Early Fire Research Articles

Alberta’s 2023 wildfires: context, factors, and futures

Wildfires burned an estimated 2.2 million hectares in Alberta in 2023. We describe key attributes of the fires relative to historical fires and fire seasons and offer a perspective on potentially influential factors. Thirty-six large fires ≥10 000 ha generated 95% of annual area burned. Individually, these fires exhibited sizes, fire weather, and behaviour consistent with historical fires; there were simply far more of them in 2023. Thirteen fires reported in early May were ignited by lightning and reached final sizes ≥10 000 ha, revealing a previously unrecognized threat. Historically, large lightning-ignited fires reported before mid-May occur just once per decade on average. Collectively, 18 large fires reported in early May coincided with drier conditions compared with 18 large fires reported after mid-May. Early May fire weather was also warmer and drier than historical weather. The early May fire group was a temporally concentrated outbreak in west-central Alberta and coincided with extreme potential rate of fire spread. Large fires reported after mid-May were intermittent through to September, concentrated in northern regions and coincided with extreme potential for fuel consumption. Individually, these two spatiotemporal modes of fire season severity (outbreak, intermittent) produced annual burned areas on par with historical extremes. Together, the 2023 multi-modal pattern of fire season severity amplified area burned far above anything previously recorded. Potential contributing factors include climate warming, hemispheric teleconnections, phenology and exhaustion of suppression resources. Implications for future fire seasons, research and management are discussed.

Impacts of early and repeated fires on microbial activities in the wooded savannahs of Burkina Faso

Bush fires which effects depend on their intensity as well as their season of occurrence are considered as a major disturbance of savanna ecosystems. The objectives of this research were to understand the changes induced on the properties of the soil by repeated early fires on microbial activities. The study was conducted on a factorial design installed on two (02) sites, namely the classified forest of Dindéresso located in the West (firsts fires were applied in 2010) and that of Tiogo located in the Center West (first fire were applied since 1992) of the country. Measurements have been carried out before, after the passage of the fire and during the rainy season on soils. The results obtained showed that the period during which the measurements were taken played a role in the microbial dynamics, thus showing the highest rates in the rainy season for the control (3.02 C g.m-2) and the burnt plots (3.63 C g.m-2) after fire. Microbial nitrogen is the element that varied considerably, going from 1.60 for the control to 1.74 N g.m-2 for the burnt plots for the Dindéresso site. Concerning Tiogo, the microbial biomass is higher on the burned plots (6.04 C g.m-2) and (6.01 C g.m-2) for the control in rainy season sampling.

Early forest flame and smoke detection based on improved feature extraction module with enhanced image processing inspired by YOLOV7

Background The forest environment is intricate and dynamic, with its depiction influenced by various factors such as geographical location, weather conditions, and capture angles. Relying solely on flame or smoke is insufficient for precise fire information. Aims This paper proposes a method for accurate detection on forest flame and smoke based on improved feature extraction module with enhanced image processing. Methods We a fusion-guided filtering image processing method and flame segmentation strategy to augment the quality of dataset. Additionally, an outstanding extraction backbone, incorporating ghost modules and decoupled fully connected (DFC) attention modules, is developed to increase the model’s receptive field. Furthermore, the ELAN-S neck with SimAM attention mechanism is introduced to fuse features from the backbone network, facilitating the extraction of shallow and deep-level semantic information. Key results Compared to YOLOV7, our model demonstrates superior performance with a 5% increase in mean average precision (mAP), a 4.3% increase in average precision for small objects (APS), and a 3–4% enhancement in other metrics. Conclusions The proposed model achieves a good balance between detection speed and detection accuracy. The improved model performs well in real forest fire detection scenarios. Implications In the early forest fire detection, the model considers both flame and smoke information to describe the fire situation, and effectively combines the semantic information of both for fire warning.

Development and Field Deployment of a ppb-Level SO2/NO2 Dual-Gas Sensor System for Agricultural Early Fire Identification.

