Method and system for predicting catastrophe and stability of side slope in cold and cold mountainous area by considering climatic change

Abstract

The invention relates to the field of natural disaster prevention and control, and discloses an alpine mountain side slope catastrophe and stability prediction method and system considering climate change, and the method comprises the following steps: predicting future rainfall and temperature change by using output data of a plurality of regional climate models and a global climate model in combination with different climate scenes; predicting a future LULC change by using a land change model based on the historical LULC data and the climate change prediction data; an FSLAM model is adopted to calculate landslide susceptibility under different future scenes, and the probability of failure (PoF) of each grid unit is obtained; the influence of climate change and LULC change on the slope stability is evaluated by calculating PoF differences under different scenes; and uncertainty analysis is carried out on the climate model, LULC prediction and model parameters, and the uncertainty range of a prediction result is quantified. The method can provide quantitative evaluation of the landslide risk of the cold and cold mountainous area, and provides reliable data support for formulating disaster prevention and reduction measures.

Description

Technical Field

[0001] The present invention relates to the technical field of natural disaster prevention and control, and particularly to a method and system for predicting slope disasters and stability in alpine regions considering climate change. Background Art

[0002] Currently, with the intensification of global climate change, extreme weather events occur frequently. Especially in alpine regions, the occurrence frequency and severity of landslide disasters are increasing continuously. The landslide disasters in alpine regions not only cause great damage to the ecological environment, but also pose a threat to the safety of human life and property. Therefore, how to accurately predict the landslide risk in alpine regions, timely evaluate the slope stability, and take effective preventive measures has become a major problem to be solved urgently.

[0003] Existing landslide risk assessment methods mainly rely on historical data and geological exploration results. Most traditional prediction models ignore the potential impact of climate change and land use / land cover (LULC) change on landslide occurrence. Although some studies have begun to attempt to incorporate climate change into landslide risk assessment, most methods are still limited to the analysis of a single factor and fail to effectively combine multiple factors such as climate change, land use change, and soil properties, resulting in low accuracy and reliability of the assessment results.

[0004] In addition, existing landslide susceptibility prediction methods often rely on static geological conditions and historical landslide data, lacking the consideration of dynamic climate change and land use change, and it is difficult to adapt to the impact of climate change on the regional environment. Especially in alpine regions, the change of climate conditions has a profound impact on slope stability. Traditional methods fail to fully reflect this complex spatio-temporal change, resulting in large prediction errors in practical applications.

[0005] Moreover, although some models have begun to involve uncertainty analysis, most existing methods are limited to the uncertainty analysis of a single factor and fail to conduct systematic uncertainty quantification analysis on landslide susceptibility under multi-factor and complex scenarios. This makes the landslide risk assessment results lack sufficient reliability and credibility and is difficult to provide strong data support for decision-making.

[0006] Therefore, there are many limitations in the existing technology for landslide risk assessment. There is an urgent need for a landslide susceptibility prediction method that comprehensively considers climate change, LULC change, and uncertainty analysis to improve the accuracy and reliability of the assessment results and provide a more scientific decision-making basis for landslide disaster risk management in alpine regions. Summary of the Invention

[0007] In view of the deficiencies of the prior art, the present invention provides a method for predicting landslide susceptibility in alpine mountainous areas that comprehensively considers climate change and land use / land cover change, thereby improving the accuracy and reliability of landslide risk assessment.

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for predicting slope disasters and stability in alpine mountainous areas considering climate change, comprising the following steps:

[0009] Use the output data of multiple regional climate models and global climate models, and combine different climate scenarios to predict future precipitation and temperature changes;

[0010] Based on historical LULC data and climate change prediction data, use a land change model to predict future LULC changes;

[0011] Use the FSLAM model, based on the climate change prediction results and LULC change prediction data, to calculate the landslide susceptibility under different future scenarios and obtain the failure probability of each grid cell;

[0012] By calculating the differences in failure probabilities under different scenarios, evaluate the impacts of climate change and LULC change on slope stability;

[0013] Conduct uncertainty analysis on climate models, LULC predictions, and model parameters to quantify the uncertainty range of prediction results.

[0014] Preferably, the step of using the output data of multiple regional climate models and global climate models and combining different climate scenarios to predict future precipitation and temperature changes includes:

[0015] Select multiple regional climate models and global climate models, and according to different climate scenario inputs, predict future precipitation and temperature;

[0016] Adopt statistical downscaling techniques to convert the output data of regional climate models into regional-scale climate data to ensure that precipitation and temperature predictions have a high spatial resolution;

[0017] Combine the output results of multiple regional climate models and global climate models to simulate future precipitation and temperature changes under different scenarios;

[0018] Output the predicted precipitation and temperature change data as input data for subsequent LULC change prediction and landslide susceptibility assessment.

[0019] Preferably, the step of using a land change model to predict future LULC changes based on historical LULC data and climate change prediction data includes:

[0020] Obtain historical LULC data and climate change prediction data as inputs, where the LULC data includes land use / land cover types over a certain past period, and the climate change data includes prediction results of future temperature and precipitation changes;

[0021] Use a land change model to simulate the changes in land use / land cover types in different future periods based on the input historical LULC data and climate change prediction data;

[0022] Through LULC conversion probability analysis, calculate the conversion of various land use / land cover types under different future scenarios, generate a future LULC change prediction map, and evaluate the impact of different LULC types on landslide susceptibility and slope stability;

[0023] Combine the generated LULC change results with the climate change prediction results, output the predicted data of land use changes in the future slope area, and use it for subsequent landslide susceptibility assessment.

