nwfscSurvey

Introduction

The nwfscSurvey package was written by scientists at the Northwest Fisheries Science Center (NWFSC) to explore and process survey composition data for use in groundfish stockassessments. The package can be used for the NWFSC West Coast Groundfish Bottom Trawl (WCGBT) survey, the NWFSC slope survey, the Alaska Fisheries Science Center (AFSC) slope survey, and the AFSC Triennial survey. The package includes functions that query the NWFSC data warehouse, visualize, and process the data for use in groundfish assessments.

Use

nwfscSurvey can be used to:

Pull survey catch, haul, and biological data from the NWFSC data warehouse.
Calculate the design based indices based upon pre-specified stratification.
Create length- and age-composition data expanded by the design-based indices formated for use in Stock Synthesis.
Create unexpanded conditional age-at-length composition data for use in Stock Synthesis.
Create plots to visualize the data: design-based index plots, bubble plots of length and age data, observeed sex ratio, and distributions across depth and latitude.

Functions

A list of all functions in the packages can be viewed by:

ls("package:nwfscSurvey")

Examples

NWFSC WCGBT Survey

Pull data

Pull both the catch and biological data:

catch = pull_catch(
  common_name = "Pacific ocean perch", 
  survey = "NWFSC.Combo")

bio = pull_bio(
  common_name = "Pacific ocean perch", 
  survey = "NWFSC.Combo")

Initial data visualization

There are a range of functions to create visualizations of the data by examining catch rates by latitude or depth, lengths or ages by latitude and depth, presence/absence in tows, and observed sex ratios.

plot_cpue(
  catch = catch)

plot_bio_patterns(
  bio = bio, 
  col_name = "Length_cm")

wh_plot_proportion(
  data_catch = catch,
  data_bio = bio
)

Define strata

Depth and latitude strata are required to calculate a design-based index of abundance as well as length and age compositions. The strata should be specific to the range where the species in question has been observed. Note that the WCGBT Survey sampling design has changes in sampling intensity at 183 and 549 meters and north and south of 34.5 degrees latitude, so these should generally be included as strata breaks if the range spans these values. Pacific ocean perch are rarely observed south of 34.5 degrees or deeper than 540 m, so there’s no need to include strata that have very few observations (the survey extends southward to about 32 degrees and as deep as 1280 m):


strata <- CreateStrataDF.fn(
  names = c("shallow_s", "deep_s", "shallow_n", "deep_n"), 
  depths.shallow = c(  55,  183,  55, 183),
  depths.deep    = c( 183,  400, 183, 400),
  lats.south     = c(34.5, 34.5,  42,  42),
  lats.north     = c(  42,   42,  49,  49)
)

Index of abundance

Calculate the design based index of abundance:

biomass = get_design_based(
  data = catch,  
  strata = strata)

get_design_based() returns a list with elements $biomass_by_strata and $biomass, where the second element is the design based index of abundance aggregated across strata. If the dir function input is specified, the function writes a csv file inside the dir input location to a “forSS3” folder.

Plot the coastwide design based index of abundance with uncertainty intervals:

plot_index(
  data = biomass,
  plot = 1)

Plot the design based index of abundance for each strata without uncertainty intervals:

plot_index(
  data = biomass,
  plot = 2)

Length composition data

get_expanded_comps() calculates and formats the length-composition data for Stock Synthesis:

length_comps <- get_expanded_comps(
    bio_data = bio,
    catch_data = catch,
    comp_bins = seq(10, 40, 2),
    strata = strata,
    comp_column_name = "length_cm",
    output = "full_expansion_ss3_format",
    two_sex_comps = TRUE,
    input_n_method = "stewart_hamel")

The above call will calculate the length frequencies for use in Stock Synthesis. This function returns a list of sexed and unsexed length composition data formatted for Stock Synthesis. If the dir function input is specified, it will write these dataframes to separate csv files inside the dir input location to a “forSS3” folder. In the above example the input sample size is calculated based on the Stewart and Hamel approach (e.g., unique samples calculated as a function of species type and tows).

To plot the length frequency data:

plot_comps(
  data = length_comps)

If the dir function input is specified, then a “plot” folder will be created in the directory location and a png of the plot will be saved.

There is also a function to create raw or unexpanded composition data that works for either length or age data.

raw_length_comps <- get_raw_comps(
    data = bio,
    comp_bins = seq(10, 40, 2),
    comp_column_name = "length_cm",
    two_sex_comps = TRUE)

This function returns a list of sexed and unsexed length composition data formatted for Stock Synthesis. The input sample size is set equal to the number of samples in the data frame.

Marginal age composition data

age_comps <- get_expanded_comps(
    bio_data = bio,
    catch_data = catch,
    comp_bins = 1:40,
    strata = strata,
    comp_column_name = "age",
    output = "full_expansion_ss3_format",
    two_sex_comps = TRUE,
    input_n_method = "stewart_hamel")

The above call will calculate the marginal age-composition data for the age data in a format for Stock Synthesis. This function returns a list of sexed and unsexed age composition data formatted for Stock Synthesis. If the dir function input is specified, it will write these dataframes to separate csv files as well. In the above example the input sample size is calculated based on the Stewart and Hamel approach (e.g., unique samples calculated as a function of species type and tows).

To plot the age frequency data:

plot_comps(
  data = age_comps)

If the dir function input is specified, then a “plot” folder will be created in the directory location and a png of the plot will be saved.

There is also a function to create raw or unexpanded composition data that works for either length or age data.

raw_age_comps <- get_raw_comps(
    data = bio,
    comp_bins = 1:40,
    comp_column_name = "age",
    two_sex_comps = TRUE)

This function returns a list of sexed and unsexed marginal age composition data formatted for Stock Synthesis. The input sample size is set equal to the number of samples in the data frame.

Conditional-age-at-length data

To calculate conditional-age-at-length data formatted for Stock Synthesis:

caal <- get_raw_caal(
  data = bio, 
  len_bins = seq(10, 40, 2),
  age_bins = 1:40)

Creates unexpanded conditional-age-at-length data for each sex with input sample sizes based on the observed number of fish in each length bin by year. It returns a dataframe with conditional-age-at-length data by sex (sex = 0, sex = 1, sex = 2). If the dir function input is specified, it will write the dataframe to a csv file.

Maps

To make a map showing the distribution of density in aggregate and by year:

PlotMap.fn(
  dat = catch)

Additional data visualization

There are a couple of additional plotting functions that are included in the package:PlotVarLengthAtAge.fn() and PlotSexRatio.fn().

Chantel Wetzel

2025-02-14