How We Used Google's Perch v2 to Build a Bird-Based H5N1 Early Warning System in Patagonia

A small conservation foundation in Chilean Patagonia applied Google's Perch v2 bioacoustic model to 4,998 field recordings from a remote nature sanctuary — and detected all 6 sentinel seabird species across 49,383 acoustic events, including a Vulnerable endemic duck absent from previous model versions. The goal: an AI-powered One Health biosurveillance system for H5N1 and antimicrobial resistance in fjord ecosystems under pressure from salmon aquaculture.