Sulfur dioxide (SO2) and nitrogen dioxide (NO2) are chemical indicators of crop straw combustion as well as significant atmospheric pollutants. It is challenging to promptly detect natural "wildfires" during agricultural production, which often lead to uncontrollable and substantial economic losses. Moreover, both "wildfires" and artificial "straw burning" practices pose severe threats to the ecological environment and human health. Consequently, developing sensors capable of rapid and high-precision quantitative analysis of SO2/NO2 is essential and urgent for detecting early fires in agricultural activities. Here, we demonstrate an incoherent broadband cavity-enhanced absorption spectroscopy (IBBCEAS) sensing system utilizing a 366 nm ultraviolet light emitting diode, designed for real-time, high-precision monitoring of SO2 and NO2 and is used for early fire detection validation. The optical resonant cavity is constructed within a 60 mm cage system mechanical structure, achieving a maximum optical path length of nearly 2 km with a length of ∼460 mm. The output light carrying information about the species and concentration of the analyte molecules is coupled into the miniaturized grating spectrometer via a fiber, and continuous spectral fitting and concentration inversion are performed on the computer. We propose a spectral analysis and concentration inversion model based on an improved particle swarm optimization-support vector machine (IPSO-SVM) algorithm. By discrimination of the absorption spectral characteristics of SO2/NO2, we achieve superior prediction accuracy. Experimental results indicate that the detection limits of SO2 and NO2 under the optimized averaging time are 77.5 parts per billion by volume (ppbv) and 0.037 ppbv, respectively. The field deployment of the sensor in scenarios such as continuous outdoor air pollution monitoring, in situ combustion feature identification, and early fire mobile detection has demonstrated the superior reliability and sensitivity of this sensor system.

Fire return intervals and recruitment affect population growth rate of canopy trees in tall open forest in humid savanna

AbstractSavannas are the major biome in tropical regions of the globe, defined as sparsely wooded regions with a continuous herbaceous layer of mainly C4 grasses where rainfall is distinctly seasonal. Fire is a common feature of most savannas. The largest protected areas of savannas are found in sparsely populated monsoonal northcentral Australia with strong annual wet and dry seasons. The most common vegetation type is relatively intact, tall (<15 m), open forests where Eucalyptus canopy trees form the basic structure. Over the past half century, traditional indigenous fire regimes were largely replaced by contemporary fires where individual trees may experience fire as often as 3 out of 5 years. The potential for long‐term persistence of the canopy tree populations is an open question. A stage‐based population model of the canopy trees was previously developed to address this question, drawing on data from three decades of experimental field studies wherein the survival, growth, and reproduction of individual marked trees were recorded under different seasonal fires and understory types to produce transition matrices among eight life history stages, and used to calculate population growth rates (λ). Here, we apply that model to determine how λ varies across a range of fire return intervals from 1 to 12 years for both early and late dry season fires, in two different understory types. We also explore the sensitivity of λ to two key life history parameters: recruitment and seedling survival. Minimum fire return intervals of 2–5 years were generally required for λ ≥1 that would allow populations to persist; these were shorter with stochastic year‐to‐year timing of fires and with higher recruitment rates. Uniquely, under certain conditions, there was also a maximum fire return interval above which λ <1, creating a “window” of fire return intervals that allowed canopy tree populations to persist. Mechanisms underpinning results as well as implications for savanna structure, alternate states, cyclical dynamics, future research, and management by fire are discussed.

A Multi-Scale Approach to Early Fire Detection in Smart Homes

In recent years, advancements in smart home technologies have underscored the need for the development of early fire and smoke detection systems to enhance safety and security. Traditional fire detection methods relying on thermal or smoke sensors exhibit limitations in terms of response time and environmental adaptability. To address these issues, this paper introduces the multi-scale information transformer–DETR (MITI-DETR) model, which incorporates multi-scale feature extraction and transformer-based attention mechanisms, tailored specifically for fire detection in smart homes. MITI-DETR achieves a precision of 99.00%, a recall of 99.50%, and a mean average precision (mAP) of 99.00% on a custom dataset designed to reflect diverse lighting and spatial conditions in smart homes. Extensive experiments demonstrate that MITI-DETR outperforms state-of-the-art models in terms of these metrics, especially under challenging environmental conditions. This work provides a robust solution for early fire detection in smart homes, combining high accuracy with real-time deployment feasibility.

