Calculation Techniques to Acquire Displacement from Seismic Acceleration Records

Computing displacement from measured acceleration is important for understanding the characteristics of an earthquake. However, to obtain such displacement with a good accuracy is difficult. It is partly because the low-frequency component dominating the behavior of the displacement is often affected by the sensor’s tilt and the measurement error. To avoid these, a high pass filter is usually applied; however, since this operation often distorts the acceleration record itself, we should apply it in a careful manner. In this paper, one methodology to obtain displacement from seismic acceleration records is hereby proposed. This method is composed with two methods; one to identify and remove the sensor's tilt motion, and the other to integrate a record by using only a real part of the Fourier transformation in consideration of causality. The adequacy of this technique is confirmed by applying the method to estimate the ground displacement of 2008 Iwate-Miyagi Inland Earthquake.