README.md

In bentrueman/cwrshelpr: Functions for CWRS Data Analysis

cwrshelpr

cwrshelpr provides a collection of functions for data analysis tasks at Dalhousie University’s Centre for Water Resources Studies.

Installation

You can install the development version from GitHub as follows. If the remotes package is not installed, run install.packages("remotes") first. If you’re new to R, R for data science provides an excellent introduction.

remotes::install_github("bentrueman/cwrshelpr")

The examples below rely on the tidyverse family of packages. Run install.packages("tidyverse") if it’s not already installed.

ICP-MS

To read and clean ICP-MS data files generated by the Thermo Scientific iCAP-RQ, use the read_icp() function. (N.B., the code below uses external raw data files as examples; replace them with your data.)

library("tidyverse")
library("cwrshelpr")
# example path to example raw data file
files <- list.files(
  # replace this line with the path to your data:
   path = system.file("extdata", package = "cwrshelpr"), 
   full.names = TRUE,
   pattern = ".+\\.xlsx"
)
read_icp(path = files[1]) # read and clean one file
#> # A tibble: 1,440 × 6
#>    sample_name estimate_type         isotope element    value unit 
#>    <chr>       <chr>                 <chr>   <chr>      <dbl> <chr>
#>  1 Wash        Concentration average 27      Al       0.115   µg/l 
#>  2 Wash        Concentration average 31      P       -0.0408  µg/l 
#>  3 Wash        Concentration average 45      Sc      99.6     %    
#>  4 Wash        Concentration average 55      Mn      -0.00621 µg/l 
#>  5 Wash        Concentration average 56      Fe       0.0124  µg/l 
#>  6 Wash        Concentration average 65      Cu       0.0396  µg/l 
#>  7 Wash        Concentration average 115     In      99.2     %    
#>  8 Wash        Concentration average 159     Tb      98.1     %    
#>  9 Wash        Concentration average 208     Pb       0.0497  µg/l 
#> 10 Wash        Concentration per Run 27      Al       0.133   µg/l 
#> # … with 1,430 more rows
files[1:2] %>% 
  set_names() %>% 
  map_dfr(read_icp, .id = "file") # read and clean multiple files
#> # A tibble: 2,439 × 7
#>    file                           sampl…¹ estim…² isotope element    value unit 
#>    <chr>                          <chr>   <chr>   <chr>   <chr>      <dbl> <chr>
#>  1 /private/var/folders/fy/v4w9p… Wash    Concen… 27      Al       0.115   µg/l 
#>  2 /private/var/folders/fy/v4w9p… Wash    Concen… 31      P       -0.0408  µg/l 
#>  3 /private/var/folders/fy/v4w9p… Wash    Concen… 45      Sc      99.6     %    
#>  4 /private/var/folders/fy/v4w9p… Wash    Concen… 55      Mn      -0.00621 µg/l 
#>  5 /private/var/folders/fy/v4w9p… Wash    Concen… 56      Fe       0.0124  µg/l 
#>  6 /private/var/folders/fy/v4w9p… Wash    Concen… 65      Cu       0.0396  µg/l 
#>  7 /private/var/folders/fy/v4w9p… Wash    Concen… 115     In      99.2     %    
#>  8 /private/var/folders/fy/v4w9p… Wash    Concen… 159     Tb      98.1     %    
#>  9 /private/var/folders/fy/v4w9p… Wash    Concen… 208     Pb       0.0497  µg/l 
#> 10 /private/var/folders/fy/v4w9p… Wash    Concen… 27      Al       0.133   µg/l 
#> # … with 2,429 more rows, and abbreviated variable names ¹​sample_name,
#> #   ²​estimate_type

