SpaceML/UnorthoDOS

UnorthoDOS - Unorthorectified Dataset for On board Satellite methane detection

This repository contains two ML-ready datasets, comprising orthorectified and unorthorectified hyperspectral images from the Earth Surface Mineral Dust Source Investigation (EMIT) sensor, for methane detection. This dataset supports the research presented in the paper Towards Methane Detection Onboard Satellites.

• Paper: Towards Methane Detection Onboard Satellites
• Code: github.com/spaceml-org/plume-hunter

Due to storage limitations, this repository includes only the subset of images featuring methane plumes. To access the full dataset used in the associated publication, please feel free to contact us.

Citation

If you find the code or the UnorthoDOS datasets useful in your research, please consider citing our work.

📦 Dataset

@misc{UnorthoDOS,
    author       = { Maggie Chen and Hala Lambdouar and Luca Marini and Laura Martínez-Ferrer and Chris Bridges and Giacomo Acciarini },
    title        = { UnorthoDOS (Revision a370bd0) },
    year         = 2025,
    url          = { https://huggingface.co/datasets/SpaceML/UnorthoDOS },
    doi          = { 10.57967/hf/6778 },
    publisher    = { Hugging Face }
}

📄 Paper

@article{chen2025towards,
  title={Towards Methane Detection Onboard Satellites},
  author={Chen, Maggie and Lambdouar, Hala and Marini, Luca and Mart{\'\i}nez-Ferrer, Laura and Bridges, Chris and Acciarini, Giacomo},
  journal={arXiv preprint arXiv:2509.00626},
  year={2025}
}