.Rhistory

commercial_gbif %>%
filter(accepted %in% commercial_products$sciname_joinable_to_fishbase) %>%
select(accepted, family, order, class),
by = c("sciname_joinable_to_fishbase" = "accepted")
)
commercial_gbif %>%
filter(accepted %in% commercial_products$sciname_joinable_to_fishbase)
commercial_gbif
commercial_gbif <- commercial_gbif %>%
filter(accepted %in% commercial_products$sciname_joinable_to_fishbase)
commercial_gbif
commercial_products %>%
commercial_products %>%
commercial_products <- commercial_products %>%
mutate(
species = TrueSpecies,
genus = Genus,
across(
c(family, order, class),
~ map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(accepted, .x)) %>%
count(.x) %>%
arrange(.x) %>%
drop_na() %>%
pull(.x)  # Pull the respective column based on the target
if (length(result) > 0) result[1] else NA_character_
})
)) %>%
select(-c(TrueSpecies, Genus))
names(commercial_products)
commercial_products
commercial_gbif %>%
filter(str_detect(accepted, "Abalistes stellatus")) %>%
count(accepted) %>%  # Count occurrences of each accepted name
arrange(desc(n))
commercial_gbif %>%
filter(str_detect(accepted, "Abalistes stellatus")) %>%
count(genus) %>%  # Count occurrences of each accepted name
arrange(desc(n))
names(commercial_products)
commercial_products
commercial_gbif %>%
filter(str_detect(accepted, "Abramis brama")) %>%
count(genus) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(genus)
commercial_gbif %>%
filter(str_detect(accepted, "Abramis brama")) %>%
count(genus) %>%  # Count occurrences of each accepted name
arrange(desc(n))
commercial_products
commercial_gbif %>%
filter(str_detect(accepted, "Abudefduf vaigiensis")) %>%
count(genus) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(genus)
commercial_gbif %>%
filter(str_detect(accepted, "Abudefduf vaigiensis")) %>%
count(genus) %>%  # Count occurrences of each accepted name
arrange(desc(n))
commercial_gbif %>%
count(accepted)
commercial_gbif %>%
count(accepted) %>%
arrange(-n)
commercial_gbif %>%
filter(str_detect(accepted, "Abudefduf vaigiensis")) %>%
count(genus) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(genus)
commercial_gbif %>%
count(accepted) %>%
arrange(-n) %>%
drop_na() %>%
pull(accepted)
commercial_products <- commercial_products %>%
mutate(
species = TrueSpecies,
genus = Genus,
# Apply across to mutate multiple columns (family, order, class)
family = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(accepted, .x)) %>%
count(family) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(family)  # Pull the family column after filtering and processing
if (length(result) > 0) result[1] else NA_character_  # Return first result or NA
}),
order = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(accepted, .x)) %>%
count(order) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(order)
if (length(result) > 0) result[1] else NA_character_
}),
class = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(accepted, .x)) %>%
count(class) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(class)
if (length(result) > 0) result[1] else NA_character_
})
) %>%
select(-c(TrueSpecies, Genus))
commercial_products
commercial_gbif %>%
filter(accepted == "Acanthurus fowleri")
commercial_products %>%
filter(accepted == "Acanthurus fowleri")
commercial_products %>%
filter(sciname_joinable_to_fishbase == "Acanthurus fowleri")
name_lookup("Acanthurus fowleri")
commercial_gbif
commercial_gbif %>%
count(accepted)
commercial_gbif %>%
count(accepted) %>%
arrange(-n)
commercial_gbif
read_parquet("../data/rgbif_data/commercial_higher_taxa_names.parquet") %>%
count(accepted) %>%
arrange(-n)
# Pull rgbif joined data with commercial species
commercial_gbif <- read_parquet("../data/rgbif_data/commercial_higher_taxa_names.parquet")
commercial_gbif
commercial_products %>%
filter(is.na(family) & !is.na(order) & !is.na(class))
