Remedy 1: Design and implementation of unified metadata format under a common data model

To develop a sustained service, metadata, and data quality and data validation are of crucial importance. Their harmonization is a requirement which is intended to establish a common understanding of the data content, to ensure correct and proper use and interpretation of the data by its owners and users, thus maximizing the benefit for the users. To address the current heterogeneity in the metadata standards, a collaborative effort among different communities and stakeholders must be undertaken. The technical approach to adopt could be of two different types:

1. A common data model merging the metadata information provided in the various existent metadata formats (CFNetCDF, WIGOS, ISO-19115, and NASA-Ames mainly) must be adopted. This allows users to provide, as realised within GAIA-CLIM, a unified metadata format (UMDF) that retains all contributing metadata and that is extendable should new metadata elements be required. This leads to an improvement in the discoverability of data and enables an easy and comprehensive conversion into a multitude of formats desired by end users. Similar efforts include the smart extensions of existing international standards like “Climate Science Modelling Language” (CSML), developed by University of Reading on the basis of ISO19115 or the UNIDATA abstract model.

The Copernicus Climate Change Service is already extending the scope of the GAIA-CLIM work for selected Baseline and Reference in-situ observations to make metadata and data compatible with Observation Data Base (ODB) developed at ECMWF. The use of a CDM (and consequently of a UMD) could make a significant attempt to improve the metadata harmonization at the international level can also facilitate the interoperability and, if possible, the integration of the existing data repositories improving the users’ access to the data from multiple suppliers and collected with different measurement techniques.

2. A different approach is to adopt or customize one broadly used standard for both discovery and observation metadata and to provide users with a number of software converters to map the metadata onto the most commonly used international standards. To date, this has been the approach adopted by various international bodies (WMO, ESA, GCOS, GEOSS, GAW...). It must be noted that this solution, as well as being more computationally consuming, might arise substantial challenges in the metadata conversion from one format to another (often left to the users themselves), with the possibility to lose information in the conversion between standards as the element-wise mapping is often not 1-to-1.