Debris flow activity quantitative prediction method based on source intensity

Abstract

The invention relates to the technical field of geological disaster prevention and control engineering, in particular to a debris flow activity quantitative prediction method based on material source intensity. According to the specific technical scheme, the method comprises the following steps: (1) constructing a terrain disaster-pregnant index: calculating a weight based on a probability fusion method: constructing a decision matrix Z based on debris flow basic data, and calculating an upper limit value and a lower limit value of an evaluation factor by using a weight vector of the evaluation factor calculated by an equal weight method, an entropy weight method and a variable coefficient method; and (2) in combination with digital elevation model data and remote sensing images, carrying out data fitting on the material source participation degree and the terrain disaster-pregnancy index for many years to obtain a dynamic evaluation model of debris flow activity. According to the method, the limitation that a single factor quantifies topographic conditions is solved, the influence of different topographic disaster-pregnant environments on debris flow activity is more considered, and meanwhile, the object source activity intensity key index based on the time effect scale is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster prevention and control engineering, and particularly relates to a quantitative prediction method for debris flow activity based on source intensity. Background Art

[0002] As one of the most common mountain geological disasters, debris flow can release a large amount of previously accumulated substances and huge energy in an instant, often damaging transportation facilities such as highways, railways, and bridges, as well as communication infrastructure, and seriously threatening the social security and economic development of mountainous areas in the long term. According to statistics, 771 counties (cities) in China have suffered from debris flow disasters. The debris flow disasters cause an average of more than 3,700 deaths per year and direct economic losses of about 1 billion yuan per year. Therefore, to ensure the safety of production, life, and economic development in mountainous areas, it is necessary to continuously strengthen the research on debris flow. Among them, how to accurately and quickly obtain the actual activity of debris flow is one of the research focuses.

[0003] Looking back on the research history of debris flow activity at home and abroad, the research on debris flow activity can be traced back to the 1980s at the earliest. Before the 1980s, it mainly focused on field investigations of debris flow disasters and qualitative descriptions of activity, and related research was in its infancy. In the 1980s, related research began to gradually focus on using mathematical methods to reveal the research of multiple factors and debris flow activity, and the relevant analysis gradually transitioned to a semi-qualitative and semi-quantitative discrimination model. For example, the quantitative analysis of rainfall and topography, revealing the indirect impact of rainfall on slope stability through the influence of rainfall intensity and duration on the activity of shallow landslides and debris flows, that is, the I-D threshold method; and using methods such as basin area-elevation curve integration to quantitatively analyze the correlation between topographic development conditions and debris flow activity from the perspective of disaster-causing mechanisms. Since the 1990s, with the rapid development of GIS, RS, artificial intelligence, and computer technology, the evaluation method of debris flow activity has gradually developed towards a quantitative and accurate evaluation stage, making the refinement and correlation of the evaluation process more deeply reflected.

[0004] At present, the evaluation methods of debris flow activity in China can be roughly divided into four categories: 1) The evaluation method of debris flow activity using multi-source data for mathematical statistics; 2) The evaluation method of debris flow activity under the action of multi-factor correlation; 3) The regional and single-gully multi-scale evaluation method; 4) The evaluation method based on computer technologies such as GIS, RS, and machine learning. However, the current methods mostly consider the influence of terrain, rainfall, or single-source reserves, but often ignore the changes in formation conditions, that is, the evolution characteristics of the environment itself, which also leads to the fact that the existing debris flow activity evaluation methods cannot truly reflect its actual activity state.

[0005] The Chinese patent document with the publication number CN109447493A and the publication date of February 18, 2022 discloses a method for evaluating the risk of post-earthquake debris flow based on the activity intensity of the source material. Although the method for evaluating the risk of post-earthquake debris flow disclosed in this patent document considers the activity intensity of the source material, it does not consider the evolution of formation conditions such as topography and geomorphology, and all the selected samples are in earthquake areas, so the applicable range of the model is limited. In the actual development process of debris flow, driven by hydrodynamic conditions, data such as topography, source material, and elevation will also change with the change of the source material. Therefore, the actual activity of debris flow disasters cannot be obtained more accurately, and it is also impossible to provide strong support for the prevention and control of debris flow disasters. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the present invention provides a quantitative prediction method for debris flow activity based on the source material intensity, which solves the limitation of quantifying the topographic conditions by a single factor, considers more the influence of different topographic disaster-forming environments on debris flow activity, and at the same time proposes key indicators of the source material activity intensity based on the time effect scale. The activity intensity of the source material directly affects the accuracy of the evaluation results of debris flow activity on the time scale.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0008] The present invention discloses a quantitative prediction method for debris flow activity based on the source material intensity, including the following steps:

[0009] (1) Construction of the topographic disaster index

[0010] Calculate the weight based on the probability fusion method: construct a decision matrix Z based on the basic data of debris flow, and use the weight vectors of the evaluation factors calculated by the equal weight method, entropy weight method, and coefficient of variation method respectively to calculate the upper and lower limit values of the evaluation factors; then obtain the optimal weight through Monte Carlo simulation;

[0011] Assign and calculate the topographic disaster index based on the weight calculated by the probability fusion method, as shown in the following formula:

[0012]

[0013] In the formula, TDI is the topographic disaster index, w i is the weight calculated by the probability fusion method, is the normalized result of the evaluation factor;

[0014] (2) Combine the digital elevation model data and remote sensing images, and perform data fitting on the participation degree of the source material and the topographic disaster index over the years to obtain a dynamic evaluation model of debris flow activity.

[0015] Preferably, the geomorphic information entropy, gully density, longitudinal gradient of the basin, basin area, basin elevation difference, and main gully length are used as evaluation factors for constructing the terrain disaster pregnancy index.

[0016] Preferably, the calculation of the geomorphic information entropy is as follows:

[0017]

[0018] Where H is the geomorphic information entropy; S is the Strahler area-elevation integral value; f(x) is the Strahler area-elevation curve equation.

[0019] Preferably, the normalization processing steps of the evaluation factors are as follows: The decision matrix Z is linearly transformed by using min-max standardization, and the original value of Z is mapped into the interval [0,1], as follows:

[0020]

[0021] In the formula: max() represents taking the maximum value element; min() represents taking the minimum value element; the decision matrix Z = {z ij}, (i = 1, 2,..., m; j = 1, 2,..., n), i represents the number of debris flows; j represents the number of evaluation factors.

