iiinova — where your digital self keeps evolving

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/10/2026

🔥 New in IJWF: Ochoa et al. introduce ESFire30, a 30 m burned-area dataset mapping nearly four decades of wildfire across peninsular Spain. The dataset reveals strong spatial variability, frequent reburning in the northwest. 🔗 doi.org/10.1071/WF26... #IJWildlandFire

Fig. 4. Pixels that burned multiple times between 1985 and 2023: (a) burned area (BA) in large fires; (b) burned area > 500 ha for the same period. wf26035

Fig. 3. Annual burned areas classified by fire type (a), proportion of patches by size and number of fires; and (b) the total number of patches per year with frequency in logarithmic scale.

043

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 21/09/2026

🔥 New in IJWF: Morrison et al. show bushfire evacuation is rarely a simple A-to-B journey. Interviews with 52 residents across three Australian fires reveal stops for children, pets, neighbours, belongings and information, shaped by social connections. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Conceptual model explaining intermediate stop locations for evacuees during three Australian bushfires; the types of stop sites are shown in bold (surrounded by dotted-lined boxes).

043

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 16/09/2026

🔥 New in IJWF: Van Dusen et al. show fuel treatments can reduce high-severity fire, especially when recent and combining thinning with prescribed fire, while extreme fire spread can overwhelm treatment effects in complex landscapes. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

A semi-automated, empirical workflow to evaluate the effects of fuel treatments across biophysical conditions. Our workflow is built with flexibly to support the evaluation of treatment effects across multiple measures effectiveness. In this study, we focus on one response variable: vegetation burn severity (low-moderate vs high severity). The workflow includes three broad phases: Data preparation, Fire-level Matching, Landscape-level Modeling. (1) Geospatial data preparation includes the selection of previous fires and historical treatments, gathering geospatial datasets, selecting a response variable (burn severity) and data extraction, (2) fire-level matching aims to pair treated and untreated points based on fire-specific drivers of the response variable, and (3) landscape-level models use model-based inference to evaluate treatment effectiveness across multiple biophysical conditions.

The mean predicted probability of high severity among treatment classes across the Klamath-Trinity Landscape for the east and west random forest classification models. The partial dependence mean probability of high severity for each class was calculated by predicting each treatment class across all other predictors values then taking the mean response. The error bars for each treatment class represents the 95th confidence interval of the mean probability using 1000 bootstrapped datasets to fit our model.

Partial dependence plots for treatment age the west (a) and east (b) study area of the Klamath-Trinity Landscape separated by each treatment type. The dashed, navy untreated line represents the mean probability of high severity for untreated points. Importance was determined by permutation variable importance from the landscape random forest classification model. Each solid line displays the mean relationship between each treatment type and environmental variable across all other environmental variable combinations. The shaded region represents the 95th percentile across variables by treatment type through 1000 bootstrapped models.

066

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 05/09/2026

🔥 New in IJWF: Charles et al. show that satellite-derived fire maps can underestimate fire frequency, particularly in infrequently burnt areas. Integrating public land fire histories and environmental data improved landscape-scale estimates. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Fire frequency from 1987 to 2023 in southeast Queensland, Australia derived from (a) observed satellite and (b) public land fire history data. The observed fire frequencies were compared to predictions from: (c) generalised linear model (GLM), (d) generalised additive model (GAM), (e) down-weighted BRT, (f) unweighted BRT, and (g) Infinitely Weighted Logistic Regression (Infinite BRT). White areas are those mapped as unburnt. The maximum estimated fire frequency varied across model types: (a) satellite data = 29; (b) public data = 12; (c) GLM = 29; (d) GAM = 40; (e) down-weighted BRT = 130; (f) unweighted BRT = 115; (g) Infinite BRT = 9. Fewer than 1% of cells had fire frequencies >30 from 1987 to 2023 for GAM, unweighted BRT and down-weighted BRT. Thus, to aid visualisation, fire frequencies >30 are not shown but can be extracted from the database provided online (Charles and Smith 2026).

Comparisons of fire frequency estimates between public land fire history data (‘public’), raw, unmodelled satellite data (‘satellite’) and predictions from a range of model types. The right-hand panel for each model type shows cell counts below 100 to enable comparisons at high fire frequencies (fire frequencies ≥4 had very low cell counts and were difficult to visualise). All fire frequency estimates were compared against the public land fire data as a baseline, with fire frequency at presence points ranging from 0 to a maximum of 16 fires depending on the model. (a) Observed = satellite and public land, (b) generalised linear (GLM), (c) generalised additive (GAM), (d) down-weighted Boosted Regression Tree (BRT), (e) unweighted BRT, and (f) Infinitely Weighted Logistic Regression BRT (Infinite BRT) model predictions.

Generalisable workflow for improving fire frequency estimates using predictive modelling: (a) obtain and reformat fire (e.g. public land and satellite, where available) and environmental (e.g. climate, site productivity, terrain) data; (b) calculate fire frequency from fire history data; (c) run models; (d) produce spatial predictions using the modelled relationship between satellite-derived fire frequency, public land fire frequency and environmental co-variates, which can be projected outside the public estate; and (e) evaluate predictions by comparison of spatial predictions and model performance statistics. The workflow could be used with any fire regime parameter of interest, e.g. substituting fire frequency for fire return interval, time since last fire, or fire seasonality. ROC, Receiver Operating Characteristic Curve.

085

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 01/09/2026

🔥 New in IJWF: Lin et al. show that non-composted mulches spread fire faster and produce greater heat flux than composted products. Full-scale tests also reveal wind-driven spotting and particle-size effects in the WUI. 🔗 doi.org/10.1071/WF26... #IJWildlandFire

Progression and key fire behavior phenomena of tests using fir bark with different particle size distributions.

Fire spread rate calculated based on the moving speed of the fire front. Note: the fire spread rate was calculated as the moving speed of the burning region (including the flame front or smoldering front), and the effect of spotting ignition by moving embers was not considered here. Moreover, the fire spread rate of screened compost was not shown, as the fire only burned locally without spreading within the mulch bed. The error bars represent the standard deviations of the repeated tests.

Comparison of peak radiation, convection, and total heat fluxes from different mulch products, measured at 1 foot (~30 cm) beyond and ~4.6 inches (11.8 cm) above the bed boundary as part of a sensor package. Note: these should be considered conservative estimates of the true fire exposure potential because the actual fluxes experienced near the mulch surface are expected to be substantially higher. The error bars represent the standard deviations of the repeated tests.

