Search

ABSTRACTRecently, the positive unlabeled (PU) learning algorithms have proven highly effective in generating accurate landslide susceptibility maps. The algorithms categorize samples exclusively into positive samples (landslides) and unlabeled samples for training, eliminating random or subjective selection of non-landslide samples. However, existing PU learning algorithms face limitations in capturing correct negative samples in multi-genesis landslide areas, leading to lower prediction accuracy. To address this issue, the PU-pullbaggingDT algorithm was proposed in this study. This integrated method combines the strengths of two techniques: the superior ranking performance of the PU-baggingDT algorithm and the low bias of the contrastive learning approach. The contrastive learning introduces a new contrastive loss of PU learning (PUpull loss), which pulls the distance of similar landslide samples in the projection space closer based on the validation accuracy threshold, without the need for data augmentation and class probability. The PUpull loss relaxes the tightness toward unlabeled samples, reducing the impact of incorrectly defined non-landslide samples on prediction results in multi-genesis landslide areas. The proposed algorithm outperforms existing PU-learning and machine learning methods (support vector machine, decision tree, logistic regression, AdaBoost, and XGBoost) with random selection of negative samples for predicting landslide susceptibility in China’s Zigui County, as demonstrated by comprehensive evaluation metrics. The landslide susceptibility mapping utilizes equal interval division and ranking, achieving approximately a 90% landslide percentage in areas with very high and high susceptibility in Zigui County. This demonstrates the capability of the proposed algorithm to effectively predict landslide susceptibility in complex geological settings.

Objective With the construction of reservoir projects, the problem of deformation and instability of colluvial landslides in reservoir areas has become increasingly prominent. These landslides often experience deformation during the period of reservoir water level decline particularly exacerbated by rainfall. Therefore, the seepage-deformation mechanism of landslides under the joint action of reservoir water level decline and rainfall is one of the key scientific issues. Methods In this paper, a typical colluvial landslide in the Three Gorges Reservoir Area, the Shilongmen landslide, is used as a prototype. Physical model tests were designed to observe the macroscopic phenomena of the slope body and the pressure changes in the slope during combined scenarios of reservoir level drops and rainfall of different intensities, revealing the seepage-deformation mechanism of the colluvial landslide under complex hydrological conditions. Results The test results indicated that the decrease in reservoir level has a more obvious effect on the control of seepage at the front of the slope, while rainfall significantly raises the water head at the middle and rear of the slope, and the combination of the two will increase the hydraulic gradient in the slope, thus inducing slope deformation. Conclusion The seepage-deformation mechanism of the colluvial landslide when the reservoir level drops is as follows: In the absence of rainfall, only shallow surface scouring action occurs due to the difference between the internal and external heads of the slope. Under the condition of the worst rain in a 100-year rainstorm, cracks appear near the high water level, and then overall deformation occurs, with a maximum displacement of 58.3 mm. This study can provide a theoretical basis for the prevention and control of colluvial landslides under complex hydrological conditions.

Susceptibility evaluation is the basis of regional landslide risk early warning and stability analysis. Scientific and reasonable division of evaluation unit is the key to landslide susceptibility evaluation. For large-scale fine landslide susceptibility evaluation, the traditional slope unit division method based on hydrology and geomorphology generally results in low accuracy of the evaluation. In this paper, an improved slope unit method based on the slope geological environment is proposed. Dazhou Town was selected as an example and the obtained results from the proposed model were compared with the results from hydrological analysis method and curvature watershed method. The results show that the size uniformity of the evaluation units divided by the proposed method is better, and no fine units or deformed long strip units were generated. The overall morphological characteristics of the evaluation unit are more reasonable, and the morphological index is between 1 and 2, which generally presents circular-like or square-like shape. At the same time, the superposition degree between the results of the improved slope unit division and the range of the existing disaster boundary is the highest, which can better reflect the physical significance of landslide risk assessment. The proposed model has significant potential for improving the accuracy of regional landslide susceptibility evaluation.

