Moonquakes

The Moon quakes, and people have known it since the week of the first landing. During the first moonwalk in July 1969 Buzz Aldrin set a small, solar-powered seismometer on the ground at Tranquility Base, and it was switched on while he and Neil Armstrong were still outside. Over the next eight years instruments at four more landing sites recorded thousands of events, and between them they sorted the Moon's shaking into four kinds. Deep moonquakes happen 800 to 1,000 km down and keep time with Earth's tides. Shallow moonquakes happen near the surface and are the ones strong enough to worry anyone planning to build there. Thermal moonquakes are small cracks and slips as the ground swings between the frigid lunar night and the hot day. And impacts, from meteoroids and, for a few years, from hardware NASA crashed into the Moon on purpose, shake the ground from outside. Almost all of this is tiny. The team's 1973 report put the moonquakes it had recorded at Richter magnitude 2 or less, at rates of 600 to 3,000 a year per station. The exceptions, the strongest of 28 shallow quakes at about magnitude 5, are why a subject that was once about the Moon's interior is now also about where to land.

The listening post was more modest than the results suggest. Apollo 11's seismometer ran on solar panels, slept through its first lunar night and stopped taking commands on 27 August 1969. The long record came from the experiment packages left by Apollo 12, Apollo 14, Apollo 15 and Apollo 16, which ran on radioisotope generators and did not need the Sun. They formed a rough triangle about 1,100 km on a side across the nearside, each with three long-period sensors and one short-period sensor, sensitive enough, the team wrote, to register ground motion of 0.05 nanometres. The Apollo 12 station started work on 19 November 1969, and on 30 September 1977 the whole network was shut down. Almost everything on this page about natural moonquakes rests on those eight years and a handful of nearside instruments designed in the 1960s, and scientists are still finding new things in the tapes.

The strangest discovery came on the Apollo 12 station's second day. On 20 November 1969, after Pete Conrad and Alan Bean had rejoined Richard Gordon in lunar orbit, controllers sent the empty ascent stage of their lunar module into the surface, 2,383 kg at 1.68 km/s, skimming in at 3.7 degrees, about 76 km from the new seismometer. On Earth a jolt like that would give a sharp arrival and a quick fade. The Moon gave a signal that grew for about 7 minutes and faded so slowly that the total ran to about 55 minutes, without the familiar sequence of seismic waves an earthquake produces. "It was very peculiar," Yosio Nakamura, a member of the team, recalled in 2019, "nothing like any signal produced here on Earth." At the announcement, according to the University of Texas Institute for Geophysics, Maurice Ewing of Lamont-Doherty Geological Observatory reached for an image: "It is as though one had struck a bell, say, in the belfry of a church, and found that the reverberation from it continued for thirty minutes." That sentence is the source of the popular line that the Moon 'rang like a bell'; the words are Ewing's, and the recorded signal ran well past his thirty minutes (about 55 minutes in the team's 1970 report, over an hour in the Texas institute's retelling).

In 1970 the team weighed scattering against a near-surface waveguide, and by 1973 it had settled on an explanation that has held up. The Moon's outer rock is almost perfectly dry, so seismic waves lose very little energy as they travel, and it is intensely fractured, so the waves scatter back and forth instead of passing cleanly through. On Earth, water in rock damps vibration quickly; on the Moon there is almost none to do it. To learn more, NASA turned spent hardware into seismic sources. From Apollo 13 onward the Saturn IVB third stage, emptied after sending the crew toward the Moon, was steered into the surface: the Apollo 13 stage, nearly 14 tonnes at 2.58 km/s, hit 135 km from the Apollo 12 station on 15 April 1970 (still 14 April in US time zones), and NASA compared the Apollo 14 stage's impact to 11 tons of TNT. More lunar module ascent stages followed the one from Apollo 12. By the end, the Apollo 17 report counts nine artificial impacts recorded by the network, each with a known time, place and energy, which is exactly what a seismologist needs to calibrate a planet. Apollo 14 and 16 also carried active seismic experiments, and Apollo 17 laid out eight explosive packages of 57 to 2,722 g, detonated from Earth after the crew had gone, which mapped about 1.2 km of lava fill under the Taurus-Littrow valley.

