  ADS  Google Scholar 

  25. Head, J. W. III. Lunar volcanism in space and time. Rev. Geophys. 14, 265–300 (1976).

    Article  ADS  Google Scholar 

  26. Titus, T. N. et al. A roadmap for planetary caves science and exploration. Nat. Astron. 5, 524–525 (2021).

    Article  ADS  Google Scholar 

  27. Blank, J. G. et al. Planetary Caves as Astrobiology Targets. White Paper (National Academy of Sciences, 2018).

  28. Carrer, L. et al. Dataset of ‘Radar evidence of an accessible cave conduit on the Moon below the Mare Tranquillitatis pit’. Zenodo https://doi.org/10.5281/zenodo.11005458 (2024).

Download references

Acknowledgements

We would like to acknowledge all members of the Topical Team on Planetary Caves of the European Space Agency for the useful discussion on the interpretation of our findings. Capella Space X-band SAR imagery was provided by Capella Space under the Open Data Community programme. This work was supported by the Italian Space Agency (Contract No. 2022-23-HH.0, ‘Attività scientifiche per il radar sounder di EnVision fase B1’).

Author information

Authors and Affiliations

Authors

Contributions

L.C. formulated the concept. L.C., D.C. and L.B. developed the radar theoretical model for explaining the observations. L.C., D.C., R.P. and F.S. designed the experiments. L.C., D.C. and L.B. analysed the radar data. R.P. produced the 3D models of the pit and cave-like conduit. R.P and F.S. provided the geological interpretation of the experimental results. L.B. supervised the research and the related funding project. All authors co-wrote the paper and discussed the results and the related implications.

Corresponding authors

Correspondence to Leonardo Carrer or Lorenzo Bruzzone.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Astronomy thanks Chunyu Ding, Tyler Horvath and Matthew Perry for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Wagner and Robinson17 3D model of the Mare Tranquillitatis Pit with superimposed geometric quantities.

(a) Geometric model with pit characteristics and radar incident rays. The incident radiation rays are depicted for \({\theta }_{L}\) equal to the one of the Mini-RF acquisition. (b) Geometric model detail depicting the parameters involved in the inversion of the cave conduit characteristics. Refer to methods for the description of the variables displayed in the figures.

Extended Data Fig. 2 Comparison between the experimental X-band SAR image and the radar simulation and ground truth of a series of terrestrial analogue pits in Lanzarote, Spain (Lat = 29.165° deg, Lon = −13.454° deg).

(a) Capella Space X-band (9.65 GHz) Very High Resolution Synthetic Aperture Radar image14. Radar look direction is indicated with a white arrow. (b) 3D radar simulation24 without subsurface Lidar 3D digital model. (c) 3D radar simulation24 with subsurface Lidar 3D digital model. The red lines identify the radar response originating from the conduit interior. (d) 3D Lidar scans and drone photogrammetry of the surface (transparency) and the subsurface14. Color coding from red to green indicate a progressive increase of the points depth. (e) Superimposition of a detail of the Synthetic Aperture Radar image (Jameo Redondo and Cumplido) with the 3D Lidar scans and drone photogrammetry of the surface and the subsurface14.

Extended Data Fig. 3 Additional examples of tested models and 3D Radar Simulations Results.

3D Radar Simulation assuming (a) roof and floor slope of 10°, (b) roof and floor slope of 20°, (c) roof and floor slope of 50°, (d) roof and floor slope of 60°, (e) roof and floor slope of 80°, (f) roof and floor slope of 50° and 40°, (g) roof and floor slope of 60° and 40°, (h) roof and floor slope of 50° and 60°. The red shape marks the outline of the anomaly in the experimental data (Fig. 1a).

Extended Data Fig. 4 Examples of the evaluated models latitudinal power profiles.

3D Radar Simulation assuming (a) only the surface elevation model, (b) Wagner and Robinson’s17 3D Pit Model (surface and overhang), (c) model A (roof and floor slope of 3°), (d) model B (roof and floor slope of 55° and 45°), (e) conduit roof and floor slope of 5°, (f) conduit roof and floor slope of 50°, (g) conduit roof and floor slope of 60°, (h) conduit roof and floor slope of 70°, (i) conduit roof and floor slope of 50° and 40°, (l) conduit roof and floor slope of 60° and 40°, (m) conduit roof and floor slope of 20° and (n) conduit roof and floor slope of 80°. The normalized power profiles are evaluated at a fixed latitude of about 8.335°. The two power peaks of about 0 dB and -10 dB are the overhang and conduit response, respectively. There is a discrepancy of about 10 dB between the experimental and simulated data in the level of the power response from the lunar surface. This implies that the simulator, as expected, is correctly estimating the scattering contribution from the pit, but underestimating the diffuse scattering contribution from the lunar surface by about 10 dB. However, this does not affect the general validity of the results. The large negative peak of the simulations corresponds to the interior of the pit. This is not shown in the experimental data as due to the Mini-RF dynamic range.

Extended Data Fig. 5 Results on selection of the best-fitting model through correlation analysis between experimental and simulated radar data.

(a) Values of the correlation coefficient (see Methods) between experimental and simulated data versus the roof and floor slopes. The black arrow represents the uncertainty with respect to the best fit model denoted as B. (b) Maximum value of the correlation coefficient versus the roof’s slope. As a result of the radar ambiguity in determining the cave parameters, the two models denoted as A and B are possible. The range of plausible slopes for which the correlation coefficient yields a high value is in line with what predicted by the radar geometric model for estimating the cave conduit slope from the radar image (see Methods). The correlation coefficient value for the simulated data based on the sole Wagner and Robinson overhang model17 is equal to 0.66.

Extended Data Fig. 6 Comparison between LROC NAC image and the meshed model of the MTP.

(a) LROC NAC image M155016845R at 0.41 m/pixel resolution. Notably, two large boulders of 8–10 m of size are located in the south-western side of the MTP’s floor. These were not modelled in the procedural rock population generation as they were considered outliers in the global population and also they do not affect the outputs of the simulated Mini-RF response. (b) Shaded meshed model of the MTP with the central pit bottom populated by the procedurally generated rocks with geometry nodes with random spatial distribution and a size distribution between 1 m and 4 m. This particular range of size has been selected based on the boulder’s size that can be observed from LROC NAC images of the MTP. (c, d, e, f, g) Transparency view of the modelled conduit in plan-view and in perspective view. The LROC NAC DEM and the photogrammetric model by Wagner and Robinson17 are in orange whereas the procedurally generated pit and cave used for the simulations of the subsurface response to Mini-RF are in cyan. The presence or absence of a cave is simulated, and diameter ranges are displayed here starting from 30, 50, 100 and 200 m. The checkboxes show whether the output of the Mini-RF simulation matches with the observed data or not.

Extended Data Fig. 7 3D Radar simulations results for different values of the conduit width.

3D Radar Simulation assuming a conduit width of (a) 15 m, (b) 30 m, (c) 55 m, (d) 100 m and (e) 200 m. (f) Value of the radar measured conduit versus the simulated model cave width. The red shape marks the outline of the anomaly in the experimental data (Fig. 1a).

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Carrer, L., Pozzobon, R., Sauro, F. et al. Radar evidence of an accessible cave conduit on the Moon below the Mare Tranquillitatis pit. Nat Astron 8, 1119–1126 (2024). https://doi.org/10.1038/s41550-024-02302-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41550-024-02302-y

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing