Get occurrence (presence) data for multiple species (including adhoc records) from the africabirdmap API

get_multiple_species_data.R

# Function adapted from https://github.com/AfricaBirdData/ABAP/
# Only pulls presence data

library("tidyverse")
library("purrr")

jsonToTibble <- function(jsonfile) {
  out <- jsonfile %>%
    lapply(function(x) {
      x[sapply(x, is.null)] <- NA
      x <- unlist(x)
      x[x == "N/A"] <- NA
      cnames <- names(x)
      df <- data.frame()
      df <- rbind(df, x)
      names(df) <- cnames
      return(df)
    }) %>%
    data.table::rbindlist(fill = TRUE) %>%
    dplyr::as_tibble()
  
  return(out)
}

getAbapData <- function(.spp_code,
                        .region_type = c("country", "province", "pentad"),
                        .region,
                        .years = NULL,
                        .adhoc = FALSE) {
  if (is.null(.region_type)) {
    .region_type <- "country"
  } else if (!.region_type %in% c("country", "province", "pentad")) {
    stop(".region_type must be one of 'country', 'province', 'pentad'")
  }
  
  if (!is.null(.years)) {
    .years <- paste(.years, collapse = ",")
    .years <- paste0("&year=", .years)
  }
  
  .region <- tolower(.region)
  .region <- gsub(" ", "", .region)
  
  url <- paste0(
    "https://api.birdmap.africa/sabap2/v2/cards/species/85ee37929696cba93e1cdda4dbb3f93a/",
    .spp_code,
    "/",
    .region_type,
    "/",
    .region,
    "?format=csv",
    .years
  )
  
  if (.adhoc) {
    url <- paste0(url, "&adhoc=1")
  }
  
  # Extract data
  myfile <- httr::RETRY("GET", url) %>%
    httr::content(as = "text", encoding = "UTF-8")
  
  if (myfile == "") {
      return(NULL)
  }
  
  # Format
  out <- myfile %>%
    textConnection() %>%
    utils::read.csv(header = TRUE) %>%
    dplyr::as_tibble()
  
  # If there are multiple regions (e.g. multiple pentads), then multiple headers
  # are produced and inserted as data. Eliminate these
  out <- out %>%
    dplyr::filter(Pentad != "Pentad")
  
  out <- out %>%
    readr::type_convert(
      col_types = readr::cols(
        .default = readr::col_integer(),
        CardNo = readr::col_character(),
        StartDate = readr::col_date(format = ""),
        EndDate = readr::col_date(format = ""),
        StartTime = readr::col_character(),
        Pentad = readr::col_character(),
        Spp = readr::col_character(),
        Sequence = readr::col_character(),
        Common_name = readr::col_character(),
        Taxonomic_name = readr::col_character()
      )
    )
  
  return(out)
}


# species_list can be any arbitrary list of species IDs
# examples species_list = c(1, 196)
# In this case we've used the Kenya bird list to get all the species IDs

species_list <- read_csv("Birds_of_Kenya_2019.csv") %>%
  select(ADU) %>%
  na.omit() %>%
  pull()

occurence_data <- species_list %>%
  map_df(
    ~ getAbapData(
      .spp_code = .,
      .region_type = "country",
      .region = "kenya",
      .adhoc = TRUE
    ),
    .default = data_frame(
      CardNo = NA,
      StartDate = NA,
      EndDate = NA,
      StartTime = NA,
      Pentad = NA,
      ObserverNo = NA,
      TotalHours = NA,
      Hour1 = NA,
      Hour2 = NA,
      Hour3 = NA,
      Hour4 = NA,
      Hour5 = NA,
      Hour6 = NA,
      Hour7 = NA,
      Hour8 = NA,
      Hour9 = NA,
      Hour10 = NA,
      TotalSpp = NA,
      InclNight = NA,
      AllHabitats = NA,
      Spp = NA,
      Sequence = NA,
      Common_name = NA,
      Taxonomic_name = NA
    )
  )