Plume‐Coupled Long‐Range Spotting Drove the Explosive Spread of the 2018 Camp Fire

The aim of this paper is to propose a more practical mountain fire spread model for fire behavior prediction and management in Southwest forest area of China. These areas are covered mainly with spatial heterogeneous flammable forest and are characterized by undulating terrain and steep slopes. This model can produce more accurate fire propagation maps by combining CA (Cellular Automaton) framework with Wang Zhengfei fire physical velocity model in fine scale. Considering the inherent uncertainties of fire propagation, the model has been built on multi-dimension geophysical and environmental components and also sound knowledge of fire spread physical mechanism. Regarding small fuel patches as spatial homogenous cells, this approach makes it easier to generate higher level complex fire behavior maps from CA simple local rules and local behavior integrated with high resolution vegetation images, fine scale terrain maps and surface wind field. Because the model focuses primarily on the study of surface fire front propagation behavior, it attempts to simplify complex fuel modeling. Additionally, this Wang-Geophysical-CA model is able to analyze the time series spatial pattern of fire-front spread and model local behavior instead of the final fire spread pattern of the conventional approach. In this work, not only single influence verification tests have been made, but also simulation tests with multiple influences are carried out to demonstrate the capability of the model with fine scale vegetation maps, surface wind field, terrain, moisture content and man-made structures. Consequently, it is believable that the model predictions are in good agreement with experimental data for steady-state fire simulation. The proposed model helps to gain a greater understanding of the fire front spread local behavior and can quickly generate a sequence of complex fire front contours. It enables local managers to plan practical fire prevention activities in Southwest forest area of China as well as improve fire management skills, and will enhance the effectiveness of fire fighting strategies.

Experimental study and numerical simulation for a storehouse fire accident

In this study, we investigate a novel application of the photogrammetric monoplotting technique for assessing wildfires. We demonstrate the use of the software program WSL Monoplotting Tool (MPT) to georeference operational oblique aerial wildfire photographs taken during airtanker response in the early stages of fire growth. We located the position of the fire front in georeferenced pairs of photos from five fires taken 31–118 min apart, and calculated the head fire spread distance and head fire rate of spread (HROS). Our example photos were taken 0.7 to 4.7 km from fire fronts, with camera angles of incidence from −19° to −50° to image centre. Using high quality images with detailed landscape features, it is possible to identify fire front positions with high precision; in our example data, the mean 3D error was 0.533 m and the maximum 3D error for individual fire runs was less than 3 m. This resulted in a maximum HROS error due to monoplotting of only ~0.5%. We then compared HROS estimates with predictions from the Canadian Fire Behavior Prediction System, with differences mainly attributed to model error or uncertainty in weather and fuel inputs. This method can be used to obtain observations to validate fire spread models or create new empirical relationships where databases of such wildfire photos exist. Our initial work suggests that monophotogrammetry can provide reproducible estimates of fire front position, spread distance and rate of spread with high accuracy, and could potentially be used to characterize other fire features such as flame and smoke plume dimensions and spotting.

