Fire Occurrence Research Articles

Проверка сообщений о преступлениях, связанных с лесными пожарами

The investigation of crimes associated with forest fires is a critical endeavor aimed at determining the causative factors preceding the commencement of criminal proceedings. Key aspects under scrutiny encompass the origin of the forest fire, extent of its propagation, magnitude of material damage incurred, employed technical means and methodologies for fire suppression, and identification of individuals present within the boundaries of the forested area where ignition occurred. The investigative protocol entails a comprehensive array of measures including on-site inspection of the forest fire locus, testimonies elicited from eyewitnesses and lessees of forested parcels, procurement and scrutiny of documents sourced from meteorological services, territorial forest protection authorities, and commissioning of judicial expertise. Additionally, the judicious utilization of unmanned aerial vehicles (UAVs) in fire scene reconnaissance is advocated for its intrinsic safety benefits to operational personnel. Moreover, the significance of satellite-derived imagery depicting pre- and post-fire forested terrain conditions is underscored for facilitating causal inference regarding forest fire occurrences. In conclusion, the investigation of crimes associated with forest fires embodies a symbiotic interplay between scientific inquiry and legal scrutiny, wherein the convergence of empirical evidence and jurisprudential deliberation serves as the linchpin of investigative integrity and judicial recourse. Moving forward, continued investment in research and technological development is imperative to augment the efficacy and expediency of forest fire investigations, thereby fortifying societal resilience against the perils of environmental degradation and criminal malfeasance.

Framework to Create Inventory Dataset for Disaster Behavior Analysis Using Google Earth Engine: A Case Study in Peninsular Malaysia for Historical Forest Fire Behavior Analysis

This study developed a comprehensive framework using Google Earth Engine to efficiently generate a forest fire inventory dataset, which enhanced data accessibility without specialized knowledge or access to private datasets. The framework is applicable globally, and the datasets generated are freely accessible and shareable. By implementing the framework in Peninsular Malaysia, significant forest fire factors were successfully extracted, including the Keetch–Byram Drought Index (KBDI), soil moisture, temperature, windspeed, land surface temperature (LST), Palmer Drought Severity Index (PDSI), Normalized Vegetation Index (NDVI), landcover, and precipitation, among others. Additionally, this study also adopted large language models, specifically GPT-4 with the Noteable plugin, for preliminary data analysis to assess the dataset’s validity. Although the plugin effectively performed basic statistical analyses and visualizations, it demonstrated limitations, such as selectively dropping or choosing only relevant columns for tests and automatically modifying scales. These behaviors underscore the need for users to perform additional checks on the codes generated to ensure that they accurately reflect the intended analyses. The initial findings indicate that factors such as KBDI, LST, climate water deficit, and precipitation significantly impact forest fire occurrences in Peninsular Malaysia. Future research should explore extending the framework’s application to various regions and further refine it to accommodate a broader range of factors. Embracing and rigorously validating large language model technologies, alongside developing new tools and plugins, are essential for advancing the field of data analysis.

Efek Kebakaran Hutan Kalimantan Pada Hukum Internasional

Indonesia is an archipelagic country where almost more than half of its area has forests that store various kinds of plants, it indicates that almost all regions of Indonesia have a forest ecosystem that are very important for the survival of Indonesia and the world. Forests have many positive impacts that can be used by the community to become a livelihood. Behind the forests owned by Indonesia, there are various problems that are of international concern, one of which is forest fires. The causes of forest fires are very diverse which causes the impact of smoke from forest fires to spread to our neighboring countries such as Singapore and Malaysia which makes smoke from these forest fires a negative view of the international world on Indonesia. Human resource factors are important in the prevention and handling of forest fires, human factors such as opposing loads are positive and negative, on the one hand humans can prevent forest fires on the other hand humans can be the cause of forest fires. These things are complications experienced by Indonesia in handling and preventing forest fires, however, natural factors can also affect the occurrence of forest fires. The country has a responsibility in maintaining its natural sustainability as well as the country has a responsibility in the smoke problem that causes neighboring countries to suffer losses from both the economic sector and other sectors.