[0022] Preferably, the calculation steps of the source participation degree index are as follows:

[0023] (1) Preprocess and interpret the obtained remote sensing image data to obtain the source area and spatial distribution information of the debris flow activity area;

[0024] (2) After determining the source area and the number, use the GIS hydrological analysis method to determine the position of the main gully in the debris flow activity area, calculate the farthest distance from the main gully to the gully boundary, and divide the buffer zone;

[0025] (3) According to the source area in the buffer zone and the distance from the main gully, calculate the supply capacity of the source to the main gully, and obtain the source participation degree index.

[0026] Preferably, in step (3), the sources in the buffer zone are weighted, and the calculation formula is:

[0027] W(x) = e -kx

[0028] Where W(x) is the weight value of buffer x, k is the exponential decay constant, and x is the buffer distance ratio, with a range of 0.1 - 1;

[0029] The source area Ax of the buffer zone is multiplied by the weight W(x) of this buffer zone to obtain the weighted source area:

[0030]

[0031] Preferably, the source participation degree index is calculated by accumulating the weighted source areas of all buffers and dividing by the total source area A of the channel, and the formula is as follows: total to calculate, the formula is as follows:

[0032]

[0033] Preferably, the changes in the source areas of different buffers are compared to analyze the evolution trend of the source.

[0034] The present invention has the following beneficial effects:

[0035] 1. The present invention comprehensively calculates the Strahler area-elevation and geomorphic information entropy characteristic values of the basin by using topographic factors such as geomorphic information entropy, gully density, longitudinal gradient of the basin, basin area, basin elevation difference, and main gully length in the topographic data. Then, the multi-period actual debris flow source activity datasets in the study area are coupled with the geomorphic information entropy, and a debris flow activity discriminant formula is established based on the geomorphic information entropy required for different debris flow activities and landslide sources. Finally, the differences in the multi-period debris flow activity discriminant formulas are analyzed and integrated to establish a unified debris flow activity prediction and evaluation model based on driving factors such as source accumulation and rainfall. This model can effectively improve the accuracy of debris flow hazard assessment in earthquake areas, and can carry out multi-period hazard assessment of debris flows according to the activity intensity of the source, explore the evolution law of debris flow hazards, and provide a more powerful basis for the prediction and prevention of post-earthquake debris flow disasters.

[0036] 2. The method proposed by the present invention, which combines remote sensing interpretation and GIS hydrological analysis, quantifies debris flow activity by dynamically evaluating the comprehensive influence of the source participation degree index and the topographic disaster-bearing index. Compared with the prior art, the present invention has significant innovation and advantages in many aspects:

[0037] 1) Combination of remote sensing and GIS technologies

[0038] Most of the existing debris flow activity assessment methods focus on single topographic or climatic factors and ignore the influence of the source evolution process and spatial dynamics. The present invention obtains source data for different years through remote sensing interpretation technology and then combines GIS hydrological analysis to construct a dynamic debris flow activity evaluation model. This method breaks through the traditional static assessment method, enabling the assessment results to accurately reflect the spatio-temporal evolution law of debris flow activities, especially highlighting the dynamic influence of source evolution on debris flow activity, and having higher timeliness and spatial accuracy.

[0039] 2) Dynamic quantification of the source participation degree index

[0040] The present invention obtains the source area data through remote sensing image interpretation, combines it with the terrain disaster-bearing index for combined analysis, and proposes the "source participation degree index". By considering the source evolution in different years, the weight values of the source area in each buffer zone are calculated, so as to dynamically quantify the participation degree of the source in debris flow activity. Through the comparative analysis of the spatial distribution changes of the source in different years, the present invention can not only more accurately reveal the trend of source evolution, but also quantify the source supply capacity of each gully, providing more refined data support for the dynamic evaluation of debris flow activity.

[0041] 3) Construction of a quantitative model based on source evolution

[0042] In the prior art, the assessment of debris flow activity mostly conducts static analysis based on the outbreak frequency or a single terrain feature. The present invention proposes an innovative quantitative assessment model. By considering the dynamic changes of source evolution and the comprehensive effect of the terrain disaster-bearing index, combining historical data and remote sensing images, a dynamic evaluation model of debris flow activity is obtained by fitting with source data over the years. This model can not only evaluate the activity level of different gullies, but also verify and adjust the model through actual debris flow outbreak data, improving the reliability and accuracy of the assessment results.

[0043] 4) Construction of the terrain disaster-bearing index

[0044] In the present invention, the construction of the terrain disaster-bearing index is achieved by comprehensively considering multiple terrain factors to reflect the terrain disaster potential within the basin. Specifically, the terrain disaster-bearing index consists of five terrain factors: geomorphic information entropy, gully density, basin area, main gully length, and longitudinal gradient of the basin. These factors are closely related to the occurrence and development of debris flows. By accurately quantifying these terrain factors, the present invention can provide more reliable and systematic terrain basic data for debris flow activity assessment.

[0045] Compared with the traditional terrain evaluation method, the terrain disaster-bearing index in the present invention has the following innovations and advantages:

[0046] Comprehensiveness and multi-dimension: The present invention not only considers traditional terrain factors (such as basin area, gully density, etc.), but also introduces two new terrain factors, geomorphic information entropy and gully density of the basin, comprehensively reflecting the terrain characteristics of the basin. This multi-dimensional comprehensive analysis makes the terrain disaster-bearing index more accurately reflect the potential impact of the terrain on debris flow activity.

[0047] Weight Optimization and Dynamic Adjustment: The weight coefficients of the terrain disaster-bearing index are optimized using multiple methods. In particular, the Monte Carlo simulation method is used to optimize the optimal weights of each factor. This optimization method makes the weight calculation more scientific and reasonable, and at the same time has the ability of dynamic adjustment, which can adjust the weight parameters according to the actual conditions of different regions, improving the applicability and accuracy of the model.

[0048] Combined with the Degree of Sediment Source Participation: The terrain disaster-bearing index is combined with the degree of sediment source participation index, which can comprehensively evaluate the activity of debris flows. The two complement each other. The terrain disaster-bearing index provides basic information on the terrain conditions within the basin, while the degree of sediment source participation dynamically reflects the changes in debris flow sediment sources in the basin. The combination of the two helps to more accurately predict the occurrence and intensity of debris flows.

[0049] The terrain disaster-bearing index of the present invention has the following beneficial effects: Improving the accuracy of debris flow activity assessment: Through a comprehensive and accurate assessment of terrain factors, the activity potential of debris flows can be estimated more scientifically, providing data support for subsequent disaster prevention and control. Enhancing the reliability of debris flow prevention and control: By introducing a dynamically adjusted terrain disaster-bearing index, the present invention can adapt to different terrain changes over time, making the debris flow activity assessment more flexible and reliable. Wide range of applications: This method is not only applicable to the assessment of the activity of existing basin and gully debris flows, but can also be widely applied to the monitoring and early warning of debris flow disasters under different geological and climatic conditions.

