In karinorman/biodivTS: What the Package Does (One Line, Title Case)

knitr::opts_chunk$set(echo = TRUE)

library(tidyverse)
library(rgbif)
library(geonames)
library(countrycode)
library(taxadb)

Try with the USDA master list of invasive species

invasive <- read.csv(here::here("inst/extdata/USRIISv2csvFormat/USRIISv2_MasterList.csv"))
load(here::here("bt_traitfiltered.rda"))

# all invasive species have GBIF id's so let's match on that
ids <- bt_traitfiltered %>%
  select(species) %>% 
  distinct() %>%
  mutate(sciName = taxadb::get_names(species),
         gbif = taxadb::get_ids(sciName, provider = "gbif"))

included_invasives <- invasive %>%
  filter(locality == "L48") %>% 
  select(GBIF_taxonKey, establishmentMeans, degreeOfEstablishment) %>%
  distinct() %>% 
  mutate(gbif = paste0("GBIF:", GBIF_taxonKey)) %>%
  inner_join(ids)

Get a count of invasives for each study %>%

invasive_count <- bt_traitfiltered %>%
  select(species, study_id, rarefyid, year) %>%
  left_join(included_invasives %>% select(species) %>%
              mutate(invasive = 1)) %>%
  select(-species) %>%
  distinct() %>%
  group_by(study_id, rarefyid, year) %>%
  summarize(invasive_count = sum(invasive, na.rm = TRUE))

Get GBIF data

#get all the datasets from the Invasive Species Specialist Group
datasets <- dataset_search(publishingOrg = "cdef28b1-db4e-4c58-aa71-3c5238c2d0b5", limit = 300)
datasets_list <- datasets$data

occ <- occ_download(dataset = "b351a324-77c4-41c9-a909-f30f77268bc4")

Figure out which country each study is located in

pins::board_register_github(name = "github", repo = "karinorman/biodivTS_data", branch = "master")
options(geonamesUsername="karinorman")

obs_data <- pins::pin_get("bt-traitfiltered", board = "github") %>%
  select(study_id, species) %>%
  distinct() %>%
  mutate(scientificName = get_names(species, "itis"),
         gbifID = get_ids(scientificName, "gbif")) %>%
  mutate(gbifID = case_when(
    scientificName == "Oreortyx pictus" ~ "GBIF:2474320",
    scientificName == "Aegithalos caudatus" ~ "GBIF:2495000",
    scientificName == "Papio anubis" ~ "GBIF:5707341",
    scientificName == "Boana bischoffi" ~ "GBIF:160065393",
    scientificName == "Boana faber" ~ "GBIF:10863437",
    scientificName == "Ololygon rizibilis" ~ "GBIF:10896287",
    scientificName == "Ololygon perpusilla" ~ "GBIF:7341662",
    scientificName == "Boana pardalis" ~ "GBIF:10842532",
    scientificName == "Ololygon littoralis" ~ "GBIF:10733643",
    scientificName == "Brachycephalus pitanga" ~ "GBIF:10846405",
    TRUE ~ gbifID
  )) %>%
  separate(gbifID, c("gbif", "ID"), remove = F) %>%
  select(-gbif)

# "Boana caipora" and "Boana albomarginata" have no GBIF ID 

meta <- pins::pin_get("meta", board = "github") %>%
  select(study_id, cent_lat, cent_long) %>%
  filter(study_id %in% obs_data$study_id)

