Forest fires are natural disasters that can occur suddenly and can be very damaging,burning thousands of square kilometers.Prevention is better than suppression and prediction models of forest fire occurrence have dev...Forest fires are natural disasters that can occur suddenly and can be very damaging,burning thousands of square kilometers.Prevention is better than suppression and prediction models of forest fire occurrence have developed from the logistic regression model,the geographical weighted logistic regression model,the Lasso regression model,the random forest model,and the support vector machine model based on historical forest fire data from 2000 to 2019 in Jilin Province.The models,along with a distribution map are presented in this paper to provide a theoretical basis for forest fire management in this area.Existing studies show that the prediction accuracies of the two machine learning models are higher than those of the three generalized linear regression models.The accuracies of the random forest model,the support vector machine model,geographical weighted logistic regression model,the Lasso regression model,and logistic model were 88.7%,87.7%,86.0%,85.0%and 84.6%,respectively.Weather is the main factor affecting forest fires,while the impacts of topography factors,human and social-economic factors on fire occurrence were similar.展开更多
Parameterization is a critical step in modelling ecosystem dynamics.However,assigning parameter values can be a technical challenge for structurally complex natural plant communities;uncertainties in model simulations...Parameterization is a critical step in modelling ecosystem dynamics.However,assigning parameter values can be a technical challenge for structurally complex natural plant communities;uncertainties in model simulations often arise from inappropriate model parameterization.Here we compared five methods for defining community-level specific leaf area(SLA)and leaf C:N across nine contrasting forest sites along the North-South Transect of Eastern China,including biomass-weighted average for the entire plant community(AP_BW)and four simplified selective sampling(biomass-weighted average over five dominant tree species[5DT_BW],basal area weighted average over five dominant tree species[5DT_AW],biomass-weighted average over all tree species[AT_BW]and basal area weighted average over all tree species[AT_AW]).We found that the default values for SLA and leaf C:N embedded in the Biome-BGC v4.2 were higher than the five computational methods produced across the nine sites,with deviations ranging from 28.0 to 73.3%.In addition,there were only slight deviations(<10%)between the whole plant community sampling(AP_BW)predicted NPP and the four simplified selective sampling methods,and no significant difference between the predictions of AT_BW and AP_BW except the Shennongjia site.The findings in this study highlights the critical importance of computational strategies for community-level parameterization in ecosystem process modelling,and will support the choice of parameterization methods.展开更多
Historical forest fire risk databases are vital for evaluating the effectiveness of past forest management approaches,enhancing forest fire warnings and emergency response capabilities,and accurately budgeting potenti...Historical forest fire risk databases are vital for evaluating the effectiveness of past forest management approaches,enhancing forest fire warnings and emergency response capabilities,and accurately budgeting potential carbon emissions resulting from fires.However,due to the unavailability of spatial information technology,such databases are extremely difficult to build reliably and completely in the non-satellite era.This study presented an improved forest fire risk reconstruction framework that integrates a deep learning-based time series prediction model and spatial interpolation to address the challenge in Sichuan Province,southwestern China.First,the forest fire danger index(FFDI)was improved by supplementing slope and aspect information.We compared the performances of three time series models,namely,the autoregressive integrated moving average(ARIMA),Prophet and long short-term memory(LSTM)in predicting the modified forest fire danger index(MFFDI).The bestperforming model was used to retrace the MFFDI of individual stations from 1941 to 1970.Following this,the Anusplin spatial interpolation method was used to map the distributions of the MFFDI at five-year intervals,which were then subjected to weighted overlay with the distance-to-river layer to generate forest fire risk maps for reconstructing the forest fire danger database.The results revealed LSTM as the most accurate in fitting and predicting the historical MFFDI,with a fitting determination coefficient(R^2)of 0.709,mean square error(MSE)of0.047,and validation R^2 and MSE of 0.508 and 0.11,respectively.Independent validation of the predicted forest fire risk maps indicated that 5 out of 7 historical forest fire events were located in forest fire-prone areas,which is higher than the results determined from the original FFDI(2 out of 7).This proves the effectiveness of the improved MFFDI and indicates a high level of reliability of the historical forest fire risk reconstruction method proposed in this study.展开更多
Tropical peat comprises decomposed dead plant material and acts like a sponge to absorb water,making it fully saturated.However,drought periods dry it readily and increases its vulnerability to fire.Peat fires emit gr...Tropical peat comprises decomposed dead plant material and acts like a sponge to absorb water,making it fully saturated.However,drought periods dry it readily and increases its vulnerability to fire.Peat fires emit greenhouse gases and particles contributing to haze,and prevention by constructing fire-break canals to reduce fire spread into forest reserves is crucial.This paper aims to determine peat physical and chemical properties near a fire-break canal at different fire frequency areas.Peat sampling was conducted at two forest reserves in Malaysia which represent low fire frequency and high fire frequency areas.The results show that peat properties were not affected by the construction of a fire-break canal,however lignin and cellulose content increased significantly from the distance of the canal in both areas.The study concluded that fire frequency did not significantly influence peat properties except for porosity.The higher fibre content in the high frequency area did not influence moisture content nor the ability to regain moisture.Thus,fire frequency might contribute differently to changes in physical and chemical properties,hence management efforts to construct fire-break canals and restoration efforts should protect peatlands from further degradation.These findings will benefit future management and planning for forest reserves.展开更多
This article delves into the analysis of performance and utilization of Support Vector Machines (SVMs) for the critical task of forest fire detection using image datasets. With the increasing threat of forest fires to...This article delves into the analysis of performance and utilization of Support Vector Machines (SVMs) for the critical task of forest fire detection using image datasets. With the increasing threat of forest fires to ecosystems and human settlements, the need for rapid and accurate detection systems is of utmost importance. SVMs, renowned for their strong classification capabilities, exhibit proficiency in recognizing patterns associated with fire within images. By training on labeled data, SVMs acquire the ability to identify distinctive attributes associated with fire, such as flames, smoke, or alterations in the visual characteristics of the forest area. The document thoroughly examines the use of SVMs, covering crucial elements like data preprocessing, feature extraction, and model training. It rigorously evaluates parameters such as accuracy, efficiency, and practical applicability. The knowledge gained from this study aids in the development of efficient forest fire detection systems, enabling prompt responses and improving disaster management. Moreover, the correlation between SVM accuracy and the difficulties presented by high-dimensional datasets is carefully investigated, demonstrated through a revealing case study. The relationship between accuracy scores and the different resolutions used for resizing the training datasets has also been discussed in this article. These comprehensive studies result in a definitive overview of the difficulties faced and the potential sectors requiring further improvement and focus.展开更多
