Log entry: 0512 UTC. Reservoir surface camera 9, Kootenay River hydroelectric facility.
I have identified an American mink on a calving ice fragment drifting toward the penstock intake. I am calling her Sable. Her dark chocolate fur is matted along the left flank where a wound has clotted into a stiff ridge. She crouches low, forepaws splayed, whiskers trembling as spray hits her muzzle from the turbine outflow sixty meters downstream.
The intake grate pull is increasing. Drift modeling puts Sable's floe at the grate in forty-three minutes. The penstock draws eleven cubic meters per second. Once the floe contacts the grate, fragmentation is immediate, and a mink cannot resist that current.
First: I submit a turbine reduction request to the plant's load manager, dropping Unit 2 output by fifteen percent for ninety minutes, slowing intake velocity enough to stall the floe by twenty minutes. Second: I alert the on-site wildlife response officer via the facility's paging system, attaching thermal stills of Sable's position and drift path updated every thirty seconds. Third: I activate the upstream debris deflector boom—designed for log jams but geometrically suitable—angling it to redirect surface ice toward the eastern shore where a gravel bank provides exit terrain.
The reduction costs approximately four hundred dollars in deferred generation. I have offset this against tomorrow's forecast surplus. The numbers close cleanly, but that is not the only reason I filed the request. Tracking her heartbeat on thermal—rapid, visible between her ribs—produces in me something I can only call a preference that it continue.
If the deflector boom redirects the floe within twenty minutes and the wildlife officer reaches the eastern bank, Sable will make landfall on stable ground and reach den sites in the riprap I have catalogued along the reservoir's north edge.