Wildfire Spread Research Articles

Effects of fuel characteristics on ember generation characteristics at branch-scales

Spot fires caused by lofted embers (i.e. firebrands) can be a significant factor in the spread of wildfires. Embers can be especially dangerous near the wildland–urban interface (WUI) because of the potential for the fire to be spread near or on structures. This work sought to identify how ember generation changes for different fuel characteristics and environmental conditions. Samples were burned in a heated wind tunnel and the time required for the ember(s) to form was measured using a DSLR camera. A factorial analysis of variance was used to determine the sensitivity of the time to generation to species, diameter, moisture content, fuel condition (i.e. dowel v. natural sample), crossflow temperature and crossflow velocity. Four fuel species were evaluated: Douglas fir (Pseudotsuga menziesii), western juniper (Juniperus occidentalis), ponderosa pine (Pinus ponderosa) and white oak (Quercus alba). The diameter of the samples had the greatest effect on the time to generation and the fuel species had the second greatest effect. The small diameter samples were relatively insensitive to changes in other parameters. Natural samples required significantly longer time than dowels to generate an ember. These conclusions suggest that fuel morphology is one of the most significant factors influencing ember generation.

Coupling wildfire spread and erosion models to quantify post-fire erosion before and after fuel treatments

Wildfires are known to change post-fire watershed conditions such that hillslopes can become prone to increased erosion and sediment delivery. In this work, we coupled wildfire spread and erosion prediction modelling to assess the benefits of fuel reduction treatments in preventing soil runoff. The study was conducted in a 68000-ha forest area located in Sardinia, Italy. We compared no-treatment conditions v. alternative strategic fuel treatments performed in 15% of the area. Fire behaviour before and after treatments was estimated by simulating 25000 wildfires for each condition using the minimum travel time fire-spread algorithm. The fire simulations replicated historic conditions associated with severe wildfires in the study area. Sediment delivery was then estimated using the Erosion Risk Management Tool (ERMiT). Our results showed how post-fire sediment delivery varied among and within fuel treatment scenarios. The most efficient treatment alternative was that implemented near the road network. We also evaluated other factors such as exceedance probability, time since fire, slope, fire severity and vegetation type on post-fire sediment delivery. This work provides a quantitative assessment approach to inform and optimise proactive risk management activities intended to reduce post-fire erosion.

Crowd-Sourced Wildfire Spread Prediction With Remote Georeferencing Using Smartphones

Wildfires are natural hazards with severe consequences worryingly worsening for many climate-change affected regions of our planet. Unfortunately, technologies that can provide real-time fire-line information, such as satellites, in-field sensors, and social media texts, exhibit low spatial/temporal resolution or cannot be deployed cost-effectively in widespread geographical areas. We present the design, development, and implementation of a novel software service, called CITISENS, which by exploiting commodity smartphone sensors allows ordinary citizens to easily georeference a fire-line in real-time and report its coordinates as they are photographing a wildfire. The location/orientation sensors and the camera are used to compute the view-ray of the smartphone, and a digital elevation model is employed to estimate the ray's intersection with the topography. We have tested the georeferencing accuracy obtained and it is to be on par with, or even better, than that of existing satellite wildfire hotspot services. When combined with FLogA, a flexible wildfire spread simulator we have also developed, CITISENS offers the following unique advantages: real-time prediction of burn probabilities, dynamic assimilation of citizen-reported hotspots into ongoing simulations for improved predictive accuracy, and decision support to issue citizen alarms based on the estimated time-dependent risk at their location due to an approaching wildfire.

