Forest grassland fire risk assessment method based on COPRAS model

Abstract

The invention discloses a forest and grassland fire risk assessment method based on a COPRAS model, and the method comprises the steps: determining disaster-causing factors of a forest and grassland fire, and dividing the disaster-causing factors into a combustible index, a meteorological condition index, a geographic and topographic index and a human activity index; performing screening and preprocessing on disaster-causing factors by using a principal component analysis method, extracting key indexes and constructing an evaluation index system; calculating a weight value of each evaluation index in the evaluation index system by adopting an analytic hierarchy process; based on the weight value of each evaluation index, constructing a forest steppe fire risk evaluation model based on a COPRAS model, and dividing forest steppe fire risk grades of the research area; performing precision verification on the forest steppe fire risk assessment model; and visualizing the forest steppe fire risk grade and the region division result of the research region. Therefore, the forest and steppe fire risk can be accurately evaluated, and the Greewing can better provide effective reference for predicting and preventing forest and steppe fire and formulating risk decision theory according to local conditions.

Description

Technical Field

[0001] The present invention relates to the field of forest and grassland fire risk assessment, and in particular to but not limited to a forest and grassland fire risk assessment method based on a COPRAS model. Background Art

[0002] Fire is an important disturbance factor affecting forest and grassland ecosystems, and has a significant impact on the structure and function of the ecosystem. For example, fire helps to improve vegetation productivity, release nutrients in litter, and maintain the stability and diversity of the ecosystem. However, fire has more negative effects on forest and grassland ecology. For example, fire burns vegetation, livestock and important species, destroys soil structure, increases the probability of sandstorms, and may have a negative impact on plant biomass and biodiversity. With global warming and the increase of human activities, the frequency, scope and severity of fires are showing an increasing trend. In recent years, the average annual number of forest and grassland fires in the world has reached hundreds of thousands, and more than one million hectares of forest and grassland have been burned, which has seriously damaged the sustainability of ecosystems and posed a huge threat to social and economic development and personal safety. Accurately estimating and predicting forest and grassland fire risks is of practical significance for building a forest and grassland fire defense system, reasonably dividing fire risk periods, and allocating limited fire prevention resources. However, forest and grassland fires involve multiple factors related to combustibles, meteorological conditions, geographical terrain and human activities, and the relationship between these factors is complex. The formation mechanism of forest and grassland fire risks is unclear, resulting in low assessment accuracy of forest and grassland fire risks. In practical applications, many factors used in the assessment are difficult to obtain. In many remote areas, due to the lack of forest and grassland fire data, it is difficult to timely assess and update forest and grassland fire risks.

[0003] Therefore, there is an urgent need to develop a forest and grassland fire risk assessment model that uses satellite remote sensing data and statistical data and is based on a multi-criteria decision analysis model to accurately assess the risk of forest and grassland fires, hoping to better provide an effective reference for predicting and preventing forest and grassland fires and formulating risk decision-making theories according to local conditions. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a forest and grassland fire risk assessment method based on the COPRAS model, which at least solves the problems of low accuracy and slow assessment and update speed in existing forest and grassland fire risk assessment.

[0005] The technical solutions of the embodiments of the present invention are as follows:

[0006] The embodiment of the present invention provides a forest and grassland fire risk assessment method based on the COPRAS model, the method comprising:

[0007] Determine the disaster-causing factors of forest and grassland fires, and divide them into combustible indicators, meteorological condition indicators, geographical terrain indicators and human activity indicators; use the principal component analysis method to screen and preprocess the disaster-causing factors, extract key indicators and construct an evaluation index system; use the hierarchical analysis method to calculate the weight value of each evaluation index in the evaluation index system; based on the weight value of each evaluation index, construct a forest and grassland fire risk assessment model based on the COPRAS model, and divide the forest and grassland fire risk level of the study area; verify the accuracy of the forest and grassland fire risk assessment model; visualize the forest and grassland fire risk level and regional division results of the study area.

[0008] Furthermore, the combustible material indicators include: combustible material load, combustible material moisture content, NDVI, vegetation coverage, and vegetation type; the meteorological indicators include: FWI, surface temperature, average temperature, maximum temperature, average wind speed, maximum wind speed, average rainfall, precipitation index, relative humidity, evapotranspiration, solar radiation, and humidity index; the terrain indicators include: composite terrain index, altitude, slope, and slope aspect; the human activity indicators include: distance from roads, distance from settlements, distance from rivers, distance from railways, distance from power lines, distance from fire stations, population, density of important fire sources, and land use.

[0009] Furthermore, the disaster-causing factors are screened and preprocessed using the principal component analysis method, key indicators are extracted and an evaluation indicator system is constructed, including: using the principal component analysis method to select the main variables in the disaster-causing factors whose cumulative contribution rate reaches more than 85% as the key indicators; the key indicators are plotted in ArcGIS software as a raster layer with a unified projection system, coordinate system and grid to obtain the evaluation indicator system.

