Near real-time indicators of burn severity in the western U.S. from active fire tracking

Orlandetal. Fire Ecology (2025) 21:55 https://doi.org/10.1186/s42408-025-00407-x ORIGINAL RESEARCH Open Access © The Author(s) 2025. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Fire Ecology Near real-time indicators ofburn severity inthewestern U.S. fromactive fire tracking Elijah Orland1,2* , Tempest D. McCabe1,3, Yang Chen4, Rebecca C. Scholten4, Zeb Becker1,3, Rachel A. Loehman5, James T. Randerson4, Shane R. Coffield1,3, Tianjia Liu6, Alexey N. Shiklomanov1, Kurtis Nelson7, Birgit Peterson7, Melanie B. Follette‑Cook1 and Douglas C. Morton1 Abstract Background Timely information on wildfire burn severity is critical to assess and mitigate potential post‑fire impacts on soils, vegetation, and hillslope stability. Tracking individual fire spread and intensity using satellite active fire data provides a pathway to near real‑time (NRT) information. Here, we generated a large database (n = 2177) of wildfire events in the western United States (U.S.) between 2012 and 2021 using active fire detections from the Visible Infrared Imaging Radiometer Suite (VIIRS) sensor on the Suomi National Polar‑orbiting Partnership (SNPP) satellite and the Fire Events Data Suite (FEDS) algorithm to track large fire growth every 12 h. We integrated fire tracking data with final fire perimeters and burn severity data from the Monitoring Trends in Burn Severity (MTBS) program to evaluate the rela‑ tionship between burn severity and fire behavior metrics derived from the fire tracking approach, including the rate of fire spread and average fire radiative power (FRP) of fire detections for each 12‑h growth increment. Results When stratified by vegetation type, FRP and rate of spread metrics were positively correlated with classified burn severity for each 12‑h growth increment, highlighting the potential to rapidly identify areas of high and low severity burning. In forests, integrated measures of FRP over the fire lifetime captured persistent flaming and smolder‑ ing that compensated for initial differences between AM (01:30) and PM (13:30) fire detections. Predictive modeling of these relationships based on multiple fire behavior indicators and vegetation type from the LANDFIRE program yielded an accuracy of 78% for the separation of unburned/low and moderate/high burn severity classes. Conclusions These results demonstrate the ability to capture within‑fire differences in burn severity using NRT indicators from fire tracking to assist with emergency management and disaster preparedness for post‑fire hazards, such as landslides, debris flows, or changes in stream flow and water quality. As VIIRS data are available within min‑ utes of each satellite overpass in the U.S., rapid estimates of burn severity based on fire tracking can be made days or weeks before a large wildfire is fully contained. Keywords Remote sensing, Post fire, Machine learning, Burn severity, Fire tracking, Fire intensity Resumen Antecedentes La información oportuna sobre la severidad de los incendios es crítica para determinar y mitigar los potenciales impactos post‑fuego sobre los suelos, la vegetación, y la estabilidad de las laderas de montaña. El seguimiento de la velocidad de propagación de cada incendio activo usando datos de satélites, provee de una vía rápida para obtener información en tiempo real (NRT). En este trabajo, generamos una gran base de datos de *Correspondence: Elijah Orland [email protected]; [email protected] Full list of author information is available at the end of the article Page 2 of 18 Orlandetal. Fire Ecology (2025) 21:55 eventos de incendios (n = 2.177) en el oeste de los EEUU entre 2012 y 2021, usando detecciones activas del sensor de imágenes radiométricas infrarrojas (VIIRS) del satélite Suomi National Polar-Orbiting Partnership (SNPP), y de un algo‑ ritmo de un conjunto de datos de eventos de incendios (FEDS). Esto permite, cada 12 horas, rastrear el crecimiento de grandes incendios. Integramos los datos de rastreo de estos incendios con los perímetros finales del fuego y con las detecciones cada 12 horas sobre la severidad de estos incendios basados en el programa de monitoreo de las ten‑ dencias en severidad de las quemas (MTBS), para evaluar la relación entre la severidad de los incendios y las métricas de comportamiento del fuego derivadas de la aproximación del rastreo, incluyendo la tasa de propagación del fuego, y el promedio del poder radiante del fuego (FRP) de las detecciones para cada 12 h de incremento en el crecimiento del fuego. Resultados Cuando fueron estratificados por tipo de vegetación, el FRP y las métricas de propagación fueron positi‑ vamente correlacionadas con la clasificación de la severidad del fuego cada 12 h de incremento en el crecimiento del incendio, subrayando el potencial para identificar rápidamente áreas de alta y baja severidad del fuego. En bosques, las medidas integradas de FRP sobre la duración del incendio capturaron llamas y material en combustión de manera permanente, que compensaron las diferencias iniciales de detección entre la 01:30 AM y las 13:30 PM. El modelo pre‑ dictivo de esas relaciones basadas en indicadores múltiples de comportamiento del fuego y tipos de vegetación del programa LANDFIRE, tuvieron una exactitud del 78% para la separación de clases de severidad de “no quemado/baja”, y “moderado/alta intensidad”. Conclusiones Estos resultados muestran la habilidad para capturar las diferencias de severidad entre fuegos usando indicadores del rastreo de NRT y poder asistir con el manejo de la emergencia y la preparación del desastre y los peligros del post fuego como los deslizamientos de laderas, el flujo de residuos, o cambios en el flujo de las corrientes de los arroyos y la calidad del agua. Dado que los datos de VIIRS están disponibles dentro de pocos minutos luego de que el satélite haya sobrevolado los EEUU, las estimaciones rápidas de la severidad basadas en el rastreo del fuego puede hacerse muchos días o semanas antes de que un gran incendio sea totalmente contenido. Background Wildfires have substantial and interconnected impacts on vegetation, soils, and hydrology (Bowman etal. 2009) that vary as a function of burn extent and intensity (e.g., Adams 2013; Coop etal. 2019; Schwilk & Ackerly 2001). Fuel consumption and fire-induced vegetation mortality alter biogeochemical cycles and contribute to greenhouse gas emissions (e.g., Crutzen & Andreae 1990; Hao & Liu 1994; Kasischke etal. 1995; Seiler & Crutzen 1980; Van Der Werf etal. 2003, 2017). Post-fire changes in vegetation structure and composition also change surface albedo (Randerson etal. 2006), reduce the infiltration capacity of burned soils (Debano 2000; Letey 2001), trigger soil nutrient and chemical losses (Alexakis etal. 2021; Chen et al. 2010; Neff et al. 2005; Rovira et al. 2012), and increase the available sediment for mobilization downslope (Florsheim etal. 1991, 2016; Gabet 2003; Lamb etal. 2011, 2013). Combined, these impacts lead to longitudinal changes to the hydrologic cycle, including reductions in evapotranspiration (Ahmad etal. 2024; Bond-Lamberty etal. 2009; Kang etal. 2006; Roche etal. 2018) an increase in overland flow (Scott etal. 1998; Vega & Díaz-Fierros Viqueira 1987; Wells 1981) and greater risk of catastrophic debris flows (Cannon 2001; Cannon & DeGraff 2009; Kean etal. 2011; Lancaster etal. 2021). As wildfires in the western United States and other fire-prone regions become more frequent and intense (Abatzoglou & Williams 2016; Cunningham etal. 2024; Mueller etal. 2020; Westerling etal. 2006), the expedited delivery of burn severity data is crucial for assessing fire effects and for allocating resources to manage post-fire hazards. One common approach to assess burn severity relies on pre-fire and post-fire satellite imagery to estimate the differenced or"delta" normalized burn ratio (dNBR)–a metric sensitive to the loss of live