Select Git revision
Chelsa_Download_functions.R
Jentsch, Helge authored
230814: Update Chelsa Timeseries Download. Note: Stacking is not available for Chelsa.timeseries.download()-function at the moment.
Code owners
Assign users and groups as approvers for specific file changes. Learn more.
Chelsa_Download_functions.R 90.01 KiB
#'@title Deprecated function for downloading the CHELSA climate dataset (1979-2013)
#'@author Helge Jentsch
#'@description This function supports the download, pre-processing and management of CHELSA climate data comprising of monthly precipitation sums in mm, monthly temperature (average, minimum, maximum) in degrees Celsius, and annual chracteristics (19 bioclimatic variables). The spatial resolution of the downloaded data is 30 arc-seconds.\cr To allow pre-processing, clipping and buffering, conversion to ASCII-grids and stacking options are included.\cr Optional an output of a .bib-file of the cited literature can be retrieved.\cr For user convenience, saving directories will be created automatically. Also options to "zip" and/or delete the RAW-files are included.
#'
#'@param save.location string. Input where the datasets will be saved. \cr Default: Working Directory.
#'@param parameter string (vector). Input of parameters which should be downloaded. \cr Default: \code{c("prec", "temp", "tmax", "tmin", "bio")}
#'@param bio.var integer (vector). Input which monthly data should be downloaded. Only applicable to BIOCLIM variables. For further information see: \url{http://chelsa-climate.org/bioclim/}. \cr Default: \code{c(1:19)}
#'@param month.var integer (vector). Input which monthly data should be downloaded. Only applicable to precipitation and temperature (average, maximum, minimum). \cr Default: \code{c(1:12)}
#'@param version.var string (vector). Input which version of the dataset should be downloaded. Multiple selection is possible. \cr Default: \code{c("1.2")}
#'@param clipping logical. Input whether the downloaded data should be clipped.\cr If \code{FALSE}: clip.shapefile, buffer, clip.extent will be ignored. \cr Default: \code{FALSE}
#'@param clip.shapefile string. Input which shapefile should be used for clipping. \cr Default: \code{NULL}
#'@param clip.extent numeric (vector). Input vector with four numeric values. This is following the input order c("xleft", "xright", "ybottom", "ytop"). \cr Default: \code{c(-180, 180, -90, 90)}
#'@param buffer numeric. Input of decimal degrees of buffer around the shapefile and/or extent. \cr Default: \code{0}
#'@param convert.files.to.asc logical. Input whether files should be converted into the ASCII format.\cr If \code{TRUE}: a new subdirectory is created and the rawdata is saved there. If \code{clipping} is \code{TRUE}: the clipped raster files are also saved as ASCII grids. \cr Default: \code{FALSE}
#'@param stacking.data logical. Input whether the downloaded data should be stacked as a netCDF-rasterstack. \cr Default: \code{FALSE}
#'@param combine.raw.zip logical. Should the downloaded raw-data be "zipped". \cr Default: \code{FALSE}
#'@param delete.raw.data logical. Should the downloaded raw-data be deleted.\cr If \code{combine.raw.zip} is \code{TRUE}: raw-data is still available in the zipped file. \cr Default: \code{FALSE}
#'@param save.bib.file logical. Whether a BibTex-citation file of the dataset should be provided in the Working directory. \cr Default: \code{TRUE}
#'@return CHELSA climate datasets for the period of 1979 - 2013
#'
#'@note Please note that the downloaded data for temperature and the therefore also the first eleven bioclim-variables are processed to °C with one significant decimal without offset and factor. Processing and conversion to other file-formats on a global dataset may take some time.
#'
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. "Climatologies at high resolution for the earth's land surface areas". In: _Scientific Data_ 4.1 (Sep. 2017). DOI: 10.1038/sdata.2017.122. <URL: https://doi.org/10.1038/sdata.2017.122>.
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. _Data from: Climatologies at high resolution for the earth's land surface areas_. En. 2018. DOI: 10.5061/DRYAD.KD1D4. <URL: http://datadryad.org/stash/dataset/doi:10.5061/dryad.kd1d4>.
#'
#'@examples
#' \dontrun{
#' # Bioclim
#' Chelsa.Clim.download(parameter = "bio", bio.var = c(1,19))
#' # Precipitation
#' Chelsa.Clim.download(parameter = "prec", month.var = c(1,12))
#' }
#'
#'@import stringr
#'@import RCurl
#'@import ncdf4
#'@import terra
#'@importFrom utils unzip download.file setTxtProgressBar txtProgressBar
#'
#'
Chelsa.Clim.download.deprecated<- function(save.location = "./",
parameter = c("prec", "temp", "tmax", "tmin", "bio"),
bio.var = c(1:19),
month.var = c(1:12),
version.var = c("1.2"),
clipping = FALSE,
clip.shapefile = NULL,
clip.extent = c(-180, 180, -90, 90),
buffer = 0,
convert.files.to.asc = FALSE,
stacking.data = FALSE,
combine.raw.zip = FALSE,
delete.raw.data = FALSE,
save.bib.file = TRUE){
gc()
call.time <- stringr::str_replace_all(stringr::str_replace_all(paste0(Sys.time()), pattern = ":", replacement = "-"), pattern = " ", replacement = "_")
# initial check -----------------------------------------------------------
# normalize Path for easier application later
save.location <- normalizePath(save.location, winslash = "/")
# Check which parameters are put in and if the connected
# month/bio-variables are correctly input
if(is.element("prec", parameter)|is.element("temp", parameter)|
is.element("tmax", parameter)|is.element("tmin", parameter)){
# if month.var is just a single numeric input it is here casted into
# a vector for comparabilities
month.var <- c(month.var)
# if there is not a numeric input -> prompt error
if(!is.numeric(month.var)) stop()
# Padding of "one-digit" months with a 0
month.var <- str_pad(month.var, 2, 'left', pad = "0") }
# analog to the if-clause before - here the parameter bio.var is checked.
if(is.element("bio", parameter)){
bio.var <- c(bio.var)
if(!is.numeric(bio.var)) stop()
bio.var <- str_pad(bio.var, 2, 'left', pad = "0")
}
# Download: 1. work through all parameters -----------------------------------
for(i in parameter){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
# create intermediate strings for later use
interm <- switch(i,
"prec" = "prec/",
"temp" = "tmean/",
"tmax" = "tmax/",
"tmin" = "tmin/",
"bio" = "bio/",
stop())
variable.numbers <- switch(i,
"bio" = bio.var,
"tmin" = month.var,
"tmax" = month.var,
"temp" = month.var,
"prec" = month.var,
stop())
# if not already created, create new directory
if (!dir.exists(paste0(save.location, "/", i))){
dir.create(paste0(save.location, "/", i))
