Short-term displacement prediction for newly established monitoring slopes based on transfer learning

This study makes a significant progress in addressing the challenges of short-term slope displacement prediction in the Universal Landslide Monitoring Program,an unprecedented disaster mitigation program in China,where lots of newly established monitoring slopes lack sufficient historical deformation data,making it difficult to extract deformation patterns and provide effective predictions which plays a crucial role in the early warning and forecasting of landslide hazards.A slope displacement prediction method based on transfer learning is therefore proposed.Initially,the method transfers the deformation patterns learned from slopes with relatively rich deformation data by a pre-trained model based on a multi-slope integrated dataset to newly established monitoring slopes with limited or even no useful data,thus enabling rapid and efficient predictions for these slopes.Subsequently,as time goes on and monitoring data accumulates,fine-tuning of the pre-trained model for individual slopes can further improve prediction accuracy,enabling continuous optimization of prediction results.A case study indicates that,after being trained on a multi-slope integrated dataset,the TCN-Transformer model can efficiently serve as a pretrained model for displacement prediction at newly established monitoring slopes.The three-day average RMSE is significantly reduced by 34.6%compared to models trained only on individual slope data,and it also successfully predicts the majority of deformation peaks.The fine-tuned model based on accumulated data on the target newly established monitoring slope further reduced the three-day RMSE by 37.2%,demonstrating a considerable predictive accuracy.In conclusion,taking advantage of transfer learning,the proposed slope displacement prediction method effectively utilizes the available data,which enables the rapid deployment and continual refinement of displacement predictions on newly established monitoring slopes.

Multitemporal UAV-based photogrammetry for landslide detection and monitoring in a large area:a case study in the Heifangtai terrace in the Loess Plateau of China

With high spatial resolution,on-demand-flying ability,and the capacity for obtaining threedimensional measurements,unmanned aerial vehicle(UAV)photogrammetry is widely used for detailed investigations of single landslides,but its effectiveness for landslide detection and monitoring in a large area needs to be investigated.The Heifangtai terrace in the Loess Plateau of China is a loess terrace that is extremely susceptible to irrigation-induced loess landslides.This paper used UAV-based photogrammetry for a series of highresolution images spanning over 30 months for landslide detection and monitoring of the terrace with an area of 32 km^2.Dense and evenly distributed ground control points were established and measured to ensure the high accuracy of the photogrammetry results.The structure-from-motion(Sf M)technique was used to convert overlapping images into orthographic images,3D point clouds,digital surface models(DSMs)and mesh models.Using multitemporal differential mesh models,landslide vertical movements and potential landslides were detected and monitored.The results indicate that a combination of UAV-based orthophotos and differential mesh models can be used for flexible and accurate detection and monitoring of potential loess landslides in a large area.

Monitoring landslide associated with reservoir impoundment using synthetic aperture radar interferometry: A case study of the Yalong reservoir

The construction of large reservoirs can address the problem of uneven distribution of rivers in time and space,thereby meeting the needs of human production and living.However,the huge elevation of the water level in some areas may modify the distribution of the groundwater level,causing geological disasters,such as surface deformation and landslides.The Yalong reservoir supplies water to the downstream area of Shannan,Tibet;however,since the reservoir started storing water in 2017,the government has discovered two ancient landslides.In this study,to monitor the deformation of the Yalong reservoir since its construction in 2014,we first used synthetic aperture radar(SAR) data and the multidimensional small baseline subset(MSBAS) method to obtain the deformation in the east-west and vertical directions.The result indicated the presence of three large,slow-moving landslides:Landslides I and II,located on the right bank of the Yalong reservoir,which are consistent with the results obtained by the actual survey,and a new discovery,Landslide Ⅲ,located on the left side of the reservoir.Meanwhile,the experimental results indicated that the dam had undergone obvious deformation after impoundment,which should not be ignored.The global positioning system and interferometric SAR(InSAR) timeseries deformation residual data were used to verify the accuracy of the InSAR method.The results also showed that the deformation caused by the three landslides had te nded to accele rate after the rese rvoir’s impoundment,and that the failure mode was retrogressive landslide.We found that InSAR plays a vital role in landslide detection and failure mode research around reservoirs,and assists in providing early warning for subsequent landslide disasters.

Evaluation of the treatment effect of rear slope cutting on hydrodynamic pressure landslides:A case study

After the impoundment of the Three Gorges Reservoir,some huge ancient landslides were reactivated and deformed,showing typical hydrodynamic pressure landslide characteristics.The Baishuihe landslide was a typical hydrodynamic pressure landslide.The management department conducted slope cutting treatments from 2018 to 2019.To evaluate the treatment effect of rear slope cutting,this study analyzed the data of the surface deformation survey and field monitoring over the past 20 years and the characteristics of the reservoir water-triggered Baishuihe landslide deformation,and calculated the seepage field,displacement field,and stability coefficient before and after landslide treatment.The results showed that the deformation of the Baishuihe landslide was primarily related to a decrease in the reservoir water level.Owing to the poor permeability of the landslide soil,the decrease in the reservoir water level produced a seepage force pointing to the outside of the landslide body,leading to the step deformation of the landslide displacement.The landslide was treated by rear slope cutting,and the“step”deformation of the landslide disappeared after treatment.The hydrodynamic pressure caused by the change in reservoir water after cutting the slope did not disappear.However,as the slope cutting greatly reduced the overall sliding force of the landslide,its stability was greatly improved.Notably,high stability can still be ensured under extreme rainfall after treatment.Slope cutting is effective for treating hydrodynamic pressure landslides.This study can provide effective technical support for the treatment of reservoir landslides.