vegetation cover and soil exposure following burning (Eidenshink etal. 2007; Key & Benson 2006). As such, mapped dNBR within a burn scar based on Landsat or Sentinel-2 imagery is a common input for classifying burn severity, as used in standard products from the Monitoring Trends in Burn Severity (MTBS) or Burned Area Emergency Response (BAER) programs. These same image sources can also be used to derive alternative indices such as the Relativized dNBR (RdNBR) (Miller & Thode 2007) or the Relativized Burn Ratio (RBR) (Parks etal. 2014). Remote sensing-based metrics of burn severity vary in their ability to accurately represent field conditions, as index suitability changes based on fuel type and intended use case (Epting etal. 2005; Miller & Thode 2007; Morgan etal. 2014; Parks etal. 2014; Picotte & Robertson 2011; Whitman etal. 2018). One of the key limitations of using remote sensingbased indices for assessing burn severity is the need Page 3 of 18 Orlandetal. Fire Ecology (2025) 21:55 for post-fire imagery, given that image acquisition during or immediately following a wildfire may be delayed by clouds, smoke, or satellite revisit time. This time gap between the burn date and assessment date introduces uncertainty in the estimate of burn severity and delays the use of these data for situational awareness in response to a fire event. For groups tasked with emergency response—such as BAER teams in the United States—uncertainties tied to the availability of cloud-free imagery represent a barrier for responsive planning and management based on delays in mapping efforts to identify areas of elevated risk requiring immediate assessment and/or treatment. Additionally, longitudinal fire impacts such as delayed tree mortality may only become visible during the following growing season; for capturing these effects, “extended” MTBS assessments traditionally rely on imagery acquired 1 year after the fire to provide a more thorough picture of vegetation response (Key & Benson 2006; Eidenshink etal. 2007). Given these limitations, there is an opportunity to develop near real-time (NRT) approaches that draw on complementary satellite information to assist with burn severity assessments during and immediately following a large fire event prior to the availability of standard MTBS and BAER products. One pathway for anticipating estimates of burn severity is to leverage pre-fire information. For example, using a combination of airborne light detection and ranging (lidar) and satellite-based land surface albedo measurements, Fernández-Guisuraga et al. (2021) examined the link between pre-fire vegetation structure and burn severity, highlighting the correlations between canopy height and volume with Composite Burn Index (CBI) and dNBR values. Similarly, Staley etal. (2018) linked historical distributions of dNBR values with existing vegetation type (EVT) classifications derived from LANDFIRE products (Rollins 2009). Using machine learning or related methods, many data-driven studies also demonstrate the important control of elevation on burn severity (Dillon etal. 2011; Estes etal. 2017; Holden etal. 2009; Wu etal. 2013). Finally, fire spread simulations (e.g., Finney 2006; Finney etal. 2011; Linn etal. 2002, 2020; Mell etal. 2007) have also been used to model fire behavior and serve as “scenario-based” assessments prior to burning. To date, these approaches have not incorporated active fire information made available in NRT to account for diurnal or day-to-day variability in fire behavior or intensity. Satellite active fire detections provide information on the location and intensity of fire activity, and data are typically available within minutes to hours after each satellite overpass. For example, the Moderate Resolution Imaging Spectroradiometer (MODIS) sensors on NASA’s Terra and Aqua satellites have already provided over 20years of active fire detections at 1-km resolution (Giglio etal. 2016) and daily burned area estimates at 500-m resolution (Giglio 2018). The wealth of MODIS active fire and burned area data has spurred a range of approaches to delineate individual fire events, both on regional and global scales (Andela etal. 2019; Archibald & Roy 2009; Balch etal. 2013, 2020; Hantson etal. 2015; LizundiaLoiola etal. 2020; Loboda & Csiszar 2007; Scaduto etal. 2020; Veraverbeke etal. 2014). However, many of these products rely on datasets not available in NRT, and thus are most appropriately used for retrospective analysis. Improvements in spatial resolution, sensitivity, and geolocation accuracy of active fire detections from the Visible Infrared Imaging Radiometer Suite (VIIRS) sensors (Schroeder et al. 2014) support new approaches to track individual fire events every 12h (Andela etal. 2022; Chen et al. 2022). Recent work by Chen et al. (2022) introduced the Fire Events Data Suite (FEDS), an approach to use NRT active fire observations from the Suomi-NPP VIIRS sensor to iteratively track and reconstruct fire progression in 12-h intervals for the state of California from 2012 to 2020. The resulting FEDS data provide unprecedented insight into the variability in fire spread rate and intensity of large wildfires, thus promoting a framework to explore the relationships between fire behavior and burn severity. In this study, we applied the FEDS algorithm to create a dataset of individual fire events for the Western U.S. from 2012 to 2021, aiming to systematically investigate the relationship between active fire characteristics and burn severity. We evaluated the potential for using multiple metrics of fire behavior derived in NRT from FEDS, exploring the tradeoffs between accuracy and latency for rapid assessments of burn severity. As wildfires in the U.S. can burn for weeks or months, these NRT indicators may fill an unmet need by providing timely updates on burn severity. Such information is crucial for situational awareness and responsive action both during and immediately after wildfire events. Methods Fire tracking We used the FEDS algorithm (Chen et al. 2022) to generate 12-hourly fire progression data for the western U.S. from 2012 to 2021. The FEDS algorithm uses VIIRS 375-m active fire detections (Schroeder et al. 2014) to track individual fire progression at 12-h intervals that correspond to the cadence of VIIRS overpasses for a given area, with daily overpasses occurring at approximately 01:30 and 13:30 local time. Active fires are detected as thermal anomalies by the VIIRS sensor, where each 375-m active fire pixel indicates likely flaming or smoldering fire activity. Theoretical detection limits for sub-pixel burning in the VIIRS 375-m data product Page 4 of 18 Orlandetal. Fire Ecology (2025) 21:55 are reported to be as fine as 5-m2, with this threshold varying as a function of day/night thermal contrast between fires and background conditions at the time of overpass, as well as the level of smoke or cloud obscuration (Schroeder etal. 2014). VIIRS active fire detection data contain supplementary information such as confidence flags, infrared brightness temperatures, and estimated fire radiative power (FRP) in megawatts (MW), representing the rate of energy output for that pixel at the time of observation. FRP can be directly linked to the rate of biomass combustion (Wooster etal. 2005) and therefore is used as a snapshot indicator of fire intensity and emissions at the time of satellite overpass. To reconstruct the progression of historical fires, we used archived 375-m VNP14IMGML active fire location data to track the progression of all fires in the western U.S. from 2012 to 2021. The resulting dataset provides temporally consistent observations of fire spread in discrete 12-h periods across the study domain. Additionally, FEDS data capture multiple properties relevant to tracking active fire behavior, such as FRP, fire spread rate, and fire line length. As the FEDS algorithm was developed with a focus on tracking wildfires in California (Chen