A geographical similarity-based sampling method of non-fire point data for spatial prediction of forest fires

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.

Performance Analysis of Support Vector Machine (SVM) on Challenging Datasets for Forest Fire Detection

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.

Lightning in a Forest (Wild) Fire: Mechanism at the Molecular Level

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.

Review of state-of-the-art decision support systems (DSSs) for prevention and suppression of forest fires

Forest ecosystems are our priceless natural resource and are a key component of the global carbon budget. Forest fires can be a hazard to the viability and sustainable management of forests with consequences for natural and cultural environments, economies, and the life quality of local and regional populations. Thus, the selection of strategies to manage forest fires, while considering both functional and economic efficiency, is of primary importance. The use of decision support systems(DSSs) by managers of forest fires has rapidly increased. This has strengthened capacity to prevent and suppress forest fires while protecting human lives and property. DSSs are a tool that can benefit incident management and decision making and policy, especially for emergencies such as natural disasters. In this study we reviewed state-of-the-art DSSs that use: database management systems and mathematical/economic algorithms for spatial optimization of firefighting forces; forest fire simulators and satellite technology for immediate detection and prediction of evolution of forest fires; GIS platforms that incorporate several tools to manipulate, process and analyze geographic data and develop strategic and operational plans.

Prediction of forest fire occurrence in China under climate change scenarios

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).

A Survey of the Machine Learning Models for Forest Fire Prediction and Detection

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.

Trends in Global and Mexico Research in Wildfires: A Bibliometric Perspective

In the last two decades, unprecedented changes have taken place in the frequency and severity of wildfires;in different regions of the world, some fires were even classified as megafires. Although there are studies about the diverse effects of fire, which have made significant theoretical contributions, a comprehensive review of the changes in fire research is required to understand worldwide patterns, particularly in those countries where fire activity is on the rise, such is the case of Mexico. The objective of this study was to analyze the trends in the research on wildfires published in Mexico and worldwide over a 40-year timescale. For this purpose, the Web of Science database, bibliometric tools, and the keywords TI = Forest fire* OR TI = Wildfire* were used to extract as many articles as possible related to fires from 1980 to 2020, without being restricted to those studies whose title included any of the variants of the keywords. There were 8458 publications about fires in the vegetation cover, with a notable increase in the frequency of studies in the previous decade;52% of the studies were concentrated in five countries and 20% of the articles focused on the study of different aspects of the soil. Mexico ranks thirteenth in volume of scientific production and studies in the country have focused mainly on the description of the quantitative relationship between the size of the affected area and the number of occurrences in the landscape, meanwhile, studies on fires and the consequences on the biotic interactions have been little explored.

Estimation of the carbon storage of forest vegetation and carbon emission from forest fires in Heilongjiang Province,China

Heilongjiang 省的森林资源在六的基础国家森林库存数据预定的 china.On 有重要位置(1973-1976， 1977-1981， 1985-1988,1989-1993， 1994-1998， 1999-2003 ) 由 P. R 的林地部调查了。从 1973 ～ 2003 的中国，在 Heilongjiang 省的森林的碳存储被使用在生物资源和体积之间的每树种的 linearrelationship 的方法估计。结果证明在六个时期的 Heilongjiang 森林的 carbonstorage ( 1973-1976 ， 1977-1981 ， 1985-1988 ， 1989-1993,1994-1998 ， 1999-2003 )是 7.164x10 ~ 8 t ， 4.871x10 ~ 8 t ， 5.094x10 ~ 8 t ， 5.292x10 ~ 8 t ， 5.594x10 ~ 8 t and5.410x10~8t ，分别地它显示出在早时间减少然后增加的一个趋势。Heilongjiang 森林玩的 Itindicated 一个重要角色作为经过 30 年的大气的碳 dioxideduring 的一个水池。从 1980～1999 基于森林火的数据并且为某森林的地面 biomassestimation 录入 Heilongjiang 省， meanannual 的数量消费了森林的生物资源，这被估计是 391758.65t-522344.95t ，为来自森林的 totalnational 完成的6.4%-8.4%的财务开火，并且碳排放的数量是 176 291.39t-235055.23t ，来自森林的大约8%全部的国家排放开火。CO_2 的排放，公司，分别地，来自在 Heilongjiang 省的森林火的 CH_4 andNMHC 在 581761.6-775682.25 t， 34892.275-46523.04 t， 14091.11-18788.15 t 和 6500-9000 t 被估计每年。

Wildland fires and moist deciduous forests of Chhattisgarh, India: divergent component assessment