[0050] 5) Activity discriminant considering terrain and sediment source factors

[0051] Different from the existing single-factor assessment methods, the present invention combines two factors, the terrain disaster-bearing index and the degree of sediment source participation, and uses a weighted method for comprehensive assessment, enabling the present invention to comprehensively evaluate the activity of debris flows from multiple dimensions, thereby improving the accuracy and robustness of the discriminant. In practical applications, the stability of the terrain disaster-bearing index and the dynamic changes in the degree of sediment source participation can work together to provide a more comprehensive and dynamic quantitative evaluation system for debris flow activity.

[0052] 6) Flexible model verification and evaluation mechanism

[0053] After the model is constructed, the present invention can not only test the fitting effect of the model through methods such as scatter plots and error limit analysis, but also provides a way to verify based on the outbreak frequency and actual activity, ensuring the reliability and scientific nature of the evaluation results. In addition, the present invention calculates the relationship between the sediment source evolution situation and terrain characteristics of different gullies, and combines the historical data of debris flow outbreaks to continuously correct and optimize, forming a sustainable update and self-adaptive adjustment evaluation system.

[0054] 7) Improve the accuracy of debris flow prevention and control and reduce disaster risks

[0055] The ultimate goal of the present invention is to construct a dynamic evaluation system that can accurately evaluate the activity and evolution law of debris flows. Through the evaluation of different channels, potential debris flow disaster hazard areas can be accurately identified. Through the dynamic comprehensive analysis of source evolution and terrain disaster formation, the present invention can achieve the precision and efficiency of debris flow activity early warning, provide a scientific basis for disaster prevention and control, and thus reduce the risks and losses brought by debris flow disasters.

[0056] 8) Cost advantages and application prospects

[0057] By combining remote sensing and GIS technologies, the high costs and low efficiency of traditional field surveys and manual measurements are reduced, and at the same time, it has strong time adaptability and space adaptability. In the monitoring and evaluation of large-scale debris flow disasters, the present invention can be widely applied to various geographical environments such as mountains and hills, which helps to improve the input-output ratio of debris flow prevention and control projects and reduce the costs of subsequent disaster prevention. In addition, the data sources of this method are extensive and convenient to update, and it has strong sustainability and cross-field application value. Description of the drawings

[0058] Figure 1 It is a flowchart for calculating weights by the probability fusion method;

[0059] Figure 2 It is a flowchart of the evaluation method of the present invention;

[0060] Figure 3 It is a case map of source interpretation of landslides and avalanches;

[0061] Figure 4 It is a Strahler area-elevation curve graph;

[0062] Figure 5 It is a source interpretation map of Banzi Gully in 2010, 2013, and 2023;

[0063] Figure 6 It is a fitting graph of terrain disaster formation index and source participation degree. Detailed implementation manners

[0064] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0065] If not specifically specified, the technical means used in the implementation examples are conventional means well known to those skilled in the art.

[0066] The present invention fully combines the source materials, hydrological and topographic conditions required for debris flow outbreaks. It not only evaluates the sensitivity of single debris flow outbreaks, but also comprehensively considers that the source materials are in a dynamic evolution process every year, with strong time evolution characteristics. At the same time, it proposes a quantitative analysis method for the influence of topographic disaster-bearing conditions on debris flow activity, solves the limitation of quantifying topographic conditions by a single factor, considers more the influence of different topographic disaster-bearing environments on debris flow activity, and proposes key indicators for the activity intensity of source materials based on the time effect scale. The activity intensity of source materials directly affects the accuracy of the evaluation results of debris flow activity on the time scale.

[0067] Therefore, the present invention first interprets the source material bodies in the debris flow basin from remote sensing images of each year, then quantitatively calculates the activity intensity of the source material bodies, and further determines the evaluation factors; the participation degree index of the source material, and statistically calculates the comprehensive discrimination values of different grading conditions of each evaluation factor under each activity intensity condition by using the mathematical method of probability. The method of the present invention combines remote sensing image interpretation, GIS analysis and the data of the participation degree of source materials over the years with the topographic disaster-bearing index to establish a dynamic and quantitative debris flow activity assessment model. This model can not only effectively evaluate the debris flow activity in historical data, but also provide a scientific basis for future debris flow risk assessment. In addition, the model has strong scalability and can be applied to the activity evaluation of other debris flow-prone areas.

[0068] A quantitative prediction method for debris flow activity based on the intensity of source materials specifically includes the following steps:

[0069] A. Calculation of the information entropy of the basin geomorphology. By integrating the information entropy of the basin area and elevation, it can be used to quantitatively describe the strength of the erosion action in the debris flow basin. The information entropy of the geomorphology is calculated according to the following formula:

[0070]

[0071] Where H is the information entropy of the geomorphology; S is the Strahler area-elevation integral value; f(x) is the Strahler area-elevation curve equation.

[0072] B. Construction of the topographic disaster-bearing index

[0073] In view of the limitations of the information entropy method of geomorphology, the present invention proposes to introduce the "topographic disaster-bearing index" (TDI, Topographic Disaster Index) as a new factor to comprehensively evaluate debris flow activity. The construction of the topographic disaster-bearing index aims to make up for the deficiencies in the existing methods, and provides a more accurate and comprehensive evaluation of debris flow activity by integrating multiple topographic factors and considering their relative importance.

[0074] The terrain disaster-prone index is evaluated by comprehensively considering multiple terrain factors (such as gully density, longitudinal gradient of the basin, basin area, elevation difference of the basin, main gully length, etc.). These factors play different roles in the occurrence of debris flows, and their interrelationships are relatively complex. A single factor (such as geomorphic information entropy) often cannot fully reflect this complex interaction. By weighted integration of the effects of multiple factors, the terrain disaster-prone index can more comprehensively capture the potential impact of terrain on debris flow activity, thus providing more accurate predictions. Compared with the static analysis of the geomorphic information entropy method, the terrain disaster-prone index has strong regional adaptability and flexibility by considering the terrain diversity and spatial differences of the basin. In different basins, the weights of terrain factors can be adjusted according to the actual situation, enabling this index to provide a relatively accurate assessment of debris flow activity in different geographical environments.

[0075] Therefore, the terrain disaster-prone index can not only overcome the limitations of single-factor analysis methods but also provide evaluation results that are more in line with the actual situation. In the assessment of debris flow activity, traditional methods often rely on fixed or expert-set weights to measure the contribution of each terrain factor to the occurrence of debris flows. However, this method of weight assignment has significant subjectivity and limitations. Especially when faced with basins with complex terrain conditions and intertwined factor interactions, it often leads to assessment results that do not match the actual debris flow activity. To solve this problem, the present invention proposes a weight fusion method based on probability theory and closely combines it with the terrain disaster-prone index (TDI) to improve the accuracy and adaptability of debris flow activity prediction.

