Robert Wagner was doing the kind of work that almost never produces anything. On 24 October 2025 he was running a quality check on a global map of the Moon. The job means stacking hundreds of wide-angle frames taken years apart, so that software can grey out everything unchanged and show everything that had changed as a bright or dark patch. On a normal pass the software hands him hundreds of false alarms, most of them small shifts in lighting or shadow.
Then a bright spot appeared inside a dark halo, spanning hundreds of pixels on a camera whose pixels are roughly the size of an American football field. It marked a crater 222 metres wide on the Moon’s eastern near side that did not exist before the spring of 2024, and it is the largest newly formed impact crater found anywhere in the solar system.
“I just stopped, dropped everything, and started looking into what that spot was,” Wagner, an image-processing specialist from Intuitive Machines who works with the Lunar Reconnaissance Orbiter Camera system, told NASA in its account of the discovery. The crater is now named McGetchin, after the lunar scientist Tom McGetchin. Two papers describing it were published in Science Advances on 16 September 2026.
The ground that used to be there
Before the impact, the site was an old patch of lunar surface with a complicated boundary running through it, at 1.3536 degrees north, 67.1765 degrees east. It sits just inside the outer ring of the Crisium basin, 330 kilometres from the edge of Mare Crisium. Lava had flowed in from an ancient eruptive centre 23 kilometres to the east-southeast, lapping up against the highland flank of a 7,200-metre-wide crater called Dubyago N and leaving an irregular boundary between dark mare basalt and paler highland material.
The authors read that boundary as the reason the new crater is slightly lopsided. The discovery paper records a rim 231 metres across at its widest, on a north-northeast line, and 213 metres at its narrowest, east-southeast. Fitting the two halves of the rim separately gives 230 metres on the western side against 214 on the eastern. The authors take that to suggest excavation into loose granular regolith on one side and a coherent subsurface lava flow on the other.
The surface was also, like almost all of the Moon, thoroughly cratered already. Within the footprint of the new crater the largest existing pit was 40 metres across. None of those older craters survives. Within the continuous ejecta blanket outside the rim, only five of 103 craters between 6 and 12 metres wide are still identifiable, and 12 of 35 in the 12 to 48 metre range, all of them heavily degraded.
Sometime between 11 April and 22 May 2024
The impact happened inside a 41-day window that the before-and-after images bracket. A comet or asteroid fragment struck, and the paper puts the energy of the event at 6.5 x 10^10 kilojoules, assuming an impactor density of 3 grams per cubic centimetre and an impact speed of 15 kilometres per second. The paper calls that more than an order of magnitude beyond the energy of the previous record holder, a 70-metre crater that formed in 2012 and put at 2.4 x 10^9 kilojoules on the same assumptions. Divide the two published figures and the ratio comes out at about 27. The new crater’s event is also more than four orders of magnitude beyond a 10-metre one. NASA describes the object as the size of a three- to six-storey building; a firmer estimate of its size and force is expected in a later paper.
It left a funnel. The crater is 43 metres deep, with a depth-to-diameter ratio of 0.19 that is typical of the freshest craters this size. Its walls run at a median slope of 24 degrees and reach 41 degrees at the steepest. The floor is a hummocky patch 15 by 30 metres, and a slump boulder 16 by 32 metres sits about 10 metres below the eastern rim.
The rim itself stands a median 7.7 metres above the old surface. A scaling relation derived from terrestrial crater measurements and estimates from lunar craters puts about 4.6 metres of that down to ejecta piled on top, which leaves about 3.1 metres of target material structurally pushed upward. The largest ejected boulder, 13 by 9 by 3 metres, is perched on the southern rim, and the second, 8 by 7 by 2 metres, landed on the south-eastern flank. The crater interior holds the brightest material in the whole 9 by 27 kilometre frame, and also material darker than mature mare basalt, which the authors read as enriched in glassy, quenched impact melt, perhaps of mare origin.
The halo was the giveaway
The Wide-Angle Camera that caught it is the moderate-resolution instrument of the three-camera LROC system, imaging the whole Moon roughly once a month since 2010 from about 100 kilometres up. Individually its frames barely resolve a 222-metre hole. Stacked and ratioed against older frames, they show the splash of material around it, and the size and shape of the hole itself took the Narrow-Angle Camera.
The team had averaged around 540 observations per pixel across the site before the impact and 60 after it, then normalised everything to a common viewing and lighting geometry and divided the later mosaic by the earlier one. Noise cancels. Real change survives. In the result Wagner looks for a small fuzzy halo around a bright point, the signature of regolith splashed outward by something new.