[0024] Preferably, the step of using the FSLAM model to calculate the landslide susceptibility under different future scenarios based on the climate change prediction results and LULC change prediction data to obtain the failure probability of each grid cell includes:

[0025] Use the FSLAM model to calculate the landslide susceptibility. This model combines landslide triggering factors based on the climate change prediction results and LULC change prediction data to calculate the failure probability of each grid cell;

[0026] Obtain the failure probability of each grid cell under different scenarios through model calculation, reflecting the likelihood of landslide occurrence;

[0027] According to the calculation results of the failure probability under different scenarios, evaluate the impact of different climate changes and LULC changes on landslide susceptibility, and output a landslide susceptibility distribution map.

[0028] Preferably, the calculation formula for the failure probability is:

[0029] PoF = f(ERR, Pe, soilproperties, LULC)

[0030] Where PoF is the probability of landslide occurrence, ERR is the effective antecedent recharge, Pe is the event precipitation, soilproperties is the soil property, and LULC is the land use / land cover type.

[0031] Preferably, during the calculation process of the FSLAM model, the impact of precipitation change and temperature change on the effective antecedent recharge of the soil is further considered, and the calculation formula is:

[0032] ERR = ∫0 t (P t -E t )dt

[0033] Among them, ERR is the effective early-stage supply, P t is the precipitation at time t, and E t is the evapotranspiration at time t. The integral calculates the impact of the difference between precipitation and evapotranspiration on soil moisture.

[0034] Preferably, the steps of evaluating the impacts of climate change and LULC change on slope stability by calculating the difference in failure probabilities under different scenarios include:

[0035] Calculate the landslide failure probabilities under different scenarios, and evaluate the impacts of climate change and LULC change on slope stability based on the difference in failure probabilities for each scenario;

[0036] Compare the failure probability results between the future scenario and the reference scenario, and quantify the stability change by analyzing the difference in failure probabilities;

[0037] Output the stability change results for different scenarios, providing a quantitative analysis of the impacts of future climate change and LULC change on slope stability.

[0038] Preferably, the quantitative analysis of the stability change is further obtained by calculating the difference in failure probabilities for each grid cell and summing them to get the total change amount. The formula is:

[0039]

[0040] where ΔPoF total is the total sum of the differences in failure probabilities for all grid cells, ΔPoF i is the difference in failure probability for the i-th grid cell, and n is the total number of grid cells.

[0041] Preferably, the steps of performing uncertainty analysis on the climate model, LULC prediction, and model parameters, and quantifying the uncertainty range of the prediction results include:

[0042] Perform uncertainty analysis on the climate model, LULC prediction, and model parameters, and randomly sample the input parameters through Monte Carlo simulation or sensitivity analysis methods to simulate the prediction results under different input conditions;

[0043] Quantify the uncertainty propagation of the climate model, LULC prediction, and landslide susceptibility assessment model parameters, and calculate the range of changes in the prediction results for different scenarios;

[0044] Output the confidence interval of the results through uncertainty analysis, evaluate the stability of future landslide susceptibility prediction, and provide an uncertainty assessment report.

[0045] The present invention also provides a prediction system for slope disasters and stability in alpine regions considering climate change, including:

[0046] A climate change prediction module for predicting future precipitation and temperature changes by combining output data from multiple regional climate models and global climate models under different climate scenarios;

[0047] An LULC change prediction module for predicting future LULC changes based on historical LULC data and climate change prediction data;

[0048] A landslide susceptibility assessment module for calculating the landslide susceptibility under different future scenarios using the FSLAM model based on the climate change prediction results and LULC change prediction data, and generating a failure probability map;

[0049] A stability change prediction module for evaluating the change in landslide stability based on the difference in failure probabilities under different scenarios;

[0050] An uncertainty analysis module for performing uncertainty analysis on climate models, LULC predictions, and model parameters to quantify the uncertainty range of prediction results.

[0051] The present invention provides a method and system for predicting slope disasters and stability in alpine regions considering climate change. It has the following beneficial effects:

[0052] 1. By combining the predictions of climate change and land use / land cover (LULC) change, the present invention can comprehensively evaluate the combined effects of both on slope stability and landslide susceptibility. This multi-dimensional analysis method can accurately reflect the comprehensive impact of climate change and human activities on landslide risks in alpine regions, providing a more accurate and reliable risk assessment result than single-factor analysis.

[0053] 2. The present invention uses the FSLAM model to calculate landslide susceptibility. By considering various factors such as climate change, LULC change, and soil properties, it can effectively quantify the probability of failure (PoF) of landslides. The application of this model provides an efficient and accurate method for predicting landslide susceptibility, which can provide a scientific basis for slope stability analysis and disaster risk management.

[0054] 3. By calculating the difference in PoF under different scenarios, the present invention can deeply evaluate the change trend of landslide risks under different climate and LULC change scenarios. This method can provide a quantitative analysis of landslide risks under different environmental change scenarios, helping decision-makers identify potential high-risk areas and formulate corresponding disaster prevention and mitigation measures.

[0055] 4. The present invention conducts uncertainty analysis on climate models, LULC predictions, and model parameters through Monte Carlo simulation and sensitivity analysis methods, quantifying the uncertainty range of prediction results. This analysis method provides a confidence interval and reliability assessment for landslide susceptibility prediction, significantly improving the credibility of prediction results and helping decision-makers make more scientific decisions in the face of uncertain environments.

[0056] 5. By comprehensively considering the impacts of climate change and LULC change on landslide susceptibility, the present invention provides a practical technical means for landslide risk management in alpine regions. This method can not only identify high-risk areas but also provide a prediction basis for future climate change and land use change, helping to plan and implement effective landslide prevention and control measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a schematic flow chart of the method of the present invention;

[0058] Figure 2 is a schematic structural diagram of the system of the present invention.

