A global fire sensitivity assessment method based on climate factors

Abstract

The present invention discloses a global fire sensitivity assessment method based on climate factors, which relates to the fields of climate science and disaster management, and solves the technical problem that the existing fire sensitivity assessment method does not consider the nonlinear relationship and synergistic effect between climate variables that affect the assessment effect; the present invention includes S1, constructing a dynamic time lag effect model; S2, constructing a spatiotemporal interaction effect model; S3, constructing a correlation coefficient between the climate change intensity measurement and the fire reaction intensity measurement; S4, considering the complex impact of climate change, combining the dynamic time lag effect, the spatiotemporal interaction effect and the correlation coefficient, and using a mathematical modeling method to construct a spatiotemporal sensitivity index (STSI); S5, using the STSI to cluster the sensitivity of fire to climate change into 5 levels; the present invention can integrate multiple climate variables, including temperature, precipitation, wind speed and humidity, and consider the spatiotemporal interaction between these factors, so as to more accurately identify and classify fire sensitivity.

Description

Technical Field

[0001] The present invention relates to the fields of climate science and disaster management, and in particular to a global fire sensitivity assessment method based on climate factors. Background Art

[0002] Global climate change has become one of the major challenges facing modern society, profoundly impacting the frequency and intensity of natural disasters, particularly fires. As Earth's surface temperature rises, extreme weather events, including droughts and heatwaves, become more frequent, significantly increasing the risk of forest fires. Fires not only cause enormous economic losses but also have long-term negative impacts on ecosystems and human life. Therefore, accurately understanding and predicting the susceptibility of different regions to fire is crucial for developing effective fire prevention strategies and disaster reduction plans. Traditional fire sensitivity analysis methods often rely on basic climate data and relatively simple statistical models to assess fire risk. These methods typically analyze climate variables such as temperature, humidity, and wind speed as independent factors, ignoring their nonlinear relationships and synergistic effects. For example, the combination of high temperature and low humidity can significantly increase fire risk, while analyzing these variables individually cannot accurately reflect their combined impact. Furthermore, traditional methods often rely on static data or fixed time periods, failing to fully account for the dynamic characteristics of climate variables over time, such as seasonal variations and the increasing frequency of extreme climate events. Such static models fail to reflect the dynamic trends of climate change, limiting their effectiveness and accuracy in the context of global warming.

[0003] For example, the Chinese patent "A Forest Fire Prediction Method Based on Deep Learning and Multi-Source Remote Sensing Data" (Patent Application No.: CN202210458913.2, Publication No.: CN114998719A) uses data samples of influencing factors to train a ResNet residual neural network, and then uses the trained model to predict forest fires.

[0004] However, this patented solution only uses specific values within a certain time period as samples for model training, and does not fully consider the dynamic characteristics of climate variables over time. While it can achieve the expected prediction results for forest fires under climate conditions similar to the influencing factor data samples, it can fail to achieve the expected prediction results when faced with completely different climate conditions. Summary of the Invention

[0005] To address the aforementioned challenges in the existing technologies, the present invention provides a global fire susceptibility assessment method based on climate factors. This method integrates multiple climate variables, including temperature, precipitation, wind speed, and humidity, and considers the spatiotemporal interactions between these factors. Furthermore, given the potential time delays in the effects of climate change, a dynamic time-lag model is employed to capture these delayed effects. This comprehensive and systematic analysis allows for more precise identification and classification of fire susceptibility, providing a more scientific basis for policymaking and resource allocation.

[0006] A global fire sensitivity assessment method based on climate factors, including:

[0007] S1. Construct a dynamic time-lag effect model to capture the delayed impact of climate change parameters on fires;

[0008] S2. Construct a spatiotemporal interaction effect model to examine the variability of fire impacts across ecological zones and time periods;

[0009] S3. Construct correlation coefficients between climate change intensity measures and fire response intensity measures to quantify the association between climate change and fire;

[0010] S4. Considering the complex impacts of climate change, combining dynamic time lag effects, spatiotemporal interaction effects and correlation coefficients, a mathematical modeling method is used to construct the spatiotemporal sensitivity index (STSI);

[0011] S5. Use the STSI to cluster fire sensitivity to climate change into five levels.