033

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 27/08/2026

🔥 New in IJWF: Prescott et al. develop a Bayesian model for postfire debris-flow probability in California, quantifying uncertainty in rainfall thresholds and showing that local rainfall climatology can substantially alter watershed hazard estimates. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Map of the southern California basins used in this study. Training basin abbreviations given by Staley et al. (2017): Orange and San Diego Counties (OSD); San Gabriel, San Bernardino and San Jacinto Mountains (SGSBSJ); and Ventura (VEN). Inset at bottom shows the location of the study area within the United States. Basemaps sourced from ESRI (2026a, 2026b).

Time series of observed fifteen-minute rainfall intensity, I15 (black vertical lines), and occurrences of debris flows (black stars) at the (a) Las Lomas and (b) Bond 408 basin outlets compared with predictions of fifteen-minute rainfall intensity thresholds, I15T, computed using the SCMbayes posterior at pc = 0.30 (red histogram and shaded Kernel Density Estimate, KDE) and its Maximum Likelihood Estimate (MLE) at pc = 0.5 (black dashed line). Statistics under each subplot’s label give the mean plus or minus one standard deviation of the SCMbayes ensemble I15T predictions. Dates are formatted in (a) as YYYY-MM; (b) as YYYY-MM-DD.

First postfire year total probability analysis for the (a,c,e) Buckweed 813 and (b,d,f) Station 17816 basins. (a and b) SCMbayes ensemble predictions of the probability that a postfire debris flow (PFDF) occurs conditioned on the annual peak 15 min rainfall intensity, P(y = 1|I15). (c and d) Probability density functions, G, of the calibrated marginal distributions in Fig. 5. (e and f) the cumulative sum of the product between the PFDF probability conditional distribution, the I15 marginal distribution, and the I15 grid spacing (1 mm h−1). The total probability corresponds to the maximum value of one curve at the far right of (e and f). ‘cpi’ stands for the central prediction interval of the respective model ensembles.

054

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 17/08/2026

🔥New in IJWF: West et al. used 29 years of wildfire ignition records and Landsat-derived herbaceous fractional cover in California, finding increasing fire occurrence in areas with >40% herbaceous cover and supporting evidence for a human–grass–fire cycle. 🔗 doi.org/10.1071/WF26... #IJWildlandFire

$Herbaceous fractional cover (HFC) maps from the four Landsat image dates (YYYY/MM/DD) with NLCD ‘Developed’ areas and surface water, as well as the southeast portion not covered by Landsat, masked. Purple pixels represent locations with low HFC (<10%) and red pixels indicate locations with high HFC (>90%).$ $Annual percentage of ignitions in ‘Federal’ lands per fraction of the study area (1992–2020). Line charts represent the fluctuations in the percentage of ignitions per associated fraction of the area represented by HFC interval class (from left to right): 0–19.9%, 20–39.9%, 40–59.9%, 60–79.9%, and 80–100%. Dotted trendlines indicate whether the percentage of ignitions increased or decreased over time. Arrows indicate the HFC interval classes that exhibit increasing percentages of ignitions over time. (a) All 1031 ignitions. (b) 606 human-caused ignitions. (c) 182 naturally-caused ignitions. (d) 116 ignitions resulting in a multi-date fire (>1 day). (e) 101 ignitions resulting in a large fire (≥4.05 ha).$ $Histograms of the percentage of ‘Federal’ ignition points per fraction of the study area (1992–2020) separated by 20% HFC interval classes. Arrows indicate the HFC interval in which the highest percentage of ignitions occurred. (a) All 1031 ignitions. (b) 606 human-caused ignitions. (c) 182 naturally-caused ignitions. (d) 116 ignitions resulting in a multi-date fire (>1 day). (e) 101 ignitions resulting in a large fire (≥4.05 ha).$

084

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 15/08/2026

🔥 New in IJWF: Li et al. identify two major wildfire regions in China and reveal contrasting winter–spring climate–fire pathways: AO-linked snow cover and warming shape soil moisture in the Northeast, while ENSO-linked warming dominates in the Southwest. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Fire regime areas identified across mainland China using (a) a traditional k-means clustering method that identified cells of five identified fire activity types; and (b) the Gaussian-optimized k-means algorithm to identify two areas of high burn activity in the Northeast and Southwest fire regime areas.

Seasonal patterns in area burned and associated environmental conditions in (a) Northeast and (b) Southwest fire regime areas.

Monthly relationships between large-scale climate indices and regional hydroclimatic variables based on extended climate records from 1950–2024. Bars show Pearson correlation coefficients calculated from annual time series of monthly values. Specifically, each monthly correlation was calculated using 75 annual values for that month during 1950–2024, rather than annual-mean indices. Panels show correlations between (a) Arctic Oscillation (AO) and snow cover in the Northeast fire regime region (NFR), (b) AO and temperature in the NFR, (c) Niño3.4 and snow cover in the Southwest fire regime region (SFR), and (d) Niño3.4 and temperature in the SFR during February–April. Asterisks indicate correlations significant at P < 0.01.

043

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 12/08/2026

🔥 New in IJWF: Bennett et al. assess W.I.S.E. fire-spread predictions against 19,848 satellite-derived daily wildfire perimeters across Canada. Default simulations substantially overpredicted spread, while optimizing burn duration and wind direction. 🔗 doi.org/10.1071/WF26... #IJWildlandFire

Simulation of one wildfire burn day for three wildfires showing the true boundary from the Canadian Fire Spread Dataset (CFSDS) and simulated perimeters from scenarios 1, 2 and 3.

Distributions of normalized area difference and F1 score for scenarios 1, 2 and 3. Normalized area difference values near 0 indicate high levels of agreement, while positive values indicate overprediction and negative values indicate underprediction.

Tornado chart depicting the sensitivity of Wildfire Intelligence and Simulation Engine (W.I.S.E.) performance to input variables and fire characteristics in scenario 1. Annotations at either side of each bar provide ranges for bins associated with the worst (left) and best (right) average F1 score. BUI, Build Up Index; DMC, Duff Moisture Code; FFMC, Fine Fuel Moisture Code; FWI, Fire Weather Index; DC, Drought Code; ISI, Initial Spread Index.

064

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 08/08/2026

🔥New in IJWF: Price et al. show that 94% of houses destroyed in the 2019–20 NSW fires were within 100 m of flammable vegetation, supporting current bushfire risk mapping—although extreme fire weather pushed house losses further into surrounding landscapes. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Map of New South Wales showing Bush Fire Prone Land (BFPL) with the footprint of the 2019–20 bushfires overlaid. The locations of the 2475 destroyed houses is indicated.

Flowchart showing how each of the two building datasets (Building Impact Assessment (BIA) and Microsoft Building Footprints (MBF)) were subsetted and used in the four analyses.