[0059] Among them, 10 is a climate change prediction module; 20 is an LULC change prediction module; 30 is a landslide susceptibility assessment module; 40 is a stability change prediction module; 50 is an uncertainty analysis module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0061] Please refer to the attached Figure 1 , the present invention provides a method for predicting slope catastrophes and stability in alpine regions considering climate change, aiming to provide a scientific basis for risk assessment and management of landslide disasters in alpine regions by comprehensively analyzing the impacts of climate change and land use / land cover (LULC) change on landslide susceptibility and slope stability.

[0062] As Figure 1 shown, the method for predicting slope catastrophes and stability in alpine regions considering climate change may include the following steps:

[0063] S1. Use the output data of multiple regional climate models and global climate models, and combine different climate scenarios to predict future precipitation and temperature changes;

[0064] S2. Predict the future LULC changes using a land change model based on historical LULC data and climate change prediction data;

[0065] S3. Use the FSLAM model to calculate the landslide susceptibility under different future scenarios based on the climate change prediction results and LULC change prediction data, and obtain the failure probability of each grid cell;

[0066] S4. Evaluate the impacts of climate change and LULC changes on slope stability by calculating the differences in failure probabilities under different scenarios;

[0067] S5. Conduct uncertainty analysis on the climate model, LULC prediction, and model parameters to quantify the uncertainty range of the prediction results.

[0068] The following is a detailed implementation of the technical solution of the present invention, and each step of the method will be described in detail.

[0069] For step S1, in this embodiment, the output data of multiple regional climate models (RCMs) and global climate models (GCMs) are used to predict the future precipitation and temperature changes in combination with different climate scenarios. The core of this step is to obtain a comprehensive prediction result of future climate change through multi-model fusion and scenario simulation.

[0070] First of all, in the present invention, multiple regional climate models and global climate models are selected. These models can provide climate data at different time scales and spatial scales. Regional climate models, such as RegCM (Regional Climate Model) and REMO, are mainly used for high-resolution prediction of climate within a specific region and are suitable for alpine mountainous areas that are extremely sensitive to climate change. The global climate models (such as GISS and MPI, etc.) used in combination with them provide a wider range of climate change scenarios. These models can predict global climate change based on different greenhouse gas emission scenarios (such as RCP4.5 and RCP8.5).

[0071] The selection of climate models is not only based on the need for spatial resolution but also requires choosing models with high credibility and simulation effects according to the climate characteristics of the region. For example, regional climate models are particularly effective in alpine mountainous areas because the climate characteristics in these areas are usually different from those under the global model, and the smaller spatial scale requires higher-precision simulation.

[0072] Secondly, in combination with different climate scenarios (such as RCP4.5 and RCP8.5), this embodiment adopts a climate scenario analysis method, which obtains possible paths of climate change by analyzing future greenhouse gas emission scenarios. RCP4.5 represents a medium greenhouse gas emission scenario, while RCP8.5 represents a high emission scenario, reflecting a relatively severe scenario of greenhouse gas concentration. Under each scenario, the climate model will calculate the precipitation and temperature change data at a certain future time point (such as 2030, 2050, 2100) respectively.

[0073] In the process of processing climate data, this embodiment adopts a downscaling technique, which is mainly used to convert the low-resolution data of the global climate model into regional climate data with higher spatial resolution. Through downscaling, the simulation of climate change can more finely reflect the climate change characteristics in specific regions, especially in complex terrains such as alpine mountainous areas. For example, the precipitation data provided by the climate model may have a resolution of only 50 km at the spatial scale, while through the downscaling technique, the resolution can be increased to 1 km or even higher, making the prediction results of climate change more applicable to regions with complex terrain and sensitive to climate change.

[0074] Specifically, in this embodiment, the core variables of climate change - precipitation and temperature - are predicted under different future scenarios respectively. The change in precipitation mainly affects the change in soil moisture, while the change in temperature directly affects the freezing and thawing process of the soil, which has an important impact on the landslide susceptibility in alpine mountainous areas. Through the output data of the climate model, the change trends of precipitation and temperature at different future time points can be obtained, and these data will be used as input data in the subsequent steps, affecting the prediction of land use change (LULC) and ultimately affecting the landslide susceptibility assessment.

[0075] Through the above methods, by using multi-model fusion, scenario simulation and downscaling techniques, high-precision prediction results of future climate change in alpine mountainous areas are provided. These prediction data will be used in the subsequent steps to analyze the LULC change and landslide susceptibility, thus providing important data support for the management and prevention of landslide risks in alpine mountainous areas.

[0076] During the implementation process, for the selection of different climate scenarios, in addition to RCP4.5 and RCP8.5, other climate scenarios such as RCP2.6 (low emission scenario) can also be added according to specific needs to obtain more prediction results on climate change trends. By comparing the climate changes under different scenarios, the potential impact of climate change on landslide disasters in the region can be more comprehensively evaluated, providing more abundant information for the subsequent risk assessment and the formulation of response measures.

[0077] For step S2, in this embodiment, based on historical LULC data and climate change prediction data, a land change model (LCM) is used to predict future LULC (land use / land cover) changes. The core task of this step is to simulate land use / land cover types in different future periods according to climate change and other driving factors, so as to provide reliable data support for subsequent landslide susceptibility assessment and slope stability analysis.

[0078] First, before predicting LULC changes, the present invention collects historical LULC data from multiple data sources. These data usually come from remote sensing images, ground surveys, and existing land use databases (such as the land use dataset of USGS). Exemplary historical LULC data includes the distribution of different land types such as forests, grasslands, agricultural land, and urban land. By analyzing these historical data, the land use change trend of this area can be obtained, thus providing a basis for future LULC change prediction.

