Abstract
Abstract. Most research activities in Earth System Sciences (ESS) are data-driven. There is a growing need to establish innovative, cross-cutting data management and data analysis methods in ESS to support the collaboration of interdisciplinary research building on heterogeneous sources. Data management plans (DMPs) are structured documents that outline data handling and include for instance agreements on roles, specifications of data products, and definition of workflows. However, the structure of existing DMP templates is mostly designed for funder’s requirements and consequently address only the broad and interdisciplinary research community. Thus, these templates do lack (1) guidance on how to structure domain-specific information in a DMP – by providing domain-specific profiles, e.g. to harmonize the structure and improve the comprehensibility of DMP instances and (2) (linking into) tools enabling efficient management and reuse of information / sections of DMP instances. Therefore, we provide a concept of future DMP templates and address geo-domain-specific requirements, and the integration of DMPs into research data infrastructures. We recommend integrating structured provenance and quality information, using established concepts, and define a pathway to link tools into research data infrastructures, such that they foster automation of data management workflows and data reuse.
Highlights
Most Earth System Science (ESS) research projects are data-driven and produce datasets as main results
Linking to ground truth will improve the detailed evaluation of quality parameters and the reusability of the data. - provide and link to quality datasets, showing geo-located quality information, facilitating the detailed evaluation of certain regions and the automated usage of quality information, e.g. as input dataset in modelling tools. - describe sources of quality information to underpin the reliability of the given information
- link to workflow management systems, e.g. to better update provenance information for geospatial workflows and to support modeldriven workflow development starting from the structured Data management plans (DMPs) description in the DMP instance to generate code snippets
Summary
Most Earth System Science (ESS) research projects are data-driven and produce datasets as main results. DMPs are evolving and several communities 6 are addressing some of the challenges in the provision of dynamic DMPs as living documents, used collaboratively along the data life cycle, stored in catalogues 7 This includes requirements on being machine-actionable (Miksa et al 2019), domain specific through the Data Domain Protocols (DDP) 8 and making DMPs findable, accessible, interoperable, and reusable (FAIR) All of these challenges, in particular making DMPs FAIR, machine-actionable and domain-specific are closely linked and will improve transparency, reuse of information and provide interoperability within a data infrastructure e.g. using sections of DMP instances to fill dataset metadata (Davidson et al 2019).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