[0010] Furthermore, the use of the analytic hierarchy process to calculate the weight value of each evaluation index in the evaluation index system includes: establishing a hierarchical structure model, determining the target layer, the criterion layer and the scheme layer; the target layer is to evaluate the forest and grassland fire risk, the criterion layer is each evaluation index in the evaluation index system, and the scheme layer is a selectable scheme for decision-making; the nine-scale method is used to construct a judgment matrix for each evaluation index in the criterion layer; wherein the element a in the judgment matrix ij Indicates the importance of evaluation index i relative to evaluation index j; calculates the relative weight and combined weight of each evaluation index based on the judgment matrix; and verifies the consistency of the judgment matrix.

[0011] Furthermore, based on the weight values ​​of the evaluation indicators, a forest and grassland fire risk assessment model based on the COPRAS model is constructed, and the forest and grassland fire risk levels of the study area are divided, including: based on the evaluation indicators, a decision matrix of the COPRAS model is constructed; wherein the elements in the decision matrix are the indicator values ​​of the corresponding indicators; each element in the decision matrix is ​​normalized to obtain a normalized decision matrix, and based on the weight values ​​of the evaluation indicators, weights are assigned to each element in the decision matrix to obtain a weighted normalized decision matrix; the weighted normalized decision matrix is ​​assigned to each element in the decision matrix to obtain a weighted normalized decision matrix; The dimensionless values ​​in the unified decision matrix are divided into positive ideal solutions representing favorable factors for fire occurrence and negative ideal solutions representing unfavorable factors for fire occurrence; based on the positive ideal solutions and negative ideal solutions, the comprehensive evaluation values ​​of the alternatives in the study area are determined; wherein the alternatives represent each unit grid with a risk value in the study area; based on the comprehensive evaluation values, the utility of the alternatives is determined and ranked; based on the ranked utility, the natural breakpoint method is used to divide the forest and grassland fire assessment risk level in the study area into five levels: high, medium-high, medium, medium-low and low.

[0012] Furthermore, the accuracy verification of the forest and grassland fire risk assessment model includes: analyzing the overlap rate between the burned area and the forest and grassland fire risk level area to determine the simulation accuracy of the forest and grassland fire risk assessment model; and / or analyzing the overlap rate between historical fire points and the forest and grassland fire risk level area to determine the simulation accuracy of the forest and grassland fire risk assessment model.

[0013] Furthermore, the visualization of forest and grassland fire risk levels and regional division results in the study area includes: regional division of forest and grassland fire risk levels in the study area according to thresholds of five risk levels of forest and grassland fire high, medium-high, medium, medium-low and low; and visualization conversion of regional division results using ArcGIS software.

[0014] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0015] In the embodiment of the present invention, COPRAS and GIS technology are combined to construct a forest grassland fire risk assessment model using data on disaster-causing factors such as combustibles, meteorological conditions, geographical terrain and human activities, which can provide intuitive, detailed and high-precision technical means for forest grassland fire risk assessment and forest grassland fire risk zone delineation. The present invention uses the COPRAS model to carry out forest grassland fire risk assessment, and then divides the forest grassland fire risk level. Compared with the existing forest grassland fire risk assessment method, it can better provide effective references in guiding forest grassland fire prevention work, curbing the occurrence of forest grassland fires, and mitigating the environmental effects of forest grassland fires, thereby improving the efficiency of forest grassland fire protection and strengthening the effect of forest grassland fire protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0017] Figure 1 A schematic diagram of a forest and grassland fire risk assessment method based on the COPRAS model provided in an embodiment of the present invention;

[0018] Figure 2 A schematic diagram of the classification results of forest and grassland fire risk levels provided by an embodiment of the present invention;

[0019] Figure 3 A schematic diagram of the forest and grassland fire risk area division result verification based on the burned area of ​​the study area provided in an embodiment of the present invention;

[0020] Figure 4 A schematic diagram of verifying the forest and grassland fire risk area division results based on forest and grassland fire point data provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0023] It should be pointed out that the terms "first\second\third" involved in the embodiments of the present invention are only used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present invention described here can be implemented in an order other than that illustrated or described here.

[0024] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the field to which the embodiments of the present invention belong. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0025] The COPRAS model is one of the latest multi-criteria decision analysis models. The COPRAS model is a comprehensive evaluation model that combines mathematical methods, decision theory, expert judgment and empirical knowledge. The model has many advantages, such as simple algorithm, short calculation time, effective integration of multiple factors and decision makers' preferences, and adaptability to local empirical conditions and actual conditions. However, there is limited research on the application of the COPRAS model in forest and grassland fire risk.

[0026] Figure 1 A flowchart of a forest and grassland fire risk assessment method based on the COPRAS model provided in an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method comprises at least the following steps:

[0027] Step S110, determining the disaster-causing factors of forest and grassland fires, and dividing them into combustible material indicators, meteorological condition indicators, geographical terrain indicators and human activity indicators.

[0028] Here, according to the guidelines of forest and grassland fire prevention and control plans in various places, the disaster-causing factors of forest and grassland fires are determined by integrating existing research. For example, the content required for forest and grassland fire prevention and control and prediction in this area is extracted from the content of "prevention, monitoring and early warning" related to forest and grassland fires. By combining the existing forest and grassland fire research literature, the disaster-causing factors that affect the occurrence and development of forest and grassland fires are collected. Preferably, a total of 30 disaster-causing factors are collected as evaluation indicators.