[0076] The core idea of the weight fusion method is to regard the weight as a bounded uniform random variable and randomly generate multiple weight combinations within the specified boundary conditions through the Monte Carlo simulation strategy. Specifically, the boundary conditions are calculated by three commonly used weight assignment methods: the equal weight method, the entropy weight method, and the coefficient of variation method. These methods respectively reflect the influence degree of different terrain factors on debris flow activity and provide a reasonable initial range for the weights. Then, through repeated sampling, the Monte Carlo simulation generates a large number of weight samples within this boundary condition and selects the optimal weight combination based on the principle of minimum deviation.

[0077] The process of calculating weights by the probability fusion method is as Figure 1 shown. The equal weight method, the entropy weight method, and the coefficient of variation method can refer to the literature (Li Li, Zhang Shixin, Qiang Yue, et al. Probability Optimization Method for Weight Assignment in Debris Flow Hazard Assessment [J]. Journal of Chongqing Jiaotong University (Natural Science Edition), 2022, 41(07): 120 - 125.). The specific process is as follows:

[0078] 1. Entropy weight method

[0079] The core of the entropy weight method is to determine the objective weight according to the degree of disorder of the evaluation factors, which overcomes the subjectivity in the evaluation method. The entropy value vector Hj of the evaluation factors is defined as follows:

[0080]

[0081] The weight of the evaluation factor entropy weight method is as follows:

[0082]

[0083] In the entropy weight method, the smaller the information entropy of the evaluation factor, the greater the role it plays in the evaluation, and the greater the entropy weight; vice versa.

[0084] 2. Coefficient of variation method

[0085] The coefficient of variation method, also known as the "standard deviation rate method", is an objective assignment method. Its basic idea is to assign values to each evaluation index according to the degree of variation of the evaluation factors. The greater the variation gap, the greater the difference between the actual value and the ideal target value of the index, and the greater weight should be assigned to the index; vice versa.

[0086] Both the mean value and the standard deviation will affect the size of the coefficient of variation. Therefore, it is necessary to first calculate the mean value and the standard deviation S j , as follows:

[0087]

[0088] The ratio of the standard deviation to the mean value is the coefficient of variation. The coefficient of variation v of the evaluation factor j is calculated as follows:

[0089]

[0090] Normalize the coefficient of variation to obtain the coefficient of variation method weight w j , as follows:

[0091]

[0092] 3. Equal weight method

[0093] The equal weight method is the most widely used and convenient method for determining weights. In the analytic hierarchy process, if the importance of each evaluation factor is regarded as equal, the equal weight method can be regarded as a special analytic hierarchy process.

[0094] The equal weight method is as follows:

[0095]

[0096] 4. Weight fusion method

[0097] The decision matrix composed of the basic debris flow data in the study area was solved from different data processing perspectives by the equal weight method, entropy weight method, and coefficient of variation method, and three groups of weight vectors were obtained. According to the definition of weights, the evaluation factor weight vector \(w = [w_1, w_2, w_3, w_4, w_5, w_6, w_7, w_8, w_9, w\) 10 T can be used as weights. Considering the weights as bounded uniformly distributed random variables, the upper and lower limits of the evaluation factors were obtained through three weight determination methods. The steps are as follows:

[0098] 1) Construct a decision variable matrix \(Z\) based on the basic debris flow data, and calculate the weight factor vector \(w\) of each evaluation factor through the equal weight method, entropy method, and coefficient of variation method respectively;

[0099] 2) Calculate the upper and lower limit values of the evaluation factors according to \(w\) respectively: \(w_1\in[0.0947, 0.1667]\); \(w_2\in[0.1434, 0.1732]\); \(w_3\in[0.069, 0.1667]\); \(w_4\in[0.1667, 0.3382]\); \(w_5\in[0.0926, 0.1667]\); \(w_6\in[0.1667, 0.2227]\). That is, according to the weight values of each evaluation factor calculated by the first three weight calculation methods, each factor has a weight interval.

[0100] 3) Calculate the normalized decision matrix;

[0101] 4) Generate a sample matrix \(W = \{w\) ij \}, (\(i = 1, 2, \cdots, l\); \(j = 1, 2, \cdots, n\)) within the boundary, where: \(i\) represents the number of generated samples; \(j\) represents the number of factors.

[0102] 5) Obtain the optimal weights by solving the minimum deviation to optimize the weight assignment;

[0103] The calculation process of steps 3) - 5) is the Monte Carlo simulation strategy.

[0104] According to the analysis of the topography and geomorphology of the debris flow gullies in the study area, the present invention selects the geomorphic information entropy, gully density, longitudinal gradient of the basin, basin area, basin elevation difference, and main gully length as the evaluation factors for constructing the terrain disaster - bearing index. These basic data are defined as the decision matrix \(Z\), where \(Z=\{z\) ij \}, (\(i = 1, 2, \cdots, m\); \(j = 1, 2, \cdots, n\)). \(i\) represents the number of debris flows; \(j\) represents the number of evaluation factors. In the present invention, \(n = 6\). Since the different dimensions of each evaluation factor will affect the results, it is necessary to normalize the decision matrix. Considering the data distribution law, the min - max standardization is used to perform a linear transformation on the decision matrix, mapping the original value of \(Z\) to the interval \([0, 1]\), as shown in the following formula:​

[0105]

[0106] In the formula: max() represents taking the maximum value element; min() represents taking the minimum value element. What is enclosed in the brackets of this formula means taking the maximum and minimum values in each evaluation factor of all debris flow channels in the evaluation area. The larger the values of the six evaluation factors, the higher the degree of danger. Therefore, only the above formula needs to be used for normalization; the larger the normalized value, the higher the degree of danger.

[0107] Finally, based on the weight calculation result of the probability fusion method, the terrain disaster - bearing index is assigned and calculated as follows:

[0108]

[0109] In the formula, TDI is the terrain disaster - bearing index, w i is the weight calculated by the probability fusion method, is the normalization result of the evaluation factor.

[0110] C. Calculation of the degree of source participation index

[0111] The landslide - debris flow source is one of the important conditions for the occurrence of debris flows in strong earthquake areas. A large amount of loose source materials generated by co - seismic landslides accumulate on the slopes of each basin. Under the action of heavy rainfall, these landslide - debris flow sources gradually converge into the channels, providing favorable conditions for the outbreak of debris flows in strong earthquake areas. With the rapid development of remote sensing technology, the remote sensing monitoring of geological disasters in strong earthquake areas has never stopped. Multi - source and multi - temporal remote sensing data provide a large amount of macro - effective information for the study of post - earthquake geological disasters.

