阅读说明：本技术 一种线路工程走廊浅层滑坡危险性评估及监测预警系统 (Risk assessment and monitoring early warning system for shallow landslide of line engineering corridor ) 是由 谭衢霖 孙晗凌 周嘉琦 廖骜杰 白明洲 胡俊 于 2021-06-16 设计创作，主要内容包括：本发明公开了一种线路工程走廊浅层滑坡危险性评估及监测预警系统建立方法。首先,确定线路工程走廊评估区域,结合地形地貌、遥感信息、地震地质、土壤、植被、水文及气候和区域调查等基础与综合数据,综合分析并确定走廊区域滑坡孕灾关键影响因素,构建走廊斜坡稳定性评估模型；然后,对线路走廊区域进行滑坡危险性初步评估。其次,针对诱发浅层滑坡灾害主要特征因素降水量,对线路走廊浅层滑坡危险性进行不同降水量条件下的分段评价,确定滑坡危险区重点监测预警线路区段。最后,建立线路工程走廊浅层滑坡地质灾害危险性评估与预警WebGIS系统,实现多平台终端数据查询、更新与信息发布互通共享。(The invention discloses a method for establishing a system for assessing, monitoring and early warning the risk of shallow landslide of a line engineering corridor. Firstly, determining a line engineering corridor evaluation area, comprehensively analyzing and determining landslide and disaster pregnancy key influence factors of a corridor area by combining basic and comprehensive data such as terrain and landform, remote sensing information, seismic geology, soil, vegetation, hydrology, climate, regional investigation and the like, and constructing a corridor slope stability evaluation model; and then, carrying out initial landslide risk assessment on the line corridor area. And secondly, aiming at the precipitation of main characteristic factors inducing the shallow landslide hazard, performing sectional evaluation on the risk of the shallow landslide of the line corridor under different precipitation conditions, and determining key monitoring and early warning line sections in the landslide hazard area. And finally, establishing a WebGIS system for evaluating and early warning the geological disaster risk of the line engineering corridor shallow landslide, and realizing data query, update and information publishing, intercommunication and sharing of the multi-platform terminal.)

1. A method for establishing a system for assessing, monitoring and early warning the risk of landslide of a line engineering corridor shallow layer is characterized by comprising the following steps of: the method comprises the following steps:

a, line engineering corridor area investigation and multi-source space-time basis and comprehensive database construction; b, comprehensively analyzing and determining key influence factors of slope instability and pregnancy disaster; c, processing and analyzing the influence factor indexes of the slope pregnancy disaster and assigning standardized numerical values; d, establishing a risk evaluation model of the line corridor shallow landslide; e, primarily evaluating the instability risk of the shallow slope of the line engineering corridor; f, analyzing and determining a precipitation characteristic value for inducing the instability of the shallow slope of the line engineering corridor; g, evaluating the instability risk of the shallow slope of the line corridor in a segmented manner; h, establishing a WebGIS system for early warning of slope instability of the line engineering corridor.

2. The method for establishing the risk assessment, monitoring and early warning system for the shallow landslide of the line engineering corridor as claimed in claim 1, wherein the method comprises the following steps: the line engineering corridor area investigation and multi-source space-time foundation and comprehensive database construction comprises the steps of analyzing the geographical geological environment of the line engineering corridor, conducting survey, collecting and obtaining three-dimensional laser scanning digital elevation models, satellite-borne/airborne/unmanned aerial vehicle multiband optical remote sensing images, vegetation coverage, soil type distribution, water systems, traffic and other basic geographic element data and geographical national condition monitoring data of the line engineering corridor area, and building a line corridor multi-source space-time foundation and comprehensive database.

3. The method for establishing the risk assessment, monitoring and early warning system for the shallow landslide of the line engineering corridor as claimed in claim 1, wherein the method comprises the following steps: comprehensively analyzing and determining key influence factors of slope instability and pregnancy disaster, including verifying the significance of the influence factors by using a machine learning multi-factor variance analysis method; the potential factors influencing the shallow slope destabilization and pregnancy disaster mainly comprise: terrain grade, elevation, engineering rock mass, slope structure, vegetation cover, potential seismic source area, activity fracture, micro-relief type, ergonomic activity, precipitation, soil type, distance from river (valley), etc.

4. The method for establishing the risk assessment, monitoring and early warning system for the shallow landslide of the line engineering corridor as claimed in claim 1, wherein the method comprises the following steps: the C, D, E, F and G steps comprise the following sub-steps:

(1) establishing a line corridor topographic feature layer grid based on the line corridor digital elevation model;

(2) processing and analyzing data such as vegetation coverage, soil type, water system, precipitation and the like in the corridor area, and assigning a standardized numerical value;

(3) based on a slope stability theory, establishing a line corridor shallow slope instability risk evaluation model;

(4) calculating a safety coefficient by using an infinite slope stability theory, and further calculating a slope stability index;

(5) dividing a calibration area for a line corridor area, calibrating model parameters, performing multi-factor variance analysis on the landslide point data and parameter variables known in the area based on machine learning, and verifying the contribution degree of each parameter to the landslide hazard risk or whether the parameter plays an important role;

(6) performing model simulation calculation and evaluation, preliminarily evaluating instability risk of a shallow slope of a line engineering corridor, calculating a preliminary result by using a landslide hazard point distribution investigation data evaluation model, and appropriately adjusting parameters;

(7) analyzing and determining a rainfall characteristic value inducing line engineering corridor shallow slope instability, and performing mathematical statistical analysis on the landslide disaster point distribution, the risk grading and the change of the landslide disaster point distribution and the risk grading under different rainfall amounts;

(8) and (3) carrying out sectional evaluation on the line by using slope instability risks of corridor areas under different precipitation conditions: superposing slope instability risk evaluation graphs of the line and the corridor area, and reclassifying line sections located in different risk regions by using a buffer area analysis and partition statistical method: slope unstable region and high unstable region-high risk section; unstable zone-hazardous zone; metastable zone-metastable segment; stable zone-stable segment; and (4) analyzing the slope instability danger regions of the line under different precipitation conditions in the high stability region-safety section, carrying out sectional statistics on the length and the proportion of each danger evaluation section under each precipitation condition, and determining key monitoring line sections in different seasons.

5. The method for establishing the risk assessment, monitoring and early warning system for the shallow landslide of the line engineering corridor as claimed in claim 1, wherein the method comprises the following steps: and the H step comprises the step of establishing a WebGIS system for the line engineering corridor shallow slope instability disaster early warning based on the line landslide disaster risk segmented evaluation result information and the actual line condition of the model and combining the line engineering geographical geological environment and the line engineering information parameters, and providing command decision guidance for the line landslide disaster monitoring early warning.

6. The WebGIS system for early warning of destabilizing disasters in shallow slopes of line-engineered corridors according to claim 5, wherein: the WebGIS system is different from a traditional or general GIS-based early warning system, and is an online network system which is used for sending early warning and monitoring analysis data and landslide danger information of slope instability disasters of a line engineering corridor on line, realizing real-time intercommunication and sharing of multi-platform terminal data and information, realizing real-time scheduling of field work and real-time uploading of monitoring data by using field tool software, and providing efficient information release and sharing for line landslide disaster early warning and monitoring work.

