carbon14

The goal of carbon14 is to provide a tidy interface to radiocarbon dating, to promote its inclusion in reproducible manuscripts.

Installation

You can install carbon14 from github with:

# install.packages("devtools")
devtools::install_github("paleolimbot/carbon14")

Example

Radiocarbon ages from Long Lake, Nova Scotia, Canada:

#> # A tibble: 5 x 4
#>   sample_id       depth age_14C age_error
#>   <chr>           <dbl>   <dbl>     <dbl>
#> 1 LL082011-C2-39  94.0     9148      49.0
#> 2 UOC-0844        70.5     8582      28.0
#> 3 LL082011-C2-87  46.0     4396      55.0
#> 4 UOC-0845        37.5      575      18.0
#> 5 LL082011-C2-124  9.00     623      34.0

Calibrating dates using calibrate():

result <- dates %>%
  calibrate(
    measured_age = age_14C, 
    measured_age_error = age_error,
    name = sample_id
  )
result
#> # A tibble: 5 x 9
#>   name   measured_age measured_age_err…    df curve   cal_age   curve_name
#>   <chr>         <dbl>             <dbl> <dbl> <S3: a> <S3: cdi> <chr>     
#> 1 LL082…         9148              49.0   Inf <age_c… 10318.48… intcal13  
#> 2 UOC-0…         8582              28.0   Inf <age_c… 9539.570… intcal13  
#> 3 LL082…         4396              55.0   Inf <age_c… 5005.665… intcal13  
#> 4 UOC-0…          575              18.0   Inf <age_c… 591.8846… intcal13  
#> 5 LL082…          623              34.0   Inf <age_c… 604.1533… intcal13  
#> # ... with 2 more variables: measured_age_type <chr>, cal_age_type <chr>

The result is a tibble, which also has a plot method to examine the results:

plot(result)

And a summary of the ages can be generated using summary() on the cal_age column of the output:

summary(result$cal_age)
#> # A tibble: 5 x 8
#>   .name       weighted_mean quantile_5 quantile_25 quantile_50 quantile_75
#>   <chr>               <dbl>      <dbl>       <dbl>       <dbl>       <dbl>
#> 1 LL082011-C…         10318      10237       10263       10305       10362
#> 2 UOC-0844             9540       9528        9535        9539        9543
#> 3 LL082011-C…          5006       4870        4917        4975        5040
#> 4 UOC-0845              592        544         553         607         618
#> 5 LL082011-C…           604        555         575         599         635
#> # ... with 2 more variables: quantile_95 <dbl>, dist <chr>

Calibration Curves

The default calibration curve for calibrate() is Intcal13, but others are also incluced in this package (see ?intcal13). The curve can be changed using the curve argument, which doesn't have to be the same for each element (it can be a list of actual curves, or a list of object names). To read in another curve, you can use the read_14c() function, which will read a .14c file from a path or URL.

read_14c("http://www.radiocarbon.org/IntCal13%20files/intcal13.14c")
#> <age_calibration_curve: 'intcal13'>
#> cal_age = cal_bp, measured_age = age_14C, measured_age_error = error
#> Radiocarbon Years BP => Calibrated BP
#> 
#> # A tibble: 5,141 x 5
#>    cal_bp age_14C error delta_14C sigma
#>     <dbl>   <dbl> <dbl>     <dbl> <dbl>
#>  1  50000   46401   274       313  44.8
#>  2  49980   46381   273       313  44.6
#>  3  49960   46362   271       313  44.3
#>  4  49940   46343   270       313  44.1
#>  5  49920   46324   269       313  44.0
#>  6  49900   46305   268       313  43.8
#>  7  49880   46286   267       312  43.6
#>  8  49860   46267   267       312  43.6
#>  9  49840   46249   266       312  43.5
#> 10  49820   46230   266       312  43.4
#> # ... with 5,131 more rows

Finally there is the null_calibration_curve(), which keeps dates as they are (e.g., pre-calibrated dates), and a calibration curve of your own design using calibration_curve().

custom_curve <- calibration_curve(
  cal_age = 0:10000,
  measured_age = cal_age + 50 * sin(cal_age / 20)
)

curve_result <- calibrate(
  measured_age = 623, 
  measured_age_error = 34,
  curve = list(
    intcal04, intcal09, intcal13,
    shcal13, custom_curve, NULL
  )
)

plot(curve_result, ylim = c(450, 750), xlim = c(450, 750))