Fireline Intensity Research Articles

Visual assessments of fuel loads are poorly related to destructively sampled fuel loads in eucalypt forests

Fire managers around the world commonly use visual assessment of forest fuels to aid prediction of fire behaviour and plan for hazard reduction burning. In Australia, fuel hazard assessment guides also allow conversion of visual assessments to indicative fuel loads, which is essential for some rate of spread models and calculation of fireline intensity or emissions. The strength of correlation between fuel hazard and destructively sampled (directly measured) fuel load was tested using a comprehensive dataset of >500 points from across a range of eucalypt forests in Australia. Overall, there was poor correlation between the assigned fuel hazard rating and measured biomass for surface, near-surface and elevated fuel components, with a clear tendency for these systems to under-predict fuel load at low hazard ratings, and over-predict it at high hazard ratings. Visual assessment of surface fuels was not statistically different from a random allocation of hazard level. The considerable overlap in fuel load between hazard ratings at higher ranges suggests the need to reduce the number of hazard classes to provide clearer differentiation of fuel hazard. To accurately assess forest fuel condition, improvements in fuel hazard descriptions and calibration of visual assessment with destructively measured fuels is essential.

Hazard curve evaluation method development for a forest fire as an external hazard on nuclear power plants

A method to obtain a hazard curve of a forest fire was developed. The method has four steps: a logic tree formulation, a response surface evaluation, a Monte Carlo simulation, and an annual exceedance frequency calculation. The logic tree consists domains of “forest fire breakout and spread conditions”, “weather conditions”, “vegetation conditions”, and “forest fire simulation conditions.” Condition parameters of the logic boxes are static if stable during a forest fire or not sensitive to a forest fire intensity, and non-static parameters are variables whose frequency/probability is given based on existing databases or evaluations. Response surfaces of a reaction intensity and a fireline intensity were prepared by interpolating outputs from a number of forest fire propagation simulations by fire area simulator (FARSITE). The Monte Carlo simulation was performed where one sample represented a set of variable parameters of the logic boxes and a corresponding intensity was evaluated from the response surface. The hazard curve, i.e. an annual exceedance frequency of the intensity, was therefore calculated from the histogram of the Monte Carlo simulation outputs. The new method was applied to evaluate hazard curves of a reaction intensity and a fireline intensity for a typical location around a sodium-cooled fast reactor in Japan.

Mapping fire behaviour under changing climate in a Mediterranean landscape in Greece

Understanding how future climate periods influence fire behaviour is important for organizing fire suppression strategy and management . The meteorological factors are the most critical parameters affecting fire behaviour in natural landscapes; hence, predicting climate change effects on fire behaviour could be an option for optimizing firefighting resource management. In this study, we assessed climate change impacts on fire behaviour parameters (rate of fire growth, rate of spread and fireline intensity) for a typical Mediterranean landscape of Greece. We applied the minimum travel time fire simulation algorithm by using the FlamMap software to characterize potential response of fire behaviour for three summer periods. The results consisted of simulated spatially explicit fire behaviour parameters of the present climate (2000) and three future summer periods of 2050, 2070 and 2100, under the A1B emissions scenario. Statistical significant differences in simulation outputs among the four examined periods were obtained by using the Tukey’s significance test. Statistical significant differences were mainly obtained for 2100 compared to the present climate due to the significant projected increase in the wind speed by the end of the century. The analysis and the conclusions of the study can be important inputs for fire suppression strategy and fire management (deployment of fire suppression resources, firefighter safety and exposure, transportation logistics) quantifying the effect that the expected future climate periods can have on fire suppression difficulty in Mediterranean landscapes.

