Marjani et al. (2026) ABNextFire: A Multi-Source Deep Learning Based Dataset for Wildfire Spread Prediction

Identification

• Journal: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Mohammad Marjani, Fariba Mohammadimanesh, Masoud Mahdianpari
• DOI: 10.1109/jstars.2026.3677699

Research Groups

Not available in the provided text.

Short Summary

This paper introduces ABNextFire, a multi-source deep learning based dataset specifically designed for the prediction of wildfire spread.

Objective

• To create and present ABNextFire, a multi-source dataset suitable for training deep learning models for wildfire spread prediction.

Study Configuration

• Spatial Scale: Not available in the provided text.
• Temporal Scale: Not available in the provided text.

Methodology and Data

• Models used: Deep learning based approaches are intended to use this dataset, but specific models used in its creation or validation are not detailed in the provided text.
• Data sources: Described as "Multi-Source," but specific sources (e.g., satellite imagery, weather stations, topographical data) are not detailed in the provided text.

Main Results

Not available in the provided text. The paper's primary output is the dataset itself.

Contributions

The main contribution is the creation and release of ABNextFire, a novel multi-source dataset tailored for advancing deep learning research in wildfire spread prediction.

Funding

Not available in the provided text.

Citation

@article{Marjani2026ABNextFire,
  author = {Marjani, Mohammad and Mohammadimanesh, Fariba and Mahdianpari, Masoud},
  title = {ABNextFire: A Multi-Source Deep Learning Based Dataset for Wildfire Spread Prediction},
  journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
  year = {2026},
  doi = {10.1109/jstars.2026.3677699},
  url = {https://doi.org/10.1109/jstars.2026.3677699}
}