Study Links Nighttime Harbor Noise to Porpoise Avoidance
Quieter operational windows may matter more than average daytime decibels, researchers find.
A four-year acoustic study of the Port Meridian shipping channel suggests harbor porpoises abandon whole segments of the waterway during nights of elevated low-frequency noise, even when the twenty-four-hour average decibel readings that regulators rely on look unremarkable.
The finding, released Monday by the Basin Marine Acoustics Program, rests on 41,600 hours of hydrophone recordings gathered at nine fixed stations between the Outer Bar and the rail bridge, cross-checked against 212 boat-based visual surveys. In the three lower stations, porpoise detections fell by 68 percent on nights when noise in the 63- to 125-hertz band exceeded 118 decibels for more than ninety minutes. Daytime detections at the same stations barely moved.
Lead author Dr. Tess Morin, who has run the program since its first two-station pilot, said the pattern points toward regulating the clock rather than the meter. “The animals are telling us about the schedule,” she said. “We spent a decade arguing about caps nobody could enforce and nobody could hear. Meanwhile there is a four-hour window most nights where the channel goes quiet on its own, and the porpoises come straight back into it.”
Port officials called the findings constructive and noted that some night work is driven by tide windows and rail schedules outside local control.
Morin's team measured that return. After a noisy interval ended, median time to first detection at Station 4 was 71 minutes; at Station 7, closer to the container berths, it stretched past four hours. That asymmetry, she said, is the part of the paper she expects to be argued about longest, because it implies the animals are not simply reacting to instantaneous sound but carrying some memory of where the channel was unpleasant.
## What the sensors actually endure
The romance of the work ends at the equipment locker. Field technicians described a maintenance burden that shapes the dataset as much as any model does. Station 2 lost eleven weeks to a corroded connector. Station 6 recorded gull strikes on the surface buoy for most of one autumn, a rhythmic thumping that had to be filtered out by hand. Batteries sag in cold water, and a firmware update pushed in the second winter shifted timestamps by 400 milliseconds on three units before anyone noticed.
“People think the hard part is the math,” said senior technician Ruben Ostro, who has pulled and reseated most of the moorings personally. “The hard part is a February swap in a six-meter boat with numb fingers. If we miss a service window, we don't have a gap in a spreadsheet. We have a hole in a season.”
That candor is deliberate. The program publishes both its data and its analysis code, a condition attached by the Basin Research Trust after years of coastal science that could not be independently checked. Uncertainty bands travel with every released figure, and the program's communications officer, Delphine Okonjo, said she has stopped promising reporters a single headline number. “The honest answer has a range in it,” she said. “We decided we would rather be quoted accurately than quoted often.”
Graduate students on the project write methods sections structured as training manuals, so that a cohort arriving after a two-year grant turnover can reproduce a deployment without hunting down someone's retired laptop. Continuity, Morin argues, is itself a research output.
## The counter-case from the berths
The Port Meridian Harbor Authority did not dispute the acoustics. Its objection is operational, and it is not trivial. Deputy operations director Callum Weathersby said roughly 40 percent of night activity in the lower channel is fixed by forces the port does not control: tide windows for deep-draft vessels, and the two overnight rail departures that determine whether a container leaves the basin the next day or three days later.
“I can give you a quiet hour I already had,” Weathersby said. “What I cannot give you is a tide. If the study becomes a permit condition written by someone who has never watched a bulk carrier wait out a foot of water, we will hit the target on paper and move the same noise ninety minutes earlier.”
Weathersby also raised a displacement question the researchers concede is open: whether concentrating operations into fewer, louder blocks is better or worse for the animals than a diffuse hum. The paper does not answer it. Morin called it “the obvious next experiment” and said it would require the cooperation of at least two terminal operators to run honestly.
A working group convened by the basin's environmental council will meet in October to test whether voluntary quiet windows are workable during the June-to-September months when porpoise presence peaks. The Coastal Waters Alliance wants the windows written into berth permits at renewal. Industry representatives want a two-season voluntary pilot with published compliance data first. Neither side has moved off those positions in three meetings.
Replication is already under way. A partner lab at the inland university has agreed to re-run the detection classifier against a held-out third of the recordings without access to the original team's parameters. Morin asked for that arrangement herself, calculating that any municipal appetite for the work would evaporate the first time someone called it in-house science.
The next funding decision arrives in March, and it will turn partly on metrics and partly on narrative. Morin's wager is that a mixed result travels farther than a clean one.
She keeps a printout taped inside a cabinet door in the lab: the Station 4 spectrogram from a night in August 2024 when a generator failed at the grain terminal and the channel went unexpectedly silent for six hours. The detection trace climbs across it like a staircase. Eleven porpoises, in a stretch of water that had logged none for four nights running.
Reporting for this story was prepared for The Harbor Ledger’s science desk. Tips:newsroom@theharborledger.com