As massive underground projects have become popular in dense urban cities,a problem has arisen:which model predicts the best for Tunnel Boring Machine(TBM)performance in these tunneling projects?However,performance le...As massive underground projects have become popular in dense urban cities,a problem has arisen:which model predicts the best for Tunnel Boring Machine(TBM)performance in these tunneling projects?However,performance level of TBMs in complex geological conditions is still a great challenge for practitioners and researchers.On the other hand,a reliable and accurate prediction of TBM performance is essential to planning an applicable tunnel construction schedule.The performance of TBM is very difficult to estimate due to various geotechnical and geological factors and machine specifications.The previously-proposed intelligent techniques in this field are mostly based on a single or base model with a low level of accuracy.Hence,this study aims to introduce a hybrid randomforest(RF)technique optimized by global harmony search with generalized oppositionbased learning(GOGHS)for forecasting TBM advance rate(AR).Optimizing the RF hyper-parameters in terms of,e.g.,tree number and maximum tree depth is the main objective of using the GOGHS-RF model.In the modelling of this study,a comprehensive databasewith themost influential parameters onTBMtogetherwithTBM AR were used as input and output variables,respectively.To examine the capability and power of the GOGHSRF model,three more hybrid models of particle swarm optimization-RF,genetic algorithm-RF and artificial bee colony-RF were also constructed to forecast TBM AR.Evaluation of the developed models was performed by calculating several performance indices,including determination coefficient(R2),root-mean-square-error(RMSE),and mean-absolute-percentage-error(MAPE).The results showed that theGOGHS-RF is a more accurate technique for estimatingTBMAR compared to the other applied models.The newly-developedGOGHS-RFmodel enjoyed R2=0.9937 and 0.9844,respectively,for train and test stages,which are higher than a pre-developed RF.Also,the importance of the input parameters was interpreted through the SHapley Additive exPlanations(SHAP)method,and it was found that thrust force per cutter is the most important variable on TBMAR.The GOGHS-RF model can be used in mechanized tunnel projects for predicting and checking performance.展开更多
This paper aims to analyze and evaluate a model of forest conservation and management of ethnic minority (or indigenous) people in central Vietnam, often referred to as the spiritual forest. These forests, called sacr...This paper aims to analyze and evaluate a model of forest conservation and management of ethnic minority (or indigenous) people in central Vietnam, often referred to as the spiritual forest. These forests, called sacred forest or ghost forests by the ethnic minority people in Thua Thien Hue province, have existed for a long time among forest residents. However, they have recently declined, both in quality and quantity, due to various factors, including changes in society, economy, environment, and perception, among other reasons. Based on research conducted in A Luoi district, Thua Thien Hue province with household interviews, group discussions, and field surveys, we find that spiritual forest retains religious and human significance. They are also often among the last remaining natural forests left due to deforestation by human activities. The research results indicate challenges that conservation of spiritual forest may face, while giving recommendations derived from communities for sustainable forest development and conservation in the region.展开更多
The paper summarizes the structure and water-absorbing mechanism,classification,and preparation method of polymer fire extinguishing gel,and prospects for its application in aerial firefighting,forest ground fire exti...The paper summarizes the structure and water-absorbing mechanism,classification,and preparation method of polymer fire extinguishing gel,and prospects for its application in aerial firefighting,forest ground fire extinguishing,opening of firebreaks,and mitigating human casualties in forest fire extinguishing.展开更多
The mechanism of lightning that ignites a forest fire and the lightning that occurs above a forest fire are explained at the molecular level. It is based on two phenomena, namely, internal charge separation inside the...The mechanism of lightning that ignites a forest fire and the lightning that occurs above a forest fire are explained at the molecular level. It is based on two phenomena, namely, internal charge separation inside the atmospheric cloud particles and the existence of a layer of positively charged hydrogen atoms sticking out of the surface of the liquid layer of water on the surface of rimers. Strong turbulence-driven collisions of the ice particles and water droplets with the rimers give rise to breakups of the ice particles and water droplets into positively and negatively charged fragments leading to charge separation. Hot weather in a forest contributes to the updraft of hot and humid air, which follows the same physical/chemical processes of normal lightning proposed and explained recently[1]. Lightning would have a high probability of lighting up and burning the dry biological materials in the ground of the forest, leading to a forest (wild) fire. The burning of trees and other plants would release a lot of heat and moisture together with a lot of smoke particles (aerosols) becoming a strong updraft. The condition for creating lightning is again satisfied which would result in further lightning high above the forest wild fire.展开更多
The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was p...The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was proposed to reduce casting defects and improve production efficiency,which includes the random forest(RF)classification model,the feature importance analysis,and the process parameters optimization with Monte Carlo simulation.The collected data includes four types of defects and corresponding process parameters were used to construct the RF model.Classification results show a recall rate above 90% for all categories.The Gini Index was used to assess the importance of the process parameters in the formation of various defects in the RF model.Finally,the classification model was applied to different production conditions for quality prediction.In the case of process parameters optimization for gas porosity defects,this model serves as an experimental process in the Monte Carlo method to estimate a better temperature distribution.The prediction model,when applied to the factory,greatly improved the efficiency of defect detection.Results show that the scrap rate decreased from 10.16% to 6.68%.展开更多
Earth resource and environmental monitoring are essential areas that can be used to investigate the environmental conditions and natural resources supporting sustainable policy development,regulatory measures,and thei...Earth resource and environmental monitoring are essential areas that can be used to investigate the environmental conditions and natural resources supporting sustainable policy development,regulatory measures,and their implementation elevating the environment.Large-scale forest fire is considered a major harmful hazard that affects climate change and life over the globe.Therefore,the early identification of forest fires using automated tools is essential to avoid the spread of fire to a large extent.Therefore,this paper focuses on the design of automated forest fire detection using a fusion-based deep learning(AFFD-FDL)model for environmental monitoring.The AFFDFDL technique involves the design of an entropy-based fusion model for feature extraction.The combination of the handcrafted features using histogram of gradients(HOG)with deep features using SqueezeNet and Inception v3 models.Besides,an optimal extreme learning machine(ELM)based classifier is used to identify the existence of fire or not.In order to properly tune the parameters of the ELM model,the oppositional glowworm swarm optimization(OGSO)algorithm is employed and thereby improves the forest fire detection performance.A wide range of simulation analyses takes place on a benchmark dataset and the results are inspected under several aspects.The experimental results highlighted the betterment of the AFFD-FDL technique over the recent state of art techniques.展开更多