}
# set the 1. order temporal save location to this directory
# 1. Order -> parameter!
temp.save.location <- paste0(save.location, "/", i, "/")
# Add "10" after parameter string for all parameters except precipitation
if(i != "prec"){
i <- paste0(i, "10")
}
## Download: 2. Work through versions as given as initial parameter ---------
for (version in version.var) {
if(version == "1.1") next
# create version string
vers <- switch(version,
"1.1" = "",
"1.2" = "_V1.2",
stop())
### Download: 3. Preparation of the save location 2. order -----------------
vers_path <- str_remove(vers, pattern = "_")
# set the 2nd order temporal save location to create a
# managable directory tree in the 1st order directory
temp.temp.save.location <- paste0(temp.save.location,
stringr::str_replace_all(interm,
pattern = "/",
"_"),
vers_path, "/")
# if not already created, create new directory
if(!dir.exists(temp.temp.save.location)){
dir.create(temp.temp.save.location)
}
# normalize the path to make it work more easily
# temp.temp.save.location <- normalizePath(temp.temp.save.location,
# winslash = "/")
##### Download: 4. Check if bio is not requested -----------------------------
if(i != "bio10"){
# should years be added? necessary for the download function
years <- switch(i,
"prec" = "",
"_1979-2013")
###### Download: 5. Work through every requested month ----------------------
for(month in 1:length(month.var)){
# create a variable that is a placeholder for the produced output.
dest.temp <- paste0(temp.temp.save.location, "CHELSA_", i, "_",
month.var[month], years, vers, "_land.tif")
# check if the output already exists
# - if yes, the download will just be skipped.
# which makes a huge runtime difference!
if(!file.exists(dest.temp)){
# create a variable for the later requested Download-URL to avoid
# requireing multiple changes, if the link changes.
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/climatologies/",
interm, "CHELSA_", i, "_", month.var[month], years,
vers, "_land.tif")
# check if URL is available
if(url.exists(URL.temp)){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
# download file to save location
download.file(url = URL.temp,
destfile = dest.temp,
# overwrite is TRUE otherwise a error is caused
overwrite = TRUE,
# From the description file:
# The choice of binary transfer (mode = "wb" or "ab")
# is important on Windows, since unlike Unix-alikes
# it does distinguish between text and binary files and
# for text transfers changes
# \n line endings to \r\n (aka ‘CRLF’).
mode = 'wb',
# to show progression bar
quiet = FALSE,
cacheOK = FALSE)
if(i != "prec"){
raster.temp <- terra::rast(dest.temp)
gc()
raster.temp <- terra::clamp(raster.temp, lower = -1000,
useValues = FALSE)
raster.temp <- process.raster.int.doub(raster.temp)
gc()
terra::writeRaster(raster.temp,
dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}
}else{
# Error message if file is not available
warning(paste0("File does not exist. Did not download: \n",
URL.temp, "\n\n"),
# call is printed later or within the "warnings()"
call. = TRUE, immediate. = FALSE)
}
}
###### if-clause to include the post-download manipulations of the files -------
# if clause checks for:
# Is the month the last month of the demanded months
# Are .tif raster files downloaded?
if(month.var[month] == month.var[length(month.var)] &
length(list.files(temp.temp.save.location,
pattern = ".tif", include.dirs = FALSE)) != 0){
# if clipping is TRUE ...
if(clipping == TRUE){
# the function "clipping.tif" (found in the auxiliary.R-File)
# is executed. The clip.save.location is the same location as the
# "current" save location
clipping.tif(clip.save.location = temp.temp.save.location,
# the clip-shapefile is passed
# default "NULL" does not produce error
clip.shapefile = clip.shapefile,
# Clip.extent is passed
# default "c(-180, 180, -90, 90)" does not produce errors
# simply clips the whole world.
clip.extent = clip.extent,
# buffer is passed
# default: 0. Unit is arc-degrees
buffer = buffer,
# conversion to ASCII format here integrated into the
# clipping function. Since it can be assumed that
# they should be converted lateron anyway.
convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
# if converting.files.to.asc is TRUE ...
if(convert.files.to.asc == TRUE){
# the function "convert.to.asc" (found in the auxiliary.R-File)
# is executed. The save.location is the same location as the
# "current" save location. Also another new subdirectory will
# be created with the name "ASCII" .
convert.to.asc(save.location = temp.temp.save.location,
time.stamp.var = call.time)
}
# if stacking.data is TRUE ...
if(stacking.data == TRUE){
# the function "stacking.downloaded.data"
# (found in the auxiliary.R-File) is executed.
# The save.location is the same location as the
# "current" save location.
if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
time.stamp.var = call.time)
}
}
# if combine.raw.zip is TRUE ...
if(combine.raw.zip == TRUE){
# the function "combine.raw.in.zip"
# (found in the auxiliary.R-File) is executed.
# The save.location is the same location as the
# "current" save location. The name of the zip-file is also
# passed with the current parameter in it.
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSAClim_", i, ""),
time.stamp.var = call.time)
}
# if delete.raw.data is TRUE ...
if(delete.raw.data == TRUE){
# All .tif raster files in the current 2nd order subdirectory are
# unlinked (deleted).
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE,
full.names = T), force = TRUE)
}
# delete all temporary files
unlink(list.files(tempdir(), recursive = T, full.names =T))
}
}
}else{
# analog to other parameters
# just that bio.var instead of month.var is used.
for(bio in bio.var){
if(version == "1.1"){
warning("CHELSA BioClim Version 1.1 not available!")
next
}
dest.temp <- paste0(temp.temp.save.location,
"CHELSA_", i, "_", bio, vers, ".tif")
if(!file.exists(dest.temp)){
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/climatologies/",
interm, "CHELSA_", i, "_", bio, ".tif")
# check if URL is available
if(url.exists(URL.temp)){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
# download file to save location
download.file(url = URL.temp,
destfile = dest.temp,
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
if(bio <= 11){
raster.temp <- terra::rast(dest.temp)
# raster.values <- values(raster.temp)
# raster.values[raster.values==-32768] <- NA
# values(raster.temp) <- raster.values
# rm(raster.values)
gc()
raster.temp <- terra::clamp(raster.temp, lower = -1000,
useValues = FALSE)
raster.temp <- process.raster.int.doub(raster.temp)
terra::writeRaster(raster.temp,
dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}else{
raster.temp <- terra::rast(paste0(temp.temp.save.location, "CHELSA_",
i, "_", bio, vers, ".tif"))
gc()
raster.temp <- terra::clamp(raster.temp, lower = -1000,
useValues = FALSE)
gc()
terra::writeRaster(raster.temp,
paste0(temp.temp.save.location, "CHELSA_",
i, "_", bio, vers, ".tif"),
overwrite = TRUE)
rm(raster.temp)
gc()
}
}else{
# Error message
warning(paste0("File does not exist. Did not download: \n", URL.temp),
call. = TRUE, immediate. = FALSE)
}
}
if(bio == bio.var[length(bio.var)] & length(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent,
buffer = buffer,
convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(save.location = temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){
if(clipping == TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSAClim_", i, ""),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T),
force = TRUE)
}
# delete all temporary files
unlink(list.files(tempdir(), recursive = T, full.names =T))
}
}
}
# if no subdirectories and files are to be found in the 2nd order
# subdirectory the directory will be deleted for a better overview of
# given data.
if(length(list.files(temp.temp.save.location,
include.dirs = TRUE)) == 0){
unlink(str_sub(temp.temp.save.location, 1,
end = str_length(temp.temp.save.location)-1),
force = TRUE, recursive = TRUE)