Short-term effects of mastication on fuel moisture and thermal regime of boreal fuel beds

Mechanical mastication is becoming a common fuel management treatment to reduce vertical fuel connectivity, as well as crown fire initiation and potential fire-line intensity, but the moisture dynamics of these novel fuel types have been largely unstudied. We recorded concurrent in situ meteorological observations with moisture and temperature profiles (at depths of 5 and 13 cm) for masticated fuel beds in three treatments in a lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm.) – black spruce (Picea mariana (Mill.) B.S.P.) boreal stand located in the Upper Foothills of west-central Alberta. Mulch at the 5 cm depth remained at or near 100% moisture content for a majority of the observation period within the treatment that lacked any residual canopy cover. Only during a 10-day rain-free period was substantial drying observed at a depth of 5 cm. At the interface with the underlying duff layer, mulch remained at upwards of 150% moisture content. Drying was typically <10% of potential evaporation rates, except on days after rain, when drying equal to 25%–100% of potential evaporation was observed. Thermal properties of the mulch showed approximately five times the thermal diffusivity of soils, but less than a fuel crib, suggesting that masticated fuel beds have a thermal and moisture regime more similar to soils than fuel cribs, with a diffusion-dominated regime enhanced by minor advection. Low moisture movement observed from depth to the surface promoted highly variable surface moisture that created surface mulch with temperatures far in excess of air temperatures during periods of full sun. Such excess temperatures are shown to be an efficient indicator of surface dryness and high ignition probability.

Forest fire propagation simulations for a risk assessment methodology development for a nuclear power plant

After the Fukushima Daiichi nuclear power plant [NPP] accident, there has been an increased concern with the safety of NPPs in terms of external hazards, one of which is a forest fire which can create potential challenges to safety functions and the structural integrity of an NPP. As a part of the development of a risk assessment methodology for forest fires as an external hazard, forest fire propagation simulations have been performed by using the FARSITE simulator. These simulations have been used to evaluate two intensity parameters (i.e. fireline intensity and reaction intensity) and three other key parameters (i.e. flame length, rate-of-spread, and forest fire arrival time) which are related to “heat” and “flame” effects on an NPP. Sensitivity analyses for a wide range of weather conditions were performed in order to identify the variable ranges of the intensity and other key parameters. The location studied was selected from among areas with typical topographical and vegetation surrounding NPPs in Japan. The NPP is facing the sea and surrounded by hills, distanced from an urban area, with mostly broad leaf forests, several paddy fields and a few pasture areas.Low-to-high frequency weather conditions have been utilized in this analysis; forest fire propagation simulations were performed “with/without prevailing wind” (i.e. 0–24m/s wind speed) and “high/low values for ambient temperature and relative humidity” (−4.3 to 37°C and 5–99%, respectively) according to the recorded data ranges for the typical NPP site. The maximum values of fireline intensity and rate-of-spread are 4.7×102kW/m and 2.4m/min and they depend very much on prevailing wind speed and relative humidity (around 2.3 and 1.8 times respectively) but less on ambient temperature (around 1.1 times). Reaction intensity and flame length change within relatively narrow ranges (around 1.7 and 1.5 times respectively) even for all the variation in weather parameters. The forest fire arrival time at the site is reduced by a factor of 5 with changing prevailing wind speed from the recorded-highest to zero. The arrival time increases some 3.4 times with the highest humidity compared to the recorded-lowest conditions, although it is changed little even by varying ambient temperature.Given that this study shows that the maximum height of a flame on a canopy top is close to the range of power line height, a loss of offsite power is recognized as a possible subsequent event during a forest fire.

A mixed integer program to model spatial wildfire behavior and suppression placement decisions

Wildfire suppression combines multiple objectives and dynamic fire behavior to form a complex problem for decision makers. This paper presents a mixed integer program designed to explore integrating spatial fire behavior and suppression placement decisions into a mathematical programming framework. Fire behavior and suppression placement decisions are modeled using nodes associated with cell centers from raster landscapes. The nodes at which suppression is located are determined by control variables. Response variables include fire spread paths, arrival times, and fireline intensities for each node. Both fire arrival times and fireline intensities are necessary to address ecological objectives and fire control. Test cases for this model provide examples of fire behavior interacting with suppression placement to achieve multiple objectives.

