78
Chapter 10
2.
EXTRACTION AND EVALUATION OF
METEOROLOGICAL PRODUCTS
The meteorological geostationary satellite currently exploited by
EUMETSAT has three channels
1 with a 30 minutes image repeat cycle.
Quantitative meteorological products such as cloud motion winds (CMW),
sea surface temperatures (SST), clear sky radiances (CSR), cloud cover,
cloud top temperature and cloud top height (CTH) are automatically
extracted from these images. This operational processing chain, initiated
almost 20 years ago (ESA 1987), has reached a very high level of maturity
and automation. The main product extraction processing steps are
summarized here after:
1. Data reception. MPEF real-time input is composed of Meteosat
observations, forecast data from the European Centre for Medium-range
Weather Forecasts (ECMWF) and finally meteorological observations.
2. Pre-processing. The pre-processing phase consists in a series of data
manipulations. These include satellite data calibration based on ground
observations and radiosondes (Gube et al. 1996; van de Berg et al.
1995), the computation of scene predicted radiances (Schmetz 1986) and
finally image analysis. The latter involves cloudy pixel identification and
cloud top height estimation, accounting for the presence of semitransparent clouds (Szejwach 1982).
3. Product extraction. Meteorological products are then automatically
generated. Examples of the CMW product can be found at the following
URL address: http://www.eumetsat.de. Separate wind vectors
are extracted from the different channels. Each wind vector is
characterized by an altitude (low, medium or high), a direction, a speed
and finally a quality value (Schmetz and Holmlund 1990).
4. Quality control. The quality control step generates information
concerning product accuracy and reliability. In the case of the CMW
product, each wind is assigned a quality score based on a series of
consistency checks, which are combined to give an overall value. These
verifications concern (1) a test on the temporal direction and speed
change, (2) a test on the spatial direction and speed change, and (3) a
comparison with forecast values (Elliott and Holmlund 1996).
Consistent quality flags are of primary importance to ensure the optimal
product interpretation.
5. Product dissemination. The products are both archived in the
Meteorological Archive and Retrieval Facility (MARF) and distributed
1 Visible:
water vapor:
and thermal infrared:
Chapter 10
2.
EXTRACTION AND EVALUATION OF
METEOROLOGICAL PRODUCTS
The meteorological geostationary satellite currently exploited by
EUMETSAT has three channels
1 with a 30 minutes image repeat cycle.
Quantitative meteorological products such as cloud motion winds (CMW),
sea surface temperatures (SST), clear sky radiances (CSR), cloud cover,
cloud top temperature and cloud top height (CTH) are automatically
extracted from these images. This operational processing chain, initiated
almost 20 years ago (ESA 1987), has reached a very high level of maturity
and automation. The main product extraction processing steps are
summarized here after:
1. Data reception. MPEF real-time input is composed of Meteosat
observations, forecast data from the European Centre for Medium-range
Weather Forecasts (ECMWF) and finally meteorological observations.
2. Pre-processing. The pre-processing phase consists in a series of data
manipulations. These include satellite data calibration based on ground
observations and radiosondes (Gube et al. 1996; van de Berg et al.
1995), the computation of scene predicted radiances (Schmetz 1986) and
finally image analysis. The latter involves cloudy pixel identification and
cloud top height estimation, accounting for the presence of semitransparent clouds (Szejwach 1982).
3. Product extraction. Meteorological products are then automatically
generated. Examples of the CMW product can be found at the following
URL address: http://www.eumetsat.de. Separate wind vectors
are extracted from the different channels. Each wind vector is
characterized by an altitude (low, medium or high), a direction, a speed
and finally a quality value (Schmetz and Holmlund 1990).
4. Quality control. The quality control step generates information
concerning product accuracy and reliability. In the case of the CMW
product, each wind is assigned a quality score based on a series of
consistency checks, which are combined to give an overall value. These
verifications concern (1) a test on the temporal direction and speed
change, (2) a test on the spatial direction and speed change, and (3) a
comparison with forecast values (Elliott and Holmlund 1996).
Consistent quality flags are of primary importance to ensure the optimal
product interpretation.
5. Product dissemination. The products are both archived in the
Meteorological Archive and Retrieval Facility (MARF) and distributed
1 Visible:
water vapor:
and thermal infrared: