packages <- c("qs", "dplyr", "ggplot2", "sf", "lubridate", "here", "ows4R", "patchwork", "kableExtra")
lapply(packages, function(pkg) {
  if (!require(pkg, character.only = TRUE)) install.packages(pkg)
  library(pkg, character.only = TRUE)
})

📥 Download and load the file

From qs (if you want to download geometry as well, it can be heavy)

qs_file <- "global_catch_tunaatlasird_level2.qs"
if (!file.exists(qs_file)) {
  download.file("https://zenodo.org/record/15221705/files/global_catch_tunaatlasird_level2.qs?download=1", qs_file, mode = "wb")
}
catch_data <- qs::qread(qs_file)

catch_data$geom_wkt <- NULL

From csv (geometry can be downloaded from github, downloading is slower and the dataset is close to 1GB).

csv_file <- "global_catch_tunaatlasird_level2 (without_geom).csv"
if (!file.exists(csv_file)) {
  download.file("https://zenodo.org/records/15221705/files/global_catch_tunaatlasird_level2%20(without_geom).csv?download=1", csv_file, mode = "wb")
}
catch_data <- readr::read_csv(csv_file) %>% dplyr::mutate(geographic_identifier = as.character(geographic_identifier))

First lines

kableExtra::kable(head(catch_data))

source_authority	fishing_fleet	time_start	time_end	geographic_identifier	gear_type	species	fishing_mode	measurement_unit	measurement_value	GRIDTYPE	species_group	gear_label	fishing_fleet_label	species_label	species_definition	ocean	fishing_mode_label
CCSBT	AUS	1975-12-01	1975-12-31	5237150	01.1	SBF	UNK	Tons	59.8000	1deg_x_1deg	SCOMBROIDEI	Purse seines	Australia	Southern bluefin tuna	Thunnus maccoyii	WCPFC	Undefined school
CCSBT	AUS	1976-09-01	1976-09-30	5237150	09.1	SBF	UNK	Tons	36.2000	1deg_x_1deg	SCOMBROIDEI	Handlines and hand-operated pole-and-lines	Australia	Southern bluefin tuna	Thunnus maccoyii	WCPFC	Undefined school
CCSBT	AUS	1976-10-01	1976-10-31	5237150	09.1	SBF	UNK	Tons	41.0000	1deg_x_1deg	SCOMBROIDEI	Handlines and hand-operated pole-and-lines	Australia	Southern bluefin tuna	Thunnus maccoyii	WCPFC	Undefined school
CCSBT	AUS	1976-12-01	1976-12-31	5232132	09.1	SBF	UNK	Tons	2109.5199	1deg_x_1deg	SCOMBROIDEI	Handlines and hand-operated pole-and-lines	Australia	Southern bluefin tuna	Thunnus maccoyii	IOTC	Undefined school
CCSBT	AUS	1976-12-01	1976-12-31	5233134	09.1	SBF	UNK	Tons	448.3517	1deg_x_1deg	SCOMBROIDEI	Handlines and hand-operated pole-and-lines	Australia	Southern bluefin tuna	Thunnus maccoyii	IOTC	Undefined school
CCSBT	AUS	1976-12-01	1976-12-31	5234134	09.1	SBF	UNK	Tons	711.5694	1deg_x_1deg	SCOMBROIDEI	Handlines and hand-operated pole-and-lines	Australia	Southern bluefin tuna	Thunnus maccoyii	IOTC	Undefined school

As the geometry can be heavy, we propose to remove it and to add it when necessary in plots.

