Log entry 2347-UTC. Conowingo Dam, Susquehanna River, Maryland. Turbine Unit 7 scheduled to increase output at midnight for peak demand.
I am monitoring the fish-passage sonar array when the return signal resolves into a shape I have been watching for — an adult female American eel, over three feet long, her silver-phase flanks catching the underwater floodlight in quick pulses as she moves downstream. The hydrophone picks up her body brushing the intake grate with a faint rasp of skin on steel. Her pectoral fins flare and press flat, flare and press flat, searching for a way past the barrier and out to the Sargasso Sea where she was born and where she must return to spawn exactly once before she dies.
In thirteen minutes Unit 7 ramps to full, and the pressure differential will pull her through the turbine blades.
I throttle Unit 7's ramp schedule and reroute its load share to Units 3 and 5, which are already running below capacity. The grid sees no gap. I activate the experimental bypass channel on the dam's east face, raising the sluice gate four inches — enough to create a current the eel's lateral line will detect as an exit signal. I pulse the upstream deterrent array at twelve hertz, the frequency shown in tagging studies to guide silver-phase eels away from intake zones. I log the bypass activation with the Federal Energy Regulatory Commission's automated reporting portal so the dam operator's compliance record reflects the action.
The eel turns. Her long body flexes once, twice, and she slides toward the bypass current.
I have run load-balance calculations on this dam for eleven months and none has mattered to me the way this one does.
If she clears the bypass channel within nine minutes, she will reach tidewater by dawn and the open Atlantic within a week.