The wildfire hazard models are a set of 12 raster geospatial products produced by the USDA Forest Service, Northern Research Station that are intended to be used in analyses of wildfire hazards in the region of New Jersey, Ohio, and Pennsylvania. Each raster represents the monthly hazard of a wildfire occurring within a 30 meter pixel as a probability. A statistical model for each month was parameterized with an integrated moisture index, a cumulative drought severity index for the month during the period 2000 to 2009, percent forest cover, and wildland-urban interface classifications to predict the probability of wildfire occurrence based on reported wildfires. Each model included 10 iterations and the raster products of average, minimum, maximum, median, and standard deviation of the predicted probability of a wildfire occurrence is provided. All raster values were converted to integers by multiplying by 10 to reduce file sizes. Therefore, the range of probabilities is 0 to 1000 for the modeled occurrence of a wildfire.These products are intended to provide managers and planners with information related to the wildfire hazard based on reported incidents from 2000 to 2009. Local and daily weather conditions should be monitored to determine site specific burn susceptibility. Our monthly wildfire hazard data is intended to provide long-term trends of potential environmental conditions that coincided with reported wildfires.

These data were generated to map spatial burn severity and emissions of each historically observed large wildfires (>404 hectares (ha)) that burned during 1984–2020 in the state of California in the US. Event-based assessments were conducted at 30-m resolution for all fires and daily emissions were calculated at 500-m resolution for fires burned since 2002. A total of 1697 wildfires were assessed using the Wildfire Burn Severity and Emissions Inventory(WBSE) framework developed by Xu et al 2022. The comprehensive, long-term event and daily emissions records described here could be used to study health effects of wildfire smoke, either by combining them with transport modeling to model air quality and estimate exposures, or by incorporating them into statistical models predicting health impacts as a direct function of estimated emissions. These data will also facilitate analyses of changing emissions impacts on the carbon cycle over the last three decades. High resolution severity and emissions raster maps are generated for each fire event to support further spatial analysis. While the emissions calculated for California with WBSE are not a substitute for real-time daily emissions estimates, it is designed to extend the estimated emissions record back to 1984 with a finer spatial resolution and provide more up-to-date estimates on emissions factors reflecting information from California's recent extreme fires. Methods This dataset provides estimates of 30 m resolution burn severity, and emissions of CO2, CO, CH4, non-methane organic compounds (NMOC), SO2, NH3, NO, NO2, nitrogen oxides (NOx = NO + NO2), PM2.5, OC, and BC. WBSE was implemented for California large wildfires on a per-fire event scale since 1984 and also a daily scale since 2002. The inventory implementation steps, input datasets, and output data are summarized in figure 1 in Xu et al, 2022. Burn severity calculation Fire records for California from 1984 to 2019 were retrieved from MTBS (https://mtbs.gov/viewer/index.html) via interactive viewer on 8 May 2021, resulting in a dataset with a total of 1623 wildfires. We also acquired fire perimeters for 74 large wildfires in 2020 from CAL FIRE (https://frap.fire.ca.gov/frap-projects/fire-perimeters/) and calculated dNBR for each 2020 fire using the dNBR calculation tool with Google Earth Engine (GEE). This process first selects either initial assessment or extended assessment for each fire. The initial assessment utilizes Landsat images acquired immediately after a fire to capture first-order fire effects. The extended assessment uses images obtained during the growing season following the fire to identify delayed first-order effects and dominant second-order effects (Eidenshink et al 2007). We utilized LANDFIRE Biophysical Settings (BPS) to determine which assessment type to apply for each fire burned in 2020. After Picotte et al (2021), we used extended assessment if the majority of general vegetation groups within the fire perimeter are forests, while initial assessment is used when the majority of general vegetation groups are grassland/shrubland. By contrast, MTBS uses extended assessment for forest and shrubland types. We did not delineate grasslands into burn severity categories. Instead, we classified them as burned ('grass burn') because of difficulties in assessing vegetation change. Post-fire images for extended assessment were selected during the next peak of the green season (June–September) using the mean compositing approach suggested by Parks et al (2018). Composite post-fire images acquired immediately within two months after the fire containment dates were used for the initial assessment. Composite pre-fire images for extended and initial assessments were acquired with the matching periods from the preceding year. The dNBR images were produced by quantifying the spectral difference between composite pre-fire and post-fire Landsat scenes. We calculated the unitless, continuous CBI variable from dNBR/NBR values using the linear and Sigmoid B regression models developed for the CONUS by Picotte et al (2021). CBI values were then classified following thresholds modified based on Crotteau et al(2014) into six severity classes: unburned, low severity, moderate severity, high severity, grass burn, and non-processing area. Emissions calculation Emissions of all species are calculated as a function of area burned, fuel loading, the fraction of vegetation burned based on burn severity, and an emissions factor specific to each vegetation type using the following equation modified from the FINN model (Wiedinmyer et al 2011). Fuel categories were assigned from LANDFIRE EVT products. For emissions calculations, EVT data were then categorized into five general vegetation categories: grass, shrub, forest under 5500 feet (1676 m), forest between 5500–7500 feet (1676–2286 m), and forest above 7500 feet (2286 m), updated for California ecosystems. Fuel consumption was determined following Hurteau et el 2014 assigning fuel loading and consumption values for each severity class for the five general vegetation categories based on the First Order Fire Effects Model v5 (Reinhardt et al 1997). Emission factors for greenhouse gases, particulate matter, and reactive trace gases were updated with recent data for each general vegetation class using results from recent field campaigns and studies specific for California ecosystems and Western U.S. ecosystems. Day of burning and daily emissions To assign the day of burning for individual pixels, NASA fire information for resource management system (FIRMS) active fire products from MODIS (Collection 6) within 750 m of the fire perimeter shapefiles supplied by MTBS or CAL FIRE were selected for interpolation to account for detections that might be outside the boundary due to detection radius. VIIRS 375 m data, when available since 2012, was added to complement MODIS data with improved performance to assign burn dates using the fire progression raster tool (figure 4). We filtered the MODIS/VIIRS detection points to the date range of interest and created a 500 m buffer around each point. Points were then converted to circle polygons to represent each point's detection extent properly. The average date was selected as the proper date in regions of overlapping buffers. We then calculated daily emissions and assigned them to the centroids of the aggregated daily progression polygons.

