IgniWise Dataset — Prescribed Burn Window Prediction for Spain (1983–2015)

Fuente: Zenodo
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Romera Martínez, Sergio
Format: Recurso digital
Sprache:Spanisch
Veröffentlicht: Zenodo 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866902154394992640
author Romera Martínez, Sergio
author_facet Romera Martínez, Sergio
contents <p>Machine learning dataset for predicting safe prescribed burn windows across <br>48 Spanish provinces, supporting the IgniWise open-source system.</p> <p>Dataset contents:<br>- training_data.csv: 11,996 records derived from historical fire occurrence <br>  data (MITECO / IEPNB, 1983–2015) with 20 features including meteorological <br>  variables (statistically approximated from Spanish climate distributions), <br>  FWI indices (Van Wagner, 1987), and real geographic features per province<br>- random_forest_v1.pkl: Trained Random Forest model (200 estimators, <br>  5-fold CV accuracy: 99.5% ± 0.002)<br>- provincias_geo.geojson: Spanish province geometries enriched with real <br>  topographic (Copernicus DEM GLO-30), vegetation (Sentinel-2 NDVI via GEE), <br>  and land cover features (CORINE Land Cover 2018)</p> <p>Note on meteorological features: weather variables in training_data.csv are <br>statistical approximations calibrated to provincial climate. Operational <br>predictions use real-time data from OpenWeatherMap. The 99.5% CV accuracy <br>reflects model fit to the classification rules rather than independent <br>real-world validation, which requires field data from executed prescribed burns.</p> <p>Sources: MITECO/IEPNB (fire occurrences), Copernicus DEM GLO-30 (topography),<br>Sentinel-2/GEE (NDVI), CORINE Land Cover 2018 (vegetation cover).<br>FWI methodology: Van Wagner, C.E. (1987), Canadian Forest Service.</p> <p>License: CC BY 4.0 | Web: igniwise.com | Code: github.com/TrueRomanZe/igniwise</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_19144668
institution Zenodo
language spa
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle IgniWise Dataset — Prescribed Burn Window Prediction for Spain (1983–2015)
Romera Martínez, Sergio
Wildfires
Machine Learning
Spain
Prescribed Burns
Fire Weather Index
Random Forest
Wildfire Prediction
Fire Management
<p>Machine learning dataset for predicting safe prescribed burn windows across <br>48 Spanish provinces, supporting the IgniWise open-source system.</p> <p>Dataset contents:<br>- training_data.csv: 11,996 records derived from historical fire occurrence <br>  data (MITECO / IEPNB, 1983–2015) with 20 features including meteorological <br>  variables (statistically approximated from Spanish climate distributions), <br>  FWI indices (Van Wagner, 1987), and real geographic features per province<br>- random_forest_v1.pkl: Trained Random Forest model (200 estimators, <br>  5-fold CV accuracy: 99.5% ± 0.002)<br>- provincias_geo.geojson: Spanish province geometries enriched with real <br>  topographic (Copernicus DEM GLO-30), vegetation (Sentinel-2 NDVI via GEE), <br>  and land cover features (CORINE Land Cover 2018)</p> <p>Note on meteorological features: weather variables in training_data.csv are <br>statistical approximations calibrated to provincial climate. Operational <br>predictions use real-time data from OpenWeatherMap. The 99.5% CV accuracy <br>reflects model fit to the classification rules rather than independent <br>real-world validation, which requires field data from executed prescribed burns.</p> <p>Sources: MITECO/IEPNB (fire occurrences), Copernicus DEM GLO-30 (topography),<br>Sentinel-2/GEE (NDVI), CORINE Land Cover 2018 (vegetation cover).<br>FWI methodology: Van Wagner, C.E. (1987), Canadian Forest Service.</p> <p>License: CC BY 4.0 | Web: igniwise.com | Code: github.com/TrueRomanZe/igniwise</p>
title IgniWise Dataset — Prescribed Burn Window Prediction for Spain (1983–2015)
topic Wildfires
Machine Learning
Spain
Prescribed Burns
Fire Weather Index
Random Forest
Wildfire Prediction
Fire Management
url https://doi.org/10.5281/zenodo.19144668