Search

Detecting underground voids, such as old mine workings, solution cavities in karst terrain, or unknown tunnels such as illicit cross-border tunnels, is a challenging problem for geophysics and an important concern for geotechnical design, public safety, and domestic security. Seismic surface-wave-based detection methods have become increasingly popular for detecting relatively shallow and small targets; however, the theoretical limitations of these methods have thus far remained unclear. We use a suite of 3D numerical simulations inspired by a tunnel detection experiment carried out at the Black Diamond Mines Regional Preserve (BDM) in northern California. The geophysical anomalies predicted by our numerical simulations at BDM agree with field observations, and our estimates for the location of the primary tunnel target agree with historical records in the area. Using our calibrated numerical model, we perform a parametric study to determine the effect of tunnel size, depth of burial, filling material, and source characteristics to determine the range over which surface-wave backscattering and attenuation-based methods are effective. In addition, we perform a regression analysis to determine a relationship for the maximum depth at which a tunnel may be detected via these approaches, given the target diameter, wavelength of interest, and the signal-to-noise ratio.

To better understand the relationship between P-wave velocities and ice content in saturated, unconsolidated saline permafrost, we constructed an effective-medium model based upon ultrasonic P-wave data that were obtained from earlier laboratory studies. The model uses a two-end-member mixing approach in which an ice-filled, fully frozen end member and a water-filled, fully unfrozen end member are mixed together to form the effective medium of partially frozen sediments. This mixing approach has two key advantages: (1) It does not require parameter tuning of the mixing ratios, and (2) it inherently assumes mixed pore-scale distributions of ice that consist of frame-strengthening (i.e., cementing and/or loadbearing) ice and pore-filling ice. The model-predicted P-wave velocities agree well with our laboratory data, demonstrating the effectiveness of the model for quantitatively inferring ice content from P-wave velocities. The modeling workflow is simple and is largely free of calibration parameters -attributes that ease its application in interpreting field data sets.

Genetic algorithm (GA) is combined with finite-element method for the first time as an alternative method to invert gravity anomaly data for reconstructing the 3D density structure in the subsurface. The computational efficiency is significantly improved by storing the coefficient matrix and using it in all the forward calculations, and by dividing the region of interest into many sub-regions and applying the GA to the sub-regions in parallel. Central Taiwan - a geologically complex region - is used as an example to demonstrate the utility of the method. A crustal block, 120 x 150 km2 in area and 34 km in thickness, is represented by a finite-element model of 76,500 cubic elements, each 2x2x2 km3 in size. An initial density model is reconstructed from the regional 3D tomographic seismic velocity, using an empirical relation between velocity and density. The difference between the calculated and the observed gravity anomaly (i.e., the residual anomaly) shows an elongated minimum of large magnitude that extends along the axis of the Taiwan mountain belt. Among the interpretative models tested, the best model shows a crustal root beneath the axis of the Central Ranges and a density contrast of 400 or 500 kg/m3 across the Moho. Both predictions appear to be supported by independent seismological and laboratory evidences.

Saline permafrost is sensitive to thermal disturbances and is prone to subsidence, which renders it a major source of geohazard in Arctic coastal environments. Seismic methods could be used to map and monitor saline permafrost at scales of geotechnical interests because of the ice-content dependencies of seismic properties. We have developed a comprehensive study of the ultrasonic P-wave properties (i.e., velocity and attenuation) of synthetic saline permafrost samples for a range of salinities and temperatures, and measurements conducted on a fine-grained permafrost core obtained from Barrow, Alaska. The resulting data consist of P-wave properties presented as functions of temperature and salinity. Notable observations include the following: P-wave velocities showed marked reductions in the presence of dissolved salts and complex variations resulting from the water-to-ice phase transitions; strong P-wave attenuation was present in the temperature intervals in which the samples were partially frozen. When presented as functions of ice saturation, the data sets lead us to two key findings: (1) neither a purely cementing nor a purely pore-filling model of the pore-scale distributions of ice could adequately fit the observed velocity data and (2) although the velocities increase monotonically with increasing ice saturations, P-wave attenuation reaches a maximum at intermediate ice saturations--contrary to the ordinary expectation of decreasing attenuation with increasing velocities. The observed ice-content dependencies of P-wave properties, along with the implications on the probable pore-scale distributions of ice, provide a valuable basis for rock-physics modeling, which in turn could facilitate seismic characterizations of saline permafrost. [ABSTRACT FROM AUTHOR]