Sensitivity analysis of regional rainfall-induced landslide based on UAV photogrammetry and LSTM neural network

Rainfall stands out as a critical trigger for landslides,particularly given the intense summer rainfall experienced in Zheduotang,a transitional zone from the southwest edge of Sichuan Basin to Qinghai Tibet Plateau.This area is characterized by adverse geological conditions such as rock piles,debris slopes and unstable slopes.Furthermore,due to the absence of historical rainfall records and landslide inventories,empirical methods are not applicable for the analysis of rainfall-induced landslides.Thus we employ a physically based landslide susceptibility analysis model by using highprecision unmanned aerial vehicle(UAV)photogrammetry,field boreholes and long short term memory(LSTM)neural network to obtain regional topography,soil properties,and rainfall parameters.We applied the Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability(TRIGRS)model to simulate the distribution of shallow landslides and variations in porewater pressure across the region under different rainfall intensities and three rainfall patterns(advanced,uniform,and delayed).The landslides caused by advanced rainfall pattern mostly occurred in the first 12 hours,but the landslides caused by delayed rainfall pattern mostly occurred in the last 12 hours.However,all the three rainfall patterns yielded landslide susceptibility zones categorized as high(1.16%),medium(8.06%),and low(90.78%).Furthermore,total precipitation with a rainfall intensity of 35 mm/h for 1 hour was less than that with a rainfall intensity of 1.775 mm/h for 24hours,but the areas with high and medium susceptibility increased by 3.1%.This study combines UAV photogrammetry and LSTM neural networks to obtain more accurate input data for the TRIGRS model,offering an effective approach for predicting rainfall-induced shallow landslides in regions lacking historical rainfall records and landslide inventories.

A WD-GA-LSSVM model for rainfall-triggered landslide displacement prediction

This paper proposes a WD-GA-LSSVM model for predicting the displacement of a deepseated landslide triggered by seasonal rainfall,in which wavelet denoising(WD)is used in displacement time series of landslide to eliminate the GPS observation noise in the original data,and genetic algorithm(GA)is applied to obtain optimal parameters of least squares support vector machines(LSSVM)model.The model is first trained and then evaluated by using data from a gentle dipping(~2°-5°)landslide triggered by seasonal rainfall in the southwest of China.Performance comparisons of WD-GA-LSSVM model with Back Propagation Neural Network(BPNN)model and LSSVM are presented,individually.The results indicate that the adoption of WD-GA-LSSVM model significantly improves the robustness and accuracy of the displacement prediction and it provides a powerful technique for predicting the displacement of a rainfall-triggered landslide.

Effect of rainfall on a colluvial landslide in a debris flow valley

A colluvial landslide in a debris flow valley is a typical phenomena and is easily influenced by rainfall. The direct destructiveness of this kind of landslide is small, however, if failure occurs the resulting blocking of the channel may lead to a series of magnified secondary hazards. For this reason it is important to investigate the potential response of this type of landslide to rainfall. In the present paper, the Goulingping landslide, one of the colluvial landslides in the Goulingping valley in the middle of the Bailong River catchment in Gansu Province, China, was chosen for the study. Electrical Resistivity Tomography(ERT), Terrestrial Laser Scanning(TLS), together with traditional monitoring methods, were used to monitor changes in water content and the deformation of the landslide caused by rainfall. ERT was used to detect changes in soil water content induced by rainfall. The most significant findings were as follows:(1) the water content in the centralupper part(0~41 m) of the landslide was greaterthan in the central-front part(41~84 m) and(2) there was a relatively high resistivity zone at depth within the sliding zone. The deformation characteristics at the surface of the landslide were monitored by TLS and the results revealed that rainstorms caused three types of deformation and failure:(1) gully erosion at the slope surface;(2) shallow sliding failure;(3) and slope foot erosion. Subsequent monitoring of continuous changes in pore-water pressure, soil pressure and displacement(using traditional methods) indicated that long duration light rainfall(average 2.22 mm/d) caused the entire landslide to enter a state of creeping deformation at the beginning of the rainy season. Shear-induced dilation occurred for the fast sliding(30.09 mm/d) during the critical failure sub-phase(EF). Pore-water pressure in the sliding zone was affected by rainfall. In addition, the sliding L1 parts of the landslide exerted a discontinuous pressure on the L2 part. Through the monitoring and analysis, we conclude that this kind of landslide may have large deformation at the beginning and the late of the rainy season.

The formation of the Wulipo landslide and the resulting debris flow in Dujiangyan City, China

The Wulipo landslide, triggered by heavy rainfall on July 10, 2013, transformed into debris flow,resulted in the destruction of 12 houses, 44 deaths, and 117 missing. Our systematic investigation has led to the following results and to a new understanding about the formation and evolution process of this hazard. The fundamental factors of the formation of the landslide are a high-steep free surface at the front of the slide mass and the sandstone-mudstone mixed stratum structure of the slope. The inducing factor of the landslide is hydrostatic and hydrodynamic pressure change caused by heavy continuous rainfall. The geological mechanical model of the landslide can be summarized as "instability-translational slide-tension fracture-collapse" and the formation mechanism as "translational landslide induced by heavy rainfall". The total volume of the landslide is 124.6×104 m3, and 16.3% of the sliding mass was dropped down from the cliff and transformed into debris flow during the sliding process, which enlarged 46.7% of the original sliding deposit area. The final accumulation area is found to be 9.2×104 m2. The hazard is a typical example of a disaster chain involving landslide and its induced debris flow. The concealment and disaster chain effect is the main reason for the heavy damage. In future risk assessment, it is suggested to enhance the research onpotential landslide identification for weakly intercalated slopes. By considering the influence of the behaviors of landslide-induced debris flow, the disaster area could be determined more reasonably.

Two types of summertime heating over Asian large-scale orography and excitation of potential-vorticity forcing II.Sensible heating over Tibetan-Iranian Plateau

Based on analysis and simulation,the interaction of thermal forcing between the Tibetan Plateau(TP) and Iranian Plateau(IP) in summer is investigated.Associated influences on water vapor transport in the Asian subtropical monsoon region and the formation of a cold center in the lower stratosphere over Eurasia are also investigated.Results show that surface sensible heating(SH) over the two plateaus not only have mutual influences but also feedback to each other.SH over the IP can reduce the SH and increase the LH over the TP,whereas the SH over the TP can increase surface heating over the IP,thereby reaching quasi-equilibrium among the SH and LH over the TP,IP SH and atmosphere vertical motion.Therefore,the so-called Tibetan-Iranian Plateau coupling system(TIPS) is constructed,which influences atmosphere circulation.In the TIPS system,interaction between surface SH and LH over the TP plays a leading role.SH of the IP and TP influences on other regions not only have superimposed effects but also mutually offset.Accounting for contributions to the convergence of water vapor transport in the Asian subtropical monsoon region,TP SH contributes more than twice that of the IP.The combined influence of SH over TP and IP represents the major contribution to the convergence of water vapor transport in that region.In addition,the heating effect of TIPS increases the upper tropospheric temperature maximum and lifts the tropopause,cooling the lower stratosphere.Combined with large-scale thermal forcing of the Eurasian continent,the TIPS produces a strong anticyclonic circulation and the South Asian High that warms the upper troposphere and cools the lower stratosphere,thereby affecting regional and global weather and climate.