Retrieve per-cell ARA diagnostics from an add_mspaf() result

After calling add_mspaf() with a reference_model (or any reference), this accessor returns a tibble describing what happened in the ARA subtraction for every (sample × analyte) that was assessed. This is the primary tool for auditing the "reference higher than target" case (floored to zero) and for understanding which tier was used per cell.

Usage

ara_summary(x)

Arguments

x

A data frame returned by add_mspaf().

Value

A tibble with columns:

sample_id

Sample identifier.

analyte

Analyte name.

ref_norm

Normalised reference concentration subtracted.

C_norm

Normalised target concentration (before ARA).

C_adj

ARA-adjusted concentration (max(C_norm - ref_norm, 0)).

C_excess

Unfloored difference C_norm - ref_norm; negative values indicate the reference exceeded the target — possibly a geogenic artefact (e.g. low-pH upstream metal mobilisation).

floor_fired

Logical; TRUE when C_norm < ref_norm.

ref_source

"disabled", "matched", or "unmatched".

ref_tier

"matched", "model", "model_integrated", or "static". NA for non-temporal reference.

Returns NULL (with a message) if the attribute is absent.

Details

The attribute is stored by add_mspaf() and is dropped by most dplyr verbs, so read the summary before further wrangling.

See also

add_mspaf(), fit_reference_model()