Infrasonic wave propagation through the atmosphere
Introduction
Following the realization of the Deelen Infrasound Array (DIA), we carried out a study about wave propagation through the atmosphere,
in order to characterize and identify the events recorded by the DIA. We investigated the met orological parameters of interest and we used
a ray-tracing technique to try to explain the recorded data. We found that in some cases the method is very useful, and in other cases it
is not.
The tau - p method of Garces et al. (1998) is used to perform the ray-tracing through the atmosphere.
The use and advantages of this method are illustrated through an example from seismology and its application in wave propagation through
the atmosphere is given afterwards.
Using the tau - p method, we can calculate the time, horizontal range, and transverse range
of the arriving waves as a
function of ray parameter, under the influence of the meteorological parameters, temperature and wind vector. The complete path traveled
by the ray can be constructed, and gives information about the existence of caustics, which results in (de-)focusing of energy.
The recordings of the DIA from infrasound generated by an exploding meteor are used to determine the source location. Using the appropriate
atmospheric models for that particular day and location, we calculated traveltime curves for sources at different heights and compared
these with the data. We recorded two arriving phases, with a differential time around 55 seconds. Two possible solutions were found.
For source heights around 17 km, differential times are between 47 and 66 seconds, resulting
in source distances between 324 and 327 km.
The first arrival is a low thermospheric return for rays traveling upward from the source, the
second arrival is a low thermospheric
return for rays which are first reflected by the surface. For sources around 30 km, differential times are between 50 and 61 seconds,
resulting in source distances between 309 and 318 km. The first arrival is a low thermospheric return, the second a high thermospheric
return, both for rays which are first reflected by the surface. Deviations between the observed and actual required angle to trace back
from receiver to source, lie between 1 and 3.7 degrees, in both cases.
Data from the DIA of the S.E. Fireworks explosions cannot be explained by this method. The array was clearly in the shadow zone, and the
energy recorded was diffracted into the shadow zone due to meteorological features not present
in our models. Also, ray-tracing
cannot deal with diffraction in general. Data from the Freyung
Array are partly explained. Some of the calculated phases were not present in the data, because of the poor array response and high noise
level. Calculated and recorded arrival times differ around 110 seconds. Approximations, assumptions, and defects in the theory together
with possible solutions for the errors in the results are discussed. Topics of possible future
research are given afterwards.