Technical Field

The invention relates to the field of line engineering disaster prevention and reduction monitoring and early warning safety, in particular to a method for establishing a line engineering corridor shallow slope instability risk sectional assessment and monitoring and early warning system, which is suitable for geological disaster prevention and control in the field of line engineering (railways, highways, electric power, pipelines and the like).

Background

The line engineering such as highway, railway, electric power, pipeline engineering is the important infrastructure passageway of country, carries out regional slope unstability geological disaster danger quantitative evaluation of line engineering and is vital to line safety and unobstructed. The risk evaluation of landslide disaster is a comprehensive analysis of the possibility and the degree of risk of occurrence of slope destabilizing disaster in the corridor area. Landslide disaster development and development are affected by environmental factors such as precipitation, vegetation coverage, topography, soil type, geological conditions and even surface runoff, plate movement, surface morphology, etc. The main causes of landslide disasters are generally considered as follows: dewatering; secondly, earthquake; thirdly, scouring and soaking the surface water; and fourthly, unreasonable human activities. The rainfall is the most common, most frequent and most direct inducement of natural slope instability disasters, and is a key research influence factor for shallow landslide disaster research and prevention.

For many years, a great deal of research has been carried out on the regional slope instability risk at home and abroad, and various methods or evaluation models have been proposed. Mainly comprises the following steps: (1) expert system analysis methods typically assess the vulnerability and risk of a slope instability based on the judgment of the expert. (2) A statistical or probabilistic analysis method based on statistical analysis of various influencing factors and landslide distributions. The evaluation mode for carrying out statistical analysis on the mutual relation between the influencing factors and the slope instability ensures the objectivity of the susceptibility or the risk subarea to a great extent. The limitation of the method is mainly limited by data quality, such as incomplete data cataloging, insufficient data precision and the like. (3) Methods based on classical slope stability theory, such as infinite slope analysis, limit balancing, finite element techniques, etc. These models require standard soil mechanics parameter inputs such as soil thickness, soil strength, groundwater pressure, slope geometry, etc. The shallow slope instability stability model is widely applied, but the influence of the ground soil cohesion on the ground slope stability is not considered in the shallow slope instability stability model, and the applicability is limited. And the hydrologic dynamic modeling is further combined with an infinite slope stability model, and the soil cohesion and the plant root system function are considered in a combined manner. The models are improved to a certain extent, but have advantages and disadvantages relatively, and have low applicability.

In recent years, GIS and geospatial information models have become important support technologies for geological disaster research and analysis. Such as Quzhihua Wang, Quququlin, Yang Wu year, etc. by utilizing RS and GIS technology to develop landslide evaluation application; invar dragon and the like establish a landslide susceptibility evaluation system by using an information quantity model and a logistic regression model, and generate a landslide susceptibility mapping chart in a corridor area according to a selected disaster causing factor as an evaluation index; and factors such as landform, geology, climate and hydrology are considered based on the GIS, and the probability of occurrence of geological disasters in the corridor area is evaluated by combining a statistical method.

In conclusion, how to combine the advantages of the developed model, the technical advantages of the GIS comprehensive capability are fully exerted, various factors can be comprehensively considered, the common action of various factors such as landform, geology, soil, vegetation, hydrology and climate and the like is considered, and the problem that various parameters are uncertain in a large area can be solved, so that the rapid evaluation of the proposed model in a large similar area range has advantages, the model can be applied in different geographical and geological environments, and certain universality becomes the application requirement of landslide hazard evaluation.

One line project usually extends for hundreds of kilometers or even thousands of kilometers, passes through different geological units, particularly railways in western severe mountain areas, frequently encounters or faces geological disasters such as landslides, collapse, debris flows and the like, and is severe in prevention and treatment situation. In order to ensure smooth and safe operation of a line and prevent and treat serious loss caused by landslide geological disasters, the method provides a method for establishing a system for piecewise evaluation and monitoring and early warning of the landslide hazard of a shallow slope of a line engineering corridor. The method is suitable for preventing and treating geological disasters in the field of line engineering (railways, highways, electric power, pipelines and the like).

Disclosure of Invention

The invention aims to provide a method for establishing a system for assessing, monitoring and early warning the risk of shallow landslide of a line engineering corridor, which is realized by the following technical scheme or process for realizing the aim:

a, line engineering corridor area investigation and multi-source space-time basis and comprehensive database construction; b, comprehensively analyzing and determining key influence factors of slope instability and pregnancy disaster; c, processing and analyzing the influence factor indexes of the slope pregnancy disaster and assigning standardized numerical values; d, establishing a risk evaluation model of the line corridor shallow landslide; e, primarily evaluating the instability risk of the shallow slope of the line engineering corridor; f, analyzing and determining a precipitation characteristic value for inducing the instability of the shallow slope of the line engineering corridor; g, evaluating the instability risk of the shallow slope of the line corridor in a segmented manner; h, establishing a WebGIS system for early warning of slope instability of the line engineering corridor.

According to the scheme, the method is characterized in that the geological and geographical environment of the line engineering corridor is analyzed, a survey is conducted, basic geographic element data such as a three-dimensional laser scanning digital elevation model, satellite-borne/airborne/unmanned aerial vehicle multiband optical remote sensing images, vegetation coverage, soil type distribution, water systems, traffic and the like and geographical national condition monitoring data of the line corridor area are collected and obtained, and a multi-source space-time foundation and a comprehensive database of the line corridor are established.

According to the scheme, the method is characterized in that the second method is characterized in that the key influence factors of slope instability and pregnancy disaster are comprehensively analyzed and determined, and the significance of the influence factors is verified by using a multi-factor variance analysis method. The potential factors influencing the shallow slope destabilization and pregnancy disaster mainly comprise: terrain grade, elevation, engineering rock mass, slope structure, vegetation cover, potential seismic source area, activity fracture, micro-relief type, ergonomic activity, precipitation, soil type, distance from river (valley), etc.

According to the scheme, the method is characterized by comprising the following steps:

(1) and creating a line corridor topographic feature layer grid based on the line corridor digital elevation model.

(2) The processing analyzes data such as vegetation coverage, soil type, water system, precipitation and the like in the corridor area, and assigns standardized numerical values.

(3) And establishing a line corridor shallow slope instability risk evaluation model based on a slope stability theory.

The evaluation model is based on an infinite slope stability model as a theoretical basis, which balances the unstable components of gravity with the stable components of friction and cohesion on a plane, and the edge effects are ignored. The method is suitable for most of regions with superficial landslide of the earth surface. The stability Factor (FS) for an infinite slope is calculated by:

wherein, C_rRoot cohesion (N/m)²)；C_s: soil cohesion (N/m)²) (ii) a θ: tilt angle (°); phi: the soil internal friction angle (°); rho_s: density of wet soil (kg/m)³)；ρ_w: density of water (kg/m)³) (ii) a g: acceleration of gravity g is 9.81m/s²(ii) a D: a vertically measured soil depth (m); d_w: groundwater level depth (m).

