This article presents a new wireless seismic sensor network system, especially design for building monitoring. The designed prototype allows remote control, and remote and real-time monitoring of the recorded signals by any internet browser. The system is formed by several Nodes (based on the CC3200 microcontroller of Texas Instruments), which are in charge of digitizing the ambient vibrations registered by three-component seismic sensors and transmitting them to a central server. This server records all the received signals, but also allows their real-time visualization in several remote client browsers thanks to the JavaScript’s Node.js technology. The data transmission uses not only Wi-Fi technology, but also the existing network resources that nowadays can be found usually in any official or residential building (lowering deployment costs). A data synchronization scheme was also implemented to correct the time differences between the Nodes, but also the long-term drifts found in the internal clock of the microcontrollers (improving the quality of records). The completed system is a low-cost, open-hardware and open-software design. The prototype was tested in a real building, recording ambient vibrations in several floors and observing the differences due to the building structure.

The reactivation of very large landslides may cause severe damage to society. Its prevention and management requires detailed information on the geometry and structure of these landslides, but the use of standard techniques (boreholes) may be prohibitive from an economic point of view. To overcome these difficulties, geophysical techniques are of special interest because they allow for studying very large areas at a reasonable cost. In this paper, we present a case study wherein the analysis of ambient noise allowed us to produce a model of a large landslide near Granada (southern Spain). The geometry and location of the failure zone, as well as the assessment of the state of involved materials, were estimated by combining two available boreholes and different geophysical techniques (downhole tests and the spectral analysis of ambient noise, horizontal to vertical spectral ratios (HVSR) and the frequency-wavenumber (f-k) methods). The results have allowed us to differentiate between values within the landslide mass with respect to those of stable materials, and to perform for the first time a comprehensive geological model of this unstable mass. Differences were also observed within the landslide mass (earth flow vs. slide zones), which are attributed to differences in the degree of alteration and the disturbance of the internal structure of materials constituting the landslide mass. These results show that techniques based on the measurement of ambient noise are of special interest for studying very large, highly remolded landslide masses.

The analysis of seismic noise provides a reliable estimation of the soil properties, which supposes the starting point for the assessment of the seismic hazard. The horizontal-to-vertical spectral ratio technique calculates the resonant frequency of the soil just by using a single three-component sensor. Array measurements require at least several vertical sensors registering simultaneously and their analysis provides an estimation of the surface waves dispersion curve. Although these methods are relatively cheaper than other geotechnical techniques, the cost of the sensors and the multi-channel data acquisition system means that small research groups cannot afford this kind of equipment. In this work, two prototypes for registering seismic noise have been developed and implemented: a three-channel acquisition system, optimized for working with three-component sensors; and a twelve-channel acquisition system, prepared for working simultaneously with twelve vertical geophones. Both prototypes are characterized by being open-hardware, open-software, easy to implement, and low-cost. The main aim is to provide a data acquisition system that can be reproduced and applied by any research group. Both developed prototypes have been tested and compared with other commercial equipment, showing their suitability to register seismic noise and to estimate the soil characteristics.

The characterization of soil is essential for the evaluation of seismic hazard, because soil properties strongly influence the damage caused by earthquakes. Methods based on seismic noise are the most commonly used in soil characterization. Concretely, methods based on seismic noise array measurements allow for the estimation of Rayleigh wave dispersion curves and, subsequently, shear-wave velocity profiles. The equipment required for the application of this technique is usually very expensive, which could be a significant economic challenge for small research groups. In this work, we have developed a wireless multichannel seismic noise recorder system (Geophonino-W), which is suitable for array measurements. Each station includes a microcontroller board (Arduino), a conditioning circuit, an Xbee module, an SD card, and a GPS module. Several laboratory tests were carried out in order to study the performance of the Geophonino-W: A frequency response test (impulse response and noise); synchronization test; and battery duration test. Comparisons of Geophonino-W with the commercial systems and field measurements were also carried out. The estimated dispersion curves obtained using the proposed system were compared with the ones obtained using other commercial equipment, demonstrating the effectiveness of Geophonino-W for seismic noise array measurements. Geophonino-W is an economic open-source and hardware system that is available to any small research group or university.