Skip to content

Constellr LST30

Constellr LST30 consists of homogenized rasters of Land Surface Temperature (LST) data. These are their characteristics:

  • All thermal datasets are spectrally harmonised using a single in-house algorithm to retrieve LST.

  • The data are delivered as if they were coming from a single satellite constellation.

  • The geometry is aligned in all images: all images of an are of interest will show a perfect pixel alignment, regardless of the source.

  • Each pixel holds a temperature value in degree Kelvin.

  • Pixels that do not contain temperature values hold a ‘nodata’ value. Some zones may show empty pixels in several or all images, due to issues in the data processing from the public providers.

Specification

Have a look at our Technical specifications documentation Here

File-naming convention

Various layers are provided and follow this compact naming convention:

YYYYMMDDTHHMMSS_aa

Where

  • YYYYMMDDTHHMMSS is the datatake sensing time of the land surface temperature in isoformat. Time is based on Universal Time Coordinated (UTC), also known as Greenwich Mean Time (GMT).

  • aa indicates what contains the file

    • LST_AOI.tif: LST raster pixel value: Kelvin Degree

    • CLOUDS_AOI.tif: Cloud mask raster

      pixel values: 0=clear pixel, 1=contamined pixel

      cloud mask is made of cloud and shadow pixels

    • .json: Metadata json

Data Sources

Constellr LST30 leverages data from different missions:

  • Lisr, constellr property
  • Landsat 8 and 9 from USGS
  • Era5 and CAMS from CDSAPI
  • MODIS (soon)
  • Sentinel 3 (soon)

Legal notice: Contains modified Copernicus Sentinel data, Landsat 8 and 9 images (U.S. Geological, Survey), Copernicus Climate Change Service information, Copernicus Service information, modified Modis data (courtesy of the NASA - LP DAAC) and modified Ecostress data (courtesy of the NASA - JPL).