The deep moonquakes turned out to run like clockwork. Many signals matched each other almost exactly, as if the same patch of rock had failed again and again, and by 1971 the team had seen that they arrived in bursts once a month, near perigee, when the Moon is closest to Earth. The final summary of the experiment put them between about 800 and 1,000 km down. Earth's tide squeezes and relaxes the Moon's interior on each orbit, and these quakes are the interior answering. They are also where the Moon keeps one of its open questions. Almost all the deep sources that could be located are on the nearside. In 2005 Nakamura identified about 30 source regions that are probably on the farside, and concluded that either the deep interior on the far side is nearly aseismic, or something very deep in the Moon absorbs or bends the waves before they reach nearside stations (the Apollo team had suggested in 1973 that a deep central zone, possibly partly molten, stops or weakens shear waves from far-side sources). He leaned, cautiously, toward the second, on limited data. New instruments are meant to help settle it: explaining why Apollo heard so few far-side quakes is a stated goal of NASA's Farside Seismic Suite, and comparing the nearside and farside interiors is a goal of the seismometer on Chang'e 7, in the words of the Chinese Academy of Sciences academician Wu Fuyuan.

Thermal moonquakes are the Moon's background crackle, and part of that crackle comes from the hardware astronauts left behind. A lunar day lasts about a month, and the swing from night to day heats and cools the surface enough to crack rock and set loose regolith sliding. The Apollo team saw it early: its 1973 report described thousands of small signals on the short-period seismometers, tied closely to lunar sunrise and sunset, and listed the lunar module descent stages and other equipment left at each site among the likely sources. When Tamama, Civilini and colleagues sorted more than 12,000 thermal events in the Apollo 17 geophone record in 2025, they found that the repeating ones clustered around early sunrise and came mostly from the east, where the lunar module's descent stage had been left standing. Many of those, in other words, were probably the lander ticking as it warmed. Other events look like rock cracking or regolith sliding down crater slopes, and they linked 45 to several boulders near the site that may be cracking in the heat. Their paper warns future missions to expect the same repeating signals from their own hardware.

The shallow moonquakes are the ones that matter. The four long-running stations recorded 28 of them between 1969 and 1977, ranging from about 2 to around 5 on the Richter scale in NASA's summary, and NASA's Moon pages give the top of the range as 5.5 with shaking that can last more than 10 minutes. Watters and Schmerr call them the most energetic events the Apollo network recorded, although the team's 1973 report, with less of the record in hand, said meteoroid impacts had produced the largest signals up to then. There were probably more than the classic count: in 2024 Keisuke Onodera went back to the short-period records that had largely been set aside because of their noise, found 22,000 new events of all kinds, and concluded there were 2.6 times more tectonic events than previously thought, with the Apollo 15 site busier than the others. The weakness in all of it is geometry. Four stations on one side of the Moon locate a distant quake poorly, and for decades nobody could say with confidence where the shallow ones started.

The candidate that emerged is a shrinking Moon. The Moon has been losing heat since it formed, and as the interior cools it contracts, and a brittle crust answers by breaking along thrust faults, where one slab of crust is pushed up and over another. The result is a lobate scarp, a step-like cliff often tens of metres high. In 2010 Thomas Watters and colleagues used Lunar Reconnaissance Orbiter images to show these scarps are spread across the whole Moon and are among its youngest landforms, crisp enough to cut small craters. By 2019 NASA said the orbiter's camera had imaged more than 3,500 of them, and it puts the shrinkage at more than about 50 metres over the last several hundred million years. In 2019 Watters and five co-authors reran the 28 Apollo shallow quakes through an algorithm designed for sparse networks. Eight epicentres landed within 30 km of a scarp, close enough for strong shaking, and six of those eight occurred when the Moon was near apogee, when tidal stress adds to the compression on these faults. They wrote that this, together with fresh boulder fields and disturbed soil on the scarps, strongly suggests the Moon is tectonically active today. That is a strong case built on an inference: 20 of the 28 quakes do not fall within 30 km of a known scarp, the epicentres still carry wide uncertainty, and the argument rests on locations and timing rather than on a quake caught on a mapped fault by a nearby instrument. The sources of the stress are still being worked out. Watters's group models global contraction plus Earth's tides, and Nypaver and Thomson, mapping 1,116 recently active wrinkle ridge segments on the maria in 2022, found orientations that match predicted stresses from both contraction and the Moon's slow outward drift from Earth, plus reactivation of old structures tied to the South Pole-Aitken basin.

