Mountain torrent formation disaster-causing mechanism analysis method based on disaster-pregnant environment

Abstract

The invention discloses a mountain torrent formation disaster-causing mechanism analysis method based on a disaster-pregnant environment. The method comprises the following steps: step 1, constructing a historical mountain torrent disaster attribute set and an environment factor set in a research area; 2, determining mountain torrent disaster pregnancy environment type partitions and features thereof; step 3, determining the interpretation degree and the uncertainty interval of the environmental category and the environmental factor of each subarea on the mountain torrent disaster attribute change; step 4, determining dominant environment factors of each disaster-pregnant environment type partition; and 5, analyzing the influence of the dominant environmental factor change on the mountain torrent formation process and the disaster-causing process. The method fully considers the spatial-temporal heterogeneity of meteorological, geographic and social economic factors of the affected basin, separately quantifies the contribution and uncertainty of different environmental factors such as meteorological, geographic and social economic factors to formation of mountain torrents to cause disasters, can improve the comprehensive prevention and control level of mountain torrent disasters, and is high in applicability and easy to popularize. And a scientific basis is provided for mountain torrent disaster prevention and management.

Description

Technical Field

[0001] The present invention belongs to the technical field of mountain flood disaster management, and particularly relates to a method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-forming environment. Background Art

[0002] Mountain flood disasters are one of the most dangerous and severely loss-causing natural disasters in mountainous areas in the world. The formation and disaster-causing process of mountain floods are affected by the combined action of meteorological conditions, basin geographical conditions, and social and economic conditions, and have great uncertainties. Therefore, scientifically revealing and comparing the disaster-forming environments of mountain flood disasters and their meteorological, geographical, and social and economic environmental components is crucial for objectively understanding the formation mechanism of mountain flood disasters and formulating targeted flood management measures.

[0003] Currently, the methods for analyzing the formation and disaster-causing mechanism of mountain floods mainly include two categories: (1) Post-disaster investigation: After a major mountain flood disaster occurs, quickly conduct on-site post-disaster investigations, and clarify the causes, processes, and weak links in the disaster prevention process of mountain flood disasters through visits and research. (2) Numerical simulation: By collecting data such as topography, land use, hydrometeorology, and disaster losses before and after the disaster, construct a basin hydrological model or a basin hydrodynamic model to reproduce the physical scene during the disaster and clarify the causes of mountain flood disasters. However, the above methods mainly conduct analyses on a certain major mountain flood disaster from the event or basin scale. On the one hand, they can only depict the formation and disaster-causing mechanism of mountain floods in a specific environment, and the analysis results are difficult to be applied to basins or disaster events with different meteorological, geographical, and social and economic environments; on the other hand, they mainly analyze major disaster events that cause heavy casualties or property losses, ignoring a large number of disaster events with low mortality rates, and the analysis results are not representative. In addition, different combinations of meteorological, geographical, and social and economic environments have different effects on the formation and disaster-causing process of mountain floods, resulting in significant differences in the disaster losses caused.

[0004] Therefore, due to the complex and variable formation and disaster-causing mechanisms of mountain floods in different types of disaster-forming environments, there is an urgent need for a method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-forming environment to explore the quantitative causal relationship between various environmental factors and mountain flood disaster losses, and provide a basis for the prevention and management of mountain flood disasters. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-forming environment to solve the above technical problems.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] The present invention discloses a method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-forming environment, and the method includes the following steps:

[0008] Step 1: Construct the historical mountain flood disaster attribute set and environmental factor set for the study area: Collect and organize the environmental factor data of each environmental category in the study area and the attribute information of historical mountain flood disaster events in the study area. The environmental categories include meteorological category, geographical category, and social and economic category. The environmental factors include meteorological category factors, geographical category factors, and social and economic category factors. The attribute information of historical mountain flood disaster events includes the disaster occurrence location, occurrence time, number of deaths or missing persons, number of collapsed houses, and direct economic losses. Then, taking the basin where the historical mountain flood disaster event occurred as the unit and the occurrence time of the historical mountain flood disaster event as the benchmark, construct the historical mountain flood disaster attribute set and the corresponding environmental factor set for the study area.

[0009] Step 2: Determine the zoning of the mountain flood disaster disaster-forming environment types and their characteristics: According to the historical mountain flood disaster attribute set and the corresponding environmental factor set constructed in Step 1, determine the preliminary zoning of the mountain flood disaster disaster-forming environment types. Then, evaluate the significant differences of all environmental factors and mountain flood disaster attributes in the preliminary zoning of the mountain flood disaster disaster-forming environment types to determine the final zoning of the mountain flood disaster disaster-forming environment types. For the determined final zoning of the mountain flood disaster disaster-forming environment types, identify the meteorological category factors, geographical category factors, social and economic category factors, and mountain flood disaster attribute characteristics of each environmental type and their spatial distribution patterns.

[0010] Step 3: Determine the explanatory degree and uncertainty interval of the environmental categories and their environmental factors in each zoning on the change of mountain flood disaster attributes: Taking the final zoning of the mountain flood disaster disaster-forming environment types determined in Step 2 as the unit, randomly select the environmental categories, their environmental factors, and mountain flood disaster attributes in each zoning using the N-fold uncertainty analysis method to quantify the explanatory degree and uncertainty interval of different environmental categories on the change of mountain flood disaster attributes, and quantify the explanatory degree, uncertainty interval, and significance of each environmental factor on the change of mountain flood disaster attributes.

