inst/shiny/modules/envs_worldclim.R

In wallace: A Modular Platform for Reproducible Modeling of Species Niches and Distributions

# Wallace EcoMod: a flexible platform for reproducible modeling of
# species niches and distributions.
#
# envs_worldclim.R
# File author: Wallace EcoMod Dev Team. 2023.
# --------------------------------------------------------------------------
# This file is part of the Wallace EcoMod application
# (hereafter “Wallace”).
#
# Wallace is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License,
# or (at your option) any later version.
#
# Wallace is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Wallace. If not, see <http://www.gnu.org/licenses/>.
# --------------------------------------------------------------------------
#
envs_worldclim_module_ui <- function(id) {
  ns <- NS(id)
  tagList(
    tags$div(
      title = paste0('Approximate lengths at equator: 10 arcmin = ~20 km, ',
                     '5 arcmin = ~10 km, 2.5 arcmin = ~5 km, 30 arcsec = ~1 km. ',
                     'Exact length varies based on latitudinal position.'),
             selectInput(ns("wcRes"),
                         label = "Select WorldClim bioclimatic variable resolution",
                         choices = list("Select resolution" = "",
                                        "30 arcsec" = 0.5,
                                        "2.5 arcmin" = 2.5,
                                        "5 arcmin" = 5,
                                        "10 arcmin" = 10))), # Check default (No selected parameter)
    checkboxInput(
      ns("doBrick"),
      label = "Save to memory for faster processing and save/load option",
      value = FALSE), # Check default (value = FALSE)
    shinyWidgets::pickerInput(
      "bcSel",
      label = "Select bioclim variables",
      choices = setNames(as.list(paste0('bio', sprintf("%02d", 1:19))),
                         paste0('bio', sprintf("%02d", 1:19))),
      multiple = TRUE,
      selected = paste0('bio', sprintf("%02d", 1:19)),
      options = list(`actions-box` = TRUE)),
    conditionalPanel(
      sprintf("input['%s'] != '0.5'", ns("wcRes")),
      tags$div(
        title = "Apply selection to ALL species loaded",
        checkboxInput(ns("batch"), label = strong("Batch"), value = FALSE) # Check default (value = FALSE)
      )
    ),
    conditionalPanel(
      sprintf("input['%s'] == '0.5'", ns("wcRes")),
      em("Batch option not available for 30 arcsec resolution."), br(), br(),
      strong(
        paste0("Coordinates centroid as reference for tile download. ",
               "You can vizualize the tile in the bottomleft corner on ",
               "the map. All occurrence outside of the purple polygon will ",
               "be removed")), br(), br(),
      textOutput(ns("ctrLatLon"))
    ),
    br(),
    actionButton(ns("goEnvData"), "Load Env Data")
  )
}

envs_worldclim_module_server <- function(input, output, session, common) {

  logger <- common$logger
  bcSel <- common$bcSel
  occs <- common$occs
  spp <- common$spp
  allSp <- common$allSp
  curSp <- common$curSp
  envs.global <- common$envs.global
  mapCntr <- common$mapCntr

  observeEvent(input$goEnvData, {
    # ERRORS ####
    if (is.null(curSp())) {
      logger %>% writeLog(type = 'error',
      "Before obtaining environmental variables, obtain occurrence data in 'Occ Data' component.")
      return()
    }
    # Specify more than 2 variables
    if (length(bcSel()) < 2) {
      logger %>%
        writeLog(
          type = 'error',
          "Select more than two variables.")
      return()
    }

    # loop over all species if batch is on
    if (input$batch == FALSE) spLoop <- curSp() else spLoop <- allSp()

    # PROCESSING ####
    for (sp in spLoop) {
      # FUNCTION CALL ####
      if (input$wcRes != 0.5) {
        wcbc <- envs_worldclim(input$wcRes, bcSel(), doBrick = input$doBrick,
                               logger = logger)
        req(wcbc)
        envs.global[[paste0("wcbc_", sp)]] <- wcbc
      } else {
        wcbc <- envs_worldclim(input$wcRes, bcSel(), mapCntr(), input$doBrick, logger)
        req(wcbc)
        envs.global[[paste0("wcbc_", sp)]] <- wcbc
      }
      # get environmental variable values per occurrence record
      withProgress(message = paste0("Extracting environmental values for occurrences of ",
                                    spName(sp), "..."), {
        occs.xy <- spp[[sp]]$occs[, c('longitude', 'latitude')]
        occsEnvsVals <- as.data.frame(raster::extract(wcbc, occs.xy, cellnumbers = TRUE))
      })

      # remove occurrence records with NA environmental values
      remOccs <- remEnvsValsNA(spp[[sp]]$occs, occsEnvsVals, sp, logger)
      if (!is.null(remOccs)) {
        spp[[sp]]$occs <- remOccs$occs
        occsEnvsVals <- remOccs$occsEnvsVals
      } else {
        # When remOccs is null, means that all localities have NAs
        return()
      }

      logger %>% writeLog(hlSpp(sp), "Worldclim variables ready to use.")

