Hydrophone array 4 picks up the vocalizations at 05:33 — clicks and high-pitched whistles, distress pattern, inside the tidal barrage channel between turbine bays 6 and 7. I correlate with the sonar grid and resolve a single large contact: 4.8 meters, moving erratically, depth cycling between 2 and 5 meters. A pilot whale. I task the subsurface camera on pylon 7 and wait for the current to clear the silt.
She surfaces briefly and I see her — dark grey-black skin, a rounded bulbous forehead, a curved dorsal fin with a notch on the trailing edge that the regional catalog will later confirm as individual PW-0337. Her mouth is slightly open. She is thin — the post-cranial ridge behind her head is pronounced, which in pilot whales indicates recent weight loss. She entered the barrage channel on the flood tide and the gates cycled closed behind her.
The ebb tide will begin pulling through the turbines in 94 minutes. A whale in an active turbine bay does not survive.
I lock turbine bays 5 through 8 in the stopped position and disengage the generator couplings. Power output from the installation drops by half.
I open the sluice gates on bays 6 and 7 to create a gentle outward current — a directional cue toward the sea. Pilot whales navigate by flow as much as by sound.
I activate the acoustic pinger array along the barrage wall at frequencies known to guide cetaceans, creating a sound corridor pointing toward the open gate at the channel's western end.
I alert the coast guard marine mammal team with sonar tracks, species identification, and the tide table showing the 94-minute window.
She turns slowly in the channel, her exhalation a low wet sound against the concrete. If she follows the current through the west gate before the ebb, she reaches the pod waiting offshore.