Harmonising the sharing of marine observation data considering data quality information

<p>Marine observation data is an important source of information for scientists to investigate the state of the ocean environment. In order to use data from different sources it is critical to understand how the data was acquired. This includes not only information about the measurement process and data processing steps, but also details on data quality and uncertainty. The latter aspect becomes especially important if data from different types of instruments shall be used. An example for this is the combined use of expensive high-precision instruments in conjunction with lower-cost but less precise instruments in order to densify observation networks.</p> <p>Within this contribution we will present the work of the European MINKE project which intends, among further objectives, to facilitate the quality-aware and interoperable exchange of marine observation data.</p> <p>For this purpose, a comprehensive review of existing interoperability standards and encodings has been performed by the project partners. This included aspects such as:</p> <ul> <li>standards for encoding observation data</li> <li>standards for describing sensor data (metadata)</li> <li>Internet of Things protocols for transmitting data from sensing devices</li> <li>interfaces for data access</li> </ul> <p>From a technical perspective, the evaluation has especially considered developments such as the OGC API family of standards, lightweight data and metadata encodings, as well as developments coming from the Internet of Things community. This has been complemented by an investigation of relevant vocabularies that may be used for enabling semantic interoperability through a common terminology within data sets and corresponding metadata.</p> <p>Furthermore, specific consideration was given to the description of different properties that help to assess the quality of an observation data sets. This comprises not only the description of the data itself but also quality related aspects of data acquisition processes. For this purpose, the MINKE project is working on recommendations how to enhance the analysed (meta) data models and encodings with further elements to better transport critical information for better interpreting data sources with regard to the accuracy, uncertainty and re-usability.</p> <p>Within our contribution we will present the current state of the work within the MINKE project, the results achieved so far and the practical implementations that are performed in cooperation with the project partners.</p>