The thickness h of the soil perpendicular to the slope and the relation with the depth D are:

h＝Dcosθ， (2)

as the depth changes, the infinite slope stability safety factor is further expressed in a dimensionless form:

wherein, topographic moisture index (relative humidity):

w＝D_w/D＝h_w/h； (4)

dimensionless cohesion coefficient: c ═ C_r+C_s)/(hρ_sg)； (5)

Relative density ratio of water to soil: r ═ p_w/ρ_s。 (6)

The topographic humidity index is calculated based on the hydrological model TOPMODEL, and the following are set: (1) the lateral flow of shallow groundwater follows the terrain gradient, i.e. the direction of flow in the productive zone at any point is determined by the particular catchment area defined by the terrain. (2) The lateral flow of each point and the return water flow R in a stable state are in a balanced state. (3) The lateral flow rate per point is Tsin θ, where T is the soil hydraulic conductivity, i.e., the hydraulic conductivity multiplied by the soil thickness h. By setting, a lateral flow q per unit length is obtained, i.e. when the flow behavior of the unsaturated layer is not taken into account, this value can be regarded as the product of the steady lateral flow and the unit catchment area a: q is Ra. Unlike the commonly used TOPMODEL, it is assumed here that the hydraulic conductivity of the soil overlying the relatively impermeable bedrock is uniform. Further, because the flow distance is actually along the slope, the calculation is performed using sin θ instead of tan θ. For small angle terrain, although the difference between tan and sin is not large, for steep slopes, which tend to produce mountain landslides, the difference between the two will become very significant.

The topographic moisture index is:

the upper limit of the topographic moisture index is 1 and any supersaturated zone is considered to form surface water.

The dimensionless cohesion factor C combines cohesion (related to soil type and the characteristics of the roots of the overlying vegetation) with soil density and thickness, which can be considered as the ratio of cohesion strength to soil weight, or relative contribution to the cohesion slope stability. The second term in the numerator of equation (3) is the contribution of quantified soil friction (quantified by the friction angle phi or coefficient of friction tan phi) to stability, decreasing with increasing soil moisture (increased moisture results in increased pore pressure and decreased normal force carried by the soil matrix). The sensitivity of this variation is closely related to the density ratio r.

Substituting the humidity index of formula (7) into formula (3), and converting the safety coefficient calculation into formula (8):

the variables a and θ can be determined by calculation from topographical data, C,R and R/T are variable parameters, but the value ranges of the three parameters can be allowed by setting the lower limit and the upper limit. Let R/T equal to x, tan phi equal to T, and the value range of the parameter is: c to U (C)₁,C₂)，x～U(x₁,x₂)，t～U(t₁,t₂) The density ratio r is regarded as substantially constant (e.g., set at 0.5).

(4) And calculating a safety coefficient by using an infinite slope stability theory, and further calculating a slope stability index. Based on the calculated value of the slope stability safety factor, a slope Stability Index (SI) is defined as the probability that the earth's slope is stable, which is between 0 (least stable) and 1 (critically stable). If the most conservative (unstable) set of parameters in the model is calculated to be stable, the stability index is defined as the safety factor (ratio of stable force to unstable force) calculated at the most conservative set of parameters. When C and t take the minimum value (i.e. C)₁And t₁) And x takes the maximum value (i.e. x)₂) Then, this is the most conservative case for the parameter set. In this worst case, the region with a safety factor FS greater than 1 is unconditionally stable, defined as:

for regions with a minimum safety factor of less than 1, there is a possibility of (probabilistic) instability. This is due to the destabilizing probability caused by uncertainties (spatial variability) in the C, tan and T parameters. Since the parameter R characterizes the humidity over time, the uncertainty of the parameter x actually combines the probability of variation in space and in time.

In (FS)_minRegion definition of < 1):

SI＝Prob(FS＞1)，0＜SI＜1。 (10)

in the distribution of C, x and t, the best case is when C ═ C₂，x＝x₁And t ═ t₂Then, at this time:

when FS is used_maxWhen the number is less than 1, the following components are present:

SI＝0(FS_max＜1)。

according to the set parameters, the earth surface stability index of the area can be calculated by grid unit. The main process comprises 5 steps: (1) DEM pit filling correction; (2) calculating the gradient and the flow direction; (3) calculating the catchment area; (4) calculating a slope stability safety coefficient; (5) grading classification of stability index. Table 1 shows the shallow landslide hazard zone defined by the Stability Index (SI) classification.

TABLE 1 stability index rating and hazard zoning

Stability index	Grading	Slope instability risk zoning
			SI≥1.5	1	High stability region
1.5＞SI≥1.25	2	Stable zone
			1.25＞SI≥1	3	Quasi-stable region
1＞SI≥0.5	4	Unstable region
			0.5＞SI＞0	5	Highly unstable region
SI＝0	6	Protective zone

(5) And dividing a calibration area for the line corridor area, calibrating the model parameters, performing multi-factor variance analysis on the known landslide point related data and parameter variables of the area based on machine learning, and verifying the contribution degree of each parameter to the landslide hazard risk or playing an important role.

The essence of parameter calibration is that a group of parameters are assumed and substituted into a model to obtain a calculation result, then the calculation result is compared with actual measurement data, and if the difference between the calculation value and the actual measurement value is not large, the parameter at the moment is taken as the parameter of the model; if the difference between the calculated value and the measured value is larger, the adjustment parameter is substituted into the model for recalculation, and then comparison is carried out until the error between the calculated value and the measured value meets a certain range.

Before the calibration of the model parameters, a calibration area needs to be divided. The calibration zone is a zone in which the lower limit and the upper limit of the calibration parameter values of T/R, dimensionless cohesion C and internal friction angle in the zone are in the same range. Spatial information application models typically use soil, geological, vegetation and land use maps to define regions to identify regions with consistent calibration parameters, within which the same calibration parameters are used for simulation operations.

The parameters that need to be extracted from the data source include: soil volume weight, soil internal friction angle, dimensionless cohesion, water conductivity coefficient T of soil body, flow rate in stable state, namely effective precipitation R and terrain slope. The terrain gradient can be calculated by a DEM digital elevation model, and part of soil parameters refer to relevant research and experimental results of the area according to the acquired data source. The water diversion coefficient T of the soil body is related to the permeability coefficient K of the soil body and the thickness of the soil body, and the T/R is taken into consideration as an integral factor in model calculation. The soil Saturation (conservation) is mainly determined by two variables of unit catchment area a and T/R, wherein a can determine the value of each pixel through a digital elevation model DEM, so that the change range of T/R can be determined through the soil Saturation and the unit catchment area. Generally, the saturation of the soil mass is less than or equal to 1, but when the saturation is greater than 1, we generally consider that surface runoff is present. The value range of the parameter T/R can be determined according to the actual surface water system distribution estimation rate.

After the calibration area is established and the parameters are calibrated, landslide point data needs to be imported into the calibration area. The method comprises the steps of carrying out multi-factor variance analysis on relevant data and parameter variables of the landslide point known in the region based on machine learning, verifying the contribution degree of each parameter to the landslide disaster danger or judging whether the parameter plays an important role, and accordingly determining the applicability and the reliability of a model landslide danger division.

So-called multi-factor anova is commonly used to study the significance of the effect of two or more control variables on observed variables (the risk of landslide hazard is typically an observed variable under multi-factor effect), and is also considered to be a correlation study. However, the multi-factor analysis of variance does not simply analyze the independent influence of each control variable on the observed variable, but also includes the interaction of a plurality of control variables into an analysis range in the analysis process, so as to comprehensively analyze the influence degree of each factor. The essence of multifactorial anova is statistical inference, whose theoretical basis and steps of computation are similar to those of hypothesis testing.

