A Method for Predicting Landslide Risk Based on Connectivity and Geographic Detectors

Abstract

This invention discloses a method for predicting the risk of landslides based on connectivity and a geographic detector. It employs a discretization approach to transform the comprehensive soil erosion index and influencing factors into characteristic parameters of categorical variables. Multicollinearity analysis is performed on all characteristic parameters to filter out those exhibiting collinearity. The geographic detector is used to calculate the explanatory power of each characteristic parameter for landslide risk in the target area and the frequency of landslide occurrence at sample points within different grids. The product of the explanatory power and frequency of landslide occurrence for different characteristic parameters within each grid is accumulated to obtain a spatial distribution map of landslide risk prediction for the target area. The beneficial effects of this invention are: by combining soil erosion and sediment transport potential, and using a geographic detector to analyze the explanatory power of each characteristic parameter for landslide occurrence, highly accurate landslide risk assessment results are obtained.

Description

Technical Field

[0001] This invention relates to the field of soil erosion and disaster prevention technology, and in particular to a method for predicting the risk of gully collapse based on connectivity and geographic detectors. Background Technology

[0002] Landslides, as a common geological hazard, often have a serious impact on human life and property safety and the ecological environment. With climate change and intensified human activities, the frequency and intensity of landslides are gradually increasing, especially in mountainous and hilly areas. Traditional landslide risk assessment methods often focus on static factor analysis, such as using information content methods, bivariate entropy information methods, and logistic regression analysis. The factors considered mainly include climate (rainfall, temperature), geology, soil, vegetation, topography (slope, elevation), and human activities (land use). They lack a comprehensive understanding of the interaction between soil erosion and sediment transport, resulting in low accuracy and reliability of landslide risk assessments. Summary of the Invention

[0003] To address the aforementioned problems, this invention proposes a method for predicting the risk of hill collapse based on connectivity and geographic detectors, which mainly solves the problem of low accuracy in existing hill collapse risk assessment methods based on static factor analysis.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0005] A method for predicting the risk of hill collapse based on connectivity and geographic detectors includes the following steps:

[0006] Obtain the spatial distribution map of soil surface erosion and the sediment connectivity index of the target area, and normalize the spatial distribution map of soil surface erosion and the sediment connectivity index to obtain the soil surface erosion layer and the connectivity index layer.

[0007] After the soil surface erosion layer and the connectivity index layer are superimposed, the comprehensive soil erosion index of each region in the target area is calculated.

[0008] Obtain spatial distribution information of landslide areas in the target region, randomly collect sample points in the target region, and mark the sample points according to preset rules, defining the sample points as two types: those with landslide points and those without landslide points.

[0009] Identify the influencing factors that are strongly correlated with the occurrence of the collapse;

[0010] The comprehensive soil erosion index and the influencing factors are transformed into typological variable feature parameters by discretization. Multicollinearity analysis is performed on all the feature parameters to filter out feature parameters that are collinear.

[0011] The explanatory power of the aforementioned feature parameters for the risk of landslides in the target area and the frequency of landslide occurrence of sample points in different grids are calculated using a geographic detector. The product of the explanatory power and frequency of landslide occurrence corresponding to different feature parameters in each grid is accumulated to obtain a spatial prediction distribution map of landslide risk in the target area.

[0012] In some implementations, the process of obtaining the spatial distribution map of soil surface erosion includes: taking the digital elevation model of the target area, land use data, and meteorological data as input, using an empirical soil erosion estimation model as the framework, and outputting the spatial distribution map of soil surface erosion.

[0013] In some implementations, the process of obtaining the connectivity index includes: assessing the sediment transport potential based on the obtained data of sediment formation zones and sediment transport zones in the target area to obtain the connectivity index.

[0014] In some implementations, the preset rule is: assign a value of 1 to the sample points that have a risk of collapse, and assign a value of 0 to the sample points that have no risk of collapse.

[0015] In some implementations, the influencing factors include altitude, slope, soil type, lithology and / or land use.

[0016] In some implementations, the frequency of hill collapses represents the proportion of the sample points that have experienced hill collapses.

[0017] In some implementations, the explanatory strength is the importance of the influence factor.