Honey Mesquite (Prosopis glandulosa) Seedling Responses to Seasonal Timing of Fire and Fireline Intensity

Fire-resistant woody species have invaded many grassland and savanna ecosystems worldwide. Adults of many of these species are fire resistant because they resprout following fire; however, seedlings may be vulnerable to fire. Here we quantify effects of seasonal timing of fire and fireline intensity on seedling mortality response to fire in honey mesquite (Prosopis glandulosa Torr.), a native woody legume that has increased in abundance in southern Great Plains, USA grasslands. Scarified mesquite seeds were planted in plots with either mid-grass or tall-grass fuels in each of 2 years (Cohorts 1 and 2). Plots were burned in winter when seedlings were 10 or 22 months old or late summer when they were 17 months old. In the mid-grass fuel type, seedling mortality was greater (P ≤ 0.05) in summer-burned, 17-month-old seedlings (85%) than in winter-burned, 10-month-old seedlings (35%), suggesting summer fire may be important in limiting seedling establishment under lower-grass, fine-fuel conditions. The tall-grass fuel type yielded similar mortality responses to summer (89% and 79%) and winter (77% and 66%) fires in Cohorts 1 and 2, respectively (differences not significant at P ≤ 0.05). The study found a positive relationship between fireline intensity and seedling mortality, especially when only winter fires were included in the regression (r2 = 0.96). Summer fire may be most useful in lower-grass fuel amounts and may accomplish the same level of seedling mortality without the need for lengthy grazing deferral to accumulate sufficient grass fine fuel for a successful winter fire, as seedling mortality from winter fire depends on fireline intensity.

Adding fuel to the fire? Revegetation influences wildfire size and intensity

The regrowth of woody vegetation in cleared landscapes (i.e. revegetation) has the potential to dramatically alter the spatial characteristics of vegetation and fuels, which will potentially alter fire characteristics. Understanding how revegetation alters fire size and intensity will be critical in determining the social and environmental value of revegetation. We used simulation modelling to examine (i) whether increasing native woody vegetation extent across landscapes cleared for pasture (i.e. revegetation) affects fire size and median fireline intensity and (ii) whether fuel load in the pasture matrix, the initial extent of land clearing and weather conditions during a fire alter the direction and/or magnitude of the relationships between revegetation and fire size or intensity. Simulations revealed that fire size and intensity were altered by increasing woody vegetation extent, though the direction of change was dependent upon landscape context. Increased woody vegetation extent led to (i) increased fire size in landscapes with low pasture fuel load (2tha−1) regardless of the extent of land clearing, (ii) decreased fire size in highly cleared landscapes with moderate (4.5tha−1) and high (7tha−1) pasture fuel load, and (iii) little change to fire size in landscapes subjected to low levels of clearing when pasture fuel load was moderate or high. Similar patterns were observed for fireline intensity. The magnitude of change in fire size and intensity was greatest under extreme fire weather conditions. Revegetation rarely increased median fireline intensity beyond suppressible levels (i.e. 4000kWm−1), with fire weather and pasture fuel load being the main determinants of suppression potential. Our findings show that the response of fire size and intensity to revegetation will depend on landscape scale pasture management.

Evaluating the Effects of Simulated Silvicultural Treatments and Management on Wildfire Severity in Pinus halepensis Mill. Even-Aged Stands

Greece is a Mediterranean country where the problem of devastating wildfire occurrence is great, threatening human lives and property. In the current research, an attempt is being made to evaluate the effects of simulated silvicultural treatments in order to efficiently reduce wildfire’s’ potential severity. The evaluation is based on the establishment of 52 experimental plots of 500 m2 and the estimation of the dendrometric characteristics (diameter at breast height, basal area, total height, crown length, canopy cover, understorey composition). The total sample was composed of n=1488 Pinus halepensis individuals, while the topographic parameter of slope steepness was estimated for each plot. Based on the NEXUS wildfire simulator software, critical fire hazard indicators such as the Torching and Crowning Indices were estimated before and after the virtual establishment of silvicultural treatments. The analysis showed that combinations of simple silvicultural interventions decreased the wildfire severity in all cases. The frontal fireline intensity was reduced by more than 70%, while the flame length was reduced to 59.1% and the spread rate to 20.6%. Despite the efficiency of the silvicultural treatments, under hazardous climatic conditions fire severity remained high for firefighting operations. Keywords: Crowning index, Greece, NEXUS, pine woodlands, torching index, wildfire behavior.

