Search

Pore-fluid pressure (PP) plays an important role in bed erosion, but the mechanisms that control PP evolution and the resulting feedbacks on flow dynamics are unclear. Here, we develop a general formulation, allowing quantification of the propensity for PP evolution of saturated and unsaturated bed sediments. We conduct erosion experiments by systematically varying grain composition and water content of beds, for investigating effects of PP evolution on flow erosion. With increasing water content, PP shows a slight rise in deforming beds with drained behavior but significant larger rise in undrained beds. Regardless of bed composition, the erosion rate of beds presents a synchronous change tendency with PP evolution due to the loss in basal friction. PP instigates positive feedback that induces a remarkable gain of flow velocity and momentum on wet beds with undrained behavior. Our results help explain observations of volume growth and long run out of debris flows.

This study aims to accurately predict landslide displacement characterized by a step-like displacement curve, resulting from complex interactions among multiple factors, including periodic and variable elements. Specifically, we propose a framework based on a Convolutional Neural Network (CNN) and optimized Bidirectional Gated Recurrent Unit (BiGRU) with an Attention mechanism, designed to forecast landslide displacement with a step-like curve. Initially, landslide displacements are decomposed into periodic and trend terms utilizing complementary ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm. Subsequently, a CNN layer is constructed to extract intricate high-dimensional features, while a BiGRU layer is established to capture temporal dependencies within historical sequences. Notably, an attention mechanism module is introduced to enhance the impact of key information within rainfall, water level, and historical displacement data. Finally, the nonlinear weighted Huber loss function (NLWHL) evaluation method is introduced to assess the accuracy of the model in predicting mutation states. The results demonstrate that the proposed framework exhibited superior accuracy in predicting landslide displacement in comparison to alternative intelligent algorithms. Regarding total displacement, the CNN-BiGRU-Attention model demonstrates superior predictive capabilities, reducing RMSE values by 12.52%, 14.15%, 15.58%, 21.68%, 21.84%, 32.81%, 58.10%, and 53.81% in the test set compared to CNN-BiGRU, CNN-BiLSTM, BiGRU-Attention, SMA-GRU, SMA-LSTM, SMA-SVM, GRU, and LSTM, respectively. Furthermore, the introduction of NLWHL underscores the remarkable accuracy of the proposed framework in forecasting both creep and mutation states. Assessing the displacement mutation state provides the necessary opportunity for early detection and intervention. © 2024

Debris flows are common geological hazards in mountainous regions worldwide. Predicting the impact pressure of debris flows is of major importance for hazard mitigation. Here, we experimentally investigate the impact characteristics of debris flows by varying the concentrations of debris grains and slurry. The measured impact pressure signal is decomposed into a stationary mean pressure (SMP) and a fluctuating pressure (FP) through empirical mode decomposition. The SMP of low frequency is caused by the thrusting of bulk flow while the FP of high frequency is induced by the collision of coarse debris grains, revealed by comparing the features of impact pressure spectra of pure slurries and debris flows. The peak SMP and the peak FP first increase and then decrease with the slurry density. The basal frictional resistance is reduced by the nonequilibrium pore-fluid pressure for debris flows with low-density slurry, which can increase the flow velocity and impact pressures. In contrast, the viscous flow of high-density slurry tends to reduce the flow velocity. The peak SMPs are well predicted by the Bernoulli equation and are related to the hydrostatic pressure and Froude number of the incident flow. The peak FPs depend on the kinetic energy and degree of segregation of coarse grains. The maximum degree of segregation occurs at an intermediate value of slurry density due to the transition of flow regime and fluid drag stresses. Our results facilitate predicting the impact pressures of debris flows based on their physical properties.

Two large successive co-site landslide dams blocked the Jinsha River in the Sichuan Province of China in 2018. A quantitative risk analysis was carried out to quantify potential human and economic losses resulting from the failure of these dams, especially the interaction between the co-site landslide dams, and to further investigate the influence of digging a diversion channel on risk mitigation. Flood routing for three scenarios (i.e. after the first dam formation, when the second landslide mass was added, and with a diversion channel after failures) were simulated using HEC-RAS. The human and economic losses were evaluated using a human risk assessment model together with empirical formulations. The results show that the risk of breaching floods had increased significantly after the second co-site landslide dam formation on the pre-existing loose deposits of the first dam. This amplification effect of outburst floods was so great that the peak outflow resulting from the breaching of the second landslide dams was more important, leading to greater economic losses than those resulting from the breaching of the first dam. However, the expected loss of life caused by the breach of the two landslide dams appeared small because of the sufficient time lag provided by the long distance between the residential area and the dam site. The simulations also outline the importance of the diversion channel in decreasing the peak outflow rate and hence downstream risks. A parametric analysis shows that a deep channel with a moderate longitudinal gradient can significantly decrease the peak outflow discharge at the dam site. Flood intensity and risks to downstream towns did not change because of the relatively small attenuation rate of the peak outflow rate. However, a shorter height of the residual dam can be obtained by optimizing the diversion channel, and the amplification effects of such co-site dam can be significantly reduced in the future.