Forest fires are a significant threat to the environment, causing ecological damage, economic losses, and posing a threat to human life. Hence, timely detection and prevention of forest fires are critical to minimizin...Forest fires are a significant threat to the environment, causing ecological damage, economic losses, and posing a threat to human life. Hence, timely detection and prevention of forest fires are critical to minimizing their impact. In this paper, we review the current state-of-the-art methods in forest fire detection and prevention using predictions based on weather conditions and predictions based on forest fire history. In particular, we discuss different Machine Learning (ML) models that have been used for forest fire detection. Further, we present the challenges faced when implementing the ML-based forest fire detection and prevention systems, such as data availability, model prediction errors and processing speed. Finally, we discuss how recent advances in Deep Learning (DL) can be utilized to improve the performance of current fire detection systems.展开更多
Forest fire is one of the main natural hazards because of its fierce destructiveness. Various researches on fire real time monitoring, behavior simulation and loss assessment have been carried out in many countries. A...Forest fire is one of the main natural hazards because of its fierce destructiveness. Various researches on fire real time monitoring, behavior simulation and loss assessment have been carried out in many countries. As fire prevention is probably the most efficient means for protecting forests, suitable methods should be developed for estimating the fire danger. Fire danger is composed of ecological, human and climatic factors. Therefore, the systematic analysis of the factors including forest characteristics, meteorological status, topographic condition causing forest fire is made in this paper at first. The relationships between biophysical factors and fire danger are paid more attention to. Then the parameters derived from remote sensing data are used to estimate the fire danger variables, According to the analysis, not only PVI (Perpendicular Vegetation Index) can classify different vegetation but also crown density is captured with PVI. Vegetation moisture content has high correlation with the ratio of actual evapotranspiration (LE) to potential ecapotranspiration (LEp). SI (Structural Index), which is the combination of TM band 4 and 5 data, is a good indicator of forest age. Finally, a fire danger prediction model, in which relative importance of each fire factor is taken into account, is built based on GIS.展开更多
Climate change has an impact on forest fire patterns.In the context of global warming,it is important to study the possible effects of climate change on forest fires,carbon emission reductions,carbon sink effects,fore...Climate change has an impact on forest fire patterns.In the context of global warming,it is important to study the possible effects of climate change on forest fires,carbon emission reductions,carbon sink effects,forest fire management,and sustainable development of forest ecosystems.This study is based on MODIS active fire data from 2001-2020 and the influence of climate,topography,vegetation,and social factors were integrated.Temperature and precipitation information from different scenarios of the BCC-CSM2-MR climate model were used as future climate data.Under climate change scenarios of a sustainable low development path and a high conventional development path,the extreme gradient boosting model predicted the spatial distribution of forest fire occurrence in China in the 2030s(2021-2040),2050s(2041-2060),2070s(2061-2080),and2090s(2081-2100).Probability maps were generated and tested using ROC curves.The results show that:(1)the area under the ROC curve of training data(70%)and validation data(30%)were 0.8465 and 0.8171,respectively,indicating that the model can reasonably predict the occurrence of forest fire in the study area;(2)temperature,elevation,and precipitation were strongly correlated with fire occurrence,while land type,slope,distance from settlements and roads,and slope direction were less strongly correlated;and,(3)based on future climate change scenarios,the probability of forest fire occurrence will tend to shift from the south to the center of the country.Compared with the current climate(2001-2020),the occurrence of forest fires in 2021-2040,2041-2060,2061-2080,and 2081-2100 will increase significantly in Henan Province(Luoyang,Nanyang,S anmenxia),Shaanxi Province(Shangluo,Ankang),Sichuan Province(Mianyang,Guangyuan,Ganzi),Tibet Autonomous Region(Shannan,Linzhi,Changdu),Liaoning Province(Liaoyang,Fushun,Dandong).展开更多
Ongoing climate changes have a direct impact on forest growth;they also affect natural fire regimes,with further implications for forest composition.Understanding of how these will affect forests on decadal-to-centenn...Ongoing climate changes have a direct impact on forest growth;they also affect natural fire regimes,with further implications for forest composition.Understanding of how these will affect forests on decadal-to-centennial timescales is limited.Here we use reconstructions of past vegetation,fire regimes and climate during the Holocene to examine the relative importance of changes in climate and fire regimes for the abundance of key tree species in northeastern China.We reconstructed vegetation changes and fire regimes based on pollen and charcoal records from Gushantun peatland.We then used generalized linear modelling to investigate the impact of reconstructed changes in summer temperature,annual precipitation,background levels of fire,fire frequency and fire magnitude to identify the drivers of decadal-to-centennial changes in forest openness and composition.Changes in climate and fire regimes have independent impacts on the abundance of the key tree taxa.Climate variables are generally more important than fire variables in determining the abundance of individual taxa.Precipitation is the only determinant of forest openness,but summer temperature is more important than precipitation for individual tree taxa with warmer summers causing a decrease in cold-tolerant conifers and an increase in warmth-demanding broadleaved trees.Both background level and fire frequency have negative relationships with the abundance of most tree taxa;only Pinus increases as fire frequency increases.The magnitude of individual fires does not have a significant impact on species abundance on this timescale.Both climate and fire regime characteristics must be considered to understand changes in forest composition on the decadal-to-centennial timescale.There are differences,both in sign and magnitude,in the response of individual tree species to individual drivers.展开更多
Objective Body fluid mixtures are complex biological samples that frequently occur in crime scenes,and can provide important clues for criminal case analysis.DNA methylation assay has been applied in the identificatio...Objective Body fluid mixtures are complex biological samples that frequently occur in crime scenes,and can provide important clues for criminal case analysis.DNA methylation assay has been applied in the identification of human body fluids,and has exhibited excellent performance in predicting single-source body fluids.The present study aims to develop a methylation SNaPshot multiplex system for body fluid identification,and accurately predict the mixture samples.In addition,the value of DNA methylation in the prediction of body fluid mixtures was further explored.Methods In the present study,420 samples of body fluid mixtures and 250 samples of single body fluids were tested using an optimized multiplex methylation system.Each kind of body fluid sample presented the specific methylation profiles of the 10 markers.Results Significant differences in methylation levels were observed between the mixtures and single body fluids.For all kinds of mixtures,the Spearman’s correlation analysis revealed a significantly strong correlation between the methylation levels and component proportions(1:20,1:10,1:5,1:1,5:1,10:1 and 20:1).Two random forest classification models were trained for the prediction of mixture types and the prediction of the mixture proportion of 2 components,based on the methylation levels of 10 markers.For the mixture prediction,Model-1 presented outstanding prediction accuracy,which reached up to 99.3%in 427 training samples,and had a remarkable accuracy of 100%in 243 independent test samples.For the mixture proportion prediction,Model-2 demonstrated an excellent accuracy of 98.8%in 252 training samples,and 98.2%in 168 independent test samples.The total prediction accuracy reached 99.3%for body fluid mixtures and 98.6%for the mixture proportions.Conclusion These results indicate the excellent capability and powerful value of the multiplex methylation system in the identification of forensic body fluid mixtures.展开更多