[0011] Step 4: Determine the dominant environmental factors in each zoning of the disaster-forming environment types: Based on the explanatory degree and uncertainty interval of the environmental categories and their environmental factors in each zoning on the change of mountain flood disaster attributes determined in Step 3, use the cause analysis method and comparative analysis method to determine the dominant environmental factors in each zoning of the disaster-forming environment types and their differential effects, and characterize the mountain flood formation and disaster-causing mechanism in each zoning of the disaster-forming environment types.

[0012] Step 5. Analyze the impacts of changes in dominant environmental factors on the formation process and disaster-causing process of mountain floods: Construct a meteorological-hydrological-hydrodynamic coupled numerical model for the study area, and automatically optimize the parameters of the coupled model using the measured rainstorm-flood data and historical flood mark data of the basin. For areas lacking measured data, use parameter regionalization technology to determine the parameters of the coupled model; then, according to the dominant environmental factors of each disaster-prone environment type partition determined in Step 4, set up a scenario library of dominant environmental factors, and analyze the impacts of changes in dominant environmental factors on the formation process and disaster-causing process of mountain floods in the study area; furthermore, propose corresponding disaster prevention and mitigation measures for each disaster-prone environment type partition.

[0013] Furthermore, the meteorological category factors described in Step 1 include the humidity, flood season precipitation, and short-duration extreme rainfall characteristics in the year when the mountain flood disaster event occurs. The short-duration extreme rainfall characteristics include magnitude, variability, and concentration; the geographical category factors include the terrain relief degree, slope, proportion of forest area, proportion of grassland area, proportion of farmland area, soil saturation moisture content, river length, river slope, and river curvature in the year when the mountain flood disaster event occurs; the social and economic category factors include the GDP and population density in the year when the mountain flood disaster event occurs, and the water conservancy investment before the mountain flood disaster event.

[0014] Furthermore, the specific process of determining the preliminary disaster-prone environment type partition of mountain floods in Step 2 is as follows:

[0015] Use the clustering algorithm to classify all the environmental factors corresponding to historical mountain flood disasters in the environmental factor set of the study area, and determine the disaster-prone environment type partition CE’ of mountain floods; use the N-fold uncertainty analysis method to randomly divide the environmental factor set of the study area into N parts, select N-1 parts of the environmental factor data for clustering analysis, and repeat this process N times to determine the N-fold disaster-prone environment type partition CE of mountain floods N ; Use the hit rate to evaluate the coincidence degree of the disaster-prone environment type partition CE’ of mountain floods and the N-fold disaster-prone environment type partition CE N of mountain floods, and the results are expressed in the form of the mean and standard deviation of the hit rate; if the hit rate reaches more than 95%, then take the disaster-prone environment type partition CE’ of mountain floods as the preliminary disaster-prone environment type partition of mountain floods; the hit rate calculation formula is:

[0016]

[0017] In the formula, PAC q is the q-fold hit rate, 1 ≤ q ≤ N; PC iq is the disaster-prone environment type of the i-th historical mountain flood disaster event in the q-fold, 1 ≤ i ≤ m q , PC iq ∈CE N ; AC iis the disaster-forming environment type of the i-th historical mountain flood disaster event among all historical mountain flood disaster events, AC i ∈CE’; m q is the number of historical mountain flood disaster events in the q-fold;

[0018] Wherein:

[0019]

[0020] The specific process of determining the final zoning of the disaster-forming environment type of mountain flood disasters is as follows: Use the Kruskal-Wallis test to evaluate the significant differences of all environmental factors in each zoning of the disaster-forming environment type. If the calculated significance level p < 0.01, it indicates that there are significant differences in environmental factors among the sub-zones; when the number of zonings of the disaster-forming environment type exceeds two, further use the Nemenyi test to evaluate the significant differences between any two groups of zonings of the disaster-forming environment type. If the calculated significance level p < 0.01, it indicates that there are significant differences in environmental factors between any two sub-zones; use the similarity test to evaluate the significant differences of the mountain flood disaster attributes in each zoning of the disaster-forming environment type. If the calculated significance level p < 0.01, it indicates that there are significant differences in the mountain flood disaster attributes among the sub-zones. The zoning passing the above tests can be used as the final zoning of the disaster-forming environment type; if the above tests are not passed, re-divide the zoning of the disaster-forming environment type of mountain flood disasters until all tests are passed.

[0021] Furthermore, the specific process of quantifying the explanatory degree of different environmental categories for the change of mountain flood disaster attributes and its uncertainty interval in step 3 is as follows: For a certain zoning of the disaster-forming environment type of mountain flood disasters, randomly divide the meteorological category factors, geographical category factors, socio-economic category factors and the corresponding mountain flood disaster attributes in this zoning into N parts. Select N - 1 parts of environmental categories as independent variables and the corresponding N - 1 parts of mountain flood disaster attributes as response variables. Use detrended correspondence analysis to determine the length of the first sorting axis. Further select the appropriate restricted sorting analysis method or canonical correspondence analysis method according to the axis length to determine the explanatory degree of different environmental categories for the change of mountain flood disaster attributes respectively. Repeat this process N times to determine the uncertainty interval of the explanatory degree of each environmental category, which is expressed in the form of the average value and standard deviation of the N results;

[0022] The specific process of quantifying the explanatory degree, uncertainty interval, and significance of the changes in each environmental factor on the attributes of mountain flood disasters is as follows: For a specific mountain flood disaster disaster-forming environment type sub-region, the environmental factors and corresponding mountain flood disaster attributes within this sub-region are randomly divided into N parts. Select N - 1 parts of the environmental factors as independent variables and the corresponding N - 1 parts of the mountain flood disaster attributes as response variables, and conduct 999 Monte Carlo permutation tests to determine the explanatory degree and significance of each environmental factor on the response variable. If the significance level p < 0.01, it indicates that this environmental factor significantly affects the response variable. Repeat this process N times to determine the uncertainty interval and significance of the explanatory degree of all environmental factors in this mountain flood disaster disaster-forming environment type sub-region, which are expressed in the form of the average value and standard deviation of the N results.

