State of Wildfires 2024/25 – ConFLAME Driver Assessment - Northeastern Amizonia/Pantanal-Chiquitano

This repository contains ConFLAME driver assessment outputs for Northeastern Amizonia and Pantanal-Chiquitano regions, as used in the State of Wildfires 2024/25 report.
It provides the model ouputs required to explore the contribution of individual drivers (fuel, moisture, weather, wind, ignitions, suppression) to burned area (BA) in 2024/25, along with all intermediate data required for reproducing the reported results.

Contents

Core directories:

• figs – Automatically generated evaluation figures and selected plots from the driver assessment runs.

• time_series – Burned area (BA) time series for all driver assessment experiments.
  Structure:

   

  <<model_id>>/<<experiment>>/<<range>>/<<members or percentiles>>/<<metric>>/<<variable>>.csv

  • Model ID: _21-frac_points_0.5

  • Experiment: baseline- – Baseline driver assessment run for this region

  • Range:

    • mean – Regional mean burned area

    • pc-0.95 – Burned area for the top 5% of BA grid cells

  • Members or Percentiles:

    • members – Time series for each ensemble member from the sampled posterior

    • percentiles – Precomputed percentile ranges (e.g., 5th–95th) across members

  • Metric:

    • absolute – Burned area in km² (or equivalent units)

    • climatology – Long-term monthly mean BA

    • anomaly – Deviation from climatology for that month

    • ratio – BA divided by climatology

  • Variables:
    These files are explicitly named to indicate the driver tested and whether the BA is simulated (Evaluate), the standard limitation (Standard_[N]), or the increase in BA due to control (Potential_climatology[N]):

     

    Control.csv Evaluate.csv standard-Fuel.csv standard-Moisture.csv standard-Weather.csv standard-Wind.csv standard-Ignition.csv standard-Suppression.csv potential_climatology-Fuel.csv potential_climatology-Moisture.csv potential_climatology-Weather.csv potential_climatology-Wind.csv potential_climatology-Ignition.csv potential_climatology-Suppression.csv

    Grouping in the State of Wildfires report:

    • Fuel & Moisture → Fuel

    • Weather & Wind → Weather

    • Ignitions & Suppression → Ignitions/Human

• samples – Full spatial posterior samples for each driver and control experiment.
  Structure:

   

  <<model_id>>/<<experiment>>/<<variable>>/sample-predXXX.nc

  • Model ID: _21-frac_points_0.5

  • Experiment: baseline-

  • Variables:

    • Evaluate – Simulated BA

    • Standard_[N] – Standard limitation for driver N,  Where N is:

      • • 0: Fuel

        • 1: Moisture

        • 2: Weather

        • 3: Wind

        • 4: Ignitions

        • 5: Suppression

    • Potential_climatology[N] – Increase in BA from removal of limitation N
      (N mapping as above)

  • Pairing: Samples are paired across variables within the same optimisation run (i.e., sample-predX.nc for Evaluate corresponds to the same ensemble member as sample-predX.nc for Standard_[N]).

Note: A separate repository exists for Congo Basin/Southern California State of Wildfires 2024/25 regions here: