Description

This dataset comprises joint Sentinel-2 and Landsat-8 and Landsat-9 Satellite Image Time Series (SITS) over year 2022. It is composed of 64 Areas Of Interest (AOIs) for the training set and 41 AOIs for the testing set, each covering an area of 9.9x9.9 km² (990x990 pixels for Sentinel-2 and 330x330 pixels for Landsat) covering Europe as well as a few spots in west Africa and north of South America. All dates with more than 25% of clear pixels over the AOI are included in the dataset. This yields a total of 2 984 Sentinel-2 images and 1 609 Landsat-8 and 9 images in the training set. Additional statistics about the number of dates per AOI for each sensor are presented in the following table:

	Sentinel-2				Landsat
	Total	Average	Min	Max	Total	Average	Min	Max
Train split	2984	44	11	94	1581	26	0	61
Test split	1609	39	6	98	736	18	1	56

 

Important Notice : A multi-year, wolrdwide complement to this dataset called LS2S2MYWW is available here: https://doi.org/10.6096/1029 (it is too large to be hosted on Zenodo).

For each sensor, Top-of-Canopy surface reflectance from level 2 products are used. The spectral bands included in the dataset are presented in the following table. It can be observed that the Landsat sensor does not have Red Edge bands or wide Infra-Red, and conversely Sentinel-2 sensor does not retrieve Land Surface Temperature (LST).

 

Sentinel-2		Landsat		Description
Band	Resolution (m)	Band	Resolution (m)
		B1	30.	Deep blue
B02	10.	B2	30.	Blue
B03	10.	B3	30.	Green
B04	10.	B4	30.	Red
B05, B06, B07	20.			Red Edge
B08	20.	B5	30.	Near Infra-Red
B8a	10.			Wide Near Infra-Red
B11, B12	20.	B6, B7	30.	Short Wavelength Infra-Red
		B10	100.	Land Surface Temperature

 

In addition to the spectral bands, corresponding quality masks have been used to derive a validity mask for each date of each sensor. This dataset has been gathered through the OpenEO API, in the frame of following work:

Julien Michel, Jordi Inglada. Temporal Attention Multi-Resolution Fusion of Satellite Image Time-Series, applied to Landsat-8 and Sentinel-2: all bands, any time, at best spatial resolution. 2025. ⟨hal-05101526⟩

The source code associated with the paper, including the download script that created the dataset, is available here: https://github.com/Evoland-…

File organization

Main zip files

Two main zip files are provided: ls2s2_train.zip contains the training split, and ls2s2_test.zip contains the test split. Both zip files contains one internal zip file per AOI, organized as follows.

Note that we provide test_31TCJ_12.zip as a sample for previewing the content of the dataset before downloading the train or test split.

The dataset comprises one zip file per AOI. The naming pattern for the zip file is as follows {test/train}_{mgrs_tile}_{subtile}.zip. The {test/train} field indicates if the file is part of the training or testing set. The {mgrs_tile} field correspond to the MGRS tile from which the AOI has been sampled. The {subtile} field indicate which su-btile of the MGRS tile has been sampled. Sub-tiles correspond to 1024x1204 internal JPEG2000 tiles of the Sentinel-2 product. Their numbering follows the lexicographical order (columns then rows).

Each zip file contains the following layout:

{train/test}/{mgrs_tile}_{subtile}/ 
  {mgrs_tile}_{subtile}.json 
  {mgrs_tile}_{subtile}_sentinel2_synopsis.png 
  {mgrs_tile}_{subtile}_landsat_synopsis.png 
  sentinel2/
    index.csv
    2022mmdd/ 
      sentinel2_mask_2022mmdd.tif
      sentinel2_bands_2022mmdd.tif
    ...
  landsat/  
    index.csv
    index_pan.csv
    2022mmdd/ 
      landsat_mask_2022mmdd.tif
      landsat_bands_2022mmdd.tif
      landsat_pan_mask_2022mmdd.tif
      landsat_pan_2022mmdd.tif
    ...