[0079] Secondly, in order to accurately predict future LULC changes, this embodiment adopts a land change model (LCM). LCM is a prediction tool based on spatial analysis and dynamic simulation, which can consider various driving factors such as climate change, land demand, economic development, and social factors to dynamically simulate LULC. By inputting historical LULC data and future climate change scenario data, LCM can simulate land use changes under different scenarios.

[0080] Specifically, in the model, first, the main driving factors of land use change are analyzed. These driving factors usually include climate change (temperature and precipitation changes), population growth, policy changes, and land economic value. For the impact of climate change, combined with the climate prediction data provided in step S1, LCM estimates the potential impact of climate change on land use by simulating the changes in future climate conditions (such as temperature, precipitation, etc.). For example, rising temperatures may lead to the degradation of forest areas, while decreasing precipitation may lead to the expansion or reduction of agricultural land.

[0081] Then, a multi-objective land allocation algorithm and a neural network algorithm are used to predict LULC changes. These algorithms can generate the future LULC distribution at each time point (such as 2030, 2050, 2100) according to the current land use pattern and future driving factors. The neural network algorithm optimizes the prediction results by training historical data, so that the simulation of LULC changes can be as close as possible to the actual change trend. For each simulation time point, the model will output the spatial distribution of different land use types (such as forests, grasslands, agricultural land, urban land, etc.).

[0082] The specific LULC change model formula is:

[0083] LULC t+1 = f(LULC t , ΔClimate, ΔPopulation, ΔEconomic)

[0084] where LULC t+1 is the land use type at the predicted future time (t + 1), LULC t is the land use type at the current time (t), ΔClimate represents the impact of climate change (temperature, precipitation, etc.), and ΔPopulation and ΔEconomic respectively represent the impacts of population and economic factors on land use. This formula indicates that by comprehensively considering various factors such as climate, population, and economy, the land use / land cover status at a future time can be obtained.

[0085] To further improve the reliability of the model, in this embodiment, uncertainty analysis is added during the LULC prediction process. Due to the complexity of climate change and socio-economic factors, the future changes in LULC have a certain degree of uncertainty. Therefore, in the model, multiple scenarios (such as low, medium, and high emission scenarios) are simulated to quantify these uncertainties. The simulation results will generate multiple possible LULC change maps to show the possible change paths of land use under different scenarios.

[0086] Through this method, step S2 can not only simulate the potential impact of climate change on LULC, but also fully consider socio-economic factors, thereby providing a comprehensive prediction of future LULC changes. These prediction results will provide data support for subsequent landslide susceptibility assessment and slope stability analysis, and help formulate more scientific and effective risk management strategies.

[0087] For step S3, in this embodiment, the FSLAM (Fast Shallow Landslide Assessment Model) is used to calculate the landslide susceptibility under different future scenarios and obtain the probability of failure (PoF) for each grid cell. The purpose of this step is to systematically evaluate the future landslide susceptibility by combining the results of climate change and land use / land cover (LULC) change predictions, and thereby provide the necessary data support for slope stability analysis.

[0088] First, when conducting the landslide susceptibility assessment, the input data mainly includes: climate change prediction results, LULC change prediction results, and information such as the topography and soil properties of the area. The climate change prediction results are mainly reflected in the changes in precipitation and temperature, and the LULC change prediction results involve the changes in land use types. These data jointly affect the probability of landslide occurrence. By comprehensively analyzing these factors, the probability of failure of the landslide can be calculated for each grid cell.

[0089] Specifically, in this embodiment, we calculate landslide susceptibility based on the FSLAM model. The FSLAM model combines climate change predictions (such as precipitation and temperature changes), land use change predictions (such as forest degradation, agricultural expansion, etc.), and the effects of topography and soil properties, and uses a set of parameters to quantify the landslide susceptibility of each grid cell.

[0090] In the model, the calculation formula for landslide susceptibility is as follows:

[0091] PoF = f(ERR, Pe, soilproperties, LULC)

[0092] Where PoF represents the probability of failure (i.e., the probability of landslide occurrence) of each grid cell, ERR is the effective antecedent recharge, Pe is the event precipitation, soilproperties are the soil properties (such as soil permeability, density, etc.), and LULC is the land use / land cover type.

[0093] Furthermore, ERR (effective antecedent recharge) represents the effective amount of water stored in the soil due to factors such as precipitation and evapotranspiration. Pe (event precipitation) refers to the change in precipitation over a certain period, usually referring to the impact of extreme precipitation events on landslide susceptibility. Soil properties include characteristics such as soil type, thickness, and saturated moisture, which affect soil stability and the probability of landslide occurrence. The land use / land cover type (LULC) is related to factors such as vegetation cover and land use status, and different land use types affect soil stability and landslide susceptibility.

[0094] During the calculation process, the FSLAM model combines these input data for each grid cell and calculates the landslide failure probability of each grid cell through numerical methods. In this way, the FSLAM model can reflect the landslide susceptibility under different scenarios (including climate change and LULC change), providing data support for subsequent stability analysis.

[0095] For example, assume that in a certain area, due to climate change, the precipitation increases, and due to land use change, the forest area decreases and agricultural land expands. The FSLAM model calculates the combination of these factors to obtain the PoF value of each grid cell in this area. If the PoF value of a certain grid cell is high, it indicates that the risk of landslide occurrence in this area in the future is relatively large. Conversely, areas with lower PoF values are relatively stable.

[0096] Meanwhile, in this embodiment, multiple different climate scenarios (such as RCP4.5 and RCP8.5) and LULC change scenarios are considered. By comparing and analyzing the PoF values under each scenario, the specific impacts of climate change and LULC change on landslide susceptibility can be evaluated. These analysis results help identify high-risk areas for landslides and provide a quantitative basis for slope stability assessment.

