Curating and sharing XAS data – Metadata and Scientific quality control

With the growing volume of measured X-ray Absorption Spectroscopy (XAS) data and their need for machine-learning and reference purposes across different facilities, both spectral quality and documentation of the XAS measurements need to be standardized. The documentation is also important to avoid unnecessarily repeating reference measurements and therefore wasting precious beamtime. In this article, we have discussed the classification of quality control with respect to meta data and scientific quality, important for standard documentation and curation of the XAS data. As an example, we take the use case of the RefXAS database developed under the DAPHNE4NFDI project. Considering that the database is under development, the initial requirement for an XAS database is a comprehensive set of metadata fields, that enhances the interpretation of XAS spectra; thereby improving its reusability and the reproducibility. The metadata schema should be able to provide details about the sample, optimized equipment and measurement conditions thereby making the process of acquiring the data clear. The next important component is the evaluation of the quality of the spectra as raw data and the metadata set to ensure the accuracy and reliability of the data stored in the database. Quality criteria need to be formulated for automated initial screening of any uploaded data set followed by manual curation involving utilization of details of metadata provided during upload and scientific quality of the XAS data. In this way, metadata and scientific quality control for XAS data uploaded at a database will provide a well-defined protocol for curation of the data and strengthen its distribution under FAIR data principles.