Sharing water and sediment samples metadata: an example from the LTER-Italy marine and freshwater sites

Water and sediment samples are key observational units in environmental sciences, particularly in marine ecology and geosciences, hydrobiology and limnology, where they are essential to document and analyse many biological communities which need to be processed in the laboratory. Such samples must be maintained in the medium/long term and require to be easily discoverable. At the moment unique identification standards and metadata catalogues representing them are lacking, even though some targeted initiatives are already in place, such as: the Research Data Alliance (RDA) Interesting Group Physical Samples and Collections in the Research Data Ecosystem , the Earth Cube group , and the International symposium hosted by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) . We present here both best practices and technological solutions for representing and sharing sample descriptions, using as an example water and sediment samples from a number of marine and freshwater sites belonging to the Italian Long-Term Ecological Research network (LTER-Italy). In LTER-Italy observations and specimens are stored and shared, respectively, by ISO 19156:2011 (Observations and Measurements, O&M) and CSIRO physical samples metadata model identifying with International Geo Sample Number (IGSN) (Anusuriya et al., 2016 ). Information available for each physical sample (i.e., sediment or water) includes, among the others, the sample name, type, the sampling location and method, and the indication of the sub-samples collected, with the relative sampling methodologies. When the samples are analysed, for each of them a list of other information becomes available, such as species composition abundances, biomass and analytical methods applied for studying them. Moreover, since the samples are often collected for ecological or paleoecological studies, each sample is accompanied by environmental variables data, including measures obtained directly in the field, reported on "sampling sheets" and laboratory tables, partially organized in electronic format. We present here a proposal that provides a way for describing, archiving, visualising and sharing samples and allows connecting the different related entities, such as sampling methods, persons, instruments, and observations representations. A hyperlinking approach is tested as a valuable solution to manage the information (metadata), giving the possibility to move among them, adding also a further contribution for the maintenance of the data life cycle. Applying this approach makes it possible to refer each entity to the other, describing not only the relationships between them but also those among the different sub-entities generated from the same samples (e.g., from the sediment core to the slide with recognized foraminifera or from the water sample to the phytoplankton sample). 1 https://rd-alliance.org/ig-physical-samples-and-collections-research-data-ecosystem-rda-11th-plenary-meeting 2 https://earthcube.org/group/isamples 3 https://csiro-enviro-informatics.github.io/environmental-data-symposium-2017/ 4 https://doi.org/10.1016/j.cageo.2016.07.018