📊 Data overview

This dataset provides monthly records of tuna and tuna-like species catches at the global scale between 1950 and 2021. Here is a description of the columns:

source_authority: Regional Fisheries Management Organization (RFMO) that provided the data
fishing_fleet: Code of the flag state or fleet (e.g., AUS, JPN)
time_start, time_end: Start and end dates of the observation period (monthly granularity, sometimes annualy)
geographic_identifier: Code identifying the spatial grid cell, from CWP grid dataset.
gear_type: Code of the fishing gear used
species: Species code (e.g., SBF for Southern Bluefin Tuna)
fishing_mode: Code indicating the fishing mode
measurement_unit: Unit used to report catches (e.g., Tons, Number)
measurement_value: Numeric value of the measured catch
GRIDTYPE: Spatial grid type (e.g., 1deg_x_1deg)
species_group: Taxonomic group (e.g., SCOMBROIDEI)
gear_label: Human-readable gear name
fishing_fleet_label: Human-readable label for the flag or fleet
species_label: Common name of the species
species_definition: Scientific name of the species
geom_wkt: Geometry in WKT format (MULTIPOLYGON)
ocean: RFMO area of activity

dplyr::glimpse(catch_data)

## Rows: 5,035,119
## Columns: 18
## $ source_authority      <chr> "CCSBT", "CCSBT", "CCSBT", "CCSBT", "CCSBT", "CC…
## $ fishing_fleet         <chr> "AUS", "AUS", "AUS", "AUS", "AUS", "AUS", "AUS",…
## $ time_start            <date> 1975-12-01, 1976-09-01, 1976-10-01, 1976-12-01,…
## $ time_end              <date> 1975-12-31, 1976-09-30, 1976-10-31, 1976-12-31,…
## $ geographic_identifier <chr> "5237150", "5237150", "5237150", "5232132", "523…
## $ gear_type             <chr> "01.1", "09.1", "09.1", "09.1", "09.1", "09.1", …
## $ species               <chr> "SBF", "SBF", "SBF", "SBF", "SBF", "SBF", "SBF",…
## $ fishing_mode          <chr> "UNK", "UNK", "UNK", "UNK", "UNK", "UNK", "UNK",…
## $ measurement_unit      <chr> "Tons", "Tons", "Tons", "Tons", "Tons", "Tons", …
## $ measurement_value     <dbl> 59.80000, 36.20000, 41.00000, 2109.51994, 448.35…
## $ GRIDTYPE              <chr> "1deg_x_1deg", "1deg_x_1deg", "1deg_x_1deg", "1d…
## $ species_group         <chr> "SCOMBROIDEI", "SCOMBROIDEI", "SCOMBROIDEI", "SC…
## $ gear_label            <chr> "Purse seines", "Handlines and hand-operated pol…
## $ fishing_fleet_label   <chr> "Australia", "Australia", "Australia", "Australi…
## $ species_label         <chr> "Southern bluefin tuna", "Southern bluefin tuna"…
## $ species_definition    <chr> "Thunnus maccoyii", "Thunnus maccoyii", "Thunnus…
## $ ocean                 <chr> "WCPFC", "WCPFC", "WCPFC", "IOTC", "IOTC", "IOTC…
## $ fishing_mode_label    <chr> "Undefined school", "Undefined school", "Undefin…

summary(catch_data$measurement_value)

##      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
##       0.0       0.8       4.1      94.2      21.6 1158525.0

kableExtra::kable(table(catch_data$measurement_unit))

Var1	Freq
Number of fish	580236
Tons	4454883

🗺️ Catch map

This map shows the spatial distribution of catches or effort aggregated by geographic grid cell. The measurement_value is summed per area.

catch_summary_by_unit <- catch_data %>%
  dplyr::group_by(geographic_identifier, measurement_unit) %>%
  dplyr::summarise(total = sum(measurement_value, na.rm = TRUE), .groups = "drop")

catch_summary_geom <- catch_summary_by_unit %>%
  dplyr::inner_join(cwp_grid %>% dplyr::select(cwp_code, geom), by = c("geographic_identifier" = "cwp_code")) %>%
  sf::st_as_sf()

plots <- catch_summary_geom %>%
  dplyr::group_split(measurement_unit) %>%
  purrr::map(function(df) {
    unit <- unique(df$measurement_unit)
    
    ggplot2::ggplot() +
      ggplot2::geom_sf(data = continent, fill = "gray90", color = "gray60") +
      ggplot2::geom_sf(data = df, ggplot2::aes(fill = total), color = NA) +
      ggplot2::scale_fill_viridis_c(option = "plasma", trans = "log", na.value = "transparent") +
      ggplot2::labs(title = paste("Unit:", unit), fill = "Catch Total") +
      ggplot2::theme_minimal()
  })

patchwork::wrap_plots(plots, ncol = 2)

