67 3. Seismic tomography • Receiver: vertical geophone, Oyo GS-14 Hz, with 5 m spacing, and 10 m receiver line spacing; • Number of receiver lines: 3; • Seismograph: 96 channels; • Record length: 1 second; • Swaths: 1 using receiver lines 1-3; 2 using receiver lines 2-4. 3.1.2 Tomographic methodology Figure 3.3 provides the workflow for the transmission tomography algorithm of Mendes (Mendes, 2009) used to produce tomographic images of the subsurface, enabling a structural evaluation of the upper epikarst. Figure 3.3 Simplified workflow of the global inversion scheme. The transmission tomography methodology utilizes the picking times of first arrivals and a simplified initial velocity model to produce a more detailed velocity model for the epikarst region. The basic features of this algorithm are: • gridded model; • SIRT back-projection technique for iterative inversion; • Fresnel volume: which is the area formed by the points around the geometric ray delayed by less than half of the period of the dominant wave; • fat-ray: represents the wave path from source to receiver, with a width defined by the points belonging to the Fresnel volume. The choice of the most appropriate dominant period of the input wave, not only depends on the characteristic period of the source wavelet, but also on the scale of the experiment. A useful “rule of thumb” for choosing the input period, T, suggested by Jordi et al. (2016) is estimated by the ratio T = 0.1 × H/V, where H is the target depth of the survey and V the average of the expected subsurface velocities.
RkJQdWJsaXNoZXIy NjA3NzQ=