Understanding the mechanisms and risks of forest fires by building a spatial prediction model is an important means of controlling forest fires.Non-fire point data are important training data for constructing a model,...Understanding the mechanisms and risks of forest fires by building a spatial prediction model is an important means of controlling forest fires.Non-fire point data are important training data for constructing a model,and their quality significantly impacts the prediction performance of the model.However,non-fire point data obtained using existing sampling methods generally suffer from low representativeness.Therefore,this study proposes a non-fire point data sampling method based on geographical similarity to improve the quality of non-fire point samples.The method is based on the idea that the less similar the geographical environment between a sample point and an already occurred fire point,the greater the confidence in being a non-fire point sample.Yunnan Province,China,with a high frequency of forest fires,was used as the study area.We compared the prediction performance of traditional sampling methods and the proposed method using three commonly used forest fire risk prediction models:logistic regression(LR),support vector machine(SVM),and random forest(RF).The results show that the modeling and prediction accuracies of the forest fire prediction models established based on the proposed sampling method are significantly improved compared with those of the traditional sampling method.Specifically,in 2010,the modeling and prediction accuracies improved by 19.1%and 32.8%,respectively,and in 2020,they improved by 13.1%and 24.3%,respectively.Therefore,we believe that collecting non-fire point samples based on the principle of geographical similarity is an effective way to improve the quality of forest fire samples,and thus enhance the prediction of forest fire risk.展开更多
To prevent, detect, and protect against forest fires, forest personnel need to define rules for determining forest fire risk. In Portugal, all municipalities must annually produce forest fire risk(FFR) maps. To produc...To prevent, detect, and protect against forest fires, forest personnel need to define rules for determining forest fire risk. In Portugal, all municipalities must annually produce forest fire risk(FFR) maps. To produce more reliable FFR maps more easily, we developed an open source model using the Modeler plugin of SEXTANTE in the program QGIS version 2.0 Dufour. The model provides all the maps involved in the FFR model(susceptibility map, hazard map, vulnerability map, economic value map,and potential loss map) and was produced according to Portuguese Forest Authority's(AFN, Autoridade Florestal Nacional) rules for determining the FFR. This model was tested for the Portuguese municipality Santa Maria da Feira, where 40 % of the total municipality area falls in the category ‘‘very high'' or ‘‘high'' fire risk. The ‘‘very high''fire risk area is mainly classified as broad-leaved forest and has the steepest slopes(>15 %). The distance of burned areas to roads was also analyzed; the proportion of burned areas increased with increasing distance to the main roads.In addition, 92.6 % of the ‘‘high'' and ‘‘very high'' risk zones were located in areas with lower elevation. These results confirmed that forest fire is strongly influenced not only by environmental factors but also by anthropogenic factors. The procedure implemented here was compared with our open source application already available in QGIS and also to the same procedure implemented in GIS proprietary software. Although the results were obviously the same, the model developed here presents several advantages over the other two approaches. Besides being faster,it is easy to change the model parameters according to user needs(i.e., to the rules of different countries), and can be modified and adapted to other variables and other areas to create risk maps for different natural phenomena(e.g.,floods, earthquakes, landslides). The model is easy to use and to create risk and hazard maps rapidly in a free, open source environment that does not require any programming knowledge.展开更多
The latest advancements in computer vision and deep learning(DL)techniques pave the way to design novel tools for the detection and monitoring of forestfires.In this view,this paper presents an intelligent wild forestfi...The latest advancements in computer vision and deep learning(DL)techniques pave the way to design novel tools for the detection and monitoring of forestfires.In this view,this paper presents an intelligent wild forestfire detec-tion and alarming system using deep learning(IWFFDA-DL)model.The pro-posed IWFFDA-DL technique aims to identify forestfires at earlier stages through integrated sensors.The proposed IWFFDA-DL system includes an Inte-grated sensor system(ISS)combining an array of sensors that acts as the major input source that helps to forecast thefire.Then,the attention based convolution neural network with bidirectional long short term memory(ACNN-BLSTM)model is applied to examine and identify the existence of danger.For hyperpara-meter tuning of the ACNN-BLSTM model,the bacterial foraging optimization(BFO)algorithm is employed and thereby enhances the detection performance.Finally,when thefire is detected,the Global System for Mobiles(GSM)modem transmits messages to the authorities to take required actions.An extensive set of simulations were performed and the results are investigated interms of several aspects.The obtained results highlight the betterment of the IWFFDA-DL techni-que interms of various measures.展开更多
The numerical simulation and slope stability prediction are the focus of slope disaster research.Recently,machine learning models are commonly used in the slope stability prediction.However,these machine learning mode...The numerical simulation and slope stability prediction are the focus of slope disaster research.Recently,machine learning models are commonly used in the slope stability prediction.However,these machine learning models have some problems,such as poor nonlinear performance,local optimum and incomplete factors feature extraction.These issues can affect the accuracy of slope stability prediction.Therefore,a deep learning algorithm called Long short-term memory(LSTM)has been innovatively proposed to predict slope stability.Taking the Ganzhou City in China as the study area,the landslide inventory and their characteristics of geotechnical parameters,slope height and slope angle are analyzed.Based on these characteristics,typical soil slopes are constructed using the Geo-Studio software.Five control factors affecting slope stability,including slope height,slope angle,internal friction angle,cohesion and volumetric weight,are selected to form different slope and construct model input variables.Then,the limit equilibrium method is used to calculate the stability coefficients of these typical soil slopes under different control factors.Each slope stability coefficient and its corresponding control factors is a slope sample.As a result,a total of 2160 training samples and 450 testing samples are constructed.These sample sets are imported into LSTM for modelling and compared with the support vector machine(SVM),random forest(RF)and convo-lutional neural network(CNN).The results show that the LSTM overcomes the problem that the commonly used machine learning models have difficulty extracting global features.Furthermore,LSTM has a better prediction performance for slope stability compared to SVM,RF and CNN models.展开更多
基金This research was funded by the National Natural Science Foundation of China(grant no.32271881).
