WIT - The Witteveen Infrasound Array

Infrasound has been measured at KNMI's geophysical observatory in Witteveen for over six years. The infrasound array was reconstructed during 2003. Figure 1 shows the installation of the instrument's housing. KNMI microbarometers (mb) were installed to replace previously used electret microphones.

Figure 1: The installation of pvc tubes for housing the KNMI microbarometer.

Wind noise reduction is achieved by connecting porous hoses to each microbarometer. The final installation of one of the microbarometers is shown in Figure 2. Six porous hoses are attached in a spider-like configuration. Closed hoses are used for the first meter to sample not too close to the instrument.

Figure 2: The final installation of the microbarometer situated inside the pvc tube. Porous hoses in a spider-like layout are used for wind noise reduction.

The old array constisted of six elements placed two by two on the sides of an equilateral triangle. The inter-station distance along each side of the triangle was 12.5 meters. This array was replaced by a pentagonal array with a central element (six elements). Figure 3 shows the layout of the pentagonal array with an aperture of 37.5 meters.

Figure 3: The layout of the six mb's configuring WIT.

Figure 4 shows the response of the array to a monochromatic plane wave of 4 Hz. The circular form of the response represents the ability of WIT to uniformly sample the atmosphere.

Figure 4: The response of WIT to a 4 Hz plane wave.

A sonic event occurred soon after the construction of the array was finalized. Figure 5 shows the infrasound measurements done by the six microbarometers of WIT on 2003, May 15. The coherent infrasound makes the traces almost look identical. The origin of the energy is probably a sonic boom from a military fighter, based on the waveform.

Figure 5: The recordings of infrasound by the six microbarometers of WIT. The time axis gives time since 11h46m35.903s Local Time on 2003, May 15. The data are bandpass filtered with a second order Butterworth filter having corner frequencies of 0.5 and 15 Hz.

The data are analyzed on the basis of signal coherency and the travel time differences over the array. A scaled measure of signal-to-noise ratio is used to detect event and is called the Fisher ratio. Figure 6 displays the results of the analysis. The sonic boom is clearly detected and found to have travelled from a back azimuth of 325.9 degrees with an apparent sound speed of 356.9 m/s.

Figure 6: Fisher analysis of the infrasound data. The Fisher ratio, a scaled signal-to-noise ratio, increases around the time of the event (lower frame). This indicates the presence of a coherent signal travelling over the array. Event characteristics of 356.9 m/s for the apparent sound speed and a back azimuth of 325.9 degrees are resolved at maximum coherency, around 126 seconds. The top frame shows the best beam for the event. The best beam is the sum a the time aligned traces.

Infrasound from the fighter was also detected in the De Bilt Infrasound Array (DBN) and the Deelen Infasound Array (DIA). The source can be localized by cross bearing the back azimuths obtained in the three infrasound arrays. Figure 7 shows the location of the fighter.

Figure 7: Map showing the location of the fighter as found by cross bearing the back azimuths of the three Dutch infrasound arrays.

May 2003
Läslo Evers