On Sunday Morning The Sky Went Quiet For Forty-One Minutes

GREEN BANK, W.Va. — Four receivers. Four continents. One moment.

At 04:12 on Sunday the carrier went to zero. It stayed there for forty-one minutes and twelve seconds.

Then it came back at exactly the level it had left at. Every trace since has been ordinary.

The measurement

Odile Marchetti-Kwan has been logging the band from a shed on the Vercors plateau since 2014, on a receiver she built from a published design and calibrates monthly against a reference source. Her log for 3 August shows the carrier at its usual level until 04:12:07 UTC, at zero until 04:53:19, and back at its usual level within four seconds of returning.

Bandile Nkosi, near Sutherland in the Northern Cape, logged it going down at 04:12:09 and back at 04:53:20. Two other operators, in Saskatchewan and in Kerala, logged the same interval to within three seconds.

None of the four knew the others were listening.

Herschel Pond spent thirty-four years at Green Bank and has no view about what the carrier is. This newspaper asked him to read all four raw logs and told him before he opened them that we would print whatever he said. He confirmed the interval appears in all four data sets, at the same UTC times, and then set out four ways it could be an artefact:

• A shared external interferer that stopped rather than started, which would look identical to a signal stopping.
• A common-mode error in three home-built receivers using the same published design, which they do.
• A propagation effect at a shared ionospheric boundary, which he thinks unlikely at these frequencies but not impossible.
• Coincidence, which he was careful to say is a real explanation and not a way of declining to give one.

"I can't rank those for you," he said. "If you print that I can, you'll have made this story better and less true."

The forty-one minutes

The part of this story that resists reporting is what people say happened in the interval.

This newspaper has taken 61 accounts, all volunteered, none solicited by us before the logs were published. They come from eleven countries. They do not agree with one another and several of them contradict themselves. We are not going to summarise them into a pattern, because a summary is where a pattern gets manufactured.

We will report the two that were checked back with their sources, and the fact that both sources then asked us to be more careful than we had been.

Sister Ana Lucía Restrepo, prioress of a contemplative community in Boyacá, told us through an interpreter that Sunday's vigil, which begins at 23:00 local and runs past the interval, had been "unusually quiet." When this story first ran we put that word in the standfirst. She wrote to us four days later to say she had meant the ordinary quiet of a Sunday morning in a house of eleven women in the mountains, and that she had not intended a technical meaning and would not want one taken.

Her clarification, in full: "I told you the truth and you printed it accurately. But you printed it in a place where it meant something I did not say. Please put this here, at the same size."

Terrence Aboyade-Hill, a night nurse in Leeds who came off a fourteen-hour shift at 05:00 on Sunday, said the same in fewer words. "Quiet ward. Quiet wards happen. I shouldn't have said it to a newspaper."

We removed the word from the standfirst and we have left both accounts in the story, with their corrections attached, because deleting them would tidy the record in our favour.

What we are not saying

We are not saying the interval was a transmission stopping. We are not saying it was not. We are saying four instruments recorded a flat interval at the same time; that a career radio astronomer with no stake in it thinks the likeliest explanation is instrumental and named four candidates; that sixty-one people have told us things about those forty-one minutes; and that the two accounts we followed up were both walked back by the people who gave them, at their own initiative, without being asked.

The compiler of PX-1788, who has spent nine years assembling long-period lunar events from public archives, has a standing sentence about work like this that we asked his permission to borrow. He wrote it in his own catalogue, about his own data:

"I am not proposing an extraordinary explanation. I am only listing the events."