Curdt, Constanze (2014). Design and Implementation of a Research Data Management System: The CRC/TR32 Project Database (TR32DB). PhD thesis, Universität zu Köln.
|
PDF
Curdt_Dissertation_2014.pdf Bereitstellung unter der CC-Lizenz: Creative Commons Attribution Non-commercial. Download (41MB) |
Abstract
Research data management (RDM) includes all processes and measures which ensure that research data are well-organised, documented, preserved, stored, backed up, accessible, available, and re-usable. Corresponding RDM systems or repositories form the technical framework to support the collection, accurate documentation, storage, back-up, sharing, and provision of research data, which are created in a specific environment, like a research group or institution. The required measures for the implementation of a RDM system vary according to the discipline or purpose of data (re-)use. In the context of RDM, the documentation of research data is an essential duty. This has to be conducted by accurate, standardized, and interoperable metadata to ensure the interpretability, understandability, shareability, and long-lasting usability of the data. RDM is achieving an increasing importance, as digital information increases. New technologies enable to create more digital data, also automatically. Consequently, the volume of digital data, including big data and small data, will approximately double every two years in size. With regard to e-science, this increase of data was entitled and predicted as the data deluge. Furthermore, the paradigm change in science has led to data intensive science. Particularly scientific data that were financed by public funding are significantly demanded to be archived, documented, provided or even open accessible by different policy makers, funding agencies, journals and other institutions. RDM can prevent the loss of data, otherwise around 80-90 % of the generated research data disappear and are not available for re-use or further studies. This will lead to empty archives or RDM systems. The reasons for this course are well known and are of a technical, socio-cultural, and ethical nature, like missing user participation and data sharing knowledge, as well as lack of time or resources. In addition, the fear of exploitation and missing or limited reward for publishing and sharing data has an important role. This thesis presents an approach in handling research data of the collaborative, multidisciplinary, long-term DFG-funded research project Collaborative Research Centre/Transregio 32 (CRC/TR32) “Patterns in Soil-Vegetation-Atmosphere Systems: Monitoring, Modelling, and Data Assimilation”. In this context, a RDM system, the so-called CRC/TR32 project database (TR32DB), was designed and implemented. The TR32DB considers the demands of the project participants (e.g. heterogeneous data from different disciplines with various file sizes) and the requirements of the DFG, as well as general challenges in RDM. For this purpose, a RDM system was established that comprises a well-described self-designed metadata schema, a file-based data storage, a well-elaborated database of metadata, and a corresponding user-friendly web interface. The whole system is developed in close cooperation with the local Regional Computing Centre of the University of Cologne (RRZK), where it is also hosted. The documentation of the research data with accurate metadata is of key importance. For this purpose, an own specific TR32DB Metadata Schema was designed, consisting of multi-level metadata properties. This is distinguished in general and data type specific (e.g. data, publication, report) properties and is developed according to the project background, demands of the various data types, as well as recent associated metadata standards and principles. Consequently, it is interoperable to recent metadata standards, such as the Dublin Core, the DataCite Metadata Schema, as well as core elements of the ISO19115:2003 Metadata Standard and INSPIRE Directive. Furthermore, the schema supports optional, mandatory, and automatically generated metadata properties, as well as it provides predefined, obligatory and self-established controlled vocabulary lists. The integrated mapping to the DataCite Metadata Schema facilitates the simple application of a Digital Object Identifier (DOI) for a dataset. The file-based data storage is organized in a folder system, corresponding to the structure of the CRC/TR32 and additionally distinguishes between several data types (e.g. data, publication, report). It is embedded in the Andrew File System hosted by the RRZK. The file system is capable to store and backup all data, is highly scalable, supports location independence, and enables easy administration by Access Control Lists. In addition, the relational database management system MySQL stores the metadata according to the previous mentioned TR32DB Metadata Schema as well as further necessary administrative data. A user-friendly web-based graphical user interface enables the access to the TR32DB system. The web-interface provides metadata input, search, and download of data, as well as the visualization of important geodata is handled by an internal WebGIS. This web-interface, as well as the entire RDM system, is self-developed and adjusted to the specific demands. Overall, the TR32DB system is developed according to the needs and requirements of the CRC/TR32 scientists, fits the demands of the DFG, and considers general problems and challenges of RDM as well. With regard to changing demands of the CRC/TR32 and technologic advances, the system is and will be consequently further developed. The established TR32DB approach was already successfully applied to another interdisciplinary research project. Thus, this approach is transferable and generally capable to archive all data, generated by the CRC/TR32, with accurately, interoperable metadata to ensure the re-use of the data, beyond the end of the project.
Item Type: | Thesis (PhD thesis) | ||||||||
Translated abstract: |
|
||||||||
Creators: |
|
||||||||
URN: | urn:nbn:de:hbz:38-58820 | ||||||||
Date: | 2014 | ||||||||
Language: | English | ||||||||
Faculty: | Faculty of Mathematics and Natural Sciences | ||||||||
Divisions: | Faculty of Mathematics and Natural Sciences > Department of Geosciences > Geographisches Institut | ||||||||
Subjects: | Library and information sciences Natural sciences and mathematics Earth sciences Technology (Applied sciences) |
||||||||
Uncontrolled Keywords: |
|
||||||||
Date of oral exam: | 7 July 2014 | ||||||||
Referee: |
|
||||||||
Funders: | Deutsche Forschungsgemeinschaft (DFG) | ||||||||
Projects: | SFB/Transregio 32 "Patterns in Soil-Vegetation-Atmosphere Systems - Monitoring, Modelling and Data Assimilation" | ||||||||
Refereed: | Yes | ||||||||
URI: | http://kups.ub.uni-koeln.de/id/eprint/5882 |
Downloads
Downloads per month over past year
Export
Actions (login required)
View Item |