文摘Forest fires are natural disasters that can occur suddenly and can be very damaging,burning thousands of square kilometers.Prevention is better than suppression and prediction models of forest fire occurrence have developed from the logistic regression model,the geographical weighted logistic regression model,the Lasso regression model,the random forest model,and the support vector machine model based on historical forest fire data from 2000 to 2019 in Jilin Province.The models,along with a distribution map are presented in this paper to provide a theoretical basis for forest fire management in this area.Existing studies show that the prediction accuracies of the two machine learning models are higher than those of the three generalized linear regression models.The accuracies of the random forest model,the support vector machine model,geographical weighted logistic regression model,the Lasso regression model,and logistic model were 88.7%,87.7%,86.0%,85.0%and 84.6%,respectively.Weather is the main factor affecting forest fires,while the impacts of topography factors,human and social-economic factors on fire occurrence were similar.
基金This research was funded by the National Natural Science Foundation of China(Grant Nos.31870426).
文摘Parameterization is a critical step in modelling ecosystem dynamics.However,assigning parameter values can be a technical challenge for structurally complex natural plant communities;uncertainties in model simulations often arise from inappropriate model parameterization.Here we compared five methods for defining community-level specific leaf area(SLA)and leaf C:N across nine contrasting forest sites along the North-South Transect of Eastern China,including biomass-weighted average for the entire plant community(AP_BW)and four simplified selective sampling(biomass-weighted average over five dominant tree species[5DT_BW],basal area weighted average over five dominant tree species[5DT_AW],biomass-weighted average over all tree species[AT_BW]and basal area weighted average over all tree species[AT_AW]).We found that the default values for SLA and leaf C:N embedded in the Biome-BGC v4.2 were higher than the five computational methods produced across the nine sites,with deviations ranging from 28.0 to 73.3%.In addition,there were only slight deviations(<10%)between the whole plant community sampling(AP_BW)predicted NPP and the four simplified selective sampling methods,and no significant difference between the predictions of AT_BW and AP_BW except the Shennongjia site.The findings in this study highlights the critical importance of computational strategies for community-level parameterization in ecosystem process modelling,and will support the choice of parameterization methods.
基金the following grants:The National Key R&D Program of China(2019YFA0606600)the Natural Science Foundation of China(31971577)the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)。
文摘Historical forest fire risk databases are vital for evaluating the effectiveness of past forest management approaches,enhancing forest fire warnings and emergency response capabilities,and accurately budgeting potential carbon emissions resulting from fires.However,due to the unavailability of spatial information technology,such databases are extremely difficult to build reliably and completely in the non-satellite era.This study presented an improved forest fire risk reconstruction framework that integrates a deep learning-based time series prediction model and spatial interpolation to address the challenge in Sichuan Province,southwestern China.First,the forest fire danger index(FFDI)was improved by supplementing slope and aspect information.We compared the performances of three time series models,namely,the autoregressive integrated moving average(ARIMA),Prophet and long short-term memory(LSTM)in predicting the modified forest fire danger index(MFFDI).The bestperforming model was used to retrace the MFFDI of individual stations from 1941 to 1970.Following this,the Anusplin spatial interpolation method was used to map the distributions of the MFFDI at five-year intervals,which were then subjected to weighted overlay with the distance-to-river layer to generate forest fire risk maps for reconstructing the forest fire danger database.The results revealed LSTM as the most accurate in fitting and predicting the historical MFFDI,with a fitting determination coefficient(R^2)of 0.709,mean square error(MSE)of0.047,and validation R^2 and MSE of 0.508 and 0.11,respectively.Independent validation of the predicted forest fire risk maps indicated that 5 out of 7 historical forest fire events were located in forest fire-prone areas,which is higher than the results determined from the original FFDI(2 out of 7).This proves the effectiveness of the improved MFFDI and indicates a high level of reliability of the historical forest fire risk reconstruction method proposed in this study.
基金This research was funded by the Ministry of Higher Education Malaysia via the Fundamental Research Grant Scheme(FRGS/1/2020/WAB03/UPM/02/1)。
文摘Tropical peat comprises decomposed dead plant material and acts like a sponge to absorb water,making it fully saturated.However,drought periods dry it readily and increases its vulnerability to fire.Peat fires emit greenhouse gases and particles contributing to haze,and prevention by constructing fire-break canals to reduce fire spread into forest reserves is crucial.This paper aims to determine peat physical and chemical properties near a fire-break canal at different fire frequency areas.Peat sampling was conducted at two forest reserves in Malaysia which represent low fire frequency and high fire frequency areas.The results show that peat properties were not affected by the construction of a fire-break canal,however lignin and cellulose content increased significantly from the distance of the canal in both areas.The study concluded that fire frequency did not significantly influence peat properties except for porosity.The higher fibre content in the high frequency area did not influence moisture content nor the ability to regain moisture.Thus,fire frequency might contribute differently to changes in physical and chemical properties,hence management efforts to construct fire-break canals and restoration efforts should protect peatlands from further degradation.These findings will benefit future management and planning for forest reserves.
文摘This article delves into the analysis of performance and utilization of Support Vector Machines (SVMs) for the critical task of forest fire detection using image datasets. With the increasing threat of forest fires to ecosystems and human settlements, the need for rapid and accurate detection systems is of utmost importance. SVMs, renowned for their strong classification capabilities, exhibit proficiency in recognizing patterns associated with fire within images. By training on labeled data, SVMs acquire the ability to identify distinctive attributes associated with fire, such as flames, smoke, or alterations in the visual characteristics of the forest area. The document thoroughly examines the use of SVMs, covering crucial elements like data preprocessing, feature extraction, and model training. It rigorously evaluates parameters such as accuracy, efficiency, and practical applicability. The knowledge gained from this study aids in the development of efficient forest fire detection systems, enabling prompt responses and improving disaster management. Moreover, the correlation between SVM accuracy and the difficulties presented by high-dimensional datasets is carefully investigated, demonstrated through a revealing case study. The relationship between accuracy scores and the different resolutions used for resizing the training datasets has also been discussed in this article. These comprehensive studies result in a definitive overview of the difficulties faced and the potential sectors requiring further improvement and focus.