Amazon Wildfires and Respiratory Health: Impacts during the Forest Fire Season from 2009 to 2019.

The Brazilian Amazon, a vital tropical region, faces escalating threats from human activities, agriculture, and climate change. This study aims to assess the relationship between forest fire occurrences, meteorological factors, and hospitalizations due to respiratory diseases in the Legal Amazon region from 2009 to 2019. Employing simultaneous equation models with official data, we examined the association between deforestation-induced fires and respiratory health issues. Over the studied period, the Legal Amazon region recorded a staggering 1,438,322 wildfires, with 1,218,606 (85%) occurring during August-December, known as the forest fire season. During the forest fire season, a substantial portion (566,707) of the total 1,532,228 hospital admissions for respiratory diseases were recorded in individuals aged 0-14 years and 60 years and above. A model consisting of two sets of simultaneous equations was constructed. This model illustrates the seasonal fluctuations in meteorological conditions driving human activities associated with increased forest fires. It also represents how air quality variations impact the occurrence of respiratory diseases during forest fires. This modeling approach unveiled that drier conditions, elevated temperatures, and reduced precipitation exacerbate fire incidents, impacting hospital admissions for respiratory diseases at a rate as high as 22 hospital admissions per 1000 forest fire events during the forest fire season in the Legal Amazon, 2009-2019. This research highlights the urgent need for environmental and health policies to mitigate the effects of Amazon rainforest wildfires, stressing the interplay of deforestation, climate change, and human-induced fires on respiratory health.

The Dynamics of Fire Activity in the Brazilian Pantanal: A Log-Gaussian Cox Process-Based Structural Decomposition

We present a novel statistical methodology for analyzing shifts in spatio-temporal fire occurrence patterns within the Brazilian Pantanal, utilizing remote sensing data. Our approach employs a Log-Gaussian Cox Process to model the spatiotemporal dynamics of fire occurrence, deconstructing the intensity function into components of trend, seasonality, cycle, covariates, and time-varying spatial effects components. The results indicate a negative correlation between rainfall and fire intensity, with lower precipitation associated with heightened fire intensity. Forest formations exhibit a positive effect on fire intensity, whereas agricultural land use shows no significant impact. Savannas and grasslands, typical fire-dependent ecosystems, demonstrate a positive relationship with fire intensity. Human-induced fires, often used for agricultural purposes, contribute to an increase in both fire frequency and intensity, particularly in grassland areas. Trend analysis reveals fluctuating fire activity over time, with notable peaks in 2018–2021.

Predicting post-fire forest recovery using the 3-PG model with bi-temporal Landsat imagery in high-severity burned areas of Great Xing’an Mountain

High-severity fire altered the landscape pattern and destroyed forest resilience in the Great Xing’an Mountain. Predicting the spatiotemporal patterns of forest recovery in high-severity burned areas is critical for formulating effective post-fire management measures. Previous studies attempted to address this topic by fitting the recovery trajectory based on time-series remote sensing vegetation indices. However, these methods typically require long-term time series of satellite images for prediction and have sometimes been demonstrated as an unrealistic representation of the post-fire forest patterns due to the saturation issues of vegetation indices. In this study, we proposed a novel approach to predict post-fire forest recovery patterns by coupling a process-based Physiological Principles in Predicting Growth (3-PG) model with bi-temporal satellite imagery. First, based on tree core data, we calibrated the physiological parameters of the 3-PG model using a Model-Independent Parameter Estimation (PEST) model. Second, we used vegetation indices derived from bi-temporal Landsat data to estimate site-specific parameters of the 3-PG model and extrapolated the meteorological factors by a Mountain Microclimate Simulation Model (MTCLIM). Third, we used these variables as input parameters to drive the 3-PG model, resulting in the simulation of the post-fire 50-year regeneration dynamic of the leaf area index (LAI). The calibrated 3-PG model was validated by the independent sites, providing good estimates of DBH and biomass components with R2 of greater than 0.994 and RRMSE of less than 15.7 %. The 3-PG model predicted LAI maps were evaluated by referenced LAI retrieved from Landsat data, showing a high correlation with reference but the accuracy decreased with the prediction time increased. The model predicted forest recovery to pre-fire LAI within 32 years where the longest recovery interval was observed in the low seedling density and soil fertility. This study shows great potential for reasonable prediction and assessment of forest recovery after the fire occurrence. By combining more simulated scenarios, our approach also facilitates the exploration of the impact of artificial measures and climate on post-forest recovery.