[0097] In this embodiment, the FSLAM model can not only calculate the PoF of each grid cell, but also output the changing trends of landslide susceptibility under different scenarios, thus providing more detailed information for the assessment of landslide disaster risks. In addition, through multiple simulations and scenario comparisons, the differences in landslide risks under different climate change and LULC change scenarios can be further evaluated, providing support for decision-makers to formulate more accurate risk management strategies.

[0098] In this embodiment, during the calculation process of the FSLAM model, the impacts of precipitation change and temperature change on effective rainfall recharge (ERR) are further considered. Effective rainfall recharge (ERR) is one of the important factors affecting landslide susceptibility, and its calculation takes into account the difference between precipitation and evapotranspiration.

[0099] Specifically, in this model, the calculation formula for ERR is:

[0100] ERR = ∫ 0 t (P t - E t )dt

[0101] where ERR is the effective rainfall recharge, P t is the precipitation at time t, and E t is the evapotranspiration at time t. This formula calculates the impact of the difference between precipitation and evapotranspiration on soil moisture through integration, and then affects the soil moisture and stability. The magnitude of effective rainfall recharge directly affects the soil strength and slope stability, thus having an important impact on the probability of landslide occurrence (PoF).

[0102] During the implementation process, by considering the dynamic changes of precipitation and evapotranspiration, this calculation model can reflect the changing trends of soil moisture under different climate scenarios. An increase in precipitation usually increases soil moisture, while an increase in evapotranspiration reduces the moisture in the soil. Especially in alpine mountainous areas, an increase in temperature may lead to an increase in evapotranspiration, thus affecting the effectiveness of soil moisture and then changing slope stability.

[0103] For example, assume that in a certain region, the precipitation is relatively large, but due to the increase in temperature, the evapotranspiration also increases. Eventually, the soil moisture fails to increase effectively, and instead, the soil wetness may decrease. This change may lead to a decline in soil stability and increase the risk of landslides. By calculating the difference between precipitation and evapotranspiration, the impact of this change on soil moisture can be accurately simulated, and its specific contribution to landslide susceptibility can be quantified.

[0104] For step S4, in this embodiment, by calculating the difference in the probability of failure (PoF) under different scenarios, the impacts of climate change and LULC (land use / land cover) change on slope stability are evaluated. The main task of this step is to compare the PoF results between future scenarios (including climate change scenarios and LULC change scenarios) and the reference scenario, and by analyzing the PoF difference, the change in stability is quantified. This process provides a key quantitative analysis basis for the management and response to landslide risks.

[0105] First, in this step, calculating the difference in PoF under different scenarios is mainly achieved by comparing the landslide probability of failure (PoF) values obtained under different climate and LULC scenarios with the PoF value in the reference scenario (usually the scenario under the current or historical state). The PoF difference represents the change in landslide susceptibility under different scenarios and can reflect the specific impacts of climate change and land use change on landslide risks.

[0106] The calculation formula for the PoF difference is as follows:

[0107] ΔPoF = PoF future - PoF reference

[0108] where ΔPoF represents the difference in the landslide probability of failure, PoF future is the landslide probability of failure under future climate change and LULC change scenarios, and PoF reference is the landslide probability of failure in the reference scenario.

[0109] Secondly, in this embodiment, the PoF differences under different scenarios are calculated through the above formula, and then these difference values are statistically analyzed to further evaluate the specific impacts of climate change and LULC change on slope stability. For example, if the PoF in some regions under future scenarios is significantly higher than that in the reference scenario, it can be inferred that the landslide risks in these regions are relatively large, and this change is mainly caused by climate change or LULC change. On the contrary, if the PoF difference is small or the change is not obvious, it indicates that the landslide risks in these regions change little under different scenarios, and the slope stability is relatively stable.

[0110] In this way, this step can not only quantify the impacts of climate change and LULC change on slope stability, but also provide a quantitative basis for slope stability analysis, helping to identify areas with higher future landslide risks. This is of great significance for formulating slope stability guarantee measures to cope with climate change and land use change.

[0111] For example, assume that in a high-cold mountainous area, due to rising temperatures, the precipitation in some areas increases, resulting in increased soil moisture, which in turn increases the landslide risk in this area. At the same time, due to land use change, forest areas are converted into agricultural land, further exacerbating soil erosion and instability. In this case, the FSLAM model will calculate the PoF values of these areas under different scenarios. By comparing the PoF differences, it can clearly point out the adverse impacts of climate change and LULC change on the slope stability of this area.

[0112] Furthermore, through the PoF difference analysis, the changing trends of landslide susceptibility under different climate scenarios and LULC change scenarios can be obtained, and risk assessments can be carried out based on these changing trends. For example, if the PoF difference is large in a certain scenario, it indicates that climate change and LULC change have had a significant impact on the landslide risk in this area, and targeted disaster prevention measures may need to be taken; while if the PoF difference is small, it may indicate that the landslide risk in this area is relatively stable.

[0113] In this embodiment, during the process of calculating the PoF difference and conducting stability analysis, multiple different climate scenarios (such as RCP4.5, RCP8.5, etc.) and LULC change scenarios are considered. Through the comprehensive analysis of these scenarios, more comprehensive prediction results can be provided for slope stability assessment. These results not only help to understand the impacts of climate change and LULC change on landslide susceptibility, but also provide strong support for the management and control of future landslide risks.

[0114] In this embodiment, in step S4, the quantitative analysis of stability change is further obtained by calculating the difference in failure probability of each grid cell and summing to obtain the total change amount. The purpose of this step is to aggregate the differences in landslide failure probability (PoF) of each grid cell and quantify the overall impact of climate change and LULC change on slope stability.