etal. 2022), the algorithm’s application to the larger domain of the western U.S. in this study included minor improvements in the efficiency of the underlying clustering and merging components of the workflow to meet the increased computational demand. Additionally, fire tracking for the western U.S. region used projected coordinate systems (e.g., the US National Atlas Equal Area system, EPSG:9311) in contrast to the World Geodetic System (WGS) 84 geographic coordinate system (EPSG: 4326) used in Chen etal. (2022). See Data Availability for more information on data and code access. Because the FEDS algorithm relies on the preprocessed VIIRS-based data products outlined in Schroeder etal. (2014) and Schroeder & Giglio (2016), the same limitations discussed therein apply. This includes the possibility of false positive detections from static source hot spots and false negatives due to cloud or smoke cover. In this analysis, we included all FEDS fire objects that intersected MTBS perimeters designated as wildfires with matching ignition dates within 10 days. Because MTBS includes all fires > 1000 acres in the western U.S., smaller fires in the FEDS database were excluded from this study. We computed the intersection-over-union (IOU) for all matches to allow additional filtering based on a quantitative representation of their spatial agreement. In total, the final dataset contains 2177 matched wildfires in the western U.S. between 2012 and 2021, representing a total fire-affected area of over 166,722 km2 as mapped by the FEDS algorithm (Fig.1). Calculation offire spread rate, intensity, andpersistence We analyzed fire spread based on the individual “increments” of fire growth during each 12-h interval (Fig.2). Individual increments of fire growth were delineated by taking the geometric difference between fire perimeters derived at time t and those derived 12h later. Each individual area of fire growth was assigned a unique index such that multiple segments of fire spread (each with different directions and locations on the fire perimeter) during the same 12-h period were tracked separately. We refer to these areas as “spread increments” or “growth increments,” whose rate of growth can be expressed in units of km2/12-h. Each increment is categorized based on the timing of initial detection: increments marked as “PM” were constructed using active fire detections first observed at 13:30 local time. These increments include instantaneous measures of PM fire behavior (e.g., FRP) but nonetheless represent morning fire growth between 01:30 (the preceding overpass) and 13:30 (the current overpass) (Fig.2). Similarly, growth increments linked to the AM overpass (01:30) mark afternoon fire growth. To record information on fire intensity, we performed a spatial join between all spread increments and all VIIRS detections recorded for that fire. Notably, we recorded all pixels detected within each spread increment, including fire detections from the initial period of fire spread and any persistent burning detected within each increment over the lifetime of the fire. This approach provided multiple metrics of fire intensity and fire persistence (duration). For each increment of growth, we calculated the mean, maximum, and area normalized total FRP. Fire persistence was estimated using two metrics: (1) the number of unique 12-h periods with one or more VIIRS active fire detections within a given spread increment; and (2) the time difference between the first and last active fire detections, measured in hours. For comparison with MTBS burn severity data, we performed zonal statistics between individual spread increments and classified MTBS pixels, recording the median MTBS pixel class within each polygon. MTBS severity classes in this analysis range from low/unburned (1), low (2), moderate (3), and high (4) severity; classes not pertaining to these groups, such as those representing enhanced regrowth (5) or no data (0), were excluded. Comparisons with MTBS data included both initial assessments focused on immediate fire impacts in low biomass systems, such as grasslands or small shrublands, as well as extended assessments using remote sensing imagery 1 year after the fire to capture delayed ecosystem effects in high biomass environments like dense shrublands or forests (Key & Benson 2006; Eidenshink et al. 2007). We analyzed the combined dataset using both assessment types and separately evaluated the Page 5 of 18 Orlandetal. Fire Ecology (2025) 21:55 Fig. 1 Western U.S. wildfires from 2012 to 2021 included in the analysis (n = 2177), colored by the number of 12 h spread increments in each fire (total n = 56,700) Fig. 2 Example of the differencing method used to isolate individual areas of fire spread based on the 2021 Sugar Fire in California. a FEDS perimeter on the afternoon of July 9th, 2021 (13:30). b The FEDS perimeter 12 h later at 01:30. c The geometric difference between these two perimeters highlighting areas of fire spread in that 12 h period, referred to in this work as fire growth or spread increments. Red shading in panel b denotes active fire detections during the 01:30 overpass, and areas of individual fire spread represent all growth that occurred between t0 and t1 Page 6 of 18 Orlandetal. Fire Ecology (2025) 21:55 influence of initial versus extended assessment data on the relationships between fire behavior and burn severity. We used the LANDFIRE data products (Rollins 2009) matched to the appropriate fire year to estimate the most common existing vegetation type (EVT) prior to each fire to stratify the analysis of fire behavior and burn severity by vegetation type. Finally, to ensure proper agreement between FEDS and MTBS products, we computed the overlap between each 12-h growth increment and the corresponding MTBS fire perimeter. Only increments with at least 50% overlap with MTBS were included in the analysis. This threshold retained 84% of the total spread increment dataset (n = 47,098), demonstrating broad agreement between FEDS and MTBS despite more than an order of magnitude difference in the spatial resolution of their source data (375-m vs 30-m, respectively). Predictive modeling ofvegetation burn severity We developed two models to explore the potential to predict final MTBS burn severity class using FEDS data, where we tested different versions of the decision tree-based ensemble model, XGBoost (Chen & Guestrin 2016)—a model with recent applications in studies related to burn severity and remote sensing (e.g., He etal. 2024; Seydi etal. 2024). The first (multiclass) model predicted the median MTBS class within each increment of growth, and the second (binary) model provided the probability that the median MTBS class within each increment was moderate/high severity (1) or not (0). Each model was trained via a grid search, varying the decision tree count from 25 to 1000 and tree depth from 1 to 5. This strategy was chosen to achieve reasonable performance while being sensitive to overfitting and diminishing returns of continued model training. Input features included growth increment spread rate, the number of unique detection periods, the most commonly occurring LANDFIRE EVT value, designation of initial AM or PM observation, and FRP characteristics summarized as the mean, maximum, and area-normalized total (sum) on both the day of initial spread and over the lifetime of the fire. LANDFIRE EVT data included the four-digit EVT code, in addition to the simplified “EVT_ LF” and “EVT_PHYS” variables. Lastly, the ecoregion in which the fire occurred—as defined by Olson etal. (2001)—was included to provide a secondary, regional representation of ecological context. For each classification scheme (multiclass or binary), we compared models with three different sets of input variables: (1) a model with EVT and ecoregion characteristics only; (2) a model with active fire characteristics only; and (3) a model with the combination of all characteristics. Similar