My listening has been on the mainland but gibbons do elect some tough envir=
onments, probably because that's all we've left them.
That said, I'm looking more for consistency of measurement and reducing err=
ors than really accurate data analysis and the highest quality recording. S=
tandard practice at present is to go to fixed point at dawn for two or thre=
e days in a row and then estimate the number of gibbons based on the calls.=
Clearly not good enough.
A multichannel recording using a microphone array then later analyzing the =
recording in a lab occurred to me as well. I also thought of a variant, usi=
ng multiple Zaxcom recorders. They can be locked to time code and then move=
d fairly far from one another. Geolocate each with a GPS and one would be g=
ood to go. The microphones father apart would be much easier to analyze and=
the number of points available to evaluate would be limited only to the nu=
mber of recorders one had available in the field. But, any such a system th=
at's based on triangulating based on the relative delays among the micropho=
nes would be hard to interpret in the field and will be replete with just t=
he kinds of problems Bernie has elucidated. Also field surveys are usually =
budget operations.
I'm thinking a little lower tech: Two listening stations with very directio=
nal microphones and head phones or maybe just a passive listening tube on a=
tripod, an attached GPS with electronic compass, some field notebooks and =
walkie-talkies. Gibbons start calling: Two observers locate the calls, coor=
dinating with walkie-talkies. Using the compass and GPS location a bearing =
is taken on the call. The results can be analyzed with a map or GIS softwar=
e. Thoughts?
|