}
} # version for-loop END
# Download END
} # Parameters for-loop end
# Saving BIB File
# if save.bib.file is TRUE, the save.citation function will be called.
# it saves a .bib-file of the downloaded data-set on the highest level of
# the save.location. If the user's working directory is the desktop
# and no other save.location is specified initially, on the desktop.
if(save.bib.file == TRUE) {
save.citation(save.location = save.location, dataSetName = "CHELSA")
}
# delete all temporary files
unlink(list.files(tempdir(), recursive = T, full.names =T))
}
#'@title Function for downloading CHELSA CMIP 5 future climatologies for the years 2041-2060 and 2061-2080
#'@author Helge Jentsch
#'@description This function supports the download of CHELSA CMIP5 future climate scenarios comprising of monthly precipitation sums in mm, monthly temperature (average, minimum, maximum) in degrees Celsius, and annual chracteristics (19 bioclimatic variables).\cr To allow pre-processing, clipping and buffering, conversion to ASCII-grids and stacking options are included.\cr Optional an output of a .bib-file of the cited literature can be retrieved.\cr For user convenience, saving directories will be created automatically. Also options to "zip" and/or delete the RAW-files are included.
#'
#'@details "The downscaled data has been produced using climatological aided interpolation based on the 1979-2013 reference climatologies from CHELSA." (CHELSA Climate 2020: \url{http://chelsa-climate.org/future/})
#'
#'@note Please note that the downloaded data for temperature and the therefore also the first eleven bioclim-variables are processed to °C with one significant decimal without offset and factor. Processing and conversion to other file-formats on a global dataset may take some time.\cr For some of the datasets not all models and rcps are available. For the ones that are not supported the data will not be downloaded and a warning will be prompted. See parameter \code{model.var} for more information or check the website of CHELSA Climate (\url{http://chelsa-climate.org/future/}). Please note, that the downloaded data for temperature and the therefore also the first eleven bioclim-variables are processed to °C without offset and factor. Processing and conversion to other file-formats on a global dataset may take some time.
#'
#'@param save.location string. Input where the datasets should be saved. \cr Default: Working Directory.
#'@param parameter string (vector). Input of parameters which should be downloaded. \cr Default: \code{c("prec", "temp", "tmax", "tmin", "bio")}
#'@param bio.var integer (vector). Input which monthly data should be downloaded. Only applicable to BIOCLIM variables. For further information see: \url{http://chelsa-climate.org/bioclim/}. \cr Default: \code{c(1:19)}
#'@param month.var integer (vector). Input which monthly data should be downloaded. Only applicable to Precipitation and Temperature (average, maximum, minimum). \cr Default: \code{c(1:12)}
#'@param emission.scenario.var string (vector). Input which emission scenario dataset should be downloaded. Provided are the representative concentration pathways (RCP) 2.6, 4.5, 6.0, and 8.5.\cr Default: \code{c("rcp26", "rcp45", "rcp60", "rcp85")}
#'@param time.interval.var string (vector). Input for which time interval data should be downloaded. CHELSA provides downscaled CMIP5 climatologies for 2050 and 2070. Multiple inputs possible.\cr Default: \code{c("2041-2060", "2061-2080")}
#'@param model.var string (vector). Input which future model dataset should be downloaded. For more information see: \url{http://chelsa-climate.org/future/}.\cr For some of the datasets not all downloads are available. For the ones that are not supported the data will not be downloaded and a warning will be prompted. For an overview please try "warnings()" after execution. \cr Default: \code{c("ACCESS1-0", "bcc-csm1-1", "BNU-ESM", "CanESM2", "CCSM4", "CESM1-BGC", } \cr \code{"CESM1-CAM5", "CMCC-CESM", "CMCC-CM", "CMCC-CMS", "CNRM-CM5", "CSIRO-Mk3-6-0",} \cr \code{ "CSIRO-Mk3L-1-2", "EC-EARTH", "FGOALS-g2", "FIO-ESM", "GFDL-CM3", "GFDL-ESM2G", } \cr \code{"GFDL-ESM2M","GISS-E2-H", "GISS-E2-H-CC", "GISS-E2-R", "GISS-E2-R-CC", "HadGEM2-AO",} \cr \code{"HadGEM2-CC", "HadGEM2-ES", "inmcm4", "IPSL-CM5A-LR", "IPSL-CM5A-MR","MIROC-ESM",} \cr \code{"MIROC-ESM-CHEM","MIROC5", "MPI-ESM-LR", "MPI-ESM-MR", "MRI-CGCM3", "MRI-ESM1",} \cr \code{ "NorESM1-M","NorESM1-ME")}
#'@param clipping logical. Input whether the downloaded data should be clipped.\cr If \code{FALSE} \code{clip.shapefile}, buffer, clip.extent will be ignored. \cr Default: \code{FALSE}
#'@param clip.shapefile string. Input which shapefile should be used for clipping. \cr Default: \code{NULL}
#'@param clip.extent numeric (vector). Input vector with four numeric values. This is following the input order c("xleft", "xright", "ybottom", "ytop").\cr Default: \code{c(-180, 180, -90, 90)}
#'@param buffer numeric. Input of decimal degrees of buffer around the shapefile and/or extent. \cr Default: \code{0}
#'@param convert.files.to.asc logical. Input whether files should be converted into the ASCII format.\cr If \code{TRUE}: a new subdirectory is created and the rawdata is saved there. If \code{clipping} is \code{TRUE}: the clipped raster files are also saved as ASCII grids. \cr Default: \code{FALSE}
#'@param stacking.data logical. Input whether the downloaded data should be stacked as a netCDF-rasterstack. \cr Default: \code{FALSE}
#'@param combine.raw.zip logical. Should the downloaded raw-data be "zipped". \cr Default: \code{FALSE}
#'@param delete.raw.data logical. Should the downloaded raw-data be deleted. If the \code{combine.raw.zip} is \code{TRUE}: raw-data is still available in the zipped file. \cr Default: \code{FALSE}
#'@param save.bib.file logical. Whether a BibTex-citation file of the dataset should be provided in the Working directory. \cr Default: \code{TRUE}
#'
#'@return Downscaled CHELSA CMIP5 climatologies for 2050 and 2070.
#'
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. "Climatologies at high resolution for the earth's land surface areas". In: _Scientific Data_ 4.1 (Sep. 2017). DOI: 10.1038/sdata.2017.122. <URL: https://doi.org/10.1038/sdata.2017.122>.
