10 Purple Mountain
Lenghu, Qinghai, April 2028; Purple Mountain Observatory, Nanjing · June – August 2029.
The pipeline flagged it on 21 April, one line in a morning list of eighty: a source at 61° from the Sun, moving east at almost exactly the Sun’s own rate, tagged co-orbital with a low score. Most co-orbital tags at that elongation were ghosts off a bright star, or two unrelated detections the linker had married. Lin opened the cutouts in the canteen on the ridge, between the oxygen machine and the noodles, and put it on the follow-up queue for dusk. The scheduler would complain. Dusk was when the survey took its calibration fields, and the calibration fields belonged to Hefei.
Then four nights of wind, with dust off the Qaidam so thick the dome stayed shut and the yardangs below the mountain went brown. The 26th was clear. So were the 27th, 28th and 29th.
She took them herself from the control room, in a down coat because the heating there was set for the electronics. The 2.5 m opened on the western sky at the end of nautical twilight and had perhaps forty minutes before the survey took the telescope back. The sky that close to the Sun never got properly dark. On the 29th the seeing was 1.8 arcseconds, the object a soft V 17.9 smear that the centroider kept wanting to pull toward a background galaxy.
Five nights, counting the survey night, and nowhere near enough arc. She went to the Minor Planet Center’s public file of orphaned tracklets and found seven from the American surveys that fit her positions, February and March, unclaimed. With them the arc was ten weeks. Gravity alone would not close it. The residuals walked, all in one direction, along the track. She freed an along-track acceleration and it came back negative at 1.8σ: braking.
She thought about the Minor Planet Center report the way she thought about a paper submitted too early: it would go out with her name on it and she would spend the next season defending it. Two sigma in twilight was not a result. She kept it on her own queue, a few minutes a week through the summer, and did not report those either. By the turn of the year the arc would be long enough that, if it was still braking, it would say so at five sigma and she would not have to. The queue was hers. Nobody would ask.
On the 29th the dome came round and closed on a brown horizon, and she did not send it.
The preprint went up on 13 June 2029 and was in the astro-ph listing the next morning: Purple Mountain and USTC, fourteen authors, her name with the asterisk.
The sentence in section 4 was hers. The Hefei group had struck it twice in the shared draft and offered cannot be excluded; the deputy director had suggested moving it to the discussion, where people stopped reading. She had put it back each time, as the last sentence of the section. It was conditional in the only place it needed to be. If somebody wanted to call it something else, they could do the arithmetic and argue with the number.
The paper had waited on the American circular. Purple Mountain had recovered the object in January from Xuyi, and then the Academy had sat on the draft until 22 May, when Cambridge designated it and an amateur in America got his name on the circular. After that it took three weeks to clear. Lin had not argued with the three weeks. She had held the spring herself.
The methods said the object had been recovered in January 2029, which was true. The summer points were in the fit and not in the table. She had not taken them out. Very few people read the tables.
By the 16th it was in Nature’s news pages under a headline that used the word mystery, and the American designation, 2028 DN214, was in the first line, as if the object had been born in Cambridge, Massachusetts.
The man from the ministry came to Xianlin on 19 June. His card said State Administration of Science, Technology and Industry for National Defence. He had read section 4; there was a printout in his folder with the last sentence underlined in pencil, and nothing else marked.
He asked two questions.
“How far is it from Earth, in kilometers?”
“Today, 158 million. In January it was 141. It makes a loop around the Lagrange point once a year.” She waited to see if he wanted the error. He did not.
“Is its position stable?”
“The center of the loop has not moved by more than 3,000 kilometers since January. Our uncertainty is about 1,500.”
He wrote both numbers down, carefully, with the units, and read them back to her. He had the second one wrong by a factor of ten and corrected it himself before she could.
A man who wanted to understand the object would have asked what it was made of, or why it was fading. These were the numbers you needed if you meant to do something about it. She did not round either one.
At the door he said, “Several people in Beijing think it is artificial.”
“It braked,” Lin said. “I can tell you that much.”
“And the mass?”
“It has lost about 9% of its light since January. There is no dust.”
He went to find his driver.
That evening, in her office at the end of the fourth-floor corridor, she went back over the second number.
He had written 30,000 and struck the last zero before she could open her mouth. People copying a number they had just heard got it wrong and waited to be told. He had known what it was supposed to be. The preprint gave the orbit, not the history of the loop center. That came from the summer fit, and the summer points lived in one folder on the group server that three people could open: Lin, the postdoc who ran the Xuyi reductions, and the dynamicist down the corridor who had checked her integrator in February. Neither of them talked to Beijing about anything but money.
She had never looked at the folder’s access log. She expected her own name eighty times, and it was there, with the postdoc’s once in March. On 9 May and again on 23 May every file in the folder had been read, in order, start to finish, at 02:41 and at 03:17, by a maintenance account on the observatory’s internal server with no person’s name on it.
The postdoc had never heard of the account. The man who ran the group’s cluster said maintenance accounts belonged to the IT office and he had never been trusted with a password for one.
She filed a ticket with the IT office, three lines and the two timestamps. The answer came on the 21st, signed with the office’s name and nobody’s: a scheduled backup job, migrated to the new server during the spring upgrade. The account belonged to the job. The job ran at night so as not to load the network. They regretted any concern.
It was plausible. Backups ran at night. The spring upgrade had also taken the fourth-floor printers away for most of April. She wrote back to ask why a scheduled job had run twice in May and not since, and received the same paragraph, with the regret moved to the top.
She changed her password. She copied the folder onto the ThinkPad she had taken to Lenghu, which had not been on a network since its wireless card died in the dust, checked every file against the server copy, and deleted the server copy. The ThinkPad went in the bottom drawer of her desk, under the 2027 observing logs, with its charger coiled on top.
The Hubble archive took her a Saturday morning. The designation found nothing, so she searched for moving-target programs in late 2028 whose ephemerides passed within a degree of the trailing Lagrange point, and got one.
| PI: | Dana Okafor, Jet Propulsion Laboratory |
| Title: | Reflectance spectroscopy of a candidate Earth co-orbital |
| Orbits: | 40 |
| Instruments: STIS, WFC3/UVIS | |
| Approved: | 2028 October |
| Status: | Completed |
| Exclusive access period: 12 months | |
Forty orbits of Director’s time. Approved in October 2028, seven months before the Americans gave it a name and eight before they said anything about it. One name on the program, no co-investigators. Okafor’s publication list was orbits: linkage, astrometry, small-body catalogs, nothing with a spectrograph in it. The abstract was three sentences and called the target “of unusual dynamical interest,” which was how you described a thing braking at L5 to a committee that already knew. A linker had asked for STIS. Somebody above her had wanted to know what it was made of, and quickly.
NASA’s June release had said metallic, about two kilometers. It had not said when they learned it.
The data would come out of their exclusive period in October, at the earliest. By then the spectrum would be a year old and somebody in Washington would have decided what it meant.
She had held a spring and a summer. They had held forty orbits and a spectrum, published nothing, and then published a press release.
The letter from Foreign Affairs came at the end of August. A consultation in Vienna in November, convened by the Secretary-General’s office, closed to the press. The Americans would bring a scientist. China would bring one too, and the Academy had given them her name.
She said yes before she had finished reading it.
The Americans would bring Okafor, or someone who had read Okafor. All Lin had of her was a program number, a publication list and a name on an archive page, and none of it said how the woman sounded when she was asked a question she could answer. She wanted to see what someone knew who had asked for forty orbits in October. She wanted to see which questions Okafor answered exactly, and which ones she had been told not to answer at all.