Karst ground collapse, a geological disaster in karst areas characterized by the sudden subsidence of surface rock and soil, poses significant risks to human life and property owing to its abrupt and frequent occurrence. Karst ground collapses can be classified into soil-cave-type and hourglass-type, based on the viscosity of the overlying layer. Among these, the hourglass-type presents a higher collapse risk owing to the lack of cohesive forces in the overlying layer. This study focused on hourglass-type karst ground collapse, utilizing physical model tests and the discrete element numerical simulations to develop and validate a collapse model. The physical model tests reproduced the collapse process and provided insights into its underlying mechanism. Numerical simulations were employed to evaluate the effects of karst channel conditions and drilling-induced vibrations on hourglass-type collapses. The results indicated that although the length of the karst channel had minimal impact on collapse speed and pattern, a wider karst channel resulted in a faster collapse and a larger final collapse pit. Moreover, vibration loads increased the collapse speed, shifted the collapse pit towards the vibration source, and expanded the scale of the collapse, thereby amplifying the overall damage extent.

Highway tunnels are occasionally built under difficult ground conditions and technical limitations, in a surrounding ground mass with weak mechanical characteristics, specifically the Cretaceous soil, which is a significantly weathered and deteriorated rock exhibiting a scaly composition. These tunnels are typically dug with wide cross sections and commonly in double-tube configurations, maintaining a distinct gap between them to promote traffic flow and safety services. Tunnels with high sections often cause considerable stress changes and distortions that can lead to ground collapse under misinterpreted conditions of the ground mass, especially in freeway twin-tunnel with defined spacing. This study examines a paired 3-lane tunnel, which is a component of a nationwide freeway initiative covering approximately 1200 km, characterized by a horseshoe configuration with a cross of 190 m2 each and a clear gap of 17 m. Considering the surrounding rock conditions, at a specific phase of the project, significant displacements peaked at 41 mm/day; additionally, concrete fractures were noted before a major failure extending up to 130 m toward the tunnel. This study suggests a back assessment for main disturbance factors before and after collapse, evaluates support force, and employs numerical modeling to reconstruct ground behavior. It has been observed that the decompression was quite significant, particularly above the tunnel's crown. The monitoring activities emphasized the effectiveness of both the original and enhanced support systems, showing a decrease in displacement from 47 to 64%. Given the surrounding rock's inadequate mechanical properties, the noticeable distance between the tunnels and the extensive excavation area raises concerns about support effectiveness. As excavation advances, the need for prompt responses is also highlighted to facilitate feedback contributions.

Karst collapse as a unique environmental geological hazard in karst areas, easily causes changes in surrounding water and soil environments. Train-induced vibration is a significant inducement for shallow karst ground collapse. Previous studies on the dynamic properties of surrounding soil under train vibration loads often neglected the impact of time intermittent effects. Taking the red soil covering a typical potential karst collapse area along a high-speed railway in China as the research object, field monitoring of the vibration characteristics of the surrounding environment was conducted. A series of continuous loading and continuous-stop-continuous dynamic triaxial tests and scanning electron microscopy (SEM) tests were designed considering factors such as loading frequency, intermittent duration, and dynamic stress amplitude. The effects of loading intermittence on the dynamic response and microstructure of red soil were compared and analyzed. The experimental results show that the drainage and unloading of red soil samples during the intermittent phase dissipate the accumulated excess pore water pressure and adjust the internal particle and structure of the soil, reducing the accumulation of plastic deformation during subsequent loading stages. The residual strain under vibration loading conditions considering the time intermittent effect is significantly reduced, and the residual strain decreases significantly with the increase of time intervals. The weakening effects of both macro and micro characteristics of red soil in karst-prone areas are significantly enhanced with the increase of intermittent time. The research results are of great significance for the prevention and control of karst ground collapse in karst areas.

Leakages from damaged or deteriorated buried pipes in urban water distribution networks may cause significant socio-economic and environmental impacts, such as depletion of water resources and sinkhole events. Sinkholes are often caused by internal erosion and fluidization of the soil surrounding leaking pipes, with the formation of soil cavities that may eventually collapse. This in turn causes road disruption and building foundation damage, with possible victims. While the loss of precious water resources is a well-known problem, less attention has been paid to anthropogenic sinkhole events generated by leakages in water distribution systems. With a view to improving urban smart resilience and sustainability of urban areas, this study introduces an innovative framework to localize leakages based on a Machine learning model (for the training and evaluation of candidate sets of pressure sensors) and a Genetic algorithm (for the optimal sensor set positioning) with the goal of detecting and mitigating potential hydrogeological urban disruption due to water leakage in the most sensitive/critical locations. The application of the methodology on a synthetic case study from literature and a real-world case scenario shows that the methodology also contributes to reducing the depletion of water resources.

Mining activities can damage rock masses and easily induce ground collapse, which seriously threatens safe production in mining areas. Micro-seismic systems can monitor rock mass deformation signals in real time and provide more accurate data for rock mass deformation analysis. Therefore, in this study, the waveform characteristics of micro-seismic events induced by ground collapse in the Rongxing gypsum mine were analyzed; the occurrence of these events was introduced on the basis of Fast Fourier Transform, an established Frequency-Time-Amplitude model, in order to put forward the index of energy proportion of the main band. The results showed the following. (1) The seismic sequence type of ground collapse was foreshock-mainshock-aftershocks. The interval between the foreshock and mainshock was longer than that between the mainshock and aftershocks. (2) The deformation corresponding to the foreshock micro-seismic events was mainly that of a small-scale crack. The deformation corresponding to the micro-seismic events during the mainshock was characterized by the gradual development of small-scale cracks, and the development of large-scale cracks accelerated, accompanied by slight rock collapse. The deformation corresponding to the micro-seismic events during the aftershocks showed that almost no small-scale cracks developed, and the large-scale crack development was intense, and accompanied by numerous rock and soil mass collapses. (3) The observed decreasing frequency distribution and energy dispersion can be used as possible precursors of ground collapse.