[0012] Furthermore, the impact of climate change on fire is not only immediate, but also has a time lag effect. To capture this dynamic time lag effect, the dynamic time lag effect model described in S1 is expressed as follows:

[0013]

[0014] is the fire index at time t and ecological zone s, Lag time Climate index at the moment, and is the weight, is the error term. is the climate index at time t and ecoregion s, Lag time Fire index at the moment,

[0015] Furthermore, considering the complex relationship between climate change and fire activity: the spatiotemporal interaction effect metric considers the nonlinear interaction between climate factors and fire. The spatiotemporal interaction effect model described in S2 is expressed as follows:

[0016]

[0017] represents the spatiotemporal interaction effect model, is the climate index at time t and ecoregion s, is the fire index at time t and ecological zone s, is the spatial weight of different ecological zones S, which is determined by geographically weighted regression (GWR) to reflect the impact of climate on fire in different regions. and is the adjustment coefficient, which measures the impact of temperature changes on the fire burning area. Determined by linear regression, Determined by nonlinear fitting (logarithmic regression), The nonlinear response index was determined by analyzing the data distribution of climate index and burned area and their interaction, and using nonlinear regression model (polynomial regression) to determine the optimal nonlinear response index.

[0018] Furthermore, the S3 includes:

[0019] S31. Construct a climate change intensity metric;

[0020] S32. Construct fire reaction intensity metric;

[0021] S33. Construct the correlation coefficient between the climate change intensity measure and the fire response intensity measure.

[0022] Furthermore, the intensity measurement of climate change reflects the changes in climate factors at different time points and spatial points. The intensity of climate change is measured by calculating the change in climate factors and normalizing them to eliminate the influence of seasonal fluctuations. The formula for calculating the intensity measurement of climate change in S31 is as follows:

[0023]

[0024] represents the climate change intensity measure, and Respectively represent the i-th and value, is the climate index at time t and ecoregion s, Lag time Climate index at the moment, and represents the mean and standard deviation, is the adjustment coefficient, determined by linear regression analysis.

[0025] Furthermore, the fire reaction intensity metric is used to describe the change in the burned area (BA) over time. The fire reaction intensity is obtained by calculating the annual change in the burned area and further normalizing it. The formula for calculating the fire reaction intensity metric in S32 is as follows:

[0026]

[0027] represents the fire reaction intensity measure, and Respectively represent the jth and value, is the fire index at time t and ecological zone s, Lag time Fire index at the moment, and represents the mean and standard deviation, is the adjustment coefficient, determined by linear regression analysis.

[0028] Furthermore, in order to reveal the connection between climate change and fire intensity, the correlation between the climate change intensity measure and the fire reaction intensity measure is calculated. The formula for calculating the correlation coefficient between the climate change intensity measure and the fire reaction intensity measure in S33 is as follows:

[0029]

[0030] represents the correlation coefficient, and Respectively represent the i-th and The value of represents the fire reaction intensity measure, represents the climate change intensity measure, and Respectively and Mean.

[0031] Furthermore, to fully reflect the complex relationship between climate change and fire, a comprehensive model was designed that considers the correlation coefficient between climate change intensity and fire response intensity, spatiotemporal interaction effects, and time lag effects. This model considers both the immediate and delayed impacts of climate factors and fire burn area, providing a multidimensional perspective to analyze the dynamic relationship between them. The spatiotemporal sensitivity index (STSI) formula is as follows:

[0032]

[0033] Temporal and spatial sensitivity index, for A measure of the intensity of climate change at any given moment, for The fire reaction intensity measure at the moment, represents the attenuation of the time lag effect, is the decay rate, determined by the least squares method, and is the adjustment coefficient, represents the intensity measure of climate change, Represents a measure of fire reaction intensity, determined by linear regression analysis.