Two examples of buildings destroyed more than 400 m from flammable vegetation. (a) The destroyed buildings most distant from flammable vegetation. This area is 20 km west of Rappville, New South Wales (NSW) (maximum 1464 m from flammable vegetation), burnt by the Busbys Flat Rd fire, 8 October 2019 at Forest Fire Danger Index (FFDI) 16. The smaller red dots are destroyed outbuildings and the larger dot is a destroyed house (840 m from flammable vegetation). These buildings were burnt on 8 October 2019 by the Busby’s Flat Rd fire. The overlays are vegetation categories. (b) Cobargo, 488 m from flammable vegetation, burnt by the Badja fire, 31 December 2019 under FFDI 73. The eucalypt (Euc.)-Forest category is the Bush Fire Prone Land (BFPL) category one and the non Euc.-Forest category are the BFPL categories 2 and 3 combined. (c) Shows the location of the two communities and the two fires in NSW.

096

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/08/2026

🔥 New in IJWF: Ren et al. present a high-accuracy workflow for georeferencing drone thermal imagery during a fast-moving prescribed canyon fire. Using fire anatomy, terrain-aware transformation and strategic tie point. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Synchronous Red, Green and Blue (RGB) and Thermal Infrared (TIR) keyframes after 2 min 00 s since the TIR data collection started. The RGB image (a) is covered by heavy smoke, while the TIR image (b) shows clear temperature diversity between pixels.

Georeferenced Thermal Infrared Thermal (TIR) Keyframe, showing tie points in the back burnt area and the right flank, other than in the intensive head fire (at a cumulative TIR video time of 34 min 55 s).

Thermal Infrared (TIR) images of oak trees during the head fire of the same area at different time intervals, from 3 min 00 s (3’00’) to 6 min 00 s (6’00’) since the TIR data collection started. Labels a, b, c and d indicate individual oak trees. Note: RGB = Red, Green and Blue.

053

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 23/07/2026

🔥 New in IJWF: Girardin et al. show that rising atmospheric dryness coordinates boreal forest change: higher VPD suppresses tree growth, increases annual area burned, and indirectly reduces biomass accumulation through linked growth and fire pathways. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Fig. 1. (a) Geographic depiction of the study area located within the forested regions of the Core Study Domain of NASA’s Arctic–Boreal Vulnerability Experiment (ABoVE) in northwestern boreal Canada colored in blue. Plot locations of Canada’s National Forest Inventory tree samples used for this study are indicated by white circles. Fire polygons from the Canadian National Fire Database (CNFDB) are orange colored. (b) ABoVE’s aboveground biomass estimates for 1984 (grid cells of 30 m resolution). (c) ABoVE’s average annual aboveground biomass increment (AGBI) estimates from 1985–2014 for each of the 79 grid tiles, each measuring 180 × 180 km.

Fig. 2. Time series of (a) spring and summer mean daylight vapor pressure deficit (VPD, 1951–2022); (b) annual mean basal area increment (BAI) change (1950–2009); (c) annual sums of area burned (AAB, 1950–2020); and (d) spatially averaged Landsat-derived annual aboveground biomass increments (AGBI, 1985–2014) in northwestern boreal Canada within the NASA Arctic–Boreal Vulnerability Experiment (ABoVE) Core Study Domain. Shaded areas represent 95% bootstrap confidence intervals, reflecting spatial uncertainty. For BAI, the period after 2009 was excluded due to inflated confidence intervals resulting from low sample size (n < 80 trees).

Fig. 4. Structural equation model linking climate drivers to boreal forest productivity. This diagram illustrates the hypothesized causal relationships among vapor pressure deficit (VPD), basal area increment (BAI), annual aboveground biomass increment (AGBI) and annual area burned (AAB) in boreal forests. Arrows represent fully standardized path coefficients (with values indicated) estimated using structural equation modeling (SEM). Red solid arrows indicate statistically significant relationships (P < 0.05), while gray dashed arrows denote non-significant paths.

097

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 22/07/2026

🔥 New in IJWF: Yang et al. show that accounting for uncertainty in forecast fuel moisture improves 1–4 day spatially explicit forecasts of human-caused wildfires across Ontario, reducing underprediction and producing better-calibrated prediction intervals. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Color-coded maps for the predicted fire occurrence probabilities and the observed fire incidents (blue dots) for 11 June 2018 using two-days ahead forecasted Fine Fuel Moisture Code (FFMC) values. The corresponding 95% prediction intervals for the total number of fires were [4, 16] for SIMEX and [0, 8] for naïve. Seven human-caused fires occurred.

043

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 13/07/2026

Lampman et al. combine drone thermal imagery and AI to quantify fire behaviour and accurately predict short-term fire spread during a prescribed grassland burn. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Fire progression intervals represented as Fire Radiative Power (FRPD; fine-scale colored pixels along each progression line 1–7) across the burn unit with rate of spread vectors (ROS; colored lines between Intervals 1–6) from right to left and for the multi-layer perceptron (MLP) predictions (beyond Interval 6). The wind rose indicates the direction the wind was coming from. The ROS rose is for the observed intervals.

Visual workflow for deriving rate of spread, fireline intensity and fire radiative power using UAS-based TIR images of spreading wildland fire. (a) Individual TIR images to orthomosaics using Structure from Motion (SfM); orthomosaics to Fire Radiative Power Density (FRPD) fire front interval rasters using temperature threshold and Stefan–Boltzmann law. (b) Fire front interval rasters converted to polygons and then minimally smoothed with points generated along the edges. (c) Flaming front depth (FD) vectors generated for points within fire front intervals to derive fireline intensity (FI) using the sum of Fire Radiative Power Density (FRPD) pixels along the FD vectors multiplied by the FD length (FRPD-FD method); rate of spread (ROS) vectors generated for points between fire front interval leading edges to measure spread distance and time difference between fire front interval (a) and fire front interval (b). PAEK, Polynomial Approximation with Exponential Kernel.

083

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/07/2026

Douglas: "My PhD project aimed to improve decision-support for land and fire management agencies... I continue to learn and am honoured with this award from the IJWF. Further research will make this work more reflective of the multi-dimensional nature of fire..."

020

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/07/2026

Douglas Radford ( bsky.app/profile/doug...) for "Optimising fuel treatment plans to reduce burn probability", using simulation-optimisation to boost burn probability reduction by 81–284% over existing approaches. 🔗 doi.org/10.1071/WF25080

051

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/07/2026

Kendra: "Motivated by the needs of wildland fire practitioners, this research developed a novel methodology for evaluating fuel treatment effectiveness... I sought to produce research that bridged the gap between current operational challenges and applied fire science."