[0023] Furthermore, the specific process of determining the dominant environmental factors and their differential effects in each disaster-forming environment type sub-region in step 4 is as follows: For a specific mountain flood disaster disaster-forming environment type sub-region, screen out the significant environmental factors and arrange them in descending order according to the average value of their explanatory degree on the changes in mountain flood disaster attributes. The environmental factor with the largest explanatory degree is the first dominant factor, the environmental factors with an explanatory degree greater than or equal to 0.1 and less than that of the first dominant factor are the main dominant factors, and the remaining environmental factors are the secondary dominant factors. Compare the rankings of the dominant environmental factors in each mountain flood disaster disaster-forming environment type sub-region, and analyze the effects of each environmental factor on the formation and development process of mountain floods from the perspective of hydrological genesis to determine the differential effects of the dominant environmental factors in each disaster-forming environment type sub-region.

[0024] Furthermore, the specific process of constructing the meteorological - hydrological - hydrodynamic coupled numerical model of the study area in step 5 is as follows: Taking the small watershed as the basic calculation unit and the location where the historical mountain flood disaster event occurred as the watershed outlet, couple the meteorological numerical model, watershed hydrological model, and one - two - dimensional hydrodynamic model to construct the meteorological - hydrological - hydrodynamic coupled numerical model of the study area;

[0025] The specific process of analyzing the influence of the changes in the dominant environmental factors on the formation process and disaster - causing process of mountain floods in the study area is as follows: For the values of each dominant environmental factor in each disaster-forming environment type sub-region in the study area, conduct perturbations at a step size of 10% within the range of ±100% of each factor value to set up a scenario library of dominant environmental factors. Input each scenario into the calibrated meteorological - hydrological - hydrodynamic coupled numerical model to obtain the mountain flood process line, inundation range, inundation depth, and inundation duration information of the study area under different environmental factor scenarios. Compare the environmental factors and disaster attributes during the occurrence of historical mountain flood disasters to quantify the influence of the changes in the dominant environmental factors on the formation process and disaster - causing process of mountain floods in the study area.

[0026] The beneficial effects of the present invention are as follows: The method of the present invention fully considers the spatio-temporal heterogeneity of meteorological, geographical and socio-economic factors in the disaster-affected basin, separates and quantifies the contributions and uncertainties of different environmental factors such as meteorology, geography and socio-economy to the formation and disaster-causing of mountain floods, and analyzes the formation mechanisms of mountain flood disasters under different disaster-forming environment types. The method of the present invention can achieve differential prevention and control of mountain flood disasters, improve the comprehensive prevention and control level of mountain flood disasters, has strong applicability, and provides a scientific basis for the prevention and management of mountain flood disasters.

[0027] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic flow chart of the method of the present invention;

[0029] Figure 2 It is the explanatory degree and its uncertainty interval of different environmental categories on the change of mountain flood disaster attributes in Embodiment 1;

[0030] Figure 3 It is the explanatory degree and its uncertainty interval of each environmental factor on the change of mountain flood disaster attributes in Embodiment 1. SPECIFIC EMBODIMENTS

[0031] The present invention discloses a method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-forming environment, as Figure 1 shown, the method includes the following steps:

[0032] Step 1: Construct a historical mountain flood disaster attribute set and an environmental factor set for the study area:

[0033] First, collect and organize the environmental factor data of each environmental category in the study area, as well as the attribute information of historical mountain flood disaster events in the study area. The environmental categories include meteorological categories, geographical categories and socio-economic categories, etc., and the corresponding environmental factors include meteorological category factors, geographical category factors and socio-economic category factors, etc. Among them, the meteorological category factors include the humidity, flood season precipitation, short-duration extreme rainfall characteristics (including magnitude, variability, concentration) in the year when the mountain flood disaster event occurs; the geographical category factors include the terrain undulation degree, slope, forest area ratio, grassland area ratio, farmland area ratio, soil saturation moisture content, river length, river gradient, river curvature rate in the year when the mountain flood disaster event occurs; the socio-economic category factors include the GDP and population density in the year when the mountain flood disaster event occurs, and the water conservancy investment before the mountain flood disaster event. The attribute information of historical mountain flood disaster events includes the disaster occurrence location, occurrence time, number of deaths or missing persons, number of collapsed houses, direct economic losses, etc.

[0034] Then, taking the basins where historical mountain flood disaster events occurred as units and the occurrence times of historical mountain flood disaster events as the benchmark, construct the historical mountain flood disaster attribute set and the corresponding environmental factor set in the study area.

[0035] Step 2: Determine the zoning and characteristics of the disaster-forming environment types of mountain floods:

[0036] Based on the historical mountain flood disaster attribute set and environmental factor set in the study area constructed in Step 1, preliminarily determine the zoning of the disaster-forming environment types of mountain floods; specifically: Use clustering algorithms (such as dynamic K-means clustering, hierarchical clustering, or SOM clustering) to classify all the environmental factors corresponding to historical mountain flood disasters in the environmental factor set of the study area, and determine the zoning CE' of the disaster-forming environment types of mountain floods; Use the N-fold uncertainty analysis method to randomly divide the environmental factor set of the study area into N parts, select N-1 parts of the environmental factor data for clustering analysis, and repeat this process N times to determine the N-fold zoning CE of the disaster-forming environment types of mountain floods N ; Use the hit rate to evaluate the coincidence degree between the zoning CE' of the disaster-forming environment types of mountain floods and the N-fold zoning CE of the disaster-forming environment types N The hit rate calculation formula is shown in Equation (1), and the results are expressed in the form of the mean and standard deviation of the hit rate; if the hit rate reaches more than 95%, then take the zoning CE' of the disaster-forming environment types of mountain floods as the preliminary zoning of the disaster-forming environment types of mountain floods;

[0037]

[0038] In the formula, PAC q is the hit rate of the q-fold, 1 ≤ q ≤ N; PC iq is the disaster-forming environment type of the i-th historical mountain flood disaster event in the q-fold, 1 ≤ i ≤ m q , PC iq ∈CE N ; AC i is the disaster-forming environment type of the i-th historical mountain flood disaster event among all historical mountain flood disaster events, AC i ∈CE'; m q is the number of historical mountain flood disaster events in the q-fold.

