Hi Bernie--
I looked into the physics of the variables you
raise. I'm assuming that Marc wants to locate the
Gibbons at a pretty good distance. According to
wikipedia,
http://en.wikipedia.org/wiki/Speed_of_sound
extreme differences in humidity between the mic
paths could create differences in the speed of
sound as much as .6%. At a distance of 2000
feet, that would be an error of 1.2 feet,
maximum. A 9=B0F difference in temperature between
the mic paths could change the speed of sound as
much as 3 meters per second and throw off the
measurement around 7 feet at a distance of 2000
feet. These seem to be workable variances.
>From my experience, the differences in the
landforms that shape the mic paths can actually
aid in localization by adding unique, consistent
complexities to each signal. With three mic
monitoring, you might not have to triangulate or
compute the location every time; one might learn
to recognize the timing difference/acoustic
difference patterns of frequent calling locations.
3-D localizing would raise some serious
challenges-- especially in very hilly/mountainous
settings. I suppose the X-Y lateral mics need to
be at the same altitude,.. wouldn't the Z mic
need to be a considerable distance above the
canopy to provide enough reference? A ridge
towards the middle of the triangle might be
beneficial but not so high as to obstruct distant
sounds. Rob D.
At 8:31 AM -0800 12/17/08, Bernie Krause wrote:
>Because gibbons typically call from the canopy and live in a 3-D
>environment (as opposed to a plane), Rob, one would need to set up a
>tetrahedral (pyramid w/ all mics equidistant from one another) in
>order to approximate location. But because the rainforests where they
>reside (Sumatra and Borneo in my experience) are so reverberant, one
>would get complex arrival times at each mic obviating precise analysis
>of location. To further complicate matters, the acoustics of these
>biomes are in constant flux, changing rapidly with the slightest shift
>in temperature or humidity or time of day. That said, if one has taken
>careful measurements over time under different conditions, one might
>be be able to approximate both distance and location assuming well-
>calibrated gear that one can rely on under those conditions.
>
>Bernie
>
>On Dec 17, 2008, at 8:20 AM, Rob Danielson wrote:
>
>> At 7:51 PM -0500 12/16/08, Marc Myers wrote:
>> >Anyone have experience locating animals by measuring distance and
> > >direction of their calls in natural environments? Gibbons often live
>> >in difficult terrain, at low densities and in small family groups.
>> >They can be hard to see in their native environment and are
>> >critically endangered through the majority of their range. Making
>> >population estimates is important in conservation planning, when
>> >advising local governments on forestry use and establishing risks to
>> >extinction. Gibbons have famously loud and expressive calls,
>> >typically sounded first thing in the morning. Populations are
>> >presently estimated based on the frequency, direction and apparent
>> >distance of the calls. Of course even experienced listeners are not
>> >very good at estimating distance and as a consequence population
>> >estimates vary wildly from environment to environment and researcher
>> >to researcher. Any thoughts?
>> >
>> >
>> >_
>>
>> Hi Marc--
>> Here's another "wild" scheme. If pinpointing where the calls/animals
>> are located is really important, one could record with 3 omni mics in
>> a large, equilateral triangle layout. I'm not sure what distances
>> between the mics to recommend but 500 feet or more would probably
>> provide enough accuracy to identify their favorite calling spots in
>> audible range with certainty, maybe even determine that animals are
>> on the move. You'd need to know the exact distances between the mics
>> and record all three tracks at once. Arrival time differences are
> > easier to measure (and hear) with percussive calls than sustained
>> ones and that might be an drawback with Gibbons. One could use
>> wireless transmitters (a hit, quality-wise) but an option if running
>> wire isn't possible. Rob D.
>> --
>>
>>
>>
>
>Wild Sanctuary
>POB 536
>Glen Ellen, CA 95442
>707-996-6677
><http://www.wildsanctuary.com>http://www.wildsanctuary.com
><chirp%40wildsanctuary.com>
>Google Earth zooms:
><http://earth.wildsanctuary.com>http://earth.wildsanctuary.com
>SKYPE: biophony
>
>
>
--
|