基金the National Natural Science Foundation of China(Grant 42177164)the Distinguished Youth Science Foundation of Hunan Province of China(2022JJ10073).
文摘As massive underground projects have become popular in dense urban cities,a problem has arisen:which model predicts the best for Tunnel Boring Machine(TBM)performance in these tunneling projects?However,performance level of TBMs in complex geological conditions is still a great challenge for practitioners and researchers.On the other hand,a reliable and accurate prediction of TBM performance is essential to planning an applicable tunnel construction schedule.The performance of TBM is very difficult to estimate due to various geotechnical and geological factors and machine specifications.The previously-proposed intelligent techniques in this field are mostly based on a single or base model with a low level of accuracy.Hence,this study aims to introduce a hybrid randomforest(RF)technique optimized by global harmony search with generalized oppositionbased learning(GOGHS)for forecasting TBM advance rate(AR).Optimizing the RF hyper-parameters in terms of,e.g.,tree number and maximum tree depth is the main objective of using the GOGHS-RF model.In the modelling of this study,a comprehensive databasewith themost influential parameters onTBMtogetherwithTBM AR were used as input and output variables,respectively.To examine the capability and power of the GOGHSRF model,three more hybrid models of particle swarm optimization-RF,genetic algorithm-RF and artificial bee colony-RF were also constructed to forecast TBM AR.Evaluation of the developed models was performed by calculating several performance indices,including determination coefficient(R2),root-mean-square-error(RMSE),and mean-absolute-percentage-error(MAPE).The results showed that theGOGHS-RF is a more accurate technique for estimatingTBMAR compared to the other applied models.The newly-developedGOGHS-RFmodel enjoyed R2=0.9937 and 0.9844,respectively,for train and test stages,which are higher than a pre-developed RF.Also,the importance of the input parameters was interpreted through the SHapley Additive exPlanations(SHAP)method,and it was found that thrust force per cutter is the most important variable on TBMAR.The GOGHS-RF model can be used in mechanized tunnel projects for predicting and checking performance.
文摘This paper aims to analyze and evaluate a model of forest conservation and management of ethnic minority (or indigenous) people in central Vietnam, often referred to as the spiritual forest. These forests, called sacred forest or ghost forests by the ethnic minority people in Thua Thien Hue province, have existed for a long time among forest residents. However, they have recently declined, both in quality and quantity, due to various factors, including changes in society, economy, environment, and perception, among other reasons. Based on research conducted in A Luoi district, Thua Thien Hue province with household interviews, group discussions, and field surveys, we find that spiritual forest retains religious and human significance. They are also often among the last remaining natural forests left due to deforestation by human activities. The research results indicate challenges that conservation of spiritual forest may face, while giving recommendations derived from communities for sustainable forest development and conservation in the region.
基金Central Finance Forestry Science and Technology Promotion Demonstration Project(H[2023]TG31).
文摘The paper summarizes the structure and water-absorbing mechanism,classification,and preparation method of polymer fire extinguishing gel,and prospects for its application in aerial firefighting,forest ground fire extinguishing,opening of firebreaks,and mitigating human casualties in forest fire extinguishing.
文摘The mechanism of lightning that ignites a forest fire and the lightning that occurs above a forest fire are explained at the molecular level. It is based on two phenomena, namely, internal charge separation inside the atmospheric cloud particles and the existence of a layer of positively charged hydrogen atoms sticking out of the surface of the liquid layer of water on the surface of rimers. Strong turbulence-driven collisions of the ice particles and water droplets with the rimers give rise to breakups of the ice particles and water droplets into positively and negatively charged fragments leading to charge separation. Hot weather in a forest contributes to the updraft of hot and humid air, which follows the same physical/chemical processes of normal lightning proposed and explained recently[1]. Lightning would have a high probability of lighting up and burning the dry biological materials in the ground of the forest, leading to a forest (wild) fire. The burning of trees and other plants would release a lot of heat and moisture together with a lot of smoke particles (aerosols) becoming a strong updraft. The condition for creating lightning is again satisfied which would result in further lightning high above the forest wild fire.
基金financially supported by the National Key Research and Development Program of China(2022YFB3706800,2020YFB1710100)the National Natural Science Foundation of China(51821001,52090042,52074183)。
文摘The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was proposed to reduce casting defects and improve production efficiency,which includes the random forest(RF)classification model,the feature importance analysis,and the process parameters optimization with Monte Carlo simulation.The collected data includes four types of defects and corresponding process parameters were used to construct the RF model.Classification results show a recall rate above 90% for all categories.The Gini Index was used to assess the importance of the process parameters in the formation of various defects in the RF model.Finally,the classification model was applied to different production conditions for quality prediction.In the case of process parameters optimization for gas porosity defects,this model serves as an experimental process in the Monte Carlo method to estimate a better temperature distribution.The prediction model,when applied to the factory,greatly improved the efficiency of defect detection.Results show that the scrap rate decreased from 10.16% to 6.68%.
基金The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work under Grant Number(RGP.1/172/42)Princess Nourah bint Abdulrahman University Researchers Supporting Project Number(PNURSP2023R191)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.This study is supported via funding from Prince Sattam bin Abdulaziz University Project Number(PSAU/2023/R/1444).
文摘Earth resource and environmental monitoring are essential areas that can be used to investigate the environmental conditions and natural resources supporting sustainable policy development,regulatory measures,and their implementation elevating the environment.Large-scale forest fire is considered a major harmful hazard that affects climate change and life over the globe.Therefore,the early identification of forest fires using automated tools is essential to avoid the spread of fire to a large extent.Therefore,this paper focuses on the design of automated forest fire detection using a fusion-based deep learning(AFFD-FDL)model for environmental monitoring.The AFFDFDL technique involves the design of an entropy-based fusion model for feature extraction.The combination of the handcrafted features using histogram of gradients(HOG)with deep features using SqueezeNet and Inception v3 models.Besides,an optimal extreme learning machine(ELM)based classifier is used to identify the existence of fire or not.In order to properly tune the parameters of the ELM model,the oppositional glowworm swarm optimization(OGSO)algorithm is employed and thereby improves the forest fire detection performance.A wide range of simulation analyses takes place on a benchmark dataset and the results are inspected under several aspects.The experimental results highlighted the betterment of the AFFD-FDL technique over the recent state of art techniques.
