Skip to contents

Add weight-at-length estimates to a datras_raw object

Source: R/weight.R
add_weight_at_length.Rd

Estimate catch weight from length data and add the resulting weight fields to a datras_raw / DATRASraw object.

Usage

add_weight_at_length(
  x,
  max_length = NULL,
  max_weight = NULL,
  plus_group = FALSE,
  lw_source = c("lookup", "ca"),
  lw_pars = NULL,
  lookup_as_backup = FALSE,
  verbose = TRUE
)

Arguments

x

A datras_raw object.

max_length

Optional numeric value giving the maximum length in centimetres to retain when fitting the length-weight relationship. Observations above this value are excluded.

max_weight

Optional numeric value giving the maximum individual weight in grams to retain when fitting the length-weight relationship. Observations above this value are excluded.

plus_group

Logical. If TRUE the mid length of the last length bin is not taken halfway to its upper limit, which might be Inf or arbitrarily high and result in an unrealistically high weight for that length bin. Instead the lower limit plus half of the size of the second last length bin is used. This is also done when the upper limit is Inf.

lw_source

Character string specifying the fallback source of length-weight parameters when lw_pars is NULL or does not cover a species. One of "lookup" (default, uses species_info) or "ca" (fits a log-log model to CA).

lw_pars

Optional custom length-weight parameters. Overrides lw_source for the covered species. Accepted forms:

  • A named numeric vector: c(a = 0.01, b = 3.0).

  • A named list: list(a = 0.01, b = 3.0).

  • A data frame with columns a and b (one row per species when multiple species are present), and optionally a column Valid_Aphia (or aphia) to identify which species each row applies to.

When no species identifier column is present, the data must contain exactly one species; an error is raised otherwise. When a Valid_Aphia column is present, parameters are matched by species, and lw_source is used as a fallback for any species not covered.

lookup_as_backup

Logical. If TRUE and lw_source = "ca", the species_info lookup table is used when CA data are insufficient. Default: FALSE.

verbose

Logical. If TRUE (default), progress messages are printed.

Value

A datras_raw object with estimated weight information added.

Details

The function derives weight-at-length from custom parameters supplied via lw_pars, from a log-log model fitted to the CA table (lw_source = "ca"), or from the built-in species_info lookup table (lw_source = "lookup"). Custom parameters in lw_pars take precedence over lw_source for the covered species.

Weight at length is calculated as \(W = a \times L^b\), where \(L\) is the mid-length of each length bin defined by attr(x, "cm.breaks"), e.g. 7.5 cm for the bin [7,8).

When lw_source = "ca", a linear model \(\log(W) = \alpha + b \log(L)\) is fitted to positive individual weights in CA, and the resulting parameters are used. Here \(L\) is the mid length of the measuring class of each record, i.e. LngtCm (the lower limit of the class) plus half the class width given by LngtCode, so that fitted and predicted lengths are on the same scale.

When lw_source = "lookup", parameters a and b are taken from the built-in species_info table, matched by Valid_Aphia.

See also

check_weights(), add_total_weight_by_haul()

Examples


## Add numbers at length
dab <- add_numbers_at_length(dab)
#> Warning: Mixed accuracies found in var[[3]]$LngtCode - worst chosen: 1 cm
#> Warning: NAs found in var[[3]]$LngtCode - assumed to be 1 cm

## Fitted weight-at-length from CA data
x <- add_weight_at_length(dab, lw_source = "ca")

## Lookup weight-at-length from species_info
x <- add_weight_at_length(dab, lw_source = "lookup")

## Custom parameters (single species) - named vector
x <- add_weight_at_length(dab, lw_pars = c(a = 0.00832, b = 3.09))

## Custom parameters (single species) - data frame
x <- add_weight_at_length(dab, lw_pars = data.frame(a = 0.00832, b = 3.09))

if (FALSE) { # \dontrun{
## Custom parameters for multiple species
pars <- data.frame(
  Valid_Aphia = c(127139L, 126436L),
  a = c(0.00832, 0.0045),
  b = c(3.09, 3.12)
)
x <- add_weight_at_length(x_multi, lw_pars = pars)
} # }