[0112] The present invention relates to the field of debris flow activity assessment, especially a method for calculating the degree of source participation based on remote sensing image interpretation and GIS analysis. By interpreting and spatially analyzing remote sensing image data of different years and combining with the setting of the terrain disaster - bearing index, the present invention provides an innovative method for quantitatively evaluating debris flow activity, especially highlighting the dynamics of the influence of source evolution on debris flow activity.

[0113] The specific steps are as follows:

[0114] 1. Acquisition of remote sensing images and interpretation of source area

[0115] The present invention first obtains image data of different years through remote sensing technology, and uses the spatial resolution of remote sensing images to extract the source area and spatial distribution information of the debris flow activity area. The sources of remote sensing image data include satellite images, aerial images or images taken by unmanned aerial vehicles. After obtaining the image data, image pre - processing, classification and source extraction are carried out to determine the source area within the channel area for each year. An example of the interpretation of landslide - debris flow sources is as Figure 3 shown.

[0116] 1.1 Remote sensing image data acquisition

[0117] Use satellite images (such as Landsat or Sentinel series), UAV images or aerial images to ensure the temporal consistency and spatial resolution of the images. The frequency of image acquisition should meet at least once a year to obtain data for different periods (such as different years like 2010, 2013, 2019, 2022, etc.).

[0118] 1.2 Image preprocessing and interpretation

[0119] Perform radiometric correction, geometric correction and atmospheric correction on the acquired remote sensing images to eliminate the influence caused by factors such as illumination and meteorology. Subsequently, adopt a method combining automatic classification and manual recognition and interpretation, and use land cover classification algorithms (such as maximum likelihood method, support vector machine, etc.) to extract the source area. By interpreting the images of different years, obtain the source area and spatial distribution information of each gully.

[0120] 1.3 Source area extraction and calculation

[0121] Extract the total source area and the number of sources of each gully every year to generate a source spatial distribution database. And calculate the source area and the number of sources of each gully in each year to provide basic data for subsequent analysis.

[0122] 2. Determination of main gullies and division of multi-ring buffers

[0123] After determining the source area, through GIS hydrological analysis methods, determine the position of the main gully of each gully and calculate the maximum distance from the main gully to the gully boundary. Then, divide several buffers according to the ratio of the maximum distance, and each buffer represents the source supply ability of different sources to the main gully.

[0124] 2.1 Determination of main gullies

[0125] Conduct hydrological analysis through GIS software, use watershed segmentation and watershed network analysis to determine the main gully of each gully, and calculate the maximum distance from the main gully to the gully boundary.

[0126] 2.2 Buffer division

[0127] Divide the maximum distance from the main gully to the gully boundary into ten equal parts to form multi-ring buffers. The width of each buffer is 1 / 10 of the maximum distance, and at the same time determine that the ratio of the buffer distance from the gully boundary to the ratio of the maximum distance from the main gully to the gully boundary is: 0.1 - 1.0, and conduct unified division in this way to eliminate the influence of different buffer step sizes on the calculation of the source participation degree.

[0128] 3. Calculation of the Degree of Provenance Participation and Quantification of Provenance Supply Capacity

[0129] After the buffer zones are demarcated, according to the provenance area within each buffer zone and its distance from the main gully, the supply capacity of the provenance to the main gully is calculated by a weighted method, and finally the provenance participation degree index is obtained.

[0130] 3.1 Calculation of the Provenance Area within the Buffer Zone

[0131] For each buffer zone, calculate the provenance area within it and obtain the provenance data within the buffer zone. The provenance area within each buffer zone represents the potential contribution of the provenance in this area to the main gully. The provenance area is directly calculated in ArcGIS according to the actual situation of remote sensing interpretation.

[0132] 3.2 Weight Allocation and Quantification of Provenance Supply Capacity

[0133] The exponential weight method is used to weight the provenance in different buffer zones. The specific calculation method is as follows:

[0134] W(x) = e -kx

[0135] where W(x) is the weight value of buffer zone x, k is the exponential decay constant, which is defaulted to 1 and can be adjusted according to the actual gully terrain situation, and x is the buffer zone distance ratio, with a range of 0.1 - 1.

[0136] Multiply the provenance area Ax of each buffer zone by the weight W(x) of this buffer zone to obtain the weighted provenance area:

[0137]

[0138] 3.3 Calculation of the Provenance Participation Degree Index

[0139] For each gully, the provenance participation degree index (abbreviated as MPI) is calculated by accumulating the weighted provenance areas of all buffer zones and dividing by the total provenance area A of this gully total as follows:

[0140]

[0141] D. Analysis of the Evolution Trend of Debris Flow Provenance

[0142] By comparing the changes in the provenance area on different buffer zones in different years, analyze the evolution trend of the provenance. By comparing the provenance area and spatial distribution in different years, identify the evolution trend of the provenance towards the main gully. If the evolution trend of the provenance indicates convergence towards the main gully, it indicates that the gully has relatively high disaster - forming conditions for debris flow in terms of provenance attributes.

[0143] E. Construct a dynamic evaluation model of activity through data fitting

[0144] The debris flow activity dynamic evaluation method provided by the present invention, which is based on remote sensing image interpretation, GIS spatial analysis, and the combination of the degree of source participation and terrain disaster pregnancy index, constructs a debris flow activity dynamic evaluation model through the fitting analysis of the degree of source participation and terrain disaster pregnancy index over the years, and quantitatively analyzes and evaluates the activity in combination with the grading standard of debris flow outbreak frequency.

[0145] 1. Determination and classification of debris flow outbreak frequency

[0146] An important index of debris flow activity is the outbreak frequency of debris flow, that is, the occurrence frequency of debris flow events. According to the standard of "DZ-T0220-2018 Exploration Specification for Debris Flow Disaster Prevention Engineering", the debris flow outbreak frequency is used as an important grading index for debris flow activity. The specific outbreak frequency classification is as follows:

[0147] High debris flow activity: The channels with an outbreak frequency of once in 5 years, once a year, and multiple times a year.

[0148] Medium debris flow activity: The channels with an outbreak frequency of once in 20 years - once in 5 years.

[0149] Through the above standards, the actual debris flow activity level of each channel is determined. By statistically analyzing historical debris flow events, the debris flow outbreak frequency of each channel is obtained, providing a basis for subsequent fitting analysis.