[0018] The beneficial effects of this invention are: by combining soil erosion and sediment transport potential, and using geographic detectors to analyze the explanatory strength of various characteristic parameters for the occurrence of landslides, a highly accurate landslide risk assessment result can be obtained. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the method for predicting the risk of hill collapse based on connectivity and geographic detectors disclosed in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the comprehensive soil erosion index disclosed in an embodiment of the present invention;

[0021] Figure 3 This is a spatial distribution map of landslide risk prediction based on a geographic detector, as disclosed in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the content of this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to this invention are shown in the accompanying drawings, not all of them.

[0023] This embodiment proposes a method for predicting the risk of hill collapse based on connectivity and geographic detectors, such as... Figure 1 As shown, it includes the following steps 1-6:

[0024] Step 1: Obtain the spatial distribution map of soil surface erosion and the sediment connectivity index of the target area, and perform maximum absolute value normalization on the spatial distribution map of soil surface erosion and the sediment connectivity index to obtain the soil surface erosion layer and the connectivity index layer.

[0025] In this embodiment, the target area's digital elevation model, land use data, and meteorological data (annual rainfall and annual temperature) are used as inputs. An empirical soil erosion estimation model is employed as the framework to output a spatial distribution map of soil surface erosion. The empirical soil erosion estimation model can be either the RUSLE model or the MUSLE model. RUSLE (Revised Universal Soil Loss Equation) is a modified universal soil loss equation, a widely accepted and applied empirical soil erosion estimation model. MUSLE (Modified Universal Soil Loss Equation) is a watershed-scale soil erosion calculation model developed by the Soil Conservation Institute in the United States. The RUSLE model primarily estimates soil erosion based on rainfall erosivity factors, while the MUSLE model focuses more on the impact of surface runoff on soil erosion. In this embodiment, the method for obtaining the spatial distribution map of soil surface erosion is not limited.

[0026] The process of obtaining the connectivity index includes: obtaining the connectivity index based on the acquired sediment formation zone (Dup) and sediment transport zone (Ddn) data of the target area, and assessing the sediment transport potential. Sediment connectivity refers to the physical connection of sediment from one region or location to another through a channel system, and the potential for specific particles to move within the system. It reflects the sediment cascade relationship between different geomorphic or landscape units within a watershed and is an important indicator for measuring the sediment movement potential within geomorphic units. Methods for calculating the sediment connectivity index mainly include the index method, model simulation method, and graph theory method. Among these, the connectivity index (IC) in the index method is the most widely used, as it can characterize the potential connection between sediment sources and sinks in a watershed. The model simulation method simulates the sources and sinks of eroded sediment in a watershed by constructing mathematical models, quantitatively characterizing sediment connectivity. The graph theory method can visually display the spatial distribution and dynamic changes of sediment cascades between different landscape units, but its applicable spatial scale is limited. In this embodiment, the method for obtaining the sediment connectivity index is not limited.

[0027] Step 2 involves overlaying the soil surface erosion layer and the connectivity index layer to calculate the comprehensive soil erosion index for each region within the target area. The core of this step is to incorporate the sediment transport potential, represented by sediment connectivity, into the soil erosion system. Figure 2 As shown.

[0028] Step 3: Obtain the spatial distribution information of the collapsed hill areas in the target area, randomly collect sample points in the target area, and mark the sample points according to preset rules, defining the sample points as two types: those with collapsed hills and those without collapsed hills.

[0029] The preset rule is: assign a value of 1 to sample points with the risk of collapse and a value of 0 to sample points without the risk of collapse. Treating the phenomenon of whether or not a collapse occurs as a binary variable helps to identify and quantify the key factors affecting the occurrence of collapse and facilitates decision-making.

[0030] Step 4: Obtain the influencing factors that are strongly correlated with the occurrence of the collapse.

[0031] Influencing factors include altitude, slope, soil type, lithology and / or land use.

[0032] Step 5: The comprehensive soil erosion index and influencing factors are transformed into characteristic parameters of categorical variables using a discretization method. Multicollinearity analysis is then performed on all characteristic parameters to filter out those exhibiting collinearity.

[0033] In step 5, by using collinear feature parameters, the instability of the model can be reduced, the robustness of the model can be improved, and the individual impact of each influencing factor on the subsequent output prediction distribution can be more clearly understood, thereby improving the interpretability of the model.

