R/prodes.R

In datazoompuc/datazoom.amazonia: Simplify Access to Data from the Amazon Region

#' @title PRODES - Deforestation Monitoring Project in the Legal Amazon by Satellite
#'
#' @description Loads data on deforestation in the Legal Amazon region.
#'
#' @param dataset A dataset name ("deforestation").
#' @inheritParams load_baci
#'
#' @return A \code{tibble} with the selected data.
#'
#' @examples
#' \dontrun{
#' # Download treated data (raw_data = FALSE)
#' # in portuguese (language = 'pt').
#' data <- load_prodes(
#'   raw_data = FALSE,
#'   language = "pt"
#' )
#' }
#'
#' @export

load_prodes <- function(dataset, raw_data = FALSE,
                        language = "eng") {
  ###########################
  ## Bind Global Variables ##
  ###########################

  year <- municipio <- cod_ibge <- estado <- area_km2 <- increment <- NULL
  municipality <- municipality_code <- state <- deforestation <- desmatamento2000 <- NULL

  #############################
  ## Define Basic Parameters ##
  #############################

  param <- list()
  param$source <- "prodes"
  param$dataset <- dataset
  param$raw_data <- raw_data
  param$language <- language

  # check if dataset and time_period are supported

  check_params(param)

  ###################
  ## Download Data ##
  ###################

  ## Column Names come with numbers at the side - we need to clean those

  dat <- external_download(
    dataset = param$dataset,
    source = param$source
  )

  ## Return Raw Data

  if (param$raw_data) {
    return(dat)
  }

  ######################
  ## Data Engineering ##
  ######################

  # keep only deforestation-related variables

  dat <- dat %>%
    janitor::clean_names() %>%
    dplyr::select(
      municipio, cod_ibge, estado, area_km2, desmatamento2000, dplyr::starts_with("incremento")
    )

  # change to long format with increment variable

  dat <- dat %>%
    dplyr::rename("incremento2000" = "desmatamento2000") %>%
    tidyr::pivot_longer(
      dplyr::starts_with("incremento"),
      names_prefix = "incremento",
      names_to = "year",
      values_to = "increment"
    )

  # calculating cumulative deforestation

  dat <- dat %>%
    dplyr::arrange(municipio, year) %>%
    dplyr::mutate(
      deforestation = cumsum(increment),
      .by = municipio
    )

  dat <- dat %>%
    dplyr::mutate(
      increment = dplyr::case_when(
        year == 2000 ~ NA,
        .default = increment
      )
    )

  ################################
  ## Harmonizing Variable Names ##
  ################################

  if (param$language == "eng") {
    dat_mod <- dat %>%
      dplyr::rename(
        "municipality" = municipio,
        "municipality_code" = cod_ibge,
        "state" = estado
      )
  }
  if (param$language == "pt") {
    dat_mod <- dat %>%
      dplyr::rename(
        "ano" = year,
        "cod_municipio" = "cod_ibge",
        "uf" = estado,
        "incremento" = increment,
        "desmatamento" = deforestation
      )
  }

  return(dat_mod)
}