Preprints
https://doi.org/10.5194/essd-2021-253
https://doi.org/10.5194/essd-2021-253

  12 Oct 2021

12 Oct 2021

Review status: this preprint is currently under review for the journal ESSD.

TimeSpec4LULC: A Global Deep Learning-driven Dataset of MODIS Terra-Aqua Multi-Spectral Time Series for LULC Mapping and Change Detection

Rohaifa Khaldi1, Domingo Alcaraz-Segura1,3, Emilio Guirado4, Yassir Benhammou5, Abdellatif El Afia2, Francisco Herrera5, and Siham Tabik5 Rohaifa Khaldi et al.
  • 1Dept. of Botany, Faculty of Science, University of Granada, 18071 Granada, Spain
  • 2ENSIAS, Mohammed V University, Rabat, 10170, Morocco
  • 3iEcolab, Inter-University Institute for Earth System Research, University of Granada, 18006 Granada, Spain
  • 4Multidisciplinary Institute for Environment Studies “Ramón Margalef”, University of Alicante, 03690, Spain
  • 5Dept. of Computer Science and Artificial Intelligence, Andalusian Research Institute in Data Science and Computational Intelligence, DaSCI, University of Granada, 18071, Granada, Spain

Abstract. Land Use and Land Cover (LULCs) mapping and change detection are of paramount importance to understand the distribution and effectively monitor the dynamics of the Earth’s system. An unexplored way to create global LULC maps is by building good quality LULC-models based on state-of-the-art deep learning networks. Building such models requires large global good quality time series LULC datasets, which are not available yet. This paper presents TimeSpec4LULC (Khaldi et al., 2021), a smart open-source global dataset of multi-Spectral Time series for 29 LULC classes. TimeSpec4LULC was built based on the 7 spectral bands of MODIS sensor at 500 m resolution from 2002 to 2021, and was annotated using a spatial agreement across the 15 global LULC products available in Google Earth Engine. The 19-year monthly time series of the seven bands were created globally by: (1) applying different spatio-temporal quality assessment filters on MODIS Terra and Aqua satellites, (2) aggregating their original 8-day temporal granularity into monthly composites, (3) merging their data into a Terra+Aqua combined time series, and (4) extracting, at the pixel level, 11.85 million time series for the 7 bands along with a set of metadata about geographic coordinates, country and departmental divisions, spatio-temporal consistency across LULC products, temporal data availability, and the global human modification index. To assess the annotation quality of the dataset, a sample of 100 pixels, evenly distributed around the world, from each LULC class, was selected and validated by experts using very high resolution images from both Google Earth and Bing Maps imagery. This smartly, pre-processed, and annotated dataset is targeted towards scientific users interested in developing and evaluating various machine learning models, including deep learning networks, to perform global LULC mapping and change detection.

Rohaifa Khaldi et al.

Status: open (until 07 Dec 2021)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Rohaifa Khaldi et al.

Data sets

TimeSpec4LULC: A deep learning-oriented global dataset of MODIS Terra-Aqua multi-spectral time series measured from 2002 to 2021 for LULC mapping and change detection Rohaifa Khaldi; Domingo Alcaraz-Segura; Emilio Guirado; Yassir Benhammou; Siham Tabik; Francisco Herrera; Abdellatif El Afia https://doi.org/10.5281/zenodo.5020024

Rohaifa Khaldi et al.

Viewed

Total article views: 264 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
203 58 3 264 3 4
  • HTML: 203
  • PDF: 58
  • XML: 3
  • Total: 264
  • BibTeX: 3
  • EndNote: 4
Views and downloads (calculated since 12 Oct 2021)
Cumulative views and downloads (calculated since 12 Oct 2021)

Viewed (geographical distribution)

Total article views: 233 (including HTML, PDF, and XML) Thereof 233 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 24 Oct 2021
Download
Short summary
This dataset with millions of time series for seven spectral bands was built by merging Terra and Aqua satellite data and annotated for 29 LULC by spatiotemporal agreement across 15 global LULC products. Mean accuracy 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 and change detection given the disagreement between current global LULC products.