commercial_products
commercial_products %>%
filter(is.na(family) & !is.na(order) & !is.na(class))
commercial_products
commercial_products %>%
filter(!is.na(family) & !is.na(order) & !is.na(class))
commercial_products <- commercial_products %>%
mutate(
species = TrueSpecies,
genus = Genus,
# Apply across to mutate multiple columns (family, order, class)
family = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(family) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(family)  # Pull the family column after filtering and processing
if (length(result) > 0) result[1] else NA_character_  # Return first result or NA
}),
order = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(order) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(order)
if (length(result) > 0) result[1] else NA_character_
}),
class = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(class) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(class)
if (length(result) > 0) result[1] else NA_character_
})
) %>%
select(-c(TrueSpecies, Genus))
commercial_products
# Step 4: Join full matched species only with fb/slb species of commercial interest
commercial_products <- inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase))
commercial_products <- commercial_products %>%
mutate(
species = TrueSpecies,
genus = Genus,
# Apply across to mutate multiple columns (family, order, class)
family = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(family) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(family)  # Pull the family column after filtering and processing
if (length(result) > 0) result[1] else NA_character_  # Return first result or NA
}),
order = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(order) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(order)
if (length(result) > 0) result[1] else NA_character_
}),
class = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(class) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(class)
if (length(result) > 0) result[1] else NA_character_
})
) %>%
select(-c(TrueSpecies, Genus))
commercial_products <- commercial_products %>%
mutate(
species = TrueSpecies,
genus = Genus,
# Apply across to mutate multiple columns (family, order, class)
family = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(family) %>%  # Count occurrences of each accepted name
arrange(desc(n)) %>%  # Arrange by count (descending)
drop_na() %>%
pull(family)  # Pull the family column after filtering and processing
if (length(result) > 0) result[1] else NA_character_  # Return first result or NA
}),
order = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(order) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(order)
if (length(result) > 0) result[1] else NA_character_
}),
class = map_chr(sciname_joinable_to_fishbase, ~ {
result <- commercial_gbif %>%
filter(str_detect(scientificName, .x)) %>%
count(class) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(class)
if (length(result) > 0) result[1] else NA_character_
})
) %>%
select(-c(TrueSpecies, Genus))
commercial_products
# 3914
commercial_products %>%
filter(!is.na(family) & !is.na(order) & !is.na(class))
# 3914
commercial_products %>%
filter(!is.na(family) & !is.na(order) & !is.na(class))
inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase))
inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase)) %>%
mutate(species = TrueSpecies, genus = Genus)
inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase)) %>%
mutate(species = TrueSpecies, genus = Genus)
source("../code/test_script.Rmd")
source("code/test_script.Rmd")
source("code/test_script.R")
source("../code/test_script.Rmd")
source("code/test_script.Rmd")
file.exists("code/test_script.Rmd")
file.exists("../code/test_script.Rmd")
source("../code/test_script.Rmd")
rmarkdown::render("../code/test_script.Rmd")
knitr::opts_chunk$set(echo = TRUE)
pee = "hi"
