9 Fifteen Months
Animas, New Mexico, then JPL, Pasadena · Saturday, 19 May – Monday, 18 June 2029.
Earl Voss had the roof rolled back by ten past eight, with the Sun a few minutes under the Peloncillos and the sky over the ridge still too bright for anybody’s program but his.
The big surveys gave short shrift to the sky near the Sun after sunset. Earl had heard the reasons at three different meetings in Tucson and agreed with every one of them: the sky was bright, the good part of the night was short, and a forty-million-dollar camera had better things to do than fight the twilight for an hour. Earl didn’t have better things to do, and that was his whole observing plan. In four years it had got him nine discoveries, all main-belt, all faint, and an email from a woman at the Minor Planet Center saying his timing was the cleanest she’d seen from an amateur that year. He’d read it out to his wife at supper, and again the next night in case she’d missed anything.
The telescope was a 20-inch on a fork mount he’d machined himself. He had spent thirty-eight years designing gearboxes for the haul trucks and ball mills at Morenci and Tyrone, and his right-ascension drive had less backlash than anything you could buy, which he would tell you about if you stood still long enough. He’d ground the secondary on the kitchen table over a winter, Ronchi-testing it at midnight until his wife quit asking when it would be done. The camera took its time off a GPS receiver bolted to the roof rail, every frame stamped to the microsecond. Half the people on the mailing list still used their laptop clocks and wondered why their residuals all ran two arcseconds along the track.
He was on his third field, 61° from the Sun, when the blink caught something at the edge of the noise. V 18.3, if that. He’d have thrown it out for a hot pixel except it was in the next frame too, a hair east, and in the one after.
He took twelve frames over 48 minutes and measured every one against Gaia stars, by hand where the software got lazy. It was moving 0.99° a day, east. Earl sat back and did the arithmetic again, slower. A main-belter out at that elongation ought to be crawling. Anything close enough to move that fast ought to be moving a good deal faster, and some other way. This one was keeping pace with the Sun.
He ran it against everything numbered and everything that wasn’t. Nothing within a degree. He checked his clock, which he knew was fine, and his plate solutions, which were fine, and the temperature log on the tube, because thermal drift had fooled him once in 2026 and once was all he allowed it.
By twenty to twelve it was down in the junipers on the ridge. He closed the roof and wrote it up at the kitchen counter, and read it through before he sent it.
| Subject | Odd slow mover, W twilight, 2029 May 19 |
Friends -- 12 positions attached from Animas tonight, 03:02-03:50 UT May 20. V ~18.3, elongation 61 deg. Rate 0.99 deg/day, PA 78. Nothing known within a degree. Frames GPS-timed (to the microsecond, as always), astrometry against Gaia DR3, residuals under 0.25". Near as I can tell it is moving right along with the Sun. Would welcome eyes if anybody can get that low.
Clear skies,
Earl Voss
20-inch, homemade secondary, Animas NM
The circular went out on Tuesday morning, and Glen was in Dana’s doorway before she had finished the first screen of it.
“Who,” Glen said, “is Earl Voss?”
“Retired. Animas, New Mexico.” She had read his post on Sunday, and his positions three times, and told nobody.
2028 DN214. The observations ran 18 February to 28 March 2028 from Catalina, Pan-STARRS, ATLAS and NEO Surveyor, then a fourteen-month gap, then twelve lines from E. Voss, Animas, 20 May UT. The orbit was gravity only, residuals 0.58” RMS. The 2028 survey points carried most of that, and nobody at the MPC was going to lose sleep over half an arcsecond on survey astrometry. Her braking was in there somewhere, spread thin across the old residuals. It was very nearly the clean orbit she had put on the shared drive fourteen months ago.
“He posted to the list on Saturday night,” Glen said. “On a Saturday. With GPS timing. He ground his own secondary.”
“His timing’s better than Catalina’s.”
“That isn’t funny.”
“It wasn’t a joke.”
Glen read the designation off her screen. “Twenty twenty-eight,” he said. “They kept our year.”
He stayed in the doorway. Somewhere down the hall a printer was jammed and somebody was swearing at it in a low, steady voice.
“Headquarters called,” he said. “They want a release out before the Chinese or the Europeans or some kid on Reddit does it for them. Saturday, probably. A telecon. They want me to do it.”
“And me?”
“On the line. Muted. In case there’s a technical question.” He looked at the MPEC again. “If we’d gone to Icarus in October, the credit line would be us.”
“I know, Glen.”
“Four surveys and a man in a shed.”
By Wednesday afternoon a science reporter at one of the wire services had done the subtraction. First tracklet, 18 February 2028. Circular, 22 May 2029. Earth’s new companion sat in NASA survey data for fifteen months. By Thursday it was the number in every story, and nobody bothered to say from when.
NASA’s release went out at ten in the morning Eastern on Saturday, 2 June: NASA Confirms First Known Trojan Asteroid at Earth’s L5 Point. Natural, it said, about two kilometers, metallic. It had nothing to say about the rest.
The teleconference was at one Eastern. Dana dialed in from her kitchen table with the phone on mute, as instructed: technical questions only, and only if the press officer passed them to her. Glen was in a studio at headquarters with the press officer and a man from the Planetary Science Division she’d met once.