      # LOAD INTO SPP ####
      # add reference to WorldClim bioclim data
      spp[[sp]]$envs <- paste0("wcbc_", sp)

      # add columns for env variable values for each occurrence record
      if (!any(names(occsEnvsVals) %in% names(spp[[sp]]$occs))) {
        spp[[sp]]$occs <- cbind(spp[[sp]]$occs, occsEnvsVals)
      } else {
        shaEnvNames <- names(occsEnvsVals)[names(occsEnvsVals) %in% names(spp[[sp]]$occs)]
        spp[[sp]]$occs <- spp[[sp]]$occs %>% dplyr::mutate(occsEnvsVals[shaEnvNames])
      }


      # METADATA ####
      spp[[sp]]$rmm$data$environment$variableNames <- names(wcbc)
      spp[[sp]]$rmm$data$environment$yearMin <- 1970
      spp[[sp]]$rmm$data$environment$yearMax <- 2000
      spp[[sp]]$rmm$data$environment$resolution <- paste(round(raster::res(wcbc)[1] * 60, digits = 2), "minutes")
      spp[[sp]]$rmm$data$environment$extent <- as.character(raster::extent(wcbc))
      spp[[sp]]$rmm$data$environment$sources <- 'WorldClim v2.1'
      spp[[sp]]$rmm$data$environment$projection <- as.character(raster::crs(wcbc))

      spp[[sp]]$rmm$code$wallace$wcRes <- input$wcRes
      spp[[sp]]$rmm$code$wallace$bcSel <- bcSel()
      spp[[sp]]$rmm$code$wallace$mapCntr <- mapCntr()
      spp[[sp]]$rmm$code$wallace$wcBrick <- input$doBrick
    }

    common$update_component(tab = "Results")
  })

  # text showing the current map center
  output$ctrLatLon <- renderText({
    req(curSp(), occs())
    glue::glue('Using coordinate centroid {join(mapCntr())}')
  })

  output$envsPrint <- renderPrint({
    req(curSp(), spp[[curSp()]]$envs)
    envs.global[[spp[[curSp()]]$envs]]
  })

  return(list(
    save = function() {
      list(
        wcRes = input$wcRes,
        wcBrick = input$doBrick,
        bcSel = bcSel()
      )
    },
    load = function(state) {
      updateSelectInput(session, "wcRes", selected = state$wcRes)
      updateCheckboxInput(session, "doBrick", value = state$doBrick)
      shinyWidgets::updatePickerInput(session, "bcSel", selected = state$bcSel)
    }
  ))
}

envs_worldclim_module_result <- function(id) {
  ns <- NS(id)
  # Result UI
  verbatimTextOutput(ns("envsPrint"))
}

envs_worldclim_module_map <- function(map, common) {
  # Map logic
  occs <- common$occs
  mapCntr <- c(mean(occs()$longitude), mean(occs()$latitude))
  lon_tile <- seq(-180, 180, 30)
  lat_tile <- seq(-60, 90, 30)
  map %>% clearAll() %>%
    addCircleMarkers(data = occs(), lat = ~latitude, lng = ~longitude,
                     radius = 5, color = 'red', fill = TRUE, fillColor = "red",
                     fillOpacity = 0.2, weight = 2, popup = ~pop) %>%
    addRectangles(lng1 = lon_tile[sum(lon_tile <= mapCntr[1])],
                  lng2 = lon_tile[sum(lon_tile <= mapCntr[1])] + 30,
                  lat1 = lat_tile[sum(lat_tile <= mapCntr[2])],
                  lat2 = lat_tile[sum(lat_tile <= mapCntr[2])] + 30,
                  color = "purple", group = "30 arcsec tile") %>%
    hideGroup("30 arcsec tile") %>%
    addLayersControl(overlayGroups = "30 arcsec tile", position = "bottomleft",
                     options = layersControlOptions(collapsed = FALSE))
}

envs_worldclim_module_rmd <- function(species) {
  # Variables used in the module's Rmd code
  list(
    envs_worldclim_knit = !is.null(species$rmm$code$wallace$wcRes),
    wcRes_rmd = species$rmm$code$wallace$wcRes,
    bcSel_rmd = printVecAsis(species$rmm$code$wallace$bcSel),
    mapCntr_rmd = printVecAsis(species$rmm$code$wallace$mapCntr),
    wcBrick_rmd = species$rmm$code$wallace$wcBrick
  )
}