State of Wildfires 2025-26: Midwestern Canadian Shield Forests Burned Area Attribution and Projections

This repository contains the input and ouput data for Burned Area attribution and projections from ConFLAME ussd inthe State of Wildfire 2025/26 report. We have three directories: driving_data: contains input data for attribution (nrt), historical training data to optimise the model for projections (isimip3a) and future projections input data (isimip3b) driving_data All NetCDF files are on regular 0.5-degree grids, on a monthly timestep, across the three focal regions used in the report. NRT (Near Real-Time) Data Contained in the nrt directory. This data was used for near real-time attribution. Factual in dir factual/ This is the actual ConFLAME driving data, including the attribution "factual" data. Files include: Climate Variables: pr.nc, tas_max.nc, tas_mean.nc, no_of_dry_days.nc, hursmin_min.nc, wind_max.nc, cumm_dry_days_max.nc, cumm_dry_days_max.nc, cumm_dry_days_mean.nc, LI.nc, cg_strokes.nc Fuel and Vegetation: wood_HADGEM.nc, tree_HADGEM.nc, shrub_HADGEM.nc, veg_HADGEM.nc, veg_log_HADGEM.nc, shrub_HADGEM.nc Land Use and Socioeconomic Drivers: cropland_regridded_to_era5.nc, pasture_regridded_to_era5.nc, rangeland_regridded_to_era5.nc, grazing_land_regridded_to_era5.nc,urban_area_regridded_to_era5.nc, total_irrigated_regridded_to_era5.nc,population_density_regridded_to_era5.nc, urban_population_regridded_to_era5.nc, rural_population_regridded_to_era5.nc Burned Area Target burned_area.nc: Regridded 0.5° burned area data, used as the target for model training and evaluation. Counterfactuals In two dirs: counterfactual_mean/: Contains monthly counterfactuals based on the mean difference between HadGEM3 “ALL” and “NAT” ensemble members. See Barbosa et al. (submitted) for details. counterfactual/: Contains ensemble-based counterfactuals used for attribution. Climate, fuel and vegetation variables in the factual have subdirectoies: ens-<number>.nc, where <number> is the ensemble member Here's a full summery: Category File Name or Directory For factual, contained in firename.nc For Counterfactual conainted in filename/ens_<<no>>.nc where <<no>> is ensmeble number Counterfactual included Description Climate Variables pr Y Monthly total precipitation (mm). tas_max Y Monthly maximum daily temperature (°C). tas_mean Monthly mean daily temperature (°C). hursmin_min Y Monthly minimum of daily minimum relative humidity (%). no_dry_days Y Monthly count of dry days (e.g., days with precipitation < 1 mm). cumm_dry_days_max Y Monthly maximum number of consecutive dry days. cumm_dry_days_mean Y Monthly mean number of consecutive dry days. LI N Monthly mean Lifted Index (°C), a convective instability indicator. cg_strokes N Cloud to ground lightning count Fuel and Vegetation Moisture wood_HADGEM N Fractional woody vegetation (tree and shurb) cover. tree_HADGEM N Fractional tree cover shrub_HADGEM N Fractional shrub cover veg_HADGEM N Fractional vegetation cover veg_log_HADGEM N log transformed fractional vegetation cover Land Use and Socioeconomic cropland_regridded_to_era5.nc N Monthly cropland cover (fraction), regridded to match climate data. pasture_regridded_to_era5.nc N Monthly pasture area (fraction), regridded. rangeland_regridded_to_era5.nc N Monthly rangeland area (fraction), regridded. grazing_land_regridded_to_era5.nc N Monthly total grazing land (pasture + rangeland), regridded. urban_area_regridded_to_era5.nc N Monthly urban area (fraction), regridded. total_irrigated_regridded_to_era5.nc N Monthly total irrigated land (fraction), regridded. population_density_regridded_to_era5.nc N Monthly gridded population density (people/km²), regridded. urban_population_regridded_to_era5.nc N Monthly gridded urban population (people), regridded. rural_population_regridded_to_era5.nc N Monthly gridded rural population (people), regridded. Burned Area Target burned_area.nc Tranining data Monthly burned area at 0.5° resolution, used as the target variable in model training and testing. ISIMIP3a Data (Attribution) Directory structure: <<experiment>>/<<reanalysis_source>>/period_yyyy_yyyy/ experiment: obsclim: Observed climate (factual) counterclim: Detrended (counterfactual) reanalysis_source: Currently only GSWP3-W5E5 Time periods used in the report: Factual: obsclim/GSWP3-W5E5/period_2002_2019/ Counterfactual: counterclim/GSWP3-W5E5/period_2002_2019/ Early Industrial: counterclim/GSWP3-W5E5/period_1901_1920/ ISIMIP3b Data (Future Projections) Directory structure: <<experiment>>/<<GCM>>/period_yyyy_yyyy/ experiment: historical, ssp126, ssp370, ssp585 GCM: Refers to the General Circulation Model used ISIMIP Variable Table NetCDF file Variable Used? Source Notes consec_dry_mean.nc Max. consecutive dry days Yes ISIMIP3a/3b Based on precipitation crop_jules-es.nc Cropland Yes ISIMIP3a/3b Interpolated monthly from annual values filled_debiased_nonetree_cover_jules-es.nc Non-tree vegetated cover Yes JULES-ES Bias-corrected using MODIS VCF filled_debiased_tree_cover_jules-es.nc Tree cover Yes JULES-ES Bias-corrected using MODIS VCF filled_debiased_vegCover_jules-es.nc Total vegetation cover