[0115] Specifically, by calculating the PoF difference of each grid cell, the change in landslide susceptibility of each grid cell under different scenarios is obtained. Then, the sum of the failure probability differences of all grid cells is obtained through the following formula:

[0116]

[0117] where, ΔPoF totalis the sum of the differences in failure probabilities for all grid cells, ΔPoF i is the difference in failure probability for the i-th grid cell, and n is the total number of grid cells.

[0118] The function of this formula is to accumulate the PoF differences of each grid cell to obtain the change in landslide risk in the entire area. The PoF difference of each grid cell reflects how the likelihood of landslides changes in this area under the influence of climate change and LULC change. Summing up the differences of all grid cells can obtain the total change in landslide risk in this area under different scenarios.

[0119] For example, assume a study area in a high-cold mountainous region where the PoF differences in some areas are large (i.e., the change in landslide risk is large), while those in other areas are small. By summing up the PoF differences of all grid cells, the change in the overall landslide risk in this area can be quantified, providing a quantitative basis for risk management.

[0120] Furthermore, by calculating the total PoF difference, this embodiment can evaluate the impact of climate change and LULC change on the landslide susceptibility of the entire area. For example, if ΔPoF total is large, it indicates a significant increase in landslide risk caused by climate and land use changes, and more urgent disaster prevention measures may be required; if the total difference is small, it may indicate that the change in landslide risk in this area is relatively stable, and the risk management strategy can be adjusted appropriately.

[0121] For step S5, in this embodiment, uncertainty analysis is performed on the climate model, LULC prediction, and model parameters to quantify the uncertainty range of the prediction results. The goal of this step is to identify and analyze the possible uncertainty factors in the prediction process of climate change and land use / land cover (LULC) change, and evaluate the impact of these factors on the landslide susceptibility prediction results through uncertainty analysis, so as to provide more robust decision-making support for risk management.

[0122] First of all, uncertainty analysis quantifies the impact of these changes on the landslide susceptibility prediction results by simulating different possible values of climate change scenarios, LULC change scenarios, and other input parameters. In this embodiment, mainly Monte Carlo simulation or sensitivity analysis methods are used. Monte Carlo simulation is a commonly used uncertainty analysis method that obtains the probability distribution of output results by randomly sampling the possible values of different input variables and performing multiple simulations. Sensitivity analysis is used to evaluate the relative impact of each input variable on the prediction results, so as to identify the most important influencing factors.

[0123] Specifically, in Monte Carlo simulation, for each input variable (such as the output data of climate models, LULC prediction data, and other model parameters), random sampling will be carried out according to its possible value range. The random values of these variables will be input into the model for multiple simulations. In this way, the distribution of landslide susceptibility prediction results under different input conditions can be obtained.

[0124] The results of uncertainty analysis usually include a probability distribution or confidence interval, which can represent the range of variation of the prediction results. For example, for the prediction of the probability of failure (PoF) of a landslide, the output result may be an interval range rather than a single definite value. This confidence interval can be used to represent the reliability and uncertainty of the landslide susceptibility prediction results. For example, if the confidence interval of PoF is wide, it indicates that the prediction result of the model has a large uncertainty and further analysis or adjustment of model parameters may be required; while if the confidence interval of PoF is narrow, it indicates that the prediction result has a high certainty.

[0125] Secondly, in this embodiment, not only the uncertainty analysis of the predictions of climate change scenarios and LULC change scenarios is carried out, but also the uncertainty of model parameters is considered. These parameters include soil properties, topographic features, precipitation patterns, etc. in the FSLAM model. There may also be certain uncertainties in the values of these model parameters. Therefore, when conducting simulations, these parameters need to be appropriately varied and their impacts on the final prediction results need to be analyzed.

[0126] For example, assume that the soil permeability parameter involved in the FSLAM model has a certain uncertainty. Then, through sensitivity analysis, the impact of this parameter on the landslide failure probability prediction result can be evaluated. If the change in soil permeability has a large impact on the difference in PoF, then the accuracy of this parameter needs to be focused on and optimized in further research.

[0127] In this embodiment, through uncertainty analysis, the reliability range of the landslide susceptibility prediction results can be identified under different uncertainty conditions of climate change, LULC change, and model parameters. Through this method, more comprehensive and robust prediction results can be provided for slope stability assessment, and thus reliable data support can be provided for the management of landslide disaster risks.

[0128] In the implementation process, the specific process of uncertainty analysis may include the following steps: First, select the relevant input parameters of climate models, LULC predictions, and the FSLAM model; then, generate different values of the input parameters through Monte Carlo simulation or sensitivity analysis methods; next, run multiple simulations and analyze the distribution of the output results; finally, quantify the uncertainty of the results and evaluate the prediction results through indicators such as confidence intervals and standard deviations.

[0129] For example, in a certain area, due to the impact of climate change, the predicted results of precipitation and temperature have relatively large uncertainties. Through uncertainty analysis, the impacts of these uncertainties on landslide susceptibility prediction can be evaluated, thus providing more valuable data support for landslide risk assessment in this area.

[0130] Generally speaking, the present invention provides a scientific basis for risk assessment and management of landslide disasters in alpine regions by comprehensively analyzing the impacts of climate change and land use / land cover (LULC) change on landslide susceptibility and slope stability. The method includes using regional climate models and global climate models to predict future climate change, combining with the LULC change model to predict land use change, using the FSLAM model to calculate landslide susceptibility, and quantifying the impacts of climate change and LULC change on landslide risk through uncertainty analysis. By calculating the differences in the probability of failure (PoF) of landslides under different scenarios, the specific impacts of climate change and land use change on slope stability can be evaluated, and reliable data support can be provided for early warning and risk management of landslide disasters.

[0131] The alpine slope disaster and stability prediction system considering climate change described below can be correspondingly referred to the alpine slope disaster and stability prediction method considering climate change described above.