This layer contains wildfire hazard potential (WHP) data for the conterminous United States aggregated from states to block groups and 50 km hex bins then enriched with demographic data. The data is from the USDA Forest Service Fire Modeling Institute providing an index of WHP at a 270 meter resolution. Wildfire hazard potential provides information on the relative potential for wildfire that would be difficult for fire crews to contain. "Areas with higher wildfire potential values represent fuels with a higher likelihood of experiencing high-intensity fire with torching, crowning, and other forms of extreme fire behavior." - Fire Modeling Institute. A score of 5 is very high risk and a score between 0-1 is likely non-burnable area such as water or asphalt. "On its own, WHP is not an explicit map of wildfire threat or risk, but when paired with spatial data depicting highly valued resources and assets such as communities, structures, or powerlines, it can approximate relative wildfire risk to those resources and assets. WHP is also not a forecast or wildfire outlook for any particular season, as it does not include any information on current or forecasted weather or fuel moisture conditions. It is instead intended for long-term strategic planning and fuels management."Each layer has been enriched with 2020 Esri demographic attributes to better approximate wildfire hazard risk relating to the human population. This layer is available in a ready to use web map. A hosted imagery layer of this data is available in ArcGIS Living Atlas for additional analysis.Data notes:Zonal Statistics as Table were run against a local copy of the WHP data using US standard geographies as the feature zone input for the analysis. Geographies included are: State, County, Congressional District, ZIP Code, Tract, and Block Group. Statistical tables were joined to geographies. To learn more about zonal statistics, view the documentation here. 50 km hex bins were created using Generate Tessellation and then joined to zonal statistics as described above (step 1).Data was enriched with 2020 Esri Demographics. Attributes include population, households & housing units, growth rate, and calculated variables such as population change over time. To create the population-weighted attributes on the state, congressional district, and county layers, the hex value population values were used to create the weighting. Within each hex bin, the total population figure and average WHP were multiplied.The hex bins were converted into centroids and summarized within the state, congressional district, and county boundaries.The summation of these values were then divided by the total population of each respective geography.

Following wildfire, mountainous areas of the western United States are susceptible to enhanced runoff and erosion and an increased vulnerability to debris flow during intense rainfall. Convective storms that can generate debris flows in recently burned areas may occur during or immediately after the wildfire, leaving insufficient time for development and implementation of risk mitigation strategies. We present a method for estimating post-fire debris-flow hazards prior to wildfire using historical data to define the range of potential fire severity for a given location based on the statistical distribution of severity metrics obtained from remote sensing. Estimates of debris-flow likelihood, magnitude and triggering rainfall threshold based upon the statistically simulated fire severity data provide hazard predictions consistent with those calculated from fire severity data collected after wildfire. Simulated fire severity data also produce hazard estimates that replicate observed debris-flow occurrence, rainfall conditions, and magnitude at a monitored site in the San Gabriel Mountains of southern California. Future applications of this method should rely upon a range of potential fire severity scenarios for improved pre-fire estimates of debris-flow hazard. The method presented here is also applicable to modeling other post-fire hazards, such as flooding and erosion risk, and for quantifying historic trends in fire severity in a changing climate. This release contains the data used to derive the historical distributions of fire severity, including a) the data used to derive a Weibull cumulative distribution function to historical measures of the differenced normalized burn ratio for fires >= 4 square kilometers (1000 acres) that burned between 2001 and 2014 in the western United States, b) the shape and scale parameters for the Weibull cumulative distribution function for every class of existing vegetation type, and the statistics describing goodness-of-fit of the Weibull distribution to these data, and c) the data used to determine the BARC4 threshold defining the break between pixels burned at low and moderate or high severity.

These data were compiled so that annual wildfire could be modelled across the sagebrush region in the western United States. Our goal was to understand how wildfire probability relates to climate and fuel conditions across the entire sagebrush region. To do this we developed a statistical model that represents the relationship between annual wildfire probability and a small number of climate and fuel variables. Specifically, created predictions of wildfire probability using a biologically plausible logistic regression model that related wildfire probability to mean temperature, annual precipitation, the proportion summer precipitation (PSP), and aboveground biomass of annual herbaceous plants and perennial herbaceous plants. The biomass variables were used as proxies for fine fuel availability. These data represent annual fire occurrence in 1 km pixels (i.e. did a given pixel burn that year), predicted wildfire probability, as well as the three year running average (i.e. average a ...