[0034] Step 6: Utilize the geographic detector to calculate the explanatory power of feature parameters for landslide risk in the target area and the frequency of landslide occurrence for sample points within different grids. Accumulate the product of the explanatory power and frequency of landslide occurrence for different feature parameters within each grid to obtain a spatial prediction distribution map of landslide risk for the target area, such as... Figure 3 As shown.

[0035] Wherein, the frequency of collapse (r) represents the proportion of sample points with collapse, and the explanatory strength is the importance (q value) of the influencing factor.

[0036] In this embodiment, by detecting the explanatory power of different characteristic parameters for geographical phenomena, a geographic detector is used to effectively identify key factors influencing ridge collapse. Specifically, the geographic detector analyzes the relationship between each factor and ridge collapse occurrence, and the risk assessment result is obtained by multiplying the importance q-value (explanatory strength) of each factor by the frequency of ridge collapse occurrence at each grid location. By multiplying the q-value by the frequency of ridge collapse occurrence, the risk of ridge collapse occurrence in each grid can be quantified, aiming to spatially identify areas with high ridge collapse risk and provide a scientific basis for ridge collapse prevention and decision-making.

[0037] Taking Huacheng Town as an example, frequent landslides in this area have severely impacted the local ecological environment and economic development. Soil erosion in this region is mainly caused by hydraulic and gravitational forces, resulting in significant soil loss. To effectively assess the risk of landslides, this embodiment utilizes a landslide risk prediction method based on connectivity and geographic detectors. High-resolution remote sensing imagery was used to identify 2146 landslide sample points through a machine learning model, and 1806 non-landslide sample points were manually added to establish a database of landslide-affected and non-landslide samples. Influencing factors included altitude, slope, aspect, lithology, rainfall, temperature, vegetation, and the comprehensive soil erosion index. The results show that the comprehensive soil erosion index, aspect, and elevation ranked in the top three, with an area under the participant characteristic curve (AUC) value of 0.851. The research results can provide a reference for soil and water conservation and landslide prevention work in this area.

[0038] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for predicting the risk of gully occurrence based on connectivity and geographic probes, characterized in that, The method comprises the following steps: obtaining a soil surface erosion spatial distribution map and a sediment connectivity index of a target region, and performing normalization processing on the soil surface erosion spatial distribution map and the sediment connectivity index to obtain a soil surface erosion layer and a connectivity index layer; after superimposing the soil surface erosion layer and the connectivity index layer, a comprehensive soil erosion index of each region in the target region is calculated; obtaining spatial distribution information of a gully region in the target region, randomly collecting sample points in the target region, and marking the sample points according to a preset rule, wherein the sample points are defined as two types of gully points and non-gully points; obtaining an influence factor that is strongly correlated with gully occurrence; using a discretization method to convert the comprehensive soil erosion index and the influence factor into characteristic parameters of type variables, performing multiple collinearity analysis on all the characteristic parameters, and filtering characteristic parameters that have collinearity; using a geographic detector to calculate the explanatory strength of the characteristic parameters in the target region in terms of gully occurrence risk and the gully occurrence frequency of sample points in different grids, and accumulating the product of the gully occurrence explanatory strength and the gully occurrence frequency of different characteristic parameters in each grid to obtain a gully risk spatial prediction distribution map of the target region; the process of obtaining the soil surface erosion spatial distribution map comprises: using a digital elevation model, land use data and meteorological data of the target region as input, using an empirical soil erosion estimation model as a skeleton, and outputting the soil surface erosion spatial distribution map; the influence factor comprises altitude, slope, soil type, lithology and / or land use.

2. The connectivity and geographic probe-based risk prediction method of gully erosion occurrence according to claim 1, wherein, the process of obtaining the sediment connectivity index comprises: evaluating sediment transport potential according to obtained sediment formation area and sediment transport area data of the target region to obtain the sediment connectivity index.

3. The connectivity and geographic probe-based risk prediction method of gully erosion occurrence according to claim 1, wherein, the preset rule is that the sample points with gully risk are assigned a value of 1, and the sample points without gully risk are assigned a value of 0.

4. The connectivity and geographic probe-based risk prediction method of gully erosion occurrence according to claim 1, wherein, the gully occurrence frequency represents the proportion of the gully points among the sample points.

5. The connectivity and geographic probe-based risk prediction method of gully erosion occurrence according to claim 1, wherein, the explanatory strength is the importance of the influence factor.

Images