A large number of high voltage transmission tower foundations crossing mountainous and hilly areas are often located in high-prone slope areas of landslide disasters. Applying appropriate protective measures to improve their stability is the key to ensuring the continuous and safe operation of transmission lines. To study the protection effect of different protection measures on the tower foundation landslide, this paper takes the Yanzi landslide in Badong County, Hubei Province as a geological prototype, designs and produces a physical test model, and carries out physical model tests of the landslide under extreme rainfall conditions(50, 100 mm/h) without protection, applying anti-slide piles and lattice protection. The deformation and failure characteristics of the landslide and the protective effect of different protective measures are revealed from the experimental point of view. The results show that under two extreme conditions, the unprotected landslide experienced the evolution process of slope surface erosion, crack propagation, local collapse and deformation, and overall sliding. The anti-slide pile measures have a significant effect on the overall protection of the landslide. The landslide is in a stable state, the deformation of the tower foundation is small, and the inclination rate of the tower meets the specification, but the slope surface will be scoured and collapsed. Lattice slope protection measures can effectively reduce the risk of slope erosion and slope toe collapse, but the overall stabilization of the tower foundation under continuous heavy rainfall is slightly weaker. The model test results are consistent with the historical deformation of the landslide and the actual treatment effect. The test conclusions can provide a reference for the failure mechanism research and protection engineering design of similar tower foundation landslides.

Landslide susceptibility mapping (LSM) is significant for landslide risk assessment. However, there remains no consensus on which method is optimal for LSM. This study implements a dynamic approach to landslide hazard mapping by integrating spatio-temporal probability analysis with time-varying ground deformation velocity derived from the MT-InSAR (Multi-Temporal InSAR) method. Reliable landslide susceptibility maps (LSMs) can inform landslide risk managers and government officials. First, sixteen factors were selected to construct a causal factor system for LSM. Next, Pearson correlation analysis, multicollinearity analysis, information gain ratio, and GeoDetector methods were applied to remove the least important factors of STI, plan curvature, TRI, and slope length. Subsequently, information quantity (IQ), logistic regression (LR), frequency ratio (FR), artificial neural network (ANN), random forest (RF), support vector machine (SVM), and convolutional neural network (CNN) methods were performed to construct the LSM. The results showed that the distance to a river, slope angle, distance from structure, and engineering geological rock group were the main factors controlling landslide development. A comprehensive set of statistical indicators was employed to evaluate these methods’ effectiveness; sensitivity, F1-measure, and AUC (area under the curve) were calculated and subsequently compared to assess the performance of the methods. Machine learning methods’ training and prediction accuracy were higher than those of statistical methods. The AUC values of the IQ, FR, LR, BP-ANN, RBF-ANN, RF, SVM, and CNN methods were 0.810, 0.854, 0.828, 0.895, 0.916, 0.932, 0.948, and 0.957, respectively. Although the performance order varied for other statistical indicators, overall, the CNN method was the best, while the BP-ANN and RBF-ANN method was the worst among the five examined machine methods. Hence, adopting the CNN approach in this study can enhance LSM accuracy, catering to the needs of planners and government agencies responsible for managing landslide-prone areas and preventing landslide-induced disasters.

Machine learning models are regarded as efficient and popular models for natural disaster susceptibility prediction. However, few studies have focussed on the applications of the latest popular machine learning models in collapse susceptibility assessment (CSA). This paper proposes a 3S (RS, GPS and GIS) technology-based support vector machine (SVM) to map collapse susceptibility in the Nantian area of China. The 3S technology-based back-propagation neural network (BPNN) and radial basis function neural network (RBFNN) models are also proposed for comparison. First, 44 recorded collapses are identified through field investigation, and fourteen collapse-related causal factors are acquired using 3S technology. Second, among these recorded collapses and randomly selected ‘non-collapses’, 70% of the collapse and non-collapse grid cells are used to train the three models, while the remaining 30% are used to test the models. Third, the collapse susceptibility maps of the Nantian area are produced using the three models. Finally, the prediction accuracies of these models are evaluated. The results indicate that the SVM model has the highest prediction accuracy, while the RBFNN model has the lowest prediction accuracy for CSA. In addition, the distribution characteristics of collapse susceptibility in the Nantian area are produced well by all three models.