Geospatial forest fire risk assessment and zoning by integrating MaxEnt in Gorkha District, Nepal

Forest fires are an imminent danger to natural forest ecosystems, and carrying out zoning studies and forest fire risk assessments are of great practical significance in steering fire prevention, minimizing fire incidents, and limiting the environmental consequences of fire. Using the Gorkha district of Nepal as a case study, this study used remotely sensed high-temperature fire data as the forest fire sample. Nine parameters related to topography, climatic conditions, vegetation, and human intervention were used as environmental variables affecting fire occurrence. Next, a MaxEnt forest fire risk assessment model was generated with GIS and R, which analysed the contribution, significance, and responses of environmental variables to the forest fire in Gorkha District. The findings demonstrate that (1) following a test of sample locations for forest fires, the MaxEnt model has excellent relevance and practicality when applied to fire risk assessment; (2) Out of 2747 fire points in the forest, only 110 Spatio-temporally independent fire points were used for the model building having high and normal confidence level. Regarding Area Under Curve (AUC) values, the training data yielded results of 0.875, while the test data produced acceptable results of 0.861 with a standard deviation of 0.0322; (3) the importance of climatic and Land Use Land Cover (LULC) variables to forest fire are 56.2 % and 32.9 %, respectively, and their contribution to forest fire are 32 % and 47.6 %, respectively. (4) There are numerous and intricate ways that environmental factors influence forest fires. The forest fire response curves to the nine chosen environmental variables are complex and nonlinear rather than linear; Maximum temperature of the warmest month (bio_5), Isothermality (bio_3), Precipitation of Driest Quarter (bio_17) and mean Diurnal Range (bio_2) bear a nonlinear positive link with the possibility of forest fires. In contrast, elevation, slope, temperature seasonality (bio_4), distance from the settlement, and LULC have a favorable stimulating response to the possibility of forest fires within an appropriate interval. (5) In Gorkha, there are geographical differences in the risk of forest fires. Only 12.83 % of the whole area is made up of areas at significantly high risk or above, compared to 87.17 % for high-risk and below.

Historical association between respiratory diseases hospitalizations and fire occurrence in the Pantanal wetland, Brazil

The occurrence of fires in the Pantanal wetland increases during drought conditions, resulting in higher aerosol emissions and deterioration of air quality. For the first time, a comprehensive assessment of the climatic effects on monthly Respiratory Disease Hospitalizations (RDH) was conducted in the Pantanal region. We investigated correlations among climatic variables and health indicators, focusing on three groups: RDH5 (<5 years old), RDH60 (≥60 years old), and RDH (total records) between 2003 and 2019 in all 17 municipalities of the region. We reported a 26–34% increase in RDH during drought years, as represented by the negative relationship between RDH and precipitation. Precipitation and humidity exert a primary control on fire occurrence and, hence, on respiratory illnesses in the Pantanal. Particulate material emissions are strongly related to increased fires and temperature across the region. Incidence rates among children and the elderly were four and three times higher, respectively, than general population, confirming the higher vulnerability of these groups. To effectively mitigate the risks and vulnerabilities associated with droughts, policies and action plans should prioritize vulnerable populations and implement targeted measures to reduce their exposure.