[0029] The attributes of each indicator factor in the disaster-causing factors of forest and grassland fires are clarified, and the collected assessment indicators that affect the occurrence and development of forest and grassland fires are divided into four categories: combustibles, meteorological conditions, geographical terrain, and human activities. Among them, combustibles are the basis for fire combustion and spread. Meteorological conditions affect air humidity and combustible dryness, which further determines flammability. Topographic factors are important influencing factors before and during fires, determining the spread rate of fire. Fire sources are divided into man-made fire points and natural fire points. Studies have shown that more than 90% of fires are caused by human factors. The environmental conditions and fire characteristics of different regions vary greatly. This method realizes the cross-regional application of fire risk assessment models by considering regional specific environmental factors.

[0030] Step S120, using the principal component analysis method to screen and preprocess the disaster-causing factors, extract key indicators and construct an evaluation indicator system.

[0031] Here, in the evaluation process, a large amount of raw data corresponding to the disaster-causing factors needs to be preprocessed and key indicators need to be screened. The implementation process of the principal component analysis (PCA) method is as follows: First, the collected data is standardized to eliminate the dimensional influence between different indicators so that the data is at the same level for subsequent analysis. Then, the principal component analysis method is used to calculate the correlation coefficient matrix to select the principal component as the key indicator according to the size of the eigenvalue, and then the key indicator is used as a new evaluation indicator to establish an evaluation indicator system for forest and grassland fire risks. In this way, multicollinearity between evaluation indicators is avoided through screening and preprocessing, that is, different indicators are highly correlated, resulting in unstable or difficult to interpret model results. Prevent model overfitting, that is, the model is too complex and fits the training data too perfectly, but has poor generalization ability for new data.

[0032] Step S130, using the analytic hierarchy process to calculate the weight value of each evaluation index in the evaluation index system.

[0033] Here, in the multi-criteria decision analysis, determining the weight of each indicator is a key step. The embodiment of the present invention adopts the analytic hierarchy process (AHP) to calculate the weight value of each evaluation indicator in the evaluation indicator system. It mainly includes: establishing a hierarchical structure model, constructing a judgment matrix, calculating eigenvalues ​​and eigenvectors, calculating weight values, and consistency testing. Through these steps, a quantitative weight value can be provided for each evaluation indicator in the forest and grassland fire risk assessment indicator system. These weight values ​​reflect the relative importance of each indicator in risk assessment and provide a scientific basis for subsequent risk assessment and decision-making.

[0034] Step S140, based on the weight values ​​of the evaluation indicators, construct a forest and grassland fire risk assessment model based on the COPRAS model, and divide the forest and grassland fire risk levels in the study area.

[0035] Here, forest and grassland fire risk assessment involves multiple factors, such as climate conditions, vegetation conditions, topography, etc. COPRAS (Compromise Optimization and Programming under Ratio and interval Scales) is a multi-criteria decision analysis method that ranks each alternative solution by calculating its proximity to the ideal solution, thereby helping decision makers make decisions under many decision criteria and solving the complexity problem of multi-criteria decision analysis. The model construction steps include: determining the evaluation index; standardizing the index; calculating the index weight; combining the weight values ​​of each index, constructing the COPRAS model, and calculating the risk score of each evaluation object (such as different regions) in the study area.

[0036] The COPRAS model is a comprehensive evaluation model that combines mathematical methods, decision-making theory, expert judgment and empirical knowledge. It has the advantages of simple algorithm, short calculation time, effective integration of multiple factors and decision-makers' preferences, and adaptability to local experience conditions and actual conditions. In the evaluation event, the COPRAS model calculates the comprehensive score of the evaluation object by using the weighted benefits and costs of each indicator. Even if it is necessary to weigh multiple influencing factors, it can still simulate a more reasonable result. The spatial distribution of forest and grassland fires is affected by many indicators such as combustibles, meteorological conditions, geographical terrain and human activities. Therefore, the COPRAS model can be used to evaluate and zoning such complex situations with multiple risk factors.

[0037] Step S150, verifying the accuracy of the forest and grassland fire risk assessment model.

[0038] Here, accuracy verification of the assessment model is an important step to ensure the reliability of the assessment results. In some embodiments, the accuracy of the forest and grassland fire risks in the simulated area is verified, for example, by using the historical burned area and / or forest and grassland fire point data of the study area, and comparing the zoning results output by the forest and grassland fire risk assessment model to verify the credibility of the model assessment results. In other embodiments, the accuracy of the model can also be evaluated using confusion matrices and receiver operating characteristic curves, which can be used for accuracy verification.

[0039] Step S160, visualizing the forest and grassland fire risk level and regional division results of the study area.

[0040] Here, the fire risk coefficient data calculated by the COPRAS model is imported into ArcGIS, and ArcGIS is used for visualization. The mapping function of ArcGIS is used to generate a forest and grassland fire risk map containing different risk levels, and then the generated risk level map is displayed to decision makers and relevant personnel through the display function of ArcGIS. This effectively visualizes the results of forest and grassland fire risk assessment, helps relevant personnel better understand and manage fire risks, and thus improves the efficiency of fire prevention and response.

[0041] The embodiment of the present invention combines COPRAS and GIS technology, and uses data on disaster-causing factors such as combustibles, meteorological conditions, geographical terrain and human activities to construct a forest and grassland fire risk assessment model, which can provide intuitive, detailed and high-precision technical means for forest and grassland fire risk assessment and forest and grassland fire risk zone delineation.