Biomass and fire dynamics in a temperate forest-grassland mosaic: Integrating multi-species herbivory, climate, and fire with the FireBGCv2/GrazeBGC system

Landscape fire succession models (LFSMs) predict spatially–explicit interactions between vegetation succession and disturbance, but these models have yet to fully integrate ungulate herbivory as a driver of their processes. We modified a complex LFSM, FireBGCv2, to include a multi-species herbivory module, GrazeBGC. The system is novel in that it explicitly accommodates multiple herbivore populations, inter- and intra-specific spatial forcing of their forage demands, and site-specific dietary selectivity to interactively modify biomass, fuels and fire behavior across a landscape and over time. A factorial experiment with five grazing regimes, three climates and two fire-management scenarios generated interactive influences on undergrowth biomass (shrub, herb, total), surface-fire (fire-line intensity; flame length; scorch height; soil heat; CO, CO2, CH4, and PM2.5 emissions), and the landscape’s fire-return interval. Herbivory’s effects increased with biophysical site potential and herbivore forage demand, but its effects were also contingent on climate and fire-suppression. Multi-species grazing modified biomass and fire within stands and biophysical sites, but regimes involving only wildlife or livestock were less effectual. Multi-species herbivory affected the landscape’s fire-return interval, but otherwise it did not “scale up” to significantly modify total landscape respiration, primary production, carbon, or the total area burned by individual fires. As modeled here, climate change and the effectiveness of future fire suppression exerted stronger effects on landscape metabolism and carbon than did herbivory.

Carbon loss from planned fires in southeastern Australian dry Eucalyptus forests

This study reports the immediate impact of planned fires on carbon distribution and storage in six forest types of south-eastern Australia. Aboveground carbon (AGC) ranged from 56 to 183MgCha−1 where between 4 and 11Mgha−1 (or 6% of AGC) was lost in planned fire. Between 3.6 and 5MgCha−1 was redistributed within the forest to the soil surface as char and partly combusted organic matter. Most carbon was lost from litter while near surface vegetation and dead trees either standing or lying were the next largest C loss groups. Overstorey tree biomass (MgCha−1) was a significant predictor of carbon loss in major fuel categories, explaining 45% of total carbon loss. The loss of coarse woody debris (CWD, %) was strongly correlated with losses in litter (%, R2=0.44) and with fireline intensity (%, R2=0.38).These C loss estimates in planned fire establish an empirical basis for further modeling to identify management options for mitigating fire-induced C losses from SE Australian Eucalyptus forests.

Wind Effects, Unsteady Behaviors, and Regimes of Propagation of Surface Fires in Open Field

The subject of this article concerns the unsteady effects (fire intensity, wind) upon the propagation and, more generally, the behavior of surface fires in open fields. The study focused on two sources of unsteadiness: the first one resulting from the regime of propagation (wind driven or plume dominated), which can affect greatly the behavior of the flame front and consequently the fire intensity, the second one resulting from the wind gusts associated with the conditions of flow of wind in real conditions. The study was based on numerical simulations, using a multiphase formulation, and on spectral analysis of the time evolution of the fire line intensity. The calculations were performed in 2D for a homogeneous vegetation layer (grassland) and for a large interval of wind conditions (10 m open wind velocity U10 ranged between 1 m/s and 25 m/s). The results have highlighted the link between the unsteady character of flame front behavior and the regime of propagation (plume dominated, wind driven). A particular interest was focused on the role played by two potential sources of instabilities, namely the Kelvin-Helmholtz instability (wind effects) and the thermo-convective instability (plume effects), upon the behavior of fires. A second set of simulations has been carried out using unsteady wind conditions, reproduced using sinusoidal boundary conditions for the streamwise velocity, with a frequency ranging between 0.5 Hz and 3 Hz.

