Chang'e-7's Landing Ellipse Moved 41 Kilometres. Nobody Announced It.
Nine published accounts of the mission give a target centre. Six give one point; three give another, 41 kilometres away. The new point first appears eleven days after a cluster of long-period events under the old one.
BREMEN, Germany — Two points on the Moon, both published by the same agency, nine months apart.
88.8°S, 123.4°E.
87.5°S, 135.4°E.
They are the target centre of the Chang'e-7 landing ellipse. On a sphere of radius 1,737.4 kilometres they are 41.1 kilometres apart, which is further than the ellipse is long.
Six published accounts of the mission give the first. Three give the second. There has been no erratum, no note, and no announcement in any language we can read.
What we counted
Nine documents describe where this spacecraft is meant to come down: two conference papers, four refereed articles, and three successive editions of the mission's own English press kit.
We pulled the target centre out of each one by hand. Two people did it, separately, neither told what the other had got. They agree on all nine.
Six accounts, from June 2023 to May 2025, give the same point to one decimal place. The three from February 2026 onward give a different one.
The change appears without comment. The third press kit does not say the site has moved. It simply gives a different number in the same table, in the same typeface, on the same page as the second edition.
The eleven days
On 2 February 2026 the lunar seismometer catalogue this newspaper keeps in The Vault, PX-1788, logs six long-period events inside eleven hours. The compiler's own coordinates put them under the original target centre, to within thirty kilometres.
Eleven days later, a conference paper is submitted carrying the new centre for the first time.
That is the finding. It is also the whole of the finding, and this desk has spent six weeks trying to make it smaller rather than larger.
What the cluster is when you re-derive it
We did not trust our own catalogue. That is not modesty; this desk published a story last month establishing that four of forty entries we spot-checked in it were wrong.
Dr. Bo-Kyung Ferrar, a planetary seismologist paid by this newspaper, went back to the public archive and re-derived the 2 February events from the raw traces without looking at the compilation.
Six becomes four. Two of the catalogue entries are the same event, listed twice, eighteen minutes apart, from two stations.
"Four long-period events in eleven hours is in the top two per cent of that catalogue," she said. "It is not in the top hundredth of a per cent. I have four better clusters in the same decade and nobody moved anything after those."
Three ordinary reasons
Dr. Hyacinth Oduya-Renke compared the two sites on public elevation and illumination data. She was not told which one the mission had started with.
The new site is flatter — mean slope 4.1 degrees against 8.6. It holds continuous sunlight for a larger fraction of the year, 0.83 against 0.78. And the Earth sits 2.4 degrees higher above its horizon, which is worth a great deal to a direct radio link at that latitude.
"You have found three good engineering reasons and one coincidence," she said, "and you are only excited about the coincidence."
The room
The new coordinates were presented at a session in Bremen in June. Our correspondent was in the room.
Afterwards he asked eleven people who work on polar landing missions whether they had noticed that the target had moved.
One had. She had noticed in April, from the press kit, and had assumed there was a note somewhere and had not gone looking for it.
The postponement
On 19 August four maritime navigation warnings covering the range danger areas were cancelled inside eleven hours, and the launch did not happen.
We checked that pattern against the previous three postponements of Chinese lunar launches. Two of the three look the same, and both were later attributed to propellant loading. The third does not look the same at all.
Two out of three is not an instrument. It is a coin with a chip in it, and we are printing it only so that nobody has to take our word that we looked.
What this desk takes from it
We can establish that the target moved. We can establish that it was never announced. We can establish, to the day, when the new number first appeared in public.
We cannot establish why, because the agency does not answer letters, and the most useful sentence in this story came from somebody who told us we should have telephoned two years ago.
She is right. The reply is printed below, entire, and the part that stings is the last paragraph.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We did not ask anybody where the mission is going. We read every published account of it we could find — six years, four venues, three editions of the press kit — and pulled the target coordinates out of each one by hand, twice, without telling either checker what the other had got. Then we paid a selenologist to compare the two sites on public data with no indication of what we hoped she would find, and a seismologist to re-derive the February cluster from the raw archive rather than from the compilation this newspaper itself publishes.
Who we spoke to
- Prof. Ines Kwarteng-Vogl, Chair, an international lunar landing-site coordination panel. Interviewed at the conference and again by video call; reply printed in full June and August 2026 Disputes the framing of this story in terms this desk accepts.
- Dr. Hyacinth Oduya-Renke, Selenologist; illumination and slope modelling. Compared both sites on public elevation and illumination data at our request and expense July 2026 Found three ordinary engineering reasons to prefer the new site.
- Dr. Bo-Kyung Ferrar, Planetary seismologist. Re-derived the 2 February cluster from the public archive without reference to the compilation August 2026 Reduces six events to four. Two entries are one event listed twice.