Yes JULES-ES Bias-corrected lightning.nc Lightning climatology Yes ISIMIP3a Climatology only pasture_jules-es.nc Pasture Yes ISIMIP3a/3b Monthly interpolated pr_mean.nc Monthly precipitation Yes ISIMIP3a/3b tas_max.nc Monthly maximum temperature Yes ISIMIP3a/3b Max of daily max temps tas_mean.nc Mean temperature Yes ISIMIP3a/3b Daily mean urban_jules-es.nc Urban area fraction Yes JULES-ES vpd_max.nc Max monthly vapour pressure deficit Yes ISIMIP3a/3b vpd_mean.nc Mean monthly VPD Yes ISIMIP3a/3b ... ... No ... Other vegetation and VCF files not used directly Variables with _change in the filename represent the difference from the previous year to the current year. ISIMIP data is detailed in Frieler et al. (2024) and can be accessed at https://data.ISIMIP.org.VCF data used for bias correction is from MODIS Vegetation Continuous Fields Collection 6.0 (DiMiceli et al., 2015).JULES-ES simulations used ISIMIP climate drivers as described in Mathison et al. (2023). Hadgem3 This contains the files necessary for rerunning the model without redoing the optimisation.While you may not need to view these directly, they are useful for setting up your own ConFLAME model experiments (see the ConFLAME configuration settings). Core files: model_graph.png – Graphical representation of the trained model scalers-*.csv – Scaling parameters used for model input features trace-*.nc – Posterior traces from the Bayesian optimisation none_trace-params-*.txt – Parameter values without the trace variables_info-*.txt – List of variables used in the model and their definitions namelist_*.txt – Model configuration files Core directories: figs – Automatically generated evaluation figures and selected output plots time_series – Burned area (BA) time series for attribution experiments.Structure: <<model_id>>/<<experiment>>/<<range>>/<<members or percentiles>>/<<metric>>/<<variable>>.csv Model ID: currently only _19-frac_points_0.5 Experiments: factual – Simulations using observed climate conditions counterfactual – Simulations with climate change signal removed counterfactual-metmean – Counterfactuals using mean meteorological differences (HadGEM3 ALL–NAT) as in Barbosa et al. (submitted) Range: mean – Regional mean burned area pc-0.95 – Burned area for grid cells in the top 5% of BA values Members or Percentiles: members – Time series for individual ensemble members from the sampled posterior percentiles – Precomputed percentile ranges (e.g., 5th–95th) over all members Metric: absolute – Burned area in km² or equivalent units climatology – Long-term monthly mean burned area anomaly – Deviation from climatology for that month ratio – Burned area divided by climatology Variables: Evaluate.csv – BA simulated with stochasticity (used in the State of Wildfires report) Control.csv – BA simulated without stochasticity samples – Contains full spatial samples from the posterior for each experiment: <<model_id>>/<<experiment>>/<<variable>>/sample-predXXX.nc Variables align with those in the time_series directory. Pairing: Samples are paired across experiments within the same optimisation, meaning sample-predX.nc in the factual experiment matches sample-predX.nc in the counterfactual experiment. isimip Each region directory contains time series outputs from ConFLAME simulations driven by ISIMIP climate data, covering both historical and future projection experiments. The data structure is similar to the attribution outputs under "hadgem" above, but differs in experiment types and the inclusion of multiple climate models. Directory Structure <<region>>/time_series/<<experiment>>/... 1. Attribution-style experiments (ISIMIP3a; 2000–2019) For the following experiments: early_industrial – Simulations under pre-industrial climate counterfactual – Simulations with climate change signal removed factual – Simulations with observed historical climate The structure continues as: <<range>>/<<members or percentiles>>/<<metric>>/<<variable>>.csv Range: mean – Regional mean burned area pc-0.95 – Burned area for the grid cells with the highest 5% BA Members or Percentiles: members – Time series for all ensemble members from the posterior percentiles – Precomputed percentile ranges over all members Metric: absolute – Burned area in km² or equivalent climatology – Long-term monthly mean burned area anomaly – Deviation from climatology for that month ratio – Burned area divided by climatology 2. Future projections (ISIMIP3b; Historical: 2000–2014, SSPs: 2015–2100) For the following experiments: historical – Simulations with observed climate up to 2014 ssp126 – Low-emissions scenario ssp370 – Middle-of-the-road scenario ssp585 – High-emissions scenario The structure continues as: model/<<model>>/<<range>>/<<members or percentiles>>/<<metric>>/<<variable>>.csv Models: GFDL-ESM4 IPSL-CM6A-LR MPI-ESM1-2-HR MRI-ESM2-0 UKESM1-0-LL Important Notes This repo does not include the samples directory (full spatial posterior draws) because these can exceed 1 TB per region. You can easily regenerate samples by installing ConFLAME from:https://github.com/douglask3/Bayesian_fire_modelsand running the provided namelist for the desired experiment.