Use the read_lims() function to read/clean files in the new LIMS format:

read_lims(files[3], work_order = "00A123456")
#> # A tibble: 12 × 7
#>    param                             unit  rdl     value date       sample bdl  
#>    <chr>                             <chr> <chr>   <dbl> <date>     <chr>  <lgl>
#>  1 Total Lithium Solid Digestion ug… ug/L  0.4   8.54e+2 2022-05-11 SAMPL… FALSE
#>  2 Total Lithium Solid Digestion ug… ug/L  0.4   1.01e+3 2022-05-11 SAMPL… FALSE
#>  3 Total Lithium Solid Digestion ug… ug/L  0.4   8.14e+2 2022-05-11 SAMPL… FALSE
#>  4 Total Lithium Solid Digestion ug… ug/L  0.4   4   e-1 2022-05-11 SAMPL… TRUE 
#>  5 Total Sodium Solid Digestion mg/L mg/L  0.1   3.44e+0 2022-05-11 SAMPL… FALSE
#>  6 Total Sodium Solid Digestion mg/L mg/L  0.1   4.28e+0 2022-05-11 SAMPL… FALSE
#>  7 Total Sodium Solid Digestion mg/L mg/L  0.1   3.52e+0 2022-05-11 SAMPL… FALSE
#>  8 Total Sodium Solid Digestion mg/L mg/L  0.1   2.78e-1 2022-05-11 SAMPL… FALSE
#>  9 Total Magnesium Solid Digestion … mg/L  0.1   3.64e+1 2022-05-11 SAMPL… FALSE
#> 10 Total Magnesium Solid Digestion … mg/L  0.1   4.37e+1 2022-05-11 SAMPL… FALSE
#> 11 Total Magnesium Solid Digestion … mg/L  0.1   5.36e+1 2022-05-11 SAMPL… FALSE
#> 12 Total Magnesium Solid Digestion … mg/L  0.1   1   e-1 2022-05-11 SAMPL… TRUE

Fluorescence excitation-emission matrices (FEEM)

To read and clean FEEM data files generated by the Horiba Aqualog, use the read_feem() function. The argument truncate = TRUE may be useful if the corrected first order Rayleigh scattering line includes negative intensities.

files <- list.files(
   path = system.file("extdata", package = "cwrshelpr"), # replace this line with the path to your data
   full.names = TRUE,
   pattern = ".+\\.csv"
)
read_feem(path = files[1]) # read and clean one file
#> # A tibble: 15,125 × 4
#>    emission em_regular excitation intensity
#>       <dbl>      <dbl>      <dbl>     <dbl>
#>  1     212.       212.        600         0
#>  2     212.       212.        597         0
#>  3     212.       212.        594         0
#>  4     212.       212.        591         0
#>  5     212.       212.        588         0
#>  6     212.       212.        585         0
#>  7     212.       212.        582         0
#>  8     212.       212.        579         0
#>  9     212.       212.        576         0
#> 10     212.       212.        573         0
#> # … with 15,115 more rows
feem_dat <- files %>% 
  set_names() %>% 
  map_dfr(~ read_feem(.x, truncate = TRUE), .id = "file") # read and clean multiple files

Plot FEEM data using the function ggplot2::geom_raster() (among others). Use the column em_regular to avoid horizontal striping due to irregular spacing of the emission wavelengths.

feem_dat %>% 
  ggplot(aes(excitation, em_regular, fill = intensity)) +
  facet_wrap(vars(file)) + 
  geom_raster() + 
  scale_fill_viridis_c()

Here is an example using ggplot2::stat_contour(geom = "polygon"):

feem_dat %>% 
  ggplot(aes(excitation, em_regular, z = intensity, fill = stat(level))) +
  facet_wrap(vars(file)) + 
  stat_contour(geom = "polygon") +
  scale_fill_viridis_c("intensity")
#> Warning: `stat(level)` was deprecated in ggplot2 3.4.0.
#> ℹ Please use `after_stat(level)` instead.