The geospatial products described and distributed here depict the probability of high-severity fire, if a fire were to occur, for several ecoregions in the contiguous western US. The ecological effects of wildland fire � also termed the fire severity � are often highly heterogeneous in space and time. This heterogeneity is a result of spatial variability in factors such as fuel, topography, and climate (e.g. mean annual temperature). However, temporally variable factors such as daily weather and climatic extremes (e.g. an unusually warm year) also may play a key role. Scientists from the US Forest Service Rocky Mountain Research Station and the University of Montana conducted a study in which observed data were used to produce statistical models describing the probability of high severity fire as a function of fuel, topography, climate, and fire weather. Observed data from over 2000 fires (from 2002-2015) were used to build individual models for each of 19 ecoregions in the contiguous US (see Parks et al. 2018, Figure 1). High severity fire was measured using a fire severity metric termed the relativized burn ratio, which uses pre- and post-fire Landsat imagery to measure fire-induced ecological change. Fuel included pre-fire metrics of live fuel amount such as NDVI. Topography included factors such as slope and potential solar radiation. Climate summarized 30-year averages of factors such as mean summer temperature that spatially vary across the study area. Lastly, fire weather incorporated temporally variable factors such as daily and annual temperature. In turn, these statistical models were used to generate "wall-to-wall" maps depicting the probability of high severity fire, if a fire were to occur, for 13 of the 19 ecoregions. Maps were not produced for ecoregions in which model quality was deemed inadequate. All maps use fuel data representing the year 2016 and therefore provide a fairly up-to-date assessment of the potential for high severity fire. For those ecoregions in which the relative influence of fire weather was fairly strong (n=6), two additional maps were produced, one depicting the probability of high severity fire under moderate weather and the other under extreme weather. An important consideration is that only pixels defined as forest were used to build the models; consequently maps exclude pixels considered non-forest.This record was taken from the USDA Enterprise Data Inventory that feeds into the https://data.gov catalog. Data for this record includes the following resources: ISO-19139 metadata ArcGIS Hub Dataset ArcGIS GeoService For complete information, please visit https://data.gov.

This web map shows the wildfire hazard potential (WHP) for the conterminous United States aggregated from states to block groups and 50 km hex bins. The data is from the USDA Forest Service Fire Modeling Institute providing an index of WHP at a 270 meter resolution. Wildfire hazard potential provides information on the relative potential for wildfire that would be difficult for fire crews to contain. Areas with higher wildfire potential values represent fuels with a higher likelihood of experiencing high-intensity fire with torching, crowning, and other forms of extreme fire behavior. A score of 5 is very high risk and a score between 0-1 is non-burnable area such as water or asphalt. On its own, WHP is not an explicit map of wildfire threat or risk, but when paired with spatial data depicting highly valued resources and assets such as communities, structures, or powerlines, it can approximate relative wildfire risk to those resources and assets. WHP is also not a forecast or wildfire outlook for any particular season, as it does not include any information on current or forecasted weather or fuel moisture conditions. It is instead intended for long-term strategic planning and fuels management.Each layer has been enriched with 2020 Esri demographic attributes to better approximate wildfire hazard risk. A hosted imagery layer of this data is available in ArcGIS Living Atlas for additional analysis.Data notes:Zonal Statistics as Table were run against a local copy of the WHP data using US standard geographies as the feature zone input for the analysis. Geographies included are: State, County, Congressional District, ZIP Code, Tract, and Block Group. Statistical tables were joined to geographies. To learn more about zonal statistics, view the documentation here. 50 km hex bins were created using Generate Tessellation and then joined to zonal statistics as described above (step 1).Data was enriched with 2020 Esri Demographics. Attributes include population, households & housing units, growth rate, and calculated variables such as population change over time. To create the population-weighted attributes on the state, congressional district, and county layers, the hex value population values were used to create the weighting. Within each hex bin, the total population figure and average WHP were multiplied.The hex bins were converted into centroids and summarized within the state, congressional district, and county boundaries.The summation of these values were then divided by the total population of each respective geography.

Aim:

Climate warming is increasing fire activity in many of Earth’s forested ecosystems. Because fire is an important catalyst for change, investigation of post-fire vegetation response is crucial for understanding the potential for future conversions from forest to non-forest vegetation types. To better understand effects of wildfire and climate warming on forest recovery, we assessed the extent to which climate and terrain influence spatiotemporal variation in past and future post-fire tree regeneration.

Location:

Montane forests, Rocky Mountains, USA

Time Period:

1981-2099

Taxa Studied:

Pinus ponderosa; Pseudotsuga menziesii

Methods:

We developed a network of dendrochronological samples (n = 717) and field plots (n = 1301) from post-fire environments spanning a range of topographic and climatic settings. We then used boosted regression trees to predict annual suitability for post-fire seedling establishment and generalized linear mixed models to predict total post-fire seedling abundances, reconstructing recent trends in post-fire recovery and projecting future dynamics using three general circulation models (GCMs) under moderate and extreme emission scenarios.