[0150] 2. Fitting analysis of the degree of source participation and terrain disaster pregnancy index over the years

[0151] Taking the average value of the degree of source participation over the years of each channel as the dependent variable and the terrain disaster pregnancy index as the independent variable, data fitting is carried out. The specific operation steps are as follows:

[0152] 2.1 Data preparation

[0153] Collect the data of the degree of source participation of each channel in different years, calculate it as the average degree of source participation over the years, as the dependent variable, and the terrain disaster pregnancy index (assumed not to change with years in the present invention) as the independent variable. Each channel will have a corresponding value of the degree of source participation and a value of the terrain disaster pregnancy index.

[0154] 2.2 Selection of fitting method

[0155] Select a fitting method suitable for the data characteristics, such as polynomial fitting. The goal of fitting is to establish a discriminant formula that can reflect debris flow activity by analyzing the relationship between the terrain disaster pregnancy index and the degree of source participation.

[0156] 2.3 Establishment of fitting model

[0157] Visualize the dependent variable (degree of multi-year provenance participation) and the independent variable (terrain disaster-prone index) through a scatter plot. Use statistical software (such as Origin, Excel, SPSS, etc.) for fitting, select polynomial fitting, and evaluate the accuracy of the fitting results. During the fitting process, focus on the trend of the fitting curve and the 2 value of R to ensure the rationality of the model.

[0158] 2.4 Error Analysis and Model Optimization

[0159] Error analysis in the fitting results is one of the key steps. Use a residual plot to check the fitting quality of the fitting curve. If the fitting error is large, the model can be further optimized according to the actual data situation, such as increasing the number of fittings, adjusting the selection of independent variables, etc., to optimize the accuracy and predictive ability of the model.

[0160] 3. Selection and Discriminant Construction of High-Activity Channels

[0161] According to the fitting results and the definition of the actual debris flow outbreak frequency, select the channels with high activity and perform fitting. The specific steps are as follows:

[0162] 3.1 Selection of High-Activity Channels

[0163] Select the channels that meet the high debris flow activity criteria based on the historical debris flow outbreak frequency and the fitting results. These channels are important data points in the fitting model and have a high debris flow occurrence frequency. Through statistical analysis, select the channels with a high outbreak frequency as the fitting samples.

[0164] 3.2 Fitting and Discriminant Generation

[0165] Through polynomial fitting or other appropriate methods for fitting the provenance participation degree and terrain disaster-prone index of high-activity channels, obtain a discriminant that reflects the relationship between debris flow activity, provenance participation degree, and terrain disaster-prone index. This discriminant will be able to predict the debris flow activity of other unevaluated channels based on the new provenance participation degree and terrain disaster-prone index.

[0166] 3.3 Model Validation and Evaluation

[0167] To verify the effectiveness and reliability of the model, use methods such as leave-one-out cross-validation to verify the model, ensuring that the discriminant has good adaptability and predictive ability for new data. Calculate the accuracy, precision, and goodness of fit of the model (such as 2 value of R, mean square error, etc.), and compare with the actual debris flow activity data.

[0168] 4. Application of the Dynamic Debris Flow Activity Evaluation Model

[0169] The constructed discriminant for debris flow activity can be used to evaluate the debris flow activity of other channels. By inputting the degree of source participation and the terrain disaster - pregnant index of a new channel into the model, the debris flow activity level of that channel can be predicted, providing a scientific basis for debris flow prevention and control.

[0170] The present invention will be further elaborated below in conjunction with specific embodiments.

[0171] Embodiment 1

[0172] Banzi Gully is located southwest of Wenchuan County, about 10 km away from Wenchuan County seat. This area is located in the Longmenshan Fault Zone between the Qinghai - Tibet Plateau and the Sichuan Basin, with strong tectonic activity. There are 2 faults passing through the downstream, and the gully mouth is the Wenchuan - Maoxian Fault (WMF). The Wenchuan - Maoxian Fault was one of the main active faults during the 2008 Wenchuan earthquake, with a strike of N25° - E45°. The Wenchuan earthquake induced a large number of coseismic landslides in the basin, providing abundant solid materials for debris flows. The exposed lithology in Banzi Gully is relatively simple. The lithology exposed in the middle and upper reaches of the basin is mainly Mesoproterozoic plagiogranite. The lithology exposed in the downstream of the basin is mainly Mesoproterozoic carbonate rocks, Sinian metamorphic rock sandstones and dolomites. Quaternary collapse - slope deposits and debris - flow deposits are distributed throughout the basin. The drainage area of Banzi Gully is 54.7 km 2 , the main gully length is 14.1 km, the relative elevation difference of the basin is 3994 m, and the longitudinal gradient of the main gully is 206‰. The terrain of Banzi Gully is mainly rugged mountains, with the main slope range being 30° - 50°, and the average is 36°. The basin geomorphology belongs to the tectonic erosion deeply incised middle - high mountain geomorphology. The basin plane is in the shape of a broad leaf, the gully plane form is "dendritic", and there is water in the gully all year round. There are 4 larger tributary gullies developed in the Banzi Gully basin, among which 2 tributary gullies are located in the north of the main gully and 2 tributary gullies are located in the south of the main gully. The cross - section of the main gully gradually transitions from a "V" shape to a "U" shape from top to bottom. The relevant data are shown in Table 1 below.

[0173] Table 1 Relevant data of Banzi Gully

[0174] Total number of grids 351030 Total basin area Ai <![CDATA[54.681132km 2 <!-- 11 -->]]> Total basin elevation difference Hi 3998m Integral value of Strahler area-elevation curve 0.46 Geomorphic information entropy 0.236528789 Total gully length 19.318 km Gully density <![CDATA[0.353284km / km 2 >

[0175] A. Calculation of the information entropy of the basin geomorphology. By integrating the information entropy of the basin area and elevation, it can be used to quantitatively describe the intensity of erosion in the debris - flow basin. The information entropy of the geomorphology is calculated according to the following formula, and the results are shown in Table 2 below:

[0176]

[0177] Among them, H is the information entropy of the geomorphology; S is the Strahler area - elevation integral value; f(x) is the Strahler area - elevation curve equation, as Figure 4 shown.

[0178] Table 2 Calculation Results of the Information Entropy of the Banzigou Geomorphology

[0179]

[0180] B. Calculation of the Terrain Disaster-Prone Index

[0181] 1. Calculation of the Weights of the Evaluation Factors by the Entropy Weight Method

[0182] 1) Normalize the influence factors, and calculate the entropy value as follows. The results are shown in Table 3 below.