Integrate regions of the FEEM using integrate_regions(). This function uses the regions defined in Chen et al. (2003) by default, but you can supply your own as well.

feem_dat %>% 
  group_by(file) %>% 
  nest() %>% 
  ungroup() %>% 
  mutate(regions = map(data, integrate_regions)) %>% 
  unnest(regions)
#> # A tibble: 12 × 8
#>    file                       data     name  ex_min ex_max em_min em_max integ…¹
#>    <chr>                      <list>   <chr>  <dbl>  <dbl>  <dbl>  <dbl>   <dbl>
#>  1 /private/var/folders/fy/v… <tibble> regi…    200    250    250    330    46.6
#>  2 /private/var/folders/fy/v… <tibble> regi…    200    250    250    380   247. 
#>  3 /private/var/folders/fy/v… <tibble> regi…    200    250    250    550  2421. 
#>  4 /private/var/folders/fy/v… <tibble> regi…    250    340    340    380   745. 
#>  5 /private/var/folders/fy/v… <tibble> regi…    250    400    400    550 13312. 
#>  6 /private/var/folders/fy/v… <tibble> total     NA     NA     NA     NA 21597. 
#>  7 /private/var/folders/fy/v… <tibble> regi…    200    250    250    330    46.1
#>  8 /private/var/folders/fy/v… <tibble> regi…    200    250    250    380   261. 
#>  9 /private/var/folders/fy/v… <tibble> regi…    200    250    250    550  2687. 
#> 10 /private/var/folders/fy/v… <tibble> regi…    250    340    340    380   827. 
#> 11 /private/var/folders/fy/v… <tibble> regi…    250    400    400    550 14921. 
#> 12 /private/var/folders/fy/v… <tibble> total     NA     NA     NA     NA 24147. 
#> # … with abbreviated variable name ¹​integrated

The humification and biological indices of each FEEM, as described in Tedetti et al. (2011), can be calculated using calculate_indices():

feem_dat %>% 
  group_by(file) %>% 
  nest() %>% 
  ungroup() %>% 
  mutate(ix = map(data, calculate_indices)) %>% 
  unnest(ix)
#> # A tibble: 4 × 4
#>   file                                                     data     param  value
#>   <chr>                                                    <list>   <chr>  <dbl>
#> 1 /private/var/folders/fy/v4w9p72s7c996w8l8qfthxq40000gn/… <tibble> bix    0.517
#> 2 /private/var/folders/fy/v4w9p72s7c996w8l8qfthxq40000gn/… <tibble> hix   14.6  
#> 3 /private/var/folders/fy/v4w9p72s7c996w8l8qfthxq40000gn/… <tibble> bix    0.508
#> 4 /private/var/folders/fy/v4w9p72s7c996w8l8qfthxq40000gn/… <tibble> hix   13.6

Additional resources

Field flow fractionation data

Read, clean, and analyze field flow fractionation data using fffprocessr, available here.

Coagulation modeling

An implementation of the Edwards (1997) model in R is available here, or via install.packages("edwards97").

Geochemical modeling

Dewey Dunnington’s tidyphreeqc (available here) provides an interface to PHREEQC in R, and pbcusol is designed specifically for lead and copper solubility modeling (available here).

References

Chen, W., Westerhoff, P., Leenheer, J. A., & Booksh, K. (2003). Fluorescence excitation−emission matrix regional integration to quantify spectra for dissolved organic matter. Environmental science & technology, 37(24), 5701-5710.

Edwards, M. (1997). Predicting DOC removal during enhanced coagulation. Journal American Water Works Association, 89(5), 78-89.

Tedetti, M., Cuet, P., Guigue, C., & Goutx, M. (2011). Characterization of dissolved organic matter in a coral reef ecosystem subjected to anthropogenic pressures (La Réunion Island, Indian Ocean) using multi-dimensional fluorescence spectroscopy. Science of the total environment, 409(11), 2198-2210.

bentrueman/cwrshelpr documentation built on July 1, 2023, 4:29 a.m.