Page 14 - April 2008

The temporary stations typ- ically consisted of arrays of three to five elements, with a spatial separation of roughly 100 to 300 m (Fig. 6). Teams were deployed to these recording sites one or two days before each experi- ment to set up equipment and record the stable, pre-event noise levels. A variety of acoustic transducers were used. These ranged from commercially available infrasound sensors, traditional laboratory grade microphones, and experimental transducers. While the detailed instrument responses varied somewhat between stations, all sites were capable of recording frequencies ranging from audi- ble to sub-audible (infrasonic).
Arrays provide significant
advantages over single sensors.
The multiple time-synchro-
nized recordings from sensors
distributed across an area can
be processed to estimate the azimuth of incident signals as well as their speed across the ground—parameters essential for evaluating atmospheric models. Combining multiple recordings also increases the ratio of coherent signal to inco- herent noise due to wind and thus can be essential for extracting weak signals from noise at the more distant sta-
Fig. 4. As for Figure 3, but for the experiment of March 25, 2006 (WSMR2).
tions or improving understanding of the signal structure at all distances. Wind speeds generally increase after sunrise, due to solar heating of the surface, and thus each of the six explosions were set-off before dawn.
Nearly all of the sites used some type of noise reduction mechanism (i.e., windscreens) attached directly to the transduc- ers. One of the simpler schemes is the use of porous hose (i.e., garden “soaker” hose) to provide a means for filtering out short wavelength pressure fluctuations (Fig. 7). However, multiple noise-reduction systems were employed. For the Optical Fiber Infrasound Sensors (OFIS) and dense microphone arrays, the increased spatial extent of the sensors themselves provides noise reduction as an integral element of the instrument design. In general, the diverse suite of stations operated well,
with few recording failures.
The explosions
Three experiments were carried out, in the fall of 2005 and in the spring and summer of 2006. Two explosive charges, of approximately 30 kg TNT equivalent, were launched and detonated during each experi- ment, with launches roughly
Fig. 5. As for Figure 4, but for the experiment of July 21, 2006 (WSMR3).
12 Acoustics Today, April 2008