SETD3-mediated histidine methylation of MCM7 regulates DNA replication by facilitating chromatin loading of MCM.

The minichromosome maintenance complex (MCM) DNA helicase is an important replicative factor during DNA replication. The proper chromatin loading of MCM is a key step to ensure replication initiation during S phase. Because replication initiation is regulated by multiple biological cues, additional changes to MCM may provide better understanding towards this event. Here, we report that histidine methyltransferase SETD3 promotes DNA replication in a manner dependent on enzymatic activity. Nascent-strand sequencing (NS-seq) shows that SETD3 regulates replication initiation, as depletion of SETD3 attenuates early replication origins firing. Biochemical studies reveal that SETD3 binds MCM mainly during S phase, which is required for the CDT1-mediated chromatin loading of MCM. This MCM loading relies on histidine-459 methylation (H459me) on MCM7 which is catalyzed by SETD3. Impairment of H459 methylation attenuates DNA synthesis and chromatin loading of MCM. Furthermore, we show that CDK2 phosphorylates SETD3 at Serine-21 during the G1/S phase, which is required for DNA replication and cell cycle progression. These findings demonstrate a novel mechanism by which SETD3 methylates MCM to regulate replication initiation.

임베디드 플랫폼을 위한 화재 조기 감지 시스템: 오경보 최소화를 위한 딥러닝 접근 방식

임베디드 플랫폼을 위한 화재 조기 감지 시스템: 오경보 최소화를 위한 딥러닝 접근 방식

Real-Time Fire Classification Models Based on Deep Learning for Building an Intelligent Multi-Sensor System

Fire detection systems are critical for mitigating the damage caused by fires, which can result in significant annual property losses and fatalities. This paper presents a deep learning-based fire classification model for an intelligent multi-sensor system aimed at early and reliable fire detection. The model processes data from multiple sensors that detect various parameters, such as temperature, humidity, and gas concentrations. Several deep learning architectures were evaluated, including LSTM, GRU, Bi-LSTM, LSTM-FCN, InceptionTime, and Transformer. The models were trained on data collected from controlled fire scenarios and validated for classification accuracy, loss, and real-time performance. The results indicated that the LSTM-based models (particularly Bi-LSTM and LSTM) could achieve high classification accuracy and low false alarm rates, demonstrating their effectiveness for real-time fire detection. The findings highlight the potential of advanced deep-learning models to enhance the reliability of sensor-based fire detection systems.

Vegetation dynamics after fire in the Brazilian Campo Rupestre: Effects on native plant communities and flower harvesting

We investigated the impact of both early and late fires on native plant communities of the Campo Rupestre in the Sempre-Vivas National Park (PNSV, Brazil). Everlasting flower harvesters use late fires to stimulate flowering, while park managers have been implementing early fires to reduce flammable biomass and, therefore, the risk of wildfires. We aimed to explore the effects of fire on species composition, vegetation cover, and plant and flower stalks height to evaluate post-fire vegetation recovery, especially considering Comanthera species, which are highly valued by flower harvesters. The experimental design involved two areas (A1 and A2) in PNSV from May/2019 to January/2021. We installed eight 50 × 50 m plots in each area, being half submitted to experimental burnings and the other half unburned (control plots). A1 experienced early fire in May, and A2 a late fire in September. Initial phytosociological surveys revealed differences between A1 and A2, therefore, fire effects were treated separately for each area. In both A1 and A2, fire initially impacted species richness and abundance but the effect dissipated over time, with vegetation becoming similar to unburned plots. Fire also affected vegetation cover, which returned to its original condition within a year, influenced by seasonality and plant phenology. Plant communities experienced a temporary reduction in height (∼4 cm) in the months following fire, and recovered in the subsequent rainy season. However, a tendency for smaller plants persisted, and the average height of flower stalks took almost two years to fully recover. In general, the late fire conducted in A2 led to a slower recovery trajectory. These findings indicate rapid post-fire biomass recovery and minimal impact on plant species composition, highlighting the resilience of Campo Rupestre to single fires. Further studies are crucial to understand plant response to fires at different fire frequencies.