文摘Forest fires are a significant threat to the environment, causing ecological damage, economic losses, and posing a threat to human life. Hence, timely detection and prevention of forest fires are critical to minimizing their impact. In this paper, we review the current state-of-the-art methods in forest fire detection and prevention using predictions based on weather conditions and predictions based on forest fire history. In particular, we discuss different Machine Learning (ML) models that have been used for forest fire detection. Further, we present the challenges faced when implementing the ML-based forest fire detection and prevention systems, such as data availability, model prediction errors and processing speed. Finally, we discuss how recent advances in Deep Learning (DL) can be utilized to improve the performance of current fire detection systems.
文摘Forest fire is one of the main natural hazards because of its fierce destructiveness. Various researches on fire real time monitoring, behavior simulation and loss assessment have been carried out in many countries. As fire prevention is probably the most efficient means for protecting forests, suitable methods should be developed for estimating the fire danger. Fire danger is composed of ecological, human and climatic factors. Therefore, the systematic analysis of the factors including forest characteristics, meteorological status, topographic condition causing forest fire is made in this paper at first. The relationships between biophysical factors and fire danger are paid more attention to. Then the parameters derived from remote sensing data are used to estimate the fire danger variables, According to the analysis, not only PVI (Perpendicular Vegetation Index) can classify different vegetation but also crown density is captured with PVI. Vegetation moisture content has high correlation with the ratio of actual evapotranspiration (LE) to potential ecapotranspiration (LEp). SI (Structural Index), which is the combination of TM band 4 and 5 data, is a good indicator of forest age. Finally, a fire danger prediction model, in which relative importance of each fire factor is taken into account, is built based on GIS.
基金funded by the National Postdoctoral Innovative Talents Support Plan China Postdoctoral Science Foundation (BX20220038)Key R&D Projects in Hainan Province (ZDYF2021SHFZ256)。
文摘Climate change has an impact on forest fire patterns.In the context of global warming,it is important to study the possible effects of climate change on forest fires,carbon emission reductions,carbon sink effects,forest fire management,and sustainable development of forest ecosystems.This study is based on MODIS active fire data from 2001-2020 and the influence of climate,topography,vegetation,and social factors were integrated.Temperature and precipitation information from different scenarios of the BCC-CSM2-MR climate model were used as future climate data.Under climate change scenarios of a sustainable low development path and a high conventional development path,the extreme gradient boosting model predicted the spatial distribution of forest fire occurrence in China in the 2030s(2021-2040),2050s(2041-2060),2070s(2061-2080),and2090s(2081-2100).Probability maps were generated and tested using ROC curves.The results show that:(1)the area under the ROC curve of training data(70%)and validation data(30%)were 0.8465 and 0.8171,respectively,indicating that the model can reasonably predict the occurrence of forest fire in the study area;(2)temperature,elevation,and precipitation were strongly correlated with fire occurrence,while land type,slope,distance from settlements and roads,and slope direction were less strongly correlated;and,(3)based on future climate change scenarios,the probability of forest fire occurrence will tend to shift from the south to the center of the country.Compared with the current climate(2001-2020),the occurrence of forest fires in 2021-2040,2041-2060,2061-2080,and 2081-2100 will increase significantly in Henan Province(Luoyang,Nanyang,S anmenxia),Shaanxi Province(Shangluo,Ankang),Sichuan Province(Mianyang,Guangyuan,Ganzi),Tibet Autonomous Region(Shannan,Linzhi,Changdu),Liaoning Province(Liaoyang,Fushun,Dandong).
基金This work was supported by the National Nature Science Foundation of China(awards 42,271,162,41,971,100)the Natural Science Foundation of Jilin Province(award 20220101149JC)the Scholarship Fund from China Scholarship Council(award 202,206,620,038).
文摘Ongoing climate changes have a direct impact on forest growth;they also affect natural fire regimes,with further implications for forest composition.Understanding of how these will affect forests on decadal-to-centennial timescales is limited.Here we use reconstructions of past vegetation,fire regimes and climate during the Holocene to examine the relative importance of changes in climate and fire regimes for the abundance of key tree species in northeastern China.We reconstructed vegetation changes and fire regimes based on pollen and charcoal records from Gushantun peatland.We then used generalized linear modelling to investigate the impact of reconstructed changes in summer temperature,annual precipitation,background levels of fire,fire frequency and fire magnitude to identify the drivers of decadal-to-centennial changes in forest openness and composition.Changes in climate and fire regimes have independent impacts on the abundance of the key tree taxa.Climate variables are generally more important than fire variables in determining the abundance of individual taxa.Precipitation is the only determinant of forest openness,but summer temperature is more important than precipitation for individual tree taxa with warmer summers causing a decrease in cold-tolerant conifers and an increase in warmth-demanding broadleaved trees.Both background level and fire frequency have negative relationships with the abundance of most tree taxa;only Pinus increases as fire frequency increases.The magnitude of individual fires does not have a significant impact on species abundance on this timescale.Both climate and fire regime characteristics must be considered to understand changes in forest composition on the decadal-to-centennial timescale.There are differences,both in sign and magnitude,in the response of individual tree species to individual drivers.
基金supported by the grants from the Natural Science Foundation of Hubei Province(No.2020CFB780)the Fundamental Research Funds for the Central Universities(No.2017KFYXJJ020).