Conductivity characteristics of landslide considering porosity,saturation,temperature and ion concentration

High-density resistivity imaging method is widely used in landslide monitoring.The resistivity of rock and soil is closely related to factors,such as porosity,moisture content,saturation and temperature.In this study,the resistivity test was designed to investigate the influence of physical factors and pore solution components on the resistivity of landslide soil.Experimental and analytical results find that both moisture content and volumetric water content varies greatly under the same resistivity.At different temperatures,soil resistivity exhibits great changes.Under the same temperature,the ion concentration and species in pore solutions have great influence on soil resistivity.Based on the test results and grey correlation analysis,this study established a resistivity model by considering porosity,saturation,temperature and ion concentration.The study lays a foundation for the high-density resistivity method to measure the moisture content of landslides.

Prediction of landslide displacement with dynamic features using intelligent approaches

Landslide displacement prediction can enhance the efficacy of landslide monitoring system,and the prediction of the periodic displacement is particularly challenging.In the previous studies,static regression models(e.g.,support vector machine(SVM))were mostly used for predicting the periodic displacement.These models may have bad performances,when the dynamic features of landslide triggers are incorporated.This paper proposes a method for predicting the landslide displacement in a dynamic manner,based on the gated recurrent unit(GRU)neural network and complete ensemble empirical decomposition with adaptive noise(CEEMDAN).The CEEMDAN is used to decompose the training data,and the GRU is subsequently used for predicting the periodic displacement.Implementation procedures of the proposed method were illustrated by a case study in the Caojiatuo landslide area,and SVM was also adopted for the periodic displacement prediction.This case study shows that the predictors obtained by SVM are inaccurate,as the landslide displacement is in a pronouncedly step-wise manner.By contrast,the accuracy can be significantly improved using the dynamic predictive method.This paper reveals the significance of capturing the dynamic features of the inputs in the training process,when the machine learning models are adopted to predict the landslide displacement.

Numerical modeling of deep-seated landslides interacting with man-made structures

This paper describes the interaction between deep-seated landslides and man-made structures such as dams, penstocks, viaducts, and tunnels. Selected case studies are reported first with the intent to gain insights into the complexities associated with the interaction of these structures with deep-seated landslides(generally referred to as deep-seated gravity slope deformations, DSGSDs). The main features, which characterize these landslides, are mentioned together with the interaction problems encountered in each case. Given the main objective of this paper, the numerical modeling methods adopted are outlined as means for increase in the understanding of the interaction problems being investigated. With the above in mind, the attention moves to an important and unique case history dealing with the interaction of a large-size twin-tunnel excavated with an earth pressure balance(EPB)tunnel boring machine(TBM) and a deep-seated landslide, which was reactivated due to the stress changes induced by tunnel excavation in landslide shear zone. The geological and geotechnical conditions are described together with the available monitoring data on the landslide movements, based on the advanced and conventional monitoring tools used. Numerical modeling is illustrated as an aid to back-analyze the monitored surface and subsurface deformations and to assist in finding the appropriate engineering solution for putting the tunnel into service and as a follow-up means for future understanding and control of the interaction problems. The simulation is based on a novel time-dependent model representing the landslide behavior.

Stability analysis of reference station and compensation for monitoring stations in GNSS landslide monitoring

The Real-Time Kinematic(RTK)positioning method of the Global Navigation Satellite System(GNSS)has been widely used for landslide monitoring.The stability of its reference station is crucial to obtain accurate and reliable monitoring results.Unstable reference stations due to the geological environment and human activities are difficult to detect and in practical applications often ignored.As a result,it affects the positioning solutions and subsequently the interpretation and detection of landslide motions,which must be addressed in GNSS landslide monitoring.To solve this problem,we propose using the Precise Point Positioning(PPP)technique to analyze the stability of the reference station by verifying its position.The deformations of the monitoring stations are then compensated.First,the reference station coordinates are obtained by the PPP technique and tectonic motion is considered in data processing.The change or breakout of the reference station position is then determined using a cumulative sum control chart method.Finally,each monitoring station’s displacements are compensated according to the displacements of the reference station.According to the results of the Tengqing landslide experiment,the PPP technique can be used in GNSS landslide monitoring to analyze the stability of reference stations.With PPP,millimeter-level accuracy for the coordinates of reference stations is achieved.Compared to the traditional deformation series,the compensated displacement series more reliably reflects the landslide motions.This study will increase the reliability of monitoring results and contribute to implementing GNSS in monitoring landslides.

A low-cost PDGNSS-based sensor network for landslide monitoring
challenges,possibilities,and prospects

Simple navigation receivers can be used for positioning with sub-centimeter accuracy in a wireless sensor network if the read-out of the carrier phase(CP)data is possible and all data are permanently broadcast to a central processing computer.At this base station an automated near real-time processing takes place and a precise differential GNSS-based positioning of the involved sensor nodes is computed.The paper describes the technical principles of such a system with its essential demands for the sensing,the communication,and the computing components.First experiences in a research project related to landslide monitoring are depicted.Of course the developed system can also be embedded for location finding in a widespread multifunctional geo sensor network.The quality of the obtained result is restricted due to the fact that the CP measurements must be recorded over a certain time span,usually a few minutes for every independent position solution.As far as possible a modular structure with commercial off-theshelf components,e.g.standard wireless local area network for communication,and in cooperation of existing proofed and powerful program tools is chosen.Open interfaces are used as far as possible.