Influence of the Periodicity of Ground Fires in Middle-Аged Pine Forests on the Combustible Materials Stocks

Significant areas of forest are annually exposed to fires in the Krasnoyarsk Territory. Fires transform forest areas and change the stocks and structure of combustible materials. Information on the dynamics of forest fuel accumulation can serve as a basis for managing the pyrogenic resistance of forest areas. The purpose of the research is to assess the impact of early spring ground fires of different frequencies on the structure and stocks of combustible materials in forests. Experimental fires were carried out in middle-aged mixed-grass-green-moss pine forests in the forest-steppe zone of the Krasnoyarsk Territory. Experimental fires were simulated with different periodicities (annual, with an interval of 2–3 years and a singular one). Regardless of periodicity, the experimental fires led to an increase of combustible materials stocks in 2–3 years after fires. Repeated burnings reduced the stocks of combustible materials close to pre-fire values. The structure of combustible materials has changed, which has led to a decrease in potential flammability of pine forests. The proportion of needles and bark in litter has decreased, and the proportion of cones has increased. Fires led to a redistribution of deadwood classes towards larger diameters. During the research period (2014–2018), the largest reserve of forest fuel loads was noted on the plot with triennial fires (68.2 t ha–1).

Experimental Study on Early Fire Smoke Characteristics in a High-Volume Space: A Fire Detection Perspective

High-volume space structures are characterized by high combustible loads, rapid fire development, difficulty in firefighting, and potential building collapse risks, making early fire detection particularly crucial. The effectiveness of early fire detection technologies relies on their ability to adapt to the characteristics of smoke-dominant combustion products in the protected space. However, there is a lack of targeted research on the characteristics of the smoke generated during the early low-power stages of fires in high-volume spaces, which has not supported the development of early fire detection technologies for such environments. To address this, this paper presents an experimental study that, for the first time, collects data on fire smoke parameters such as temperature, velocity, concentration, and particle size at heights ranging from 6.5 m to 18.5 m using lightweight sensors. The study analyzes the characteristic and correlations of these parameters and their impact on early fire detection in high-volume spaces for the first time, presenting variation patterns in the plume velocity and particle size distribution of early fire smoke with height. It identifies three patterns of particle size distribution, contrasting with previous studies, and offers a qualitative explanation for these findings. This research enhances the understanding of early fire smoke signals in large spaces and offers valuable insights for developing more accurate and efficient fire detection strategies and technologies.