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. _Data from: Climatologies at high resolution for the earth's land surface areas_. En. 2018. DOI: 10.5061/DRYAD.KD1D4. <URL: http://datadryad.org/stash/dataset/doi:10.5061/dryad.kd1d4>.
#'
#'@examples
#' \dontrun{
#' # Bioclim
#' Chelsa.CMIP_5.download(parameter = "bio",
#' bio.var = c(1,19),
#' emission.scenario.var = "rcp26",
#' time.interval.var = "2041-2060",
#' model.var = "MPI-ESM-LR")
#' # Precipitation
#' Chelsa.CMIP_5.download(parameter = "prec",
#' month.var = c(1,12),
#' emission.scenario.var = "rcp26",
#' time.interval.var = "2041-2060",
#' model.var = "MPI-ESM-LR")
#' }
#'
#'@import stringr
#'@import RCurl
#'@import ncdf4
#'@import terra
#'@import httr
#'@importFrom utils unzip download.file setTxtProgressBar txtProgressBar
#'
#'
#'@export
Chelsa.CMIP_5.download <- function(save.location = "./",
parameter = c("prec", "temp", "tmax", "tmin", "bio"),
bio.var = c(1:19),
month.var = c(1:12),
emission.scenario.var = c("rcp26", "rcp45",
"rcp60", "rcp85"),
time.interval.var = c("2041-2060", "2061-2080"),
model.var = c("ACCESS1-0", "bcc-csm1-1",
"BNU-ESM", "CanESM2", "CCSM4",
"CESM1-BGC", "CESM1-CAM5",
"CMCC-CESM", "CMCC-CM",
"CMCC-CMS", "CNRM-CM5",
"CSIRO-Mk3-6-0", "CSIRO-Mk3L-1-2",
"EC-EARTH", "FGOALS-g2",
"FIO-ESM", "GFDL-CM3",
"GFDL-ESM2G", "GFDL-ESM2M",
"GISS-E2-H", "GISS-E2-H-CC",
"GISS-E2-R", "GISS-E2-R-CC",
"HadGEM2-AO", "HadGEM2-CC",
"HadGEM2-ES", "inmcm4",
"IPSL-CM5A-LR", "IPSL-CM5A-MR",
"MIROC-ESM", "MIROC-ESM-CHEM",
"MIROC5", "MPI-ESM-LR",
"MPI-ESM-MR", "MRI-CGCM3",
"MRI-ESM1", "NorESM1-M",
"NorESM1-ME"),
clipping = FALSE,
clip.shapefile = NULL,
clip.extent = c(-180, 180, -90, 90),
buffer = 0,
convert.files.to.asc = FALSE,
stacking.data = FALSE,
combine.raw.zip = FALSE,
delete.raw.data = FALSE,
save.bib.file = TRUE
){
# requireNamespace("stringr")
# requireNamespace("RCurl")
# requireNamespace("ncdf4")
gc()
save.location <- normalizePath(save.location, winslash = "/")
call.time <- stringr::str_replace_all(stringr::str_replace_all(paste0(Sys.time()), pattern = ":", replacement = "-"), pattern = " ", replacement = "_")
# the build of this function is very similar to the Chelsa.Clim.download function.
if(is.element("prec", parameter)|is.element("temp", parameter)|
is.element("tmax", parameter)|is.element("tmin", parameter)){
if(!is.numeric(month.var)) stop()
}
if(is.element("bio", parameter)){
if(!is.numeric(bio.var)) stop()
}
for(i in parameter){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
variable.numbers <- switch(i,
"bio" = bio.var,
"tmin" = month.var,
"tmax" = month.var,
"temp" = month.var,
"prec" = month.var,
stop())
for (time.interval in time.interval.var) {
interm <- switch(i,
"prec" = paste0("cmip5/", time.interval,
"/prec/CHELSA_pr_mon_"),
"temp" = paste0("cmip5/", time.interval,
"/temp/CHELSA_tas_mon_"),
"tmax" = paste0("cmip5/", time.interval,
"/tmax/CHELSA_tasmax_mon_"),
"tmin" = paste0("cmip5/", time.interval,
"/tmin/CHELSA_tasmin_mon_"),
"bio" = paste0("cmip5/", time.interval,
"/bio/CHELSA_bio_mon_"),
stop())
# create new directory
if(!dir.exists(paste0(save.location, "/", i))){
dir.create(paste0(save.location, "/", i))
}
temp.save.location <- paste0(save.location, "/", i, "/") # print(temp.save.location)
for (model in model.var) {
for (emission.scenario in emission.scenario.var) {
temp.temp.save.location <- paste0(temp.save.location,
stringr::str_replace_all(interm,
pattern = "/",
"_"),
model,"_",
emission.scenario, "/")
if(!dir.exists(str_sub(temp.temp.save.location,
end = str_length(temp.temp.save.location)-1))){
dir.create(str_sub(temp.temp.save.location,
end = str_length(temp.temp.save.location)-1))
}
if(i != "bio"){
for(month in month.var){
dest.temp <- paste0(temp.temp.save.location, "CHELSA_", model,
"_", emission.scenario, "_", i, "_", month,
"_", time.interval, ".tif")
if(!file.exists(dest.temp)){
URL.temp <- paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/",
interm, model, "_", emission.scenario,
# "_r1i1p1_g025.nc_", month, "_", time.interval,
"_V1.2.tif")
if(!http_error(URL.temp)){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
download.file(url = URL.temp,
destfile = dest.temp,
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
if(i != "prec"){
gc()
raster.temp <- terra::rast(dest.temp)
raster.temp <- terra::clamp(raster.temp, lower = -1000,
useValues = FALSE)
gc()
raster.temp <- process.raster.int.doub(dest.temp)
gc()
terra::writeRaster(x = raster.temp,
filename = dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}
}else{
warning(paste0("File does not exist. Did not download: \n",
URL.temp), call. = TRUE, immediate. = FALSE)
}
}
if(month.var[month] == month.var[length(month.var)] &
length(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent,
buffer = buffer,
convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(save.location = temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){ if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSACMIP5_", i, ""),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T), force = TRUE)
}
}
}
}else{
for(bio in bio.var){
dest.temp <- paste0(temp.temp.save.location, "CHELSA_", model,
"_", emission.scenario, "_", i, "_", bio,
"_", time.interval,".tif")
if(!file.exists(dest.temp)){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
URL.temp <- paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/",
interm, model, "_",
emission.scenario, "_r1i1p1_g025.nc_",
bio, "_", time.interval, "_V1.2.tif")
if(!http_error(URL.temp)){
download.file(url = URL.temp,
destfile = dest.temp,
overwrite = FALSE,
mode = 'wb',
quiet = FALSE)
raster.temp <- terra::rast(dest.temp)
# raster.values <- values(raster.temp)
# raster.values[raster.values==-32768] <- NA
# values(raster.temp) <- raster.values
# rm(raster.values); gc()
raster.temp <- terra::clamp(raster.temp,
lower = -1000,
useValues = FALSE)
gc()
if(bio <= 11){
gc()
raster.temp <- process.raster.int.doub(raster.temp)
}
terra::writeRaster(x = raster.temp,
filename = dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}else{
warning(paste0("File does not exist. Did not download: \n", URL.temp),
call.=TRUE, immediate. = FALSE)
}
}
if(bio == bio.var[length(bio.var)] &
length(list.files(temp.temp.save.location, pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent,
buffer = buffer,
convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(save.location = temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){
if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSACMIP5_", i, ""),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T), force = TRUE)
}
}
}
}
if(length(list.files(temp.temp.save.location,
include.dirs = TRUE)) == 0){
unlink(str_sub(temp.temp.save.location, 1,
end = str_length(temp.temp.save.location)-1),
force = TRUE,
recursive = TRUE)
}
}
}
}
}
# Saving BIB File ---------------------------------------------------------
if(save.bib.file == TRUE) save.citation(save.location = save.location, dataSetName = "CHELSA")
}
#'@title Function for downloading CHELSA Last Glacial Maximum datasets
#'@author Helge Jentsch
#'@description This function supports a download of the CHELSA Last Glacial Maximum Climate datasets (21.000 BP). This includes monthly precipitation sums in mm, monthly temperature (average, maximum, minimum) in degree Celsius, annual characteristics (19 bioclimatic parameters), and a global digital elevation model. For further information, please regard \url{http://chelsa-climate.org/last-glacial-maximum-climate/}.\cr To allow pre-processing, clipping and buffering, conversion to ASCII-grids and stacking options are included.\cr Optional an output of a .bib-file of the cited literature can be retrieved.\cr For user convenience, saving directories will be created automatically. Also options to "zip" and/or delete the RAW-files are included.