The multi-factor analysis of variance comprises the following steps: (1) the original hypothesis is put forward, that is, the control variables and the interaction thereof do not have significant influence on the observed variables (assuming that the mean values of the populations of the observed variables at different levels of the control variables have no significant difference, and the effects of the control variables and the interaction effect are 0 at the same time); (2) calculating and analyzing the decomposition of the variance of the variables; (3) comparing the square of the total dispersion of the observation variables with the proportion of each part, and calculating the observation value of the test statistic and the accompanying probability P value; (4) determine significance level and make a judgment: the calculated values of the association probability P of each factor are compared with the significance level (typically 0.05). If the accompanied probability P value of a certain factor is greater than the significance level, the original assumption is established, and the factor is determined to have no significant influence on the observed variable; on the other hand, if the accompanying probability P value of a certain factor is equal to or less than the significance level, the original assumption is not satisfied, and the factor is considered to have a significant influence on the observed variable.

(6) And (3) performing model simulation calculation and evaluation, and primarily evaluating the instability risk of the shallow slope of the line engineering corridor. And calculating a preliminary result by using the landslide disaster point distribution investigation data evaluation model and properly adjusting parameters.

After all necessary parameters of the model are calibrated, the model is used for carrying out simulation operation. Firstly, the stability of the surface shallow slope under the annual average precipitation condition of the corridor area is evaluated, and an evaluation result graph is output. And meanwhile, comparing and analyzing the evaluation result by utilizing the actual terrain distribution and the landslide point data. And further utilizing actual landslide data and the slope-catchment area to manufacture an evaluation graph and a regional landslide risk partition statistical table. And evaluating the coincidence degree of the regional stability evaluation result based on the geographic space information model and the actual landslide point field evaluation result, and further verifying the reliability and the applicability of the model simulation result.

(7) And analyzing and determining the characteristic value of the precipitation inducing the instability of the shallow slope of the line engineering corridor. And carrying out mathematical statistical analysis on the distribution, the risk grading and the change of the landslide disaster points under different rainfall amounts.

Precipitation is the most common direct cause of shallow landslide disasters, but precipitation under all conditions does not cause landslide disasters, so that it is very necessary to analyze landslide disaster risks under different precipitation conditions of a line corridor. The model is mainly used for evaluating a landslide area of a shallow layer of the earth surface caused by precipitation, and the precipitation R in a calibration area is an important factor for evaluation. When strong precipitation is carried out for a short time or continuous precipitation is carried out for a long time, water permeates into a soil slope body, and simultaneously, the water level of underground water rises to soften a rock-soil body, so that landslide is caused. Meanwhile, the precipitation can cause the water level of surface water to rise, so that the soil body is washed to cause landslide. In order to provide an actual reference basis for monitoring, early warning and prevention of landslide disasters of a line, landslide risk evaluation needs to be carried out under the conditions of different rainfall amounts, and line sections needing important monitoring are further determined.

In general, in summer (same as rainy season), the rainfall is increased rapidly under the influence of seasonal wind, so that the average daily rainfall in rainy season is selected as one of the conditions. In addition, stability evaluation under the worst condition needs to be considered, and the average maximum daily rainfall for many years needs to be selected as a reference for simulation calculation so as to provide guidance for disaster early warning work under extreme weather. In addition, the condition of 15mm/d of daily rainfall under the condition of medium rain (defined as the precipitation of 10-24.9 mm within 24 hours by meteorology) is introduced. And (3) carrying out risk evaluation on the shallow landslide on the earth surface of the corridor area by using different precipitation conditions, and calculating and analyzing the area and proportion of each partition and the number and proportion of landslide points in the area under different precipitation conditions.

(8) And (4) carrying out sectional evaluation on the line by utilizing the slope instability risk of the corridor area under different precipitation conditions. Superposing slope instability risk evaluation graphs of the line and the corridor area, and reclassifying line sections located in different risk regions by using a buffer area analysis and partition statistical method: slope unstable region and high unstable region-high risk section; unstable zone-hazardous zone; metastable zone-metastable segment; stable zone-stable segment; high stability zone-safe section. And analyzing the slope instability danger regions of the line under different precipitation conditions, carrying out sectional statistics on the length and the proportion of each danger evaluation section under each precipitation condition, and determining key monitoring line sections in different seasons.

The monitoring and early warning scheme generally adopts manual monitoring, and a wireless sensor network, a distributed optical fiber sensing technology and the like are arranged in an important dangerous area.

Firstly, manual monitoring

The long and large linear engineering manual monitoring cost is high, certain dangerous sections (such as sections close to residential quarters or section stations) which are convenient to monitor manually can be determined according to the risk assessment result, and monitoring and data collection are carried out on key sections regularly or the monitoring frequency is increased in a special period. Furthermore, since the route extends along rivers and their branches in mountainous areas, such road sections serve as key areas for daily manual inspection and monitoring, especially in rainy seasons and short periods of heavy rainfall. In order to reduce the cost, a mode of cooperative monitoring with related departments (such as village offices) can be adopted for certain high-risk road sections close to residential sites.

② monitoring by wireless sensor network

The wireless sensor network monitoring technology needs to lay monitoring sensors at a landslide point and establish a network, and the early investment is large but the daily operation cost is low. The wireless sensor network is arranged, so that the labor cost can be greatly reduced, and the life safety of workers in a dangerous period can be guaranteed. Should be arranged mainly in red high-risk section, only need set up measuring point at the entrance to a cave for the high-risk section that the tunnel passes through.

According to the scheme, the method is characterized in that a WebGIS system for early warning of the line engineering corridor shallow slope instability disaster is established based on the line landslide disaster danger subsection assessment result information and the line actual condition of the model and combined with the line engineering geographical geological environment and the line engineering information parameters, and command decision guidance is provided for line landslide disaster monitoring and early warning.

According to the scheme, the WebGIS system is different from a traditional or general GIS-based early warning system, and is an online network system which is used for sending early warning and monitoring analysis data and landslide danger information of slope instability disasters of a line engineering corridor, realizing real-time intercommunication and sharing of multi-platform terminal data and information, realizing real-time scheduling of field work and real-time uploading of monitoring data by using field tool software and efficiently publishing and sharing information for the early warning and monitoring work of the line landslide disasters.

Conventional manual monitoring and early warning work is often in an inefficient mode of operation. Generally, field workers need to record related information in a book after completing field patrol and survey in a responsible section, and then, data and information are summarized after returning to a station, so that a field monitoring task is completed. In the process, field workers often cannot accurately mark points or sections where landslide is likely to occur in real time, the whole process is time-consuming and labor-consuming, and the real-time early warning effect cannot be achieved under emergency conditions. The efficiency and accuracy of landslide disaster monitoring and early warning work are important. The field inspection mode is urgently required to be changed, a convenient real-time digital and networked means is required to improve the efficiency and accuracy of field monitoring and early warning, and the conversion can be realized by means of WebGIS technical application. The WebGIS technology is different from a conventional GIS-based early warning system (generally a single-machine single-platform terminal system), and the core is to embed an HTTP standard application system in the GIS so as to expand and perfect the application of the GIS technology and realize the management and the release of spatial information under the Internet environment. With the continuous popularization of smart phones, the Web browser at the mobile phone end can realize all functions of the WebGIS at present, any internet user in the global range can access various GIS services provided by the WebGIS, and data can be updated by multiple people at the same time to realize real-time interaction. The traditional field work based on paper reports or photo shooting cannot respond to sudden events such as landslide disasters in time due to low efficiency and low accuracy, and the field work based on the WebGIS technology has obvious advantages in line corridor landslide disaster monitoring and early warning in cooperation. The method adopts a free open source WebGIS series software platform: openlayers, GeoServer, PostGIS, and QGIS.

