Get started

3.1. Format your CSV file

You’ll need to add some special columns to your data file before loading it into R. The easiest way to do this is to use a spreadsheet editor like Excel. We recommend saving a seperate copy of your raw data file for mimsy (add “_mimsy" to the file name) to prevent any accidents.

Figure 1. An example of a correctly formatted raw data file.

CSV file format:

• You can save as CSV within Excel from the File > Save as menu, choosing the CSV (Comma delimited) (.csv) option.

Columns:

• Type
  • If the row contains data for a standard, enter “Standard”. If the row contains data for a sample, enter “Sample”.
  • The name of this column must be “Type” (check for correct capitalization)
    
• Group
  • Each block of standards and the samples run directly afterwards belong to a group. You’ll enter “1” for the first block of standards and the first block of samples. You’ll enter “2” for the second block of standards and the second block of samples, etc.
    
  • The name of this column must be “Group” (check for correct capitalization)
    
• CollectionTemp
  • The temperature of samples or standards at the time of field collection, in degrees C
  • The name of this column must be “CollectionTemp” (check for correct capitalization)
    
• RunDate
  • The date (M/D/YYYY) that samples were run on the MIMS.
    
  • The name of this column must be “RunDate” (check for correct capitalization)
• Label or other sample identifier columns
  • mimsy will preserve all labelling columns in the final output, so you should add whatever labels or notes columns you like
• Index, Time, 28, 32, 40, 99, N2/Ar, O2/Ar columns
  • This is the default output from the MIMS

3.2. Load your CSV file into R using read.csv()

# Load data into R
data <- read.csv(file = "data.csv", header = TRUE, stringsAsFactors = FALSE)

# Check out the structure
str(data, vec.len = 2)
#> 'data.frame':    42 obs. of  14 variables:
#>  $ Type          : chr  "Standard" "Standard" ...
#>  $ Group         : int  1 1 1 1 1 ...
#>  $ CollectionTemp: num  24.9 24.9 24.9 26.3 26.3 ...
#>  $ RunDate       : chr  "2/1/2018" "2/1/2018" ...
#>  $ Label         : chr  NA NA ...
#>  $ Index         : int  3029 3079 3128 3212 3261 ...
#>  $ Time          : chr  "10:32:53 AM" "10:34:07 AM" ...
#>  $ X28           : num  1.20e-08 1.20e-08 1.20e-08 1.19e-08 1.18e-08 ...
#>  $ X32           : num  5.75e-09 5.75e-09 5.75e-09 5.66e-09 5.66e-09 ...
#>  $ X40           : num  2.98e-10 2.98e-10 2.98e-10 2.93e-10 2.93e-10 ...
#>  $ X99           : num  5.51e-07 5.49e-07 5.49e-07 5.41e-07 5.41e-07 ...
#>  $ N2.Ar         : num  40.3 40.3 ...
#>  $ O2.Ar         : num  19.3 19.3 ...
#>  $ Notes         : chr  "Barometric pressure =  977.2 hPa" "" ...

3.3. Run the mimsy() function

You must specify the barometric pressure (as baromet.press) and its units in the function argument. Units must be one of "atm", "hPa", "psi", "bar", or "Torr". All other inputs, such as background corrections or standard salinity, are optional. Check out ?mimsy for more information.

# Run the function
results <- mimsy(data, baromet.press = 977.2, units = "hPa")
#> Calculated dissolved concentrations based on a two-point temperature standard.
#> Standard 1: 24.9 C, Standard 2: 26.3 C

3.4. Explore the results

You’ll see that mimsy() returns a list containing five seperate dataframes (results, solubility.Concentrations, calibration.Factors, calibration.DriftCorrection, and results.full). Check out ?mimsy() for more specific information on those outputs and how they were calculated.

# Check out the structure of the output
summary(results)
#>                             Length Class      Mode
#> results                     11     grouped_df list
#> solubility.Concentrations    3     data.frame list
#> calibration.Factors          3     data.frame list
#> calibration.DriftCorrection 12     data.frame list
#> results.full                30     grouped_df list

# See the summarized results dataframe
str(results$results, give.attr = FALSE)
#> Classes 'grouped_df', 'tbl_df', 'tbl' and 'data.frame':  24 obs. of  11 variables:
#>  $ CollectionTemp: num  25.1 25.1 25.1 26.1 26.1 26.1 25.3 25.3 25.3 24.9 ...
#>  $ Label         : chr  "Sample1" "Sample2" "Sample3" "Sample4" ...
#>  $ Notes         : chr  "" "" "" "" ...
#>  $ data$Type     : chr  "Sample" "Sample" "Sample" "Sample" ...
#>  $ data$Group    : int  1 1 1 1 1 1 1 1 1 1 ...
#>  $ N2_uMol       : num  469 469 465 443 447 ...
#>  $ O2_uMol       : num  301 301 300 290 293 ...
#>  $ Ar_uMol       : num  12 12 11.9 11.6 11.6 ...
#>  $ N2_mg         : num  13.1 13.1 13 12.4 12.5 ...
#>  $ O2_mg         : num  9.63 9.62 9.61 9.29 9.37 ...
#>  $ Ar_mg         : num  0.48 0.479 0.478 0.464 0.464 ...

3.5. Save the results

# Save output to an Excel workbook
mimsy.save(results, file = "results.xlsx")

# Save output to an RData file
save(results, file = "results.RData")

We don’t reccomend saving results dataframes to CSV files (although you can), as you’ll need multiple CSV’s to preserve all of the outputs, and that gets kind of messy. A good alternative is to save both an Excel workbook copy and an RData copy, that way all of your output is preserved every time.

You can load RData files back into R using load("results.RData"). Check out ?load() for more info.