[0034] Furthermore, STSI is used in S5 to cluster the sensitivity of fire to climate change into five levels, including L1 low sensitivity, L2 lower sensitivity, L3 medium sensitivity, L4 higher sensitivity, and L5 high sensitivity.

[0035] The beneficial effects of the present invention include:

[0036] First, by integrating dynamic time lag effects, spatiotemporal interaction effects, and the correlation coefficient between climate change and fire, the complex influences of climate variables on fire sensitivity can be more accurately captured and analyzed. Second, by constructing a spatiotemporal sensitivity index (STSI) and categorizing fire sensitivity in various regions around the world into five levels, this provides governments and relevant agencies with a scientific decision-making support tool, helping them to more effectively allocate resources and manage risks, thereby more effectively preventing and responding to fire disasters. Furthermore, the application of this invention can help improve public safety, reduce loss of life and property caused by fire, and promote the implementation of environmental protection and sustainable development strategies. In summary, this invention is not only technologically innovative but also has extensive social, economic, and environmental benefits in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flowchart of a global fire sensitivity assessment method based on climate factors involved in an embodiment of the present application.

[0038] Figure 2 This is a statistical diagram of the spatiotemporal sensitivity index involved in the embodiments of the present application.

[0039] Figure 3 Figure 1 shows the sensitivity ranking statistics of fires to climate change in the embodiments of this application. (a) shows the sensitivity ranking statistics based on the spatiotemporal clustering model constructed by the present invention, and (b) shows the sensitivity ranking statistics obtained using the traditional K-means clustering model. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0041] Example 1

[0042] Global ecoregions were selected as the units of spatial analysis. These ecoregions were based on the work of Dinerstein et al. (2017) and are used to describe biogeographic trends in extreme wildfire events. Based on MODIS data, the average burned area (BA) was calculated using ecoregions as the spatial unit. This paper also used the ERA5 dataset as the primary data source. ERA5 provides high-resolution meteorological data, including temperature, humidity, wind speed, and precipitation. Using this data, this paper further calculated meteorological indices such as the Fire Weather Index (FWI), Standardized Precipitation Index (SPI), Standardized Surface Temperature Index (SLTI), and wind speed (W). Data preprocessing included resampling, outlier removal, and image registration.

[0043] A global fire sensitivity assessment method based on climate factors, such as Figure 1 Shown, including:

[0044] S1. Construct a dynamic time-lag effect model to capture the delayed impact of climate change parameters on fires;

[0045] S2. Construct a spatiotemporal interaction effect model to examine the variability of fire impacts across ecological zones and time periods;

[0046] S3. Construct correlation coefficients between climate change intensity measures and fire response intensity measures to quantify the association between climate change and fire;

[0047] S4. Considering the complex impacts of climate change, combining dynamic time lag effects, spatiotemporal interaction effects and correlation coefficients, a mathematical modeling method is used to construct the spatiotemporal sensitivity index (STSI);

[0048] S5. Use the STSI to cluster fire sensitivity to climate change into five levels.

[0049] In another embodiment, the impact of climate change on fire is not only immediate, but also has a time lag effect. Figure 2 The spatiotemporal sensitivity index statistics of four climate factors, namely the Fire Weather Index (FWI), the Standardized Precipitation Index (SPI), the Standardized Surface Temperature Index (SLTI), and the Wind Speed (W), are shown. To capture this dynamic time lag effect, the dynamic time lag effect model described in S1 is expressed as follows:

[0050]

[0051] is the fire index at time t and ecological zone s, and the average value of BA is calculated according to different ecological zones. Lag time Climate index at the moment, and is the weight, is the error term, is the climate index at time t and ecological zone s, and is obtained by calculating the average value of the meteorological index according to different ecological zones. Specifically, C is a different climate factor, and C=FWI, C=SPI, C=SLTI, C=W are taken in turn. Lag time Fire index at the moment,