# get_GNcountry <- possibly(GNcountryCode, otherwise = NA_character_)
# 
# get_country <- function(study_id, cent_lat, cent_long){
#   #con_data <- get_GNcountry(lat = cent_lat, lng = cent_long, lang = "en", radius = 50)
#   con_data <- get_GNcountry(lat = cent_lat, lng = cent_long, radius = 50)
#   name <- ifelse(length(con_data) == 1, con_data, con_data$countryName)
#   
#   return(tibble(study_id = study_id, country = name))
# }
# 
# country_data <- pmap_dfr(meta, get_country)
# missing_cd <- country_data %>% filter(is.na(country)) %>% pull(study_id)
# 
# country_data2 <- pmap_dfr(meta %>% filter(study_id %in% missing_cd), get_country)
# missing_cd2 <- country_data2 %>% filter(is.na(country)) %>% pull(study_id)
# 
# country_data_bind <- bind_rows(country_data %>% filter(!study_id %in% country_data2$study_id),country_data2) %>%
#   mutate(country = case_when(
#   study_id == 108 ~ "Namibia",
#   study_id %in% c(166, 169) ~ "United States",
#   study_id == 172 ~ "Morocco",
#   TRUE ~ country
# )) %>%
#   mutate(code = countryname(country, destination = "genc2c")) %>%
#   mutate(code = case_when(
#     code == "CA" ~ "US",
#     code == "PR" ~ "DO",
#     TRUE ~ code
#   ))

country_data_bind <- pins::pin_get("country-data-bind", board = "github")

obs_data <- obs_data %>%
  left_join(country_data_bind)

Read in Invasive Data, first using rgbif

#rgbif invasive id's 
invasive_search <- occ_data(taxonKey = na.omit(obs_data$ID), establishmentMeans = "INVASIVE;INTRODUCED", limit = 3000)
invasive_data <- map_dfr(invasive_search, ~.x$data) %>%
  select(acceptedTaxonKey, countryCode, establishmentMeans)

gbif_invasive <- obs_data %>%
  mutate(ID = as.integer(ID)) %>%
  left_join(invasive_data, by = c("ID" = "acceptedTaxonKey", "code" = "countryCode")) %>%
  distinct() %>%
  filter(!is.na(establishmentMeans))

Read in invasive data from hand downloading from GBIF

#invasive data from manual download  
# invasive_files <- list.dirs(here::here("data", "invasive")) %>%
#   select(acceptedTaxonKey, countryCode, establishmentMeans, occurrenceStatus) %>% 
#   distinct()
# 
# invasive_data <- map_dfr(invasive_files[2:21], function(x){
#   taxon <- read_tsv(paste0(x, "/", "taxon.txt"))
#   profile <- read_tsv(paste0(x, "/", "speciesprofile.txt"))
#   dist <- read_tsv(paste0(x, "/", "distribution.txt")) %>%
#     select(id, countryCode)
#   
#   data <- left_join(taxon, profile) %>% left_join(dist)
#   return(data)
# }) %>%
#   mutate(cleanScientificName = clean_names(scientificName, lowercase = FALSE),
#          itisID = get_ids(cleanScientificName, "itis")) 

invasive_table <- pins::pin_get("invasive-data", board = "github") %>%
  mutate(cleanScientificName = clean_names(scientificName, lowercase = FALSE),
           itisID = get_ids(cleanScientificName, "itis"),
         gbifID = get_ids(cleanScientificName, "gbif"))

manual_invasive <-  obs_data %>%
  left_join(invasive_table %>% 
              select(itisID, countryCode, ITISisInvasive = isInvasive), 
            by = c("species" = "itisID", "code" = "countryCode")) %>%
  left_join(invasive_table %>% 
              select(gbifID, countryCode, GBIFisInvasive = isInvasive), 
            by = c("species" = "gbifID", "code" = "countryCode")) %>%
  distinct() %>% 
  filter(GBIFisInvasive == "invasive" | ITISisInvasive == "invasive")

Only six invasive species across three studies! "Mus musculus" "Rattus rattus" "Rattus norvegicus" "Sus scrofa" "Microtus arvalis" "Apodemus agrarius"

Substituting years: - orig: Canada, data: United States - orig: Puerto Rico, data: Dominican Republic

karinorman/biodivTS documentation built on Dec. 23, 2024, 10 p.m.