How are products generated
The data products generation is the last step of a process made of several phases. As first there is the data collection, then data are quality checked at different levels and when needed they are aggregated. At the end of these steps data are ready for the generation of different kind of products depending on their features.
The quality of the data is a key issue when merging heterogeneous data coming from different sources, periods and geographic areas. Within EMODnet chemistry community, commonly agreed and standardized data quality control (QC) protocols have been defined to guarantee consistency among comprehensive databases which include data from different and/or unknown origin and covering long time periods.
As a first step, the data are checked and completed by collators with a standard set of metadata that provide the basic information necessary for their long term use. Afterwards, data undergo a validation loop which consists in several validation steps. The first is done by data collators, prior to the inclusion in the decentralized infrastructure and the second step, which consists in regional quality control, is performed at regional scale on aggregated datasets. The first quality controls (QC) ensure that position and time of data are realistic and compare measurements with broad ranges and specific regional ranges. Whenever available, data are also compared with climatology. As a result of the first QC step, all data are archived with a quality flag value that provides information about their reliability.
At this point, data aggregation and regional quality control are performed by regional leaders, following a common protocol.
On top of the quality checked data different kind of data products are generated.
Interpolated maps are now generated, mainly for nutrients, with 10 years moving window in order to find a balance between the duration of the environmental evaluation cycle for Member States (to provide maps with a time frame near to the 6 years process of the MS evaluation) and the number of years that guarantee a sufficient data coverage. Guidelines for data products generation have been implemented to provide a set of common rules to obtain visualization products homogeneous for all European seas.
Profiles and time series plots are automatically generated starting from the data available in the Regional buffers, produced thanks to a service bases on WPS OGC standard, and can be dynamically customized.