[0052] In another embodiment, considering the complex relationship between climate change and fire activity: the spatiotemporal interaction effect metric considers the nonlinear interaction between climate factors and fire. The spatiotemporal interaction effect model in S2 is expressed as follows:

[0053]

[0054] represents the spatiotemporal interaction effect model, is the climate index at time t and ecoregion s, is the fire index at time t and ecological zone s, is the spatial weight of different ecological zones S, which is determined by geographically weighted regression (GWR) to reflect the impact of climate on fire in different regions. and is the adjustment coefficient, which measures the impact of temperature changes on the fire burning area. Determined by linear regression, Determined by nonlinear fitting (logarithmic regression), The nonlinear response index was determined by analyzing the data distribution of climate index and burned area and their interaction, and using nonlinear regression model (polynomial regression) to determine the optimal nonlinear response index.

[0055] In another embodiment, the S3 includes:

[0056] S31. Construct a climate change intensity metric;

[0057] S32. Construct fire reaction intensity metric;

[0058] S33. Construct the correlation coefficient between the climate change intensity measure and the fire response intensity measure.

[0059] In another embodiment, the intensity measurement of climate change reflects the changes in climate factors at different time points and spatial points. The intensity of climate change is measured by calculating the change in climate factors and normalizing the change to eliminate the influence of seasonal fluctuations. The formula for calculating the intensity measurement of climate change in S31 is as follows:

[0060]

[0061] represents the climate change intensity measure, and Respectively represent the i-th and value, is the climate index at time t and ecoregion s, Lag time Climate index at the moment, and represents the mean and standard deviation, is the adjustment coefficient, which is determined by linear regression analysis. Where C represents a different climate factor (one of the climate factors FWI, SPI, SLTI, and W). C = FWI, C = SPI, C = SLTI, and C = W are calculated in a loop. Specifically, when C = FWI, C i It represents the value of the i-th FWI; when C=W, C i It represents the value of the i-th W.

[0062] In another embodiment, the fire reaction intensity metric is used to describe the change in the burned area (BA) over time. The fire reaction intensity is obtained by calculating the annual change in the burned area and further normalizing it. The formula for calculating the fire reaction intensity metric in S32 is as follows:

[0063]

[0064] represents the fire reaction intensity measure, and Respectively represent the jth and value, is the fire index at time t and ecological zone s, Lag time Fire index at the moment, and represents the mean and standard deviation, is the adjustment coefficient, determined by linear regression analysis.

[0065] In another embodiment, in order to reveal the connection between climate change and fire intensity, the correlation between the climate change intensity measure and the fire reaction intensity measure is calculated. The formula for calculating the correlation coefficient between the climate change intensity measure and the fire reaction intensity measure in S33 is as follows:

[0066]

[0067] represents the correlation coefficient, and Respectively represent the i-th and The value of represents the fire reaction intensity measure, represents the intensity measure of climate change, and Respectively and Mean.

[0068] In another embodiment, to fully reflect the complex relationship between climate change and fire, a comprehensive model was designed that considers the correlation coefficient between climate change intensity and fire response intensity, spatiotemporal interaction effects, and time lag effects. This model considers both the immediate and delayed impacts of climate factors and fire burn area, providing a multi-dimensional perspective to analyze the dynamic relationship between them. The spatiotemporal sensitivity index (STSI) formula is as follows:

[0069]

[0070] Temporal and spatial sensitivity index, for A measure of the intensity of climate change at any given moment, for The fire reaction intensity measure at the moment, represents the attenuation of the time lag effect, is the decay rate, determined by the least squares method, and is the adjustment coefficient, represents the intensity measure of climate change, Represents a measure of fire reaction intensity, determined by linear regression analysis.