We studied moist deciduous forests of Chhattisgarh, India （1） to assess the effect of four levels of historic wildland fire frequency （high, medium, low, and no-fire） on regeneration of seedlings in fire affected areas during pre and post-fire seasons, （2） to evaluate vegetation struc- ture and diversity by layer in the four fire frequency zones, （3） to evalu- ate the impact of fire frequency on the structure of economically impor- tant tree species of the region, and （4） to quantify fuel loads by fire fre- quency level. We classified fire-affected areas into high, medium, low, and no-fire frequency classes based on government records. Tree species were unevenly distributed across fire frequency categories. Shrub density was maximum in zones of high fire frequency and minimum in low- frequency and no-fire zones. Lower tree density after fires indicated that regeneration of seedlings was reduced by fire. The population structure in the high-frequency zone was comprised of seedlings of size class （A） and saplings of size class （B）, represented by Diospyros melanoxylon, Dalbergia sissoo, Shorea robusta and Tectona grandis. Younger and older trees were more abundant for Tectona grandis and Dalbargia sis- soo after fire, whereas intermediate-aged trees were more abundant pre- fire, indicating that the latter age-class was thinned by the catastrophic effect of fire. The major contributing components of fuel load included duff litter and small woody branches and twigs on the forest floor. Total fuel load on the forest floor ranged from 2.2 to 3.38 Mg/ha. The netchange in fuel load was positive in high- and medium-frequency fire zones and negative under low- and no-fire zones. Repeated fires, how- ever, slowly reduced stand stability. An ecological approach is needed for fire management to restore the no-fire spatial and temporal structure of moist deciduous forests, their species composition and fuel loads. The management approach should incorporate participatory forest manage- ment. Use of controlled fire, fire lines and mapping of fire prone areas are fundamental principles of fire hazard reduction in these areas.

Drought and Spatiotemporal Variability of Forest Fires Across Mexico

Understanding the spatiotemporal links between drought and forest fire occurrence is crucial for improving decision-making in fire management under current and future climatic conditions. We quantified forest fire activity in Mexico using georeferenced fire records for the period of 2005–2015 and examined its spatial and temporal relationships with a multiscalar drought index, the Standardized Precipitation-Evapotranspiration Index(SPEI). A total of 47975 fire counts were recorded in the 11-year long study period, with the peak in fire frequency occurring in 2011. We identified four fire clusters, i.e., regions where there is a high density of fire records in Mexico using the Getis-Ord G spatial statistic. Then, we examined fire frequency data in the clustered regions and assessed how fire activity related to the SPEI for the entire study period and also for the year 2011. Associations between the SPEI and fire frequency varied across Mexico and fire-SPEI relationships also varied across the months of major fire occurrence and related SPEI temporal scales. In particular, in the two fire clusters located in northern Mexico(Chihuahua, northern Baja California), drier conditions over the previous 5 months triggered fire occurrence. In contrast, we did not observe a significant relationship between drought severity and fire frequency in the central Mexico cluster, which exhibited the highest fire frequency. We also found moderate fire-drought associations in the cluster situated in the tropical southern Chiapas where agriculture activities are the main causes of forest fire occurrence. These results are useful for improving our understanding of the spatiotemporal patterns of fire occurrence as related to drought severity in megadiverse countries hosting many forest types as Mexico.

Mapping forest fires by nonparametric clustering analysis

Fires have a noteworthy role to play with regards to ecological and environmental losses in Mediterranean forests. In addition to ecological impacts, fire may create economic, social as well as cultural changes. The detection of fire-scars has critical importance to help decrease losses.In the present study, forest fires recorded in Antalya, one of the most important ecological and tourist regions within the Western Mediterranean, were clustered and mapped. Since the dominant factors and devastation records derived from the cases had nominal-scaled properties, a categorical databased nonparametric clustering algorithm was performed in this evaluation. The proposed tool, k-modes algorithm,uses modes instead of means for clustering. The algorithm may be implemented quickly and does not make distributional assumptions concerning the available data. It uses a frequency-based method to update the modes of the fires.The derived modes from the maps may be useful information for local authorities to manage. In conclusion, the proposed nonparametric clustering procedure may be employed to build a decision-support system to monitor and identify fire activities and to enhance fire management efficiency.

Modeling forest fires in Mazandaran Province, Iran

We used a spafio-temporal shot-noise Cox process to study the distribution of forest fires reported between 2006 and 2010 in the Mazandaran Province＇s forests. The fitted model shows that daily temperature, altitude, and slope-exposure impacted fire occurrence. Forest fire occurred in the region had an aggregated behavior, which increased in radius below 1-km away from fired areas; a periodic pattern of fire occurrence in the region was verified. The risk of forest fire is significantly higher for areas with southern exposure and slope between 30° and 50°, northern exposure and slope between 0° and 50°, and eastern exposure and slope between 0° and 30°. The risk of fire was also significantly higher at altitudes between 1350 and 3000 m asl. Human causes were the main ignition source for forest fires in the region. The fire occurrence rate stayed above average during the drought period from September 2008 to September 2009. Our findings could lead to the development of fire-response and fire-suppression strategies appropriate to specific regions.