to the area-based filtering threshold used for the data analysis, model training data included spread increments with at least 50% overlap with MTBS perimeters from fires occurring between 2012 and 2020 (n = 39,460, across 1949 wildfires) with testing occurring on fires in 2021 (n = 9000, across 221 wildfires). Evaluation metrics included accuracy, precision, recall, f1-score, and the area under the curve (AUC, binary only). No overlap criteria were applied to the testing data to simulate NRT application, consistent with higher expected uncertainty regarding up-to-date reference perimeter data availability at the time of satellite acquisition. Results Active fire properties andburn severity At the event level, filtered and matched FEDS fires between 2012 and 2021 (n = 2177, Fig.1) in the western U.S. primarily burned conifer forests (76%), with smaller contributions from areas dominated by shrublands (14%) and grasslands (3.3%). Within each dominant vegetation type, median MTBS class values and FRP for each 12-h growth increment were positively correlated (Fig.3, Table 1). In spread increments dominated by conifer forests, median total FRP per unit area (MW/km2) was 28% higher in areas designated as high burn severity (4) as compared to moderate severity (3) (Fig.3a). This difference was even greater when comparing the distribution of mean FRP values for each increment, where FRP in high severity spread increments was 41% higher than in moderate severity increments (Fig.3b). Overall, distributions of both the mean and area-normalized total FRP values for conifer spread increments were statistically different when comparing neighboring severity classes (Mann–Whitney U test, two-sided, p < 0.01). Shrubland and grassland dominated fire growth increments exhibited similar relationships between FRP and MTBS burn severity (Fig.3a–b, Table1). For both vegetation types, the stepwise positive relationship between mean FRP and burn severity was more consistent than for areanormalized total FRP. Shrublands had the highest mean FRP per burn severity class of the three vegetation types, consistent with evidence for hotter fires in shrub ecosystems based on fuel characteristics (e.g., Burger & Bond 2015; De Luis etal. 2004; Keeley etal. 1999). Mean FRP was not statistically different between high and moderate MTBS classes in shrublands or grasslands. Small sample sizes for high burn severity increments in grasslands and shrublands may partially contribute to this finding (see Table1); heterogeneity in vegetation cover may also lead to less consistent relationships between FRP and burn severity in grasslands. Notably, metrics of total FRP exhibited greater differences across class categories when grouped into unburned/low (1 and 2) and Page 7 of 18 Orlandetal. Fire Ecology (2025) 21:55 moderate/high (3 and 4) severity classes, illustrating the potential to tailor NRT metrics to support specific information needs for emergency response. Higher MTBS severity consistently corresponded to faster spread rates for conifer and shrubland vegetation types from unburned/low (0) to moderate (3) severity (Fig.3c). For shrubland-dominant increments, those classified at high (4) severity (n = 39) were not considered statistically different from moderate severity (n = 1206). Differences between sample sizes likely affect these results. For conifers, median spread rates were marginally slower at high severity (0.74 km2/12 h) vs moderate severity (0.88 km2/12 h), and these differences were statistically different. For grassland environments, rates of spread increased between the unburned/low (1) and low (2) severity classes, and differences between the moderate (3) and low (2) categories were not statistically different. Median spread rates declined by 71%—the highest among all classes—between moderate and high severity categories. In addition to the limited sample size in the highest severity class (n = 12), the 12-h revisit time of the VIIRS sensor may not be sufficient to capture the fastmoving nature of grassland fires. Differences in fire behavior by vegetation type underscore the value of fire tracking for assessing ecological impacts of fire activity (Fig.4). For example, shrubland and grassland fires spread faster than fires in coniferdominated landscapes (Fig. 4a). The median values of VIIRS-based spread rates for grassland and shrubland classes were 53% and 106% higher, respectively, than for fire spread increments in conifers. Conifer-dominated growth increments burned longer than those dominated by other vegetation types, with median fire persistence of five 12-h periods (Fig.4b). Fire persistence was also more variable in conifer forests, measured as the hours between the first and last active fire pixel within each increment (Fig.4c), where the median duration was 96 h (interquartile range: 168 h) compared to 12 h in both shrubland and grassland spread increments (interquartile range: 60 h). As such, more persistent fire activity in conifer environments boosted total FRP per unit area, consistent with the expected influence of elevated fuel loading, fuelbed depth, and fuel particle heat content in forested ecosystems on fire behavior (Rothermel 1972). The influence of initial versus extended MTBS assessment type on the relationship between burn severity and metrics of fire behavior varied by vegetation type. Estimated burn severity for conifer increments was largely sourced from extended assessments (n = 30,345 of 35,740, or 85%). As a result, the relationships between burn severity and fire behavior metrics were comparable between extended assessments (Fig. S1) and the combined data shown inFig.3. For conifer increments with initial assessment data, the overall patterns remain unchanged, but the distributions of mean FRP and fire spread rate were higher across all severity classes than in the combined dataset (Fig. S2). By contrast, most MTBS data for shrubland and grassland growth increments were drawn from initial assessments (58% and 60%, respectively). For shrublands, the overall relationships were Fig. 3 FEDS properties delineated by dominant vegetation type and MTBS burn severity class. “ns” designation indicates non‑significant variable difference between the assigned burn severity class and the measurements in the class directly below it. a Area normalized cumulative FRP measurements, per increment, over the lifetime of the fire. Mean increment FRP values over the lifetime of the fire. c 12‑h spread rate as measured by the increment’s area. Whiskers represent the 5th and 95th percentiles, and outliers are not shown Page 8 of 18 Orlandetal. Fire Ecology (2025) 21:55 Table 1 Differences in spread rate, area normalized total FRP, and mean FRP across conifer, shrubland, and grassland environments for each MTBS class Spread rate (km2/12h) Total FRP/spread area (MW/km2) Mean FRP (MW) 5th percentile 50th percentile 95th percentile 5th percentile 50th percentile 95th percentile 5th percentile 50th percentile 95th percentile Sample size Dominant vegetation type Median MTBS class Conifer Low/ unburned (1) 0.12 0.54 4.84 3.96 55.48 362.41 0.86 4.87 22.97 2112 Low (2) 0.13 0.71 7.49 10.39 115.01 639.6 1.47 6.85 34.17 21,205 Moderate (3) 0.13 0.88 14.19 28.72 243.06 1101.3 2.66 12.17 65.98 10,672 High (4) 0.12 0.74 16.01 22.05 310.62 1478.78 2.73 17.14 104.36 1751 Shrubland Low/ unburned (1) 0.11 0.68 10.82 3.64 67.19 557.51 0.96 7.27 36.21 513 Low (2) 0.11 1.57 21.13 4.48 69.07 599.68 1.25 10.32 59.77 4697 Moderate (3) 0.18 2.08 30.78 7.97 173.39 905.88 2.71 20.54 112.93 1206 High (4) 0.11 1.11 8.43 8.17 182.41 849.17 1.44 27.11 126.31 39 Grassland Low/ unburned (1) 0.11 0.5 2.55 3.16 56.75 384.46 0.96 6.44 26.97 185 Low (2) 0.11 1.31 16.57 3.36 55.51 482.89 1.09 8.01 47.04 1151 Moderate (3) 0.11 1.27 17.53 6.97 121.49 1041.39 2.18 13.24 74.35 204 High (4) 0.1 0.37 11.76 20.7 403.34 1688.52 1.94 15.35 83.4 12 Page 9 of 18 Orlandetal. Fire Ecology (2025) 