Comparison of point and array-computed rotations for the TAIGER explosions of 4 March 2008

Abstract

Two large explosions were recorded by a dense array of strong-motion accelerometers and rotational seismometers in northeastern Taiwan associated with a Taiwan Integrated Geodynamics Research long-range refraction experiment. The objective of this experiment was to test the response of the experimental eentec rotational seismometers against calculated array rotations. Computed array rotation rates are seen to have little variation across the array, but point rotation rate measurements obtained from individual rotational seismometers show significant deviations with each other and with the array rotation rates in the ranges of 3–5 Hz and, especially, 3–50 Hz. A cross-correlation method was used to compare array-computed rotation rates and point rotation rate measurements in the frequency band of 3–5 Hz with the result that the absolute value of the normalized maximum correlation coefficient for each station set varied from 0.45 to 0.97 with an average of 0.84. Amplitude differences of the point rotation rate measurements are seen to be factors of 0.2 to 1.8 times the array rotations as well. It is not likely that the differences seen in the point and array-computed rotation rates are due to nonlinear or heterogeneous site conditions under each array element since these effects should also be seen in the acceleration data used to determine rotation rate. A rigorous method for accurately calibrating rotation rate instruments is needed to understand their response in the field.