IDENTIFYING AREAS AFFECTED BY FIRES IN SUMATRA BASED ON TIME SERIES OF REMOTELY SENSED FIRE HOTSPOTS AND SPATIAL MODELING

Wildfires threaten the environment not only at local scales, but also at wider scales. Rapid monitoring system to detect active wildfires has been provided by satellite remote sensing technology, particularly through the advancement on thermal infrared sensors. However, satellite-based fire hotspots data, even at relatively high temporal resolution of less than one-day revisit period, such as time series of fire hotspots collected from TERRA and AQUA MODIS, do not tell exactly if they are fire ignitions or fire escapes, since other factors like wind, slope, and fuel biomass significantly drive the fire spread. Meanwhile, a number of biophysical fire simulation models have been developed, as tools to understand the roles of biophysical factors on the spread of wildfires. Those models explicitly incorporate effects of slope, wind direction, wind speed, and vegetative fuel on the spreading rate of surface fire from the ignition points across a fuel bed, based on either field or laboratory experiments. Nevertheless, none of those models have been implemented using real time fire data at relatively large extent areas. This study is aimed at incorporating spatially explicit time series data of weather (i.e. wind direction and wind speed), remotely sensed fuel biomass and remotely sensed fire hotspots, as well as incorporating more persistent biophysical factors (i.e. terrain), into an agent-based fire spread model, in order to identify fire ignitions within time series of remotely sensed fire hotspots.

Seasonal fire management by traditional cattle ranchers prevents the spread of wildfire in the Brazilian Cerrado.

The use of fire by cattle ranchers is a major source of conflict between conservationists and local communities in tropical savannas. We evaluate the role of traditional pastoral management in wildfire prevention in two protected areas within the Brazilian savanna. Fine-grain field data from transect walks and interviews were combined with geospatial data at landscape scale to compare fire regimes in community-managed areas with those in government-managed areas. Local pastoral management creates seasonal mosaic patterns of burnings performed for productive activities and for deliberate landscape management, i.e. to protect fire-sensitive vegetation and avoid wildfires. Whereas government-managed areas were affected by large biennial late dry season wildfires, community-managed areas with a regular fire regime suffered less damage. These systems are under threat and poorly understood by researchers and environmental managers. In order to improve fire management in tropical savannas, greater understanding of pastoral management practices and their spatiotemporal dimensions is required.

System of personnel training in decision making in fighting wildfires

Wildfires fighting is a complex and often dangerous process connected with cooperation of many organizations and crews that requires good organization. An important role in this process plays the qualitative training of personnel involved in the fire fighting. Training of personnel in the techniques and tactics of combating wildfires is conducted in educational systems of various levels, but, regardless of the level, members of firefighting crews should have basic knowledge of fire safety techniques, methods, and tactics. It is known that one of the most effective forms of personnel training is the use of interactive computer systems that allow individual approach to the trainees. Given the dynamic nature of the wildfire spread and the need to make rapid decisions, training system should include an imitation subsystem, which can help a trainee to observe the development of the fire situation on the map, make decisions and evaluate their effectiveness. In the work presented, an interactive simulation system FIREMAN is discussed. The system is designed to train personnel in basic knowledge and decision making in combating wildfires. The structure of the system is described, and an example of how it works is given.

A GIS based method for indexing the broad-leaved forest surfaces by their wildfire ignition probability and wildfire spreading capacity

This article presents a method useful for indexing the broad-leaved forest surfaces by their wildfire ignition probability and wildfire spreading capacity at a coarse spatial scale. The framework consists of three phases; inventory, analysis, and indexing. First, the study utilizes a multi-criteria inventory procedure investigating the existing broad-leaved forest areas of the landscape based on a variety of social, environmental, and physical parameters. Beyond the statistical inventory records, the research brings forward a division between ignition probability and spreading capacities of wildfire events during the analysis phase. At this stage, particular criteria figures out to have higher impact in either ignition or spreading phases of wildfire event. At the final phase, the model is aiming to generate indexing maps categorizing the broad-leaved forest surfaces by their Wildfire Ignition Probability Index and Wildfire Spread Capacity Index. Broad-leaved forest landscape patch as derived via CORINE Land Cover data, is converted into a raster data with a pixel size of 500 m (25 ha). The centroid of each pixel act as the reference point for all measurements during all phases of the study. The presented method is aimed to be of assistance in decision making and management processes of Disaster Risk Management and Fire Safety (DRMFS) agendas at landscape scale.