Two large successive co-site landslide dams blocked the Jinsha River in the Sichuan Province of China in 2018. A quantitative risk analysis was carried out to quantify potential human and economic losses resulting from the failure of these dams, especially the interaction between the co-site landslide dams, and to further investigate the influence of digging a diversion channel on risk mitigation to better evacuate upstream waters. Flood routing for three scenarios (i.e. after the first dam formation, when the second landslide mass was added, and with a diversion channel after failures) were simulated using HEC-RAS. The human and economic losses were evaluated using a human risk assessment model together with empirical formulations. The results show that the risk of breaching floods had increased significantly after the second co-site landslide dam formation on the pre-existing loose deposits of the first dam. This amplification effect of outburst floods was so great that the peak outflow resulting from the breaching of the second landslide dams was more important, leading to greater economic losses than those resulting from the breaching of the first dam. However, the expected loss of life caused by the breach of the two landslide dams appeared small because of the sufficient time lag provided by the long distance between the residential area and the dam site. The simulations also outline the importance of the diversion channel in decreasing the peak outflow rate and hence downstream risks. A parametric analysis shows that a deep channel with a moderate longitudinal gradient can significantly decrease the peak outflow discharge at the dam site. Flood intensity and risks to downstream towns did not change because of the relatively small attenuation rate of the peak outflow rate. However, a shorter height of the residual dam can be obtained by optimizing the diversion channel, and the amplification effects of such co-site dam can be significantly reduced in the futur

Two large successive co-site landslide dams blocked the Jinsha River in the Sichuan Province of China in 2018. A quantitative risk analysis was carried out to quantify potential human and economic losses resulting from the failure of these dams, especially the interaction between the co-site landslide dams, and to further investigate the influence of digging a diversion channel on risk mitigation. Flood routing for three scenarios (i.e. after the first dam formation, when the second landslide mass was added, and with a diversion channel after failures) were simulated using HEC-RAS. The human and economic losses were evaluated using a human risk assessment model together with empirical formulations. The results show that the risk of breaching floods had increased significantly after the second co-site landslide dam formation on the pre-existing loose deposits of the first dam. This amplification effect of outburst floods was so great that the peak outflow resulting from the breaching of the second landslide dams was more important, leading to greater economic losses than those resulting from the breaching of the first dam. However, the expected loss of life caused by the breach of the two landslide dams appeared small because of the sufficient time lag provided by the long distance between the residential area and the dam site. The simulations also outline the importance of the diversion channel in decreasing the peak outflow rate and hence downstream risks. A parametric analysis shows that a deep channel with a moderate longitudinal gradient can significantly decrease the peak outflow discharge at the dam site. Flood intensity and risks to downstream towns did not change because of the relatively small attenuation rate of the peak outflow rate. However, a shorter height of the residual dam can be obtained by optimizing the diversion channel, and the amplification effects of such co-site dam can be significantly reduced in the future.

Numerous landslide dams have been induced in recent years as a result of frequent earthquakes and extreme climate hazards. Landslide dams present serious threats to lives and properties downstream due to potentially breaching floods from the impounded lakes. To investigate the factors influencing the stability of landslide dams, a large database has been established based on an in-depth investigation of 1,737 landslide dam cases. The effects of triggers, dam materials, and geomorphic characteristics of landslide dams on dam stability are comprehensively analyzed. Various evaluation indexes of landslide dam stability are assessed based on this database, and stability evaluation can be further improved by considering the dam materials. Stability analyses of aftershocks, surges, and artificial engineering measures on landslide dams are summarized. Overtopping and seepage failures are the most common failure modes of landslide dams. The failure processes and mechanisms of landslide dams caused by overtopping and seepage are reviewed from the perspective of model experiments and numerical analyses. Finally, the research gaps are highlighted, and pathways to achieve a more complete understanding of landslide dam stability are suggested. This comprehensive review of the recent advances in stability and failure mechanisms of landslide dams can serve as a key reference for stability prediction and emergency risk mitigation. © Copyright © 2021 Zheng, Shi, Shen, Peng, Hanley, Ma and Zhang.

Numerous landslide dams have been induced in recent years as a result of frequent earthquakes and extreme climate hazards. Landslide dams present serious threats to lives and properties downstream due to potentially breaching floods from the impounded lakes. To investigate the factors influencing the stability of landslide dams, a large database has been established based on an in-depth investigation of 1,737 landslide dam cases. The effects of triggers, dam materials, and geomorphic characteristics of landslide dams on dam stability are comprehensively analyzed. Various evaluation indexes of landslide dam stability are assessed based on this database, and stability evaluation can be further improved by considering the dam materials. Stability analyses of aftershocks, surges, and artificial engineering measures on landslide dams are summarized. Overtopping and seepage failures are the most common failure modes of landslide dams. The failure processes and mechanisms of landslide dams caused by overtopping and seepage are reviewed from the perspective of model experiments and numerical analyses. Finally, the research gaps are highlighted, and pathways to achieve a more complete understanding of landslide dam stability are suggested. This comprehensive review of the recent advances in stability and failure mechanisms of landslide dams can serve as a key reference for stability prediction and emergency risk mitigation. © Copyright © 2021 Zheng, Shi, Shen, Peng, Hanley, Ma and Zhang.