TimeSpec4LULC: a global multispectral time series database for training LULC mapping models with machine learning

Khaldi, Rohaifa; Alcaraz-Segura, Domingo; Guirado, Emilio; Benhammou, Yassir; El Afia, Abdellatif; Herrera, Francisco; Tabik, Siham

doi:https://doi.org/10.5194/essd-14-1377-2022

Articles | Volume 14, issue 3

https://doi.org/10.5194/essd-14-1377-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

Benchmark datasets and machine learning algorithms for Earth...

https://doi.org/10.5194/essd-14-1377-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 14, issue 3

Data description paper

|

30 Mar 2022

Data description paper |

| 30 Mar 2022

TimeSpec4LULC: a global multispectral time series database for training LULC mapping models with machine learning

Rohaifa Khaldi, Domingo Alcaraz-Segura, Emilio Guirado, Yassir Benhammou, Abdellatif El Afia, Francisco Herrera, and Siham Tabik

Data sets

TimeSpec4LULC: A Smart-Global Dataset of Multi-Spectral Time Series of MODIS Terra-Aqua from 2000 to 2021 for Training Machine Learning models to perform LULC Mapping Rohaifa Khaldi, Domingo Alcaraz-Segura, Emilio Guirado, Yassir Benhammou, Siham Tabik, Francisco Herrera, and Abdellatif El Afia https://doi.org/10.5281/zenodo.5913554

Short summary

This dataset with millions of 22-year time series for seven spectral bands was built by merging Terra and Aqua satellite data and annotated for 29 LULC classes by spatial–temporal agreement across 15 global LULC products. The mean F1 score was 96 % at the coarsest classification level and 87 % at the finest one. The dataset is born to develop and evaluate machine learning models to perform global LULC mapping given the disagreement between current global LULC products.