#'
#'@details "The CHELSA LGM data is based on a implementation of the CHELSA algorithm on PMIP3 data." (CHELSA Climate 2020: \url{http://chelsa-climate.org/last-glacial-maximum-climate/})
#'
#'@note Please note that the downloaded data for temperature and the therefore also the first eleven bioclim-variables are processed to °C with one significant decimal without offset and factor. Processing and conversion to other file-formats on a global dataset may take some time.\cr For some of the datasets not all models are available. For the ones that are not supported the data will not be downloaded and a warning will be prompted. See parameter \code{model.var} for more information or check the website of CHELSA Climate (\url{http://chelsa-climate.org/last-glacial-maximum-climate/}).
#'
#'
#'@param save.location string. Input where the datasets should be saved. \cr Default: Working Directory.#'@param parameter string (vector). Input of parameters which should be downloaded. \cr Default: \code{c("prec", "temp", "tmax", "tmin", "bio")}
#'@param bio.var integer (vector). Input which bioclim data should be downloaded. Only applicable to BIOCLIM variables. For further information see: \url{http://chelsa-climate.org/bioclim/}. \cr Default: \code{c(1:19)}
#'@param month.var integer (vector). Input which monthly data should be downloaded. Only applicable to Precipitation and Temperature (average, maximum, minimum). \cr Default: \code{c(1:12)}
#'@param model.var string (vector). Input which future model dataset should be downloaded. For more information see: \url{http://chelsa-climate.org/last-glacial-maximum-climate/}. For some of the datasets not all downloads are available. For the ones that are not supported the data will not be downloaded and a warning will be prompted. For an overview please try "warnings()" after execution. \cr Default: \code{c("CCSM4", "MRI-CGCM3", "CNRM-CM5", } \cr \code{"FGOALS-g2", "IPSL-CM5A-LR", "MIROC-ESM", "MPI-ESM-P")}
#'@param download.dem logical. Input whether a LGM digital elevation model should be downloaded. \cr Default: \code{FALSE}
#'@param clipping logical. Input whether the downloaded data should be clipped.\cr If \code{FALSE}; clip.shapefile, buffer, clip.extent will be ignored. \cr Default: \code{FALSE}
#'@param clip.shapefile string. Input which shapefile should be used for clipping. \cr Default: \code{NULL}
#'@param clip.extent numeric (vector). Input vector with four numeric values. This is following the input order c("xleft", "xright", "ybottom", "ytop").\cr Default: \code{c(-180, 180, -90, 90)}
#'@param buffer numeric. Input of decimal degrees of buffer around the shapefile and/or extent. \cr Default: \code{0}
#'@param convert.files.to.asc logical. Input whether files should be converted into the ASCII format. If \code{TRUE}: a new subdirectory is created and the rawdata is saved there. \cr If \code{clipping} is \code{TRUE}: the clipped raster files are also saved as ASCII grids. \cr Default: \code{FALSE}
#'@param stacking.data logical. Input whether the downloaded data should be stacked as a netCDF-rasterstack. \cr Default: \code{FALSE}
#'@param combine.raw.zip logical. Should the downloaded raw-data be "zipped". \cr Default: \code{FALSE}
#'@param delete.raw.data logical. Should the downloaded raw-data be deleted. If the "combine.raw.zip"-option is \code{TRUE}, raw-data is still available in the zipped file.\cr Default: \code{FALSE}
#'@param save.bib.file logical. Whether a BibTex-citation file of the CHELSA dataset should be provided in the Working directory. \cr Default: \code{TRUE}
#'
#'@return Downscaled global climatological data from the last glacial maximum.
#'
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. "Climatologies at high resolution for the earth's land surface areas". In: _Scientific Data_ 4.1 (Sep. 2017). DOI: 10.1038/sdata.2017.122. <URL: https://doi.org/10.1038/sdata.2017.122>.
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. _Data from: Climatologies at high resolution for the earth's land surface areas_. En. 2018. DOI: 10.5061/DRYAD.KD1D4. <URL: http://datadryad.org/stash/dataset/doi:10.5061/dryad.kd1d4>.
