When NASA astronauts Gene Cernan and Harrison Schmitt of the Apollo 17 mission departed the lunar surface on December 14, 1972, they left a few things behind — including a US flag, a Moon buggy, and the lunar module's descent stage.
A new study by researchers from the California University of Technology has revealed that the latter regularly causes extra, tiny "moonquakes" which shake the lunar surface.
This finding comes, ironically, thanks to another thing Cernan and Schmitt left near their landing site — an array of four geophones used to conduct seismic experiments.
Reactivated between October 1976 and May 1977 for passive listening, these seismometers recorded thousands of subtle tremors on the Moon, the result of daily temperature variations.
Until now, however, the poor quality of the data had made a comprehensive analysis difficult — hiding the fact that some of the moonquakes were not quite what they initially seemed.
[...] In their study, geophysicist Professor Allen Husker of the California University of Technology and his colleagues used techniques not available in the seventies — like machine learning — to clean up the Apollo 17 passive seismic data and undertake a more robust analysis.
The team found that thermal moonquakes occur with the regularity of clockwork, every morning and afternoon. The latter are the result of the Sun leaving its peak position in the sky, allowing the lunar surface to begin to cool off.
However, the team's artificial intelligence model revealed that the seismic activity detected in the morning has a different profile — and are not regular thermal moonquakes at all.
Using the data from the seismometer array to triangulate the source of the morning quakes, Husker and his team found that they were coming from the left-behind descent stage of the Apollo 17 lunar module.
[...] Seismological studies of the Moon are — as on Earth — also a great way to get a glimpse of the structure beneath the surface. This is because seismic waves travel at different speeds.
As Husker adds, using moonquakes, "we will hopefully be able to map out the subsurface cratering and to look for deposits."
[...] Husker continued: "There are also certain regions in craters at the Moon's South Pole that never see sunlight; they are permanently shadowed.
"If we could put up a few seismometers there, we could look for water ice that may be trapped in the subsurface. Seismic waves travel slower through water."
Journal Reference:
F. Civilini, R. Weber, A. Husker, Thermal Moonquake Characterization and Cataloging Using Frequency-Based Algorithms and Stochastic Gradient Descent, JGR Planets, 2023. DOI: 10.1029/2022JE007704
(Score: 1, Funny) by Anonymous Coward on Monday September 11 2023, @10:24PM (1 child)
It's like the earthquake caused by a leaf falling to the ground.
(Score: 2) by Freeman on Tuesday September 12 2023, @01:42PM
That must be one big tree to make a leaf so big.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 4, Interesting) by shrewdsheep on Tuesday September 12 2023, @09:58AM (1 child)
I was curious how machine learning techniques could be used to model such relative sparse data representing complex physical processes. As a matter of fact, ML was not used. Heuristic and physical modeling was used, carefully taking into account previous tries. For such data, generic ML will not be useful for quite some time to come, IMO.
(Score: 3, Informative) by Unixnut on Wednesday September 13 2023, @04:46PM
Yes, but "AI" is the new hype, so every single press release must shoehorn "AI" in somewhere, to make it seem like they are cutting edge and hip. Plus I guess it may help with future project funding applications. Whether Machine Learning is actually used anywhere in the research is somewhat beside the point.