to conduct active source and ambient noise seismic acquisitions (Figure 1). From active source records we produce in-depth images of P wave velocity (Vp) and S wave apparent velocity (Vs) using seismic refraction and spectral analysis of surface waves (SW) methods, respectively. First, the study area was discretized with cells proportional to receiver separation. Then, in the Vp image case, the procedure makes a linear adjustment of refracted time arrivals that cross each cell. As a result, the slope inverse is Vp, and the intercept allows an estimation of the bedrock depth \cite{cardenas2022pseudo}. In the case of Vs, we calculate dispersion curves \cite{herrmann2013computer} of all those records with a source-receiver distance greater than 15 m, and we use the group velocity times to elaborate tomographies in the frequency range of 6 to 24 Hz \cite{rucker2017pygimli}. Subsequently, we use ANT to build Vs images in the same frequency range. To do this, we normalize the records by 1-bit and spectral whitening \cite{bensen2007processing}, and conduct cross-correlations between all pairs of receivers in 8s time windows over 30 min to obtain the so-called EGFs \cite{shapiro2004emergence}. Subsequently, we stack acausal and causal parts of these functions and calculate group velocity dispersion curves for inter-distances greater than 15 m. Again,
Results
Preliminary results were obtained using seismic refraction and electrical resistivity tomography. The bedrock was found at an average depth of 10 m. The materials susceptible to sliding correspond to a filling deposit, as shown by the spectral H/V ratio with a relative amplitude of 6 in 6 Hz. Furthermore, the ERT section shows that the entire subsoil is highly saturated.
One way to cover a more significant area extension is by seismic tomography. We took advantage of a semi-closed array in the study area to create a Vp tomography of refracted arrivals and another of Vs using dispersion curves, both active source tomographies. For the case of the Vp image, we select the refracted arrivals and represent their velocity and intercept time in a discretized model with cells proportional to the separation of receivers. In the case of the Vs image, we reproduce it by obtaining the dispersion curves between pairs of receptors, using the same discretized model. The figures show the P-wave tomography, which reflects the velocity and irregularity of the bedrock (approximately 1500 m/s and 8 m depth). The S-wave image is at 24 Hz, where low-velocity zones can be observed at the edges of the array.
Figure 1 shows a section of seismic refraction on the most stable side of the park. Deposits of soft materials are observed at the center of the line with Vp values less than 400 m/s (typical velocities of weathered materials). The line extension allows defining a second layer with an irregular structure with poorly consolidated materials (Vp=700 m/s). Vp values greater than 1200 m/s can be associated with the substratum at depths greater than 10 m. Figure 2a shows the selection of the first arrivals of all refraction records using 24 sources and 48 receptors. Direct arrivals show that Vp in the first layer is approximately 400 m/s. Refracted arrivals (after a critical distance of 20 m) exhibit large dispersion, indicating the substratum is irregular with a Vp average of 1200 m/s. Figure 2b shows the velocities representation on the discretized model. The average depth of the substratum is 8 m. A higher velocity zone is observed in the northern part of the array, between 8 and 10 m depth (according to the refraction section shown in Figure 1), with Vp reaching up to 3000 m/s.