[0071] In another embodiment, STSI is used in S5 to cluster fire sensitivity to climate change into five levels: L1: low sensitivity, L2: relatively low sensitivity, L3: medium sensitivity, L4: relatively high sensitivity, and L5: high sensitivity. Data-driven adaptive classification does not require setting a fixed threshold. The parameters that need to be set are as follows:

[0072] n_slusters=5·The number of clusters corresponds to 5 risk levels;

[0073] Max_itec=300 · Maximum number of iterations to run;

[0074] tol=1e-4·convergence threshold (distance of centroid moving between two iterations)

[0075] N_init = 10·Number of random initializations of the algorithm (to obtain the best result).

[0076] The final classification statistical effect is as follows Figure 3 shown.

[0077] Level 1 (Very Low Sensitivity): Fires in these areas are less sensitive to climate change. Even if the climate changes (such as rising temperatures), the frequency of fires in these areas will change little, and the fire risk is relatively low.

[0078] Level 2 (Low Sensitivity): Fires in these areas are less sensitive to climate change, but climate change may affect fire frequency to some extent compared to Level 1 areas, but the changes will be smaller.

[0079] Level 3 (Moderate Sensitivity): Fires in these areas are moderately sensitive to climate change. Climate change (such as rising temperatures and reduced precipitation) is likely to significantly alter the fire risk in these areas, with fire frequency either increasing or changing.

[0080] Level 4 (High Sensitivity): Fires in these areas are highly sensitive to climate change. As climate change intensifies, the frequency of fires will increase significantly, and the fire risk is high.

[0081] Level 5 (Very High Sensitivity): These areas are extremely sensitive to climate change. Even small changes in climate (such as rising temperatures or reduced precipitation) could lead to significant increases in the frequency and intensity of fires. These areas present a very high fire risk and require special attention.

[0082] In the present invention, step 5 specifically involves using the spatiotemporal clustering model developed by the present invention and the conventional K-means clustering model to classify the fire sensitivity in different ecological zones into five levels (L1 low sensitivity, L2 relatively low sensitivity, L3 moderate sensitivity, L4 relatively high sensitivity, L5 high sensitivity). In order to evaluate the performance of these two clustering methods, the present invention uses three accuracy evaluation indicators, namely the silhouette coefficient, the Calinski-Harabasz index, and the Davies-Bouldin index, to conduct an in-depth comparative analysis. The silhouette coefficient is used to evaluate the compactness and separation of clusters, with higher values indicating that the clustering results have better cohesion and discrimination. The Calinski-Harabasz index evaluates the effectiveness of clustering by measuring the ratio of intra-cluster cohesion to inter-cluster separation, with higher index values reflecting better clustering quality. The Davies-Bouldin index evaluates the separability of clusters by measuring the average similarity of clusters, with lower values indicating better cluster separation.

[0083] The evaluation results show that the method of the present invention is superior to the traditional K-means clustering in terms of silhouette coefficient, Calinski-Harabasz index and Davies-Bouldin index. Specifically, the average silhouette coefficient of the method of the present invention is 0.65, showing a high degree of cohesion and good separation; while the average silhouette coefficient of K-means clustering is 0.45, indicating that its cluster separation is relatively low. In terms of the Calinski-Harabasz index, the present invention achieved a high score of 250, which is significantly better than the 150 points of K-means clustering, which further demonstrates the significant advantages of the clustering method of the present invention in distinguishing between clusters and within clusters. In addition, the Davies-Bouldin index provides the present invention with a score of 1.2, which is significantly lower than the 2.1 of K-means clustering, thereby confirming the superiority of the present invention in clustering quality.

[0084] Experimental results show that the method of the present invention can effectively classify and analyze fire-sensitive areas. Specifically:

[0085] The Calinski-Harabasz index and Davies-Bouldin index show that our method can clearly distinguish fire-sensitive areas, making the risk differences between different areas more obvious.

[0086] The silhouette coefficient indicates that the classification is good, the fire sensitivity within the area is similar, and it is also very different from other areas.