Early polar-orbiting satellite-based fire remote sensing during the 1960s

ABSTRACT Satellite-based sensors such as the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS) now routinely provide a detailed picture of global fire activity on a daily basis. These systems explicitly included biomass burning as a phenomenon of interest, and their design was heavily influenced by experience gained with the comparatively primitive Advanced Very High Resolution Radiometer (AVHRR) first deployed on the TIROS-N satellite in the late 1970s. The AVHRR is generally recognized as the earliest spaceborne sensor to provide a middle infrared (MWIR) channel sensitive to actively burning vegetation fires and other high-temperature, sub-pixel thermal anomalies, forming the basis of countless fire studies before higher quality MODIS data became available. Here, we consider in the context of active fire monitoring a sensor much older than the AVHRR, namely the High Resolution Infrared Radiometer (HRIR) on board NASA’s Nimbus-I, -II, and -III satellites, launched in 1964, 1966, and 1969, respectively. This early imaging sensor had a single MWIR channel primarily intended for night-time meteorological observations (a complementary daytime near-infrared channel was added for Nimbus-III) that could potentially serve as a source of pre-AVHRR active fire data. To this end, we inspected night-time HRIR imagery acquired over the western United States and found unambiguous evidence of multiple confirmed wildfires in 1966. In addition, we found strong evidence of actively burning fires in Africa, specifically the Democratic Republic of the Congo, western Zambia, and northern Namibia, and in northern Australia, commensurate with their fire seasons. Our findings indicate that it was actually the HRIR that acquired the earliest MWIR imagery that could be used for identifying biomass burning from space, well over a decade before the first AVHRR was deployed. We found in addition that the retrieval of fire radiative power (FRP) from higher dynamic-range HRIR observations was potentially possible pending refined calibration.

Early Smoke Recognition Algorithm for Forest Fires

Forest fires require rapid and precise early smoke detection to minimize damage. This study focuses on employing smoke recognition methods for early warning systems in forest fire detection, identifying smoke as the primary indicator. A significant hurdle lies in the absence of a large-scale dataset for real-world early forest fire smoke detection. Early smoke videos present characteristics such as smoke plumes being small, slow-moving, and/or semi-transparent in color, and include images where there is background interference, posing critical challenges for practical recognition algorithms. To address these issues, this paper introduces a real-world early smoke monitoring video dataset as a foundational resource. The proposed 4D attention-based motion target enhancement network includes an important frame sorting module which adaptively selects essential frame sequences to improve the detection of slow-moving smoke targets. Additionally, a 4D attention-based motion target enhancement module is introduced to mitigate interference from smoke-like objects and enhance recognition of light smoke during the initial stages. Moreover, a high-resolution multi-scale fusion module is presented, incorporating a small target recognition layer to enhance the network’s ability to detect small smoke targets. This research represents a significant advancement in early smoke detection for forest fire surveillance, with practical implications for enhancing fire management.

Towards early forest fire detection and prevention using AI-powered drones and the IoT

Wildfires are a significant natural hazard, resulting in financial losses, human deaths, and environmental damage. Due to the rising severity and frequency of wildfires, wildfire management and detection have recently received increased attention worldwide. Monitoring potential risk areas and early fire detection are critical factors for shortening the reaction time and reducing the potential damage. Conventional wildfire detection techniques like satellite imaging and remote camera-based sensing need more latency and low reliability. To tackle these limitations, this paper proposes a novel airborne UAV-based IoT (UIoT) system for wildfire sensing, detection, and extinguishing. It presents the design of low-cost and low-maintenance fire-detecting IoT nodes for large-scale deployment. It also proposes, investigates, and reports on several connectivity architectures using the LoRaWAN protocol for UIoT systems. We geolocate forest trees in a terrain-mapping exercise for precise fire localization. We then deploy an autonomous drone with a visual camera that utilizes our novel classification network to detect the presence of fire followed by fire detection and particle filter-based tracking to center fire at the center of the frame. We use a cloud back-end server for monitoring, analysis, and reporting. Our results show that our low-cost and long-range IoT nodes accurately detect fire within 1-5 min after fire ignition. Our fire classification network achieved an accuracy of 99.46% and a mean average precision of 99.64%. Numerical results suggest that the proposed UAV-IoT-based fire classification offers a fast and reliable wildfire recognition and extinguishing solution while minimizing power consumption and increasing the battery lifespan.