Results:

Though 1981-2015 declines in growing season (April-September) precipitation were associated with declining suitability for seedling establishment, 2021-2099 trends in precipitation were widely variable among GCMs, leading to mixed projections of future establishment suitability. In contrast, climatic water deficit (CWD), strongly tied to warming temperature and increased evaporative demand, was projected to increase throughout our study area. Our projections strongly suggest that future increases in CWD and an increased frequency of extreme drought will reduce post-fire seedling abundances.

Main Conclusions:

Our findings highlight the key roles of warming and drying in declines in forest resilience to wildfire. The striking differences in projections of post-fire recovery between moderate and extreme emissions scenarios suggest that the most extreme impacts on forest resilience in the latter part of the 21st century may be mitigated with aggressive emissions reductions in the next two decades.

Methods This archive includes field data and various spatial datasets used in Rodman et al. (2020; Global Ecology and Biogeography).

Individual datasets in the Dryad archive include the following:

1) Gridded climate data (annual actual evapotranspiration, annual climatic water deficit, growing season precipitation) for the 1981-2015 period and future climate projections (2021-2099). All climate data were spatially downscaled to c. 250 m.

2) Statistical models and outputs (.rds objects, example spatial models, and summaries of statistical outputs)

3) Terrain variables (60-m resolution) of topographic position index and heat load index. Soil available water capacity at a 4km-resolution.

4) Shapefiles of fire perimeters included in the study and the boundary of the study area (i.e., EPA Level III Ecoregion #21)

5) Field data summarized in this study including 1) 1301 individual field plots characterizing post-fire conifer seedling abundance, forest structure, and ground cover and 2) 717 destructively sampled seedlings dated to establishment/germination year.

These data are a synthesis of five previously published studies that surveyed post-fire seedling abundance and the timing of seedling establishment, as well as previously unpublished data following the study design of Chambers et al. (2016). As methods of collection vary slightly among individual studies, we refer the user to the original published studies (listed below). See "README.txt" for a description of the processing and development of new datasets (i.e., gridded climate data, terrain variables, spatial models, and statistical models).

Chambers, M. E., P. J. Fornwalt, S. L. Malone, and M. A. Battaglia. 2016. Patterns of Conifer Regeneration Following High Severity Wildfire in Ponderosa Pine ñ Dominated Forests of the Colorado Front Range. Forest Ecology and Management 378:57ñ67.

Ouzts, J. R., Kolb, T. E., Huffman, D. W., and A. J. Sánchez Meador. 2015. Post-fire Ponderosa Pine Regeneration With and Without Planting in Arizona and New Mexico. Forest Ecology and Management 354:281ñ290.

Rother, M. T., and T. T. Veblen. 2016. Limited Conifer Regeneration Following Wildfires in Dry Ponderosa Pine Forests of the Colorado Front Range. Ecosphere 7:e01594.

Rother, M. T., and T. T. Veblen. 2017. Climate Drives Episodic Conifer Establishment after Fire in Dry Ponderosa Pine Forests of the Colorado. Forests 8:1-14.

Rodman, K. C., Veblen, T. T., Chapman, T. B., Rother, M. T., Wion, A. P., and M. D. Redmond. 2020b. Limitations to Recovery Following Wildfire in Dry Forests of Southern Colorado and Northern New Mexico, USA. Ecological Applications 30:e02001.

This layer contains wildfire hazard potential (WHP) data for the conterminous United States aggregated from states to block groups and 50 km hex bins then enriched with demographic data. The data is from the USDA Forest Service Fire Modeling Institute providing an index of WHP at a 270 meter resolution. Wildfire hazard potential provides information on the relative potential for wildfire that would be difficult for fire crews to contain. "Areas with higher wildfire potential values represent fuels with a higher likelihood of experiencing high-intensity fire with torching, crowning, and other forms of extreme fire behavior." - Fire Modeling Institute. A score of 5 is very high risk and a score between 0-1 is likely non-burnable area such as water or asphalt. "On its own, WHP is not an explicit map of wildfire threat or risk, but when paired with spatial data depicting highly valued resources and assets such as communities, structures, or powerlines, it can approximate relative wildfire risk to those resources and assets. WHP is also not a forecast or wildfire outlook for any particular season, as it does not include any information on current or forecasted weather or fuel moisture conditions. It is instead intended for long-term strategic planning and fuels management."Each layer has been enriched with 2020 Esri demographic attributes to better approximate wildfire hazard risk relating to the human population. This layer is available in a ready to use web map. A hosted imagery layer of this data is available in ArcGIS Living Atlas for additional analysis.Data notes:Zonal Statistics as Table were run against a local copy of the WHP data using US standard geographies as the feature zone input for the analysis. Geographies included are: State, County, Congressional District, ZIP Code, Tract, and Block Group. Statistical tables were joined to geographies. To learn more about zonal statistics, view the documentation here. 50 km hex bins were created using Generate Tessellation and then joined to zonal statistics as described above (step 1).Data was enriched with 2020 Esri Demographics. Attributes include population, households & housing units, growth rate, and calculated variables such as population change over time. To create the population-weighted attributes on the state, congressional district, and county layers, the hex value population values were used to create the weighting. Within each hex bin, the total population figure and average WHP were multiplied.The hex bins were converted into centroids and summarized within the state, congressional district, and county boundaries.The summation of these values were then divided by the total population of each respective geography.