030

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/07/2026

Kendra Fallon for "A novel methodology to assess fuel treatment effectiveness: application to California's forests", 61% of treated areas modified fire behavior, with wildland fire/fuel removal treatments most effective. 🔗 doi.org/10.1071/WF24220

051

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/07/2026

to the inaugural winners of the International Journal of Wildland Fire Student Paper Prize! This new award celebrates outstanding student research advancing wildland fire science. 🔥🌲 #WildfireScience #Research

020

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 18/06/2026

🔥New in IJWF: Theodori et al. link reconstructed flame and ember exposure from the 2017 Tubbs Fire with structure-level damage data, showing how WUI codes, spacing and vegetation clearance shape destruction probability. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Modeled hazard metrics from the 2017 Tubbs Fire and observed structure outcomes. (Left) Maximum flame length (m) and (right) cumulative ember count, both derived from the ELMFIRE reconstruction over the first 12 h post-ignition at 30 m resolution. Observed structures are overlaid, with green points indicating not destroyed and magenta points indicating destroyed buildings. The final CAL FIRE perimeter (red outline, mapped 31 October 2017) is shown for context. These hazard outputs provide spatially explicit exposure metrics used to develop structure fragility functions.

Location of structures affected by the 2017 Tubbs Fire in relation to California Fire Hazard Severity Zones (FHSZ). The maps distinguish between (left) structures not subject to the California Building Code (CBC) Chapter 7A and (right) structures subject to the code (built post-2008 in an FHSZ). Points indicate structure outcomes: green, not destroyed; magenta, destroyed, with 46.5% destroyed (left) and 47.7% destroyed (right). Shaded areas denote FHSZ categories (very high, high, moderate), and the red outline marks the final Tubbs Fire perimeter.

084

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 10/06/2026

🔥New in IJWF: Johnson et al. surveyed Carolina residents and found support for prescribed burning, with prescribed fire smoke perceived as less concerning than wildfire smoke. The study highlights why smoke-risk communication matters as prescribed fire use expands 🔗 10.1071/WF25193 #IJWildlandFire

(a) Survey responses and locations of state parks conducting prescribed burning. Responses grouped by North Carolina ZIP code. Responses from ZIP codes intersecting a 20 km radius around parks are classified as ‘20 km’ and those within a 20 km radius of a park denoted by a star are classified as ‘High Burn.’ (b) Wildland–Urban Interface (WUI) classification resampled to the North Carolina ZIP code level. WUI areas include ‘intermix’ and ‘interface’ designations. Urban areas include ‘high’ and ‘medium housing density non-vegetated’ designations. Low population density vegetated (LDV) areas include ‘very low density, vegetated’ and ‘uninhabited vegetated’ designations. (c) Total Visible Infrared Imaging Radiometer Suite (VIIRS) 375 m Active Fire and Thermal Anomalies product detections summarized by North Carolina ZIP code for 2015–2019. (d) Average daily smoke PM2.5 concentrations for 2015–2019, as predicted by the National Oceanic and Atmospheric Administration (NOAA) operational smoke forecast. Concentrations are resampled to the North Carolina ZIP code level.

Acceptability of smoke from different sources among survey respondents and respondent categories. Percentage of respondents that selected either ‘acceptable’ or ‘somewhat acceptable’ is listed next to each color bar. Shading indicates respondent category types. Differences in responses significant at α ≤ 0.05 by A NC State Parks burning activity (20 km and High Burn) and land type (Wildland–Urban Interface (WUI), low population density in vegetated areas (LDV), Urban); B land type; and C NC State Parks burning activity, land type and fire detects (High and Low Detects).

Actions respondents are willing to take to reduce exposure to smoke among respondent categories. Participants could choose more than one response. Colored rings indicate the percentage of the respondent group that chose a statement and ring colors correspond to different statements. Gray rings show reference percentages. Shading indicates respondent category types. Note: WUI = Wildland–Urban Interface, LDV = low population density in vegetated areas, AQ = air quality, IAQ = indoor air quality.

293

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 07/06/2026

🔥New in IJWF: Cerovski-Darriau et al. tracked post-fire soil infiltration across northern California fires, finding chaparral soils showed increased hydraulic conductivity after burning, while most vegetation–lithology units recovered to pre-fire conditions. 🔗 10.1071/WF25141 #IJWildlandFire

Fig. 1. Location of 41 field sites that spanned four fires and four simplified lithology types (granitic (i.e. intrusive), volcanic, serpentinite (i.e. peridotite and serpentinite) and sedimentary (i.e. mélange, clastic and meta-clastic sedimentary). The infiltration measurement sites are shown as gray (burned) and white (unburned) semi-circles. Rainfall measurements (Fig. 8) were collected at a subset of the infiltration sites, as indicated by the blue triangles. The rock types are indicated by color, adapted from the US Geological Survey (USGS) State Geologic Map Compilation (SGMC) (Horton et al. 2017), with the continuous extent shown on the statewide inset. The burn perimeters are outlined (black) and labeled by fire abbreviation (LNU, 2020 LNU; WAL, 2020 Walbridge; GLA, 2020 Glass; DIX, 2021 Dixie) (CAL FIRE 2024). All the sites are in northern California (USA), as shown in the inset location map. Location (a) shows the northern California Bay Area sites in predominately sedimentary rocks with some volcanic and serpentinite sites. Location (b) shows the Dixie fire sites in the northeastern Sierra Nevadas, which are mostly granitic with some volcanic and serpentinite sites. The base map is 90-m shaded relief from the Shuttle Radar Topographic Mission (SRTM) digital elevation model (DEM).

Fig. 3. Hydraulic conductivity (Kfs) (mm h−1) for (a) mini disk (MD), and (b) bottomless bucket (BB) measurements in the months after fire separated by lithology. The individual measurements are overlain by boxplots. The boxplots show the interquartile range (25th–75th percentile) of the data with a horizontal line at the median. The boxplots are binned by post-fire number of months with paired burned (orange) and unburned (green) sites grouped with median values denoted with horizontal lines. The number of samples is shown in the corresponding color. Statistically significant differences between burned and unburned groups are shown by *P < 0.05, **P < 0.01, or ***P < 0.001. Note: low sample sizes will decrease statistical power.

Fig. 5. Boxplots of hydraulic conductivity (Kfs) (mm h−1) for (a) mini disk (MD), and (b) bottomless bucket (BB) by hydrologic response units (HRUs). The boxplots show the interquartile range (25th–75th percentile) of the data with a horizontal line at the median. The HRUs group lithology and vegetation classes with boxplots showing burned (orange) and unburned (green) site variability and median values over time. The number of samples is shown in the corresponding color. Statistically significant differences between burned and unburned groups are shown by *P < 0.05, **P < 0.01, or ***P < 0.001. The variability in the early months post-fire is shown with a large spread in the data, including large outliers for the granitic–chaparral (1) sites, which generally decreases through time. BB values are typically order(s) of magnitude higher. Note: not all burned HRUs have paired unburned sites; low sample sizes will decrease statistical power.