rmarkdown::render("../code/test_script.Rmd")
print(poo)
rmarkdown::render("../code/test_script.Rmd")
# Step 4: Join full matched species only with fb/slb species of commercial interest
commercial_products <- inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase))
commercial_products
# Helper function to get the most frequent taxonomy value
get_taxonomy_value <- function(sciname, column_name, gbif_data) {
result <- gbif_data %>%
filter(str_detect(scientificName, sciname)) %>%
count(!!sym(column_name)) %>%
arrange(desc(n)) %>%
drop_na() %>%
pull(!!sym(column_name))
if (length(result) > 0) result[1] else NA_character_
}
# Apply across family, order, and class in a more efficient way
commercial_products <- commercial_products %>%
mutate(
species = TrueSpecies,
genus = Genus,
# Use map to apply the helper function to multiple taxonomy columns
family = map_chr(sciname_joinable_to_fishbase, ~ get_taxonomy_value(.x, "family", commercial_gbif)),
order = map_chr(sciname_joinable_to_fishbase, ~ get_taxonomy_value(.x, "order", commercial_gbif)),
class = map_chr(sciname_joinable_to_fishbase, ~ get_taxonomy_value(.x, "class", commercial_gbif))
) %>%
select(-c(TrueSpecies, Genus))
commercial_products
knitr::opts_chunk$set(echo = TRUE)
file.exists("../data/fb_slb_data")
file.exists("../data/fb_slb_datad")
file.exists("../data/fb_slb_data")
file.exists("../data/fb_slb_data/")
if (!file.exists("../data/fb_slb_data/"))
!file.exists("../data/fb_slb_data/")
!file.exists("../data/fb_slb_data/")
file.exists("../data/rgbif_data/commercial_higher_taxa_names.parquet")
commercial_products
# Step 4: Join full matched species only with fb/slb species of commercial interest
commercial_products <- inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase))
commercial_products
if (file.exists("../data/rgbif_data/commercial_products.parquet"))
file.exists("../data/rgbif_data/commercial_products.parquet")
file.exists("../data/rgbif_data/commercial_products.parquet")
!file.exists("../data/rgbif_data/commercial_products.parquet")
!file.exists("../data/rgbif_data/commercial_higher_taxa_names.parquet")
# Step 4: Join full matched species only with fb/slb species of commercial interest
commercial_products <- inner_join(pa_excluding_artis, fb_slb_commercial_interests, by = c("sciname_joinable_to_fishbase" = "Species", "countries_found" = "country")) %>%
mutate(presence_or_absence = 1) %>%
mutate(sciname = str_to_lower(sciname_joinable_to_fishbase))
# commercial_products is fed through this script (takes awhile hence in other script)
rmarkdown::render("../code/QuirozConnor_SGL_Obtain_gbif_Data.Rmd")
commercial_products
write_parquet("../data/rgbif_data/commercial_products.parquet")
write_parquet
write_parquet(commercial_products, "../data/rgbif_data/commercial_products.parquet")
!file.exists("../data/rgbif_data/commercial_products.parquet")
commercial_products
pa_excluding_artis <- read_parquet("../data/rgbif_data/commercial_products.parquet")
pa_excluding_artis
merged_presence_absence_data_country
names(merged_presence_absence_data_country)
names(pa_excluding_arits)
pa_excluding_artis <- read_parquet("../data/rgbif_data/commercial_products.parquet")
names(merged_presence_absence_data_country)
names(pa_excluding_artis)
bind_rows(pa_only_artis, pa_excluding_artis)
pa_only_artis <- bind_rows(pa_only_artis, pa_excluding_artis)
presence_absence_data_country_all_taxa <- data.frame()
for (i in unique(pa_only_artis$eez_iso3c)) {
# Group data by eez country
sliced_data <- pa_only_artis %>%
filter(eez_iso3c == sym(i))
# Create updated presence on sliced data
sliced_data_for_merging <- sliced_data %>%
mutate(unique_vals = str_replace_all(paste0(filter(., presence_or_absence == 1) %>% select(sciname, species, genus, subfamily, family, order, class, superclass, phylum, kingdom) %>%
unlist() %>%