[0087] The combination of these indicators demonstrates that our method is not only applicable over a wide range, but can also efficiently and accurately identify fire-sensitive areas, thus providing accurate support for relevant decision-making. In particular, our method can maintain high classification accuracy in complex areas.

[0088] These experimental results validate the ability of the method of the present invention to effectively classify and accurately analyze fire sensitivity data on a global scale, demonstrating its significant improvement in scientificity and accuracy compared to traditional methods.

[0089] The above-described embodiments merely represent specific implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of protection of the present application. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the technical concept of the present application, and all such variations and improvements fall within the scope of protection of the present application.

Claims

1. A global fire sensitivity assessment method based on climate factors, characterized in that: The following steps are involved: S1. Construct a dynamic time-lag effect model to capture the delayed impact of climate change parameters on fires; S2. Construct a spatiotemporal interaction effect model to examine the variability of fire impacts across ecological zones and time periods; S3. Construct correlation coefficients between climate change intensity measures and fire response intensity measures to quantify the association between climate change and fire; S4. Considering the complex impacts of climate change, combining dynamic time lag effects, spatiotemporal interaction effects and correlation coefficients, a mathematical modeling method is used to construct the spatiotemporal sensitivity index (STSI); S5. Clustering fire sensitivity to climate change into five levels using the STSI; The dynamic time lag effect model described in S1 is expressed as follows: ； is the fire index at time t and ecological zone s, Lag time Climate index at the moment, and is the weight, is the error term, is the climate index at time t and ecoregion s, Lag time Fire index at the moment; The spatiotemporal interaction effect model described in S2 is expressed as follows: ； represents the spatiotemporal interaction effect model, is the climate index at time t and ecoregion s, is the fire index at time t and ecological zone s, is the spatial weight of different ecological zones S, and is the adjustment coefficient, It is a nonlinear response index, determined by analyzing the data distribution of climate index and burned area and their interaction using a nonlinear regression model; The formula for calculating the correlation coefficient between the climate change intensity measure and the fire response intensity measure is as follows: ； represents the correlation coefficient, and Respectively represent the i-th and The value of represents the fire reaction intensity measure, represents the intensity measure of climate change, and Respectively and Mean.

2. A global fire sensitivity assessment method based on climate factors according to claim 1, characterized in that: The S3 includes: S31. Constructing a climate change intensity metric ; S32. Constructing fire reaction intensity metrics ; S33. Constructing the correlation coefficient between climate change intensity measure and fire response intensity measure .

3. A global fire sensitivity assessment method based on climate factors according to claim 2, characterized in that: The formula for calculating the climate change intensity metric in S31 is as follows: ； represents the intensity measure of climate change, and Respectively represent the i-th and value, is the climate index at time t and ecoregion s, Lag time Climate index at the moment, and represents the mean and standard deviation, is the adjustment coefficient, determined by linear regression analysis.

4. A global fire sensitivity assessment method based on climate factors according to claim 2, characterized in that: The fire reaction intensity measurement formula of S32 is as follows: ； represents the fire reaction intensity measure, and Respectively represent the jth and value, is the fire index at time t and ecological zone s, Lag time Fire index at the moment, and represents the mean and standard deviation, is the adjustment coefficient, determined by linear regression analysis.

5. The global fire sensitivity assessment method based on climate factors according to claim 2, characterized in that: The formula for constructing the spatiotemporal sensitivity index STSI is as follows: ； Temporal and spatial sensitivity index, for A measure of the intensity of climate change at any given moment, for The fire reaction intensity measure at the moment, represents the attenuation of the time lag effect, is the decay rate, and is the adjustment coefficient, represents the climate change intensity measure, Represents a measure of fire reaction intensity, determined by linear regression analysis.

6. The global fire sensitivity assessment method based on climate factors according to claim 1, characterized in that: The STSI is used in S5 to cluster the sensitivity of fire to climate change into five levels, including L1 low sensitivity, L2 lower sensitivity, L3 medium sensitivity, L4 higher sensitivity, and L5 high sensitivity.

Images