Predicting the Occurrence of Forest Fire in the Central-South Region of China

Understanding the spatial and temporal patterns of forest fires, along with the key factors influencing their occurrence, and accurately forecasting these events are crucial for effective forest management. In the Central-South region of China, forest fires pose a significant threat to the ecological system, public safety, and economic stability. This study employs Geographic Information Systems (GISs) and the LightGBM (Light Gradient Boosting Machine) model to identify the determinants of forest fire incidents and develop a predictive model for the likelihood of forest fire occurrences, in addition to proposing a zoning strategy. The purpose of the study is to enhance our understanding of forest fire dynamics in the Central-South region of China and to provide actionable insights for mitigating the risks associated with such disasters. The findings reveal the following: (i) Spatially, fire incidents exhibit significant clustering and autocorrelation, highlighting areas with heightened likelihood. (ii) The Central-South Forest Fire Likelihood Prediction Model demonstrates high accuracy, reliability, and predictive capability, with performance metrics such as accuracy, precision, recall, and F1 scores exceeding 85% and AUC values above 89%, proving its effectiveness in forecasting the likelihood of forest fires and differentiating between fire scenarios. (iii) The likelihood of forest fires in the Central-South region of China varies across regions and seasons, with increased likelihood observed from March to May in specific provinces due to various factors, including weather conditions and leaf litter accumulation. Risks of localized fires are noted from June to August and from September to November in different areas, while certain regions continue to face heightened likelihood from December to February.

Fire danger in the supply areas of hydroelectric reservoirs under the restoration process in the south of Minas Gerais, Brazil

The partial or complete loss of vegetation cover triggers an increase in surface runoff, erosion, and sedimentation of water bodies, including reservoirs for hydroelectric power generation, reducing their life expectancy. To control or mitigate this issue, ecological restoration interventions should prioritize the recovery of areas most vulnerable to these processes, such as springs. Conversely, wildfires cause damage to vegetation cover and hinder ecological restoration and/or natural regeneration processes. Therefore, this study aimed to identify - with the aid of Geographic Information Systems - the temporal and spatial likelihood of fire occurrences in spring recharge areas undergoing ecological restoration and contributing to hydroelectric reservoirs in southern Minas Gerais, Brazil. The findings indicated that the months with the highest probability of wildfire occurrences were August and September (accounting for 66% of cases), requiring increased attention to prevention efforts. Furthermore, locations most susceptible to fires (steeper slopes, more flammable vegetation, and higher anthropogenic use) should be treated as priorities for both prevention and ecological restoration actions.

Impact of Hill Fires on Dissolved Organic Matter in Watersheds of Karst Areas Based on Three-Dimensional Fluorescence-Parallel Factor Analysis

Hill fires have the potential to influence dissolved organic matter (DOM) in water bodies, yet fewer studies have investigated the effects of hill fires on DOM within watersheds in karst areas. In this study, we employed the three-dimensional fluorescence-parallel factor analysis (EEM-PARAFAC) method to analyze the DOM fluorescence peaks, component compositions, fluorescence indices, and sources within the water body of the Yuanteng River sub-basin, which was impacted by the hill fire, serving as our primary research focus. The results indicate the presence of three primary fluorescent fractions in the water body of the Yuanteng River: C1, resembling humic acid (fulvic acid); C2, consisting of biopolymers and microbial by-products; and C3, containing proteins such as tyrosine and tryptophan. The Yuanteng River exhibited elevated levels of humus-like substances, diminished concentrations of protein-like substances, and demonstrated higher biogenic, freshness, and humification indices compared to unaffected water samples, reflecting the impact of the hill fire. Elevated levels of exogenous humic acid-like inputs into the waters of the Yuanteng River, along with exogenous inputs of DOM, were primarily influenced by stable, high-molecular-weight organic matter. Additionally, agricultural effluent, domestic sewage, and anthropogenic activities contributed to these inputs to a lesser degree. The impacts of endogenous inputs are mainly related to the restoration of ecosystems. The occurrence of hill fires has significantly influenced the composition of dissolved organic matter in the waters of the Yuanteng River. A comprehensive analysis of the impacts of hill fires on dissolved organic matter in water bodies can serve as a valuable reference for characterizing DOM in the water bodies of the Yuanteng River. Furthermore, it can inform strategies for environmental protection, facilitate the traceability of pollutants in water bodies, and contribute to environmental and ecological restoration efforts following hill fires in the region.

