12. Changelog

All notable changes to this project will be documented in this file. The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.9.0] - 2026 / 05 / 18

Added

Add the firebench command line interface for benchmark discovery, data download, and benchmark execution:
- firebench list lists available benchmark cases.
- firebench data versions lists available data versions for a benchmark case.
- firebench data get downloads benchmark data.
- firebench run runs a benchmark case against a model output in FireBench standard HDF5 format.
Add regression tests covering import, CLI, benchmark discovery, and deprecated environment-variable behavior without requiring local environment configuration.
Added official PyPI/TestPyPI packaging and distribution support for firebench.
Python 3.14 support

Benchmarks

Add the FB001 2021 Caldor Fire benchmark to the benchmark registry.
Add Caldor benchmark data version metadata for CLI downloads.

Documentation

Add a CLI tutorial showing how to download the Caldor Fire case and run it.
Update setup instructions to describe default FireBench data and local database paths instead of requiring environment variables.

Changed

Remove the mandatory dependency on FIREBENCH_DATA_PATH and FIREBENCH_LOCAL_DB for import, installation, CLI usage, tests, and benchmark discovery.
Use default paths when legacy environment variables are not set:
- FireBench data is resolved from the package or repository data directory.
- The local database defaults to ~/.firebench/local_db.
Add explicit data_path and local_db_path arguments where local paths are needed.
Stop using Git LFS for repository-managed files.

Deprecated

Deprecate FIREBENCH_DATA_PATH. Use the FireBench data configuration or pass data_path explicitly instead.
Deprecate FIREBENCH_LOCAL_DB. Use the FireBench local database configuration or pass local_db_path explicitly instead.
Legacy environment-variable support remains available temporarily and now emits DeprecationWarning.

[0.8.1] - 2026 / 03 / 15

Fixes

bias metric: computes x1 mean only where x2 is not Nan

Miscellaneous

FireBench Standard file format 1.0
litting
add tests for metrics
fix security warnings

[0.8.0] - 2026 / 01 / 14

Added

2021 Caldor case FB001 documentation and benchmarks (See Zenodo FireBench for release)
package metrics: contains kpi functions, metrics functions for perimeters, 1D datasets, confusion matrix.
package standardize: contains standardization functions for landfire, mtbs, ravg, synoptic, geotiff.
package signing: contains functions for certification (hardware encryption) and verification of certificates (verify_certificate_in_dict, verify_certificates_in_h5). Verification functions require gpg (not needed for benchmarking functions).
Public Key for certificates verification

Documentation

FireBench Standard file format
Add Key Performance Indicators, Metrics, Score and Normalization information

[0.7.0] - 2025 / 08 / 09

Added

anderson_2015_stats: Plot statistics from the Anderson 2015 dataset.
array_to_geopolygons: Convert an array field into geospatial polygons at a given iso-value, preserving holes.
auto_bins: Automatically generate histogram bin edges for plotting, based on data range.
CS505_cl: Compute the half-width of the confidence interval for measurement error of the Campbell Scientific CS505 Fuel Moisture Sensor.
CS506_cl: Compute the half-width of the confidence interval for measurement error of the Campbell Scientific CS506 Fuel Moisture Sensor.
current_datetime_iso8601: Get the current datetime as an ISO 8601 formatted string (YYYY-MM-DDTHH:MM[:SS]±HH:MM).
datetime_to_iso8601: Convert a given datetime to an ISO 8601 formatted string (YYYY-MM-DDTHH:MM[:SS]±HH:MM).
jaccard_binary: Compute the IoU, i.e. Jaccard Index, between two fire perimeters described as 2D binary masks.
jaccard_polygon: Compute the Intersection over Union (IoU), i.e. Jaccard Index, between two fire perimeters described as geospatial polygons.
sorensen_dice_binary: Compute the Sorensen-Dice index between two fire perimeters described as 2D binary masks.
sorensen_dice_polygon: Compute the Sorensen-Dice index between two fire perimeters described as geospatial polygons.
read_quantity_from_fb_dataset: Read a dataset from an HDF5 file, group, or dataset node and return it as a Pint Quantity according to the FireBench I/O standard.
rmse: Compute the Root Mean Square Error (RMSE) between two arrays, ignoring NaNs.
nmse_range: Compute the Normalized Mean Square Error (NMSE) between two arrays, using the range of the reference signal as normalization.
nmse_power: Compute the Normalized Mean Square Error (NMSE) between two arrays, using the product of their mean values as normalization.

Documentation

Updated developer documentation
Refactoring of Benchmarks page. Benchmarks are now sorted by model context instead of metric context.
Tags have a colorful badge.
Add documentation for FireBench standard file format version 0.1.

Benchmarks

Improve Anderson 2015 Validation benchmark document.
Refactoring of ROS validation using Anderson 2015 dataset to use the FireBench I/O standard.

Miscellaneous

Support Python 3.13
Support numpy version >= 2.0
Dependency to matplotlib > 3.8

Fix

Comma missing in two row of Anderson 2015 dataset.