[0039] Among them:

[0040]

[0041] Then, evaluate the significant differences of all environmental factors and mountain flood disaster attributes in the preliminary zoning of mountain flood disaster - forming environment types to determine the final zoning of mountain flood disaster - forming environment types. Specifically: Use the Kruskal - Wallis test to evaluate the significant differences of all environmental factors in each zoning of mountain flood disaster - forming environment types. If the calculated significance level p < 0.01, it indicates that there are significant differences in environmental factors among different zones; when the number of zoned mountain flood disaster - forming environment types exceeds two, further use the Nemenyi test to evaluate the significant differences between any two groups of zoned mountain flood disaster - forming environment types. If the calculated significance level p < 0.01, it indicates that there are significant differences in environmental factors between any two groups of zones; use the Analysis of Similarities to evaluate the significant differences of mountain flood disaster attributes in each zoning of mountain flood disaster - forming environment types. If the calculated significance level p < 0.01, it indicates that there are significant differences in mountain flood disaster attributes among different zones. The zones passing the above - mentioned tests can be used as the final zoning of mountain flood disaster - forming environment types; if the above - mentioned tests are not passed, re - divide the zoning of mountain flood disaster - forming environment types until all tests are passed.

[0042] For the determined final zoning of mountain flood disaster - forming environment types, identify the meteorological category factors, geographical category factors, socio - economic category factors, and the characteristics and spatial distribution patterns of mountain flood disaster attributes in each environmental type zone.

[0043] Step 3: Determine the explanatory degree and its uncertainty interval of each zoning environmental category and its environmental factors on the change of mountain flood disaster attributes:

[0044] Taking the final zoning of mountain flood disaster - forming environment types determined in Step 2 as the unit, randomly select the environmental categories, their environmental factors, and mountain flood disaster attributes in each zone using the N - fold uncertainty analysis method to quantify the explanatory degree and its uncertainty interval of different environmental categories on the change of mountain flood disaster attributes, quantify the explanatory degree and its uncertainty interval and significance of each environmental factor on the change of mountain flood disaster attributes.

[0045] The specific process of quantifying the explanatory degree of different environmental categories for the changes in the attributes of mountain flood disasters and their uncertainty intervals is as follows: For a specific zoning of the disaster-forming environment types of mountain flood disasters, randomly divide the meteorological category factors, geographical category factors, socio-economic category factors, and the corresponding mountain flood disaster attributes within this zoning into N parts. Select N - 1 parts of the environmental categories as independent variables, and the corresponding N - 1 parts of the mountain flood disaster attributes as response variables. Use Detrended correspondence analysis to determine the length of the first sorting axis. Further select the appropriate restricted sorting analysis method (Redundancy analysis or Canonical correspondence analysis) according to the axis length to determine the explanatory degree of different environmental categories for the changes in the attributes of mountain flood disasters. Repeat this process N times to determine the uncertainty intervals of the explanatory degrees of each environmental category, and represent them in the form of the average value and standard deviation of the N results.

[0046] The specific process of quantifying the explanatory degree, uncertainty intervals, and significance of each environmental factor for the changes in the attributes of mountain flood disasters is as follows: For a specific zoning of the disaster-forming environment types of mountain flood disasters, randomly divide the environmental factors and the corresponding mountain flood disaster attributes within this zoning into N parts. Select N - 1 parts of the environmental factors as independent variables, and the corresponding N - 1 parts of the mountain flood disaster attributes as response variables. Conduct 999 Monte Carlo permutation tests to determine the explanatory degree and significance of each environmental factor for the response variable. If the significance level p < 0.01, it indicates that this environmental factor significantly affects the response variable. Repeat this process N times to determine the uncertainty intervals and significance of the explanatory degrees of all environmental factors in this zoning of the disaster-forming environment types of mountain flood disasters, and represent them in the form of the average value and standard deviation of the N results.

[0047] Step 4: Determine the dominant environmental factors for each zoning of the disaster-forming environment types.

[0048] Based on the degree of explanation and its uncertainty interval of the changes in the attributes of mountain flood disasters by each sub-region environmental category and its environmental factors determined in Step 3, the cause analysis method and the comparative analysis method are adopted to determine the dominant environmental factors and their differential effects in each type of disaster-prone environment sub-region. Specifically: for a certain type of mountain flood disaster-prone environment sub-region, significant environmental factors are screened and arranged in descending order according to the average value of the degree of explanation of the changes in the attributes of mountain flood disasters. The environmental factor with the largest degree of explanation is the first dominant factor, the environmental factors with a degree of explanation greater than or equal to 0.1 and less than that of the first dominant factor are the main dominant factors, and the remaining environmental factors are the secondary dominant factors. By comparing the sorting of the dominant environmental factors in each type of mountain flood disaster-prone environment sub-region, the effects of each environmental factor on the formation and development process of mountain floods are analyzed from the perspective of hydrological causes, and the differential effects of the dominant environmental factors in each type of disaster-prone environment sub-region are determined; then, the formation and disaster-causing mechanisms of mountain floods in each type of disaster-prone environment sub-region are characterized.