Thermal analysis of fire effects on fire doors using the finite element method

The goal of this paper is to use the finite element method and computer simulation to prove the effectiveness in assessing the fire resistance of building construction elements, specifically on the example of the fire resistance of fire doors. For that purpose, data from the building where fire doors with defined characteristics were installed were used. For simulation purposes, numerical modelling with thermal and structural analysis will be used. In this way, the parameters of the temperature distribution on the fire door and the contact material due to fire will be obtained, as well as the distribution of strain and stresses, which will indicate the fire resistance of the used structure. Computer simulation with numerical modelling offers a number of advantages both in the speed of providing results of a larger number of variants of the simulation model and in the accuracy of the results obtained once the model is calibrated. Also, the mentioned type of predicting the effect of fire can be applied to other elements of the structure of the construction object, which can significantly influence the decision-making that will prevent the negative consequences of the occurrence of a possible fire.

Influence of wildfires on the conflict (2006–2022) in eastern Ukraine using remote sensing techniques (MODIS and Sentinel-2 images)

Humans primarily alter fire regimes in three ways: changing the distribution and density of ignitions, shifting the seasonality of burning, or altering available fuels. Remote sensing techniques are a highly feasible and effective tool for monitoring wildfires, delineating burn perimeters, describing patterns of fire occurrence, characterizing the degree of intensity, or fire risk mapping. In this study MODIS and Sentinel-2 satellite images were used to analyze the burned areas in the Luhansk and Donetsk Regions of Ukraine (2006–2022), and specifically how the political and military hostilities in 2014 impacted fire activity. The burned areas detected by the MODIS sensor were selected using a size (32 ha) and temporal (May–October) filter. The Relativized Burn Ratio (RBR) index was used to identify burned areas, while the Normalized Difference Water Index (NDWI) was employed to discriminate the water bodies in Sentinel-2 images. We analyzed the burned area data over the 17-year period, eight of them during the years before the conflict and the other nine after the conflict. The results show that burned area has increased from the pre-conflict period (2008–2013) by 689,898 ha to the post-conflict period (2014–2022) around the contact line. In addition, the frequency of fires has increased in the post-war period, especially towards the south of Donetsk. Finally, Sentinel-2 imagery was used to map Large Wildfires (LAWF), which means fires with a burned area >500 ha. Our study concluded that MODIS MCD64A1 underestimates the burned area with a recorded difference of up to 40% of the area in one of the analyzed fires. In addition, an analysis of the burned areas classified according to the main land use has been performed for both sensors. The results obtained show a regression for the total area of R2 = 0,97 and both sensors show high correlation between them for each land use analyzed. Grasslands and agricultural land covers were most affected by LAWFs during the conflict period (2014–2022), noting that fires near urban entities were also more frequent in the war period.

Zagros Grass Index—A New Vegetation Index to Enhance Fire Fuel Mapping: A Case Study in the Zagros Mountains

Annually, the oak forests of the Zagros Mountains chains in western Iran and northeastern Iraq face recurring challenges posed by forest fires, particularly in the Kurdo–Zagrosian forests in western Iran and northeastern Iraq. Assessing fire susceptibility relies significantly on vegetation conditions. Integrating in situ data, Remote Sensing (RS) data, and Geographical Information Systems (GIS) integration presents a cost-effective and precise approach to capturing environmental conditions before, during, and after fire events, minimizing the need for extensive fieldwork. This study refines and applies the Zagros Grass Index (ZGI), a local vegetation index tailored to discern between grass-covered surfaces and tree canopies in Zagros forests, identifying the grass masses as the most flammable fuel type. Utilizing the Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) product as input from 2013 to 2022, the ZGI aims to mitigate the influence of tree canopies by isolating NDVI values solely attributable to grass cover. By incorporating phenological characteristics of forest trees and grass species, the ZGI outperforms NDVI in mapping grass-covered areas crucial for the study region’s fire susceptibility assessment. Results demonstrate a substantial overlap between ZGI-based maps and recorded fire occurrences, validating the efficacy of the index in fire susceptibility estimation.