21:55 consistent between initial assessment data and the combined dataset, but with a clearer separation of median fire intensity and higher spread rates by initial assessment class. Initial assessment data for grasslands also provided greater separability by burn severity class for mean FRP and cumulative FRP metrics. Remaining shrubland and grassland data sourced from extended assessments exhibited higher intensity measures and lower spread rates than those shown in (Fig. S1). Diurnal behavior The 12-h cadence of the VIIRS observations further allows for the comparison of differences in fire behavior metrics between nighttime and daytime overpasses, including differences in spread rate and intensity across severity classes. MTBS burn severity class distributions separated by AM/PM overpass designation were statistically different from one another (p < 0.01), where spread increments tied to PM VIIRS active fire detections exhibited higher severity classes overall. Indeed, when limiting normalized total FRP values to only the time of the initial VIIRS overpass, intensity values were consistently higher for initial PM observations than initial AM observations across all vegetation types and burn severity classes (Fig.5a–c). Observed differences between AM and PM VIIRS overpasses are consistent with the expected diurnal cycle of fire intensity, with more intense burning during afternoon hours due to higher temperatures, lower relative humidity, and often higher wind speeds (Andela etal. 2015; Giglio 2007). Integrating over the lifetime of each fire event resulted in more even estimates of FRP (Fig. 5d–f). Considering the full lifetime of the fire, the ratio of PM/AM cumulative FRP aggregated across all increments varied by a factor of approximately two or less in conifer systems, with higher observed ratios in shrubland (approximately 2–4×) and grassland (approximately 3×, excluding outliers) ecosystems (Table2). For conifers, high severity increments were relatively evenly distributed across both periods of morning and afternoon growth, highlighting the influence of fire persistence on burn severity, where longer duration burning leads to more complete fuel consumption in higher fuel load systems. Conversely, the time of initial fire spread may have a stronger influence on burn severity in shrubland and grassland ecosystems—especially those considered for initial assessments only. We also observed diurnal variation in spread rates, with larger afternoon spread across nearly all burn severity classes (Fig.6a–c). In conifer forests, elevated afternoon fire spread rates are consistent with expected behavior and supported by the strong differences in the initial AM and PM FRP measurements. For example, higher PM (13:30) FRP values track daily meteorological conditions (e.g., higher temperatures and lower humidity) amenable to greater afternoon (13:30 to 01:30) fire spread. Furthermore, the increasing ratio between afternoon and morning spread rates across MTBS classes in conifer environments points to the contributions of diurnal variability in behavior on burn severity, where higher burn severity classes are observed to coincide with periods of increasingly faster afternoon spread (Fig.6d). For shrublands burned at moderate severity, afternoon spread rates were about one and a half times as fast as morning spread rates (Fig.6e), before dropping to below 1 × at the highest severity class. This subsequent decrease may be the Fig. 4 a Distributions of increment spread rate for all fires stratified by dominant vegetation type (km2/12 h). b Distributions of the number of unique periods in which one or more VIIRS pixel(s) were detected within an individual spread increment, stratified by dominant vegetation type. c Distribution of the hours between the first and last active fire detections within a spread increment, stratified by dominant vegetation type. Whiskers represent the 5th and 95th percentiles, and outliers are not shown Page 16 of 18 Orlandetal. Fire Ecology (2025) 21:55 Ahmad, S.K., T.R. Holmes, S.V. Kumar, T.M. Lahmers, P.‑W. Liu, W. Nie, A. Getirana, E. Orland, R. Bindlish, A. Guzman, C.R. Hain, F.S. Melton, K.A. Locke, and Y. Yang. 2024. Droughts impede water balance recovery from fires in the Western United States. Nature Ecology & Evolution 8 (2): 229–238. https:// doi. org/ 10. 1038/ s41559‑ 023‑ 02266‑8. Alexakis, D., I. Kokmotos, D. Gamvroula, and G. Varelidis. 2021. Wildfire effects on soil quality: Application on a suburban area of West Attica (Greece). Geosciences Journal 25 (2): 243–253. https:// doi. org/ 10. 1007/ s12303‑ 020‑ 0011‑1. Andela, N., J.W. Kaiser, G.R. Van Der Werf, and M.J. Wooster. 2015. New fire diur‑ nal cycle characterizations to improve fire radiative energy assessments made from MODIS observations. Atmospheric Chemistry and Physics 15 (15): 8831–8846. https:// doi. org/ 10. 5194/ acp‑ 15‑ 8831‑ 2015. Andela, N., D.C. Morton, L. Giglio, R. Paugam, Y. Chen, S. Hantson, G.R. Van Der Werf, and J.T. Randerson. 2019. The global fire atlas of individual fire size, duration, speed and direction. Earth System Science Data 11 (2): 529–552. https:// doi. org/ 10. 5194/ essd‑ 11‑ 529‑ 2019. Andela, N., Morton, D.C., Schroeder, W., Chen, Y., Brando, P.M., and Randerson, J.T. 2022. Tracking and classifying Amazon fire events in near real time. Science Advances, 8(30):eabd2713. https:// doi. org/ 10. 1126/ sciadv. abd27 13 Anderson, H.E. 1982. Aids to determining fuel models for estimating fire behavior. U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station. https:// doi. org/ 10. 2737/ int‑ gtr‑ 122 Archibald, S., and Roy, D.P. 2009. Identifying individual fires from satellite‑ derived burned area data. 2009 IEEE International Geoscience and Remote Sensing Symposium, III‑160‑III–163. https:// doi. org/ 10. 1109/ IGARSS. 2009. 54179 74 Balch, J.K., St. Denis, L.A., Mahood, A.L., Mietkiewicz, N.P., Williams, T.M., McGlinchy, J., and Cook, M.C. 2020. Fired (Fire events delineation): An open, flexible algorithm and database of us fire events derived from the modis burned area product (2001–2019). Remote Sensing, 12(21):1–18. https:// doi. org/ 10. 3390/ rs122 13498 Balch, J. K., B. A. Bradley, C. M. D’Antonio, and J. Gómez‑Dans. 2013. Introduced annual grass increases regional fire activity across the arid western USA (1980–2009). Global Change Biology 19 (1): 173–183. https:// doi. org/ 10. 1111/ gcb. 12046. Balch, J. K., J. T. Abatzoglou, M. B. Joseph, M. J. Koontz, A. L. Mahood, J. McGlinchy, M. E. Cattau, and A. P. Williams. 2022. Warming weakens the night‑time barrier to global fire. Nature 602 (7897): 442–448. https:// doi. org/ 10. 1038/ s41586‑ 021‑ 04325‑1. Balch, J.K., V. Iglesias, A.L. Mahood, M.C. Cook, C. Amaral, A. DeCastro, S. Leyk, T.L. McIntosh, R.C. Nagy, L. St. Denis, T. Tuff, E. Verleye, A.P. Williams, and C.A. Kolden. 2024. The fastest‑growing and most destructive fires in the US (2001 to 2020). Science 386 (6720): 425–431. https:// doi. org/ 10. 1126/ scien ce. adk57 37. Birch, D.S., P. Morgan, C.A. Kolden, A.T. Hudak, and A.M.S. Smith. 2014. Is proportion burned severely related to daily area burned? Environmental Research Letters 9 (6): 064011. https:// doi. org/ 10. 1088/ 1748‑ 9326/9/ 6/ 064011. Bond‑Lamberty, B., S. D. Peckham, S. T. Gower, and B. E. Ewers. 2009. Effects of fire on regional evapotranspiration in the central Canadian boreal for‑ est. Global Change Biology 15 (5): 1242–1254. https:// doi. org/ 10. 1111/j. 1365‑ 2486. 2008. 01776.x. Bowman, D.M.J.S., J.K. Balch, P. Artaxo, W.J. Bond, J.M. Carlson, M.A. Cochrane, C.M. D’Antonio, R.S. DeFries, J.C. Doyle, S.P. Harrison, F.H. Johnston, J.E. Keeley, M.A. Krawchuk, C.A. Kull, J.B. Marston, M.A. Moritz, I.C. Prentice, C.I. Roos, A.C. Scott, Thomas W.. Swetnam, Guido R.. van der Werf, and S.J. Pyne. 2009. Fire in the Earth system. Science 324 (5926): 481–484. https:// doi. org/ 10. 