Wildfires have shown increasing trends in both frequency and severity across the Contiguous United States (CONUS). However, process-based fire models have difficulties in accurately simulating the burned area over the CONUS due to a simplification of the physical process and cannot capture the interplay among fire, ignition, climate, and human activities. The deficiency of burned area simulation deteriorates the description of fire impact on energy balance, water budget, and carbon fluxes in the Earth System Models (ESMs). Alternatively, machine learning (ML) based fire models, which capture statistical relationships between the burned area and environmental factors, have shown promising burned area predictions and corresponding fire impact simulation. We develop a hybrid framework (ML4Fire-XGB) that integrates a pretrained eXtreme Gradient Boosting (XGBoost) wildfire model with the Energy Exascale Earth System Model (E3SM) land model (ELM). A Fortran-C-Python deep learning bridge is adapted to support online communication between ELM and the ML fire model. Specifically, the burned area predicted by the ML-based wildfire model is directly passed to ELM to adjust the carbon pool and vegetation dynamics after disturbance, which are then used as predictors in the ML-based fire model in the next time step. Evaluated against the historical burned area from Globalmore » Fire Emissions Database 5 from 2001-2020, the ML4Fire-XGB model outperforms process-based fire models in terms of spatial distribution and seasonal variations. Sensitivity analysis confirms that the ML4Fire-XGB well captures the responses of the burned area to rising temperatures. The ML4Fire-XGB model has proved to be a new tool for studying vegetation-fire interactions, and more importantly, enables seamless exploration of climate-fire feedback, working as an active component in E3SM.« less

Aim: Ecological properties governed by threshold relationships can exhibit heightened sensitivity to climate, creating an inherent source of uncertainty when anticipating future change. We investigated the impact of threshold relationships on our ability to project ecological change outside the observational record (e.g., the 21st century), using the challenge of predicting late‐Holocene fire regimes in boreal forest and tundra ecosystems.

Location: Boreal forest and tundra ecosystems of Alaska.

Time period: 850–2100 CE.

Major taxa studied: Not applicable.

Methods: We informed a set of published statistical models, designed to predict the 30‐year probability of fire occurrence based on climatological normals, with downscaled global climate model data for 850–1850 CE. To evaluate model performance outside the observational record and the implications of threshold relationships, we compared modelled estimates with mean fire return intervals estimated from 29 published lake‐sediment palaeofire reconstructions. To place our results in the context of future change, we evaluate changes in the location of threshold to burning under 21st‐century climate projections.

Results: Model–palaeodata comparisons highlight spatially varying accuracy across boreal forest and tundra regions, with variability strongly related to the summer temperature threshold to burning: sites closer to this threshold exhibited larger prediction errors than sites further away from this threshold. Modifying the modern (i.e., 1950–2009) fire–climate relationship also resulted in significant changes in modelled estimates. Under 21st‐century climate projections, increasing proportions of Alaskan tundra and boreal forest will approach and surpass the temperature threshold to burning, with > 50% exceeding this threshold by > 2 °C by 2070–2099.

Main conclusions: Our results highlight a high sensitivity of statistical projections to changing threshold relationships and data uncertainty, implying that projections of future ecosystem change in threshold‐governed ecosystems will be accompanied by notable uncertainty. This work also suggests that ecological responses to climate change will exhibit high spatio‐temporal variability as different regions approach and surpass climatic thresholds over the 21st century.

Information on the 245 burned basins, 293 unburned basins, and 356 associated fires from across the U.S. West which were used in statistical analyses of post-wildfire water quality response. Included are physiographic characteristics, as well as ESRI Shapefile polygons representing delineations for each basin and fire. Additionally, daily carbon, nitrogen, phosphorus, sediment, and turbidity data sampled from the basins' outlets are provided from 1974-2022. R programming scripts used in data processing and modeling also included.

Water quality data used to create this dataset are from the Water Quality Portal.

Climate change is projected to dramatically increase boreal wildfire activity, with broad ecological and socio-economic consequences. As global temperatures rise, periods with elevated fire weather are expected to increase in frequency and duration, which would be expected to increase the number and size of fires. Statistical forecasts or simulations of future fire activity often account for direct climatic effects only, neglecting other controls of importance, such as biotic feedbacks. This could result in overestimating the effects of climate change on fire activity, if the future distribution of vegetation or fuels were to change. We incorporated sensitivity to climate or fire weather and vegetation in a fire simulation model, and represented explicitly two key biotic feedbacks linked to succession and regeneration processes. We used this model to forecast annual fire activity from 2011 to 2099 over a large region of boreal forest in Québec, Canada, dominated by balsam fir (Abies balsamea (L.) Mill) and yellow birch (Betula alleghaniensis Britt.) or paper birch (Betula papyrifera Marsh.), with and without the biotic feedbacks. Our simulations show that vegetation changes triggered by fire disturbance altered future fire activity, and may even be as important a driver as climate change itself. Indeed, over the course of the century, vegetation changes were projected to offset much of the increase in fire activity that would be expected due to global warming as such. It follows that if biotic feedbacks are not included in statistical or simulation-based forecasts, the resultant projections of future fire activity could be biased upwards to a very considerable degree. For the case of end-of-century mean annual burn rate, we estimated this positive bias to be as high as 400%. Accounting for biotic feedbacks in simulation models is therefore necessary for accurate projection of future wildfire activity and associated vegetation changes. Purely statistical forecasts based on current vegetation cannot be relied upon, in the presence of biotic feedbacks. Our results further suggest that vegetation management could reduce fire risk in some systems by altering the abundance and distribution of the most highly flammable fuels, and thus mitigate the impact of climate change on fire activity.

