CanadaFiresD: Downsampled Wildfire Data for Statistical Analysis

CanadaFiresD is a subset of wildfire data referenced in the SOAK paper by Hocking et al. published in Statistical Analysis and Data Mining (2026). The dataset is described as a time/space subset and is a downsampled version of a larger collection. It is shared under a CC-BY-4.0 license on the OpenML platform.

Use Cases

• Modeling wildfire occurrence patterns based on the described time/space subset structure.
• Analyzing spatial correlations in fire events based on the geospatial nature of the data.
• Benchmarking statistical or machine learning methods for environmental event detection as referenced in the SOAK paper.
• Studying climate change impacts on fire frequency using the downsampled temporal records.

Strengths

• Dataset is associated with a peer-reviewed publication (SOAK paper in Statistical Analysis and Data Mining, 2026).
• Released under a permissive CC-BY-4.0 license, facilitating reuse and sharing.
• Platform tags indicate a clear focus on wildfire, climate change, and geospatial analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file size, and sample data are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

OpenML platform, associated with the SOAK paper by Hocking et al.

Collection Method

Likely derived from original wildfire monitoring sources and downsampled for analysis.

Time Range

Temporal coverage is not specified in the provided metadata.

Freshness

Publication year is 2026, but the specific data collection period is unknown.

Geography

The dataset title suggests a focus on Canada.