[0042] In some embodiments, the combustible material indicators include: combustible material load, combustible material moisture content, NDVI, vegetation coverage, and vegetation type; the meteorological indicators include: FWI, surface temperature, average temperature, maximum temperature, average wind speed, maximum wind speed, average rainfall, precipitation index, relative humidity, evapotranspiration, solar radiation, and humidity index; the terrain indicators include: composite terrain index, altitude, slope, and slope aspect; the human activity indicators include: distance from roads, distance from settlements, distance from rivers, distance from railways, distance from power lines, distance from fire stations, population, density of important fire sources, and land use.

[0043] Here, the Normalized Difference Vegetation Index (NDVI) is a widely used remote sensing indicator for assessing the condition of surface vegetation. NDVI can be used to assess the health of forests. High NDVI values ​​usually indicate that the forest cover is large and healthy, while low values ​​may indicate that the forest cover is small or has health problems. The slope and aspect indicators in the terrain indicators can be calculated using the spatial analysis tools in ArcGIS software. The Fire Weather Index (FWI) is a system for assessing the likelihood and behavior of forest fires. It is mainly used in fire management, especially in predicting and preventing forest fires.

[0044] Human activity indicators are one of the important factors in assessing forest and grassland fire risks, because they directly affect the probability of fire occurrence, the speed of fire spread, and the efficiency of fire management. When constructing a forest and grassland fire risk assessment model, these human activity indicators can be spatially analyzed and quantified using GIS (Geographic Information System) technology. By combining these indicators with other natural factors (such as meteorological conditions, topography, vegetation type, etc.), a more comprehensive assessment of fire risk can be made, providing a scientific basis for fire prevention, emergency response, and resource allocation.

[0045] In some embodiments, the above-mentioned step S120 "using the principal component analysis method to screen and preprocess the disaster-causing factors, extract key indicators and construct an evaluation indicator system" includes the following steps: using the principal component analysis method to select the main variables in the disaster-causing factors with a cumulative contribution rate of more than 85% as the key indicators; drawing the key indicators into a raster layer with a unified projection system, coordinate system and grid in ArcGIS software to obtain the evaluation indicator system.

[0046] Here, the implementation process of the principal component analysis method is as follows: First, the correlation coefficient matrix is ​​calculated for the standardized disaster factor data to reflect the correlation between the indicators. Then the correlation coefficient matrix is ​​used to calculate the eigenvalues ​​and the corresponding eigenvectors. According to the size of the eigenvalues, the first few principal components are selected, and the cumulative contribution rate of these principal components reaches more than 85% to ensure that most of the key information is retained. The selected principal components are used as new evaluation indicators to construct a forest and grassland fire risk assessment index system. Each principal component is a linear combination of the original indicators, and the contribution of the original indicators to each principal component can be determined by the coefficient of the eigenvector. Finally, each principal component is interpreted to determine their practical significance in forest and grassland fire risk assessment. Through principal component analysis, multiple highly correlated variables are merged into a few principal components to achieve data dimensionality reduction and simplify the model complexity. Selecting principal components with a cumulative contribution rate of more than 85% ensures that most of the key information is retained, while eliminating redundant variables. Reducing collinearity can improve the stability of the model and the accuracy of prediction.

[0047] It is worth noting that because the coordinate system and spatial resolution of each indicator data are different, the resolution and geographic coordinate system of all data must be unified. The embodiment of the present invention achieves the unification of data format and coordinate system by drawing the selected key indicators into a raster layer with a unified projection system, coordinate system and grid in ArcGIS software, ensuring that all indicator factors have the same resolution and alignment in space for accurate spatial overlay and analysis. The generated raster layer can be directly used for spatial analysis, such as spatial overlay, buffer zone analysis, etc., to provide necessary spatial data support for forest and grassland fire risk assessment models. The unified raster layer format simplifies the data processing process and improves the efficiency and accuracy of data processing.

[0048] In some embodiments, the above step S130 "using the analytic hierarchy process to calculate the weight value of each evaluation indicator in the evaluation indicator system" includes the following steps: establishing a hierarchical structure model, determining the target layer, the criterion layer and the solution layer; the target layer is to evaluate the forest and grassland fire risk, the criterion layer is each evaluation indicator in the evaluation indicator system, and the solution layer is a solution available for decision-making; using the nine-scale method to construct a judgment matrix for each evaluation indicator in the criterion layer; wherein the element a in the judgment matrix ij Indicates the importance of evaluation index i relative to evaluation index j; calculates the relative weight and combined weight of each evaluation index based on the judgment matrix; and verifies the consistency of the judgment matrix.

[0049] Here, after the nine-scale method is used to construct the judgment matrix, the weights of each evaluation index are calculated using formulas (1) and (2):

[0050]

[0051] In the formula, a ij is the element in the i-th row and j-th column of the judgment matrix, n is the order of the judgment matrix; M i is the geometric mean of the elements in the i-th row; w i is the weight value of each evaluation indicator. If there are multiple criterion layers, it is necessary to calculate the combined weight of each factor in the criterion layer, which usually involves multiplying the weight of each criterion layer by the weight of its subordinate factors and then summing them.

