I created the “Bird Identification” model (and
prototype database) a few years ago. It was inspired by bird field guides, but
with the idea that a GIS (Geographic Information System) and GPS (Global
Positioning System) on a PDA (Personal Digital Assistant), a field biologist
could get much more specific clues for identifying wildlife. Technically
this is all quite possible, but managing the associated data is a large task: assembling
sighting statistics and land cover and translating these into probabilities of
finding wildlife.
The concept is this: You are in the field. The GPS will tell
(within a few meters) where you are and give accurate date and time.
The GIS will identify the habitat, thus reducing the number of species you are
likely to see. The time of day and season will further constrain (or possibly
expand, in the case of migrating species) the species list. The historical
sighting statistics will allow the software to make a list in order of probable
sighting. Anatomical characteristics (such as “red head”, "long tail") can further restrict
the possibilities. Photos and flight videos (for birds, anyway) help positive
identification. “Illustration” could even be expanded to include voice
recordings, and the microphone on your cell phone could pick up the bird song
or call, with software to compare it to the recordings for probable species.
This is all wrapped up with a record of actual sightings in
the “Observation” section of the model.