Dynamic design of hazardous chemical storage in high-rise R&amp;D building based on numerical simulation

With the increasing demand for research on hazardous chemicals in high-rise research and development (R&amp;D) buildings, the use of a large number of flammable chemicals increases the fire risk significantly. Therefore, the research on the storage of hazardous chemicals in high-rise R&amp;D buildings becomes more and more important. This paper presents a maximum dynamic design method for hazardous chemical storage. A set of fire risk indicators has been established in the research, including temperature, carbon monoxide concentration, and visibility. The representative hazardous chemicals are selected through the investigation of the distribution and quantity of hazardous chemicals. Through numerical simulation of high-rise building fire with different stocks of dangerous chemicals and evacuation simulation of different fire floors, the time when the risk factors of fire occurrence reach the critical threshold of human body (ASET) and the necessary time for human evacuation (RSET) are obtained. By adjusting the amount of hazardous chemicals stored, the time to reach the hazardous critical threshold (RSET) and the time to evacuate (ASET) are changed accordingly. When RSET &lt; ASET, the storage capacity of hazardous chemicals in the high-rise R&amp;D building meets the target requirements. In the case study, ethanol was selected as the representative substance, and numerical simulation and example calculation were carried out on a 16-story R&amp;D building. The amount of ethanol stored on each floor was calculated. Among them, the eighth layer has the largest ethanol storage capacity (1.81 L/m2) and the ninth layer has the smallest ethanol storage capacity (0.89 L/m2).

CLIMATE ELEMENTS AND FIRE HAZARD INDEXES IN SANTA CATARINA

Forest fires have caused significant environmental and economic damage over the years. In this context, efforts have been made to establish relationships between climate and control measures for fire occurrence. Assigning a fire danger index based on meteorological conditions is a way to provide tools for organizations responsible for forest fire prevention and combat. In this study, computational and statistical techniques were employed to analyze the performance of three fire hazard indices: the Monte Alegre Formula, the Altered Monte Alegre Formula, and the Ängstrom Risk Factor, for the state of Santa Catarina, located in the Southern region of Brazil. The study and validation period covered from 2010 to 2022, using variables such as precipitation, relative humidity, wind speed, and air temperature, with values obtained from the National Institute of Meteorology (INMET), encompassing 23 meteorological gauges. These data were compared with fire detection records obtained from satellite imagery and provided by the National Institute for Space Research (INPE). After analyzing the data using various statistical techniques, it became evident that both Monte Alegre Formulas exhibited a high accuracy rate for days with detected fire occurrences, with over 80% accuracy, compared to only around 35% for Ängstrom. Concerning the overall number of days computed, Ängstrom achieved a higher accuracy rate (approximately 85%), compared to values around 50% for both Monte Alegre Formulas.

How do invasive predators and their native prey respond to prescribed fire?

Fire shapes animal communities by altering resource availability and species interactions, including between predators and prey. In Australia, there is particular concern that two highly damaging invasive predators, the feral cat (Felis catus) and European red fox (Vulpes vulpes), increase their activity in recently burnt areas and exert greater predation pressure on the native prey due to their increased exposure. We tested how prescribed fire occurrence and extent, along with fire history, vegetation, topography, and distance to anthropogenic features (towns and farms), affected the activity (detection frequency) of cats, foxes, and the native mammal community in south-eastern Australia. We used camera traps to quantify mammal activity before and after a prescribed burn and statistically tested how the fire interacted with these habitat variables to affect mammal activity. We found little evidence that the prescribed fire influenced the activity of cats and foxes and no evidence of an effect on kangaroo or small mammal (<800 g) activity. Medium-sized mammals (800-2000 g) were negatively associated with prescribed fire extent, suggesting that prescribed fire has a negative impact on these species in the short term. The lack of a clear activity increase from cats and foxes is likely a positive outcome from a fire management perspective. However, we highlight that their response is likely dependent upon factors like fire size, severity, and prey availability. Future experiments should incorporate GPS-trackers to record fine-scale movements of cats and foxes in temperate ecosystems immediately before and after prescribed fire to best inform management within protected areas.