[0049] Step 5. Analyze the impacts of changes in dominant environmental factors on the formation process and disaster-causing process of mountain floods:

[0050] Construct a meteorological-hydrological-hydrodynamic coupled numerical model for the study area, specifically: taking the small watershed as the basic calculation unit and the location where historical mountain flood disaster events occurred as the watershed outlet, coupling the meteorological numerical model, the watershed hydrological model, and the one-dimensional and two-dimensional hydrodynamic models to construct a meteorological-hydrological-hydrodynamic coupled numerical model for the study area. The parameters of the coupled model are automatically optimized using the measured storm-flood data and historical flood mark data of the watershed. For areas lacking measured data, parameter regionalization techniques are used to determine the parameters of the coupled model.

[0051] Then, according to the dominant environmental factors in each type of disaster-prone environment sub-region determined in Step 4, a scenario library of dominant environmental factors is set up, and the impacts of changes in dominant environmental factors on the formation process and disaster-causing process of mountain floods in the study area are analyzed. Specifically: for the values of each dominant environmental factor in each type of disaster-prone environment sub-region in the study area, perturbations are carried out at a step size of 10% within the range of ±100% of the values of each factor to set up a scenario library of dominant environmental factors. Each scenario is input into the calibrated meteorological-hydrological-hydrodynamic coupled numerical model to obtain information such as the mountain flood process line, inundation range, inundation depth, and inundation duration in the study area under different environmental factor scenarios. By comparing the environmental factors and disaster attributes during historical mountain flood disasters, the impacts of changes in dominant environmental factors on the formation process and disaster-causing process of mountain floods in the study area are quantified.

[0052] Furthermore, corresponding disaster prevention and mitigation measures are proposed for each type of disaster-prone environment sub-region.

[0053] Example 1

[0054] This example is an application example of the above method.

[0055] In this embodiment, the middle and upper reaches of the Liaohe River Basin are taken as the study area, and a method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-bearing environment is disclosed, including the following steps:

[0056] Step 1: Construct the historical mountain flood disaster attribute set and environmental factor set of the study area:

[0057] Collect and organize the data of 17 environmental factors such as meteorological category factors, geographical category factors, and social and economic category factors in the study area, as shown in Table 1, and the attribute information of 45 historical mountain flood disaster events in the study area from 1949 to the present, including the disaster occurrence location, occurrence time, number of deaths or missing persons, number of collapsed houses, and direct economic losses; taking the basin where the historical mountain flood disaster event occurs as the unit and the occurrence time of the historical mountain flood disaster event as the benchmark, construct the historical mountain flood disaster attribute set and the corresponding environmental factor set of the study area.

[0058] Table 1 List of collected environmental factor data

[0059]

[0060]

[0061] Step 2: Determine the disaster-bearing environment type zoning of mountain floods and its characteristics:

[0062] According to the historical mountain flood disaster attribute set and environmental factor set of the study area constructed in Step 1, use the dynamic K-means clustering method to initially determine 3 disaster-bearing environment type zonings of mountain floods, and use the ten-fold uncertainty analysis method to randomly sample for clustering analysis. The hit rate statistics are shown in Table 2, all of which are greater than 95%. Therefore, the 3 disaster-bearing environment type zonings of mountain floods can be used as the preliminary disaster-bearing environment type zonings of mountain floods.

[0063] Table 2 Hit rate statistics table

[0064]

[0065]

[0066] Based on the preliminary disaster-bearing environment type zonings of mountain floods, use the Kruskal-Wallis test and Nemenyi test to evaluate that there are significant differences (p<0.01) in the 17 environmental factors among the 3 preliminary zonings and between any two groups of zonings, and use the similarity test to evaluate that there are significant differences (p<0.01) in the mountain flood disaster attributes among the preliminary zonings. Therefore, determine the 3 disaster-bearing environment type zonings of mountain floods as the final disaster-bearing environment type zonings of mountain floods.

[0067] For the final zoning of the disaster-forming environment types of flash floods, identify the meteorological category factors, geographical category factors, socio-economic category factors, and the attribute characteristics and spatial distribution patterns of flash floods for each environmental type. Among them, Type 1 is mainly distributed in the northwestern region of the study area, Type 2 is mainly distributed in the central part of the study area, and Type 3 is mainly distributed in the southeastern part of the study area; the disaster intensity of Type 2 is the highest, followed by Type 3 and Type 1; the humidity, proportion of forest land area, and water conservancy investment show an increasing trend from Type 1 to Type 3; the 1-hour rainfall variability, river length, and proportion of farmland area show a decreasing trend from Type 1 to Type 3; the 1-hour rainfall concentration and proportion of grassland area of Type 1 are the largest, and Type 2 is the smallest; the flood season precipitation, 1-hour rare heavy rain, river curvature, GDP, and population density of Type 2 are the largest, and Type 1 is the smallest; the terrain undulation, slope, soil saturation moisture content, and river gradient of Type 3 are the largest, and Type 2 is the smallest.

[0068] Step 3: Determine the explanatory degree and its uncertainty interval of the environmental category and its environmental factors in each partition on the change of flash flood disaster attributes:

[0069] Taking the final zoning of the disaster-forming environment types of flash floods determined in Step 2 as the unit, randomly select the environmental category, its environmental factors, and flash flood disaster attributes in each partition by the ten-fold uncertainty analysis method. The result of the detrended correspondence analysis shows that the length of the first sorting axis is less than 1.65 times the standard deviation. Therefore, the redundancy analysis method is selected to quantify the explanatory degree and its uncertainty interval of different environmental categories on the change of flash flood disaster attributes, as Figure 2 shown. The explanatory degree, its uncertainty interval, and significance of each environmental factor on the change of flash flood disaster attributes are as Figure 3 shown.