unique(), collapse = " "), "[A-Z]", ""), presence_or_absence_all_taxa = case_when(str_detect(unique_vals, sciname) ~ 1, TRUE ~ 0))
# Combine rows of each country
presence_absence_data_country_all_taxa <- bind_rows(presence_absence_data_country_all_taxa, sliced_data_for_merging)
}
unique(pa_only_artis$eez_iso3c)
presence_absence_data_country_all_taxa
pa_only_artis <- bind_rows(pa_only_artis, pa_excluding_artis)
presence_absence_data_country_all_taxa <- data.frame()
for (i in unique(pa_only_artis$eez_iso3c)) {
# Group data by eez country
sliced_data <- pa_only_artis %>%
filter(eez_iso3c == i)
# Create updated presence on sliced data
sliced_data_for_merging <- sliced_data %>%
mutate(unique_vals = str_replace_all(paste0(filter(., presence_or_absence == 1) %>% select(sciname, species, genus, subfamily, family, order, class, superclass, phylum, kingdom) %>%
unlist() %>%
unique(), collapse = " "), "[A-Z]", ""), presence_or_absence_all_taxa = case_when(str_detect(unique_vals, sciname) ~ 1, TRUE ~ 0))
# Combine rows of each country
presence_absence_data_country_all_taxa <- bind_rows(presence_absence_data_country_all_taxa, sliced_data_for_merging)
}
# Count Number of presences/absences in filled in presence data
presence_absence_data_country_all_taxa <- presence_absence_data_country_all_taxa %>%
relocate(presence_or_absence_all_taxa, .after = presence_or_absence)
presence_absence_data_country_all_taxa %%>
presence_absence_data_country_all_taxa %>%
count(presence_or_absence_all_taxa)
commercial_products
commercial_products %>%
mutate(genus = str_to_lower(genus),
family = str_to_lower(family),
order = str_to_lower(order),
class = str_to_lower(class),
phylum = str_to_lower(phylum),
kingdom = str_to_lower(kingdom))
commercial_products <- commercial_products %>%
mutate(genus = str_to_lower(genus),
family = str_to_lower(family),
order = str_to_lower(order),
class = str_to_lower(class),
phylum = str_to_lower(phylum),
kingdom = str_to_lower(kingdom))
write_parquet(commercial_products, "../data/rgbif_data/commercial_products.parquet")
pa_excluding_artis <- read_parquet("../data/rgbif_data/commercial_products.parquet")
pa_excluding_artis
only_artis_1 <- merged_presence_absence_data_country %>%
filter(joined %in% c("only in artis", "both"))
only_artis_1
pa_only_artis
pa_excluding_artis
presence_absence_data_country_all_taxa %>%
count(presence_or_absence_all_taxa)
```
pa_excluding_artis
pa_excluding_artis %>% distinct(sciname_joinable_to_fishbase)
only_artis_1 <- merged_presence_absence_data_country %>%
filter(joined %in% c("only in artis", "both"))
pa_only_artis <- bind_rows(only_artis_1, pa_excluding_artis)
presence_absence_data_country_all_taxa <- data.frame()
for (i in unique(pa_only_artis$eez_iso3c)) {
# Group data by eez country
sliced_data <- pa_only_artis %>%
filter(eez_iso3c == i)
# Create updated presence on sliced data
sliced_data_for_merging <- sliced_data %>%
mutate(unique_vals = str_replace_all(paste0(filter(., presence_or_absence == 1) %>% select(sciname, species, genus, subfamily, family, order, class, superclass, phylum, kingdom) %>%
unlist() %>%
unique(), collapse = " "), "[A-Z]", ""), presence_or_absence_all_taxa = case_when(str_detect(unique_vals, sciname) ~ 1, TRUE ~ 0))
# Combine rows of each country
presence_absence_data_country_all_taxa <- bind_rows(presence_absence_data_country_all_taxa, sliced_data_for_merging)
}
# Count Number of presences/absences in filled in presence data
presence_absence_data_country_all_taxa <- presence_absence_data_country_all_taxa %>%
relocate(presence_or_absence_all_taxa, .after = presence_or_absence)
presence_absence_data_country_all_taxa %>%
count(presence_or_absence_all_taxa)
presence_absence_data_country_all_taxa
bind_rows(only_artis_1, pa_excluding_artis)
pa_excluding_artis
commercial_products
commercial_products %>%
mutate(eez_iso3c = countries_found) %>%
select(-countries_found)