Using Benefit Transfer to Estimate Average Relative Marginal Values for Wildland Fire Program Planning

We developed a set of generalized value categories and average relative marginal values using resource and valuation information collected at seven federal land management planning units in the USA. The categories and average values are intended to be used for rapid strategic wildland fire program planning or as foundation values for a more extended planning effort. To divide the original information into logical and statistically valid value categories, we used a k-means cluster analysis combined with expert knowledge of how each resource type is managed with respect to fire. After we established the value categories, we performed a regression tree analysis to estimate an average relative marginal value for each category under different combinations of ecosystem condition and fireline intensity. Average values for each category are as expected and the regression tree fits are excellent with R2 values ranging from .895 to .948. The results of this study can be used to identify general resource types for planning or other valuation analyses and they provide a low cost way to transfer nonmarket value information between similar planning sites.

Atmospheric PM and volatile organic compounds released from Mediterranean shrubland wildfires

Wildfires produce a significant release of gases and particles affecting climate and air quality. In the Mediterranean region, shrublands significantly contribute to burned areas and may show specific emission profiles. Our objective was to depict and quantify the primary-derived aerosols and precursors of secondary particulate species released during shrubland experimental fires, in which fire-line intensity values were equivalent to those of moderate shrubland wildfires, by using a number of different methodologies for the characterization of organic and inorganic compounds in both gas-phase and particulate-phase. Emissions of PM mass, particle number concentrations and organic and inorganic PMx components during flaming and smouldering phases were characterized in a field shrubland fire experiment. Our results revealed a clear prevalence of K+ and SO42− as inorganic ions released during the flaming–smouldering processes, accounting for 68–80% of the inorganic soluble fraction. During the residual-smouldering phases, in addition to K+ and SO42−, Ca2+ was found in significant amounts probably due the predominance of re-suspension processes (ashes and soil dust) over other emission sources during this stage. Concerning organic markers, the chromatograms were dominated by phenols, n-alkanals and n-alkanones, as well as by alcohol biomarkers in all the PMx fractions investigated. Levoglucosan was the most abundant degradation compound with maximum emission factors between 182 and 261 mg kg−1 in PM2.5 and PM10 respectively. However, levoglucosan was also observed in significant amounts in the gas-phase. The most representative organic volatile constituents in the smoke samples were alcohols, carbonyls, acids, monocyclic and bicyclic arenes, isoprenoids and alkanes compounds. The emission factors obtained in this study may contribute to the validation and improvement of national and international emission inventories of this intricate and diffuse emission source.

Fire behavior in masticated fuels: A review

Mastication is an increasingly common fuels treatment that redistributes “ladder” fuels to the forest floor to reduce vertical fuel continuity, crown fire potential, and fireline intensity, but fuel models do not exist for predicting fire behavior in these fuel types. Recent fires burning in masticated fuels have behaved in unexpected and contradictory ways, likely because the shredded, compact fuel created when trees and shrubs are masticated contains irregularly shaped pieces in mixtures quite different from other woody fuels. We review fuels characteristics and fire behavior in masticated fuels across the United States. With insights from the few laboratory and field burning experiments conducted, we highlight the variation likely to occur across different ecosystems in which these treatments are being widely implemented. Masticated debris has a propensity to flame and smolder for long durations. Fuel variability and vegetation response will likely influence whether or not treatments reduce long-term fire hazard. We identify key science needs that will better elucidate fire behavior and effects in these treatments. With mastication widely applied in an expanding wildland–urban interface it is crucial to understand how such fuels burn. What we learn about combustion in these fuels will inform effective fuels management in these and other mixed fuels.