Introduction Extreme wildfires are increasingly prevalent worldwide, driving significant forest area loss and severe environmental and socioeconomic impacts (Cunningham et al. 2024). The Mediterranean, in particular, is projected to face heightened fire risks due to climate change-induced drier conditions and lower fuel moisture (de Rivera et al. 2020). However, the drivers of extreme wildfires remain poorly understood. Current fire models, typically calibrated on global fire datasets, are primarily designed to estimate annual total burned areas and struggle to capture the unique behaviours of extreme wildfires (Forrest et al. 2024). Furthermore, correlation-based approaches, which dominate current modelling efforts, may fail to identify the underlying causal drivers of these events and are poorly suited for extrapolation to changing conditions. Causal discovery methods, which aim to identify cause-and-effect relationships from observational data, offer a promising pathway to uncover the mechanisms driving extreme wildfires. While increasingly applied in environmental sciences, their use in wildfire prediction remains limited (de Rivera et al. 2020, Zhang et al. 2024, Zhao et al. 2024).This study will use causal discovery to identify key drivers of extreme wildfire in the Mediterranean, and further integrate the causal graphs into a stand-alone model of wildfire spread. This approach aims to move beyond correlation-based models, improve our understanding of extreme wildfire behaviour and inform more robust mitigation strategies. Study Area and Data We will use the Mesogeos dataset (Kondylatos et al. 2023), designed for wildfire modelling in the Mediterranean region. Spanning 17 years (2006–2022) at a 1 km² spatial and daily temporal resolution, it includes meteorological variables (e.g., temperature, wind speed), vegetation indices (e.g., NDVI, LAI), and human activity indicators (e.g., population density, road proximity). Wildfire data include MODIS fire ignitions and burned areas from EFFIS. Methods Extreme Wildfire Definition and Sampling In this study, we define extreme wildfires as those that are exceptionally large in size. To identify these events, we will first extract the final burned areas associated with each fire ignition recorded in the Mesogeos dataset. Since the classification of large fires is inherently subjective and varies by region, we will adopt a data-driven approach based on an absolute quantitative threshold. Specifically, we will define extreme wildfires as those exceeding the 99th percentile of fire sizes, though this threshold may be adjusted to align with extreme fire events documented in national fire reports. While this method provides a straightforward and reproducible way to define extreme events, we acknowledge its limitations. Future work will refine this approach by incorporating region-specific thresholds and additional contextual factors to improve geographic relevance. Phase I: Causal Discovery Using local variables from Mesogeos, averaged over final burned areas and lagged to time t, we will estimate causal graphs for extreme events via Python’s Tigramite library with the PCMCI method (Runge et al. 2019). PCMCI detects time-lagged causal associations in large nonlinear datasets through iterative conditional independence testing. To ensure robustness, we will assess graph stability across hyperparameters and selected drivers, and validate graphs through expert knowledge. Phase II: Causal Fire Spread Model We will develop a fire spread model incorporating causal mechanisms from Phase I. This model will integrate spatiotemporal fire dynamics, causal dependencies constraining fire spread, and dynamic weather and fuel inputs. By explicitly modeling causal interactions, it aims to improve early warning systems and risk assessments under future climate scenarios. The causal model’s performance will be benchmarked against statistical models to evaluate its predictive accuracy and robustness. Expected Results We expect that the data-driven approach proposed in this study will enhance the predictability of extreme wildfires by reducing confounding effects and capturing key drivers of extreme fire events. Compared to purely statistical approaches, incorporating causal structures should lead to more reliable predictions, particularly in out-of-sample applications or under changing environmental conditions. Furthermore, the causal fire spread model will provide insights into how climate, vegetation, and anthropogenic factors interact to drive fire spread, supporting fire prevention and mitigation strategies.

We present and demonstrate the application of a systematic methodology for predicting fire spread and growth and for a relative fire hazard classification of materials for any scale and fire environment. This methodol ogy consists of three steps: (1) select laboratory test methods to perform flam mability measurements; (2) based on these measurements, obtain key flamma bility material properties which are precisely defined in this work; and (3) use these properties in a mathematical model of fire spread and growth to predict fire hazards. The complementary test methods we have selected and used are: (a) a general flammability test apparatus (such as NIST or FMRC) [1,2] modified to also provide pyrolysis measurements in an inert N2 atmosphere; (b) the Limited Oxygen Index (LOI) apparatus, which is used here as a tool for ob taining properties needed for creeping flame spread and extinction, including vitiated environments; and (c) a solid material smoke-point height apparatus [8], which is used to characterize the smokiness of the burning material needed to determine the radiation and smoke yield for arbitrary fire situations (wall fires, pool fires or ceiling fires) [8]. The use and proper interpretation of the Limited Oxygen Index apparatus can replace the LIFT [10] apparatus for deter mining in a more accurate and direct way the material properties required for creeping (vertical downward, lateral, horizontal) flame spread. The present methodology has been compared well with experiments in this work and else where [9], and it has been used to predict critical conditions for fire spread [11], not empirically as it is usually done, but based on first principles of fire spread, fire growth and burning, together with material flammability properties syste matically deduced from small-scale test measurements.