“The first Earth Trojan ever found at L5,” Glen said, and paused where somebody had told him to pause. He’d been saying that sentence to doorframes since March of last year. Over the phone it sounded as if legal had read it first.
Size, about two kilometers. Composition, iron-nickel, from Hubble. Mr. Voss, a remarkable amateur observer, and NASA was grateful. Somebody from Sky & Telescope asked why it had taken fifteen months, and Glen said that confirming a Trojan wanted a long arc, which was true of Trojans and had nothing to do with this one. Then a reporter whose name the press officer didn’t quite catch: “Dr. Farris, you’ve had this object since early 2028. Has the orbit changed in that time?”
“No. It’s a very well-behaved orbit,” Glen said. “It’s come to rest near L5, essentially, and—”
“Rest relative to what? Trojans librate. They don’t rest. And since when?”
There was a pause on the line. Dana’s thumb was on the mute key.
“Rest is the wrong word,” Glen said. “I mean it’s in a stable configuration. A tadpole orbit. Those are stable for very long times. We’d expect it’s been there—”
“You’d expect, or you know? Your team’s data goes back fifteen months. Was it always where it is now?”
“The orbit in the circular is a fit to all the data, and it’s consistent with—”
“I’m asking about your data, not the circular.”
“Let’s take the next question,” the press officer said.
The next question was whether NEO Surveyor’s budget was safe. Dana kept her thumb where it was.
Rita’s statement went up on the ESA site on Tuesday morning, under her own name, three paragraphs. The one that mattered was the second. ESA’s Planetary Defence Office observed 2028 DN214 from the Teide Observatory in spring 2028 and shared its results with NASA at the time.
Rita had written the one sentence that was good for Darmstadt and bad for Pasadena, and stopped there. By evening it was ESA: We Told NASA in 2028, and the reporter from the teleconference had sent JPL media relations a single question, which they forwarded to Dana without comment: What results?
Kessler’s email came on Thursday, 14 June, at a quarter past six in the morning her time. It was the first thing he had sent her since Why do you ask?
Have you read the Chinese preprint?
She hadn’t. It had gone up on arXiv overnight: Purple Mountain Observatory and the University of Science and Technology of China, fourteen authors, corresponding author Lin Wen. Non-gravitational deceleration of the Earth Trojan 2028 DN214 in Lenghu observations of 2028.
The Wide Field Survey Telescope, the 2.5-meter on the Qinghai plateau. Five nights in April 2028. Recovered in January 2029, the methods section said, and linked to the public isolated-tracklet file. Braking along the track at 2.1σ, from the nights in the table. A linear fit to the along-track acceleration, and its zero: 25 August ± 10 days.
Hers was 23 August ± 12, from a summer of weekly solutions, and the note in her bag had said the 21st. Theirs was two days away, with smaller error bars, from a telescope nobody outside China had known was looking.
Section 4 was one paragraph, and its last sentence had no hedge in it but the first word. If the decline in brightness reflects a loss of mass, the implied rate is of order 10⁵ kg s⁻¹.
That was her own number, the one in Priya’s column. It was sitting under a NOFORN banner in nine inboxes in Washington, and here it was in English on a public server, ranked fourth in astro-ph.EP by lunchtime.
Table 1 was the Lenghu astrometry. She read down the date column, then went back to the top and read it again.
By Friday the hashtag was #FifteenMonths. A cable network ran a chyron that said WHAT ELSE IS NASA HIDING? over footage of the Kennedy Space Center gift shop. Media relations forwarded her another question, this one from the Washington Post: Can Dr. Okafor confirm or deny the Chinese mass-loss figure? A man on a video with two million views held Earl’s post up to the camera, printed out, with moving right along with the Sun circled twice in red. He was entirely sure. “They’ve had it since 2028,” he said. “Ask yourself what keeps pace with the Sun.”
Her sister Ada called on Sunday a little before nine at night.
“Is it coming here?”
“No.”
“Don’t just say no. They had a physicist on, Dana. He said it’s been parked out there for a year and a half and NASA knew the whole time.”
“It isn’t coming here. It’s where it’s been.”
“Where’s that?”
“About as far from us as the Sun is.”
“Is that far?”
“It’s far, Ada.”
“The man said it’s metal. A whole mountain of metal. He said if it hit us it would be the end of everything.”
“It isn’t going to hit us.”
“How do you know?”
“Because I’ve had its orbit to a centimeter a second for a year.”
A child said something in the background, and Ada covered the phone and said something back. When she came on again her voice was lower. “How long have you known about it?”
“I can’t talk about work, Ada.”
“That’s a yes.”
“It’s a can’t.”
Ada didn’t say anything for long enough that Dana checked the screen to see if the call had dropped.
“Should I be scared?”
“I’ll call you if you should.”
“That’s not an answer.”
“It’s the one I’ve got.”
On Monday she had the preprint open again before seven, Table 1 full width on the screen.
Eleven rows on five nights. A survey came back to a field when its cadence said to. Four nights running was follow-up. She had asked Catalina for exactly that in March 2028, on the phone, because she’d already known what she had.
Five nights in April and a recovery in January could not put anybody’s zero on 25 August to ten days. Somebody had kept looking all summer, and left it out of the table.
The methods section said January.
26 April 2028. 27 April. 28 April. 29 April.