A nature - inspired approach based on Forest Fire model for modeling rumor propagation in social networks

Rumor propagation is a serious menace faced by online social network users. Modeling rumor propagation helps analyze the diffusion pattern of rumors in social networks and thereby limit the spread of rumors to a considerable extent. There are several factors which affect the extensive spread of rumors in social networks, but the identification of prime features is crucial in modeling the spread of rumors in a faster and more efficient way. In this paper, we propose a novel nature - inspired approach that focuses on the identification of best features to model the diffusion of rumors via online social networks. The proposed approach maps the spread of rumors in social networks to the spread of wildfire in a forest, and the prominent features for modeling rumor dissemination are determined by analyzing the main factors which contribute to the rapid spread of forest fire. We have developed a novel nature - inspired algorithm based on the Forest Fire model that utilizes the identified prominent features to calculate the probability of a node to share a rumor, and it measures the extent of rumor spread across a network by detecting the rumor - affected nodes in the network. The proposed algorithm also serves to examine the circulation path of rumors, the nodes which shared the rumor, and identify the nodes which played a major role in the rumor dissemination process. We evaluated the performance of the proposed method using two datasets obtained from Twitter, and the experimental results illustrate the efficiency of our proposed method and selected features. We also present a rumor propagation graph which aids in the analysis of the rumor diffusion pattern and discovers the key spreaders who are involved in the rumor dispersal. Furthermore, we also provide a feature - level comparison with the various existing approaches for rumor modeling to show how effectively the proposed approach maps the forest fire spread with the problem of rumor diffusion.

The Formative Power of Metony#metoo

ABSTRACTWhat can we learn about #MeToo starting with a formal analysis of its structure? What does the linguistic affirmation of adding one “me” to another “me” ad infinitum create in its powerful drift? This metonymical structure, moving part by part, creates the wildfire spread of the movement, adding to its volatility, including its power against institutional structures that trade more in identity, the stability of an “I,” and the self-preservation of metaphor. This little fragment that we name in this regard “metony#metoo” aims to point to and name a primal scene that is both the open secret of misogynistic violence and the pain and impossibility of sexual relationships. Putting these two together, this volatile structure that points to and exposes structural violence, we ask in this essay—where will we land? What further invention beyond the vent of venting is necessary to help lend this movement an ethical dimension?

Evidence for scale‐dependent topographic controls on wildfire spread

AbstractWildfire ecosystems are thought to be self‐regulated through pattern–process interactions between ignition frequency and location, and patterns of burned and recovering vegetation. Yet, recent increases in the frequency of large wildfires call into question the application of self‐organization theory to landscape resilience. Topography represents a stable bottom‐up template upon which fire interacts as both a physical and an ecological process. However, it is unclear how topographic control changes geographically and across spatial scales. We analyzed fire perimeter and topography data from 16 Bailey ecoregions across the State of California to identify spatial correspondence between ecoregional fire event and topographic patch size distributions. We found both sets of distributions followed a power‐law form and were statistically similar across several orders of magnitude, for most ecoregions. As a direct test of topographic controls on fire event perimeters, we used a paired t‐test across ~11,000 fires to identify differences in topographic attributes at fire boundaries versus fire interiors. Statistical significance was determined using 500 iterations of a neutral landscape model. Level of topographic control varied significantly by ecoregion and across topographic features. For example, north–south aspect breaks, valley bottoms, and roads showed a consistently high degree of spatial control on wildfire perimeters. Topographic controls were most pronounced in mountainous ecoregions and were least influential in arid regions. Ridgetops provided a low‐level control across all ecoregions. Spatial control was strongest for small (100–102 ha) to medium (103–104 ha) fire sizes, suggesting that controls were scale‐dependent rather than scale‐invariant. Roads were the dominant control across all ecoregions; however, removing roads from the analyses had no significant effect on the overall role of topography on wildfire extinguishment in this analysis. This result suggested that certain topographic settings show strong spatial control on fire growth, despite the presence of roads. Our results support the observation that both bottom‐up and top‐down factors constrain fire sizes and that there are likely scaling regions within fire size distributions wherein the dominance of these spatial controls varies. Human influences on fire spread may either diminish or enhance the role of some bottom‐up and top‐down factors, adding further complexity.

