There are a number (will work to append a list to this issue) of datasets that appear to be composite d18O, where the mean has been removed. This causes problems when looking at large scale patterns in mean d18O, but the composite has it's own utility. Should we flag these records somehow? @Mattriks - if you have a list of sites where you think this is relevant would add that to this issue?
nickmckay
commented
5 years ago
There are a number (will work to append a list to this issue) of datasets that appear to be composite d18O, where the mean has been removed. This causes problems when looking at large scale patterns in mean d18O, but the composite has it's own utility. Should we flag these records somehow? @Mattriks - if you have a list of sites where you think this is relevant would add that to this issue?