A demodulation method of spatial domain based on low-coherence interferometry is proposed. The original signal of spatial domain acquired by linear CCD is filtered with all-phase filter to obtain the fringe pattern, and the phase signal of the spatial frequency domain is retained after filtering. The fringe pattern analysis is employed to avoid phase ambiguity and realize large measurement range in demodulation. Afterwards, using the centroid position of the fringe, the peak position of a single fringe pattern can be obtained, and finally the accurate peak position is calculated with seven-step phase-shifting interferometry. The demodulation method is demonstrated experimentally with Fabry-Perot sensing system. At room temperature, it is proved that the full-scale measurement error is less than 0.019% in the range of 11 kPa to 290 kPa. From -20 °C to 70 °C, the accuracy is stable and the maximum error is less than 0.054 kPa. The demodulation method has the potential to enhance the measurement accuracy and adaptability in actual environment for universal applications.

Displacement predictions are essential to landslide early warning systems establishment. Most existing prediction methods are focused on finding an individual model that provides a better result. However, the limitation of generalization that is inherent in all models makes it difficult for an individual model to predict different cases accurately. In this study, a novel coupled method was proposed, combining the long short-term memory (LSTM) neural networks and support vector regression (SVR) algorithm with optimal weight. The Shengjibao landslide in the Three Gorges Reservoir area was taken as a case study. At first, the moving average method was used to decompose the cumulative displacement into two components: trend and periodic terms. Single-factor models based on LSTM neural networks and SVR algorithms were used to predict the trend terms of displacement, respectively. Multi-factors LSTM and SVR models were used to predict the periodic terms of displacement. Precipitation, reservoir water level, and previous displacement are considered as the candidate factors for inputs in the models. Additionally, ensemble models based on the SVR algorithm are used to predict the optimal weight to combine the results of the LSTM and SVR models. The results show that the LSTM models display better performance than SVR models; the ensemble model with optimal weight outperforms other models. The prediction accuracy can be further improved by also considering results from multiple models.

Abstract Purpose Fat infiltration (FI) of the deep neck extensor muscles has been shown to be associated with poor outcomes in cervical injury, mechanical neck pain, and axial symptoms after cervical spine surgery. However, information is scarce on the severity of FI in cervical extensors associated with different clinical syndromes in patients with cervical spondylosis. Objective To investigate the relationship between the severity of FI in the cervical multifidus musculature and its clinical correlates in the syndromes and sagittal alignment of patients with cervical spondylosis. Methods This study was conducted as a retrospective study of twenty-eight healthy volunteers (HV) together with sixty-six patients who underwent cervical radiculopathy (CR), degenerative myelopathy (DM), and axial joint pain (AJP) from January 2020 to March 2022. MRI was used to measure the fat cross-sectional area (FCSA), functional muscle cross-sectional area (FMCSA), total muscle cross-sectional area (TMCSA), FI ratio of the cervical multifidus musculature at each cervical level from the C3 to C6 segments and the cervical lordosis angle in the included subjects. Results The difference in the FCSA and FI ratio in patient groups with cervical spondylosis was significantly greater than that of the HV group (P

This work reports a facile and rapid method assisted by microwave irradiation for the synthesis of hemin functionalized reduced graphene oxide (hemin–RGO) materials. Our investigation confirmed that the hemin molecules were covalently grafted to the surface of graphene by the amidation reaction of the —NH2 groups on the edges of ethylenediamine functionalized graphene oxide with the —COOH groups of hemin. Hemin–RGO exhibits a homogeneous dispersion in water, dimethylformamide, and acetone after more than one month, indicating that hemin can effectively improve the dispersion and solubility of RGO in the solvent. Hemin–RGO was used as a counter electrode in dye-sensitized solar cells and exhibited preferable electrocatalytic activity for I3− to I− reduction compared with RGO. [ABSTRACT FROM AUTHOR]