When landslide hazard monitoring and early warning is carried out, field workers can use a smart phone or a tablet personal computer as a mobile terminal, obtain work tasks through a platform system, work, report progress and seek help to keep efficient work under organization of a rear end easily and conveniently, compared with the field monitoring and early warning work relying on reports in the past, the accuracy and the efficiency of work results are much better, and digital operation is completely realized. Monitoring personnel can check the attribute table of the elements of the landslide mass at the mobile terminal of the smart phone and can also change attribute information. And analyzing and calculating the element data according to the data calculation processing plug-in. By means of positioning of the mobile terminal device and application of the WebGIS platform, each worker engaged in disaster monitoring and early warning can be converted into a mobile monitoring and early warning sensor, landslide disaster data are transmitted in real time through manual monitoring, workers conducting field investigation are coordinated according to the landslide disaster data, authority for accessing required data is provided for field workers, and field work efficiency is improved by combining methods of work management, personnel scheduling and the like. The maintenance personnel can guide the working personnel to carry out on-site inspection and monitoring in real time by processing the real-time data and uploading or updating the related data or information, and the on-site working personnel can directly upload the on-site data through the mobile phone terminal, provide real-time information for decision-making personnel and make corresponding line adjustment arrangement or disaster prevention and reduction response plan.

Compared with the prior art, the invention has the following beneficial effects:

the method is particularly used for analyzing and considering the sectional evaluation of the danger of the slope instability disaster of different sections of the line corridor under different precipitation quantities aiming at the precipitation quantity triggering main factor. Under the technical support of a network geographic information system, a WebGIS system for assessing and early warning of landslide geological disaster in a line engineering corridor is established, can be used for assessing and dividing the danger of the landslide disaster in sections of the line engineering, can be used as a decision basis for a line geological disaster safety prevention and control project, and can provide more powerful technical support for monitoring and early warning of the landslide disaster, pre-assessing loss of the landslide disaster and preventing and controlling. It should be noted that the model is applicable to shallow landslide phenomena and not to deep geological formations or seismic fracture induced slope instability. In application, the model does not need accurate input of soil and climate characteristic data, and determines and maps the stability of the earth surface slope in a large range according to medium-precision data or parameter information, so that the area needing to be investigated and evaluated in more detail is rapidly determined. The output result of the model depends on the precision of a Digital Elevation Model (DEM), and the accuracy of the simulation prediction result can be improved with high precision. Furthermore, to some extent, the accuracy of known landslide survey data. Therefore, as accurate a DEM as possible should be acquired along with detailed known landslide data.

Drawings

FIG. 1 is a flow chart of example processing steps;

FIG. 2 is a diagram of an embodiment corridor area and a known landslide point;

FIG. 3 is a diagram of the soil saturation and model simulation evaluation in corridor areas;

FIG. 4 is a diagram illustrating the risk of landslide in a corridor under different precipitation levels;

FIG. 5 is a sectional evaluation diagram of the landslide risk of the line under different precipitation amounts;

FIG. 6 is a diagram of a line corridor landslide risk assessment focus monitoring section;

fig. 7 is an interface diagram of PC end and mobile end of line corridor landslide risk assessment and early warning WebGIS system.

Detailed Description

The present invention will be further described with reference to the following description and examples, which include but are not limited to the following examples.

In the embodiment, a corridor passed by a certain railway line section is used as an implementation area, and the area is high in mountains, steep in slopes, deep in canyons, complex in geology and abundant in rainwater, and is one of the areas with the most serious landslide geological disasters. Rainfall in the line section has a great influence on landslide. The effect of rainfall on the landslide is mainly shown in that a large amount of rainwater seeps downwards to cause the saturation of a soil and stone layer on the slope and even water is accumulated on a water-resisting layer at the lower part of the slope, so that the weight of the landslide body is increased, the shear strength of the soil and stone layer is reduced, and the landslide is caused. In order to ensure smooth and safe operation of railway lines and prevent landslide geological disasters, landslide geological disaster investigation and danger evaluation are carried out on an existing railway corridor. According to the method, remote sensing information, topographic and geomorphic data, geology, soil, vegetation, landslide hazard investigation data, hydrology, climate and other factor data are integrated, a corridor area evaluation model is constructed on the basis of the computational theory of landslide stability indexes in the method, landslide hazard risk assessment is carried out on a line corridor on the basis of the corridor area evaluation model, landslide hazard risk of the railway line section is evaluated in a segmented mode, a line engineering corridor landslide hazard early warning WebGIS system is built, line landslide hazard information is published and shared on line, internal and external investigation monitoring and collaborative early warning are carried out, and decision basis and method guidance are provided for railway existing line landslide hazard monitoring and early warning work.

The main implementation steps comprise the following steps (see attached figure 1): a, line engineering corridor area investigation and multi-source space-time basis and comprehensive database construction; b, comprehensively analyzing and determining key influence factors of slope instability and pregnancy disaster; c, processing and analyzing the influence factor indexes of the slope pregnancy disaster and assigning standardized numerical values; d, establishing a risk evaluation model of the line corridor shallow landslide; e, primarily evaluating the instability risk of the shallow slope of the line engineering corridor; f, analyzing and determining a precipitation characteristic value for inducing the instability of the shallow slope of the line engineering corridor; g, evaluating the instability risk of the shallow slope of the line corridor in a segmented manner; h, establishing a WebGIS system for early warning of slope instability of the line engineering corridor. The main process is detailed below.

1 line engineering corridor area investigation, data preparation and processing analysis

A corridor is an area within a certain range on both sides of a railway line section, as shown in fig. 2 a. The north is high and low, the river valley is high and deep, and the topography is big. The landform morphology is divided into tectonic erosion middle mountain areas, tectonic erosion low mountain areas and river valley mound dam areas. The surface water system is developed, the river channel is bent and extended, the river bank is greatly influenced by the river scouring, the rainfall is abundant, and geological disasters are easy to occur; the terrain in the north is relatively flat, but the terrain is rugged in three sides. The middle part is distributed along the riverside, and the topography is very flat. However, the mountain terrains on the two sides are high, and the risk of geological disasters is brought to the marginal areas of the urban area. The southwest also extends along high and deep valleys, but is more open than the northern valleys.