To demonstrate the usefulness of active remote-sensing systems in observing forest fire plume behavior, we studied two fires, one using a 3.2-cm-wavelength Doppler radar, and one more extensively, using Doppler lidar. Both instruments observed the kinematics of the convection column, including the presence of two different types of rotation in the columns, and monitored the behavior of the smoke plume. The first fire, a forest fire that burned out of control, was observed by the Doppler radar during late-morning and afternoon hours. Strong horizontal ambient winds produced a bent-over convection column, which the radar observed to have strong horizontal flow at its edges and weaker flow along the centerline of the plume. This velocity pattern implies that the column consisted of a pair of counterrotating horizontal vortices (rolls), with rising motion along the centerline and sinking along the edges. The radar tracked the smoke plume for over 30 km. It also provided circular depolarization ratio measurements, which gave information that the scattering particles were mostly flat or needle shaped as viewed by the radar, perhaps pine needles or possibly flat ash platelets being viewed edge on. The second fire, observed over a 5-h period by Doppler lidar, was a prescribed forest fire ignited in the afternoon. During the first hour of the fire the lidar observed many kinematic quantities of the convection column, including flow convergence and anticyclonic whole-column, rotation of the nearly vertical column, with a vorticity of approximately 10−2 s−1 and an estimated peak vertical velocity w of 1 5 m s−1. After the first hour ambient meteorological conditions changed, the whole-column rotation ceased, and the convection column and smoke plume bent over toward the lidar in stronger horizontal flow. At two times during this later stage, w was estimated to be 24 and 10 m s−1. Lidar observations show that the smoke plume of this second fire initially went straight up in the convection column to heights of over 2 km, so most of the smoke was injected into the atmosphere above the unstable, afternoon, convective boundary layer, or mixed layer. Later, as the horizontal winds increased, a larger friction of the smoke remained in the mixed layer. Finally, very late in the afternoon, after ignitions had ceased and the fire was smoldering, almost all of the smoke remained within the mixed layer. These analyses show that lidar and radar can provide valuable three-dimensional datasets on kinematic quantities and smoke distribution in the vicinity of fires. This kind of information should be of great value in understanding and modeling convection-column dynamics and smoke-plume behavior.

Wildland surface fire spread: Mechanism transformation and behavior transition

Fast-moving dire wildfire evacuation simulation

In spite of considerable effort to predict wildland fire behaviour, the effects of firebrand lift-off, the ignition of resulting spot fires and their effects on fire spread, remain poorly understood. We developed a cellular automata model integrating key mathematical models governing current fire spread models with a recently developed model that estimates firebrand landing patterns. Using our model we simulated a wildfire in an idealised Pinus ponderosa ecosystem. Varying values of wind speed, surface fuel loading, surface fuel moisture content and canopy base height, we investigated two scenarios: (i) the probability of a spot fire igniting beyond fuelbreaks of various widths and (ii) how spot fires directly affect the overall surface fire’s rate of spread. Results were averages across 2500 stochastic simulations. In both scenarios, canopy base height and surface fuel loading had a greater influence than wind speed and surface fuel moisture content. The expected rate of spread with spot fires occurring approached a constant value over time, which ranged between 6 and 931% higher than the predicted surface fire rate of spread. Incorporation of the role of spot fires in wildland fire spread should be an important thrust of future decision-support technologies.