Diradamenti e fuoco prescritto per la prevenzione degli incendi boschivi in rimboschimenti di pino d'Aleppo

This paper reports the results of the first fire hazard reduction experiment in Italy which integrates thinning and prescribed burning. The study was conducted in Aleppo pine stands in a high fir-risk area in the Calabrian Region. Two stands with a different stem density were selected. The experiment design included: i) prescribed burning followed by selective thinning from below in the stand with lower tree density; ii) selected thinning from below followed by prescribed burning in the stand with higher tree density. Light/moderate thinning intensity treatments were applied at both sites in June-July 2013. Prescribed burning treatments were conducted in May 2013 and March 2014. Data on fuel characteristics were collected before and after treatments. A working time study was also carried out during thinning operations. During burn operations, fireline intensity never exceeded 100 kW/m. Maximum temperature in the organic soil layer was 130°C, and on average temperatures above 60°C lasted less than 4 minutes. The organic soil layer was not consumed, and no damages to adult trees were observed. The results highlight the positive effect of prescribed burning and thinning on the reduction of fuel load and continuity, and then on potential fire behaviour, thus reducing the susceptibility and vulnerability of Aleppo pine stands to summer wildfires. The integrated approach of the study shows how thinning and prescribed burning, the latter a prevention practice surprisingly not very common in a fire-prone environment like Southern Italy, could be a sustainable practice to be included in the ordinary forest management planning.

BUILDING FUEL MODELS AND SIMULATING THEIR SURFACE FIRE BEHAVIOR IN THE “SERRA DE ITABAIANA” NATIONAL PARK, SERGIPE, BRAZIL

The objectives of this research were, for scientific and management purposes, to build fuel models and simulate their fire behavior. Three different vegetation types (shrublands, grass fields and tropical forests) of the “Serra de Itabaiana” National Park were analyzed. The fuel models were developed by destructive sampling and the collected data was inserted in the software “BehavePlus 5.0”. The results revealed that the fuel model for the shrublands presented the longest flame length, the highest fireline intensity and the greatest heat release per unit area. The fuel model for the grass fields presented the fastest surface rate of spread; and the fuel model for the tropical forests the lower fire intensity.Keywords: Fire simulator; BehavePlus; conservation units.ResumoConstruindo modelos de material combustível e simulando o seu comportamento de fogo no Parque Nacional Serra de Itabaiana, SE. Este estudo foi desenvolvido com o objetivo de construir modelos de material combustível para três diferentes formações vegetacionais (florestas arbustivas, campos graminosos e florestas tropicais), localizadas dentro do Parque Nacional Serra de Itabaiana, e simular o comportamento do fogo dentro dessas fitofisionomias, para efeitos de pesquisa e manejo. Para tal, modelos foram construídos através de amostragem destrutiva e os dados coletados inseridos no programa "BehavePlus 5.0". De acordo com os resultados, o modelo para as florestas arbustivas apresentou o maior comprimento de chamas, a maior intensidade do fogo e a maior quantidade de energia liberada por unidade de área. O modelo para os campos graminosos apresentou maior velocidade de propagação do fogo e, no modelo para as florestas tropicais, o fogo simulado apresentou menor intensidade.Palavras-chave: Manejo do fogo; BehavePlus; unidades de conservação.

Radiant, convective and heat release characterization of vegetation fire

Fire spread across forest fuel is usually characterized by the rate of spread or the fireline intensity. The determination of the fireline intensity represents an essential aspect for understanding the behaviour of the fire and the involved combustion processes. The heat released during fire spread cannot be a-priori estimated from the fundamental properties of the fuel material and experiments need to be carried out to determine it. This paper presents a global characterization of horizontal fire spread in still air across fuel beds in terms of heat release, rate of spread, flame geometry and radiant and convective fractions. The influence of the fuel load on these main fire properties is investigated. A series of experiments was conducted using a large scale heat release apparatus. The fire tests were carried out on a combustion table located on a load cell. The fuel consisted in a 2 m long and 1 m wide bed of pine needles. The fireline intensity was accurately estimated by means of oxygen consumption calorimetry and some other methods to assess this quantity were also tested. Combustion efficiency and effective heat of combustion were discussed. The heat fluxes emitted during the fire spread were also investigated. In the studied configuration, radiation was the dominant heat transfer mechanism in the preheating zone; whereas some transfers combining radiation and convection were highlighted closer to the flame front. The radiant (from flame and embers) and convective fractions of the fire front were then calculated. The data exhibit that the fuel load has a significant influence on the thermal degradation of the forest material and on the resulting fire properties. The results also suggest that the measurement of the mass loss rate can be a good alternative to estimate accurately the fireline intensity when the measure of oxygen consumption is not possible like in field-scale scenarios.

