DESCRIPTION: PR-OBS-3
Precipitation rate at ground by GEO/IR supported by LEO/MW

Instantaneous precipitation maps generated by IR images from operational geostationary satellites "calibrated" by precipitation measurements from MW images in sun-synchronous orbits, processed soon after each acquisition of a new image from GEO ("Rapid Update") and presented in the natural projection of the image from GEO properly arranged so as to implement animation.

  • Coverage: The rectangular area of the Meteosat field of view that includes the H-SAF area limited to 60° N [i.e. 25-60°N lat instead of 25-75°N lat, 25°W-45°E long]
  • Cycle: 15 min
  • Resolution: Average over Europe: 8 km (controlled by the IR pixel size)
  • Accuracy: 40-80 % (> 10 mm/h), 80-160 % (1-10 mm/h), not applicable for low rate (more suitable for convective precipitation)
  • Timeliness: Within 5 min from the end of (real time) acquisition
  • Dissemination: By dedicated lines to centres connected by GTS - By EUMETCast to most other users, especially scientific
  • Formats: Primary: values in fixed grid points of the Meteosat projection - Also animations of image-like files

Short description of the basic principles for product generation

The relationship linking IR brightness temperature and precipitation is very much indirect, since IR is only sensitive to the cloud top structure. Measurements are qualitative and mostly applicable to convective precipitation. After an initial start-up phase of NHOURS (NHOURS is a tunable parameter set to 24 h), needed to build meaningful statistical relationships over the entire study area), every time that a MW overpass is available, the corresponding IR image is "calibrated" against the precipitation measurement from MW. The "calibration" is thereafter propagated to follow-on IR images, till the next MW image is available. Any sort of MW image (SSMIS, AMSU, ...) are useful for this purpose and, in effect, any precipitation measurement from any source could be used. The "Rapid-update" method is being used operationally in NOAA and experimentally in Europe. It is useful for nowcasting but, for hydrological purposes (computation of accumulated precipitation), the quality may be insufficient because of the bias towards convective precipitation.