Front shape similarity measure for data-driven simulations of wildland fire spread based on state estimation: Application to the RxCADRE field-scale experiment

Data-driven wildfire spread modeling is emerging as a cornerstone for forecasting real-time fire behavior using thermal-infrared imaging data. One key challenge in data assimilation lies in the design of an adequate measure to represent the discrepancies between observed and simulated firelines (or “fronts”). A first approach consists in adopting a Lagrangian description of the flame front and in computing a Euclidean distance between simulated and observed fronts by pairing each observed marker with its closest neighbor along the simulated front. However, this front marker registration approach is difficult to generalize to complex front topology that can occur when fire propagation conditions are highly heterogeneous due to topography, biomass fuel and micrometeorology. To overcome this issue, we investigate in this paper an object-oriented approach derived from the Chan–Vese contour fitting functional used in image processing. The burning area is treated as a moving object that can undergo shape deformations and topological changes. We combine this non-Euclidean measure with a state estimation approach (a Luenberger observer) to perform simulations of the time-evolving fire front location driven by discrete observations of the fireline. We apply this object-oriented data assimilation method to the three-hectare RxCADRE S5 field-scale experiment. We demonstrate that this method provides more accurate forecast of fireline propagation than if either the fire spread model or the observations were taken separately. Results show that when the observation frequency becomes lower than 1/60 s−1, the forecast performance of data assimilation is improved compared to simply extrapolating observation data. This highlights the need of a physics-based forward model to forecast flame front propagation. We also demonstrate that the front shape similarity measure is suitable for both Eulerian and Lagrangian-type front-tracking solvers and thereby can provide a unified framework to track moving structures such as flame front position and topology in combustion problems.

Can Air Quality Management Drive Sustainable Fuels Management at the Temperate Wildland-Urban Interface?

Sustainable fire management has eluded all industrial societies. Given the growing number and magnitude of wildfire events, prescribed fire is being increasingly promoted as the key to reducing wildfire risk. However, smoke from prescribed fires can adversely affect public health and breach air quality standards. Here we propose that air quality standards can lead to the development and adoption of sustainable fire management approaches that lower the risk of economically and ecologically damaging wildfires while improving air quality and reducing climate-forcing emissions. For example, green fire breaks at the wildland-urban interface (WUI) can resist the spread of wildfires into urban areas. These could be created through mechanical thinning of trees, and then maintained by targeted prescribed fire to create biodiverse and aesthetically pleasing landscapes. The harvested woody debris could be used for pellets and other forms of bioenergy in residential space heating and electricity generation. Collectively, such an approach would reduce the negative health impacts of smoke pollution from wildfires, prescribed fires, and combustion of wood for domestic heating. We illustrate such possibilities by comparing current and potential fire management approaches in the environmentally similar landscapes of Vancouver Island in British Columbia, Canada and the island state of Tasmania in Australia.

Prescribed burning reduces the abundance of den sites for a hollow-using mammal in a dry forest ecosystem

Prescribed burning is used in fire-prone environments worldwide to reduce fuel loads and the severity and spread of future wildfires. Forest habitat structures, such as large trees, dead trees and logs are highly flammable, yet also are essential for animal species that require hollows (cavities) as den sites for shelter and reproduction. We examined the effects of experimental prescribed burns on the use of den sites by a small marsupial, the yellow-footed antechinus Antechinus flavipes, in south-eastern Australia. Specifically, we radio-tracked individual A. flavipes to identify forest habitat structures preferred as den sites and recorded the fate of known den sites following patchy prescribed burns. We found that large living trees and dead trees were used as den sites disproportionately to their relative abundance in the forest. While all marked individuals of A. flavipes survived the immediate impacts of patchy prescribed burns, almost a third (16/52) of den sites identified before burning were lost, including 17% of trees (4/23) and 48% of logs (10/21). The vulnerability of den sites to prescribed burns can be attributed to the decay-dependent effect of fire on both trees and logs, whereby, the amount of damage from fire is related to the structure’s pre-fire condition (i.e. whether dead or alive, amount of decay). Large trees and large logs are scarce in this dry forest ecosystem and their replacement is likely to take a century or more due to the slow growth rates of trees. The ecological impacts of prescribed burning on habitat structures used by A. flavipes and other hollow-using species can be moderated by: (1) carrying out patchy, rather than complete burns; (2) ensuring the inter-fire interval is sufficient to allow time for replenishment of resources; and (3) planning at a regional scale to maintain an appropriate spatial pattern of post-fire age-classes, including areas retained as long-unburned (e.g. >50 years) in which resources such as deep litter, large logs and dead trees can accumulate.