Probabilistic techniques deal with the randomness of variables and reliability of safety system but their application in fire safety engineering is limited due to the lack of data related to real structures subjected to real fires. This can be overcome by analysis of national fire statistics provided by fire departments. Fire statistics databases are a collection of data from real structures subjected to real fires and provides an understanding of real effectiveness of different fire safety measures (i.e. compartmentation) which influence the spread and growth of fire, and ultimately their monetary consequence. The ability to understand the realistic responses of buildings in fire is the fundamental basis of British Standards PD 7974-7, which provides data to perform probabilistic risk assessments for fire. However, the current data presented by BS PD 7974-7:2003 (referred to as PD 7974-7 within this paper) was developed between 1966 and 1987. This research has used the USA fire statistic database of 2014 to recreate the tables present in the PD 7974-7, compare the results, and understand their evolution in time. The comparison between PD 7974-7 and the USA fire statistics introduced in this paper shows that modern fire frequency can be up to more than 10 times smaller than presented in PD 7974-7; area damage in m2 and spread of fire are linked to automatic extinguish systems effectiveness and greater in the USA fire statistics than predicted by PD 7974-7. This clearly demonstrate the need of updates to PD 7974-7 and feeds towards a better understanding of the robustness, and potentially the resilience, of real structures in fire.

Firebrands are recognized as a major source of wildland fuel ignition and a critical driver of fire spread in wildland–urban interface (WUI). This study experimentally examines the ignition behaviour of two widely present vegetative fuel beds in the WUI, pine needle and eucalyptus, when exposed to glowing firebrands under both no wind and wind conditions. Key ignition parameters, including fuel consumption rate, rate of spread, and flame development, were evaluated. Pine needle beds consistently exhibited more intense burning behavior than eucalyptus, with higher fuel consumption rates, faster fire spread, and greater flame heights. For the same fuel load, the average peak values of mass loss rate, rate of spread, and flame height in pine needle fuel beds were approximately 4.5, 1.5, and 1.85 times greater, respectively, than in eucalyptus. Increasing fuel load resulted in increased mass loss and flame height by factors of approximately 1.7 and 1.3 times, respectively, while reducing the rate of spread to about 0.9 times. A notable flame separation phenomenon was also observed during spot fire, where the flame front detached and subsequently created two flame zones. These findings highlight the importance of fuel structure in determining ignition intensity and fire spread under firebrand exposure. • Fuel type, load, and wind govern ignition and fire propagation dynamics • Pine needle beds burn faster with higher consumption rates, and greater flames • Once spot ignition was initiated in the fuel, it leads to sustained fire propagation • Fires in fuel beds with lower loads spread faster but produce smaller flames • Spot fires cause flame separation, forming two flame zones

A simple field experiment was conducted to measure and quantify fire–atmosphere interactions during a grass fire spreading up a hill under a moderate cross-slope wind. The observed fire intensity measured by passive radiometers and calculated sensible heat fluxes ranged between 90 and 120 kW m\(^{-2}\). Observations from this experiment showed that convective heat generated from the fire front was transported downwind in the lowest 2 m and the highest plume temperatures remained in this shallow layer, suggesting the fire spread was driven primarily by the advection of near-ignition temperature gases, rather than by radiation of the tilted flame. Fire-induced circulations were present with upslope flows occurring during the fire-front passage helping to transport heat up the slope and perpendicular to the fire front. A decrease in atmospheric pressure of 0.4 hPa occurred at the fire front and coincided with a strong updraft core of nearly 8 m s\(^{-1}\). These observations provide evidence that, even under moderately windy conditions, the pressure minimum in the fire remains rather close to the combustion zone and plume. The turbulence associated with the fire front was characterized by isotropic behaviour at 12.0 m above the ground, while less isotropic conditions were found closer to the ground due to higher horizontal variances associated with fire-induced flow at the fire front. From analysis of the turbulence kinetic energy budget terms, it was found that buoyancy production, rather than shear generation, had a larger contribution to the generation of turbulence kinetic energy, even during a highly sheared and moderate ambient wind.