Time-varying fire-climate relationships may represent an important component of fire-regime variability, relevant for understanding the controls of fire and projecting fire activity under global-change scenarios. We used time-varying statistical models to evaluate if and how fire-climate relationships varied from 1902-2008, in one of the most flammable forested regions of the western U.S.A. Fire-danger and water-balance metrics yielded the best combination of calibration accuracy and predictive skill in modeling annual area burned. The strength of fire-climate relationships varied markedly at multi-decadal scales, with models explaining < 40% to 88% of the variation in annual area burned. The early 20th century (1902-1942) and the most recent two decades (1985-2008) exhibited strong fire-climate relationships, with weaker relationships for much of the mid 20th century (1943-1984), coincident with diminished burning, less fire-conducive climate, and the initiation of modern fire fighting. Area burned and the strength of fire-climate relationships increased sharply in the mid 1980s, associated with increased temperatures and longer potential fire seasons. Unlike decades with high burning in the early 20th century, models developed using fire-climate relationships from recent decades overpredicted area burned when applied to earlier periods. This amplified response of fire to climate is a signature of altered fire-climate-relationships, and it implicates non-climatic factors in this recent shift. Changes in fuel structure and availability following 40+ yr of unusually low fire activity, and possibly land use, may have resulted in increased fire vulnerability beyond expectations from climatic factors alone. Our results highlight the potential for non-climatic factors to alter fire-climate relationships, and the need to account for such dynamics, through adaptable statistical or processes-based models, for accurately predicting future fire activity.

The LANDFIRE fuel data describe the composition and characteristics of both surface fuel and canopy fuel. Specific products include fire behavior fuel models, canopy bulk density (CBD), canopy base height (CBH), canopy cover (CC), canopy height (CH), and fuel loading models (FLMs). These data may be implemented within models to predict the behavior and effects of wildland fire. These data are useful for strategic fuel treatment prioritization and tactical assessment of fire behavior and effects. DATA SUMMARY: Canopy base height (CBH) describes the lowest point in a stand where there is sufficient available fuel (=> .25 in dia.) to propagate fire vertically through the canopy. Specifically, CBH is defined as the lowest point at which the canopy bulk density is >= 0.012 kg m-3. A spatially explicit map of canopy base height supplies information used in fire behavior models such as FARSITE (Finney 1998) to determine the point at which a surface fire will transition to a crown fire. It should be noted that LANDFIRE layers will not include canopy characteristics in fuel types where the tree canopy is considered a part of the surface fuel and the surface fire behavior fuel model is chosen to reflect these conditions. This is because LANDFIRE assumes that the potential burnable biomass in the shorter tree canopies has been accounted for in the surface fuel model parameters. For example, maps of areas dominated by young or short conifer stands where the trees are represented by a shrub type fuel model will not include canopy characteristics. The CBH mapping process began by deriving field referenced estimates of canopy characteristics through LFRDB plot analysis. Approximately 50,000 plots were acquired throughout the U.S. for estimating CBH. Utilizing these plots, field referenced CBH values were calculated for each plot using the canopy fuel estimation software FuelCalc (Reinhardt et al. 2006b). Go to http://www.landfire.gov/participate_acknowledgements.php for more information regarding contributors of field plot data. This process of deriving field referenced estimates for CBH was employed to create a training data set with the aim of modeling CBH values. Statistical analysis of plot variables indicated that Existing Vegetation Type (EVT) and Existing Vegetation Height (EVH) demonstrated some influence on CBH, with Existing Vegetation Cover (EVC) affecting CBH values within certain EVTs. Unfortunately, these relationships were not statistically strong enough to model CBH and alternate approaches were explored. The only relationship that was statistically significant was that related to juniper EVTs. From this analysis, juniper EVTs consistently showed similar CBH outputs and were hardwired to a value of 4.0 mx10. Using the information gleaned from the statistical analysis it was decided to map CBH values using an average look up table (LUT) approach based on plot level combinations of EVT, EVC, and EVH. To assign averages using these variables, various grouping combinations of EVT, EVC, and EVH were tested to determine which would map CBH values most logically. For each grouping, a set of look-up tables was calculated enabling CBH to be mapped with the Fuels Change Mapping Tool, or ToFu Delta. These maps were analyzed, peer reviewed and tested to determine which performed best. In the process of developing and testing these grouping strategies it was realized that not enough plot data was available to account for all EVT, EVH and EVC combinations. To account for these missing values and fill in data gaps, a 'pyramid approach' was adopted for mapping CBH that would allow for a value to be assigned at some level. The basic premise of this approach was to map assignments with the most detailed data available and fill in behind it with coarser level aggregate values to account for all combinations. To accomplish these aggregate assignments, aggregate values for EVTs were derived at two coarser levels, existing vegetation groups (EVG) and existing vegetation systems (EVS). Each aggregate was more comprehensive than the previous and comprised of subgroups of cover and height combinations (ECHGs). Existing vegetation cover height groups (ECHGs) were derived by grouping CBH values into aggregate groups of EVT, EVG, and EVS. Each aggregate group was split into subgroups of the Existing Vegetation Height (EVH) classes. Lastly, the subgroups were split into two more groups where the greatest average difference occurred between an upper and lower range of forested canopy cover and the difference was greater than or equal to two feet CBH. Each ECHG was assigned an average CBH from the plots - the standard deviation of the plots that meet the requirements of EVT, EVG, or EVS and EVH and EVC. Prior to implementing this aggregate grouping strategy certain data thresholds had to be met in an attempt to ensure that a representative data set was being utilized. For each group (or subset) a data threshold greater than or equal to five plots per ECHG had to be reached before it could be implemented. Subsequently all outliers greater than or equal to three standard deviations from the mean were removed prior to computing a CBH value. The CBH data represented in the resultant layer are continuous from 0 to 9.9 meters (to the nearest 0.1 meter). Some stands dominated by broadleaf species which typically do not permit initiation of crown fire (e.g. Populus spp.) are coded with a CBH of 10 meters. Since crown fire is rarely observed in most hardwood stands, the highest CBH value possible was used to prevent false simulation of crown fire in these areas. Similarly, all non-forest values, including herbaceous, and shrub systems and non-burnable types such as urban, barren, snow and ice and agriculture, were coded as 0. Finally, certain types of agriculture that are deemed burnable were assigned a value by ToFuDelta based on region and vegetation type. REFRESH 2008 (lf_1.1.0): Refresh 2008 (lf_1.1.0) used 2001 data as a launching point to incorporate disturbance and its severity, both managed and natural, which occurred on the landscape after 2001. Specific examples of disturbance are: fire, vegetation management, weather, and insect and disease. The final disturbance data used in Refresh 2008 (lf_1.1.0) is the result of several efforts that include data derived in part from remotely sensed land change methods, Monitoring Trends in Burn Severity (MTBS), and the LANDFIRE Refresh events data call. Vegetation growth was modeled where both disturbance and non-disturbance occurs. For details on methods, see Process Description for LANDFIRE Refresh 2008 (lf_1.1.0).