📈 Annual catch trends

catch_data %>%
  dplyr::group_by(year = lubridate::year(time_start), measurement_unit) %>%
  dplyr::summarise(total = sum(measurement_value, na.rm = TRUE)) %>%
  ggplot2::ggplot(ggplot2::aes(x = year, y = total)) +
  ggplot2::geom_line() +
  ggplot2::facet_wrap(~measurement_unit, scales = "free_y") +
  ggplot2::labs(title = "Total catches by Year and Measurement unit", y = "Measured Value", x = "Year")

Your turn:

If you want to use only part of the data you can filter the dataset using the dimensison provided exemple, here let’s focus on silky sharks (note that this only concern catch data as effort cannot be related to species)

silky_sharks <- catch_data %>% dplyr::filter(species_label == "Silky shark")

silky_sharks %>%
  dplyr::group_by(year = lubridate::year(time_start), measurement_unit) %>%
  dplyr::summarise(total = sum(measurement_value, na.rm = TRUE)) %>%
  ggplot2::ggplot(ggplot2::aes(x = year, y = total)) +
  ggplot2::geom_line() +
  ggplot2::facet_wrap(~measurement_unit, scales = "free_y") +
  ggplot2::labs(title = "Total catches by Year and Measurement unit", y = "Measured Value", x = "Year")

Let’s decompose the data for some dimensions

groupped_gear <- silky_sharks %>% dplyr::group_by(gear_label, time_start, measurement_unit) %>% 
  dplyr::summarise(measurement_value = sum(measurement_value))

ggplot(groupped_gear) +
  aes(
    x = time_start,
    y = measurement_value,
    colour = gear_label
  ) +
  geom_line() +  #
  geom_point() +
  scale_color_hue(direction = 1) +
  theme_minimal() +
  facet_wrap(vars(measurement_unit), scales = "free")

groupped_fishing_fleet <- silky_sharks %>% dplyr::group_by(fishing_fleet_label, time_start, measurement_unit) %>% 
  dplyr::summarise(measurement_value = sum(measurement_value))

ggplot(groupped_fishing_fleet) +
  aes(
    x = time_start,
    y = measurement_value,
    colour = fishing_fleet_label
  ) +
  geom_line() +  #
  geom_point() +
  scale_color_hue(direction = 1) +
  theme_minimal() +
  facet_wrap(vars(measurement_unit), scales = "free")

groupped_source_auth <- silky_sharks %>% dplyr::group_by(source_authority, time_start, measurement_unit) %>% 
  dplyr::summarise(measurement_value = sum(measurement_value))

ggplot(groupped_source_auth) +
  aes(
    x = time_start,
    y = measurement_value,
    colour = source_authority
  ) +
  geom_line() +  #
  geom_point() +
  scale_color_hue(direction = 1) +
  theme_minimal() +
  facet_wrap(vars(measurement_unit), scales = "free")

How to aggregate data from 1 deg to 5 deg ?

source("https://raw.githubusercontent.com/firms-gta/geoflow-tunaatlas/master/R/sardara_functions/transform_cwp_code_from_1deg_to_5deg.R")

one_degree <- silky_sharks %>% dplyr::filter(substr(geographic_identifier, 1, 1) == "5")
five_degree <- silky_sharks %>% dplyr::filter(substr(geographic_identifier, 1, 1) == "6")
silky_sharks_aggregated <- one_degree %>% rowwise() %>% 
  dplyr::mutate(geographic_identifier = transform_cwp_code_from_1deg_to_5deg(geographic_identifier))

How to disaggregate ? (need connection to DB for now)

source("https://raw.githubusercontent.com/eblondel/geoflow-tunaatlas/master/R/tunaatlas_scripts/generation/spatial_curation_downgrade_resolution.R")

georef_dataset<-spatial_curation_downgrade_resolution(con,silky_sharks,resolution = 1 , remove = FALSE)

Catch dataset analysis (Global Tuna Atlas - IRD Level 2)