This is version 2.0 of MARHYS (MARine HYdrothermal Solutions) database. The database is a global compilation of marine hydrothermal vent fluid compositions. The database includes data presented in journal publications, book chapters and monographs and includes compositions of hydrothermal fluids, end members and background seawater. Detailed sample information (metadata) is provided along with the chemical composition, which enables the unique identification of discrete vent fluid samples. The database contains concentrations of major-, minor- and trace elements, as well as, reduced carbon compounds and isotope compositions. All concentrations are converted into units per kg of vent fluid and chemical data of unique samples reported in different publications are merged into unique sample entries.
In brief, the following changes have been applied in the course of compilation of MARHYS database version 2.0 compared to version 1.0: A large number of new vent sites from the following regions is now incorporated in the database: southern Mid Atlantic Ridge, Arctic Mid Ocean Ridge, Juan de Fuca Ridge, Okinawa Trough, Lau Basin, Manus Basin, and some more (The database has grown about nearly 50% and now contains 3394 merged entries, compared to 2374 merged entries in Version 1.0). The complete sample information section for a larger number of entries was overhauled to achieve more consistency between different datasets and within individual vent sites (additional Sample IDs have been adjusted to fit the database naming scheme and to ensure unique sample entries). Some new species were included. A small number of faulty concentrations were detected and corrected. A number of vent areas in the Manus Basin were erroneously classified in Version 1.0 and are now re-classified as back-arc spreading centres (Susu, Desmos, PACMANUS). The shape of the spreadsheet was improved by freezing the "Sample ID" column, as well as, the parameter and unit rows to provide users more comfort when browsing the database. The built-in Excel filter function is now turned on by default (row 14), enabling users to easier generate sub-datasets of interest.