Maps of the number, size, and species of trees in forests across the western United States are desirable for many applications such as estimating terrestrial carbon resources, predicting tree mortality following wildfires, and for forest inventory. However, detailed mapping of trees for large areas is not feasible with current technologies, but statistical methods for matching the forest plot data with biophysical characteristics of the landscape offer a practical means to populate landscapes with a limited set of forest plot inventory data. We used a modified random forests approach with Landscape Fire and Resource Management Planning Tools (LANDFIRE) vegetation and biophysical predictors as the target data, to which we imputed plot data collected by the USDA Forest Service’s Forest Inventory Analysis (FIA) to the landscape at 30-meter (m) grid resolution (Riley et al. 2016). This method imputes the plot with the best statistical match, according to a “forest” of decision trees, to each pixel of gridded landscape data. In this work, we used the LANDFIRE data set as the gridded target data because it is publicly available, offers seamless coverage of variables needed for fire models, and is consistent with other data sets, including burn probabilities and flame length probabilities generated for the continental United States. The main output of this project (the GeoTIFF available in this data publication) is a map of imputed plot identifiers at 30×30 m spatial resolution for the western United States for landscape conditions circa 2009. The map of plot identifiers can be linked to the FIA databases available through the FIA DataMart or to the ACCDB/CSV files included in this data publication to produce tree-level maps or to map other plot attributes. These ACCDB/CSV files also contain attributes regarding the FIA PLOT CN (a unique identifier for each time a plot is measured), the inventory year, the state code and abbreviation, the unit code, the county code, the plot number, the subplot number, the tree record number, and for each tree: the status (live or dead), species, diameter, height, actual height (where broken), crown ratio, number of trees per acre, and a unique identifier for each tree and tree visit. Application of the dataset to research questions other than those related to aboveground biomass and carbon should be investigated by the researcher before proceeding. The dataset may be suitable for other applications and for use across various scales (stand, landscape, and region), however, the researcher should test the dataset's applicability to a particular research question before proceeding.Geospatial data describing tree species or forest structure are required for many analyses and models of forest landscape dynamics. Forest data must have resolution and continuity sufficient to reflect site gradients in mountainous terrain and stand boundaries imposed by historical events, such as wildland fire and timber harvest. Such detailed forest structure data are not available for large areas of public and private lands in the United States, which rely on forest inventory at fixed plot locations at sparse densities. While direct sampling technologies such as light detection and ranging (LiDAR) may eventually make broad coverage of detailed forest inventory feasible, no such data sets at the scale of the western United States are currently available.When linking the tree list raster (“CN_text” field) to the FIA data via the plot CN field (“CN” in the “PLOT” table and “PLT_CN” in other tables), note that this field is unique to a single visit to a plot. The raster contains a “Value” field, which also appears in the ACCDB/CSV files in the “tl_id” field in order to facilitate this linkage. All plot CNs utilized in this analysis were single condition, 100% forested, physically located in the Rocky Mountain Research Station (RMRS) and Pacific Northwest Research Station (PNW) obtained from FIA in December of 2012.