Whatever the exact mix, the faults are where people want to go. In January 2024 Watters and colleagues mapped a cluster of scarps in de Gerlache Rim 2, then one of NASA's candidate Artemis III landing regions near the south pole, and linked the largest to one of the strongest shallow moonquakes Apollo recorded, whose poorly located epicentre may fall there. Their model says a quake of about moment magnitude 5.3 could have built that scarp, with strong to moderate shaking out to at least 40 km, and that most steep slopes inside Shackleton crater could shed their regolith in light shaking if it is loose. In July 2025 Watters and Nicholas Schmerr turned the Apollo 17 valley into a record of past quakes. Boulders that rolled down the valley walls and a landslide across the floor, all sampled by Eugene Cernan and Harrison Schmitt and dated by cosmic-ray exposure, point to repeated shaking from the nearby Lee-Lincoln fault in quakes of at least magnitude 2.9 to 3.3. They estimate a magnitude 3 quake on that fault about once every 5.6 million years, so the chance of a hazardous one during any single day of a short visit is about 1 in 20 million, with the caveat that this is a lower limit. The warning is for things that stay: a 50 m tall lander like SpaceX's Starship would need about ten times less ground motion than a boulder to be at risk, so the advice is to keep landed assets and long-lived infrastructure away from young faults and away from slopes with perched boulders. In December 2025 Nypaver, Watters and colleagues published what they describe as the first global map of young small mare ridges, adding another inventory of possible shallow epicentres across the maria.

For all that, the Moon has had almost no seismometers since 1977. The attempts predate Apollo: Ranger 3, 4 and 5 in 1962 each carried a seismometer in a crash-cushioned capsule, and none got a working one onto the surface. The instrument that broke the silence, in the description of researchers who later reanalysed its data, was India's ILSA, built by ISRO's Laboratory for Electro-Optics Systems around silicon micro-machined accelerometers. The first unit was lost when Chandrayaan-2's lander crashed in 2019; the second landed on Chandrayaan-3 in August 2023 and recorded in the south polar region until 2 September. Its team reported more than 250 distinct signals, about 200 matched to the Pragyan rover's movements and instrument operations and about 50 unexplained, possibly from nearby micrometeorite hits, the soil heating, or the lander itself adjusting. An independent check by Keisuke Onodera and Taichi Kawamura in 2025 examined about 100 candidate natural events and found almost all tied to the rover, the instruments or the resonances of the lander and rover, with no positive evidence for a natural quake or impact. As of September 2026 it is fair to say ILSA proved a small seismometer can work on the Moon and heard a great deal of its own spacecraft, and that whether it caught the Moon itself is unresolved.

The next instruments are in hand or in development, and as of 23 September 2026 none has a firm ride. On 11 August 2026 NASA declared its Lunar Environment Monitoring Station complete: a self-contained seismometer suite from Goddard, with sensors from the University of Arizona and Silicon Audio, designed to work through the two-week lunar night without outside power or a heat source for between three months and two years. In August 2025 NASA described it as a payload for potential flight on Artemis III, but Artemis III is now planned as a test flight in low Earth orbit in 2027, with the first crewed landing moved to Artemis IV, and LEMS will stay in a Goddard clean room until it is assigned to a landing mission. The Farside Seismic Suite, whose very broadband sensor was built as a flight spare for NASA's InSight Mars lander, was due to go to Schrödinger basin on the far side, until Draper and NASA ended that delivery contract on 15 July 2026. Fleet Space's SPIDER seismometer is on Firefly's Blue Ghost Mission 2 to the far side, now set for no earlier than 2027. China's Chang'e 7 is to carry a seismometer to the south pole, and on 23 August 2026 China's crewed spaceflight office, in a notice reposted by CNSA, announced it could not launch in this year's window, without giving a new date. Fifty-seven years after Aldrin set his seismometer down, the Moon's quakes are better understood than ever, and for now, as far as the public record shows, nobody is listening to them.