That comparison also revealed how far the disturbance reached. A brighter zone extends about 15 kilometres from the impact site. A fainter, patchier darker zone runs further still, and the discovery paper gives its reach two different ways. The text puts it more than 120 kilometres beyond the outer edge of the bright zone, about a thousand crater radii out; the paper’s own figure caption says about 100 kilometres from the crater, with localised patches to 140. In places it breaks into a delicate lacy texture. That probably formed where sparse clumps of ejected material came down far from home. Eight kilometres west, a ridge stands 700 metres above the crater and drops steeply to Mare Spumans beyond it. The mare at the base of that ridge is up to 2.5 percent darker than it was, which means material cleared the ridge. The authors note that the ejection angles required sit awkwardly with the conventional picture of jetting at less than two degrees above the surface.
A patch of colder ground seven kilometres across
Once the crater was found, the orbiter’s thermal instrument, Diviner, was pointed at it. Its first observations came on 1 November 2025, roughly a year and a half after the impact and far too late to catch any residual heat, which after the LCROSS impact faded to near-undetectability within about four hours.
What Diviner found instead was a cold spot about seven kilometres across in which the surface cools 8 to 9 kelvin further through the lunar night than the ground around it, with a measured peak anomaly of 8.61 kelvin. Decompaction is the explanation on offer. An impact decompacts the top centimetres to decimetres of soil, and looser material holds less heat through the night. In situ evidence from Apollo 16 supports that reading. The mechanism doing the decompacting is not settled, and the paper lists a granular flow set off by a cascade of secondary impacts, an impact-generated gas flow, and seismic shaking among the proposed mechanisms that McGetchin may finally separate. Fitting the cooling curve requires, on one model, that density climb with depth over a scale height of about 39 centimetres, against about 6 centimetres for the untouched regolith beside it. On a two-layer model it requires a discrete decompacted layer about 3.2 centimetres thick. The real profile probably sits between the two.
This is not a peculiarity of a very large impact. Of the 21 new craters larger than 20 metres that LROC has identified, every one above about 30 metres has a detectable cold spot, and the size of the temperature anomaly scales with the crater, though that scaling begins to flatten above about 100 metres, where McGetchin is the only new crater to sit. The Apollo 16 crew landed on a faint old example of one, and their bootprints there were deeper than at other sites. Cold spots fade over something like 100,000 to two million years, which is why many of the ones visible today are worn-down versions of what McGetchin now shows in its original state.
The seventeen-year window behind the superlative
The superlative is doing less work than it looks like it is doing. McGetchin is the largest crater found to have formed while we were watching, which is a claim about a seventeen-year archive of orbital images, set against a surface billions of years old. Craters kilometres wide and larger cover the surface. The relevant comparison set is the at least 1,000 new craters LRO has catalogued since 2009, in which the previous largest was 70 metres. The discovery paper and NASA both reach wider than that, to the largest newly formed crater found anywhere in the solar system, and this article’s headline carries the same claim. None of the sources says what the baseline for that wider comparison is.
The century framing comes out of a model. Crater production models predict an impact of this size on the Moon about once every 132 years on average. Robinson’s team reads that as a statistically rare, effectively once-in-a-lifetime observation. Read the other way, an average interval of that length says nothing about whether the previous one fell 12 years ago or 400. A numerical estimate of the impactor’s size and force has not yet been published. The three- to six-storey building is NASA’s characterisation of its size, and NASA says the estimate derived from the narrow-angle images is expected in a future paper.
The impact date is a window, not a moment. None of the three sources reports an observed flash, and the crater’s age comes entirely from the gap between two images. The site’s position on a mare-highland boundary also matters. Some of what makes the crater unusual, the lopsided rim and the odd tongue of darker ejecta running north, may say more about what it landed on than about what hit it. The authors searched for definitive evidence of a grazing impact and did not find any.
The crater will not stay this way
Its value right now is that it has barely changed. Powell’s team calls the cold spot virtually unaltered, which is the whole reason the site is worth this much instrument time, and that condition is temporary. The distal darkening around the 70-metre crater that formed in 2012 has already measurably faded, and earlier work suggests it may disappear entirely on the order of fifty years.
The cold spot will last far longer. Watching it fade from a known starting point should give the best chance yet at a calibration for the fading timescale. The authors say that may let cold spot properties serve as a rough age estimate for some of the youngest craters on the Moon.
As of 18 February 2026 Diviner had observed the site across nine campaigns and 43 orbits, twenty of them at night. Robinson’s team ends on what would help next, which is a visit: exploration by a mobile surface asset, and in-situ measurements of surface properties that would speak directly to what the cold spots are. NASA adds that ground loosened this far from a crater could matter for how rover wheels behave.