R/data_sapwld.R

In ppcadelina/bucs: Bakawans Under Constrained Survey (BUCS) package

#' Saplings and Wildlings data processing
#'
#' Read an excel file with tabs containing number of observed saplings and wildings using plot-method survey.
#' Format of the worksheet should be:
#' First column containing Tags or Codes for Site/Location,
#' Second column are names/numbers of plots
#' First row containing labels for species while
#' Rest of the body should contain the observed number of the species per plot
#'
#'
#'
#'
#' @param excelpath Location of the excel file
#' @param sheet Name/s of the excel tabs where saplings and/or wildlings data are stored
#' @param clustlvl String to categorize at which spatial level of grouping the data analysis should be.
#'
#'
#' @return Outputs include
#' 'm.saps', a data frame for "Saplings" and/or 'm.wilds', a data frame for "Wildlings"
#'
#' @keywords saplings, wildlings
#'
#'
#'
#'
#' @export

data_sapwld<- function(excelpath = excelpath,
                         sheet = sheet,
                         clustlvl = clustlvl){

  # Defines the `%>%` operator to the current environment
  `%>%` <- dplyr::`%>%`

  # Retrieves columns for selected cluster and SITE
  clust.site<- plyr::ddply(m.data,
                           plyr::.(m.data[[clustlvl]], SITE),
                           plyr::summarize,
                           n = dplyr::n_distinct(`PLOT #`))

  ## Removes last column which is unnecessary
  clust.site<- clust.site[,-3]

  # Rename first column back to the cluster label
  colnames(clust.site)[1]<- clustlvl

#####

  ## Checks if "Saplings" are included in the list
  if("Saplings" %in% sheet){

    ## Reads "Sapling" sheet from excel and declares as a data frame
    sap<- as.data.frame(readxl::read_excel(excelpath, sheet="Saplings", col_names = T))

    ## Replaces NAs of first column
    sap[,1]<- zoo::na.locf(sap[,1])

    ## Replace NAs to zeroes
    sap[is.na(sap)] <- 0

    ## Subset data to include numeric data with column sums not equal to zero
    sap.num<- sap %>%
      dplyr::select_if(~ !is.numeric(.) || sum(.) != 0)

    ## Transforms data into a long format for
    sap.melt<- reshape::melt(sap.num, id=c(1,2))

    ## Renames column names of melted data frame so it can merge
    colnames(sap.melt)<- c("SITE", "PLOT", "Species", "count")

    ## Merges melted data frame with cluster data frame by SITE codes
    clust.melt<- merge(clust.site,sap.melt, by="SITE")

    ## Summarizes data per species
    saps<- plyr::ddply(clust.melt,
                       plyr::.(clust.melt[[clustlvl]], Species),
                       plyr::summarize,
                       Counts = sum(count),
                       SE = round(sd(count) / sqrt((length(Species))), digits=2))

    ## Rename first column name to reflect cluster title
    colnames(saps)[1]<- clustlvl

    ## Transforms data frame to long format
    saps.melt<- reshape::melt(saps, id=c(clustlvl, "Species"))

    ## Merging cluster element name with variable name
    saps.melt$variable<- paste(saps.melt[[clustlvl]], saps.melt$variable, sep = "_")

    ## Transforms data frame to wide format to tabulate easier
    saps.spread<- reshape::cast(saps.melt, Species ~ variable, fill = "-")

    ## Outputs "m.saps" from the generated data frame
    assign("m.saps", saps.spread, pos = .GlobalEnv)
  }

  ## Checks if "Wildlings" are included in the list
  if("Wildlings" %in% sheet){

    ## Reads "Wildlings" sheet from excel and declares as a data frame
    wild<- as.data.frame(readxl::read_excel(excelpath, sheet="Wildlings", col_names = T))

    ## Replaces NAs of first column
    wild[,1]<- zoo::na.locf(wild[,1])

    ## Replace NAs to zeroes
    wild[is.na(wild)] <- 0

    ## Subset data to include numeric data with column sums not equal to zero
    wild.num<- wild %>%
              dplyr::select_if(~ !is.numeric(.) || sum(.) != 0)

    ## Transforms data into a long format for
    wild.melt<- reshape::melt(wild.num, id=c(1,2))

    ## Renames column names of melted data frame so it can merge
    colnames(wild.melt)<- c("SITE", "PLOT", "Species", "count")

    ## Merges melted data frame with cluster data frame by SITE codes
    clust.melt<- merge(clust.site,wild.melt, by="SITE")

    ## Summarizes data per species
    wilds<- plyr::ddply(clust.melt,
                        plyr::.(clust.melt[[clustlvl]], Species),
                        plyr::summarize,
                       Counts = sum(count),
                       SE = round(sd(count) / sqrt((length(Species))), digits=2))

    ## Rename first column name to reflect cluster title
    colnames(wilds)[1]<- clustlvl

    ## Transforms data frame to long format
    wilds.melt<- reshape::melt(wilds, id=c(clustlvl, "Species"))

    ## Merging cluster element name with variable name
    wilds.melt$variable<- paste(wilds.melt[[clustlvl]], wilds.melt$variable, sep = "_")

    ## Transforms data frame to wide format to tabulate easier
    wilds.spread<- reshape::cast(wilds.melt, Species ~ variable, fill = "-")


    ## Outputs "m.saps" from the generated data frame
    assign("m.wilds", wilds.spread, pos = .GlobalEnv)
  }

}