[0052] The consistency index and ratio are calculated by formulas (3) and (4), and the consistency of the judgment matrix is ​​tested to avoid contradictions in subjective judgment.

[0053]

[0054] Among them, λ max is the maximum eigenvalue in the judgment matrix, CI is the consistency index, RI is the random consistency index, CR is the consistency ratio. When CR<0.1, it indicates that the judgment matrix meets the consistency requirements. If CR>0.1, the judgment matrix needs to be readjusted to improve consistency.

[0055] Through these steps, the hierarchical analysis method can provide a quantitative weight value for each evaluation indicator in forest and grassland fire risk assessment. These weight values ​​reflect the importance of each indicator relative to the overall goal, thus providing a scientific basis for decision-making.

[0056] In some embodiments, the above step S140 "based on the weight values ​​of each evaluation index, constructing a forest and grassland fire risk assessment model based on the COPRAS model, and dividing the forest and grassland fire risk level of the study area" includes the following steps: based on each evaluation index, constructing a decision matrix of the COPRAS model; wherein the elements in the decision matrix are the index values ​​of the corresponding indexes; normalizing each element in the decision matrix to obtain a normalized decision matrix, and assigning weights to each element in the normalized decision matrix based on the weight values ​​of each evaluation index to obtain a weighted normalized decision matrix; Matrix; divide the dimensionless values ​​in the weighted normalized decision matrix into positive ideal solutions representing favorable factors for fire occurrence and negative ideal solutions representing unfavorable factors for fire occurrence; determine the comprehensive evaluation value of each alternative option in the study area based on the positive ideal solution and the negative ideal solution; wherein the alternative option represents each unit grid with a risk value in the study area; based on the comprehensive evaluation value, determine the utility degree of each alternative option and rank them; based on the ranked utility degree, use the natural breakpoint method to divide the forest and grassland fire assessment risk level in the study area into five levels: high, medium-high, medium, medium-low and low.

[0057] Here, first of all, it is necessary to normalize the element data in the decision matrix to eliminate the dimensional influence between different indicators. Then, a forest and grassland fire risk assessment model based on the COPRAS model is constructed, and the utility of each option in the study area is comprehensively evaluated through the relevant formulas of the COPRAS model.

[0058] The natural breakpoint method (Jenks) is a classification method that divides data into several categories so that the values ​​within each category are as similar as possible, while the differences between different categories are as large as possible. The utility output by the COPRAS model is sorted from high to low, and then the risk level is divided into five levels: high, medium-high, medium, medium-low, and low according to the score range. For example, the thresholds of the risk coefficient are set to 0-0.2 (low risk), 0.2-0.4 (medium-low risk), 0.4-0.6 (medium risk), 0.6-0.8 (medium-high risk), and 0.8-1.0 (high risk).

[0059] In some embodiments, the above-mentioned step S150 "verifying the accuracy of the forest and grassland fire risk assessment model" is implemented through the following process: analyzing the overlap rate between the burned area and the forest and grassland fire risk level area to determine the simulation accuracy of the forest and grassland fire risk assessment model; and / or, analyzing the overlap rate between historical fire points and forest and grassland fire risk level areas to determine the simulation accuracy of the forest and grassland fire risk assessment model.

[0060] Here, burned area data: Collect burned area data after actual fires occur, which can be obtained through remote sensing satellite images, on-site surveys or historical records.

[0061] Risk level area data: Use the output from the forest and grassland fire risk assessment model to divide the study area into different risk level areas.

[0062] Historical fire point data: Collect historical fire point data over a period of time. These data can be used to verify the simulation accuracy of the model.

[0063] Analyzing the overlap rate between the burned area and the risk level area includes: In GIS, spatially overlaying the burned area layer with the risk level area layer. Using the analysis tools of the GIS software, calculate the overlap rate between the burned area and the high-risk area. Evaluate model accuracy: If the overlap rate is high, it means that the model can better predict the high-risk area and the simulation accuracy is high.

[0064] Analyze the overlap rate between historical fire points and risk level areas: Analyze the historical fire point layer and the risk level area layer spatially. Count the number of historical fire points that fall in high-risk areas and calculate the overlap rate. Evaluate model accuracy: If the overlap rate is high, it means that the model can better predict the risk area of ​​fire occurrence and the simulation accuracy is high.

[0065] In this way, the simulation accuracy of the forest and grassland fire risk assessment model can be effectively verified, the reliability of the model results can be ensured, and a scientific basis can be provided for fire prevention and resource allocation.

[0066] In some embodiments, the above-mentioned step S160 "visualizing the forest and grassland fire risk levels and regional division results of the study area" is implemented through the following process: dividing the forest and grassland fire risk levels of the study area into regions based on the thresholds of five risk levels of forest and grassland fire high, medium-high, medium, medium-low and low; and using ArcGIS software to visualize the regional division results.

[0067] Here, the fire risk coefficient data calculated by the COPRAS model is imported into ArcGIS, and the natural breakpoint method function of ArcGIS is used to generate a forest and grassland fire risk level map to intuitively display the risk levels of different areas. The "reclassify" tool of ArcGIS is used to reclassify the risk coefficient data into five risk levels according to a determined threshold. In some embodiments, in ArcGIS, the study area is colored according to the risk level to generate a forest and grassland fire risk level map. In other embodiments, the symbols, colors, and layout of the map are adjusted as needed to ensure that the map information is clear and easy to understand.