1126/ scien ce. 11638 86. Burger, N., and W.J. Bond. 2015. Flammability traits of Cape shrubland species with different post‑fire recruitment strategies. South African Journal of Botany 101: 40–48. https:// doi. org/ 10. 1016/j. sajb. 2015. 05. 026. Calkin, D., C. O’Connor, M. Thompson, and R. Stratton. 2021. Strategic wildfire response decision support and the risk management assistance pro‑ gram. Forests 12 (10): 1407. https:// doi. org/ 10. 3390/ f1210 1407. Cannon, S.H. 2001. Debris‑flow generation from recently burned watersheds. Environmental and Engineering Geoscience. https:// doi. org/ 10. 2113/ gseeg eosci.7. 4. 321. Cannon, S. H., and J. DeGraff. 2009. The increasing wildfire and post‑fire debris‑ flow threat in western USA, and implications for consequences of climate change. Landslides - Disaster Risk Reduction. https:// doi. org/ 10. 1007/ 978‑3‑ 540‑ 69970‑5_9. Chatzopoulos‑Vouzoglanis, K., K.J. Reinke, M. Soto‑Berelov, and S.D. Jones. 2024. Are fire intensity and burn severity associated? Advancing our understanding of FRP and NBR metrics from Himawari‑8/9 and Senti‑ nel‑2. International Journal of Applied Earth Observation and Geoinformation 127: 103673. https:// doi. org/ 10. 1016/j. jag. 2024. 103673. Chen, T., and C. Guestrin. 2016. XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. https:// doi. org/ 10. 1145/ 29396 72. 29397 85. Chen, Y., J. T. Randerson, G. R. Van Der Werf, D. C. Morton, M. Mu, and P. S. Kasib‑ hatla. 2010. Nitrogen deposition in tropical forests from savanna and deforestation fires. Global Change Biology 16 (7): 2024–2038. https:// doi. org/ 10. 1111/j. 1365‑ 2486. 2009. 02156.x. Chen, Y., S. Hantson, N. Andela, S.R. Coffield, C.A. Graff, D.C. Morton, L.E. Ott, E. Foufoula‑Georgiou, P. Smyth, M.L. Goulden, and J.T. Randerson. 2022. California wildfire spread derived using VIIRS satellite observations and an object‑based tracking system. Scientific Data 9 (1): 249. https:// doi. org/ 10. 1038/ s41597‑ 022‑ 01343‑0. Coop, J.D., T.J. DeLory, W.M. Downing, S.L. Haire, M.A. Krawchuk, C. Miller, M.‑A. Parisien, and R.B. Walker. 2019. Contributions of fire refugia to resilient ponderosa pine and dry mixed‑conifer forest landscapes. Ecosphere 10 (7): e02809. https:// doi. org/ 10. 1002/ ecs2. 2809. Crutzen, P.J., and M.O. Andreae. 1990. Biomass burning in the tropics: Impact on atmospheric chemistry and biogeochemical cycles. Science 250 (4988): 1669–1678. https:// doi. org/ 10. 1126/ scien ce. 250. 4988. 1669. Cunningham, C.X., G.J. Williamson, and D.M.J.S. Bowman. 2024. Increas‑ ing frequency and intensity of the most extreme wildfires on earth. Nature Ecology & Evolution 8 (8): 1420–1425. https:// doi. org/ 10. 1038/ s41559‑ 024‑ 02452‑2. De Luis, M., M.J. Baeza, J. Raventós, and J.C. González‑Hidalgo. 2004. Fuel characteristics and fire behaviour in mature Mediterranean gorse shrublands. International Journal of Wildland Fire 13 (1): 79. https:// doi. org/ 10. 1071/ WF030 05. Debano, L.F. 2000. The role of fire and soil heating on water repellency in wildland environments: A review. Journal of Hydrology. https:// doi. org/ 10. 1016/ S0022‑ 1694(00) 00194‑3. Dillon, G. K., Z. A. Holden, P. Morgan, M. A. Crimmins, E. K. Heyerdahl, and C. H. Luce. 2011. Both topography and climate affected forest and woodland burn severity in two regions of the western US, 1984 to 2006. Ecosphere 2 (12) : 130. https:// doi. org/ 10. 1890/ ES11‑ 00271.1. Doerr, S. H., W. H. Blake, R. A. Shakesby, F. Stagnitti, S. H. Vuurens, G. S. Hum‑ phreys, P. Wallbrink, S. H. Doerr, W. H. Blake, R. A. Shakesby, F. Stagnitti, S. H. Vuurens, G. S. Humphreys, and P. Wallbrink. 2004. Heating effects on water repellency in Australian eucalypt forest soils and their value in estimating wildfire soil temperatures. International Journal of Wildland Fire 13 (2): 157–163. https:// doi. org/ 10. 1071/ WF030 51. Eidenshink, J., B. Schwind, K. Brewer, Z.‑L. Zhu, B. Quayle, and S. Howard. 2007. A project for monitoring trends in burn severity. Fire Ecology 3 (1): 3–21. https:// doi. org/ 10. 4996/ firee cology. 03010 03. Epting, J., D. Verbyla, and B. Sorbel. 2005. Evaluation of remotely sensed indices for assessing burn severity in interior Alaska using Landsat TM and ETM+. Remote Sensing of Environment 96 (3–4): 328–339. https:// doi. org/ 10. 1016/j. rse. 2005. 03. 002. Estes, B.L., E.E. Knapp, C.N. Skinner, J.D. Miller, and H.K. Preisler. 2017. Factors influencing fire severity under moderate burning conditions in the Klamath Mountains, northern California, USA. Ecosphere 8 (5): e01794. https:// doi. org/ 10. 1002/ ecs2. 1794. Fernández‑Guisuraga, J.M., S. Suárez‑Seoane, P. García‑Llamas, and L. Calvo. 2021. Vegetation structure parameters determine high burn severity likelihood in different ecosystem types: A case study in a burned Medi‑ terranean landscape. Journal of Environmental Management. https:// doi. org/ 10. 1016/j. jenvm an. 2021. 112462. Finney, M. 2006. An Overview of FlamMap Fire Modeling Capabilities (pp. 213– 220). U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. https:// www. fs. usda. gov/ resea rch/ trees earch/ 25948 Finney, M. A., C. W. McHugh, I. C. Grenfell, K. L. Riley, and K. C. Short. 2011. A simulation of probabilistic wildfire risk components for the continental United States. Stochastic Environmental Research and Risk Assessment 25 (7): 973–1000. https:// doi. org/ 10. 1007/ s00477‑ 011‑ 0462‑z. Page 17 of 18 Orlandetal. Fire Ecology (2025) 21:55 Florsheim, J. L., E. A. Keller, and D. W. Best. 1991. Fluvial sediment transport in response to moderate storm flows following chaparral wildfire, Ventura County, southern California. Geological Society of America Bulletin. https:// doi. org/ 10. 1130/ 0016‑ 7606(1991) 103% 3c0504: FSTIRT% 3e2.3. CO;2. Florsheim, J. L., A. Chin, L. S. O’Hirok, and R. Storesund. 2016. Short‑term post‑ wildfire dry‑ravel processes in a chaparral fluvial system. Geomorphology. https:// doi. org/ 10. 1016/j. geomo rph. 2015. 03. 035. The National Strategy: The Final Phase in the Development of the National Cohesive Wildland Fire Management Strategy. 2014. U.S. Forest and Rangelands. Gabet, E.J. 2003. Sediment transport by dry ravel. Journal of Geophysical Research: Solid Earth. https:// doi. org/ 10. 1029/ 2001j b0016 86. Giglio, L. 2007. Characterization of the tropical diurnal fire cycle using VIRS and MODIS observations. Remote Sensing of Environment 108 (4): 407–421. https:// doi. org/ 10. 1016/j. rse. 2006. 11. 018. Giglio, L. 2018. MODIS collection 4 active fire product user’ s guide table of contents. Revisión B. Nasa 1 (June): 64. Giglio, L., W. Schroeder, and C.O. Justice. 2016. The collection 6 MODIS active fire detection algorithm and fire products. Remote Sensing of Environment 178: 31–41. https:// doi. org/ 10. 1016/j. rse. 2016. 02. 054. Hantson, S., S. Pueyo, and E. Chuvieco. 2015. Global fire size distribution is driven by human impact and climate. Global Ecology and Biogeography 24 (1): 77–86. https:// doi. org/ 10. 1111/ geb. 12246. Hantson, S., N. Andela, M. L. Goulden, and J. T. Randerson. 2022. Human‑ ignited fires result in more extreme fire behavior and ecosys‑ tem impacts. Nature Communications. https:// doi. org/ 10. 1038/ s41467‑ 022‑ 30030‑2. Hao, W. M., and M.‑H. Liu. 1994. Spatial and temporal distribution of tropical biomass burning. Global Biogeochemical Cycles 8 (4): 495–503. https:// doi. org/ 10. 1029/ 94GB0 2086. He, K., X. Shen, C. Merow, E. Nikolopoulos, R. V. Gallagher, F. Yang, and E. N. Anagnostou. 2024. Improving fire severity prediction in south‑eastern Australia using vegetation‑specific information. Natural Hazards and Earth System Sciences 24 (10): 3337–3355. https:// doi. org/ 10. 5194/ nhess‑ 24‑ 3337‑ 2024. Heward, H., A.M.S. Smith, D.P. Roy, W.T. Tinkham, C.M. Hoffman, P. Morgan, and K.O. Lannom. 2013. Is burn severity related to fire intensity? Observa‑ tions from landscape scale remote sensing. International Journal of Wildland Fire 22 (7): 910–918. https:// doi. org/ 10. 1071/ WF120 87. Holden, Z.A., P. Morgan, and J.S. Evans. 