[0183]

[0184] Table 3 Entropy Values of Each Evaluation Factor

[0185]

[0186] 2) The weights of the evaluation factors by the entropy weight method are as follows. The results are shown in Table 4 below.

[0187]

[0188] Table 4 Weights of Each Evaluation Factor

[0189]

[0190] 2. Calculation of the Weights of the Evaluation Factors by the Coefficient of Variation Method

[0191] 1) Calculate the mean and standard of the evaluation factors as follows. The results are shown in Table 5 below.

[0192]

[0193] Table 5 Mean and Standard Values of Each Evaluation Factor

[0194]

[0195] 2) Calculate the coefficient of variation as follows. The results are shown in Table 6 below.

[0196]

[0197] Table 6 Coefficient of Variation of Each Evaluation Factor

[0198]

[0199] 3) Calculate the weights as follows. The results are shown in Table 7 below.

[0200]

[0201] Table 7 Weights of Each Evaluation Factor

[0202]

[0203] 3. Equal - weight method, as shown in the following formula, and the results are shown in Table 8 below.

[0204]

[0205] Table 8 Weights of each evaluation factor

[0206]

[0207] 4. By solving the decision matrix composed of the debris flow basic data in the study area from different data - processing perspectives through the equal - weight method, entropy - weight method, and coefficient of variation method, 3 groups of weight vectors are obtained. According to the definition of weights, the evaluation factor weight vector \(w = [w_1, w_2, w_3, w_4, w_5, w_6, w_7, w_8, w_9, w\) 10 T can be used as weights. Regarding the weights as bounded uniformly distributed random variables, the upper and lower limits of the evaluation factors are obtained through 3 weight - determination methods. The steps are as follows:

[0208] 1) Based on the debris flow basic data, construct the decision variable matrix \(Z\), and calculate the weight factor vectors \(w\) of each evaluation factor through the equal - weight method, entropy method, and coefficient of variation method respectively.

[0209] 2) Calculate the upper and lower limit values of the evaluation factors according to \(w\) respectively as: \(w_1\in[0.0947, 0.1667]\); \(w_2\in[0.1434, 0.1732]\); \(w_3\in[0.069, 0.1667]\); \(w_4\in[0.1667, 0.3382]\); \(w_5\in[0.0926, 0.1667]\); \(w_6\in[0.1667, 0.2227]\).

[0210] 3) Calculate the normalized decision matrix;

[0211] 4) Generate a weight random variable sample matrix \(W=\{w\) ij \}, (\(i = 1, 2,\cdots, l\); \(j = 1, 2,\cdots, n\)), where: \(i\) represents the number of generated samples; \(j\) represents the number of factors. By solving the minimum - deviation optimization weight assignment, the optimal weights are obtained, and the results are shown in Table 9 below.

[0212] Table 9 Weights of each evaluation factor

[0213]

[0214] 5. Calculation of terrain disaster - inducing index

[0215] Based on the weight calculation results of the probability fusion method, assign values to calculate the terrain disaster - inducing index, as shown in the following formula. The calculation result of Banzi Gully is: 0.569647697. ​

[0216]

[0217] C. Calculation of Provenance Participation Degree

[0218] First, obtain image data of different years through remote sensing technology. For Banzi Gully, select three-phase remote sensing images in 2010, 2013, and 2019, and use the spatial resolution of the remote sensing images to extract the provenance area and spatial distribution information of the debris flow activity area. The sources of remote sensing image data include satellite images, aerial images, or images taken by drones. After obtaining the image data, perform image preprocessing, classification, and provenance extraction to determine the provenance area within the gully area for each year.

[0219] 1. The remote sensing data in this embodiment includes Google Earth images with a resolution of 1.5 m in April 2010, September 2013, and August 2019 after the earthquake. In the target study area, the regional coverage rates of the images in 2010, 2013, and 2019 are about 90%.

[0220] 2. Based on the existing post-earthquake landslide interpretation by predecessors (Fan et al., 2018), as Figure 5 shown, combining digital elevation model (DEM) data with remote sensing images, accurately locate the morphology, structure, and elements of the landslide and avalanche bodies in three-dimensional space. By comparing multi-temporal remote sensing images in detail, digitize the post-earthquake landslide and avalanche provenance, and improve it through on-site verification. The shapes, tones, and image structures of the landslides and gully deposits in the debris flow basin of the study area show certain differences from the surrounding background in the remote sensing images. Therefore, the morphology, tone, and texture of the loose provenance bodies in the basin can be directly interpreted and delineated from the remote sensing images, and their identification signs are mainly as follows:

[0221] 1) Morphological characteristics: Earthquake-induced landslides and avalanche bodies usually appear as long strips, spoons, or semi-circles in remote sensing images, and a few are irregular. Landslide features such as landslide walls, landslide benches, and landslide tongues are clearly visible in remote sensing images. The surface of an active landslide body is generally uneven, and the slope is steep and long. In terms of the direction criterion, the longitudinal axis of the landslide should be consistent with the direction of gravity as the interpretation sign.

[0222] 2) Tone characteristics: The tone of the landslide body is quite different from that of the surrounding stable terrain. The newly formed landslide body is composed of loose accumulated materials, and its surface has a strong spectral reflection ability, showing an obvious light tone in the remote sensing image; the periphery of the landslide body in the deformation stage usually has a lighter color ring than the plane form of the main body, or there are light-colored lines at the rear edge of the slope body.

[0223] 3) Texture features: The surface of an active landslide is broken, and the texture on the unstable image is relatively rough. Large patchy blocks can be seen on the images of some rocky landslides.

[0224] 3. Determine the position of the main gully in the debris flow activity area through Gis hydrological analysis.

[0225] 1) Calculate the flow direction:

[0226] Use the tool "Flow Direction" to calculate the flow direction of each pixel unit. Flow direction analysis determines the path and direction of water flow through the height difference of the DEM. Common flow direction calculation methods include the D8 method.

[0227] Tool path: ArcToolbox -> Spatial Analyst Tools -> Hydrology -> FlowDirection

[0228] 2) Calculate the flow accumulation:

[0229] Flow accumulation analysis calculates the amount of water flowing into each pixel unit. Areas with larger flow values usually represent areas where water converges, that is, possible main gullies or water systems. The main water flow paths can be identified through the cumulative value of the flow.

[0230] Tool path: ArcToolbox -> Spatial Analyst Tools -> Hydrology -> FlowAccumulation

[0231] 3) Extract the main gully

[0232] Set the threshold to extract the main gully:

[0233] In the flow accumulation map (Flow Accumulation), use a threshold to identify areas with larger flow values, which generally represent the main gullies. The threshold can be set by observing the water flow paths under different flow values and selecting an appropriate cumulative flow value.