[0132] Please refer to the append Figure 2 , the present invention also provides an alpine slope disaster and stability prediction system considering climate change, including multiple modules. Each module cooperates with each other through data flow to finally realize the prediction and assessment of slope stability and landslide risk. The following are the specific implementation manners of each module:

[0133] Climate change prediction module 10

[0134] The main function of the climate change prediction module 10 is to predict the future changes in precipitation and temperature according to the output data of multiple regional climate models (RCMs) and global climate models (GCMs), combined with different climate scenarios (such as RCP4.5, RCP8.5, etc.). This module works through the following steps:

[0135] Climate model selection: Select regional climate models (such as RegCM, REMO) applicable to alpine regions and global climate models (such as GISS, MPI). These models provide high-spatial-resolution climate data and can consider different greenhouse gas emission paths of global climate change scenarios.

[0136] Climate scenario input: Based on different greenhouse gas emission scenarios (such as RCP4.5, RCP8.5), calculate the precipitation and temperature changes in the coming decades. The selection of climate scenarios can be adjusted according to the prediction requirements of the specific study area.

[0137] Downscaling technology: Use downscaling technology to convert the low-resolution GCM output data into high-resolution climate data that can be provided by the regional climate model, ensuring that the data accuracy adapts to the complex geographical environment of the alpine mountainous area.

[0138] Output: This module outputs the precipitation and temperature change data under different future scenarios, and these data will be used as input data for LULC change prediction and landslide susceptibility assessment.

[0139] LULC Change Prediction Module 20

[0140] LULC Change Prediction Module 20 predicts the future changes in land use / land cover (LULC) based on historical LULC data and climate change prediction data. The specific implementation steps are as follows:

[0141] Data input: This module first obtains historical LULC data, usually from remote sensing images or ground survey data. Then, the climate change prediction results (such as precipitation, temperature, etc.) are used as important factors affecting LULC changes.

[0142] Land change model: Use the land change model (LCM), which combines historical data and future climate change scenarios to simulate the land use / land cover changes in different future periods. LCM dynamically simulates land use changes through neural network algorithms and multi-objective land allocation algorithms.

[0143] Model output: This module predicts the LULC change results at multiple future time nodes (such as 2030, 2050, 2100), including the distribution of different land types such as forests, grasslands, and agricultural land.

[0144] Output: This module outputs the LULC change data under different future scenarios, and these data will be used as input for the landslide susceptibility assessment module.

[0145] Landslide Susceptibility Assessment Module 30

[0146] Landslide Susceptibility Assessment Module 30 calculates the landslide susceptibility under different future scenarios using the FSLAM model based on the climate change prediction results and LULC change prediction data. The specific steps include:

[0147] Data input: This module inputs climate change prediction data (precipitation, temperature, etc.) and LULC change prediction data (land use type changes), and combines data such as terrain and soil properties.

[0148] FSLAM model calculation: The FSLAM (Fast Shallow Landslide Assessment Model) is used for landslide susceptibility assessment. The model calculates the probability of failure (PoF) of each grid cell based on the input data.

[0149] Generation of probability of failure map: Based on the calculated PoF values, a landslide probability of failure map of the area is generated to visually display the landslide susceptibility of each grid cell.

[0150] Output: This module outputs the probability of failure maps under different future scenarios, providing data support for subsequent prediction of stability changes.

[0151] Stability change prediction module 40

[0152] The stability change prediction module 40 evaluates the landslide stability changes based on the differences in the probability of failure under different scenarios. The specific steps are as follows:

[0153] Calculation of PoF difference: This module first calculates the PoF differences under different scenarios, that is, compares the PoF values between the future scenario and the reference scenario to obtain the probability of failure difference for each grid cell.

[0154] Calculation of total change: By calculating and summing the PoF differences of all grid cells, the total amount of landslide stability change in the entire area is obtained.

[0155] Assessment of stability change: Through the total change amount, the overall impacts of climate change and LULC change on slope stability are evaluated.

[0156] Output: This module outputs the quantitative analysis results of stability changes, providing a decision-making basis for the risk management of landslide disasters.

[0157] Uncertainty analysis module 50

[0158] The uncertainty analysis module 50 is used to conduct uncertainty analysis on the parameters in the climate model, LULC prediction, and FSLAM model, and quantify the uncertainty range of the prediction results. The specific implementation is as follows:

[0159] Monte Carlo simulation: This module uses the Monte Carlo simulation method to conduct multiple random samplings on the input data (such as precipitation and temperature changes in the climate model, LULC prediction data, soil properties of the FSLAM model, etc.), so as to generate different combinations of input data.

[0160] Simulation calculation: These randomly generated input data are input into the model for multiple simulations, and the changes in the landslide susceptibility prediction results under different scenarios are calculated.

[0161] Confidence Interval Assessment: Through statistical simulation results, calculate the confidence interval of the landslide susceptibility prediction results, thereby quantifying the uncertainty range and providing a reliable risk assessment for decision-making.

[0162] Output: This module outputs the confidence interval of the landslide susceptibility prediction results, helping users evaluate the uncertainty of the prediction results and enhancing the reliability of decision-making.

[0163] In this embodiment, the slope disaster and stability prediction system in alpine regions considering climate change provides comprehensive technical support for landslide risk assessment in alpine regions through the close combination of modules such as climate change prediction, LULC change prediction, landslide susceptibility assessment, stability change prediction, and uncertainty analysis. This system can dynamically reflect the impact of climate change and land use change on slope stability, quantify landslide risk, and provide a scientific basis for the prevention and risk management of landslide disasters.