The section is a transition zone from a mountain region to a basin, the altitude change is large, and the gradient is large. The climate is a typical subtropical monsoon humid climate, rainfall is abundant and concentrated, and the soil scouring effect is obvious; in addition, the water system is developed and the side-stream watershed thereof has abundant underground water, thereby providing sufficient conditions for the development of landslide disasters. Meanwhile, the vegetation coverage in the corridor area is more, and the forest coverage rate exceeds 56.81%. Due to abundant rainfall in the corridor area, the lithology of geological strata is poor, and in addition, seismic activity in the area is frequent, so that geological disasters are frequent. The main causes for inducing unfavorable geological phenomena in the area are two, one is that the strength of rock and soil bodies is reduced due to strong precipitation or groundwater activity, hydrodynamic pressure and pore water pressure are generated to erode rock and soil bodies, the volume weight is increased, buoyancy or supporting force is generated on permeable rock strata, slope displacement is caused, and geological disasters are caused. The other is that the earth crust movement generates power to cause geological disasters, and the common phenomenon is earthquake action.

The landslide risk assessment model for the shallow layer of the earth surface considers factors such as vegetation coverage, soil type, rainfall, hydrological conditions, gradient and the like. The DEM, soil, vegetation coverage, precipitation and climate characteristic data in the area are required to be utilized, and meanwhile, the actual landslide point survey data in the area is required to be verified. Data preparation and processing analysis for the corridor area of this example is as follows:

digital elevation model DEM

The high-precision line corridor area DEM can utilize helicopter/unmanned aerial vehicle airborne three-dimensional laser scanning point cloud to generate a digital elevation model, and can obtain a high-resolution multiband optical remote sensing image. In the embodiment, the digital elevation DEM data with the spatial resolution of 12.5m is adopted, and the DEM in the range of two sides of the railway line corridor area is extracted through splicing, coordinate conversion and a buffer area. And subsequently, pit filling correction needs to be carried out on the original DEM, and the slope and the flow direction of the corridor area are calculated by using the corrected DEM. The gradient can be directly calculated according to the DEM, the region with larger gradient in the northern mountain area is more, and the gradient change of the northern region is larger; the slope of the urban area in the middle is small, and the slope change is small. And calculating the flow direction by using the gradient data after the gradient calculation is finished. According to the flow diagram, obvious valleys and mountains, approximate water distribution lines and water collection lines of the mountains can be seen, and the trends of the mountains and the rivers are basically consistent with those of the mountains and the rivers in the actual remote sensing image diagram.

(ii) landslide survey data

Landslide data is derived from a geological disaster information management platform developed by the geological survey bureau of China, and comprises historical landslide points and landslide points in monitoring. And inquiring the geological disaster information management platform, and inquiring and generating a landslide point data set (shown in figure 2b) at 61 positions in the line corridor by utilizing a geographic information technology. According to the field investigation and evaluation of landslide points by a geological disaster information management platform, stability classification is carried out on 61 landslide points, and different code values are given according to classification: there were 13 sites with good stability, 40 sites with poor stability, and 8 sites with poor stability.

③ type of superficial soil of earth's surface

The data of soil and vegetation types in the region are from the resource and environment science data center of the Chinese academy of sciences. Wherein the soil type is based on the spatial distribution data of the Chinese soil type.

The soil type in the region is mainly iron-bauxite and purple soil, and some artificial soil is slightly mixed. Wherein the iron-bauxite is mainly red soil, yellow red soil and rinsing yellow soil; the purple soil is mainly acid purple soil; artificial soils include rice soil and dredged soil. The soil texture in the area is sticky and heavy, organic matters are deficient, the long-term influence caused by regional common hydrological action erosion such as rainstorm and the like is caused, the soil layer in the area is shallow, and the superficial movement of the earth surface is easily caused by external force such as strong precipitation, earthquake action and the like.

Earth surface vegetation cover type

The vegetation types are based on the Chinese vegetation coverage zone data. The vegetation types in the regions are mainly temperate subtropical deciduous broad-leaved shrub and shrub dwarf forests, and a small amount of artificially cultivated vegetation (rice and other crops) is mixed in the vegetation types.

Years precipitation data

Precipitation data is from the China Meteorological data network (China ground climate data daily data set V3.0) under the national Meteorological science data center. Selecting daily rainfall data of three monitoring stations in the corridor as a rainfall data source of the corridor area. The precipitation in the zone has the following characteristics. 1) The sufficient precipitation provides a large amount of water source supplement for the generation of geological disasters such as landslide, mud-rock flow and the like. 2) The precipitation amount has large seasonal variation. The rainfall is more in summer and less in spring and winter, the rainfall is large and concentrated in flood season, namely, the rainfall is large and concentrated in summer, and the rainfall is small and small in spring. Precipitation is mainly concentrated in 6 to 10 months of the whole year, which accounts for about 75.3 percent of the annual precipitation, and precipitation in two months of only 7 to 8 months accounts for 46.1 percent of the annual precipitation, and approaches to one half of the annual precipitation. Therefore, the precipitation time distribution in the corridor area is extremely uneven, the precipitation amount in the flood season is very concentrated, and the distribution rule is very easy to cause short-time instantaneous storm flood disasters and further easily cause geological disasters such as landslide and debris flow. Besides the abundant surface water resources, the groundwater resources in the region are directly supplied by river water, the water quantity is abundant, and the water is mainly used for flushing the holes of the flood sand gravel layer of the fourth system to dive. Abundant water resource storage and activities in the corridor area provide conditions for the generation and development of various geological disasters, particularly landslide disasters.

The parameters that need to be extracted from the data source include: soil volume weight, soil internal friction angle, dimensionless cohesion, water conductivity coefficient T of soil body, flow rate in stable state, namely effective precipitation R and terrain slope. The terrain gradient can be calculated by a DEM digital elevation model, and partial soil parameters are obtained according to the acquired data source and by referring to the related existing research results of the area.

Because the area range of the embodiment is not large, on one hand, the area basically belongs to the same landform and has similar engineering geological conditions; on the other hand, the input values of the geological parameters in the model are determined by using the interval range values, and a region with completely consistent properties is not required, so that the corridor region of the embodiment can be divided into the same calibration region.

Table 2 shows the soil data for selected bauxite and purple soil, and table 3 shows other relevant parameters.

TABLE 2 Main soil parameters of corridor areas

TABLE 3 associated parameter definitions

Parameter(s)	Gravity (m/s)2)	Soil Density (kg/m)3)	Water density (kg/m)3)
				Value of parameter	9.81	2050	1000

Through the test and calculation of the soil saturation, the spatial distribution condition of the area with the soil saturation greater than 1 is found to be very similar to the real water system distribution of the corridor area when the T/R is about 2000-3000. Therefore, the upper and lower limits of the parameter T/R in the corridor area model are preliminarily determined to be 2000 and 3000. The values of the parameters are given in Table 4.

TABLE 4 calibration zone parameter settings

Model 2 multifactor analysis of variance

And performing machine learning multi-factor variance analysis on parameters such as catchment area, gradient and soil humidity of the landslide point. In order to verify the effect of multi-factor analysis of variance, Elevation of the landslide point to ELE are introduced as additional reference factors to be analyzed simultaneously. In the analysis, the soil humidity SA, the catchment area CA, the slope SLP and the elevation ELE are regarded as factors influencing the graded TYPE value of the landslide point (1 is known to be good in stability, 2 is known to be poor in stability, and 3 is known to be poor in stability), and variance analysis is carried out on the factors.