[0070] Step 4: Determine the dominant environmental factors in each partition of the disaster-forming environment types:

[0071] Based on the interpretability and uncertainty intervals of each partition environmental category and its environmental factors determined in Step 3 for each disaster-forming environment type partition, the genetic analysis method and the comparative analysis method are used to determine the dominant environmental factors and their differential effects of each disaster-forming environment type partition. There are 15 dominant environmental factors in Type 1. The first dominant environmental factor is the precipitation during the flood season, and the main dominant environmental factors are the 1-hour rainfall variability, soil saturation moisture content, proportion of forest land area, proportion of grassland area, wetness, slope length, proportion of farmland area, water conservancy investment, river length, GDP, population density, 1-hour rare heavy rain, and the secondary dominant environmental factors are river gradient and terrain undulation; there are 16 dominant environmental factors in Type 2. The first dominant environmental factor is the precipitation during the flood season, and the main dominant environmental factors are the 1-hour rainfall variability, 1-hour rare heavy rain, 1-hour rainfall concentration, wetness, proportion of grassland area, proportion of forest land area, proportion of farmland area, river gradient, population density, and the secondary dominant environmental factors are GDP, water conservancy investment, slope length, river length, soil saturation moisture content, and river curvature; there are 16 dominant environmental factors in Type 3. The first dominant environmental factor is the 1-hour rare heavy rain, and the main dominant environmental factors are the precipitation during the flood season, 1-hour rainfall variability, 1-hour rainfall concentration, wetness, proportion of grassland area, soil saturation moisture content, population density, GDP, river curvature, proportion of forest land area, and the secondary dominant environmental factors are river gradient, slope length, water conservancy investment, terrain undulation, and proportion of farmland area.

[0072] From the perspective of hydrological genesis, comprehensively consider the effects and their differences of each dominant environmental factor, and characterize the formation and disaster-causing mechanisms of mountain floods under each disaster-forming environment type partition. The formation and disaster-causing mechanism of mountain floods in Type 1 is as follows: In a dry climate, the water storage capacity of the basin dominated by farmland is poor. The short-duration rainfall with high altitude change and concentrated process leads to a rapid rise of floodwater in the mountain flood gullies of the basin and a small flood volume. Due to the low level of economic development and poor disaster reduction ability, the disaster intensity caused by the flood inundating the villages along the river is low; the formation and disaster-causing mechanism of mountain floods in Type 2 is as follows: In a humid climate, the water storage capacity of the basin dominated by grassland is low. The frequent high-intensity rainfall leads to a rapid rise of floodwater in the mountain flood gullies, inundating the villages along the river and the highland villages due to poor drainage. Due to the high level of economic development and medium disaster reduction ability, the disaster intensity is high; the formation and disaster-causing mechanism of mountain floods in Type 3 is as follows: In a humid climate, the water storage capacity of the steep basin dominated by forest land is strong. The frequent high-intensity concentrated rainfall leads to a rapid rise of floodwater in the mountain flood gullies, inundating the villages along the river and a compound mountain flood disaster under the action of water-sediment coupling. Due to the medium level of economic development and strong disaster reduction ability, the disaster intensity is medium.

[0073] Step 5. Analyze the influence of the change of the dominant environmental factor on the formation process and disaster-causing process of mountain floods:

[0074] With 0.5 km 2Small watersheds are extracted based on the catchment area threshold. Taking the extracted small watersheds as the basic calculation units and the occurrence locations of historical mountain flood disaster events as the watershed outlets, a coupled numerical model of meteorology - hydrology - hydrodynamic for the study area is constructed. The parameters of the coupled model are automatically optimized using 20 rain - flood data sets and 10 historical flood mark data sets in the study area. The relative errors of the average peak flood discharge, the average peak appearance time, and the average inundation depth for all data sets are 10%, 1 hour, and 8 cm respectively.

[0075] For the dominant environmental factors in each disaster - pregnant environment type partition determined in Step 4, perturbations are carried out at a step of 10% within the range of ±100% of the values of each factor, and a scenario library of dominant environmental factors is set. Each scenario is input into the calibrated meteorology - hydrology - hydrodynamic coupled numerical model to obtain information such as the mountain flood hydrograph, inundation range, inundation depth, and inundation duration in the study area under different environmental factor scenarios. By comparing the environmental factors and disaster attributes during historical mountain flood disasters, the impacts of changes in dominant environmental factors on the formation process and disaster - causing process of mountain floods in the study area are quantified. Among them, the peak flood discharge and inundation range of mountain floods of Type 1 are significantly affected by factors such as the precipitation during the flood season of the watershed, the 1 - hour rainfall variability, the soil saturation moisture content, and the proportion of forest land area; the peak flood discharge and inundation range of mountain floods of Type 2 are significantly affected by factors such as the precipitation during the flood season of the watershed, the 1 - hour rainfall variability, the 1 - hour rare heavy rain, the 1 - hour rainfall concentration, the proportion of forest land area, the proportion of farmland area, the river gradient, the population density, and GDP; the peak flood discharge and inundation range of mountain floods of Type 3 are significantly affected by factors such as the 1 - hour rare heavy rain, the precipitation during the flood season, the 1 - hour rainfall variability, and the 1 - hour rainfall concentration in the watershed.