[0.6.1] - 2025 / 05 / 20

Added

Add citation metadata in CITATION.cff file

[0.6.0] - 2025 / 05 / 08

Documentation

Add benchmark proposal and run templates
Add Call for Benchmarks section in the main page

[0.5.0] - 2025 / 04 / 10

Documentation

Remove legacy GitHub Pages & Jekyll documentation.
Move the documentation to ReadTheDocs.
Add API documentation using Sphinx

[0.4.0] - 2025 / 04 / 03

Added

Urban canyon vertical wind interpolation Masson_canyon
Fuel load per element size in Anderson Fuel Model (1h, 10h, 100h, live)
Implementation of the Santoni_2011 rate of spread model

[0.3.2] - 2024 / 12 / 16

Added

import_scott_burgan_40_fuel_model wrapper function to simplify import of Scott and Burgan fuel model
import_anderson_13_fuel_model wrapper function to simplify import of Anderson fuel model
import_wudapt_fuel_model wrapper function to simplify import of WUDAPT urban fuel model
Sensitivity workflow scripts for rate of spread models using WUDAPT urban fuel model.

Fixed

Fuel moisture of extinction unit is dimensionless instead of percent in Anderson13.json

Documentation

Add benchmark results for:
- Rate of spread model sensitivity using WUDAPT urban fuel model:
  - Hamada 1
  - Hamada 2
- Rate of spread model sensitivity using Scott and Burgan 40 fuel model:
  - Rothermel_SFIRE
  - Balbi 2022
- Rate of spread model execution time:
  - Hamada 1
  - Hamada 2
Update benchmark results for:
- Rate of spread model sensitivity using Anderson 13 fuel model:
  - Rothermel_SFIRE

[0.3.1] - 2024 / 12 / 03

Added

Efficiency workflow scripts for rate of spread model performance evaluation using Anderson fuel model.

Documentation

Add benchmark results for:
- Rate of spread model execution time:
  - Rothermel_SFIRE
  - Balbi 2022

[0.3.0] - 2024 / 11 / 24

Added

ScottandBurgan40 fuel model
Scott and Burgan utility function
- add_scott_and_burgan_total_fuel_load: aggregate the fuel load per element size to the total fuel load
- add_scott_and_burgan_total_savr: calculate the total surface area to volume ratio as weighted average of the savr of the fuel elements described in SB40.
Fuel model utility functions:
- find_closest_fuel_class_by_properties: retrieve the fuel class having the closest properties to a target set of properties.
Wind reduction factor functions
- use_wind_reduction_factor from value, fuel model dictionary, or list
- Baughman_20ft_wind_reduction_factor_unsheltered from Baughman, R. G., & Albini, F. A. (1980)
- Baughman_generalized_wind_reduction_factor_unsheltered: Generalized wind reduction factor derived from Albini (1979)
Add hash of file when copying to record
Modify the date in report automatically with copying to record

Changed

External management of wind reduction factor (no more present in rate of spread models)
Management of units for rate of spread model simplified using compute_ros_with_units
Record management to save workflow

Documentation

Add Fire Models information and Dataset and fire experiment information sections
Add pages for:
- Rothermel_SFIRE
- Balbi_2022_fixed_SFIRE
- Hamada 1
- Hamada 2
- Anderson13
- ScottandBurgan40
- WUDAPT_urban
- Wind reduction factor methods
Add benchmark results for:
- Validation Anderson 2015:
  - Rotherme_SFIRE
  - Balbi 2022
- Sensitivity to environmental variable for Anderson 13 fuel model:
  - Rotherme_SFIRE
  - Balbi 2022

[0.2.0] - 2024 / 10 / 17

Added

Hamada_1 urban rate of spread
Hamada_2 urban rate of spread
Balbi 2022 vegetation rate of spread model
WUDAPT_urban fuel database

Changed

License APACHE 2.0 is used instead of MIT

Documentation

Add dependencies, developers guide, and license pages
minor fixes to fuel models tutorial

[0.1.0] - 2024 / 07 / 09

Added

Documentation for the rate of spread sensitivity workflow
Add changelog to documentation
Archive for the rate of spread sensitivity workflow with the Rothermel_SFIRE model
Change output filename for the rate of spread sensitivity workflow

Fixes

unit issues in the rate of spread sensitivity workflow

[0.0.1] - 2024 / 07 / 08

Added

Initial release of the FireBench library.
Implementation of the Rothermel_SFIRE rate of spread model.
Fuel model Anderson13 corresponding to Anderson 13 fuel categories
Basic tools and utilities for fire modeling.
Sensitivity analysis workflow for environmental variables.
- Workflow scripts 03_01_sensitivity_env_var for performing sensitivity analysis.
Integration with GitHub Actions for continuous integration.
Documentation setup using GitHub Pages.
Initial implementation of unit tests using pytest.

Features

Support for default and custom fuel models.
Functionality for reading and validating fuel model data.
Utilities for converting units and checking input data quality.
Calculation of Sobol sensitivity indices for rate of spread models.
Saving workflow results and data to HDF5 files.
Plotting of Sobol sensitivity indices for different fuel classes.

Documentation

Overview of the FireBench project.
Tutorials for using default and custom fuel models.
How-to guides for sensitivity analysis workflows.
Setup instructions for the development environment.
Contribution guidelines and code of conduct.