In blongworth/HybridGIS: Data, Functions, and Analysis for the NOSAMS Hybrid Gas Ion Source

knitr::opts_chunk$set(message = FALSE, echo = FALSE)

library(tidyverse)
library(here)
library(openxlsx)
library(amstools)
library(HybridGIS)

options(digits = 4)
options(scipen = 3)

Data reduction and analysis for carbonate samples run on June 11, 2021. Test run of smaller samples. All samples ~8mg carbonate. Includes C-1, C-2, and NOSAMS2.

Outliers are defined as points 1.5 * interquartile range above the 3rd quartile or below the 1st quartile.

Data reduction proceeds as follows:

1. Read raw file
2. Apply 13C correction to each run of a target
3. Flag outliers by interquartile distance. Outliers are flagged, but NOT removed from the analysis.
4. Determine per-target means, internal and external errors
5. Normalize targets using mean of standards and propagate errors
6. Apply a large blank correction using the mean of blanks and propagate errors.

raw results

df <- process_hgis_results(here("data/USAMS061121R.txt"))
write_csv(df, here("data_analysed/HGIS_2021-06-11_raw.csv"))
write.xlsx(df, here("data_analysed/HGIS_2021-06-11_raw.xlsx"))

Currents and ratios vs time.

df %>% 
  filter(pos > 2) %>% 
  plot_hgis_time(norm_ratio, sig_norm_ratio, outlier = !.$ok_calc) +
  labs(title = "Sample ratio vs. time",
       y = "Fraction modern")

df %>% 
  filter(pos > 2) %>% 
  plot_hgis_time(he12C * 1E6, outlier = !.$ok_calc) +
  labs(title = "Sample current vs time",
       x = "Time (s)",
       y = "Current (uA)")

df_sum <- sum_hgis_targets(df, remove_outliers = FALSE) %>% 
  norm_hgis() %>% 
  blank_cor_hgis()

write_csv(df_sum, here("data_analysed/HGIS_2021-06-11_results.csv"))
write.xlsx(df_sum, here("data_analysed/HGIS_2021-06-11_results.xlsx"))

Results

Fm and Fm error are blank corrected Fm and error, respectively.

df_sum %>% 
  filter(pos > 2) %>% 
  mutate(he12C = he12C * 1E6) %>% 
  select(Pos = pos, Name = sample_name, `Current (uA)` = he12C, `N runs` = n_runs, Fm = fm_corr, `Fm error` = sig_fm_corr) %>% 
  arrange(Name) %>% 
  knitr::kable()
  # gt::gt()

df_sum %>% 
  filter(pos > 2) %>% 
plot_hgis_summary()

Agreement of replicates

Compare results for samples with more than one replicate.

df_sum %>% 
  filter(pos > 2) %>% 
  compare_replicates() %>% 
  select(Name, `Mean current (uA)` = he12C_mean, 
         `SD of current` = he12C_sd, `Mean Fm` = fm_corr_mean, 
         `SD of Fm` = fm_corr_sd, `Mean sample error` = sig_fm_corr_mean) %>% 
  knitr::kable()

Agreement with consensus

Comparison to expected Fm for sample types with a consensus value.

cons <- df_sum %>% 
  filter(pos > 2,
         !is.na(fm_consensus)) %>% 
  select(sample_name, fm_consensus, fm_corr, sig_fm_corr) %>% 
  mutate(Fm_diff = fm_corr - fm_consensus,
         sigma = amstools::sigma(fm_corr, fm_consensus, sig_fm_corr)) %>% 
  arrange(sample_name) %>% 
  select(Name = sample_name, 
         #`Mean current (uA)` = he12C_mean, 
         #`SD of current` = he12C_sd, 
         `Consensus Fm` = fm_consensus,
         Fm = fm_corr, 
         `Fm error` = sig_fm_corr,
         `Difference` = Fm_diff,
         Sigma = sigma)

knitr::kable(cons)

blongworth/HybridGIS documentation built on Dec. 19, 2021, 10:41 a.m.