commercial_products <- commercial_products %>%
mutate(eez_iso3c = countries_found) %>%
select(-countries_found)
write_parquet(commercial_products, "../data/rgbif_data/commercial_products.parquet")
pa_excluding_artis <- read_parquet("../data/rgbif_data/commercial_products.parquet")
pa_excluding_artis
pa_excluding_artis
only_artis_1
only_artis_1 <- merged_presence_absence_data_country %>%
filter(joined %in% c("only in artis", "both"))
pa_only_artis <- bind_rows(only_artis_1, pa_excluding_artis)
presence_absence_data_country_all_taxa <- data.frame()
for (i in unique(pa_only_artis$eez_iso3c)) {
# Group data by eez country
sliced_data <- pa_only_artis %>%
filter(eez_iso3c == i)
# Create updated presence on sliced data
sliced_data_for_merging <- sliced_data %>%
mutate(unique_vals = str_replace_all(paste0(filter(., presence_or_absence == 1) %>% select(sciname, species, genus, subfamily, family, order, class, superclass, phylum, kingdom) %>%
unlist() %>%
unique(), collapse = " "), "[A-Z]", ""), presence_or_absence_all_taxa = case_when(str_detect(unique_vals, sciname) ~ 1, TRUE ~ 0))
# Combine rows of each country
presence_absence_data_country_all_taxa <- bind_rows(presence_absence_data_country_all_taxa, sliced_data_for_merging)
}
# Count Number of presences/absences in filled in presence data
presence_absence_data_country_all_taxa <- presence_absence_data_country_all_taxa %>%
relocate(presence_or_absence_all_taxa, .after = presence_or_absence)
presence_absence_data_country_all_taxa %>%
count(presence_or_absence_all_taxa)
presence_absence_data_country_all_taxa %>%
group_by(eez_iso3c) %>%
distinct(sciname, .keep_all = TRUE) %>%
select(sciname, presence_or_absence_all_taxa) %>%
count(presence_or_absence_all_taxa) %>%
mutate(presence_or_absence_all_taxa = as.character(presence_or_absence_all_taxa)) %>%
pivot_wider(values_from = n, names_from = presence_or_absence_all_taxa, names_glue = "quantity_{presence_or_absence_all_taxa}") %>%
mutate(prop_missing =  quantity_0 / (quantity_0 + quantity_1)) %>%
filter(is.na(prop_missing) & is.na(quantity_1)) %>%
rename(number_absences = "quantity_0", number_presences = "quantity_1")
prop_absences <- presence_absence_data_country_all_taxa %>%
group_by(eez_iso3c) %>%
distinct(sciname, .keep_all = TRUE) %>%
select(sciname, presence_or_absence_all_taxa) %>%
count(presence_or_absence_all_taxa) %>%
mutate(presence_or_absence_all_taxa = as.character(presence_or_absence_all_taxa)) %>%
pivot_wider(values_from = n, names_from = presence_or_absence_all_taxa, names_glue = "quantity_{presence_or_absence_all_taxa}") %>%
mutate(prop_missing =  quantity_0 / (quantity_0 + quantity_1)) %>%
ggplot(aes(x = prop_missing)) +
geom_boxplot(fill = artis_palette(1)) +
labs(x = "Prop of countries' stock reported as absent\n(Fishbase/Sealifebase") +
theme_light() +
theme(axis.title.y=element_blank(),
axis.text.y=element_blank(),
axis.ticks.y=element_blank())
# Proportion of reports that are seen as countrywide absences according to fishbase/sealifebase
(prop_absences <- presence_absence_data_country_all_taxa %>%
group_by(eez_iso3c) %>%
distinct(sciname, .keep_all = TRUE) %>%
select(sciname, presence_or_absence_all_taxa) %>%
count(presence_or_absence_all_taxa) %>%
mutate(presence_or_absence_all_taxa = as.character(presence_or_absence_all_taxa)) %>%
pivot_wider(values_from = n, names_from = presence_or_absence_all_taxa, names_glue = "quantity_{presence_or_absence_all_taxa}") %>%
mutate(prop_missing =  quantity_0 / (quantity_0 + quantity_1)) %>%
ggplot(aes(x = prop_missing)) +
geom_boxplot(fill = artis_palette(1)) +
labs(x = "Prop of countries' stock reported as absent\n(Fishbase/Sealifebase") +
theme_light() +
theme(axis.title.y=element_blank(),
axis.text.y=element_blank(),
axis.ticks.y=element_blank()))
ggsave("../images/prop_absences.jpg", plot = prop_absences, device = "jpeg", height = 3, width = 5, units = "in")