Firstly, converting various types of data in the region into grid values with the same size, superposing landslide points on corresponding grid data sets, endowing the landslide point elements with the grid values, extracting the factor data of each point, and sorting the factor data according to the data after landslide hierarchical arrangement. The data were subjected to machine learning multi-factor analysis of variance using Python programming language based on mathematical models in the Pandas library, the results of which are listed in table 5.

TABLE 5 Multi-factor ANOVA results

According to the analysis result, the p values of the factors SA, CA and SLP of the known landslide point in the corridor area are 1.443689 multiplied by 10 respectively^-3、8.318185×10^-3And 9.705341 × 10^-7The values are far less than 0.05 (reaching significance level), which indicates that the original hypothesis can be rejected, and three factors, namely SA, CA and SLP have significant influence on the risk level of landslide; and the p-value 0.6265117 of the factor ELE>0.05, the original hypothesis cannot be rejected, indicating that the elevation of the landslide location has no significant effect on its risk rating. Through multi-factor analysis, the important influence of factors such as catchment area, gradient and soil humidity in the model on the danger of the landslide point in the corridor area is further verified, and the model has better applicability in the area.

3 simulation results

After all necessary parameters of the model are calibrated, the model is used for carrying out simulation operation. Firstly, the stability of the surface shallow slope under the annual average precipitation condition of the corridor area is evaluated, and a stability result graph is output. And meanwhile, comparing and analyzing the evaluation result by utilizing the actual terrain distribution and the landslide point data. FIG. 3a shows the saturation of the soil in the corridor area obtained by the digital elevation model DEM coupled with the TOPMODEL algorithm. The spatial distribution characteristics of the soil saturation are analyzed, so that the region with higher soil saturation (terrain humidity) is mostly present in a relatively low-lying region of the terrain, and is consistent with the distribution condition of the surface water system in the actual corridor region. An evaluation chart (fig. 3b) was further created using the actual landslide data and the slope-catchment area. The comparative analysis shows that the classification of the landslide points of the actual landslide point investigation condition is basically consistent with the stability classification of the initial evaluation by utilizing the model simulation: 8 landslide points (red dots) with poor stability are distributed in a region with SI less than 0.5, namely a highly unstable region; 40 landslide points (red squares) with poor stability are distributed in the region with 1.25 & gt SI & gt 0.5 as a whole, namely unstable and quasi-stable regions; the landslide points (red triangles) with good stability at 13 are generally distributed in the area with SI > 1.25, namely a stable area and a high-stability area.

FIG. 4a is a calculation result of model simulated landslide stability index. As can be seen from the map of the surface stability index zone, the terrain is flat in the middle area and is basically a highly stable area with SI more than or equal to 1.5; the northern area is mountainous and is mostly a highly unstable area and an unstable area; the middle part of the southwest area is along the river valley, so that the stability is better; and the landforms of the two side mountains are complex and have poor stability. Therefore, the coincidence degree of the slope stability evaluation result of the corridor area and the actual landslide point is high, the simulation result is further verified to have high reliability, and the method has applicability.

Table 6 is a statistical table of the risk zones of landslide in the corridor area. According to statistics, the area of the highly unstable region in the corridor region accounts for 10.6% of the total area, the distribution is dispersed, and the highly unstable region is mainly concentrated in a mountain region with a large gradient. The area of the unstable region accounts for 45 percent of the total area, is the region with the largest proportion, is mainly distributed at the periphery of the highly unstable region, and has wider coverage area. The proportion of the quasi-stable area to the stable area is respectively 13.3 percent and 7.9 percent, the distribution is dispersed, and the distribution is mainly at the junction of the unstable area and the high-stability area. The high stability zone area accounted for 23.2% of the total zone area, mostly concentrated in the urban zone in the middle of the corridor zone. It can be seen that the stability regions calculated by simulation have good consistency with the actual terrain and disaster spatial distribution conditions.

Table 6 line corridor landslide danger zoning statistical table

Grading	Are highly unstable	Instability of the film	Quasi-stability	Stabilization	High stability	Total up to
							Area of area (KM)2)	226.2	962.8	285.5	169.5	497.0	2141
Ratio (%)	10.6	45.0	13.3	7.9	23.2	100
							Number of landslides	8	42	8	8	13	79
Ratio (%)	9.9	51.9	9.9	9.9	16.0	97.6

4 line corridor landslide hazard under different precipitation

According to a rainfall data of a Chinese ground climate data daily value data set (V3.0), annual average daily rainfall, summer average daily rainfall, annual maximum daily rainfall and medium rainfall 15mm day rainfall are extracted, T/R values under various rainfall conditions are calculated, and statistical data and calculation results are listed in a table 7.

TABLE 7 different precipitation conditions and T/R values

And (3) carrying out risk evaluation on the shallow landslide of the earth surface in the corridor area by using different precipitation conditions, wherein evaluation results output by the model are shown in a figure 4 (the daily average precipitation is arranged from small to large in sequence). It can be seen that as the precipitation amount increases, the black unstable area gradually increases, and shallow landslide of the ground surface is highly likely to occur. The cyan region, i.e., the highly stable region, is reduced, but the central region thereof is not greatly changed. The area and the proportion of the area and the number and the proportion of the landslide points in the area under four rainfall conditions are analyzed and calculated, and the area proportion of the area landslide area of the area where the superficial layer landslide of the earth surface occurs can be predicted along with the increase of the rainfall, and is increased from 0.8% of the average daily rainfall per year to 6% of the average maximum daily rainfall per year. Once the soil body of the landslide is damaged and flows, external force damage is often caused to the area through which the soil body flows, and therefore secondary landslide of a highly unstable area or an unstable area is caused. The area ratio of the highly unstable area is also changed greatly, the area ratio is increased from 10.5% to 24.3%, the amplitude is increased to 131.4%, and the influence of precipitation on the landslide risk of the superficial layer of the earth surface in the corridor area is very obvious.

The area ratio of the unstable region increased gradually from 44.6% to 48.1% at the beginning with the increase in precipitation (2.6mm/d to 5.7mm/d), but decreased back to 44.6% at the maximum precipitation. And combining the area change of other areas to show that the highly unstable area with larger gradient is converted into the protective area and part of the unstable area is converted into the highly unstable area along with the increase of precipitation. Correspondingly, the three more stable regions gradually decrease and transform into unstable regions, resulting in substantially unchanged overall area of the unstable regions.

When the precipitation amount is just increased (2.6mm/d-15mm/d), the area of the height stable area is greatly reduced (23% -15.8%), but as the gradient of the height stable area is not large, the terrain is relatively flat and the like, the area ratio change is basically small (15.8% -15.1%) along with the increase of the precipitation amount (15mm/d-21.8 mm/d). The reason for this is that the water content of the soil is continuously increased with the increase of the precipitation amount, but the stability change caused by the increase of the precipitation amount is very small because the gradient of the soil body in these areas is small as a whole.

