Ambient noise monitoring

     The objective of ambient noise monitoring is to determine the dominant site frequencies throughout Dubai. Ground noise contains important information on the dynamic characteristics of sub-surface soil layers. In this study we have used ambient ground noise to determine the dominant site frequencies, but not the site amplification. Actual site amplifications are determined numerically from the one-dimensional soil models developed from the geotechnical borehole logs and calibrated with the results of ambient noise analysis. The calibration involves modifying the shear wave velocities of the models so that they have the same predominant frequency as the recorded ground noise.

     To record ambient ground noise, we have used 10 sets of portable broadband seismic recorders with built-in three-component acceleration sensors Figure 1. The equipment used in the investigation are the Q330 digital recorder and the Epi Sensor. The city is divided into four regions, as shown in Figure 2, and deployed 10 instruments in each region to record ambient ground noise continuously over a month (example Figure 3).

     The predominant site amplification periods in Dubai are determined by recording and analyzing ambient ground noise from 40 locations throughout the city. The records were analyzed by using the standard H/V technique (Figure 4), as well as more advanced optimal filtering techniques (Figure 5). The analysis is aimed to calculate the predominant site periods, but not the site amplification. The results indicate that the soil in Dubai is fairly stiff in most places, with dominant periods around 0.05 second. The soil is slightly softer around Dubai Creek area with periods around 0.07 second. The calculated periods are used to calibrate the one-dimensional site amplification models developed from the geotechnical logs by manipulating the shear wave velocities of the layers so that the calculated periods match the recorded ones. The actual site amplifications are determined numerically from these soil models.