Files description 

Here is a description of the different files:

File name	Description
{mgrs_tile}_{subtile}.json	A json file describing the AOI.
{mgrs_tile}_{subtile}_sentinel2_synopsis.png	A synopsis PNG file allowing to see all Sentinel-2 images and mask of the AOI at a glance.
{mgrs_tile}_{subtile}_landsat_synopsis.png	A synopsis PNG file allowing to see all Landsat images and mask of the AOI at a glance.
index.csv	The csv file indexing the Sentinel-2 or Landsat data for the AOI.
index_pan.csv	The csv file indexing the Landsat panchromatic data for the AOI.
sentinel2_mask_2022mmdd.tif	990x990 pixels GeoTIFF file containing the validity mask for the current date (0 for valid and 1 for invalid).
sentinel2_bands_2022mmdd.tif	990x990 pixels GeoTIFF file containing the Sentinel-2 spectral bands are in surface reflectance * 10 000. Band order is B2, B3, B4, B5, B6, B7, B8, B8A, B11, B12. 20m bands are up-sampled to 10m resolution by means of bicubic interpolation. No data pixels have -10 000 value.
landsat_mask_2022mmdd.tif	330x330 pixels GeoTIFF file containing the validity mask for the current date (0 for valid and 1 for invalid). Spatial resolution is 30m.
landsat_bands_2022mmdd.tif	330x330 pixels GeoTIFF file containing the Landsat spectral bands are in reflectance * 10 000, except for B10 which is in LST (in °Kelvin) *10. Band order is B1, B2, B3, B4, B5, B6, B7, B10. Spatial resolution is 30m (though B10 native resolution is 100m).  No data pixels have -10 000 value.
landsat_pan_mask_2022mmdd.tif	660x660 pixels GeoTIFF file containing the panchromatic validity mask for the current date (0 for valid and 1 for invalid). Spatial resolution is 15m.
landsat_pan_2022mmdd.tif	660x660 pixels GeoTIFF file containing the Landsat panchromatic channel top of atmosphere reflectance * 10 000. Spatial resolution is 15m.  No data pixels have -10 000 value.

csv index files

  The index.csv files serve as index to the available images. It contains one row per acquisition date for the AOI, and provides the following fields:
 

Field Name	Description
product_id	A string formatted as follows: {sensor}_{2022mmdd}
acquisition_date	The acquisition date formatted as {2022-mm-dd}
valid	Dates that have clear pixel rate below 25% are not included in the dataset, hence the valid flag with be False for these dates, and True for the other.
bands	Relative path from csv file to the image file for this date and sensor: {2022mmdd}/{sensor}_bands_{2022mmdd}.tif
mask	Relative path from csv file to the image file for this date and sensor: {2022mmdd}/{sensor}_mask_{2022mmdd}.tif

  Note that a dedicated csv index file exist for Landsat panchromatic images.

Licence

Contains modified Copernicus Sentinel data [2022]. Copernicus Sentinel data are free and open access data regulated by the Sentinel_Data_Legal_Notice.pdf that can be found in the dataset repository.

Contains modified Landsat data from Collection 2 archive [2022]. Landsat data are in public domain.
Crawford, C.J., Roy, D.P., Arab, S., Barnes, C., Vermote, E., Hulley, G., Gerace, A., Choate, M., Engebretson, C., et al. 2023. “The 50-Year Landsat Collection 2 Archive.” Science of Remote Sensing (2023)  doi.org.

In respect of the above, the LS2S2 dataset is licensed under the following:

Creative Commons Attribution 4.0 International
The Creative Commons Attribution license allows re-distribution and re-use of a licensed work on the condition that the creator is appropriately credited.

Acknowledgments

This work was partly performed using HPC resources from GENCI-IDRIS (Grant 2023-AD010114835)

This work was partly performed using HPC resources from CNES Computing Center.

The authors acknowledge funding from the EvoLand project (Evolution of the Copernicus Land Service portfolio, grant agreement No 101082130) funded from the European Union's Horizon Europe research and innovation program.