8. GRAS Products Validation and Monitoring
For more recent information, please check for relevant documents on the EUMETSAT webpage Product Validation Reports.
The overall objective of the GRAS characterisation, calibration and validation activities is to ensure that, after the commissioning phase and thereafter during the mission lifetime, the GRAS instrument achieves its expected performance with respect to the GRAS requirements specification [RD7], and that the products satisfy the EPS end-user requirements specified in [RD1]. A further objective is that the GRAS product accuracy will continuously improve as far as possible to satisfy the evolving state-of-the-art user requirements. The following specific objectives determine the product Cal/Val activities for GRAS:
It can be expected that corrections to the GRAS processing will periodically require a reprocessing of the complete data set during routine operations to ensure the consistency of the long-term data record.
GRAS In-Orbit Verification (GRAS IOV) has been carried out under the responsibility of the European Space Agency (ESA). GRAS passed the test review board on 23 January 2007 and the hardware is considered operational. The primary objective of GRAS IOV was to verify that the instrument meets its functional and performance requirements. This has been achieved by exercising specific instrument operations, first via manual commanding and then using dedicated test timelines, and by analysis of raw data from the S bands using dedicated test tools. Demonstration of this nominal instrument performance is a prerequisite for successful GRAS IOV and Commissioning Phase Hand-over Reviews. In addition, GRAS IOV activities provide significant input to the planning of commissioning phase and routine operations. The specific functional and performance tests carried out during IOV are fully detailed in [RD4].
Verification of the correct functioning of the GRAS instrument requires continuous instrument monitoring activities. These activities start during the commissioning phase, specifically during in-orbit verification of the instrument function and performance, and continue during the remainder of the commissioning phase (which partly ends with a Product Validation Board review) and during routine operations [RD4]. Furthermore, instrument characteristics might change during the GRAS lifetime, e.g. caused by software upgrades. Operational instrument monitoring is performed in the Cal/Val Facility (CVF) at EUMETSAT, a dedicated calibration and validation network for all Metop instruments, as part of the calibration and validation activities. GRAS-specific parameters (such as tracking state and number of tracked occultations) are monitored to assure the instrument's health. A very basic instrument monitoring is also performed by the Quality Control Facility within the Core Ground Segment at EUMETSAT.
A general overview of the GRAS calibration and validation activities is given in [RD8]. Validation can be separated into two parts, the so-called Precise Orbit Determination (POD) and the profile validation.
The data processing requires accurate knowledge of the GPS and Metop satellite positions and velocities.
GPS satellite positions and velocities are provided by the Ground Support Network (GSN), a service that provides GPS state vectors, clock offset estimates and ground-based measurements from a network of globally distributed fiducial stations. This service is provided by ESA/ESOC. The accuracy of the GPS orbits provided by the GSN is validated by comparing them to International GNSS Service (IGS) orbits. Additionally, the timeliness and completeness of the GSN data is monitored.
Metop satellite positions are calculated based on the GRAS navigation measurements and the GSN data in near real-time. Validation of this calculated orbit is done in the CVF within a batch process, using the ESOC Navigation Package for Earth Observation Satellites (NAPEOS) software. Additionally, several further validation approaches are performed/planned:
For a good description of the validation strategy for GRAS
see EUMETSAT Technical Memorandum 12 [SCD6].
Level 1b profiles as provided by GRAS contain, among other parameters, bending angle over impact parameters. These bending angles represent the main measurements that are assimilated at NWP centres. (Note: NWP centres might also choose to assimilate refractivity profiles as provided by the GRAS SAF.) Except for other radio occultation instruments, bending angles are generally not measured, thus a continuous, operational validation of these profiles is problematic. Hence the core validation process is based on a one-dimensional variational assimilation (1DVar) approach. This provides bending angles along with robust statistics for the GRAS validation. Additionally, temperature, pressure and humidity profiles (which present the main information that is extracted from radio occultation measurements by NWP centre assimilation) are provided by the 1DVar. The availability of these profiles allows further validation of the GRAS measurements by comparing them to instruments providing these profiles, such as radiosondes or Lidar. Hence, the validation is partly based on the variables which are of main interest to the NWP centres.
The tool for these activities is the Radio Occultation Processing Package (ROPP) software tool [GS3]. It uses a 1DVar approach to derive temperature, water vapour and pressure profiles from bending angles; background information is taken generally from the ECMWF. A so-called forward model maps the atmospheric state (ECMWF profiles) onto measurement state (bending angle profiles), hence validation can be performed in measurement space and in atmospheric state space for all available occultations.
Routine instrument monitoring and validation are based on the 1DVar output:
This validation is performed continuously by collocating GRAS measurements with other available external radio occultation missions. Possible missions are the COSMIC constellation, see e.g. [SCD3], or the CHAMP mission [SCD2]. Direct validation of bending angles is possible with this approach, but it also allows the EUMETSAT processing to be validated by using raw measurements from these external missions, running them through the POD and the profile generation, and validating the generated profiles with the external ones. Note that this validation will not allow all GRAS measurements to be accommodated since it relies on collocation.
Radiosondes provide temperature, pressure and humidity profiles. These could be used to generate a refractivity and bending angle profile for validation. Although, as mentioned above, the 1DVar output can also be used to validate directly in atmospheric parameter space, note that this validation will not allow all GRAS measurements to be accommodated since it relies on collocation.
Lidar instruments generally provide temperature profiles in the upper atmosphere. These can be used for temperature validation. Note that this validation will not allow all GRAS measurements to be accommodated since it relies on collocation.
As noted above, validation of GRAS SAF products is under the responsibility of the GRAS SAF itself. The validation activities planned by the GRAS SAF are described in detail at [GS1], [GS2]. They comprise:
Validation results from these activities, and also the diagnostic quantities produced in the generation of Level 2 products, provide valuable feedback to the Level 1 validation activities. Similarly, Level 1 verification and validation activities provide necessary input to the GRAS SAF validation activities.
An EPS/Metop Research Announcement of Opportunity (RAO) to be coordinated by EUMETSAT and ESA was started. The primary objectives of the Announcement of Opportunity are: