The radio tower had been dead for eleven years when Mara first heard it.
She was hiking the ridge trail above Carrizo Springs, the one that follows the old fence line where the Alvarez ranch used to end and BLM land began. It was late October, the light going copper, and she had her field recorder running because she’d been tracking a pair of zone-tailed hawks that nested somewhere in the bluffs. The recorder was set to capture everything above and below human hearing — ultrasonics for bat passes, infrasonics for weather fronts. She wouldn’t listen to the raw files until she got home.
At home, scrubbing through three hours of wind and bird calls and the occasional crunch of her own boots, she found it. Starting at the 1:47:22 mark and continuing for exactly four minutes and thirty-three seconds: a carrier wave at 162.4 MHz, amplitude-modulated with what sounded like — but could not possibly be — someone reading a list of numbers in Spanish.
The tower was a NOAA weather station, decommissioned in 2015 when the newer digital system came online. Mara knew this because her father had installed the original equipment in 1987, back when he worked for the National Weather Service out of the San Angelo office. She’d been up to the tower as a kid, had watched him swap out the backup battery packs and check the antenna alignment. It was a squat concrete blockhouse with a lattice mast, unremarkable, the kind of infrastructure you stop seeing after you’ve driven past it enough times.
The tower had no power. The transmitter had been removed. The antenna was still there but disconnected, its feedline coiled and zip-tied to a bracket near the base. There was nothing to generate a signal and nothing to radiate it.
Mara went back the next day with better equipment.
The signal was there again. Same frequency, same modulation. But the content was different. This time it was a woman’s voice reading what sounded like weather observations — temperature, humidity, barometric pressure, wind speed and direction — but for dates that hadn’t happened yet. November 3rd. November 4th. November 5th.
Mara wrote down the numbers.
On November 3rd, the high was 74°F, humidity 31%, barometric pressure 30.12 inHg, wind SSW at 8 mph. The signal had predicted 74, 31, 30.12, SSW 8. Exact.
On November 4th: 68, 44, 29.98, NNW 12. Exact.
On November 5th, the signal had predicted 71, 38, 30.04, S 6. The actual readings were 71, 38, 30.04, S 6.
She told no one. This was not a decision she made consciously; it was more like the secret told her that it wanted to be kept. She drove out to the ridge every three or four days, set up her receiver, and recorded. The signal was not always there. When it was, it lasted exactly four minutes and thirty-three seconds. Always.
The content varied. Sometimes weather. Sometimes the numbers without context, just a voice counting in sequences that Mara couldn’t decode. Once, a long passage that sounded like someone describing a landscape — the arroyo bends south here and there are three cottonwoods, the middle one split by lightning, and under the split trunk there is a flat stone — which Mara recognized, with a slow chill, as the wash behind her parents’ old house. She hadn’t been there in six years. She drove out the next morning. The three cottonwoods were there. The middle one was split. The flat stone was there.
Under the stone: nothing. Just earth, red and undisturbed, and a single millipede curling away from the light.
Mara had a background in signal processing. She’d done her master’s thesis on matched filtering for weak astronomical signals, and before the funding dried up she’d spent two years at the VLA helping characterize RFI — radio frequency interference, the bane of radio astronomy. She knew how signals worked. She knew how they didn’t work.
This signal could not exist. There was no transmitter. The frequency was wrong for any natural emission. The modulation was structured, intentional, human. But when she brought a spectrum analyzer to the tower itself, there was nothing. The signal only appeared on the ridge trail, a kilometer and a half to the southeast, at a specific bend where the trail crested between two limestone outcrops.
She set up a small antenna array — four dipoles in a square, spaced a quarter-wavelength apart — and tried to triangulate. The signal appeared to come from everywhere and nowhere. It had no direction. It was simply present in the space between the rocks, like temperature or humidity, like something the air itself was saying.
In December she recorded a transmission that was different from the others. The same woman’s voice, but halting, uncertain, as if reading something she didn’t fully understand:
The station will be decommissioned on March 15, 2015. The equipment will be removed. The antenna will be disconnected. But the frequency will remain. A frequency is not a thing. It is a relationship between events. You cannot remove a relationship by removing the things that once expressed it. The 162.4 will continue because the space learned it. The rocks learned it. Thirty years of weather broadcasts taught the ridge what that frequency sounds like, and now the ridge remembers.
Mara listened to this recording fourteen times. On the fifteenth listen, she noticed something she’d missed: under the voice, very faintly, the sound of wind through a lattice structure. The exact sound the antenna mast made on days when the wind came from the north.
She checked her field notes. The day she’d made the recording, the wind had been from the south.
January was cold and dry. Mara went to the ridge less often. When she did, the signal was shorter — three minutes, then two, then forty-five seconds. The content became spare. Just numbers, no context. Then just the carrier wave, the steady hum of a frequency with nothing riding on it.
On February 1st she recorded the last transmission. Twelve seconds long. The woman’s voice said: Gracias por escuchar. Thank you for listening. Then silence.
Mara went back every day for two weeks. Nothing. She expanded her search — drove the back roads with a scanner, hiked every trail within five miles of the tower, set up a remote monitoring station powered by a solar panel and a Raspberry Pi that logged the spectrum around 162.4 MHz continuously.
In six months of monitoring: nothing.
She published a paper. Not about the signal — she wasn’t ready for that, and she knew what it would sound like — but about the acoustic properties of limestone canyon formations and their effects on VHF propagation. It was a solid paper, well-received, cited fourteen times in the first year. Buried in the methodology section was a single sentence: Anomalous carrier artifacts were occasionally observed at 162.4 MHz in the vicinity of a decommissioned NOAA weather station, likely attributable to residual cavity resonance in the blockhouse structure.
It was the most dishonest sentence she’d ever written. It was also the closest she could come to telling the truth in a form that anyone would accept.
Years later, on a panel at a conference in Albuquerque, someone asked her what had first drawn her to study radio propagation in desert environments. She gave the standard answer — her father’s work, childhood visits to the tower, a general interest in how landscapes shape signals.
But driving home afterward, through the long empty stretch south of Socorro where the VLA dishes stand in their Y-shaped array like enormous flowers turned toward the sky, she thought about the real answer. The real answer was that she had once heard a landscape remember something it had been told, and she had spent the rest of her career trying to understand how that was possible.
She never heard the signal again. But sometimes, hiking in canyon country, she would stop and hold very still and listen to the specific quality of silence that fills a narrow space between rocks. It was never quite silent. There was always something — wind, insects, the faint thermal creak of stone expanding in the sun. And under all of it, if she listened with the right kind of attention, something that was almost but not quite a frequency. Something the air knew but couldn’t say.
The three cottonwoods are still there. The middle one is still split. Under the flat stone, there is still nothing but earth.
But the millipede is gone.