Clear-sky profile database for the development of Land Surface Temperature algorithms

This dataset includes clear sky atmospheric profiles from the European Centre for Medium Range Forecast (ECMWF) version-5 reanalysis (ERA5), specially selected to support the development of algorithms of Land Surface Temperature (LST) retrieval from Earth observation (EO) data. The profiles were re-sampled from an ERA5 dataset covering the 2009-2019 period, with a 1x1 degree spatial resolution, hourly sampling and using the full vertical resolution (137 model levels). The re-sampling technique is based on a dissimilarity criterion applied to profiles of temperature and specific humidity, in order to obtain regular distributions of atmospheric variables of relevance for LST retrieval in the Thermal Infrared (TIR) spectral range. The database is limited to clear-sky conditions over land, being therefore suitable for the development of satellite land products relying on optical and thermal infrared imagery in general, despite targeting especially LST.

Dataset description:

The dataset is divided in multiple netCDF4 files based on the range of skin temperature (Tskin; Kelvin) and the range of total column water vapour (TCWV; mm). Each file includes the following variables:

• Time
• Longitude
• Latitude
• 2-m temperature (t2m)
• Surface pressure (sp)
• Total cloud cover (tcc)
• Total column water vapour (tcwv)
• Skin temperature (skt)
• Surface emissivity (emis)
• Land cover classification (lcc)
• Temperature profile (t)
• Specific humidity profile (q)
• Ozone profile (o3)
• Pressure profile (p)

All profiles are provided on model levels. For each profile, 6 values of skin temperature and 25 values of emissivity are provided (see publication for details). Emissivity values correspond to the wavelengths of ~11 and ~12 µm.

Credit:

To use this data please cite this dataset and the respective journal publication:

Ermida, S.L.; Trigo, I.F. (2022) A Comprehensive Clear-Sky Database for the Development of Land Surface Temperature Algorithms. Remote Sens., 14, 2329. https://doi.org/10.3390/rs14102329 

Access:

Currently, Zenodo does not provide a simple way to download datasets with a large number of files. We recomend trying the Zenodo_get to simplify the download.