2009. A predictive model of burn severity based on 20‑year satellite‑inferred burn severity data in a large southwestern US wilderness area. Forest Ecology and Management 258 (11): 2399–2406. https:// doi. org/ 10. 1016/j. foreco. 2009. 08. 017. Johnston, J.M., N. Jackson, C. McFayden, L. Ngo Phong, B. Lawrence, D. Davignon, M.J. Wooster, H. Van Mierlo, D.K. Thompson, A.S. Cantin, D. Johnston, L.M. Johnston, M. Sloane, R. Ramos, and T.J. Lynham. 2020. Development of the user requirements for the Canadian WildFireSat satellite mission. Sensors 20 (18): 5081. https:// doi. org/ 10. 3390/ s2018 5081. Kang, S., J.S. Kimball, and S.W. Running. 2006. Simulating effects of fire distur‑ bance and climate change on boreal forest productivity and evapotran‑ spiration. Science of the Total Environment 362 (1–3): 85–102. https:// doi. org/ 10. 1016/j. scito tenv. 2005. 11. 014. Kasischke, E.S., N.L. Christensen Jr., and B.J. Stocks. 1995. Fire, global warming, and the carbon balance of boreal forests. Ecological Applications 5 (2): 437–451. https:// doi. org/ 10. 2307/ 19420 34. Kean, J.W., D.M. Staley, and S.H. Cannon. 2011. In situ measurements of post‑ fire debris flows in southern California: Comparisons of the timing and magnitude of 24 debris‑flow events with rainfall and soil moisture conditions. Journal of Geophysical Research 116 (F4): F04019. https:// doi. org/ 10. 1029/ 2011J F0020 05. Keeley, J.E., C.J. Fotheringham, and M. Morais. 1999. Reexamining fire suppres‑ sion impacts on brushland fire regimes. Science 284 (5421): 1829–1832. https:// doi. org/ 10. 1126/ scien ce. 284. 5421. 1829. Key, C.H., and Benson, N.C. 2006. Landscape Assessment (LA) sampling and analysis methods. USDA Forest Service - General Technical Report RMRS-GTR . Lamb, M. P., J. S. Scheingross, W. H. Amidon, E. Swanson, and A. Limaye. 2011. A model for fire‑induced sediment yield by dry ravel in steep landscapes. Journal of Geophysical Research. https:// doi. org/ 10. 1029/ 2010J F0018 78. Lamb, M. P., M. Levina, R. A. Dibiase, and B. M. Fuller. 2013. Sediment storage by vegetation in steep bedrock landscapes: Theory, experiments, and implications for postfire sediment yield. Journal of Geophysical ResearchEarth Surface. https:// doi. org/ 10. 1002/ jgrf. 20058. Lancaster, J.T., B.J. Swanson, S.G. Lukashov, N.S. Oakley, J.B. Lee, E.R. Spangler, J.L. Hernandez, B.P.E. Olson, M.J. Defrisco, D.N. Lindsay, Y.J. Schwartz, S.E. Mccrea, P.D. Roffers, and C.M. Tran. 2021. Observations and analyses of the 9 January 2018 debris‑flow disaster, Santa Barbara County, Califor‑ nia. Environmental and Engineering Geoscience, no. 1. https:// doi. org/ 10. 2113/ EEG‑D‑ 20‑ 00015. Letey, J. 2001. Causes and consequences of fire‑induced soil water repellency. Hydrological Processes. https:// doi. org/ 10. 1002/ hyp. 378. Li, F., X. Zhang, S. Kondragunta, and D. P. Roy. 2018. Investigation of the fire radiative energy biomass combustion coefficient: A comparison of polar and geostationary satellite retrievals over the conterminous United States. Journal of Geophysical Research: Biogeosciences 123 (2): 722–739. https:// doi. org/ 10. 1002/ 2017J G0042 79. Li, F., X. Zhang, S. Kondragunta, X. Lu, I. Csiszar, and C. C. Schmidt. 2022. Hourly biomass burning emissions product from blended geostationary and polar‑orbiting satellites for air quality forecasting applications. Remote Sensing of Environment 281 : 113237. https:// doi. org/ 10. 1016/j. rse. 2022. 113237. Linn, R., J. Reisner, J. J. Colman, and J. Winterkamp. 2002. Studying wildfire behavior using FIRETEC. International Journal of Wildland Fire 11 (4): 233. https:// doi. org/ 10. 1071/ WF020 07. Linn, R. R., S. L. Goodrick, S. Brambilla, M. J. Brown, R. S. Middleton, J. J. O’Brien, and J. K. Hiers. 2020. QUIC‑fire: A fast‑running simulation tool for pre‑ scribed fire planning. Environmental Modelling & Software 125 : 104616. https:// doi. org/ 10. 1016/j. envso ft. 2019. 104616. Liu, T., J.T. Randerson, Y. Chen, D.C. Morton, E.B. Wiggins, P. Smyth, E. Foufoula‑ Georgiou, R. Nadler, and O. Nevo. 2024. Systematically tracking the hourly progression of large wildfires using GOES satellite observations. Earth System Science Data 16 (3): 1395–1424. https:// doi. org/ 10. 5194/ essd‑ 16‑ 1395‑ 2024. Lizundia‑Loiola, J., G. Otón, R. Ramo, and E. Chuvieco. 2020. A spatio‑temporal active‑fire clustering approach for global burned area mapping at 250 m from MODIS data. Remote Sensing of Environment 236 : 111493. https:// doi. org/ 10. 1016/j. rse. 2019. 111493. Loboda, T.V., and I.A. Csiszar. 2007. Reconstruction of fire spread within wild‑ land fire events in Northern Eurasia from the MODIS active fire product. Global and Planetary Change 56 (3–4): 258–273. https:// doi. org/ 10. 1016/j. glopl acha. 2006. 07. 015. Luo, K., X. Wang, M. De Jong, and M. Flannigan. 2024. Drought triggers and sustains overnight fires in North America. Nature 627 (8003): 321–327. https:// doi. org/ 10. 1038/ s41586‑ 024‑ 07028‑5. Mell, W., M. A. Jenkins, J. Gould, and P. Cheney. 2007. A physics‑based approach to modelling grassland fires. International Journal of Wildland Fire 16 (1): 1. https:// doi. org/ 10. 1071/ WF060 02. Miller, J.D., and A.E. Thode. 2007. Quantifying burn severity in a heterogeneous landscape with a relative version of the delta normalized burn ratio (dNBR). Remote Sensing of Environment 109 (1): 66–80. https:// doi. org/ 10. 1016/j. rse. 2006. 12. 006. Morgan, P., R. E. Keane, G. K. Dillon, T. B. Jain, A. T. Hudak, E. C. Karau, P. G. Sik‑ kink, Z. A. Holden, and E. K. Strand. 2014. Challenges of assessing fire and burn severity using field measures, remote sensing and modelling. International Journal of Wildland Fire 23 (8): 1045. https:// doi. org/ 10. 1071/ WF130 58. Mu, M., J. T. Randerson, G. R. Van Der Werf, L. Giglio, P. Kasibhatla, D. Morton, G. J. Collatz, R. S. DeFries, E. J. Hyer, E. M. Prins, D. W. T. Griffith, D. Wunch, G. C. Toon, V. Sherlock, and P. O. Wennberg. 2011. Daily and 3‑hourly variability in global fire emissions and consequences for atmospheric model predictions of carbon monoxide: Daily and hourly global fire emissions. Journal of Geophysical Research: Atmospheres 116 (D24): n/a– n/a. https:// doi. org/ 10. 1029/ 2011J D0162 45. Mueller, S.E., A.E. Thode, E.Q. Margolis, L.L. Yocom, J.D. Young, and J.M. Iniguez. 2020. Climate relationships with increasing wildfire in the southwestern US from 1984 to 2015. Forest Ecology and Management. https:// doi. org/ 10. 1016/j. foreco. 2019. 117861. Neff, J. C., J. W. Harden, and G. Gleixner. 2005. Fire effects on soil organic matter con‑ tent, composition, and nutrients in boreal interior Alaska. Canadian Journal of Forest Research 35 (9): 2178–2187. https:// doi. org/ 10. 1139/ x05‑ 154. O’Connor, C. D., D. E. Calkin, and M. P. Thompson. 2017. An empirical machine learning method for predicting potential fire control locations for pre‑fire planning and operational fire management. International Journal of Wildland Fire 26 (7): 587. https:// doi. org/ 10. 1071/ wf161 35. Page 18 of 18 Orlandetal. Fire Ecology (2025) 21:55 Olson, D. M., E. Dinerstein, E. D. Wikramanayake, N. D. Burgess, G. V. N. Powell, E. C. Underwood, J. A. D’amico, I. Itoua, H. E. Strand, J. C. Morrison, C. J. Loucks, T. F. Allnutt, T. H. Ricketts, Y. Kura, J. F. Lamoreux, W. W. Wettengel, P. Hedao, and K. R. Kassem. 2001. Terrestrial ecoregions of the world: A new map of life on Earth. BioScience 51 (11): 933. https:// doi. org/ 10. 1641/ 0006‑ 3568(2001) 051[0933: TEOTWA] 2.0. CO;2. Parks, S., G. Dillon, and C. Miller. 2014. A new metric for quantifying burn severity: The relativized burn ratio. Remote Sensing 6 (3): 1827–1844. https:// doi. org/ 10. 3390/ rs603 1827. Picotte, J.J., and K.M. Robertson. 2011. Validation of remote sensing of burn sever‑ ity in south‑eastern US ecosystems. International Journal of Wildland Fire 20 (3): 453. https:// doi. org/ 10. 