Page 12 - April 2008

Fig. 1. Static sound speed profiles for WSMR2 (black line) and WSMR3 (red line) at 33.2oN, 106.5oW, near the center of the region in which the detonations took place (left). Zonal winds (positive from west to east) for WSMR2 (black) and WSMR3 (red) (middle). Right panel is same as for middle, but for the mean meridional wind speed profile (positive northward).
involving six rockets, each carrying a small payload of chem-
ical explosives. The rockets were launched from White Sands
Missile Range (WSMR) in southern New Mexico during
1,2,3
southwestern US to distances of nearly 1000 km.
The WSMR tests have provided a high-quality set of meas- urements of travel-time, signal amplitude and frequency to help address specific challenges in infrasound propagation modeling and source location. First, as increased numbers of infrasound events have been analyzed during the past decade, a systematic tendency to overestimate observed travel times
4
has been clearly identified. Data from the WSMR tests will
provide precise travel time data to address this issue. Second, the significance of internal wave scattering of acoustic energy in the stratospheric and thermospheric ducts has also been
5,6,7
Two rockets were launched during each of three WSMR experiments. The carefully tracked rockets flew a northward trajectory tens of kilometers into the strato- sphere, where the explosives were detonated. The resulting infrasonic signals were recorded at sites throughout the
2005-2006.
identified but is not completely understood.
Scattering is
often invoked to explain observations of energy leakage from
elevated ducts and possibly signals in some classic zones of
8
silence. The spatial coverage of the WSMR data provide a
means for direct observation of scattered acoustic energy.
Another challenge addressed through analysis of the WSMR
data includes a better understanding of thermospheric attenu-
9
ation. Finally, the WSMR experiments also provided an
opportunity to validate the scaling relationships between yield and dominant frequency as well as between yield and pressure amplitude for elevated sources. This article describes the gen- eral characteristics and preliminary results of the experiments. Experiment participants are preparing more detailed analyses of the large quantity of data collected.
Experiment design considerations
The scheduling of the experiments, as well as the geo- graphic distribution of the stations, was intended to maxi- mize the probability of observing signals under differing atmospheric conditions. Long-range infrasound propagation is primarily controlled by high-altitude winds and by the stat- ic sound speed that depends on the air temperature. Vertical gradients in the static sound speed and high-altitude wind profiles enhance or diminish atmospheric ducting between the ground and the lower, middle, and upper atmosphere, allowing infrasound waves to propagate to distances of hun- dreds to thousands of kilometers.
Tropospheric infrasound arrivals result from acoustic energy propagating in lower-atmosphere ducts. These ducts are a transient phenomenon involving temperature inversions in the lower atmosphere that may arise early in the day due to cool ground-level air temperatures or the tropospheric jet stream. Stratospheric arrivals, caused by ducting between the ground and stratopause, are significantly impacted by season- al variations in the zonal (east-west) stratospheric winds. In the northern hemisphere, these winds flow to the east in the winter and the west in the summer. Spring and fall are transi- tion periods. This feature results in directional ducting of the sound. For example, summertime conditions favor long-range
acoustic observations to the west of a source, but not to the east. Thermospheric arrivals, resulting from downward refrac- tion of acoustic energy by the steep sound speed gradients of the upper atmosphere, are more rarely observed due to high
9 acoustic absorption within the thin upper atmosphere. More
generally, the significance of natural atmospheric variability on infrasound propagation characteristics has been investigated
10,11,12
for the WSMR experiments a series of computations was per- formed. In Fig. 1, profiles of static sound speed as well as zonal and meridional (north-south) wind components are shown as a function of altitude for dates and locations corre- sponding to the second and third WSMR experiments (WSMR2 and WSMR3, respectively). These profiles are based on the Naval Research Laboratory Mass Spectrometer and Incoherent Scatter Radar Model-00/Horizontal Wind Model-93 (NRLMSISE-00/HWM-93) upper atmospheric
13,14
tations using the atmospheric profiles illustrated in Fig. 1. To highlight the direct arrivals and stratospherically ducted arrivals, only the lower 60 km of the atmospheric pro- files (Fig. 1) were used in the computations. Ray tracing for acoustic sound transmission in a windy environment relied on the physics governing acoustic refraction of rays in an
15
advected media. Rays were launched over a series of
azimuths and declination angles from the source point and the locations at which the rays intersect with the ground sur- face are marked by dots in Fig. 2, color-coded by time of arrival after the detonation. As shown, enhanced propagation
10 Acoustics Today, April 2008
and presented by several authors.
To evaluate the likely existence of stratospheric ducting
As shown, static sound speeds at the ground were predicted to be greater than those within the stratosphere for these dates, and one would not predict stratospheric ducting. However, the stratospheric winds must also be considered, and this was done via ray tracing compu-
empirical models.