142

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 31/05/2026

🔥New in IJWF: Ng et al. show that early dry-season burning supports greater long-term woody carbon accumulation than late dry-season fires across northern Australian savanna fire trials, while site conditions outweigh fire frequency effects. 🔗 doi.org/10.1071/WF26... #IJWildlandFire

042

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 29/05/2026

🔥New in IJWF: White argues that wildfire personnel health is core fire management infrastructure, proposing a readiness–response–recovery framework to better support workforce safety, performance and long-term sustainability. 🔗 doi.org/10.1071/WF26... #IJWildlandFire

Conceptual readiness–response–recovery cycle for treating wildfire personnel health as a managed component of fire operations.

052

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 22/05/2026

🔥New in IJWF: Provencher et al. show that spatial simulations can change vegetation reference conditions, especially in small or long-fire-interval systems, with implications for fuels management. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

042

Reposted by International Journal of Wildland Fire

CSIRO Publishing @csiropublishing.bsky.social · 28/04/2026

We're having a great time at #FBF2026! Visit the CSIRO Publishing stand to learn about our books, journals & writing workshops, plus chat to @pattedplants.bsky.social about publishing in @ijwildlandfire.bsky.social, the #OpenAccess journal of the Int'l Association of Wildland Fire. #IJWildlandFire

The cover of International Journal of Wildland Fire, with the caption "The open access journal of the International Association of Wildland Fire; connectsci.au/wf". In the background is a photo of a firefighter wearing protective gear in a smokey scrubland environment.

032

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 20/04/2026

🔥New in IJWF: Zhao et al. evaluate two common tools for estimating fine dead fuel moisture. The Wiltronics meter performed best at low FMC, while the 10-h fuel stick often underestimated moisture and responded more slowly to environmental change. 🔗https://doi.org/10.1071/WF25174 #IJWildlandFire

055

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 17/04/2026

Congratulations to Dr Xinyan Huang on receiving the 2025 Outstanding Associate Editor Award from the IJWF. An Associate Editor since 2021, Dr Huang has made outstanding contributions to wildfire research and rigorous peer review. A well-deserved honour. 🔗 doi.org/10.1071/WFv3... #IJWildlandFire

063

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 11/04/2026

🔥New in IJWF: Schenk et al. validate WEPPcloud after Arizona’s 2022 Pipeline Fire, showing the model can reasonably estimate post-fire sediment and ash delivery to the WUI, while spatial erosion predictions still need improvement. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

053

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 02/04/2026

🔥New in IJWF: Chakraborty et al. use bivariate LISA to map where wildfire risk and social vulnerability overlap across the continental US, showing priority hotspots mainly in western and southern states. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

041

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 24/03/2026

🔥New in IJWF: Ondei et al. show that airborne LiDAR can reliably capture vegetation cover and spatial arrangement in gardens at the wildland–urban interface, supporting scalable wildfire hazard assessments of defensible space, though hazards still require checks. 🔗 10.1071/WF25218 #IJWildlandFire

Fig. 1. Location of the gardens assessed in this study showing (a) distribution of dry and wet eucalypt forest and
woodland (data from TASVEG 4.0; Department of Natural Resources and Environment Tasmania 2020), and (b)
distribution of the wildland–urban interface (WUI) in the area (data from Chen et al. 2024). To avoid disclosing
the exact location of the gardens, the outline of the statistical areas in which they are located, and the number if
gardens for each statistical area, are shown. The inset shows the location of the study area within Australia.

Fig. 6. Comparison between hazard scores obtained using a combination of field and high point density Light Detection
and Ranging (LiDAR) (Combined HD Hazard) or field and low point density LiDAR (Combined LD LiDAR). Results are shown
for the (a) overall garden, as well as the (b) Fuel-free Zone (FFZ), (c) Open Zone (OZ), and (d) Tree Zone (TZ).

052

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 17/03/2026

🔥New in IJWF: Ellis et al. show that prescribed burns are most likely to limit later wildfire spread when burns are larger, more recent, and linked with roads or active control lines, while extreme fire weather reduces their effectiveness. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

Fig. 1. (a) Distribution of treatments (1999–2019) highlighting the treatment time since fire in years for all encounters with wildfires.
(b) Spatial distribution of treatments that encountered wildfires across the state of Victoria, Australia, with fill colour representing the
equivalent treatment time since fire in (a). (c) Distribution of treatment–wildfire encounters by wildfire season. (d) Spatial distribution of relevant wildfires (2003–2020) across the state of Victoria, Australia, with fill colour representing the equivalent wildfire season in (c). Hatching in (a, c) represents the subset of case study encounters.

Fig. 2. Example of our method to calculate the through-burn percentage of a treatment as a measure of its effectiveness. (a) A treatment–wildfire encounter, with a 50 m buffer around the treatment perimeter. (b) The 50 m buffer is segmented into 1 m lengths (vertices), which are intersected with the relevant wildfire. Wildfire encounters are scored as 1, no wildfire as 0. (c) For each vertex, we calculate the percentage of burnt vertices in the half-perimeter centred on that vertex using a smoothing operation. (d) The through-burn percentage for the treatment is defined as the minimum of these values, which occurs in the central vertex of the half-perimeter furthest from the wildfire. Note that one or more vertices can share this value (in this example, the circled vertices all share the value of 28%). See Supplemental material for the code for this example encounter.

Fig. 5. Shapley value distributions for a subset of predictors in the Victorian fire history and case study record models. Distributions
were averaged for each observation across 25 permutations. Negative values reflect lower likelihood of predicted through-burn for the given predictor value, while positive values reflect a higher likelihood of predicted through-burn. Bars represent the mean values for each variable across all observations, while points represent the median. The range represents 50 and 89% highest density continuous intervals across all observations’ mean Shapley values. Selected predictors include (a) control line presence, (b) road presence, (c) the interaction between treatment time since fire and three key dominant fuel types, (d) landscape time since fire, (e) total treatment area, (f) the Forest Fire Danger Index, (g) the Keetch–Byram Drought Index, and (h) the total daily precipitation. Note that panels (f) through (h) are only for the case study records model.

Fig. 6. Relationship between the through-burn probability and the fire return interval in (a) the adjusted predictions when controlling for
the effects of all predictors for both the Victorian fire history and case study GAMs, and predicted outcomes over the full range of
treatment time since fire values using (b) worst- and (c) best-case scenario parameters for both the Victorian fire history and case study
records models. Note that the dominant fuel type is set to the most representative forest type, forest with shrub, to ensure
the scenarios reflect comparable vegetation classes. (d, e) show the distribution of permuted (B = 200) Shapley values for the worst- and
best-case scenarios, respectively. Bars represent the mean Shapley value for each predictor, while points represent the median. Range
lines represent 50 and 89% highest density continuous intervals for each predictor distribution.