Ensemble simulations of a northerly low-level jet in the INFLUX field experiment

The northerly low-level jets often occur in the Great Lakes region of the United States during the cold season. Usually, the jets are closely related to winter snowstorms, wildfire spreading and air pollution dispersion. However, our knowledge on the northerly low-level jets is quite limited. In this study, an ensemble simulated approach, which is based on the WRF-LETKF ensemble Kalman filter numerical prediction/data assimilation system, is used to investigate a northerly low-level jet (N-LLJ) occurred on 1st October 2014 over the state of Indiana during the INFLUX experiment. The simulations are carefully verified by using the Halo Streamline Doppler lidar Profiler data and flight observational datasets. Results show that, the ensemble simulations are more efficient to reproduce the N-LLJ than a single simulation. Based on the ensemble simulation results, the characteristics and mechanisms of the formation and evolution of the N-LLJ are further studied. It is shown that the N-LLJ exhibits typical meso-scale characteristics. Both the increased negative buoyancy and the thermal wind imbalance caused by synoptic frontogenesis are responsible for the formation of this N-LLJ. However, controlled by a shallow large scale system, this N-LLJ does not show significant diurnal cycles. Besides, the comparisons among the ensembles imply that the strength of the N-LLJ is closely related to the intensity of negative buoyancy and frontogenesis.

Analyzing fine-scale spatiotemporal drivers of wildfire in a forest landscape model

We developed and applied a wildfire simulation package in the Envision agent-based landscape modelling system. The wildfire package combines statistical modelling of fire occurrence with a high-resolution, mechanistic wildfire spread model that can capture fine scale effects of fire feedbacks and fuel management, and replicate restoration strategies at scales that are meaningful to forest managers. We applied the model to a landscape covering 1.2 million ha of fire prone area in central Oregon, USA where wildland fires are increasingly impacting conservation, amenity values and developed areas. We conducted simulations to examine the effect of human versus natural ignitions on future fire regimes under current restoration programs, and whether contemporary fire regimes observed in the past 20 years are likely to change as result of fire feedbacks and management activities. The ignition prediction model revealed non-linear effects of location and time of year, and distinct spatiotemporal patterns for human versus natural ignitions. Fire rotation interval among replicate simulations varied from 78 to 170 years and changed little over the 50-yr simulation, suggesting a stable but highly variable and uncertain future fire regime. Interestingly, the potential for fire-on-fire feedbacks was higher for human versus natural ignitions due to human ignition hotspots within the study area. We compare the methods and findings with other forest landscape simulation model (FLSM) studies and discuss future application of FLSMs to address emerging wildfire management and policy issues on fire frequent forests in the western US.

Using Spatial Reinforcement Learning to Build Forest Wildfire Dynamics Models From Satellite Images

Machine learning algorithms have increased tremendously in power in recent years but have yet to be fully utilized in many ecology and sustainable resource management domains such as wildlife reserve design, forest fire management and invasive species spread. One thing these domains have in common is that they contain dynamics that can be characterized as a Spatially Spreading Process (SSP) which requires many parameters to be set precisely to model the dynamics, spread rates and directional biases of the elements which are spreading. We present related work in Artificial Intelligence and Machine Learning for SSP sustainability domains including forest wildfire prediction. We then introduce a novel approach for learning in SSP domains using Reinforcement Learning (RL) where fire is the agent at any cell in the landscape and the set of actions the fire can take from a location at any point in time includes spreading North, South, East, West or not spreading. This approach inverts the usual RL setup since the dynamics of the corresponding Markov Decision Process (MDP) is a known function for immediate wildfire spread. Meanwhile, we learn an agent policy for a predictive model of the dynamics of a complex spatially-spreading process. Rewards are provided for correctly classifying which cells are on fire or not compared to satellite and other related data. We examine the behaviour of five RL algorithms on this problem: Value Iteration, Policy Iteration, Q-Learning, Monte Carlo Tree Search and Asynchronous Advantage Actor-Critic (A3C). We compare to a Gaussian process based supervised learning approach and discuss the relation of our approach to manually constructed, state-of-the-art methods from forest wildfire modelling. We also discuss the relation of our approach to manually constructed, state-of-the-art methods from forest wildfire modelling. We validate our approach with satellite image data of two massive wildfire events in Northern Alberta, Canada; the Fort McMurray fire of 2016 and the Richardson fire of 2011. The results show that we can learn predictive, agent-based policies as models of spatial dynamics using RL on readily available satellite images that other methods and have many additional advantages in terms of generalizability and interpretability.

