Sensitivity analysis for DOC, depth, and time of year

Enter a single location and PAH, and a range of values for DOC or Kd_ref, water depth, and dates and get back a data.frame of narcotic benchmark, Pabs, and phototoxic benchmark values.

Usage

sens_kd_depth(
  pah = NULL,
  lat = NULL,
  lon = NULL,
  elev_m = NULL,
  DOC = NULL,
  Kd_ref = NULL,
  depth_m = NULL,
  date = NULL,
  time_multiplier = 2,
  ...
)

Arguments

pah: The PAH of interest, which is used to calculate the narcotic benchmark value.
lat: latitude of the site, decimal degrees. Required.
lon: longitude of the site, decimal degrees. Required.
elev_m: elevation of the site above sea level, in metres. If NULL (default) it is looked up using a digital elevation model using lat and lon.
DOC: dissolved organic carbon concentration, in mg/L. Ignored if Kd_ref is set directly.
Kd_ref: Light attenuation coefficient at reference wavelength. Can be set directly, or calculated from DOC.
depth_m: depth at which to calculate the light attenuation coefficient. Required.
date: date of the calculation, as Date object, or a character in a standard format that can be converted to a Date object (e.g., "YYYY-MM-DD"). Required.
time_multiplier: If x is a tuv_results data frame, this is the multiplier to get the total exposure time. I.e., if the tuv_results contains 24 hours of data, and you need a 48h exposure, the multiplier would be 2 (this is the default). Ignored if x is a numeric value of light absorption.
...: other parameters passed on to the tuv model. See tuv()

Value

A data.frame of all of the input parameters, plus a list column containing TUV results and narcotic benchmark, Pabs, and phototoxic benchmark values

Details

You can add other variables beyond those listed explicitly, but if there are too many combinations it will create many runs of the TUV model, which can take a long time. Explicit variable checking is only performed on pah, lat, lon, elev_m, DOC, Kd_ref, depth, and months. Passing invalid values of other parameters may cause cryptic errors or unexpected results. Note that combinations of many DOC, Kd_ref, depth_m, and date values will result in many runs of the TUV model and thus take a long time.

Examples

sens_kd_depth(
  "Anthracene",
  lat = 52,
  lon = -113,
  Kd_ref = 40:45,
  depth_m = c(0.25, 0.5, 1),
  date = c("2023-07-01", "2023-08-01")
)
#> # A tibble: 36 × 11
#>      lat   lon elev_m depth_m date       Kd_ref tuv_res               pah       
#>    <dbl> <dbl>  <dbl>   <dbl> <date>      <int> <list>                <chr>     
#>  1    52  -113    880    0.25 2023-07-01     40 <tv_rslts [421 × 28]> anthracene
#>  2    52  -113    880    0.5  2023-07-01     40 <tv_rslts [421 × 28]> anthracene
#>  3    52  -113    880    1    2023-07-01     40 <tv_rslts [421 × 28]> anthracene
#>  4    52  -113    880    0.25 2023-08-01     40 <tv_rslts [421 × 28]> anthracene
#>  5    52  -113    880    0.5  2023-08-01     40 <tv_rslts [421 × 28]> anthracene
#>  6    52  -113    880    1    2023-08-01     40 <tv_rslts [421 × 28]> anthracene
#>  7    52  -113    880    0.25 2023-07-01     41 <tv_rslts [421 × 28]> anthracene
#>  8    52  -113    880    0.5  2023-07-01     41 <tv_rslts [421 × 28]> anthracene
#>  9    52  -113    880    1    2023-07-01     41 <tv_rslts [421 × 28]> anthracene
#> 10    52  -113    880    0.25 2023-08-01     41 <tv_rslts [421 × 28]> anthracene
#> # ℹ 26 more rows
#> # ℹ 3 more variables: narcotic_benchmark <dbl>, pabs <dbl>,
#> #   phototoxic_benchmark <dbl>