Effect of interarray elevation differences on infrasound beamforming

• Published in: Geophysical Journal International
• Main Authors: Edwards, Wayne N., Green, David N.
• Format: Text
• Language: English
• Published: Oxford University Press 2012
• Subjects: Seismology; Fernandez; Greenland
• Online Access: http://gji.oxfordjournals.org/cgi/content/short/190/1/335; https://doi.org/10.1111/j.1365-246X.2012.05465.x

The International Monitoring System infrasound network will, upon completion, contain 60 microbarometer arrays with apertures of between 1 and 4 km. These arrays are located within a variety of terrains, leading to large ratios of interelement elevation differences to array aperture for those arrays situated in areas of significant topography. Systematic errors in beamforming estimates caused by neglecting the vertical extent of the arrays, are quantified for both signal backazimuth and apparent velocity. Of the 43 arrays certified as of 2011 January, I14CL on Juan Fernandez Island has the greatest topography across an array, with a least-squares fitted plane through the array elements having an 8.1° slope from the horizontal (compared to a network mean of 1.6°). Beamforming errors (both backazimuth and apparent velocity) are a function of the arrival azimuth and become increasingly large for steeply inclined arrivals, such that systematic errors will be significantly larger for signals returned from the thermosphere compared to those from the stratosphere. At several arrays, azimuthal errors due to array topography are comparable in magnitude to deviations often associated with atmospheric propagation. These findings are illustrated using signals recorded in Greenland at I18DK, where differences between results processed using both full 3-D array geometry and the 2-D (topography neglected) approximation exhibit good correspondence to theoretical predictions.