173

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 09/03/2026

🔥 New in IJWF: Baker et al. assess soil heating beneath prescribed burns by UK teams trained to first vegetation fire standards, showing minimal heating at 2–3 cm depth and little evidence of below-ground organic matter loss under mild burning conditions. 🔗 doi.org/10.1071/WF25224 #IJWildlandFire

Fig. 1. Map illustrating the site locations of the management burns
tested, and photographs of the sites in this study. 1. The Cawdor
Estate, Scotland (photographs a – 23 March 2023 burn; b – Fire 1;
c – Fire 2; d – Fire 3); 2. Spaunton Moor, England (photographs a – Fire 1; b – Fire 2); 3. Rempstone Forest, England; 4. Corfe Common, England.

Fig. 2. Lighting patterns and wind direction for each of the prescribed burns monitored. Measurements represent the area covered by thermocouples.

Fig. 3. Maximum soil temperatures recorded from thermocouples during each of the burns. Orange arrows depict the lighting pattern.

Fig. 4. Heat maps demonstrating maximum temperatures reached at the surface during each of the
prescribed burns monitored.

042

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 24/02/2026

This collection will bring together cutting‑edge work from across the conference themes, amplifying the insights of researchers, practitioners, and policymakers working in this space.

010

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 24/02/2026

IJWF is proud to be partnering with the organising team of the 10th International Conference on Forest Fire Research to produce a special collection highlighting the diverse research showcased at this year’s event. Conference Website: events.adai.pt/en/10th-icffr #FireResearch #Science

events.adai.pt

10th International Conference on Forest Fire Research

063

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 15/02/2026

🔥New in IJWF: Chen et al. use CSIRO Pyrotron wind-tunnel burns in eucalypt litter to model fine woody debris consumption (6–50 mm). A two-step ML pipeline predicts full vs partial burnout (74% accuracy) and estimates combustion factor. 🔗 doi.org/10.1071/WF25255 #IJWildlandFire

Fig. 1. (a) Diagram of the dimensions and division of the fuel bed; (b) picture of the fuel bed before fire (under a FWD load of
1.2 kg m−2); and schematic of (c) heading, and (d) backing fire spread modes, where the black rectangles indicate the area of the fuel bed.

Fig. 4. Box plot with overlaid data distribution of the fire behaviour variables and FWD combustion factor in the dataset:
(a) ignition delay and residence time, (b) duration of flaming and smouldering combustion, (c) interval and cumulative flame
rate of spread, (d) interval and cumulative fireline intensity, (e) charring intensity, and (f) FWD combustion factor. In panels,
the number next to the hollow square in the box plot represents the mean and the other quantity presented is the median.

053

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 05/02/2026

🔥New in IJWF: Rhea et al. examine the 2002 Hayman Fire (Colorado) and show stream nitrate export can stay ~19× higher even 17 years later—driven by reduced vegetation N demand and subsurface transport, not higher mineralization. 🔗 doi.org/10.1071/WF25145 #IJWildlandFire

(a) Sampling locations in watersheds
within and adjacent to the 2002 Hayman fire,
Colorado, USA. Triangles mark upland sites
(n = 20). Circles mark near-stream networks,
each of which contained six sites (i.e. riparian,
toeslope and midslope on both stream banks),
shown here as single points due to their close
proximity. Burned sites are orange and unburned
sites are blue. The Cheesman weather station is
denoted by a black square ( WRCC 2021). (b)
Conceptual diagram of the four topographic
positions along a hillslope used to structure
site placement and represent landscape transitions from riparian to upland environments.
USFS, United States Forest Service.

Ion exchange resin (IER) (a) nitrate (NO3 ) and (b) ammonium (NH4
+) accumulation rates separated by summer monsoon and spring snowmelt seasons. This represents plant-available inorganic nitrogen (N) in the top 5 cm of mineral soil. Upslope positions only have winter data. The centerline of the boxplots denotes the median values, the upper and lower limits span the interquartile range, the whiskers include data within 1.5 times the interquartile range and the dots beyond the whiskers are outliers. Fire effect significance is
denoted by *P < 0.1 and **P < 0.05.

Annual terrestrial net primary productivity (NPP) (kg C/ha.year) by functional type for (a) unburned and (b) burned sites. The upper panels represent near-stream sites and lower panels upland sites. NPP is partitioned by plant functional type which is illustrated with color shading. The vertical black line represents the year of the 2002 Hayman Fire.

083

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 05/02/2026

🔥New in IJWF: Finney et al. show wind-driven flame spread along single horizontal fuel particles follows an elliptical relationship with orientation—supporting why wildfire perimeters often approximate ellipses, even at particle scale. 🔗 doi.org/10.1071/WF25135 #IJWildlandFire

Experimental results from quadruple 3.175 mm fuel samples showing: (a) flame spread data and polar ellipse model ( Eqn 2,
Table 3) for each windspeed and wind direction (α); and (b) data and elliptical shapes of fires from the fitted model in Cartesian
coordinates. Note that data are displayed offset from actual α to distinguish different wind speeds.

Composite of photographs showing flame angle from vertical as a function of wind speeds for: (a) single 3.175 mm cardboard
fuel samples, and (b) quadruple 3.175 mm cardboard fuel samples, and examples of parallelogram analysis used for determining flame
angle from vertical for different wind speeds (c) 0.23 ms−1, (d) 0.36 ms−1, and (e) 0.62 m s−1 (aligned with spread direction).

073

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 23/01/2026

🔥New in IJWF: Edwards et al. present a biome-scale fire severity mapping framework for Australia’s tropical savannas using MODIS data and 6478 field sites, achieving 93% accuracy. A major step for biodiversity, emissions, and Indigenous-led fire programs. 🔗 doi.org/10.1071/WF25044 #IJWildlandFire

Fig. 1. Workflow for biome-scale fire severity
mapping framework in northern Australia. The
process integrates MODIS surface reflectance,
MODIS-derived burnt area mapping, active fire
data, field data and classification and validation
steps to generate annual fire severity maps and a
confusion matrix.

Fig. 2. North Australia depicted as four main zones, from north to south: the High Rainfall, Low Rainfall, Southern Savannas and
Australian Mainland. Also delineated are the three north Australian pyro-geographic regions defining the Early and Late Dry Season
thresholds: (R1) Kimberley/Top End West (1 January to 30 June); (R2) Top End East/West Queensland (to 30 July); and (R3) Northeast
Queensland (to 31 August) delineated by the lines of longitude at 132°E and 142°E, respectively. The validation waypoints collected
from 2011 to 2016 are included.

Fig. 4. Scatterplots illustrating the Fire Severity class values of post-fire NIR vs RdNIR (relativised difference
of pre and post NIR) derived from the supervised Random Forest classification in the Early Dry Season (EDS)
on the left, and the Late Dry Season (LDS) on the right, 2016. Although there appears to be considerable
overlap, there are distinct individual values of Severe and Mild reflectance supported by the validation.

