Must be periodically moved or geese realize they are not a threat.
Geese eventually figure out the difference between decoys and the real thing.
Certain decoys can disrupt human enjoyment of the space.
Can be effective at chasing away geese.
Dogs are brought on site for short period(s) of time. Geese may visit at other times.
Dogs need to be trained to avoid harm to themselves and others.
Dogs need to be cared for by humans.
Can be effective for short periods of time.
Designed to be put on a goose's food source. The problem area may be where geese roost, not
where
they eat.
Need to be reapplied as the food source grows.
Can be washed away by rain.
Unknown long term effects to geese, humans, other animals, and the environment.
Systems tend to be dumb. They can NOT tell the difference between motion caused by geese,
humans,
or trees blowing in the wind. They trigger deterrents when they should not.
People disable systems when they are tired of misfires.
People disable systems when they use the area, then forget to turn them back on.
Systems tend to be coupled with deterrents like sprinklers, flashing lights, recorded noises,
...
Geese become accustomed to certain deterrents.
Certain deterrents may not be "well received" by neighbors.
Systems need to be coupled with deterrents like sprinklers, flashing lights, recorded noises, ...
Systems tend to be good at identifying people; OK at identifying geese.
Not all models produce satisfactory results.
System performance should be evaluated periodically.
Models should be retrained periodically.
AI and IR are not enough; pre and postprocessing logic will greatly improve system results.