文摘Objective Body fluid mixtures are complex biological samples that frequently occur in crime scenes,and can provide important clues for criminal case analysis.DNA methylation assay has been applied in the identification of human body fluids,and has exhibited excellent performance in predicting single-source body fluids.The present study aims to develop a methylation SNaPshot multiplex system for body fluid identification,and accurately predict the mixture samples.In addition,the value of DNA methylation in the prediction of body fluid mixtures was further explored.Methods In the present study,420 samples of body fluid mixtures and 250 samples of single body fluids were tested using an optimized multiplex methylation system.Each kind of body fluid sample presented the specific methylation profiles of the 10 markers.Results Significant differences in methylation levels were observed between the mixtures and single body fluids.For all kinds of mixtures,the Spearman’s correlation analysis revealed a significantly strong correlation between the methylation levels and component proportions(1:20,1:10,1:5,1:1,5:1,10:1 and 20:1).Two random forest classification models were trained for the prediction of mixture types and the prediction of the mixture proportion of 2 components,based on the methylation levels of 10 markers.For the mixture prediction,Model-1 presented outstanding prediction accuracy,which reached up to 99.3%in 427 training samples,and had a remarkable accuracy of 100%in 243 independent test samples.For the mixture proportion prediction,Model-2 demonstrated an excellent accuracy of 98.8%in 252 training samples,and 98.2%in 168 independent test samples.The total prediction accuracy reached 99.3%for body fluid mixtures and 98.6%for the mixture proportions.Conclusion These results indicate the excellent capability and powerful value of the multiplex methylation system in the identification of forensic body fluid mixtures.
基金financially supported by the National Natural Science Fundation of China(Grant Nos.42161065 and 41461038)。
文摘Understanding the mechanisms and risks of forest fires by building a spatial prediction model is an important means of controlling forest fires.Non-fire point data are important training data for constructing a model,and their quality significantly impacts the prediction performance of the model.However,non-fire point data obtained using existing sampling methods generally suffer from low representativeness.Therefore,this study proposes a non-fire point data sampling method based on geographical similarity to improve the quality of non-fire point samples.The method is based on the idea that the less similar the geographical environment between a sample point and an already occurred fire point,the greater the confidence in being a non-fire point sample.Yunnan Province,China,with a high frequency of forest fires,was used as the study area.We compared the prediction performance of traditional sampling methods and the proposed method using three commonly used forest fire risk prediction models:logistic regression(LR),support vector machine(SVM),and random forest(RF).The results show that the modeling and prediction accuracies of the forest fire prediction models established based on the proposed sampling method are significantly improved compared with those of the traditional sampling method.Specifically,in 2010,the modeling and prediction accuracies improved by 19.1%and 32.8%,respectively,and in 2020,they improved by 13.1%and 24.3%,respectively.Therefore,we believe that collecting non-fire point samples based on the principle of geographical similarity is an effective way to improve the quality of forest fire samples,and thus enhance the prediction of forest fire risk.
文摘To prevent, detect, and protect against forest fires, forest personnel need to define rules for determining forest fire risk. In Portugal, all municipalities must annually produce forest fire risk(FFR) maps. To produce more reliable FFR maps more easily, we developed an open source model using the Modeler plugin of SEXTANTE in the program QGIS version 2.0 Dufour. The model provides all the maps involved in the FFR model(susceptibility map, hazard map, vulnerability map, economic value map,and potential loss map) and was produced according to Portuguese Forest Authority's(AFN, Autoridade Florestal Nacional) rules for determining the FFR. This model was tested for the Portuguese municipality Santa Maria da Feira, where 40 % of the total municipality area falls in the category ‘‘very high'' or ‘‘high'' fire risk. The ‘‘very high''fire risk area is mainly classified as broad-leaved forest and has the steepest slopes(>15 %). The distance of burned areas to roads was also analyzed; the proportion of burned areas increased with increasing distance to the main roads.In addition, 92.6 % of the ‘‘high'' and ‘‘very high'' risk zones were located in areas with lower elevation. These results confirmed that forest fire is strongly influenced not only by environmental factors but also by anthropogenic factors. The procedure implemented here was compared with our open source application already available in QGIS and also to the same procedure implemented in GIS proprietary software. Although the results were obviously the same, the model developed here presents several advantages over the other two approaches. Besides being faster,it is easy to change the model parameters according to user needs(i.e., to the rules of different countries), and can be modified and adapted to other variables and other areas to create risk maps for different natural phenomena(e.g.,floods, earthquakes, landslides). The model is easy to use and to create risk and hazard maps rapidly in a free, open source environment that does not require any programming knowledge.
文摘The latest advancements in computer vision and deep learning(DL)techniques pave the way to design novel tools for the detection and monitoring of forestfires.In this view,this paper presents an intelligent wild forestfire detec-tion and alarming system using deep learning(IWFFDA-DL)model.The pro-posed IWFFDA-DL technique aims to identify forestfires at earlier stages through integrated sensors.The proposed IWFFDA-DL system includes an Inte-grated sensor system(ISS)combining an array of sensors that acts as the major input source that helps to forecast thefire.Then,the attention based convolution neural network with bidirectional long short term memory(ACNN-BLSTM)model is applied to examine and identify the existence of danger.For hyperpara-meter tuning of the ACNN-BLSTM model,the bacterial foraging optimization(BFO)algorithm is employed and thereby enhances the detection performance.Finally,when thefire is detected,the Global System for Mobiles(GSM)modem transmits messages to the authorities to take required actions.An extensive set of simulations were performed and the results are investigated interms of several aspects.The obtained results highlight the betterment of the IWFFDA-DL techni-que interms of various measures.
基金funded by the National Natural Science Foundation of China (41807285)。
文摘The numerical simulation and slope stability prediction are the focus of slope disaster research.Recently,machine learning models are commonly used in the slope stability prediction.However,these machine learning models have some problems,such as poor nonlinear performance,local optimum and incomplete factors feature extraction.These issues can affect the accuracy of slope stability prediction.Therefore,a deep learning algorithm called Long short-term memory(LSTM)has been innovatively proposed to predict slope stability.Taking the Ganzhou City in China as the study area,the landslide inventory and their characteristics of geotechnical parameters,slope height and slope angle are analyzed.Based on these characteristics,typical soil slopes are constructed using the Geo-Studio software.Five control factors affecting slope stability,including slope height,slope angle,internal friction angle,cohesion and volumetric weight,are selected to form different slope and construct model input variables.Then,the limit equilibrium method is used to calculate the stability coefficients of these typical soil slopes under different control factors.Each slope stability coefficient and its corresponding control factors is a slope sample.As a result,a total of 2160 training samples and 450 testing samples are constructed.These sample sets are imported into LSTM for modelling and compared with the support vector machine(SVM),random forest(RF)and convo-lutional neural network(CNN).The results show that the LSTM overcomes the problem that the commonly used machine learning models have difficulty extracting global features.Furthermore,LSTM has a better prediction performance for slope stability compared to SVM,RF and CNN models.