093

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 19/01/2026

🔥 Call for Papers: Wildland–Urban Interface Fires 🌳🏠 Submission deadline extended to 1 March 2026. We welcome research on fire dynamics, risk modelling, exposure & impacts, mitigation, recovery, and community resilience. 🔗: connectsci.au/wf/pages/cal... #FireScience #WUI #IJWildlandFire

0138

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 15/01/2026

🔥New in IJWF: Little et al. reveal that persistent high-pressure systems drive spring vegetation fires in England, while summer fires are less weather-dependent. Forecasting fire risk? Look to both surface and synoptic signals. 🔗 doi.org/10.1071/WF25158 #IJWildlandFire

Graphic showing the three layers of
relationships investigated in this paper, from
top to bottom: synoptic weather patterns –
surface fire weather – vegetation fires. Mean
sea level pressure (MSLP) anomalies and
Canadian Fire Weather Index (FWI) values are
for June 26, 2018, the day the Saddleworth Moor
Fire in England was declared a major incident,
which remains one of the largest fires experienced in the UK at 18 km2 ( Graham et al. 2020).
The vegetation fire layer shows all spring (blue)
and summer (orange) vegetation fires >1 ha
recorded between April 2009 and April 2020.

Summary of fire data from the incident recording system (IRS) database. (a) Total number of fires (orange line) and total
burned area (hectares; pale orange bars) of crop, grassland and heathland/moorland vegetation fires in England, 2009–2020. (b) Total
number of fires across the entire data period (2009–2020) in each land cover category during spring (March–May) and summer
(June–August). (c) Total number of fires across the entire data period (2009–2020), on each day of the year. Dashed line represents
the division between spring and summer (31st May).

Ranked percentile curve score (intercept of Theil–Sen regression ± 95% confidence interval) for surface weather metrics on fire
days. Weather variables obtained from E-OBS (‘raw’ weather metrics) are MnT = mean daily temperature; MxT = maximum daily
temperature; GR = global radiation; RH = relative humidity. Weather indices obtained from the Canadian Fire Weather Index System
(CFWIS) are FWI = fire weather index; FFMC = fine fuel moisture code; DMC = duff moisture code; DC = drought code; ISI = initial spread
index; BUI = build-up index. Higher values indicate that the variable performs better at predicting fire occurrence.

0104

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 27/12/2025

🔥New in IJWF: Accary et al. argue that fire-induced wind—often ignored in models—plays a key role in extreme fire behaviour. They call for new experiments and simulations to quantify its feedback and improve predictions. 🔗 doi.org/10.1071/WF25258 #IJWildlandFire

Fig. 1. Characteristic velocities associated to the forces governing
wildfire behaviour on a sloping terrain: wind inertia (parallel to the
ground), buoyancy (vertical) having two components that are parallel
and normal to the direction of fire propagation. Ue: effective wind
speed, Uw: prevailing wind speed, UB: buoyancy characteristicvelocity.

1125

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 18/12/2025

🔥New in IJWF: Cardil et al. present an Initial Attack Assessment (IAA, 1–5) to flag fires likely to escape initial suppression. Analysis of 26,907 California ignitions shows higher IAA(especially terrain and fire-behavior)lead lower initial-attack success. 🔗 doi.org/10.1071/WF24160 #IJWildlandFire

Initial Attack Assessment index (IAA) for Californian fires retrieved by IRWIN from 2020 to 2023
(n = 26,907) considering the initial attack success (left: success (fire size <4 ha); right: fail (fire size >4 ha).
Wildfires were independently simulated using WFA to obtain their corresponding IAA. The actual reported fire
size is represented by graduated circles.

Schematic modeling process that includes the three main research objectives related with the
wildfire data subsets employed and their corresponding univariate logistic regression models.

(a) Fire simulations (n = 360) were automatically conducted for the period of 6–9 January
2025. The colored dots represent the IAA category assigned to each simulation. (b) The number of
incidents per IAA level, along with the IA success rate (%), and the names of escaped (>4 ha) and large
wildfires are shown across IAA classes (1–5). Fire names are followed by their respective burned area in
hectares.

073

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 14/12/2025

🔥 New in IJWF: Ren et al. model future wildfire susceptibility in Portugal with climate and vegetation change. Results show climate-only projections miss local shifts, while climate-driven eucalyptus expansion can increase risk even under low-emission. 🔗 doi.org/10.1071/WF25092 #IJWildlandFire

Predicted wildfire susceptibility maps generated by the Graph Convolutional Network (GCN) using climate data of the
different Shared Socioeconomic Pathways (SSPs) for the year 2060. The vegetation is unchanged. The overall geographical distributions of susceptibility are consistent across all three scenarios. An area of focus is analyzed in depth to understand local effects of
climate predictions (hashed area).

Time series of climate variables average over the northern region across different Shared Socioeconomic Pathways (SSPs)
scenarios. SSP1–2.6 displays a higher average temperature, average minimum temperature, average humidity and precipitation, but a
lower average maximum temperature.

Predicted 2060 species suitability distribution maps generated by the MaxEnt using bioclimate data of the different Shared
Socioeconomic Pathways (SSPs) for the year 2060. The vegetation adapts to the climate scenarios.

095

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 12/12/2025

🔥New in IJWF: Menick et al. test phenological offset corrections for Landsat dNBR/RdNBR across CONUS CBI plots, showing when offsets improve burn-severity–CBI relationships and when automated buffers bias severity low. 🔗 doi.org/10.1071/WF25066 #IJWildlandFire

Example of an identified fire event (a), and corresponding burned (b, c), and unburned (d–f) areas
used in the calculation of burn severity. (a) Landsat 5 post-fire mean image composite collected
June–September 2001, displayed with a false color composite using shortwave infrared 2, near-infrared and
red bands to highlight fire effects. Burned area polygons for the fire, delineated by (b) the Landsat Burned
Area (LBA) product, and (c) Monitoring Trends in Burn Severity (MTBS) datasets. Examples of unburned areas,
commonly used to calculate burn severity offset values, are shown in black and include (d) a manually
delineated unburned polygon of similar vegetation composition to the fire event, as well as an automated
ring buffer (180 m outer, 0 m inner) surrounding (e) LBA, and (f) MTBS fire perimeters.

Performance of models relating field-collected Composite Burn Index (CBI) and satellite-derived burn
severity measures. We contrast model RMSE from (a) dNBR (differenced Normalized Burn Ratio), and (b) RdNBR
(Relativized dNBR) values with and without offset correction across spectral indices, image selection method and
offset type used. Offsets generated manually (white) and from ring buffers surrounding Monitoring Trends in Burn
Severity (MTBS; black) and Landsat Burned Area (LBA; gray) fire perimeter datasets are evaluated.