[0164] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting slope disaster and stability in high-cold mountainous areas considering climate change, characterized in that: The following steps are involved: Use output from multiple regional climate models and global climate models to predict future changes in precipitation and temperature in combination with different climate scenarios; Based on historical LULC data and climate change prediction data, land change models are used to predict future LULC changes; Using the FSLAM model, based on climate change prediction results and LULC change prediction data, the landslide susceptibility under different future scenarios is calculated to obtain the failure probability of each grid cell; The impact of climate change and LULC changes on slope stability was evaluated by calculating the differences in failure probabilities under different scenarios; Uncertainty analysis of climate models, LULC predictions and model parameters was performed to quantify the uncertainty range of the prediction results.

2. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 1 is characterized in that: The step of using the output data of multiple regional climate models and global climate models and combining different climate scenarios to predict future precipitation and temperature changes includes: Select multiple regional climate models and global climate models and make predictions of future precipitation and temperature based on different climate scenario inputs; Statistical downscaling techniques are used to convert regional climate model output data into regional-scale climate data to ensure that precipitation and temperature forecasts have high spatial resolution; Combine outputs from multiple regional climate models and a global climate model to simulate future changes in precipitation and temperature under different scenarios; The predicted precipitation and temperature change data are output as input data for subsequent LULC change prediction and landslide susceptibility assessment.

3. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 1 is characterized in that: The steps of using the land change model to predict future LULC changes based on historical LULC data and climate change prediction data include: Obtain historical LULC data and climate change prediction data as input, where LULC data includes land use / land cover types within a certain period of time in the past, and climate change data includes the prediction results of future temperature and precipitation changes; Use land change models to simulate land use / land cover type changes in different periods in the future based on historical LULC data and climate change prediction data; Through LULC conversion probability analysis, the conversion of various land use / land cover types under different future scenarios is calculated, the future LULC change prediction map is generated, and the impact of different LULC types on landslide susceptibility and slope stability is evaluated; The generated LULC change results are combined with the climate change prediction results to output the predicted data of land use change in the future slope area and used for subsequent landslide susceptibility assessment.

4. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 1 is characterized in that: The steps of using the FSLAM model to calculate the landslide susceptibility under different future scenarios based on climate change prediction results and LULC change prediction data to obtain the failure probability of each grid unit include: The FSLAM model is used to calculate landslide susceptibility. The model is based on climate change prediction results and LULC change prediction data, combined with landslide triggering factors, to calculate the failure probability of each grid cell. The failure probability of each grid unit under different scenarios is calculated through the model, reflecting the possibility of landslide occurrence; According to the calculation results of failure probability under different scenarios, the impact of different climate changes and LULC changes on landslide susceptibility is evaluated, and the landslide susceptibility distribution map is output.

5. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 4 is characterized in that: The calculation formula of the failure probability is: PoF=f(ERR,Pe,soilproperties,LULC) Among them, PoF is the probability of landslide occurrence, ERR is the effective antecedent recharge, Pe is the event precipitation, soilproperties is the soil properties, and LULC is the land use / land cover type.

6. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 4 is characterized in that: In the calculation process of the FSLAM model, the influence of precipitation change and temperature change on effective initial soil recharge is further considered, and the calculation formula is: ERR=∫0 t (P t -E t )dt Among them, ERR is the effective previous supply, P t is the precipitation at time t, E t is the evapotranspiration at time t, and the integral calculates the effect of the difference between precipitation and evapotranspiration on soil moisture.

7. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 1 is characterized in that: The steps of evaluating the impact of climate change and LULC change on slope stability by calculating the difference in failure probability under different scenarios include: Calculate the probability of landslide failure under different scenarios and evaluate the impact of climate change and LULC changes on slope stability based on the difference in failure probability under each scenario; Compare the failure probability results under the future scenario with the reference scenario, and quantify the stability changes by analyzing the difference in failure probability; Output the stability change results under different scenarios and provide quantitative analysis of the impact of future climate change and LULC changes on slope stability.

8. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 7 is characterized in that: The quantitative analysis of the stability change is further performed by calculating the failure probability difference of each grid unit and summing them up to obtain the total change, the formula is: Among them, ΔPoF total is the sum of the failure probability differences of all grid cells, ΔPoF i is the failure probability difference of the ith grid unit, and n is the total number of grid units.

9. The method for predicting slope disaster and stability in high-cold mountainous areas considering climate change according to claim 1 is characterized in that: The steps of performing uncertainty analysis on the climate model, LULC prediction and model parameters to quantify the uncertainty range of the prediction results include: Conduct uncertainty analysis on climate models, LULC predictions and model parameters, randomly sample input parameters through Monte Carlo simulation or sensitivity analysis methods, and simulate prediction results under different input conditions; Quantify the uncertainty propagation of climate models, LULC predictions, and landslide susceptibility assessment model parameters, and calculate the range of prediction results under different scenarios; The confidence interval of the uncertainty analysis output results is used to evaluate the stability of future landslide susceptibility predictions and provide an uncertainty assessment report.

10. A system for predicting slope disasters and stability in high-cold mountainous areas taking into account climate change, used to execute the method for predicting slope disasters and stability in high-cold mountainous areas taking into account climate change as claimed in any one of claims 1 to 9, characterized in that: include: The climate change prediction module is used to predict future precipitation and temperature changes based on the output data of multiple regional climate models and global climate models and combine different climate scenarios; LULC change prediction module, used to predict future LULC changes based on historical LULC data and climate change prediction data; The landslide susceptibility assessment module is used to calculate the landslide susceptibility under different future scenarios based on the climate change prediction results and LULC change prediction data using the FSLAM model and generate failure probability maps; The stability change prediction module is used to evaluate the landslide stability change according to the difference in failure probability under different scenarios; The uncertainty analysis module is used to perform uncertainty analysis on climate models, LULC predictions and model parameters, and to quantify the uncertainty range of the prediction results.

Images