With the increasing of the precipitation amount, the number of the landslide points which are divided into the highly unstable area and the landslide area is increased continuously, the occupied proportion is increased gradually, the trend of landslide disaster is further increased, and the danger of the whole area is increased gradually. The area ratio of the unstable area is not changed greatly, the areas of the highly unstable area and the landslide supporting area are increased, and the areas of the highly stable area, the stable area and the quasi-stable area are reduced. This is mainly due to precipitation causing a large amount of water to seep into the sloping soil in a short period of time and causing a redistribution of the water content in the soil. The increase of the water content of the soil body and the infiltration process thereof have double influences on the volume weight and the stability coefficient of the soil, and can cause the change of the strength and the effective stress of the soil. When the saturated permeability coefficient of the slope soil body is reached, the soil layer above the underground water level is sheared and damaged. Although there are cases where the volume weight of the soil body is increased due to the increase of the water content of the soil, and the stability is increased in a small amplitude, the effect is generally much smaller than the stability loss caused by other factors.

The method comprises the following steps of utilizing the landslide danger of corridor areas under different precipitation conditions to carry out subsection evaluation analysis on railway lines, superposing a certain railway line and the landslide danger evaluation graph of the corridor areas, and reclassifying line sections located in different danger zones by utilizing a buffer area analysis and zone statistical method: landslide area and high unstable area-high risk section; unstable zone-hazardous zone; metastable zone-metastable segment; stable zone-stable segment; high stability zone-safe section. The obtained evaluation of the risk of landslide of the line segment is shown in fig. 5 (in ascending order of precipitation).

Comparing the line danger evaluation graphs under four rainfall conditions in sequence, it can be seen that the line length of the high-risk section is gradually increased along with the increase of 24h rainfall, but the high-risk section is mainly concentrated on the northern mountain area, namely a section from a beach station to a section outside a Guangyuan city area, and a section from a bamboo garden dam station at the tail end of a southwest line to the west. The main stations are basically located in the safe section except the beach station in the northern mountain area, and are slightly influenced by landslide. The bamboo garden dam station, the sword door closing station and the Guang Yuan station are flat in terrain and basically located in a safety section, and even if the precipitation is continuously increased, the probability of landslide disasters around the line is very small. Further, the length and the occupation ratio of the risk assessment segment under each precipitation condition can be counted, for example, table 8 is a line risk segment statistics under the annual average precipitation condition.

TABLE 8 segmented statistics of line dangerousness under average precipitation conditions

	High risk segment	Danger section	Quasi-stable segment	Stabilization segment	Safety section	Total up to
							Length (m)	17404	35331	12759	10900	41533	117927
Ratio (%)	14.8	30.0	10.8	9.2	35.2	100

It can be seen that even the evaluation result under the annual average precipitation condition, the high-risk section and the dangerous section occupy 44.8% of the total length, namely, the section of a certain railway line in the corridor area is greatly influenced by landslide disasters. Therefore, when the landslide hazard risk segmentation is performed on the line, the landslide hazard change of the line under various rainfall conditions or even extreme rainfall conditions should be considered.

5 determining line landslide danger key monitoring section

And in combination with the actual condition of the line, monitoring and early warning of landslide disasters of the line are carried out, and key monitoring sections are correspondingly determined according to precipitation in different seasons.

And analyzing and determining a key monitoring section according to the line dangerous area under the precipitation of 15mm day by combining the actual condition of line survey. In fig. 6a, the cyan safety segment, the blue stability segment and the yellow metastable segment are all in a flat terrain region, and no large slope is formed at the periphery, and no landslide point is known. Typically such sections do not require any side slope protection or monitoring. Fig. 6b, the section moves forward along the riverside, a long slope is arranged beside the section, and the whole section is in a high-risk section. By carefully comparing the red high-risk section with the actual line of the line section, it can be seen that a tunnel crossing mode is adopted in part of the high-risk section lines, which indicates that the railway considers the easiness of geological disasters of the road sections in the construction stage, so that the tunnel crossing mode is selected. Because the model aims at the stability of the shallow slope of the earth surface, the influence of landslide on the railway line in the tunnel is small. Although the entrance and exit of the tunnel are also in the high-risk section, the general entrance part of the tunnel is manually supported, so that the high-risk section does not need to be monitored daily, but the entrance is required to be monitored in a rainy season or in heavy rainfall by paying attention to the entrance protection. This again illustrates that the landslide hazard assessment output by the model is realistic.

And (3) combining the landslide risk evaluation result of the line and the actual line image map, excluding a high-risk section which adopts a tunnel to pass through the risk avoidance, determining 7 sections of line key monitoring and early warning sections, and sequencing the sections from north to south and from (i) to (c) along the line, wherein fig. 6c is a position map and an image map of the issued line landslide disaster key monitoring and early warning sections. Generally, manual monitoring is adopted, and a wireless sensor network monitoring technology, a distributed optical fiber sensing technology and the like are arranged in an important dangerous area if necessary.

6 landslide danger information online publishing and sharing

WebGIS publishes and applies geographic spatial data through the Internet to realize the sharing and interoperation of the spatial data, such as publishing the spatial data, spatial query and retrieval, spatial model service, organization of Web resources, making thematic maps, and performing various spatial retrieval and spatial analysis. WEBGIS can adopt a plurality of hosts and a plurality of databases to carry out distributed deployment, interconnection is realized through Internet/Intranet, and the WEBGIS is a browser/server (B/S) structure. The users and servers may be distributed in different locations and on different terminal platforms.

In this embodiment, the releasing of the line landslide hazard information and the field work are cooperatively realized by using a WebGIS technology. Firstly, a Shapefile of line element data of a landslide disaster danger zone of a certain railway line under average daily rainfall, line element data of a landslide disaster danger zone of a wide element section of the certain railway line under 15mm daily rainfall and point element data of a known landslide point of a corridor area is converted into a ZIP file. And importing a WebGIS page taking a 'world image map' as a base map (note that data are unified into a projection coordinate system of the 'world image map' as the base map, otherwise, the data cannot be imported or have deviation), and then selecting the attributes of elements to be displayed according to needs, changing the styles of the elements, and facilitating display or identification. Uploading the landslide disaster danger areas under different precipitation characteristics in the Web Map, adjusting the opacity of the area Map, and using the Web Map as a base Map for other WebGIS applications.

The user can open at any client side following the HTTP network transmission protocol through the webpage link, and the application terminal comprises intelligent equipment such as a smart phone, a tablet, a PC and the like. All lines, landslide point positions and information of each key monitoring section of the embodiment can be checked on line. For example, the specific attributes of the landslide point, the attributes and stability conditions of the section are checked, and the area, distance and position can be measured. The method can search landslide places and positions, perform amplification, default range, reduction, search, printing and sharing, change style buttons, adjust transparency, adjust styles of various attributes of landslide points and sections, link or embed maps, perform spatial analysis, perform cluster analysis on landslide point distribution and the like. The mobile end interface with computer and smartphone are shown in fig. 7(a), (b), respectively.

The embodiment provides a method for establishing a system for assessing, monitoring and early warning of the risk of shallow landslide of a line engineering corridor, and can provide more powerful technical support for prevention and control of disaster of shallow landslide of the line engineering corridor, design of a line engineering channel and safe construction. As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions.

It should be noted that the above-described embodiments may enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the drawings and the embodiments, it will be understood by those skilled in the art that the present invention may be modified and equivalents thereof; all technical solutions and modifications thereof which do not depart from the spirit and technical essence of the present invention should be covered by the scope of the present patent.