#'
#'@examples
#' \dontrun{
#' # Bioclim
#' Chelsa.lgm.download(parameter = "bio",
#' bio.var = c(1,19),
#' model.var = "MPI-ESM-P")
#' # Precipitation
#' Chelsa.lgm.download(parameter = "prec",
#' month.var = c(1,12),
#' model.var = "MPI-ESM-P")
#' }
#'
#'
#'@import stringr
#'@import RCurl
#'@import ncdf4
#'@import terra
#'@importFrom utils unzip download.file setTxtProgressBar txtProgressBar
#'
#'
#'@export
Chelsa.lgm.download <- function(save.location = "./",
parameter = c("prec", "temp", "tmax", "tmin", "bio"),
bio.var = c(1:19),
month.var = c(1:12),
model.var=c("CCSM4", "MRI-CGCM3", "CNRM-CM5",
"FGOALS-g2", "IPSL-CM5A-LR",
"MIROC-ESM", "MPI-ESM-P"
),
download.dem = FALSE,
clipping = FALSE,
clip.shapefile = NULL,
clip.extent = c(-180, 180, -90, 90),
buffer = 0,
convert.files.to.asc = FALSE,
stacking.data = FALSE,
combine.raw.zip = FALSE,
delete.raw.data = FALSE,
save.bib.file = TRUE
){
# requireNamespace("stringr")
# requireNamespace("RCurl")
# requireNamespace("ncdf4")
gc()
call.time <- stringr::str_replace_all(stringr::str_replace_all(paste0(Sys.time()), pattern = ":", replacement = "-"), pattern = " ", replacement = "_")
# initial check -----------------------------------------------------------
# normalize Path for easier application later
save.location <- normalizePath(save.location, winslash = "/")
# Check which parameters are put in and if the connected
# month/bio-variables are correctly input if(is.element("prec", parameter)|is.element("temp", parameter)|
is.element("tmax", parameter)|is.element("tmin", parameter)){
month.var <- c(month.var)
if(!is.numeric(month.var)) stop("month.var needs to be a numeric vector")
# month.var <- str_pad(month.var, 2, 'left', pad = "0")
}
if(is.element("bio", parameter)){
bio.var <- c(bio.var)
if(!is.numeric(bio.var)) stop("bio.var needs to be a numeric vector")
bio.var <- as.character(bio.var)
bio.var <- str_pad(bio.var, 2, 'left', pad = "0")
# print(bio.var)
}
# through all given parameters --------------------------------------------
# work through paramerters
for(i in parameter){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
# create intermediate strings for later use
interm <- switch(i,
"prec" = "prec/",
"temp" = "tmean/",
"tmax" = "tmax/",
"tmin" = "tmin/",
"bio" = "bioclim/",
stop())
variable.numbers <- switch(i,
"bio" = bio.var,
"tmin" = month.var,
"tmax" = month.var,
"temp" = month.var,
"prec" = month.var,
stop())
# create new directory
dir.create(paste0(save.location, "/", i))
temp.save.location <- paste0(save.location, "/", i, "/")
for (model in model.var) {
# download of the requested datasets -------------------------------------
temp.temp.save.location <- paste0(temp.save.location,
stringr::str_replace_all(interm,
pattern = "/",
"_"),
"LGM_PMIP_", model, "/")
if(!dir.exists(temp.temp.save.location)){
dir.create(temp.temp.save.location)
}
# temp.temp.save.location <- normalizePath(temp.temp.save.location,
# winslash = "/")
# Check if bio is not requested
if(i != "bio"){
if(i == "temp") {
month.var <- str_pad(month.var, 2, 'left', pad = "0")
}else{
month.var <- as.integer(month.var)
}
# work through every requested month
for(month in 1:length(month.var)){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
dest.temp <- paste0(temp.temp.save.location, "CHELSA_PMIP_", model, "_", i,"_",month.var[month],".tif")
if(!file.exists(dest.temp)){
if(i != "temp") {
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/pmip3/",
interm, "CHELSA_PMIP_", model,"_",i,"_", month.var[month],"_1.tif")
}else{
if(model == "CCSM4"){
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/pmip3/",
interm, "CHELSA_PMIP_", model, "_tmean_", month.var[month],".tif")
}else{
month.var <- as.integer(month.var)
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/pmip3/",
interm, "CHELSA_PMIP_", model,"_tmean_", month.var[month],"_1.tif")
}
}
# check if URL is available
if(url.exists(URL.temp)){
# download file to save location
download.file(url = URL.temp,
destfile = dest.temp,
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
if(i != "prec"){
raster.temp <- terra::rast(dest.temp)
raster.temp <- terra::clamp(raster.temp, lower = -1000, useValues= FALSE)
gc()
# Conversion Float
raster.temp <- process.raster.int.doub(raster.temp)
# umrechnung Kelvin - Celsius
gc()
raster.temp <- process.raster.offset(raster.layer= raster.temp)
terra::writeRaster(x = raster.temp,
filename = dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}else{
# for precipitation as http://chelsa-climate.org/last-glacial-maximum-climate/ says
raster.temp <- terra::rast(dest.temp)
raster.temp <- terra::clamp(raster.temp, upper = 30000, useValues= FALSE)
gc()
raster.temp <- process.raster.int.doub(raster.temp)
terra::writeRaster(x = raster.temp,
filename = dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}
}else{
# Error message
warning(paste0("File does not exist. Did not download: \n", URL.temp, "\n\n"),
call. = TRUE, immediate. = FALSE)
}
}
if(month.var[month] == month.var[length(month.var)] &
length(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent,
buffer = buffer, convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(save.location = temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){
if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSALGM_", i, ""),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T), force = TRUE)
}
}
}
}else{
# analog to other parameters
# just that bio.var instead of month.var is used.
for(bio in bio.var){
dest.temp <- paste0(temp.temp.save.location, "CHELSA_PMIP_", model,
"_BIO_", bio,".tif")
if(!file.exists(dest.temp)){
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/pmip3/",
interm, "CHELSA_PMIP_", model, "_BIO_",bio,".tif")
# check if URL is available
if(url.exists(URL.temp)){
# download file to save location
download.file(url = URL.temp,
destfile = dest.temp,
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
# Casting into floats and deleting NA values
raster.temp <- terra::rast(dest.temp)
raster.temp <- terra::clamp(raster.temp, upper = -1000, useValues= FALSE)
gc()
if(bio <= 11){
# values(raster.temp) <- as.numeric(values(raster.temp)/10)
raster.temp <- process.raster.int.doub(raster.temp)
}
terra::writeRaster(x = raster.temp,
filename = dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
}else{
# Error message
warning(paste0("File does not exist. Did not download: \n", URL.temp),
call. = TRUE, immediate. = FALSE) }
}
if(bio == bio.var[length(bio.var)] &
length(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent,
buffer = buffer,
convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(save.location = temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){
if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSALGM_", i, ""),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T), force = TRUE)
}
}
}
}
if(length(list.files(temp.temp.save.location,
include.dirs = TRUE)) == 0){
unlink(str_sub(temp.temp.save.location, 1, end = str_length(temp.temp.save.location)-1),
force = T, recursive = TRUE)
}
}
}
if(download.dem == TRUE){
if(!dir.exists(paste0(save.location, "/elev"))){
dir.create(paste0(save.location, "/elev"))
}
if(!dir.exists(paste0(save.location, "/elev/CHELSA_LGM_Elevation_Grid"))){
dir.create(paste0(save.location, "/elev/CHELSA_LGM_Elevation_Grid"))
}
if(!file.exists(paste0(save.location, "/elev/CHELSA_LGM_Elevation_Grid/", "CHELSA_PMIP_dem_global.tif"))){
download.file(url = "https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/pmip3/DEM/high_longlat.tif",
destfile = paste0(save.location, "/elev/CHELSA_LGM_Elevation_Grid/", "CHELSA_PMIP_dem_global.tif"),
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
}
}
# Saving BIB File
if(save.bib.file == TRUE) save.citation(save.location = save.location, dataSetName = "CHELSA")}
#'@title CHELSA Timeseries Download
#'@author Helge Jentsch
#'@description This function supports a download of the CHELSA Timeseries dataset (Jan. 1979 - Dec. 2013). This includes precipitation sums (mm) and temperature (average, maximum, minimum; °C) parameters. For further information, please regard \url{http://chelsa-climate.org/timeseries/}.\cr To allow pre-processing, clipping and buffering, conversion to ASCII-grids and stacking options are included.\cr Optional an output of a .bib-file of the cited literature can be retrieved.\cr For user convenience, saving directories will be created automatically. Also options to "zip" and/or delete the RAW-files are included.