[0068] For example, open the ArcGIS SpatialAnalyst Tools toolbox, select the Reclassify tool, use the risk factor raster layer as input, set the reclassification field and threshold, and create a new classification layer. Symbolize the reclassified risk level layer and select different colors or patterns for each level. Select "Symbology" in the layer properties and select a color scheme based on the risk level, such as using a gradient or different color fills. Add a legend to the map layout to clearly show the risk levels represented by different colors or patterns. Add annotations to important areas to provide additional risk level information.

[0069] The forest and grassland fire risk assessment method based on the COPRAS model is described below in conjunction with a specific embodiment. However, it should be noted that this specific embodiment is only for better illustrating the present invention and does not constitute an improper limitation to the present invention.

[0070] The study area of ​​this specific embodiment is located in the northwest of Xinjiang Uygur Autonomous Region, China, with a geographical location between 79°57′-87°09′E and 41°58′-49°09′N, from the Tianshan Mountains in the south to the Altai Mountains in the north, and is located on the west side of the Junggar Basin. The study area has a typical temperate continental climate, with hot and dry weather and a large temperature difference between day and night. The altitude of the study area is between 183 and 4384 meters, with an average temperature of -9-12°C over many years, and a low average precipitation over many years, with an annual average precipitation of no more than 200 mm. Due to the unique terrain conditions, geographical location and arid climate, the potential risk of forest and grassland fires in the study area continues to rise.

[0071] Step S1: According to the guidelines of forest and grassland fire prevention and control plans in various places, the disaster-causing factors of forest and grassland fires are determined by integrating existing research: According to the guidelines of the forest and grassland fire prevention and control plan of a certain area, the content required for forest and grassland fire prevention and control and prediction in the region is extracted from the content of "prevention, monitoring and early warning" of forest and grassland fires. By combining the existing forest and grassland fire research literature, the indicator factors that affect the occurrence and development of forest and grassland fires are collected (a total of 30 evaluation indicators are collected).

[0072] Step S2: clarify the attributes of each indicator factor in the forest and grassland fire disaster factors in step S1, and divide the collected indicator factors that affect the occurrence and development of forest and grassland fires into four categories: combustibles, meteorological conditions, geographical terrain and human activities. Combustibles are the basis for the combustion and spread of fires. Meteorological conditions affect air humidity and the dryness of combustibles, which further determines the flammability. Topographic factors are important influencing factors before and during the occurrence of fires, and determine the spread rate of the fire. Fire sources are divided into man-made fire points and natural fire points. Studies have shown that more than 90% of fires are caused by human factors.

[0073] Combustible material indicators include: combustible material load, combustible material moisture content, NDVI, vegetation coverage, and vegetation type.

[0074] Meteorological indicators include: FWI, surface temperature, average temperature, maximum temperature, average wind speed, maximum wind speed, average rainfall, precipitation index, relative humidity, evapotranspiration, solar radiation, and humidity index.

[0075] Topographic indicators include: composite topographic index, altitude, slope, and aspect. The slope and aspect indicators of the forest-grassland fire risk study area were calculated using the spatial analysis tools in ArcGIS software.

[0076] Human activity indicators include: distance to roads, distance to settlements, distance to rivers, distance to railways, distance to power lines, distance to fire stations, population, density of important fire sources, and land use.

[0077] Step S3: Use the principal component analysis method to screen the indicators that affect the occurrence of grassland and forest fires, and select the main variables with a cumulative contribution rate of more than 85%. This study selected 13 representative key indicators as forest and grassland fire risk influencing factors from four aspects: combustibles (combustible load), meteorological conditions (monthly average wind speed, monthly average rainfall, monthly average temperature, monthly maximum temperature and monthly minimum relative humidity), geographical terrain (altitude, slope, slope aspect) and human activities (density of important fire sources, population, distance from roads, distance from residential areas), and generated a forest and grassland fire risk assessment indicator system.

[0078] A detailed description of the key indicators is shown in Table 1 below. The required data are from the National Earth System Science Data Center (http: / / www.geodata.cn / data / ), Geospatial Data Cloud Platform (http: / / www.gsccloud.cn), Open Street Map (https: / / www.openstreetmap.org / ), and Open Spatial Demographic Data and Research (https: / / hub.worldpop.org).

[0079] Table 1 Key indicators for forest and grassland fire risk assessment

[0080]

[0081]

[0082] Step S4: In this specific embodiment, when determining the weight of each indicator, the weight value of each indicator in the forest and grassland fire risk assessment is calculated mainly based on the characteristics of the study area using the hierarchical analysis method, which can be adjusted later according to business requirements.

[0083] The specific implementation process is as follows: construct a pairwise comparison judgment matrix for the 13 key indicators of forest and grassland fire risk assessment described in step S3. For forest and grassland fire risk, compare the relative importance of each pair of factors (each two forest and grassland fire risk assessment indicators). The hierarchical analysis method pairwise judgment matrix and the weight value of each indicator are shown in Table 2:

[0084] Table 2 Comparative judgment matrix

[0085]

[0086] Step S5: Construct a forest and grassland fire risk assessment model based on the COPRAS model to divide the forest and grassland fire risk levels in the study area.