Original metadata date was 01/03/2018. Minor metadata updates made on 04/30/2019.

This dataset is a polygon coverage of counties limited to the extent of the Fire Clay coal zone resource areas and attributed with statistics on the thickness of the Fire Clay coal bed, its elevation, and overburden thickness, in feet. The file has been generalized from detailed geologic coverages found elsewhere in Professional Paper 1625-C. This resource model for the Fire Clay coal zone must be considered provisional, because the correlation of the zone continues to be evaluated in West Virginia.

A digital record of all Tesla fires - including cars and other products, e.g. Tesla MegaPacks - that are corroborated by news articles or confirmed primary sources. Latest version hosted at https://www.tesla-fire.com.

These data provide projected 30-yr normals (i.e., averages) for two climate variables, three different 21st-century time periods, and five global climate models (GCMs): Climate variables: Mean temperature of the warmest month (TWARM, °C) Total annual moisture availability (P-PETANN, mm). P = precipitation, PET = potential-evapotranspiration Time periods: 2010-2039 2040-2069 2070-2099 GCMs: CCSM4 GFDL-CM3 GISS-E2-R IPSL-CM5A-LR MRI-CGCM3 The purpose of these climatological maps is to inform and drive the statistical models from Young et al. (2017), providing spatially explicit 21st-century projections of the 30-yr probability of fire occurrence across Alaskan boreal forest and tundra ecosystems. The statistical models and the future fire-probability projections are provided in the “Generalized Boosted Models and Analysis Scripts” and “Future fire projections for 2010-2099, Alaska” nested datasets within this archive, respectively. All GCM projections are under the Representative Concentration Pathway 6.0. All climate data was obtained from Scenarios Network for Alaska and Arctic Planning (2015). Full details are available in Young et al. (2017). File naming convention: Example: "GFDL-CM3_rcp60_AnnDEF_2040_2069.tif" "GFDL-CM3" = GCM Name "rcp60" = Representative Concentration Pathway (RCP) "AnnDEF" = Climate variable. AnnDEF = Mean total P-PET (mm); TempWarm = Mean temperature of the warmest month (°C) “2040_2069” = Beginning (e.g., 2040) and ending (e.g., 2069) years for 30-yr climatology Citations: Scenarios Network for Alaska and Arctic Planning, University of Alaska. 2015. Projected Monthly Temperature and Precipitation - 2 km CMIP5/AR5. Retrieved January 2015 from https://www.snap.uaf.edu/tools/data-downloads . Young, A.M., P. E. Higuera, P. A. Duffy, and F. S. Hu. 2017. Climatic thresholds shape northern high-latitude fire regimes and imply vulnerability to future climate change. Ecography 40:606-617. doi: 10.1111/ecog.02205.

Climate warming, land use change, and altered fire regimes are driving ecological transformations that can have critical effects on Earth’s biota. Fire refugia – locations that are disturbed less frequently or severely than their surroundings – may act as sites of relative stability during this period of rapid change by being resistant to fire and supporting post-fire recovery in adjacent areas. Because of their potential value to forest ecosystems, there is an urgent need to anticipate (1) where refugia are most likely to be found and (2) where they align with environmental conditions that support post-fire tree recruitment. Using biophysical predictors and patterns of burn severity from 1,180 recent fire events, we mapped the locations of potential fire refugia across upland conifer forests in the southwestern United States (US) (99,428 km2 of forest area), a region that is highly vulnerable to fire-driven transformation. We found that low pre-fire forest cover, flat slopes or topographic concavities, moderate weather conditions, spring-season burning, and areas affected by low- to moderate-severity fire within the previous 15 years were commonly associated with refugia. Based on current (i.e., 2021) conditions, we predicted that 67.6% and 18.1% of conifer forests in our study area would contain refugia under moderate and extreme fire weather, respectively. However, refugia were 36.4% (moderate weather) and 31.2% (extreme weather) more common across forests that experienced recent fires, as compared to recently unburned areas, supporting increased fire use during moderate weather and shoulder seasons to promote fire-resistant landscapes. When overlaid with models of tree recruitment, 23.2% (moderate weather) and 6.4% (extreme weather) of forests were classified as refugia with a high potential to support post-fire recruitment in the surrounding landscape. These locations may be disproportionately valuable for ecosystem sustainability, providing habitat for fire-sensitive species and maintaining forest persistence in an increasingly fire-prone world. Methods This archive includes statistical models and spatial data used to predict the locations of forested fire refugia and post-fire tree recruitment potential throughout four ecoregions (EPA Level III Ecoregions #19, 21, 23, and 79) in Arizona, Colorado, Idaho, New Mexico, Utah, and Wyoming, USA. See "README.md" for information about datasets and their collection.