 PRODUCT LEVELDEFINITION |
DATA PRODUCT CONCEPT |
ALGORITHMS |
     |
Product level definition |
 |
The data from all the Envisat instruments are
classified according to a hierarchical data product scheme. This scheme
consists of: Raw Data is data as received from the satellite (serial data stream, not demultiplexed) Level 0 is reformatted, time-ordered satellite data (no overlap), in computer-compatible format Level 1b consists of geolocated engineering calibrated products:
Level 2 consists of geolocated geophysical products:
|
Top of the page
 |
Data Product Concept |
To offer a simple, flexible products format and structure featuring unique product format as well as stripline processing and a floating scene concept.
Envisat product data sets will extend over a five year period, and will include data from a number of instruments and operating modes collected in continuous operation or discrete segments. In order to handle this volume and range of data, it is necessary to separate the data into manageable sizes and deliver them as product files.
The product strategy is based on:
Furthermore, as a design rule, the number of facilities generating and using product information has been minimized to limit the effect of any changes in the contents of the product header in the application software.
One product => One file
Data sets:
Auxiliary data are not timely related to the MDS record time entries. All data sets are filled with data set records (DSR), each DSR starts with a time entry. The time entry permits correlation between all data sets at record level and permits subset extraction as defined hereafter. The GADS is an exception to this rule.
Product generation and extraction
Stripline products are generated along a complete segment or an orbit. Geometric and radiometric continuity are ensured along the complete segment or orbit. Scenes can be ordered from anywhere within a stripline, then extracted and distributed via the PDS User Services (no framing constraint).
The complete product is called parent product. A user can request a subset, "child product," to be extracted from the parent product. Extraction is possible by specifying:
There is no reason why a user's region of interest should be limited to matching the archived product, either in terms of size or geographical position. For these reasons, production has been dissociated from the dissemination, and according to the floating scene concept, a multiple of the minimum scene size can only be disseminated if necessary. The size of a product can be anything from the whole data downlink segment to the minimum scene size, without any constraint on the beginning of the product. Regardless of the size, the extracted product will have a size equal to a multiple of the minimum scene size.
This flexibility is obtained by two means :
The first mean corresponds to the child product concept. A child is the result of the extraction of a set of data from a parent product (Level 0, Level 1B, Level 2). Generation of child products can be formulated in terms of time, data set (DS), or ISP selection.
The result of a child product extraction can be formulated in terms of either segments or scenes. For generation of scenes, the only extraction criteria available are << times >>. When performing child extraction, it is possible to output either a segment corresponding to the time window, or a set of scenes containing the time window.
|
|
Child product extraction |
When a child extraction is required against consecutive slices archived at the ARF, the child product generation can apply to a several parent products. These parent products are identified through their physical names and their stripline continuity indicators (SCI).
Again, the result of such a child product extraction could be formulated in term of segment or scenes.
|
|
Child product extraction from consecutive slices |
The ARF will be able to generate and distribute scenes associated to the concatenation or the extraction of the different consecutive slices it has received. For NRT mode distribution, the ARF will transmit as quickly as possible each scene to the DF, not waiting for the reception of the set of complete slices before building and transmitting the associated scenes.
Each scene will be transmitted immediately as it becomes complete.
The NRT scene generation is illustrated
below:
Scene generation in NRT
 |
Algorithms |
ESA has defined a large set of products covering the full range of Envisat-1 instruments. These products are at processing levels up to Level 2, and are either produced systematically or on request.
The algorithm development and validation activities are performed by expert support laboratories (ESL) as well as supporting industries.
The algorithm implementation activities are performed within
the payload data segment contract for the PDS operational processors.
For
more information regarding the payload data segment, please connect here to
the PDS,
found under Mission & System.
Envisat-1 instrument system simulators are developed within the Envisat-1 programme by expert support laboratories and some supporting industries. For specific information regarding these simulators, please refer to each individual instrument section.
ESA is responsible for providing the algorithm specifications and test data to the PDS contractor.
In the case of ASAR, with the exception of wave mode cross spectra, the algorithm specification has been produced directly by ESA and the PDS subcontractor is directly developing the ASAR processor and test data.
