Suggestions and to dos from the workshop at CAA2026

[archaeothommy]

• pre-made bdl strategy to set negative values to NA (might occur from using standard curves in LA or pXRF)
• as_ASTR: write documentation/vignette clearer that it is not needed when importing with read_ASTR()
• copper classifications: handle NA: replace with 0 while issuing a warning
• lead isotope age models: drop columns with all ratios being NA (and maybe more fine-grained handling of single NA values)