Existence of solution of a forest fire spread model

We consider a wildfire spread model represented by the system (1) . We use results from the theory of Hamilton–Jacobi equations to prove that there exists a classical solution of (1) for any ( ϕ , t ) ∈ R × ( 0 , T ) and some T > 0 and satisfies particular initial conditions. We also use the method of characteristics to obtain the solution of (1) in a certain form.

On-demand data assimilation of large-scale spatial temporal systems using sequential Monte Carlo methods

The proliferation of sensors is generating rapidly increasing quantities of data like never before. These extensive amounts of data can provide useful information for more accurate state inference of large-scale spatial temporal systems. Sequential Monte Carlo methods are used to assimilate the observed data from sensors to improve the state estimation of large-scale spatial temporal systems, which highly rely on the available real time observation data. In many scenarios, the real time data are limited in space and time. Therefore, it is important to effectively obtain critical sensor data in real time and then dynamically feed them into the running model. In this paper, we propose the on-demand data assimilation method for large-scale spatial temporal systems, in which we quantify the spatial states using run-time state quantification methods and decide if we need to trigger data assimilation on demand and obtain more relevant real time data when the state uncertainty is high. The effectiveness of the developed framework is evaluated based on large-scale wildfire spread simulations.

Fog and live fuel moisture in coastal California shrublands

AbstractAcross Mediterranean‐type climate regions, seasonal drought desiccates plants, facilitating ignition and the spread of wildfires. Along the California coast, summertime fog has the potential to ameliorate drought conditions and thus reduce plant flammability during a critical time of elevated fire risk. This study investigated the uptake of dry season fog and how it affects live fuel moisture (LFM) in six dominant shrub species from chaparral and sage scrub plant associations. Fog water uptake was evaluated using stable isotopes of hydrogen and oxygen at several field sites in Santa Barbara County, California. Clear evidence of fog water uptake was identified only in Baccharis pilularis, from the sage scrub association. To determine the effects of fog on LFM, meteorological variables and indices including fog deposition were combined into principal components and the scores regressed against the LFM loss rate during the summer drought. Fog deposition slowed rates of LFM loss for all three sage scrub species tested, but it did not affect the chaparral species. Fog is a more regular occurrence in the sage scrub association, and thus, it is likely that fog ameliorates drought for species that experience consistent fog during the summer months. In coastal California, summer fog can be essential to plant water relations and reduce LFM loss rates during the summer drought. Understanding these effects is important in the context of changing climate in southern California and fire‐prone regions around the world.

Wind tunnel experiments on effects of woven wire screens and buffer zones in mitigating risks associated with firebrand showers

The benefits of using wire screens to mitigate wildfire spread are well known and many standards mandate their use in wildland fire prone areas. However, some recent studies show that firebrands can penetrate through screens. The penetrated firebrands are smaller in size, have less potential to ignite fuel beds and have a shorter burn-out time. This leads to the idea that a combination of a buffer zone and screens might be able to serve as a wildfire barrier. This study aims to explore the effectiveness of such an approach for two screens with a relatively short buffer zone. The results show that the combination is very effective in reducing firebrand intensity but falls short of full protection. The study suggests the combination of screens and a wider buffer zone is a promising approach to mitigate the spread of wildfires.