Spotting ignition by lofted firebrands is a significant mechanism of fire spread, as observed in many large-scale fires. The role of firebrands in fire propagation and the important parameters involved in spot fire development are studied. Historical large-scale fires, including wind-driven urban and wildland conflagrations and post-earthquake fires are given as examples. In addition, research on firebrand behaviour is reviewed. The phenomenon of spotting fires comprises three sequential mechanisms: generation, transport and ignition of recipient fuel. In order to understand these mechanisms, many experiments have been performed, such as measuring drag on firebrands, analysing the flow fields of flame and plume structures, collecting firebrands from burning materials, houses and wildfires, and observing firebrand burning characteristics in wind tunnels under the terminal velocity condition and ignition characteristics of fuel beds. The knowledge obtained from the experiments was used to develop firebrand models. Since Tarifa developed a firebrand model based on the terminal velocity approximation, many firebrand transport models have been developed to predict maximum spot fire distance. Combustion models of a firebrand were developed empirically and the maximum spot fire distance was found at the burnout limit. Recommendations for future research and development are provided.

Geovisualization tools can supplement the statistical analyses of landscape-level wildfire behavior by enabling the discovery of nuanced information regarding the relationships between fire spread, topography, fuels, and weather. The objectives of this study were to develop and evaluate the effectiveness of geovisualization tools for analyzing wildfire behavior and specifically to apply those tools to study portions of the Thomas and Detwiler wildfire events that occurred in California in 2017. Fire features such as active fire fronts and rate of spread (ROS) vectors derived from repetitive airborne thermal infrared (ATIR) imagery sequences were incorporated into geovisualization tools hosted in a web geographic information systems application. This geovisualization application included ATIR imagery, fire features derived from ATIR imagery (rate of spread vectors and fire front delineations), growth form maps derived from NAIP imagery, and enhanced topographic rasters for visualizing changes in local topography. These tools aided in visualizing and analyzing landscape-level wildfire behavior for study portions of the Thomas and Detwiler fires. The primary components or processes of fire behavior analyzed in this study were ROS, spotting, fire spread impedance, and fire spread over multidirectional slopes. Professionals and researchers specializing in wildfire-related topics provided feedback on the effectiveness and utility of the geovisualization tools. The geovisualization tools were generally effective for visualizing and analyzing (1) fire spread over multidirectional slopes; (2) differences in spread magnitudes within and between sequences over time; and (3) the relative contributions of fuels, slope, and weather at any given point within the sequences. Survey respondents found the tools to be moderately effective, with an average effectiveness score of 6.6 (n = 5) for the visualization tools on a scale of 1 (ineffective) to 10 (effective) for postfire spread analysis and visualizing fire spread over multidirectional slopes. The results of the descriptive analysis indicate that medium- and fine-scale topographic features, roads, and riparian fuels coincided with cases of fire spread impedance and exerted control over fire behavior. Major topographic features such as ridges and valleys slowed, or halted, fire spread consistently between study areas. The relationships between spotting, fuels, and topography were inconclusive.

The paper deals with a Wireless Sensor Network (WSN) as a reliable solution for capturing the kinematics of a fire front spreading over a fuel bed. To provide reliable information in fire studies and support fire fighting strategies, a Wireless Sensor Network must be able to perform three sequential actions: 1) sensing thermal data in the open as the gas temperature; 2) detecting a fire i.e., the spatial position of a flame; 3) tracking the fire spread during its spatial and temporal evolution. One of the great challenges in performing fire front tracking with a WSN is to avoid the destruction of motes by the fire. This paper therefore shows the performance of Wireless Sensor Network when the motes are protected with a thermal insulation dedicated to track a fire spreading across vegetative fuels on a field scale. The resulting experimental WSN is then used in series of wildfire experiments performed in the open in vegetation areas ranging in size from 50 to 1,000 m2.