Category talk:Marine Sediment Working Group
OK, I hope I'm doing this right. Somehow I find the wiki thing not very intuitive and I'm too lazy to find out how it really works...
Here's what I use for my own project; I don't distinguish between stable isotopes, Mg/Ca and trace elements when it comes to metadata. Since I'm working with legacy datasets there are only a few things that I consider essential (indicated with (E)), but I think we should push to make the remaining metadata fields at least recommended:
Parameter.type (E): measured or inferred
Parameter.analytical.error: based on repeat measurement of standard, 1SD
Parameter.reproducibility: based on repeat measurement of sample, 1SD
Species (E): in case of foraminifera
Nshells.measured: number of shells used
Size.fraction: size fraction of foraminifera shells
Habitat.Season: from paper
Habitat.Depth: from paper. add epifaunal, infaunal for benthics
Equilibrium.offset: from paper
Calibration.eq: for inferred parameters
Publication.DOI: publication(s) where data were reported
Data.source (E): to allow tracing of legacy data, either DOI, link or ...
For trace elements cleaning method would be a good one to include.
-- Andreasschmittner (talk) 13:43, 26 October 2017 (PDT)
I wonder if we need a standard for core names.
With regard to depths reporting I would argue that having depth_top and depth_bottom precisely defines the interval and if only the mean depth is known one can indicate that by having the same value in both depth_top and depth_bottom.