[0087] Here, the following steps S501 to S506 are implemented:

[0088] S501: Construct the decision matrix of the COPRAS model based on the selected 13 key indicators that affect the occurrence of forest and grassland fires.

[0089] S502: normalizing each indicator element in the decision matrix to obtain a normalized decision matrix.

[0090] Here, for each indicator element X in the decision matrix ij , normalized by formula (5), and then determined by formula (6) as the weighted normalized decision matrix V ij :

[0091]

[0092] V ij =ω i × ij Formula (6);

[0093] Where M is the total number of grids, r ij is the dimensionless value of each indicator in the normalized decision matrix, ω i is the weight value corresponding to each indicator.

[0094] S503: Divide each dimensionless index in the weighted normalized decision matrix into a positive ideal solution representing favorable factors for fire occurrence and a negative ideal solution representing unfavorable factors for fire occurrence.

[0095] Here, the positive ideal solution includes: combustible load, average wind speed, average temperature, maximum temperature, slope, slope direction, fire source density, and population density. The negative ideal solution includes: average rainfall, minimum relative humidity, altitude, distance to roads, and distance to residential areas.

[0096] The values ​​of the positive ideal solution and the negative ideal solution are calculated by formulas (7) and (8) respectively:

[0097]

[0098] In the formula, m is the number of indicators included in the positive ideal solution; n is the total number of indicators, nm is the number of indicators included in the negative ideal solution; S +i is the value of the positive ideal solution of favorable factors for fire occurrence; S -i is the value of the negative ideal solution of the unfavorable factors for fire occurrence.

[0099] S504: The spatial distribution of forest and grassland fires is affected by both favorable factors and unfavorable factors. The COPRAS model fully considers the values ​​of favorable factors (maximization) and unfavorable factors (minimization) and calculates the comprehensive evaluation value Q of the alternatives according to formula (9): i :

[0100]

[0101] S505: Based on the comprehensive evaluation value Q i Calculate the utility U of each alternative i , and sort them. The calculation formula is as follows:

[0102]

[0103] In the formula, Q max Represents the comprehensive evaluation value Q in the study area i The maximum value of U i Indicates the utility of the alternative, with a value between 0 and 1. i The larger the value, the higher the risk of forest and grassland fires.

[0104] S506: According to the utility level U i Carry out forest and grassland fire risk zoning.

[0105] The natural breakpoint method is used to divide the forest and grassland fire assessment risk values ​​in the study area into five levels, such as Figure 2As shown, there are low, medium-low, medium, medium-high and high risk areas, accounting for 10.38%, 23.61%, 28.75%, 25.52% and 11.74% of the total area of ​​the study area, respectively. The areas at medium-low risk and low risk levels account for 33.99% of the total area, while the areas at high and medium-high risk levels account for 37.26% of the total area. It can be seen that the possibility of forest and grassland fires in the study area is high.

[0106] From the spatial distribution of forest and grassland fire risk levels, the forest and grassland fire risk is generally high in the southwest and low in the northeast. High-risk and medium-high-risk areas are mainly distributed in the southwest and central and western parts of the study area. Low-risk areas are mainly distributed in the northeast and southwest of the study area near the Tianshan Mountains. This distribution pattern is basically consistent with the spatial distribution of the burned area extracted from MODIS satellite data.

[0107] Step S5: Verify the accuracy of forest and grassland fire risks in the simulated area.

[0108] The burned area in the study area from 2000 to 2021 was used to verify the results of the forest and grassland fire risk area division, such as Figure 3 The results show that the proportions of the observed burned areas falling into low, medium-low, medium, medium-high and high risk levels are 2.55%, 4.77%, 8.98%, 41.78% and 41.92% respectively, and 83.70% of the burned areas are distributed in high-risk and medium-high-risk areas for forest and grassland fires.

[0109] The forest and grassland fire point data of the study area from 2003 to 2023 were used to verify the results of the forest and grassland fire risk area division, such as Figure 4 The results show that the proportions of historical fire points observed falling into low, medium-low, medium, medium-high and high risk areas are 12.46%, 11.28%, 18.99%, 24.04% and 33.23% respectively, and 57.27% of historical forest and grassland fire points are distributed in high-risk, medium-high and high-risk areas for forest and grassland fires.

[0110] This shows that the forest and grassland fire risk zoning based on the COPRAS model can accurately simulate the high-incidence and prone areas of forest and grassland fires, which to a large extent verifies the reliability of the model's regional division results.

[0111] Step S6: Use grid graphic results display to make the evaluation results more intuitive. Set different colors for different forest and grassland fire risk level areas to complete the forest and grassland fire risk level zoning. The research results help fire prevention workers and other relevant personnel to directly understand the judgment of the forest and grassland fire risk level of the target corresponding area through the grid color.