- Nine published accounts of the mission, Conference papers, refereed articles and three editions of the press kit. Downloaded, dated against the publisher's own record, and each coordinate extracted by hand and checked twice 2023–2026
- China National Space Administration, Beijing. Written questions on 3 July 2026 and again on 5 August, in English and Mandarin, through two channels July–August 2026 No reply of any kind. We will print one whenever it arrives.
Documents
- PX-1788 — Lunar seismometer network, long-period event catalogue, unofficial compilation accepted
What we could not confirm
- Why the site moved. The agency has not said, and will not say to us. Three engineering explanations fit the data and so does the seismic cluster, and this newspaper cannot choose between them and has not tried.
- Whether the postponement has anything to do with either. We have four cancelled navigation warnings and a pattern that matched two of the previous three Chinese lunar postponements, both of which were later attributed to propellant loading. That is a weak instrument and we are not leaning on it.
- The cluster itself. [[PX-1788]] is an unofficial compilation, and this desk published last month that four of forty entries we spot-checked were wrong. Dr. Ferrar's reprocessing takes six events down to four. We do not know what a third reading would do.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Meridian TelegraphTHEY MOVED THE LANDING SITE AFTER THE MOON SHOOK
Ran the eleven days as causation in the first sentence and in the headline.
Had that we did not
Both sets of coordinates, correctly, which is more than most outlets managed.
Left out
That six of the long-period events are four; that the new site is flatter, better lit and better placed for the Earth link; and that a conference submission deadline fell on 15 February.
-
The Continental WireChina Postpones Chang'e-7 Lunar Launch
Four hundred words on the postponement, sourced to the agency's own announcement.
Had that we did not
The announcement, in full, and the previous launch dates.
Left out
The site change. A reader of the wire item does not learn that the mission is now aimed somewhere else.
-
They Buriedthis newspaperChang'e-7's Landing Ellipse Moved 41 Kilometres. Nobody Announced It.
Read all nine published accounts, extracted the coordinates from each, and had the separation and the site comparison worked independently.
Had that we did not
The nine accounts, the arithmetic, a reprocessing of the seismic cluster that halves it, and a reply that takes the story apart.
Left out
We put the site change and the postponement in the same story and gave the reader no help separating them. They are two facts eleven days and six months apart, and running them under one headline invites exactly the inference we then spend four hundred words refusing. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Prof. Ines Kwarteng-Vogl, chair of an international lunar landing-site coordination panel on 28 July 2026 by our reply editor, in writing, with the coordinate table, Dr. Oduya-Renke's comparison and a draft of this story attached, and twenty-one days to respond. Replied on 16 August 2026. Printed in full and unedited, at her insistence and ours.
Your coordinate table is correct. I have checked all nine and you have not made an error. I want to say that first, because everything after it is going to sound like a complaint.
Here is the complaint.
You have not found when the site changed. You have found when a conference deadline fell.
The submission date you are treating as a discovery is 13 February. The deadline for that session was 15 February. It is 15 February every year and it has been 15 February since 2011. What you have identified is the first moment at which a working group was obliged, by an external calendar with no interest whatever in the Moon, to write down what it had been assuming for some months. Papers are not decisions. They are what people type in the four days before a deadline.
I can be more specific, and I will, because vagueness here would be a kind of cowardice. Landing-site selection for a polar mission of this class runs on a cycle of eighteen months to three years. It involves slope, illumination, communications geometry, thermal environment, the descent guidance envelope, and — always, and usually decisively — the fact that somebody has re-run the terrain model with a better elevation product and the numbers have moved. Nothing about that process could complete in eleven days. Your own selenologist tells you the new site is flatter by four and a half degrees and better lit by five percentage points, and you print it, and then you print the eleven days next to it in a larger size.
I am not going to tell you that agencies announce these changes well. They do not. The absence of an erratum between the second press kit and the third is a real failing and you are right to say so, and I have said so myself in a room where it was less welcome than it is here.
But consider what you are asking for. You are asking a mission to publish a note every time an engineering trade closes differently than it did a year ago. If it did that, the note would run to hundreds of entries, nobody would read it, and the one entry that mattered would be at line four hundred. I would rather have the note. I am telling you that having it would not have helped you, because you would not have read it either, and neither would the outlet that ran your finding as a headline about the Moon shaking.
One last thing, and it is the thing I actually mind.
Six people work on this panel. Any of them would have taken your telephone call in 2024, or 2025, or in March of this year, and told you the site was moving, because it was not a secret, it was merely uninteresting. You rang in July. By then it had become a story, and a story is much harder to correct than a fact.
Ask earlier. That is the whole of my advice and you may print it as rudely as it reads.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.