The relationship between field-collected and satellite-derived burn severity data for the bestperforming automatically generated buffer offset correction methodology. (a) Observed Composite Burn
Index (CBI) burn severity and associated differenced Normalized Burn Ratio (dNBR) values calculated from
mean annual image composites, offset with the unburned dNBR from a 100 m buffered ring surrounding Landsat
Burned Area (LBA) fire perimeters. The red dashed line indicates the fitted exponential relationship between CBI
and dNBRoffset, with associated root mean squared error (RMSE) shown. (b) Exponential model-predicted dNBR
compared with associated observed dNBRoffset values, with red dashed 1:1 identity line and associated R2 value
overlaid.

Mean dNBR (differenced Normalized Burn Ratio) offset values, dNBRunburned, across tested image
selection and offset delineation methodologies for 141 fire events. We compare the influence of inner and
outer ring buffer sizes derived from Monitoring Trends in Burn Severity (MTBS) and the Landsat Burned
Area product (LBA) fire perimeters. The outer buffer indicates the maximum distance from the fire
perimeter and the inner buffer defines the minimum distance from the fire perimeter, in between which
are the pixels used to calculate the offset value.

062

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 09/12/2025

🔥 New in IJWF: Johnston et al. assess four fuel treatments using the Avoided Wildfire Emissions framework, showing that underburning and thinning + underburning meaningfully reduce future wildfire emissions, especially where annual fire probability is high. 🔗doi.org/10.1071/WF25026 #IJWildlandFire

Flowchart showing how each TreeMap2016 forest stand was
assigned a hazard level based on simulated fire behavior and effects
outputs from the Fire and Fuels Extension to the Forest Vegetation
Simulator.

Simulated treatment pattern on a synthetic landscape comprised of stands from TreeMap2016. Each treatment unit is ~49 ha and
the total treated area is about 22% of the total area.

052

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 03/12/2025

🔥New in IJWF: Mackey et al. The 2023 Québec fires burned 4.5 M ha of boreal forest. Using Sentinel-2 CBI mapping the authors show burn severity peaks on dry topographic positions, under extreme fire weather, and in 20–40-year forests. 🔗 doi.org/10.1071/WF24175 #IJWildlandFire

Composite burn index map for fire perimeters analysed in this study (a); inset map of the fire location (b);
classified composite burn severity classes across the study area and severity for large patches are presented in insets
(i), (ii) and (iii).

Partial dependence plots of each variable with the respective relative contributions to the model
in parentheses. The dashed lines represent the partial dependence values, while the solid line shows a
fitted curve with loess smoothing applied. The density distribution of the sampled test data is shown at
the bottom of each plot for continuous variables and alongside predicted response for a categorical
variable. The red dots represent the predicted response for each vegetation type. CBI: Composite Burn
Index.

Bar plot showing the mean forest cover (%) for different forest age classes in Quebec. Analysis is based on
forest cover data from four studies ( Hansen et al. 2013; Sexton et al. 2013; Matasci et al. 2018; Feng et al. 2022) and
age class data from Maltman et al. (2023). The error bars represent the standard deviation (s.d.) of the mean cover
for each age class.

094

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 20/11/2025

🔥New in IJWF: Ohlson et al. map fine dead fuel moisture across a Colorado mixed-conifer forest using 1-h fuel samples from 80 plots. They reveal strong fine-scale spatiotemporal variability shaped by canopy, understory, and aspect. 🔗 doi.org/10.1071/WF25... #IJWildlandFire

(a) Post-treatment forest structure and two people sampling fuels from the 0 to 1.7 m and 2.8 to 4.5 m plots within
the clustered design. (b) A plot placed between two whiskers to mark the distance intervals and the strings stretched
across the PVC sampling frame to denote the randomly generated 20 × 20 cm subplot for collecting the plot’s
observation. Black markings on the PVC sampling frame are at 20 cm intervals to create the grid of 36 subplots. (c)
Two people collecting a dead fine fuel sample into a pre-labeled and pre-weighed polyethylene resealable bag.

The observation period was organized into three phases: early summer (Julian day 138–189), mid-summer (Julian day
192–236) and late summer (Julian day 243–287) for the following plots. The dashed lines on the axes depicting FMC indicate
the commonly used 30% moisture of extinction threshold for dead understory forest fuels ( Rothermel et al. 1986). (a)
Empirical cumulative density functions of the three distributions (early summer, mid-summer, late summer) of observed
fuel moisture content. (b) Kernel density plots of the three intraseasonal period distributions. (c) Boxplots colored by
intraseasonal period depicting the distribution of FMC on each observation day. The box indicates the inter-quartile range
(25th to 75th percentiles). The mid-band indicates the median, and the whiskers indicate points within 1.5 times the interquartile range. Points outside the whiskers are outliers. Note: to visualize the data more effectively, the axes depicting
FMC (%) were visually constrained to 150%, preserving all values while focusing on a more relevant range.

The smoothed, marginal effects of understory cover, canopy cover, heat load index and precipitation on FMC
transformed to show the fitted GAM function on the response scale. The x axis shows values of the covariate, the rug
indicates the distribution of covariate observations, and the y axis shows expected FMC values. The gray bands
correspond to the 95% confidence interval and represent model uncertainty in the transformed (response scale)
estimate. Though negative values are not possible in FMC or under a Gamma distribution, the confidence interval
below 0 in the precipitation plot is an artifact of transformation and reflects wide uncertainty.

094

International Journal of Wildland Fire @ijwildlandfire.bsky.social · 15/11/2025

🔥New in IJWF Sanjuan et al. analyse 33 years of fires in Amazonian canga, showing stark contrasts between long-protected Forest and recently protected Forest. They find land-use history drives fire occurrence while dry-season rainfall controls intensity. 🔗 doi.org/10.1071/WF24... #IJWildlandFire

Fire frequency between 1989 and 2021 and land use and land cover (LULC) in 2021 from the Carajás National Forest, Eastern
Amazon, Pará, Brazil. Forest areas are not delineated; i.e. they correspond to areas outside the anthropised and canga areas. The
embedded pie chart shows the accumulated fire scars per LULC class; note that profound changes in LULC during the observation
period ( Fig. 1) were considered for construction of the chart.

Time series data of the annual proportion of burned areas (%) within Campo Ferruginosos National Park and Carajás National
Forest and possible fire drivers (Amazonian and regional deforestation, annual precipitation and climatic severity) for the period
1989–2021 (a) and correlations between potential fire drivers and the annual proportion of burned areas, separated by area (b). The
dashed lines in (a) represent the 5-year trends. The dashed lines in the scatterplots (b) represent the linear relationships between the
rainfall, climatic severity, and deforestation trends and the annual proportion of burned areas.

092