#'
#'@note Please note that the downloaded data for temperature are processed to °C with one significant decimal without offset and factor. Processing and conversion to other file-formats on a global dataset may take some time.
#'
#'@param save.location string. Input where the datasets should be saved. \cr Default: Working Directory.
#'@param parameter string (vector). Input of parameters which should be downloaded. \cr Default: \code{c("prec", "tmax", "tmin")}
#'@param start.year.var integer. Input year the download timeseries starts. \cr Default: 1979 (minimum)
#'@param start.month.var integer. Input month the download timeseries starts. \cr Default: 1 (minimum)
#'@param end.year.var integer. Input year the download timeseries ends. \cr Default: 2013 (maximum)
#'@param end.month.var integer. Input month the download timeseries ends. \cr Default: 12 (maximum)
#'@param include.month.var integer (vector). Input which monthly data should be downloaded. \cr Default: \code{c(1:12)}
#'@param clipping logical. Input whether the downloaded data should be clipped.\cr If \code{FALSE}: \code{clip.shapefile}, \code{buffer}, \code{clip.extent} will be ignored. \cr Default: \code{FALSE}
#'@param clip.shapefile string. Input which shapefile should be used for clipping. \cr Default: \code{NULL}
#'@param clip.extent numeric (vector). Input vector with four numeric values. This is following the input order c("xleft", "xright", "ybottom", "ytop").\cr Default: \code{c(-180, 180, -90, 90)}
#'@param buffer numeric. Input of decimal degrees of buffer around the shapefile and/or extent. \cr Default: \code{0}
#'@param convert.files.to.asc logical. Input whether files should be converted into the ASCII format.\cr If \code{TRUE}: a new subdirectory is created and the rawdata is saved there. \cr If \code{clipping} is \code{TRUE}: the clipped raster files are also saved as ASCII grids. \cr Default: \code{FALSE}
#'@param combine.raw.zip logical. Should the downloaded raw-data be "zipped". \cr Default: \code{FALSE}
#'@param delete.raw.data logical. Should the downloaded raw-data be deleted. If \code{combine.raw.zip} is \code{TRUE}: raw-data is still available in the zipped file. \cr Default: \code{FALSE}
#'@param save.bib.file logical. Whether a BibTex-citation file of the dataset should be provided in the Working directory. \cr Default: \code{TRUE}
#'
#'@return Custom dataset of CHELSA Timeseries for a chosen timeseries.
#'
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. "Climatologies at high resolution for the earth's land surface areas". In: _Scientific Data_ 4.1 (Sep. 2017). DOI: 10.1038/sdata.2017.122. <URL: https://doi.org/10.1038/sdata.2017.122>.
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. _Data from: Climatologies at high resolution for the earth's land surface areas_. En. 2018. DOI: 10.5061/DRYAD.KD1D4. <URL: http://datadryad.org/stash/dataset/doi:10.5061/dryad.kd1d4>.
#'
#'@examples
#' \dontrun{
#' Chelsa.timeseries.download(parameter = "prec",
#' start.year.var = 2000,
#' start.month.var = 1,
#' end.year.var = 2002,
#' end.month.var = 12,
#' include.month.var = c(1,12))
#' }
#'
#'@import stringr
#'@import RCurl
#'@import ncdf4
#'@import terra
#'@import httr
#'@importFrom utils unzip download.file setTxtProgressBar txtProgressBar
#'
#'
#'
Chelsa.timeseries.download <- function(save.location = "./",
parameter = c("prec", "tmax", "tmin"),
start.year.var = 1979,
start.month.var = 1,
end.year.var = 2013,
end.month.var = 12,
include.month.var = c(1:12),
# bio.var = c(1:19),
clipping = FALSE,
clip.shapefile = NULL,
buffer = 0,
clip.extent = c(-180, 180, -90, 90),
convert.files.to.asc = FALSE,
stacking.data = FALSE,
combine.raw.zip = FALSE,
delete.raw.data = FALSE,
save.bib.file = TRUE){
# requireNamespace("stringr")
# requireNamespace("RCurl")
# requireNamespace("ncdf4")
gc()
call.time <- stringr::str_replace_all(stringr::str_replace_all(paste0(Sys.time()), pattern = ":", replacement = "-"), pattern = " ", replacement = "_") # initial check -----------------------------------------------------------
# normalize Path for easier application later
save.location <- normalizePath(save.location, winslash = "/")
# Check which parameters are put in and if the connected
# month/bio-variables are correctly input
if(is.element("prec", parameter)|
is.element("tmax", parameter)|is.element("tmin", parameter)){
include.month.var <- c(include.month.var)
if(!is.numeric(include.month.var)) stop()
include.month.var <- str_pad(include.month.var, 2, 'left', pad = "0")
# print(include.month.var)
}
if(start.year.var < 1979 | end.year.var > 2013 | end.year.var < 1979 | start.year.var > 2013) {
stop("Timeseries only available from 01.1979 to 12.2013. \n Please check input!")
}
# check for consistent timeseries
if(end.year.var < start.year.var) stop("Endyear is before the startyear. Please correct the input!")
if(start.year.var == end.year.var){
if(start.month.var > end.month.var) stop("End is before the start. Please correct the input!")
