Shallow Microtremor Array Survey Using Miniature and Small Arrays: Strategy for Efficient and Feasible Dense Survey

Abstract We investigated the applicability of miniature microtremor arrays with a radius of a few meters or less to shallow surveys (up to a few tens of meters). It is shown that the upper limit wavelength normalized by the seismometer separation distance s that is analyzable by a miniature array does not depend on the observation instrument as long as the instrument has a self‐noise level that is sufficiently lower than the microtremor intensities; however, it generally depends on the average S‐wave velocity of the ground. This means that a miniature array is useful at soft‐soil sites but not hard‐soil sites. A statistical study in central Japan showed that the penetration depth by a miniature array with s = 1 m ranges from 6 to 12 m; specifically, the depth exceeds 12 m in one quarter of the cases but remains below 6 m in one quarter of the cases. This large variation is due to the variation in the average S‐wave velocity. A miniature array should thus be used in combination with a larger array with an s value of several meters to 20 m. In urbanized areas with high industrial activity, where the microtremor wavefield is likely isotropic, these arrays can be replaced by a linear array. In environments with extremely low signal‐to‐noise ratios, a less efficient zero‐crossing method should be applied to the larger array. An appropriate observation strategy should be selected for a given geoenvironment. Field examples show that our strategy enables efficient and feasible dense surveys.