Traditional seismic methods for subsurface characterization include seismic refraction and spectral analysis of surface waves. Modern acquisition and processing techniques allow obtaining, more accurately, the bedrock irregularity. However, to cover large areas or acquire data in topographically complicated terrain, those methods turn out to be impractical, and the results are only a sample of the magnitude of the problem in the case of landslides. Ambient Noise Tomography (ANT) to characterize the subsoil structure has become popular in the last decade. The principle of the method is based on Seismic Interferometry (SI), the cross-correlation of recorded seismic noise to extract the so-called Empirical Green Function (EGF). A summary of historical background and various applications in science and engineering fields at different scales can be found in \cite{larose2015environmental} and \cite{schuster2016seismic}. For example, in the case of landslides, SI has been used to identify the extent of the elastic properties contrast between the soft materials and the bedrock \cite{renalier2010shear,pilz2014combining,chavez2021landslide}. Additionally, the coda of the EGFs has allowed monitoring the changes of forces before the potential mass motions \cite{mainsant2012ambient,del2013new,le2021landslide}.
Mexico City is a densely populated city. Its western side is topographically abrupt, and the geological risk is increased because the subsoil structure is composed of silt-sandy materials interspersed with clasts and tuffs, originated by pyroclastic-detritus flows and ash deposits susceptible to landslides. A small ecological-sustainable park is at risk of disappearing in that area due to landslides caused by underground runoff and leaks in drainage systems. The park is located on the bank of a piedmont that was reforested, bordered by a river, and a residential area where vegetable planting is carried out and the habitants use it as a recreational area (Figure 1). This work aims to determine the lateral extent of materials prone to landslides by analyzing velocity images obtained from surface waves produced by seismic refraction and ambient seismic noise.