[0234] Tool path: Spatial Analyst -> Reclassify (reclassification)

[0235] Use the "Reclassify" tool to classify the flow data according to the set threshold and mark the areas with larger flow values as the main gully areas.

[0236] 4) Extract the main gully line (Stream Order):

[0237] By flowing data, the river system hierarchy (watershed classification) of the watershed can be further analyzed, and the main channels can be extracted according to the size of the river hierarchy. The commonly used watershed classification method is the Strahler watershed hierarchy method, which can help identify the main channels in the watershed.

[0238] Tool path: ArcToolbox -> Spatial Analyst Tools -> Hydrology -> StreamOrder

[0239] After extraction, the core part of the main channel can be further determined according to the level of the watershed hierarchy.

[0240] 5) Selection and trimming of the main channel

[0241] Select the main watershed:

[0242] According to the flow accumulation map and the watershed classification results, select the river channel with higher flow as the main channel. Usually, the main channel refers to the river channel with the largest flow, and further analysis can be carried out along this channel.

[0243] Trimming of the main channel:

[0244] According to the need, use tools such as "Clip" or "Erase" to trim the main channel line so that it coincides with other river systems or topographic features.

[0245] Extract the coordinates of the main channel:

[0246] Through the "Extract by Mask" or similar tools, the extracted main channel is separated from the entire watershed data for subsequent analysis and calculation.

[0247] 6) Analysis of the maximum distance between the main channel and the boundary

[0248] Calculate the maximum distance from the channel to the boundary:

[0249] Use the "Euclidean Distance" tool to calculate the maximum distance from the main channel to the watershed boundary. This tool will generate a distance map of each pixel to the main channel, and finally obtain the distance information between the main channel and the watershed boundary.

[0250] Tool path: ArcToolbox -> Spatial Analyst Tools -> Distance -> EuclideanDistance

[0251] Create a buffer:

[0252] Use the "Buffer" tool to generate buffers around the main channel, divide the maximum distance from the channel to the boundary into ten equal parts, and obtain multi-ring buffers. These buffers will be used for subsequent calculation of the source participation degree.

[0253] Tool path: Arc Toolbox->Analysis Tools->Proximity->Buffer, as shown in Table 10 below.

[0254] Table 10

[0255]

[0256] E. By means of data fitting, fit the Banzi Gully on the basis of the original activity discriminant formula to judge its activity. See Figure 6 as shown. The result shows that R 2 = 0.77, and the fitting degree is good.

[0257] The discriminant formula obtained from the high activity of the 47 selected debris flow gullies in the study area is:

[0258] y = 0.59877 - 0.46409x + 0.9174x 2

[0259] Substitute the obtained terrain disaster pregnancy index of 0.569647697 of the Banzi Gully into the discriminant formula to calculate the actual source participation degree of the Banzi Gully as 0.63129. According to the source interpretation situation in 2023 and the calculated source participation degree of the Banzi Gully in that year, the calculation result is 0.684586, which is close to the actual situation. It is determined that the debris flow activity of the Banzi Gully in 2023 is high. Through actual investigation, a large-scale debris flow broke out in the Banzi Gully in June 2023. Therefore, this model has a relatively accurate ability to judge debris flow activity.

[0260] The embodiments described above are only descriptions of the preferred modes of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A quantitative prediction method for debris flow activity based on source intensity, characterized in that: It includes the following steps: (1) Construction of terrain disaster-bearing index Calculating weights based on the probability fusion method: Constructing a decision matrix Z based on the basic debris flow data, and using the weight vectors of evaluation factors calculated by the equal weight method, entropy weight method, and coefficient of variation method respectively to calculate the upper and lower limit values of the evaluation factors; then obtaining the optimal weight through Monte Carlo simulation; Assigning and calculating the terrain disaster-bearing index based on the weight calculated by the probability fusion method, as shown in the following formula: Where TDI is the terrain disaster index, w i is the weight calculated by the probability fusion method, Normalized results for evaluation factors; (2) Combining digital elevation model data and remote sensing images, fitting the source participation degree and terrain disaster-bearing index over the years to obtain a dynamic evaluation model of debris flow activity.

2. A quantitative prediction method for debris flow activity based on source intensity according to claim 1, characterized in that: Taking geomorphic information entropy, gully density, longitudinal gradient of the basin, basin area, basin elevation difference, and main gully length as evaluation factors for constructing the terrain disaster-bearing index.

3. The quantitative prediction method for debris flow activity based on source intensity according to claim 2, wherein: The calculation of geomorphic information entropy is as follows: Where H is the geomorphic information entropy; S is the Strahler area-elevation integral value; f(x) is the Strahler area-elevation curve equation.

4. A quantitative prediction method for debris flow activity based on source intensity according to claim 1, characterized in that: The steps for normalizing the evaluation factors are: Performing a linear transformation on the decision matrix Z using min-max normalization, mapping the original value of Z to the interval [0,1], as shown in the following formula: In the formula: max() represents taking the maximum value element; min() represents taking the minimum value element; the decision matrix Z = {z ij}, (i = 1, 2,..., m; j = 1, 2,..., n), where i represents the number of debris flows; j represents the number of evaluation factors.

5. A quantitative prediction method for debris flow activity based on source intensity according to claim 1, characterized in that: The calculation steps of the source participation degree index are as follows: (1) Preprocessing and interpreting the obtained remote sensing image data to obtain the source area and spatial distribution information of the debris flow activity area; (2) After determining the source area and number, using the GIS hydrological analysis method to determine the position of the main gully in the debris flow activity area, calculating the farthest distance from the main gully to the gully boundary, and dividing the buffer zone; (3) According to the source area in the buffer zone and the distance from the main gully, calculating the supply capacity of the source to the main gully and obtaining the source participation degree index.

6. A quantitative prediction method for debris flow activity based on source intensity according to claim 5, characterized in that: In step (3), weighting the sources in the buffer zone, and the calculation formula is: W(x) = e -kx Where W(x) is the weight value of buffer x, k is the exponential decay constant, and x is the buffer distance ratio, with a range of 0.1 - 1; Multiplying the source area Ax in the buffer zone by the weight W(x) of this buffer zone to obtain the weighted source area:

7. A quantitative prediction method for debris flow activity based on source intensity according to claim 3, characterized in that: The source participation degree index is calculated by accumulating the weighted source areas of all buffers and dividing by the total source area A of the channel total as follows:

8. A quantitative prediction method for debris flow activity based on source intensity according to claim 5, characterized in that: Comparing the changes in the source areas of different buffer zones to analyze the evolution trend of the sources.

Images