FSNet: Enhancing Forest-Fire and Smoke Detection with an Advanced UAV-Based Network

Forest fires represent a significant menace to both the ecological equilibrium of forests and the safety of human life and property. Upon ignition, fires frequently generate billowing smoke. The prompt identification and management of fire sources and smoke can efficiently avert the occurrence of extensive forest fires, thereby safeguarding both forest resources and human well-being. Although drone patrols have emerged as a primary method for forest-fire prevention, the unique characteristics of forest-fire images captured from high altitudes present challenges. These include remote distances, small fire points, smoke targets with light hues, and complex, ever-changing background environments. Consequently, traditional target-detection networks frequently exhibit diminished accuracy when handling such images. In this study, we introduce a cutting-edge drone-based network designed for the detection of forest fires and smoke, named FSNet. To begin, FSNet employs the YOCO data-augmentation method to enhance image processing, thereby augmenting both local and overall diversity within forest-fire images. Next, building upon the transformer framework, we introduce the EBblock attention module. Within this module, we introduce the notion of “groups”, maximizing the utilization of the interplay between patch tokens and groups to compute the attention map. This approach facilitates the extraction of correlations among patch tokens, between patch tokens and groups, and among groups. This approach enables the comprehensive feature extraction of fire points and smoke within the image, minimizing background interference. Across the four stages of the EBblock, we leverage a feature pyramid to integrate the outputs from each stage, thereby mitigating the loss of small target features. Simultaneously, we introduce a tailored loss function, denoted as Lforest, specifically designed for FSNet. This ensures the model’s ability to learn effectively and produce high-quality prediction boxes. We assess the performance of the FSNet model across three publicly available forest-fire datasets, utilizing mAP, Recall, and FPS as evaluation metrics. The outcomes reveal that FSNet achieves remarkable results: on the Flame, Corsican, and D-Fire datasets, it attains mAP scores of 97.2%, 87.5%, and 94.3%, respectively, with Recall rates of 93.9%, 87.3%, and 90.8%, respectively, and FPS values of 91.2, 90.7, and 92.6, respectively. Furthermore, extensive comparative and ablation experiments validate the superior performance of the FSNet model.

Building Characteristics and Incompatible Land Uses as Drivers of Fire Disaster Risk

Risk-based land use planning is regarded as a controlling tool for reducing environmental disasters in urban settlements. However, unsustainable land use practices allow building and infrastructural developments in fire disaster-risk areas. Losses to urban fires are largely dependent on building and land use characteristics across settlements. This paper examines the underlying characteristics that increase fire risks in urban buildings by analyzing the relationship between land use patterns, types of building use, and fire disaster risk in the metropolitan area of Ibadan, Nigeria. Data were collected through a field survey using a structured questionnaire supplemented by focus group discussion. A multistage sampling procedure was used to select 88 neighborhoods and 1,803 building occupants were selected using systematic random sampling. Findings indicate incompatible land use developments in many areas of the city. The study revealed that one out of five buildings were not accessible to fire-fighting vehicles owing to inadequate road widths (43.7%) and pot-holes (39.8%). One out of seven buildings (14.1%) had no water supply. About 10% of the buildings had emergency doors on permanent lock. About half of the building occupants relied on power supply sources that increased risks of fire occurrence (52%) and 49.4% used unsustainable waste disposal methods that threatened fire safety. The paper concluded that land use attributes, such as accessibility, compatibility of uses, water supply sources, sources of electricity supply and solid waste disposal method, and building characteristics, such as burglar proofs, nature of entrance/exit, type of lighting fuel are factors that are relevant in fire disaster risk analysis.

Temporal Characterization of Fire Accidents in China

Based on the fire accident statistics, this study uses the seasonal index method to comprehensively analyze the temporal pattern of fire occurrence in China from 2010 to 2015. The study scientifically reflects that the number of fire occurrences has an extremely close relationship with time, and summarizes the temporal pattern and characteristics of fire occurrences. This study concludes that the high incidence of large fire accidents in China occurs in the time period of 3:00 a.m. to 5:00 a.m. The probability distribution model of the average total number of accidents in each time period was established and the Kolmogorov test was carried out, which shows that the time characteristics of fire accidents in China can be understood with the probability distribution model as a whole. The results of the analysis of the time characteristics of the fire accidents can provide decision-making information for combating fire accidents.