[0112] Combining COPRAS and GIS technology, using data on disaster-causing factors such as combustibles, meteorological conditions, geographical terrain and human activities to construct a forest grassland fire risk assessment model can provide intuitive, detailed and high-precision technical means for forest grassland fire risk assessment and forest grassland fire risk zoning. The present invention uses the COPRAS model to carry out forest grassland fire risk assessment, and then divides the forest grassland fire risk level. Compared with the existing forest grassland fire risk assessment method, it can better provide effective references in guiding forest grassland fire prevention work, curbing the occurrence of forest grassland fires, and mitigating the environmental effects of forest grassland fires, thereby improving the efficiency of forest grassland fire protection and strengthening the effect of forest grassland fire protection.

[0113] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present invention, the size of the serial number of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention. The serial numbers of the above-mentioned embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments.

[0114] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0115] In the several embodiments provided by the present invention, it should be understood that the disclosed methods can be implemented in other ways. The methods disclosed in the several method embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments. The features disclosed in the several method embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0116] The above is only an embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A forest and grassland fire risk assessment method based on the COPRAS model, characterized in that: include: Determine the factors causing forest and grassland fires and divide them into fuel index, meteorological condition index, geographical terrain index and human activity index; Use principal component analysis to screen and preprocess the disaster-causing factors, extract key indicators and construct an evaluation indicator system; The weight value of each evaluation index in the evaluation index system is calculated by using the hierarchical analysis method; Based on the weight values ​​of the evaluation indicators, a forest and grassland fire risk assessment model based on the COPRAS model is constructed, and the forest and grassland fire risk levels in the study area are divided; Verify the accuracy of the forest and grassland fire risk assessment model; Visualize the forest and grassland fire risk levels and regional division results in the study area.

2. The method according to claim 1, characterized in that The combustible material indexes include: combustible material load, combustible material moisture content, normalized difference vegetation index NDVI, vegetation coverage, and vegetation type; The meteorological indicators include: fire weather index FWI, surface temperature, average temperature, maximum temperature, average wind speed, maximum wind speed, average rainfall, precipitation index, relative humidity, evapotranspiration, solar radiation, and humidity index; The terrain indicators include: composite terrain index, altitude, slope, and slope aspect; The human activity indicators include: distance from roads, distance from settlements, distance from rivers, distance from railways, distance from power lines, distance from fire stations, population, density of important fire sources, and land use.

3. The method according to claim 1, characterized in that The principal component analysis method is used to screen and preprocess the disaster-causing factors, extract key indicators and construct an evaluation indicator system, including: Using the principal component analysis method, select the main variables with a cumulative contribution rate of more than 85% among the disaster-causing factors as the key indicators; The key indicators are plotted in ArcGIS software as a raster layer with a unified projection system, coordinate system and grid to obtain the evaluation indicator system.

4. The method according to claim 1, characterized in that: The method of using the analytic hierarchy process to calculate the weight value of each evaluation index in the evaluation index system includes: Establish a hierarchical structure model to determine the target layer, criterion layer and solution layer; the target layer is to assess the forest and grassland fire risk, the criterion layer is the evaluation indicators in the evaluation indicator system, and the solution layer is the available options for decision-making; The nine-scale method is used to construct a judgment matrix for each evaluation index in the criterion layer; wherein the element a in the judgment matrix ij Indicates the importance of evaluation index i relative to evaluation index j; Calculate the relative weight and combined weight of each evaluation index based on the judgment matrix; The consistency of the judgment matrix is ​​checked.

5. The method according to any one of claims 1 to 4, characterized in that: Based on the weight values ​​of the evaluation indicators, a forest and grassland fire risk assessment model based on the COPRAS model is constructed, and the forest and grassland fire risk levels in the study area are divided, including: Based on the evaluation indicators, a decision matrix of the COPRAS model is constructed; wherein the elements in the decision matrix are the indicator values ​​of the corresponding indicators; Normalizing each element in the decision matrix to obtain a normalized decision matrix, and assigning a weight to each element in the normalized decision matrix based on the weight value of each evaluation index to obtain a weighted normalized decision matrix; Dividing the dimensionless values ​​in the weighted normalized decision matrix into positive ideal solutions representing favorable factors for fire occurrence and negative ideal solutions representing unfavorable factors for fire occurrence; Based on the positive ideal solution and the negative ideal solution, a comprehensive evaluation value of each candidate in the study area is determined; wherein the candidate represents each unit grid having a risk value in the study area; Based on the comprehensive evaluation value, determining the utility of each of the alternatives and ranking them; Based on the utility degree after sorting, the forest and grassland fire assessment risk level in the study area is divided into five levels: high, medium-high, medium, medium-low and low using the natural breakpoint method.

6. The method according to any one of claims 1 to 4, characterized in that: The accuracy verification of the forest and grassland fire risk assessment model includes: Analyze the overlap rate between the burned area and the forest and grassland fire risk level area to determine the simulation accuracy of the forest and grassland fire risk assessment model; and / or, The overlap rate between historical fire points and forest and grassland fire risk level areas is analyzed to determine the simulation accuracy of the forest and grassland fire risk assessment model.

7. The method according to any one of claims 1 to 4, characterized in that: The visualization of the forest and grassland fire risk level and regional division results of the study area includes: The forest and grassland fire risk levels in the study area are divided into regions according to the thresholds of five risk levels of forest and grassland fire: high, medium-high, medium, medium-low and low; ArcGIS software was used to visualize the regional division results.

Images