Facilitating Future Reuse of Magnetotelluric (MT) Data: Moving to an Online Dirt-to-Desktop (D2D) Data Path to Standardise Data Collection, Curation and Publication.
Abstract
New demands for data cited in publications to comply with the FAIR principles and be stored in trusted repositories over the longer term is proving to be quite a new challenge for geoscience researchers. Currently magnetotelluric (MT) geophysicists use multiple disconnected systems to acquire, store and process data and metadata starting from the recording of the field acquisition parameters, capturing and then processing their primary data and finally curating and storing in a trusted data repository to enable FAIR publication. The Incorporated Research Institutions for Seismology (IRIS) has developed the Dirt-To-Desktop (D2D) concept that highlights the need to unify the procedures, infrastructure, protocols and standards to move data and metadata from the sensor in the "dirt" and then to make available to a user's "desktop", with as little human intervention as possible.
In Australia, there is no single agreed mechanism for capturing MT data or designing a consistent D2D processing pipeline that informs how new MT data are acquired, curated, stored and processed by the MT academic research community, particularly those using publicly funded research infrastructures (e.g. instruments, data storage, HPC computing). The Geoscience Data Enhanced Virtual Laboratories (GeoDeVL) project is working towards stabilising and increasing the consistency and quality of MT D2D publishing pipelines that start with the capture of data in the field through to processing, curation and publication. The processing path from raw time series data to the MT transfer functions to the final models is quite complex and in order to provide provenance tracking, all data in this process will need to have persistent identifiers to help enable transparent data management and support reusability and reproducibility of workflows. One specific gap is that there are no internationally agreed vocabularies for MT data and metadata, which is hindering the development of QA/QC protocols that would ensure seamless, programmatic access to the data. Hence, any version of vocabularies or standards used to currently describe the data will also need to be recorded to ensure that any MT data survey, regardless of the year it was collected, remains coherent and enduring to enable effective long term use by the research, government and industry communities.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMIN24A..08R
- Keywords:
-
- 1912 Data management;
- preservation;
- rescue;
- INFORMATICS;
- 1930 Data and information governance;
- INFORMATICS;
- 1982 Standards;
- INFORMATICS;
- 1998 Workflow;
- INFORMATICS