GNGTS 2019 - Atti del 38° Convegno Nazionale

696 GNGTS 2019 S essione 3.2 SURFACE-WAVE TOMOGRAPHY AT MINING SITES FOR P-WAVE STATICS M. Papadopoulou 1 , S. Hu 2 , L.V. Socco 3 1 Department of Environment, Land and Infrastructure Engineering (DIATI), Politecnico di Torino, Torino, Italy 2 School of Ocean and Earth Science, Tongji University, Shanghai, China 3 Inserire x Socco Introduction. Seismicmethods are becomingmore popular inmining exploration (Malehmir et al. 2012). Therefore, the development of cost-effective processing methods suitable for these sites, is important. A critical step in the seismic processing workflow are the static corrections, which are applied to the seismic data to correct for the effect of the near-surface and improve the imaging of the deeper exploration targets. Typically, body-wave tomography is used to estimate the statics. A different, less popular approach is the use of surface waves (SW) which are usually treated as noise in exploration. However, it has been shown (e.g. Foti et al. , 2014) that their sensitivity to the near-surface properties can lead to high accuracy in the statics estimation. Moreover, since SW already exist in the exploration, no extra, costly data acquisition is required, increasing the value of SW methods to industrial applications. The need for accurate statics is even more significant in mineral exploration sites, where the existence of various near-surface heterogeneities, such as lithological variations (Eaton et al. , 2003), fractures, faults and man-made structures is common (Hollis et al. , 2018). In these sites, a suitable SW method is SW tomography; due to its high data coverage and multiple illumination, it can provide superior results in terms of resolution, compared to the multichannel SW methods, typically applied in exploration. The method has been developed and widely applied in global seismology and it has been shown that it can efficiently provide the V S model at exploration scale. However, it is rarely used for V P estimation, due to the lower sensitivity of SW dispersion curves (DC) to V P . A technique to overcome this limitation is the wavelength – depth (W/D) method (Socco et al. , 2017), a multichannel SWmethod which has been proven effective in V P estimation (Socco and Comina 2017). Here, we propose a workflow that combines SW tomography with the W/D method to obtain the P-wave statics. We apply the workflow to a 2D synthetic model and show that the statics are retrieved with great accuracy. Method. Both SW tomography and the W/D method use a reference multichannel DC, which represents the dispersion characteristics of a laterally homogeneous subsurface area. If lateral variability is present, more than one reference DCs are necessary. The reference DCs are extracted by applying a wavefield transform on windows of receivers and picking the energy maxima in frequency domain. The W/D method uses only the reference DCs and its corresponding time-average V S (V Sz ) profile, which can be obtained by inverting the reference DCs for 1D V S profiles and transform them into V Sz , according to: (1) where V Sz is the time-average V S at depth z, n is the number of layers down to depth z and h i and V Si are the thickness and V S of the i -th layer, respectively. For each reference DC, the method finds couples of wavelength and depth, for which the phase velocity of the DC is equal to the velocity of the V Sz profile, providing a relationship between the propagation wavelength and the investigation depth of SW (Wavelength – Depth relationship or W/D). The W/D is sensitive to the Poisson’s ratio, and applying a sensitivity analysis, the apparent Poisson’s ratio (υ z ) of the different homogeneous areas can be estimated. For SW tomography, the necessary input is a number of path-average DCs, which are extracted from the seismic data with the two-station method (Yao et al. , 2008); the recordings