Who are the target users of this service?
Primarily public bodies concerned with environmental protection and water quality. This includes both large regional organisations such as the European Environment Agency, and local institutions such as local agencies for environmental protection
How will these users access this service?
Users will access the service via a dedicated web portal. This portal will provide state of the art tools for discovering, accessing and analysing the existing data and a platform to produce new data according to the users' specific needs.
What products does this service provide?
The products generated by this service are mainly Water Quality Indicators that are related with well-established policies, such as the Marine Framework Directive. The main data products which will be generated are maps of:
- Water temperature
- Water salinity
- Mixing indicators
- Upwelling indicator
- Kinetic energy
- Water density
These cover the Mediterranean at the moment - though the approach has global potential - and a variety time-ranges.
How will these products benefit users?
The general aim of the products generated by this service is to provide the users with information to support ocean monitoring in line with policy objectives.
Depending on the type of users and the areas/parameters of interest, the service allows a user to search and/or create customised versions of the products in order to best focus on their own objectives. These could include, for example, a specific sub-area, time range, customisation of parameters, trends etc.
Which Open Data sources drive this service?
Our main data sources are:
What processing is performed on this data?
Processing mainly consists of a re-analysis of the existing data in order to produce maps or plots of the indicators specified above. For example, this could be a map of average monthly salinity across the Mediterranean from 1986 to 2015, or a graph showing the trend in the average temperature of the Mediterranean between 2001-2010.
Thus the re-analysis consists of extracting input data and performing statistical analyses, aggregation, resampling etc in order to compute the desired indicator. It is also possible to re-compute the same indicator with only slight variations in the parameterisation.
How does this service use Linked Open Data?
The Open data sources described above are extracted and used as inputs to the reanalysis computations. The output of such computations is made available under the form of Linked Open Data.
How Open Data has improved this service
The open data sources are at the foundation of this service. As these data are readily available and accessible by standard means, this allows us to create a seamless workflow towards creating the service products.
How the Shared Platform has improved this service
The service is inherently based on the simple yet massive processing of data and cloud processing will allow us to dramatically reduce the computation time. This increase in computational power will allow us to execute massive data re-processing in a reasonable time and allow end users to perform a customised processing in near real time.
How LOD and/or visualisation tools have improved this service
As explained above, our putputs are made available under the form of LOD. The visualisation tools provided by MELODIES partners are also very important - this service uses some common tools such as the ncWMS server provided by the University of Reading.
Our biggest challenges so far...
- ... have been:
- Integrating our processing functions on the cloud environment. This has been achieved successfully owing to the well-organised project timetable. We began with a preliminary integration phase into a simplified environment for testing and then followed this with a scaling-up phase.
- The technical challenges of providing a discovery service , providing easy access to non-domain users. This has been implemented using the no-SQL database Solr which allows for easy and effective data search.
- Integrating display tools to the developing web platform. This has been achieved for multidimensional raster data thanks to the integration of the afore-mentioned ncWMS tool.