[0076] Furthermore, disaster prevention and mitigation measures for each disaster - pregnant environment type partition are proposed. Among them, for Type 1, afforestation should be emphasized to improve the soil saturation moisture content of the watershed, and the monitoring ability of the watershed should be enhanced by building a rainfall and mountain flood monitoring network in the upstream watershed of the village; for Type 2, returning farmland to forest should be emphasized to improve the soil water storage capacity of the watershed and reduce the runoff generation capacity. A mountain flood disaster forecasting and early warning model should be developed for mountain flood disaster forecasting and early warning, which should be applicable to frequently occurring heavy rainfall events, extend the mountain flood early warning lead time, improve the emergency preparedness, reduce the mortality rate due to disasters, and the river should be straightened to improve the river discharge capacity; for Type 3, remote sensing and radar rainfall measurement should be carried out in high - mountain areas where it is difficult to carry out conventional rainfall monitoring to eliminate monitoring blind spots, the development of comprehensive prevention and control technologies for multi - hazard and compound mountain flood disasters should be carried out, and floodgates should be built to store floodwaters.

[0077] Finally, it should be noted that the above is only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred layout scheme, those of ordinary skill in the art should understand that the technical solution of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-bearing environment, characterized in that, The method includes the following steps: Step 1: Construct the historical mountain flood disaster attribute set and environmental factor set of the study area: Collect and organize the environmental factor data of each environmental category in the study area and the attribute information of historical mountain flood disaster events in the study area. The environmental categories include meteorological category, geographical category, and social and economic category. The environmental factors include meteorological category factors, geographical category factors, and social and economic category factors. The attribute information of historical mountain flood disaster events includes the disaster occurrence location, occurrence time, number of deaths or missing persons, number of collapsed houses, and direct economic losses. Then, taking the basin where the historical mountain flood disaster event occurs as a unit and the historical mountain flood disaster event occurrence time as a benchmark, construct the historical mountain flood disaster attribute set and the corresponding environmental factor set of the study area; Step 2: Determine the mountain flood disaster disaster-forming environment type zoning and its characteristics: According to the historical mountain flood disaster attribute set and the corresponding environmental factor set of the study area constructed in Step 1, determine the preliminary mountain flood disaster disaster-forming environment type zoning. Then, evaluate the significant differences of all environmental factors and mountain flood disaster attributes in the preliminary mountain flood disaster disaster-forming environment type zoning to determine the final mountain flood disaster disaster-forming environment type zoning. For the determined final mountain flood disaster disaster-forming environment type zoning, identify the meteorological category factors, geographical category factors, social and economic category factors, and mountain flood disaster attribute characteristics and their spatial distribution patterns of each environmental type; Step 3: Determine the explanatory degree and uncertainty interval of the environmental category and its environmental factors in each zoning on the change of mountain flood disaster attributes: Taking the final mountain flood disaster disaster-forming environment type zoning determined in Step 2 as a unit, randomly select the environmental category, its environmental factors, and mountain flood disaster attributes in each zoning by using the N-fold uncertainty analysis method, quantify the explanatory degree and uncertainty interval of different environmental categories on the change of mountain flood disaster attributes, and quantify the explanatory degree, uncertainty interval, and significance of each environmental factor on the change of mountain flood disaster attributes; Step 4: Determine the dominant environmental factors in each disaster-forming environment type zoning: Based on the explanatory degree and uncertainty interval of the environmental category and its environmental factors in each zoning on the change of mountain flood disaster attributes determined in Step 3, use the cause analysis method and comparative analysis method to determine the dominant environmental factors and their action differences in each disaster-forming environment type zoning, and characterize the mountain flood formation and disaster-causing mechanism under each disaster-forming environment type zoning; Step 5: Analyze the impact of the change of dominant environmental factors on the mountain flood formation process and disaster-causing process: Construct a meteorological-hydrological-hydrodynamic coupled numerical model of the study area, and automatically optimize the coupled model parameters by using the measured storm flood data and historical flood mark data of the basin. For areas lacking measured data, use the parameter regionalization technology to determine the coupled model parameters. Then, according to the dominant environmental factors in each disaster-forming environment type zoning determined in Step 4, set up a dominant environmental factor scenario library, and analyze the impact of the change of dominant environmental factors on the mountain flood formation process and disaster-causing process of the study area. Furthermore, propose corresponding disaster prevention and mitigation measures for each disaster-forming environment type zoning.

2. The method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-bearing environment according to claim 1, characterized in that, The meteorological category factors described in Step 1 include the humidity, flood season precipitation, and short-duration extreme rainfall characteristics in the year when the mountain flood disaster event occurred. The short-duration extreme rainfall characteristics include magnitude, variability, and concentration; the geographical category factors include the undulation degree of the river basin topography, slope, proportion of forest land area, proportion of grassland area, proportion of farmland area, soil saturation moisture content, river length, river gradient, and river curvature in the year when the mountain flood disaster event occurred; the social and economic category factors include the GDP and population density in the year when the mountain flood disaster event occurred, and the water conservancy investment before the mountain flood disaster event.

