In sharksmhi/SHARK4R: Retrieving, Analyzing, and Validating Marine Data from SHARK and Nordic Microalgae

WoRMS

The World Register of Marine Species (WoRMS) is a comprehensive database providing authoritative lists of marine organism names, managed by taxonomic experts. It combines data from the Aphia database and other sources like AlgaeBase and FishBase, offering species names, higher classifications, and additional data. WoRMS is continuously updated and maintained by taxonomists. In this tutorial, we source the R package worrms to access WoRMS data for our function. Please note that the authors of SHARK4R are not affiliated with WoRMS.

Getting Started

Installation

You can install the package from GitHub using the devtools package:

# install.packages("devtools")
devtools::install_github("sharksmhi/SHARK4R",
                         dependencies = TRUE)

Load the SHARK4R and dplyr libraries:

library(SHARK4R)
library(dplyr)
library(ggplot2)

suppressPackageStartupMessages({
  library(SHARK4R)
  library(dplyr)
  library(ggplot2)
})

Retrieve Data Using SHARK4R

Retrieve Phytoplankton Data From SHARK

Phytoplankton data, including scientific names and AphiaIDs, are downloaded from SHARK. To see more download options, please visit the Retrieve Data From SHARK tutorial.

# Retrieve all phytoplankton data from April 2015
shark_data <- get_shark_data(fromYear = 2015, 
                             toYear = 2015,
                             months = 4, 
                             dataTypes = c("Phytoplankton"),
                             verbose = FALSE)

Match Taxa Names

Taxon names can be matched with the WoRMS API to retrieve Aphia IDs and corresponding taxonomic information. The get_worms_records_name function incorporates retry logic to handle temporary failures, ensuring that all names are processed successfully.

# Find taxa without Aphia ID
no_aphia_id <- shark_data %>%
  filter(is.na(aphia_id))

# Randomly select taxa with missing aphia_id
taxa_names <- sample(unique(no_aphia_id$scientific_name), 
                     size = 10,
                     replace = TRUE)

# Match taxa names with WoRMS
worms_records <- get_worms_records_name(unique(taxa_names),
                                        fuzzy = TRUE,
                                        best_match_only = TRUE,
                                        marine_only = TRUE,
                                        verbose = FALSE)

# Print result as tibble
tibble(worms_records)

Get WoRMS records from AphiaID

Taxonomic records can also be retrieved using Aphia IDs, employing the same retry and error-handling logic as the get_worms_records_name function.

# Randomly select ten Aphia IDs
aphia_ids <- sample(unique(shark_data$aphia_id), 
                    size = 10)

# Remove NAs
aphia_ids <- aphia_ids[!is.na(aphia_ids)]

# Retrieve records
worms_records <- get_worms_records(aphia_ids,
                                   verbose = FALSE)

# Print result as tibble
tibble(worms_records)

Get WoRMS Taxonomy

SHARK sources taxonomic information from Dyntaxa, which is reflected in columns starting with taxon_xxxxx. Equivalent columns based on WoRMS can be retrieved using the add_worms_taxonomy function.

# Retrieve taxonomic table
worms_taxonomy <- add_worms_taxonomy(aphia_ids,
                                     verbose = FALSE)

# Print result as tibble
tibble(worms_taxonomy)

# Enrich data with data from WoRMS
shark_data_with_worms <- shark_data %>%
  left_join(worms_taxonomy, by = c("aphia_id", "scientific_name"))

Assign Phytoplankton Groups

Phytoplankton data are often categorized into major groups such as Dinoflagellates, Diatoms, Cyanobacteria, and Others. This grouping can be achieved by referencing information from WoRMS and assigning taxa to these groups based on their taxonomic classification, as demonstrated in the example below.

# Subset a few national monitoring stations
nat_stations <- shark_data %>%
  filter(station_name %in% c("BY31 LANDSORTSDJ"))

# Randomly select one sample from the nat_stations
sample <- sample(unique(nat_stations$shark_sample_id_md5), 1)

# Subset the random sample
shark_data_subset <- shark_data %>%
  filter(shark_sample_id_md5 == sample)

# Assign groups by providing both scientific name and Aphia ID
plankton_groups <- assign_phytoplankton_group(
  scientific_names = shark_data_subset$scientific_name,
  aphia_ids = shark_data_subset$aphia_id,
  verbose = FALSE)

# Print result
tibble(distinct(plankton_groups))

# Add plankton groups to data and summarize abundance results
plankton_group_sum <- shark_data_subset %>%
  mutate(plankton_group = plankton_groups$plankton_group) %>%
  filter(parameter == "Abundance") %>%
  group_by(plankton_group) %>%
  summarise(sum_plankton_groups = sum(value, na.rm = TRUE))

# Plot a pie chart
ggplot(plankton_group_sum, 
       aes(x = "", y = sum_plankton_groups, fill = plankton_group)) +
  geom_col(width = 1) +
  coord_polar(theta = "y") +
  labs(
    title = "Phytoplankton Groups",
    subtitle = paste(unique(shark_data_subset$station_name),
                     unique(shark_data_subset$sample_date)),
    fill = "Plankton Group"
  ) +
  theme_void() +
  theme(plot.background = element_rect(fill = "white", color = NA))

You can add custom plankton groups by using the custom_groups parameter, allowing flexibility to categorize plankton based on specific taxonomic criteria. Please note that the order of the list matters: taxa are assigned to the last matching group. For example: Mesodinium rubrum will be excluded from the Ciliates group because it appears after Ciliates in the list in the example below.

# Define custom plankton groups using a named list
custom_groups <- list(
  "Cryptophytes" = list(class = "Cryptophyceae"),
  "Green Algae" = list(class = c("Trebouxiophyceae", 
                                 "Chlorophyceae", 
                                 "Pyramimonadophyceae"),
                       phylum = "Chlorophyta"),
  "Ciliates" = list(phylum = "Ciliophora"),
  "Mesodinium rubrum" = list(scientific_name = "Mesodinium rubrum"),
  "Dinophysis" = list(genus = "Dinophysis")
)

# Assign groups by providing scientific name only, and adding custom groups
plankton_groups <- assign_phytoplankton_group(
  scientific_names = shark_data_subset$scientific_name,
  custom_groups = custom_groups,
  verbose = FALSE)

# Add new plankton groups to data and summarize abundance results
plankton_custom_group_sum <- shark_data_subset %>%
  mutate(plankton_group = plankton_groups$plankton_group) %>%
  filter(parameter == "Abundance") %>%
  group_by(plankton_group) %>%
  summarise(sum_plankton_groups = sum(value, na.rm = TRUE))

# Plot a new pie chart, including the custom groups
ggplot(plankton_custom_group_sum, 
       aes(x = "", y = sum_plankton_groups, fill = plankton_group)) +
  geom_col(width = 1) +
  coord_polar(theta = "y") +
  labs(
    title = "Phytoplankton Custom Groups",
    subtitle = paste(unique(shark_data_subset$station_name),
                     unique(shark_data_subset$sample_date)),
    fill = "Plankton Group"
  ) +
  theme_void() +
  theme(plot.background = element_rect(fill = "white", color = NA))