In order to satisfy the needs of the Observatory Natural Carbon Flux (ONC) that uses DLR as an atmospheric forcing parameter in carbon and water cycle modelling, Institute of Meteorology of Portugal (IM) has developed a strategy which consists in using jointly satellite images and Numerical Weather Prediction (NWP) data. The algorithm computes the DLR for clear-sky conditions following an hybrid method based upon a bulk parameterisation proposed by Dilley and O’Brien (1998). This scheme depends on two-meter air temperature and precipitable water content. For cloudy sky conditions, the DLR is computed using the parameterisation scheme proposed by Josey et al. (2003). It is based on the relationship of the effective temperature of a blackbody with a radiative flux equivalent to that from the atmosphere in terms of two-meter air temperature, total cloud amount and two-meter dew point temperature. Two-meter air temperature, two-meter dew point temperature and precipitable water content fields are obtained from ECMWF model forecast and the total cloudiness data are obtained from METEOSAT processing. The image below shows an example of DLR product.

The validation of the algorithms includes feasibility analyses, sensitivity studies, assessment of prototypes. The validation of the DLR product primarily relies on regular measurements to be carried out on existing permanent ground stations from CarboEurope, BSRN, SURFRAD and ARM network. The DLR retrievals may also be compared with satellite results obtained in the framework of other projects or with the output from numerical weather prediction models. The validation is performed in close cooperation with the Karlsruhe University (IMK).

The DLR product presently available covers Europe and Africa for the period 1999 - 2005, with a temporal resolution of 6 hours. It can be downloaded with an algorithmic documentation (ATBD) and a technical documentation (readme).

DLR derived from METEOSAT-7 data, 14th August 2000.