1071/ WF100 13. Ponomarev, E.I., A.N. Zabrodin, E.G. Shvetsov, and T.V. Ponomareva. 2023. Wildfire intensity and fire emissions in Siberia. Fire 6 (7): 246. https:// doi. org/ 10. 3390/ fire6 070246. Randerson, J. T., H. Liu, M. G. Flanner, S. D. Chambers, Y. Jin, P. G. Hess, G. Pfister, M. C. Mack, K. K. Treseder, L. R. Welp, F. S. Chapin, J. W. Harden, M. L. Goulden, E. Lyons, J. C. Neff, E. A. G. Schuur, and C. S. Zender. 2006. The impact of boreal forest fire on climate warming. Science 314 (5802): 1130–1132. https:// doi. org/ 10. 1126/ scien ce. 11320 75. Robichaud, P.R., and R.D. Hungerford. 2000. Water repellency by laboratory burn‑ ing of four northern Rocky Mountain forest soils. Journal of Hydrology 231: 207–219. https:// doi. org/ 10. 1016/ S0022‑ 1694(00) 00195‑5. Roche, J.W., M.L. Goulden, and R.C. Bales. 2018. Estimating evapotranspiration change due to forest treatment and fire at the basin scale in the Sierra Nevada, California. Ecohydrology 11 (7): e1978. https:// doi. org/ 10. 1002/ eco. 1978. Rollins, M.G. 2009. LANDFIRE: A nationally consistent vegetation, wildland fire, and fuel assessment. International Journal of Wildland Fire 18 (3): 235. https:// doi. org/ 10. 1071/ WF080 88. Rothermel, R. 1972. A mathematical model for predicting fire spread in wildland fuels. https:// resea rch. fs. usda. gov/ trees earch/ 32533 Rovira, P., J. Romanyà, and B. Duguy. 2012. Long‑term effects of wildfires on the biochemical quality of soil organic matter: A study on Mediterranean shrublands. Geoderma 179:9–19. https:// doi. org/ 10. 1016/j. geode rma. 2012. 02. 011. Scaduto, E., B. Chen, and Y. Jin. 2020. Satellite‑based fire progression mapping: A comprehensive assessment for large fires in northern California. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 13: 5102–5114. https:// doi. org/ 10. 1109/ JSTARS. 2020. 30192 61. Schroeder, W., and Giglio, L. 2016. Visible Infrared Imaging Radiometer Suite (VIIRS) 375 m Active Fire Detection and Characterization Algorithm Theoretical Basis Document 1.0. Schroeder, W., P. Oliva, L. Giglio, and I. A. Csiszar. 2014. The new VIIRS 375 m active fire detection data product: Algorithm description and initial assessment. Remote Sensing of Environment 143:85–96. https:// doi. org/ 10. 1016/j. rse. 2013. 12. 008. Schwilk, D. W., and D. D. Ackerly. 2001. Flammability and serotiny as strategies: Correlated evolution in pines. Oikos 94 (2): 326–336. https:// doi. org/ 10. 1034/j. 1600‑ 0706. 2001. 940213.x. Scott, J.H., and Burgan, R.E. 2005. Standard fire behavior fuel models: A comprehensive set for use with Rothermel’s surface fire spread model. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. https:// doi. org/ 10. 2737/ rmrs‑ gtr‑ 153 Scott, D.F., D.B. Versfeld, and W. Lesch. 1998. Erosion and sediment yield in relation to afforestation and fire in the mountains of the Western Cape Province, South Africa. South African Geographical Journal 80 (1): 52–59. https:// doi. org/ 10. 1080/ 03736 245. 1998. 97136 44. Seiler, W., and P. J. Crutzen. 1980. Estimates of gross and net fluxes of carbon between the biosphere and the atmosphere from biomass burning. Climatic Change 2 (3): 207–247. https:// doi. org/ 10. 1007/ BF001 37988. Seydi, S.T., J.T. Abatzoglou, A. AghaKouchak, Y. Pourmohamad, A. Mishra, and M. Sadegh. 2024. Predictive understanding of links between vegetation and soil burn severities using physics‑informed machine learning. Earth’s Future 12 (8): e2024EF004873. https:// doi. org/ 10. 1029/ 2024E F0048 73. Staley, D. M., A. C. Tillery, J. W. Kean, L. A. McGuire, H. E. Pauling, F. K. Rengers, and J. B. Smith. 2018. Estimating post‑fire debris‑flow hazards prior to wildfire using a statistical analysis of historical distributions of fire severity from remote sensing data. International Journal of Wildland Fire 27 (9): 595. https:// doi. org/ 10. 1071/ WF171 22. Thompson, M.P., C.D. O’Connor, B.M. Gannon, M.D. Caggiano, C.J. Dunn, C.A. Schultz, D.E. Calkin, B. Pietruszka, S.M. Greiner, R. Stratton, and J.T. Mori‑ sette. 2022. Potential operational delineations: New horizons for proactive, risk‑informed strategic land and fire management. Fire Ecology 18 (1): 17. https:// doi. org/ 10. 1186/ s42408‑ 022‑ 00139‑2. Wildfire Crisis Implementation Plan. 2022. U.S. Forest Service. https:// www. fs. usda. gov/ sites/ defau lt/ files/ Wildfi re‑ Crisis‑ Imple menta tion‑ Plan. pdf Van Der Werf, G.R., J.T. Randerson, G.J. Collatz, and L. Giglio. 2003. Carbon emis‑ sions from fires in tropical and subtropical ecosystems. Global Change Biology 9 (4): 547–562. https:// doi. org/ 10. 1046/j. 1365‑ 2486. 2003. 00604.x. Van Der Werf, G. R., J. T. Randerson, L. Giglio, T. T. Van Leeuwen, Y. Chen, B. M. Rog‑ ers, M. Mu, M. J. E. Van Marle, D. C. Morton, G. J. Collatz, R. J. Yokelson, and P. S. Kasibhatla. 2017. Global fire emissions estimates during 1997–2016. Earth System Science Data 9 (2): 697–720. https:// doi. org/ 10. 5194/ essd‑9‑ 697‑ 2017. Vega, J.A., and F. Díaz‑Fierros Viqueira. 1987. Wild fire effects on soil erosion. Ecologia Mediterranea 13 (4): 119–125. https:// doi. org/ 10. 3406/ ecmed. 1987. 1195. Veraverbeke, S., E.N. Stavros, and S.J. Hook. 2014. Assessing fire severity using imaging spectroscopy data from the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and comparison with multispectral capabilities. Remote Sensing of Environment 154: 153–163. https:// doi. org/ 10. 1016/j. rse. 2014. 08. 019. Wells, W.G. 1981. Some effects of brushfires on erosion processes in coastal southern California. Erosion and Sediment Transport in Pacific Rim Steeplands. Proc. Christchurch Symposium, January 1981. Wells, A. G., T. J. Hawbaker, J. K. Hiers, J. Kean, R. A. Loehman, and P. F. Steblein. 2023. Predicting burn severity for integration with post‑fire debris‑flow hazard assessment: A case study from the Upper Colorado River Basin, USA. International Journal of Wildland Fire 32 (9): 1315–1331. https:// doi. org/ 10. 1071/ wf222 00. Westerling, A.L., H.G. Hidalgo, D.R. Cayan, and T.W. Swetnam. 2006. Warming and earlier spring increase Western U.S. forest wildfire activity. Science. https:// doi. org/ 10. 1126/ scien ce. 11288 34. Whitman, E., M. Parisien, D. K. Thompson, R. J. Hall, R. S. Skakun, and M. D. Flan‑ nigan. 2018. Variability and drivers of burn severity in the northwestern Canadian boreal forest. Ecosphere 9 (2) : e02128. https:// doi. org/ 10. 1002/ ecs2. 2128. Wiggins, E. B., A. J. Soja, E. Gargulinski, H. S. Halliday, R. B. Pierce, C. C. Schmidt, J. B. Nowak, J. P. DiGangi, G. S. Diskin, J. M. Katich, A. E. Perring, J. P. Schwarz, B. E. Anderson, G. Chen, E. C. Crosbie, C. Jordan, C. E. Robinson, K. J. Sanchez, T. J. Shingler, Michael Shook, Kenneth L.. Thornhill, Edward L.. Winstead, Luke D.. Ziemba, and R. H. Moore. 2020. High temporal resolution satellite observations of fire radiative power reveal link between fire behavior and aerosol and gas emissions. Geophysical Research Letters. https:// doi. org/ 10. 1029/ 2020G L0907 07. Wooster, M. J., G. Roberts, G. L. W. Perry, and Y. J. Kaufman. 2005. Retrieval of bio‑ mass combustion rates and totals from fire radiative power observations: FRP derivation and calibration relationships between biomass consump‑ tion and fire radiative energy release. Journal of Geophysical Research 110 (D24): D24311. https:// doi. org/ 10. 1029/ 2005J D0063 18. Wu, Z., H. S. He, Y. Liang, L. Cai, and B. J. Lewis. 2013. Determining relative con‑ tributions of vegetation and topography to burn severity from Landsat imagery. Environmental Management 52 (4): 821–836. https:// doi. org/ 10. 1007/ s00267‑ 013‑ 0128‑3. Young, J. D., E. Belval, B. Gannon, Y. Wei, C. O’Connor, C. Dunn, B. M. Pietruszka, D. Calkin, and M. Thompson. 2024. The cost of operational complexity: A causal assessment of pre‑fire mitigation and wildfire suppression. Forest Policy and Economics 169 : 103351. https:// doi. org/ 10. 1016/j. forpol. 2024. 103351. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations.