We describe an experiment conducted to assess the impact of the sound generated by an acoustic harassment device (AHD) on the relative abundance and distribution of harbor porpoises (Phocoena phocoena) in Retreat Passage, British Columbia. During control periods when the AHD was inactive, the mean number of porpoises observed in the study area was 0.39 for broad area scans conducted with the naked eye and 0.48 for narrow sector scans conducted with binoculars. Abundance declined precipitously when the AHD was activated, to 0.007 porpoises per broad area scan and 0.018 per narrow sector scan. The mean number of porpoise resightings while tracking their movements also declined from 12.2 to 13.6 per sighting during control periods to 1.1–1.9 per sighting when the AHD was activated, which suggested that the few porpoises that ventured into the study area spent less time within it when the AHD was activated. The effect of the AHD diminished with distance. No porpoises were observed within 200 m of the AHD when it was activated. The number of sightings and resightings observed when it was activated was less than 0.2% of the number expected had there been no AHD effect at a range of 200–399 m, 1.4% the number expected at a range of 400–599 m, varied between 2.5% and 3.3% of the number expected at a range of 600–2,499 m, and was 8.1% the number expected at a range of 2,500–3,500 m, which suggested that the impact of the AHD extended beyond our maximum sighting range of 3.5 km.