The role of holm oak edges in the control of disturbance and conservation of plant diversity in fire-prone landscapes

Holm oak (Quercus rotundifolia) woodlands have been able to persist in fire-prone landscapes in northern Portugal where they seem to play a relevant role in community and landscape level patterns and processes, namely in the control of disturbance events and in the maintenance of biodiversity. In this research we tested the following hypotheses: (i) holm oak woodlands affect fire behavior, mainly by disrupting and delaying fire spread; (ii) disruption and delaying of fire spread is due to fire extinction at the edge of woodlands; and (iii) plant diversity is favored by frequent low-intensity fires as a result of the control of fire spread by holm oak woodland edges. We sampled herbaceous, shrub and tree structure and herbaceous plant composition across edges of holm oak woodlands with known recent contact with fire. Using BehavePlus we converted structural data into fuel models and simulated fire behavior for the locations sampled along the exterior–interior gradient of the woodlands. Our results indicate that holm oak woodlands play a relevant role in the control of fire spread and in the maintenance of plant diversity. We found differences in simulated maximum rate of fire spread (ROS) and fireline intensity (FLI) between outside and inside woodlands as well as a 100% probability of fire extinction when fuel moisture was >12% and wind speed at 10m aboveground was <40kmh−1, supporting the self-extinction hypothesis. We found also that richness and abundance of herbaceous plants varied between the interior and exterior of the woodland, with higher values outside the woodland, and that plant composition along the interior–exterior gradient had distinct patterns. These variables and the pattern of plant distribution were also significantly affected by time since last fire event. We conclude that holm oak woodlands affect fire behavior by interrupting fire spread at the edge zone and that fire-created edges affect herbaceous plant species richness, abundance and distribution by maintaining habitat diversity, which increased plant diversity at patch and landscape levels. Based on our results and other evidence, we developed a conceptual model of the dynamics of fire, landscape structure, and plant diversity in fire-prone landscapes where holm oak patches persist.

Refining thresholds in coupled fire–vegetation models to improve management of encroaching woody plants in grasslands

Summary Restoration priorities are typically established without quantitative information on how to overcome the thresholds that preclude successful restoration of desirable ecosystem properties and services. We seek to demonstrate that quantifying ecological thresholds and incorporating them into management‐oriented frameworks provide a more comprehensive perspective on how the threshold concept can be applied to achieve restoration goals. As an example, restoration actions have been largely unsuccessful when based on prevailing ecological knowledge of fire‐based thresholds in nonresprouting Juniperus woodland. We build on previous threshold‐based research and link well‐established models from applied fire physics with a widely applied ecological positive feedback model of woody plant encroachment to introduce a more comprehensive understanding of the mechanism influencing fire intensity and juniper mortality. Our coupling of physical and ecological fire models revealed a critical knowledge gap, a lack of a quantitative estimate on the critical surface fire intensity required to cause mortality of Juniperus ashei trees, which limits the linking of scientific knowledge from these two disciplines. To quantify the relationship between fire intensity and J. ashei mortality, we input data from a previous experiment into Byram's fireline intensity model. This critical surface fire intensity–mortality threshold was estimated to be Is &gt; 160 kJ m−1 s−1. This value establishes a specific threshold that managers should target when attempting to use restoration to collapse J. ashei woodlands. Synthesis and applications. For scientific information associated with the threshold concept to be useful to practitioners, specific information is needed that demonstrates how to use restoration activities to overcome thresholds and collapse the current, degraded state in favour of a more desired ecological state. With this in mind, we present a broadly applicable decision support model within a state and transition framework that identifies the ecological states where the surface fire intensity–mortality threshold is most likely to meet restoration objectives and provides examples of how fuel properties that drive fire intensity should be targeted in restoration to surpass this threshold.