R/annual_summary_from_months.R

In Tomhigg/TamsatTools: Download and process TAMSAT rainfall estimates

#' Annual; Sum Statistics
#'
#' Calculates sum annual rainfall from monthly estimate, will crop output to extent of fill_with
#'
#' @param download_folder Folder containing the downloaded monthly estimate from tamsat_download_all()
#' @param fill_with raster brick output from monthly_summary() to fill missing months
#' @param years Years as interger(s) for which statistics will be calculated
#' @param southern_hemisphere logical, should years be calculated on a march-march cycle to match southern hemisphere rainfall
#


annual_summary_from_months <- function(download_folder, years, fill_with, southern_hemisphere) {

    if (southern_hemisphere==FALSE) {


    all_files <- list.files(path = download_folder, pattern = ".nc", recursive = TRUE, full.names = TRUE)

    annual_summaries <- stack()

    for (i in years) {
        print(i)

        year_list <- grep(pattern = i, x = all_files, value = TRUE)

        year_stack <- raster::stack(sapply(year_list, FUN = raster::raster))

        year_stack <- crop(x = year_stack, y = fill_with)

        if (nlayers(year_stack) < 12) {

            # make data frame of existing months based on month code from file string
            paths.df <- data.frame(Months = substr(x = year_list, start = nchar(year_list) - 4, stop = nchar(year_list) -
                3))
            # make data fram of all months
            months.df <- data.frame(Months = c("01", "02", "03", "04", "05", "06", "07", "08", "09", "10", "11", "12"))
            # dplyr: anti_join return objects not found in both data frames
            missing_months <- dplyr::anti_join(months.df, paths.df, "Months")
            missing_months <- as.numeric(as.character(missing_months$Months))
            # subset the missing rasters from overall annual summary
            missing_months_raster <- raster::subset(x = fill_with, subset = missing_months)
            # add total average to raster stack
            year_stack <- stack(year_stack, missing_months_raster)
        }

        annual_summary <- raster::calc(x = year_stack, fun = sum)

        names(annual_summary) <- paste0("Year_", i)

        annual_summaries <- stack(annual_summaries, annual_summary)

    }

    return(annual_summaries)
    }

    if (southern_hemisphere==TRUE) {


    all_files <- list.files(path = download_folder, pattern = ".nc", recursive = TRUE, full.names = TRUE)

    annual_summaries <- stack()

    for (i in years) {
      print(i)

      #make list of year = i, month= April-December
      year_list_1 <- grep(pattern = i, x = all_files, value = TRUE)

      year_list_1 <- grep(pattern = paste(c("04", "05", "06", "07", "08","09","10", "11","12"),
                                          collapse = "|"), x = year_list_1, value = TRUE)

      #make list of year = i+1, month= Jan-march
      i2 <- i+1
      year_list_2 <- grep(pattern = i2, x = all_files, value = TRUE)
      year_list_2 <- grep(pattern = paste(c("01", "02", "03", "04"),
                                          collapse = "|"), x = year_list_2, value = TRUE)

      year_list <- c(year_list_1, year_list_2 )

      year_stack <- raster::stack(sapply(year_list, FUN = raster::raster))

      year_stack <- crop(x = year_stack, y = fill_with)

      if (nlayers(year_stack) < 12) {

        # make data frame of existing months based on month code from file string
        paths.df <- data.frame(Months = substr(x = year_list, start = nchar(year_list) - 4, stop = nchar(year_list) -
                                                 3))
        # make data fram of all months
        months.df <- data.frame(Months = c("01", "02", "03", "04", "05", "06", "07", "08", "09", "10", "11", "12"))
        # dplyr: anti_join return objects not found in both data frames
        missing_months <- dplyr::anti_join(months.df, paths.df, "Months")
        missing_months <- as.numeric(as.character(missing_months$Months))
        # subset the missing rasters from overall annual summary
        missing_months_raster <- raster::subset(x = fill_with, subset = missing_months)
        # add total average to raster stack
        year_stack <- stack(year_stack, missing_months_raster)
      }

      annual_summary <- raster::calc(x = year_stack, fun = sum)

      names(annual_summary) <- paste0("Year_", i)

      annual_summaries <- stack(annual_summaries, annual_summary)

    }

    return(annual_summaries)
  }


}