Environmental and Economic Impact of Forest Fires in Puerto Rico 2013-2014

This study presents an analysis of the impact of forest fires in Puerto Rico for the period from 2013-2014. The climatological factors analyzed included precipitation, temperature, relative humidity, and wind. Several factors have combined to the increase of these forest fires, among others, a decrease in precipitation during this period, as well as an increase in the human involvement in these fires from approximately 40% occurs in the night period (5:00 pm to 8:00 am), where the weather conditions do not favor the appearance of these phenomena. An increase in fires of 44% occurred in 2013 compared to 2014, causing an economic loss of $13.8 million. Fire also adversely affected the flora and fauna of the island, but this was not evaluated in this paper.

Forest Fires and Climate Correlation in Mexico State: A Report Based on MODIS

Forest fires are one of the most important threats for forests in the State of Mexico. Therefore, understanding their geographical patterns is a priority for the design of forest management strategies. We processed the records obtained with the MOD14A2 product (for thermal anomalies and fire) of MODIS sensor. Such scenes correspond to dry seasons (from March 15 to June 30) from 2000 to 2012 in the State of Mexico. We analyzed such records in a GIS environment to learn their spatial patterns and establish their geographical correlations as a first step to understand the causal agents of forest fires. As a result, forest fires in the State of Mexico showed a clustered spatial trend with a southwest tendency and a slight spatial relation with total winter precipitation and maximal temperature in summer.

Geospatial Model as Strategy to Prevent Forest Fires: A Case Study

The forests of the State of Durango have been severely affected by fires in recent years. Early detection of fires through watchtowers is essential. In this work a geospatial model was generated to optimize strategic visualization points, using a GIS environment. Analysis of the area of visibility was developed by integrating a digital model of elevation and a plant cover map. The resulting distribution generates more than 50% coverage of the studied area, in points that were not always the highest. It was concluded that this strategy would permit to increase the efficiency, mainly favoring the communities of pine, whose economic importance would justify the required investment.

Serpentinite Slurries against Forest Fires

Forest fires are one of the commonest natural hazards. Forest fires make the largest contribution to CO2 emissions after the burning of fossil fuels. Here a new technology is proposed to extinguish forest fires not with water, but with a slurry of serpentine. Serpentinites are abundantly available in many countries on every continent. If serpentine is calcined, it weathers very fast and captures CO2. Calcination, however, requires a lot of heat, which makes it counterproductive to produce calcined serpentine for CO2 capture. In cases, however, where heat is the problem, like in forest fires, one can extinguish them to greater advantage by using serpentinite slurries instead of plain water. The calcined residue that is left as a thin cake on the burning material prevents oxygen to reach the burning material. It also prevents the escape of inflammable gases, and the calcination itself withdraws large quantities of heat from the fire. After the fire is extinguished, the calcined material in contact with the atmosphere will rapidly weather and capture CO2. This compensates part of the CO2 that is produced by the fire. In tests, where the efficacy of quenching fires with serpentine slurries was compared to the effect of water, it turned out that serpentinite slurries performed far better.

Efficient Application of the Radiance Enhancement Method for Detection of the Forest Fires Due to Combustion-Originated Reflectance

The existing methods for detection of the cloud scenes are applied at relatively small spectral range within shortwave upwelling radiative wavelength flux. We have reported a new method for detection of the cloud scenes based on the Radiance Enhancement (RE). This method can be used to cover a significantly wider spectral range from 1100 nm to 1700 nm by using datasets from the space-orbiting micro-spectrometer Argus 1000. Due to high sunlight reflection of the smoke originated from the forest or field fires the proposed RE method can also be implemented for detection of combustion aerosols. This approach can be a promising technique for efficient detection and continuous monitor of the seasonal forest and field fires. To the best of our knowledge this is the first report showing how a cloud method can be generalized for efficient detection of the forest fires due to combustion-originated reflectance.

CORRELATION ANALYSIS OF FOREST FIRES AND SUNSPOTS

This paper analyses the relationship of forest fires and sunspots in Hcilongjiang Province in the past 40 ycars(1950-1989). The results indicated that each of the forest fire indexes such as forest Fire rate(y1), times of forest firc(Y2), average forest fire area for one time Y4 (equals the Y5 / Y2) and total forest fire arca(Y5) has a negative correlation with mean annual relative sunspots of the same year; total times of forest fire inside and outside of forest stands( Y3) has a positive correlation with sunspots. The 5 indexes stated above has a similar relation to last year's mean annual relative sunspots.