Sequential Learning of Flame Objects Sorted by Size for Early Fire Detection in Surveillance Videos

Sequential Learning of Flame Objects Sorted by Size for Early Fire Detection in Surveillance Videos

Design of an Internet of Things (IoT) based Early Fire Prevention System Due to LPG Gas Leakage Using NodeMCU ESP8266 Module

Design of an Internet of Things (IoT) based Early Fire Prevention System Due to LPG Gas Leakage Using NodeMCU ESP8266 Module

DESIGN OF A FIRE DETECTION SYSTEM USING FIRE AND SMOKE SENSORS BASED ON ARDUINO MICROCONTROLLER

Fires very often occur due to gas leaks, careless throwing of cigarette butts or electrical short circuits which result in the fire getting bigger, causing a lot of losses. Many fire cases are discovered when the fire has already occurred and few can be detected from the start. The research aims to develop an effective system for detecting fires by integrating fire and smoke sensors, as well as providing notifications via the Blynk platform. This research is research and development (R&D) using a prototype development model. The results of the functionality testing obtained a 100% presentation which shows that the prototype is suitable for use. Tests were carried out using three levels of fire levels and various distances, showing good response from the fire sensor with a consistent incoming notification speed of around 2-3 seconds. The results of fire measurements using the MQ-2 smoke sensor show accuracy in detecting smoke concentration levels, with timely notification of potential fires. This research has produced a detection tool that has the ability to detect early fire in a room which can provide warning notifications that are integrated with Internet of Things (IoT).

Rational Design of Amorphous Carbon-Coated Laminar-Structured Wood for Integrating Repeatable Early Fire Detection and High-Temperature Affordable Flexible Pressure Sensing in One System.

High-temperature affordable flexible polymer-based pressure sensors integrated with repeatable early fire warning service are strongly desired for harsh environmental applications, yet their creation remains challenging. This work proposed an approach for preparing such advanced integrated sensors based on silver nanoparticles and an ammonium polyphosphate (APP)-modified laminar-structured bulk wood sponge (APP/Ag@WS). Such integrated sensors demonstrated excellent fire warning performance, including a short response time (minimum of 0.44 s), a long-lasting alarm time (>750 s), and reliable repeatability. Moreover, it achieved high-temperature affordable flexible pressure sensing that exhibited an almost unimpaired working range of 0-7.5 kPa and a higher sensitivity (in the low-pressure range, maximum to 226.03 kPa-1) after fire. The high stability was attributed to reliable structural elasticity, and the wood-derived amorphous carbon is capable of repeatable fire warnings. Finally, a Ag@APP/WS-based wireless fire alarm system that realized reliable house fire accident detection was demonstrated, showing great promise for smart firefighting application.

Small‐scale fires interact with herbivore feedbacks to create persistent grazing lawn environments

Abstract Fire‐herbivory feedbacks strongly influence the formation of grazing lawns in savanna ecosystems. Preliminary findings suggest that small‐scale (<25 ha) fires can engineer grazing lawns by concentrating herbivores on the post‐burn green flush; however, the persistence of such grazing lawns over the longer term and without repeated fire is unknown. We used high‐resolution Light Detection and Ranging (LiDAR) to investigate the long‐term effects of fire manipulation on short grass structure (height, cover, volume and spatial continuity) and grazing lawn establishment in Kruger National Park, South Africa. We analysed the effects of fire exclusion and experimental burns applied over a 7‐year period (2013–2019) followed by a 1‐year cessation of burning at varying spatial scales during the early and late dry seasons. Fires contributed a fourfold increase in short grass cover, regardless of fire season or size. The distribution of grass height differed significantly between fire‐induced grazing lawns and recently unburnt parts of the landscape where controlled fires were excluded over the experimental period. The volume (corresponding to bulk density) of short grass on the landscape responded strongly to fires, with grass volume <20 cm in height increasing with both early and late dry season fires. Early dry season fires caused larger and more homogeneous short grass patches. Furthermore, early dry season fires were more influential in increasing the cover of the shortest grass height class (1–5 cm). Synthesis and applications. Our results demonstrate that fire‐induced grazing lawns can persist over the longer term, even when fires are no longer applied, leading to the creation of vertical and horizontal heterogeneity in the grass layer. Small‐scale fires, therefore, represent a feasible management approach to expanding grazing lawn extent, potentially benefiting grazer coexistence and diversity.