3. A method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-bearing environment according to claim 1, characterized in that, The specific process of determining the preliminary zoning of the mountain flood disaster disaster-forming environment types in Step 2 is as follows: The clustering algorithm is used to classify all the environmental factors corresponding to historical mountain flood disasters in the environmental factor set of the study area, and the mountain flood disaster prone environment type partition CE’ is determined; the N-fold uncertainty analysis method is used to randomly divide the environmental factor set of the study area into N parts, select N-1 parts of the environmental factor data for clustering analysis, and repeat this process N times to determine the N-fold mountain flood disaster prone environment type partition CE N ; the hit rate is used to evaluate the coincidence degree of the mountain flood disaster prone environment type partition CE’ and the N-fold disaster prone environment type partition CE N The results are expressed in the form of the mean and standard deviation of the hit rate; if the hit rate reaches more than 95%, the mountain flood disaster prone environment type partition CE’ is used as the preliminary mountain flood disaster prone environment type partition; the hit rate calculation formula is: where PAC q is the q-fold hit rate, 1 ≤ q ≤ N; PC iq is the type of disaster-forming environment of the i-th historical mountain flood disaster event in the q-fold, 1 ≤ i ≤ m q , PC iq ∈ CE N ; AC i is the type of disaster-forming environment of the i-th historical mountain flood disaster event among all historical mountain flood disaster events, AC i ∈ CE’; m q is the number of historical mountain flood disaster events in the q-fold; Where: The specific process of determining the final zoning of the mountain flood disaster disaster-forming environment types is as follows: The Kruskal-Wallis test is used to evaluate the significant differences of all environmental factors in each disaster-forming environment type zone. If the calculated significance level p < 0.01, it indicates that there are significant differences in the environmental factors between the zones; when the number of disaster-forming environment type zones exceeds two, the Nemenyi test is further used to evaluate the significant differences between any two groups of disaster-forming environment type zones. If the calculated significance level p < 0.01, it indicates that there are significant differences in the environmental factors between any two groups of zones; The similarity test is used to evaluate the significant differences of the mountain flood disaster attributes in each disaster-forming environment type zone. If the calculated significance level p < 0.01, it indicates that there are significant differences in the mountain flood disaster attributes between the zones. The zones passing the above tests can be used as the final zoning of the disaster-forming environment types; if the above tests are not passed, the zoning of the mountain flood disaster disaster-forming environment types is re-divided until all tests are passed.

4. A method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-bearing environment according to claim 1, characterized in that, The specific process of quantifying the explanatory degree of different environmental categories for the changes in mountain flood disaster attributes and their uncertainty intervals in Step 3 is as follows: For a certain mountain flood disaster disaster-forming environment type zone, the meteorological category factors, geographical category factors, social and economic category factors, and the corresponding mountain flood disaster attributes in this zone are randomly divided into N parts. Select N - 1 parts of the environmental categories as independent variables, and the corresponding N - 1 parts of the mountain flood disaster attributes as response variables. The detrended correspondence analysis method is used to determine the length of the first sorting axis. According to the axis length, a suitable restricted sorting analysis method or canonical correspondence analysis method is further selected to determine the explanatory degree of different environmental categories for the changes in mountain flood disaster attributes. Repeat this process N times to determine the uncertainty intervals of the explanatory degrees of each environmental category, and express them in the form of the average value and standard deviation of the N results; The specific process of quantifying the explanatory degree, uncertainty interval, and significance of the changes in mountain flood disaster attributes for each environmental factor is as follows: For a specific mountain flood disaster disaster-forming environment type sub-region, the environmental factors and corresponding mountain flood disaster attributes within this sub-region are randomly divided into N parts. Select N - 1 parts of the environmental factors as independent variables and the corresponding N - 1 parts of the mountain flood disaster attributes as response variables. Conduct 999 Monte Carlo permutation tests to determine the explanatory degree and significance of each environmental factor on the response variable. If the significance level p < 0.01, it indicates that this environmental factor significantly affects the response variable. Repeat this process N times to determine the uncertainty interval and significance of the explanatory degrees of all environmental factors in this mountain flood disaster disaster-forming environment type sub-region, which are expressed in the form of the average value and standard deviation of the N results.

5. A method for analyzing the formation and disaster-causing mechanism of mountain floods based on the disaster-bearing environment according to claim 1, characterized in that The specific process of determining the dominant environmental factors and their differential effects in each disaster-forming environment type sub-region described in step 4 is as follows: For a specific mountain flood disaster disaster-forming environment type sub-region, screen out the significant environmental factors and arrange them in descending order according to the mean value of their explanatory degrees for the changes in mountain flood disaster attributes. The environmental factor with the largest explanatory degree is the first dominant factor, the environmental factors with explanatory degrees greater than or equal to 0.1 and less than that of the first dominant factor are the main dominant factors, and the remaining environmental factors are the secondary dominant factors. Compare the rankings of the dominant environmental factors in each mountain flood disaster disaster-forming environment type sub-region, and analyze the effects of each environmental factor on the formation and development process of mountain floods from the perspective of hydrological causes to determine the differential effects of the dominant environmental factors in each disaster-forming environment type sub-region.

6. The method for analyzing the formation and disaster-causing mechanism of mountain torrents based on the disaster-bearing environment according to claim 1, wherein The specific process of constructing the meteorological-hydrological-hydrodynamic coupled numerical model for the study area described in step 5 is as follows: Taking the small watershed as the basic calculation unit and the location where the historical mountain flood disaster event occurred as the watershed outlet, couple the meteorological numerical model, watershed hydrological model, and one-dimensional and two-dimensional hydrodynamic models to construct the meteorological-hydrological-hydrodynamic coupled numerical model for the study area; The specific process of analyzing the impacts of changes in dominant environmental factors on the mountain flood formation process and disaster-causing process in the study area is as follows: For the values of each dominant environmental factor in each disaster-forming environment type sub-region within the study area, conduct perturbations at a step size of 10% within the range of ±100% of each factor value to set up a scenario library of dominant environmental factors. Input each scenario into the calibrated meteorological-hydrological-hydrodynamic coupled numerical model to obtain the mountain flood process line, inundation range, inundation depth, and inundation duration information of the study area under different environmental factor scenarios. Compare the environmental factors and disaster attributes during the occurrence of historical mountain flood disasters to quantify the impacts of changes in dominant environmental factors on the mountain flood formation process and disaster-causing process in the study area.

Images