}
ts_string <- seq.Date(as.Date(paste(start.year.var,
start.month.var, "01", sep = "-")),
as.Date(paste(end.year.var,
end.month.var, "01", sep = "-")),
by = "month")
ts_string <- format.Date(ts_string, format = "%Y_%m")
# ts_string <- str_sub(ts_string, 1, end = str_length(ts_string)-3)
# ts_string <- str_replace_all(ts_string, pattern = "-", replacement = "_")
if(length(include.month.var)!=12){
ts.string.temp <- c()
for (incl.month in include.month.var) {
# print(incl.month)
ts.string.temp <- c(ts.string.temp,
ts_string[grep(pattern = paste0("_", incl.month)
, ts_string)]
)
}
ts_string <- ts.string.temp
}
# Parameter and directories -----------------------------------------------
# work through paramerters
for(i in parameter){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
# create intermediate strings for later use
interm <- switch(i,
"prec" = "prec/",
"temp" = "tmean/",
"tmax" = "tmax/",
"tmin" = "tmin/",
# "bio" = "bio/",
stop())
variable.numbers <- switch(i,
# "bio" = bio.var,
"tmin" = include.month.var,
"tmax" = include.month.var,
"temp" = include.month.var,
"prec" = include.month.var,
stop())
# create new directory
if(!dir.exists(paste0(save.location, "/", i))){
dir.create(paste0(save.location, "/", i), showWarnings = FALSE)
} temp.save.location <- paste0(save.location, "/", i, "/")
# to go analog to the functions before
temp.temp.save.location <- paste0(temp.save.location,
stringr::str_replace_all(interm,
pattern = "/",
"_"),
"timeseries","/")
# print(str_sub(temp.temp.save.location, end=-2))
if(!dir.exists(temp.temp.save.location)){
dir.create(str_sub(temp.temp.save.location, end=-2))
}
# temp.temp.save.location <- normalizePath(temp.temp.save.location,
# winslash = "/")
# if(i == "temp"){
# i <- "tmean"
# }
# print(interm)
# Download ----------------------------------------------------------------
# if(i != "bio"){
for (year_month in ts_string){
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/timeseries/",
interm, "CHELSA_", i,"_",year_month,
"_V1.2.1.tif")
# check if URL is available
if(!http_error(URL.temp)){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
dest.file <- paste0(temp.temp.save.location, "CHELSA_", i,
"_", year_month, "_V1.2.1.tif")
if(!file.exists(dest.file)){
# download file to save location
download.file(url = URL.temp,
destfile = dest.file,
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
if(i != "prec"){
raster.temp <- terra::rast(dest.file)
raster.temp <- terra::clamp(raster.temp, lower = -1000, useValues = FALSE)
gc()
raster.temp <- process.raster.int.doub(raster.temp)
raster.temp <- process.raster.offset(raster.layer = raster.temp)
terra::writeRaster(x = raster.temp,
filename = dest.file,
overwrite = TRUE)
rm(raster.temp)
gc()
}
}
}else{
# Warning message
warning(paste0("File does not exist. Did not download: \n", URL.temp, "\n\n"),
call. = TRUE, immediate. = FALSE)
}
if(year_month == ts_string[length(ts_string)] &
length(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent, convert.files.to.asc = convert.files.to.asc,
buffer = buffer,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){
if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
stack.time.series = TRUE,
time.series = ts_string,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.time.series = TRUE,
time.series = ts_string,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSATimeseries_", i, ""),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T), force = TRUE)
}
}
}
# Clean up, if no data was downloaded. ------------------------------------
if(length(list.files(temp.temp.save.location,
include.dirs = TRUE)) == 0){
unlink(str_sub(temp.temp.save.location, 1, end = str_length(temp.temp.save.location)-1),
force = T, recursive = TRUE)
}
}
# Saving BIB File
if(save.bib.file == TRUE) save.citation(save.location = save.location, dataSetName = "CHELSA")
}
#'@title CHELSA CRU Timeseries Download
#'@author Helge Jentsch
#'@description This function supports a download of the CHELSA CRU Timeseries dataset (Jan. 1901 - Dec. 2016). This includes precipitation sums in mm and temperature (maximum, minimum) in degree Celsius. For further information, please regard \url{http://chelsa-climate.org/chelsacruts/}.\cr To allow pre-processing, clipping and buffering, conversion to ASCII-grids and stacking options are included.\cr Optional an output of a .bib-file of the cited literature can be retrieved.\cr For user convenience, saving directories will be created automatically. Also options to "zip" and/or delete the RAW-files are included.
#'
#'@note Please note that the downloaded data for temperature is processed to °C with one significant decimal without offset and factor. Processing and conversion to other file-formats on a global dataset may take some time.
#'
#'@param save.location string. Input where the datasets should be saved. \cr Default: Working Directory.
#'@param parameter string (vector). Input of parameters which should be downloaded. \cr Default: \code{c("prec", "tmax", "tmin")}
#'@param start.year.var integer. Input year the download timeseries starts. \cr Default: 1901 (minimum)
#'@param start.month.var integer. Input month the download timeseries starts. \cr Default: 1 (minimum)
#'@param end.year.var integer. Input year the download timeseries ends. \cr Default: 2016 (maximum)
#'@param end.month.var integer. Input month the download timeseries ends. \cr Default: 12 (maximum)
#'@param include.month.var integer (vector). Input which monthly data should be downloaded. \cr Default: \code{c(1:12)}
#'@param clipping logical. Input whether the downloaded data should be clipped.\cr If \code{FALSE}: \code{clip.shapefile}, \code{buffer}, \code{clip.extent} will be ignored. \cr Default: \code{FALSE}
#'@param clip.shapefile string. Input which shapefile should be used for clipping. \cr Default: \code{NULL}
#'@param clip.extent numeric (vector). Input vector with four numeric values. This is following the input order c("xleft", "xright", "ybottom", "ytop").\cr Default: \code{c(-180, 180, -90, 90)}
#'@param buffer numeric. Input of decimal degrees of buffer around the shapefile and/or extent. \cr Default: \code{0}
#'@param convert.files.to.asc logical. Input whether files should be converted into the ASCII format.\cr If \code{TRUE}: a new subdirectory is created and the rawdata is saved there. \cr If \code{clipping} is \code{TRUE}: the clipped raster files are also saved as ASCII grids. \cr Default: \code{FALSE}#'@param stacking.data logical. Input whether the downloaded data should be stacked as a netCDF-rasterstack. \cr Default: \code{FALSE}
#'@param combine.raw.zip logical. Should the downloaded raw-data be "zipped". \cr Default: \code{FALSE}
#'@param delete.raw.data logical. Should the downloaded raw-data be deleted. If \code{combine.raw.zip} is \code{TRUE}: raw-data is still available in the zipped file. \cr Default: \code{FALSE}
#'@param save.bib.file logical. Whether a BibTex-citation file of the dataset should be provided in the Working directory. \cr Default: \code{TRUE}
#'
#'@return Custom dataset of CHELSA CRU Timeseries for a chosen timeseries.
#'
#'
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. "Climatologies at high resolution for the earth's land surface areas". In: _Scientific Data_ 4.1 (Sep. 2017). DOI: 10.1038/sdata.2017.122. <URL: https://doi.org/10.1038/sdata.2017.122>.
#'@references D. N. Karger, O. Conrad, J. B{\"o}hner , et al. _Data from: Climatologies at high resolution for the earth's land surface areas_. En. 2018. DOI: 10.5061/DRYAD.KD1D4. <URL: http://datadryad.org/stash/dataset/doi:10.5061/dryad.kd1d4>.
#'@references D. N. Karger and N. E. Zimmermann. _CHELSAcruts - High resolution temperature and precipitation timeseries for the 20th century and beyond_. 2018. DOI: http://dx.doi.org/10.16904/envidat.159.
#'
#'@examples
#' \dontrun{
#' Chelsa.CRUts.download(parameter = "prec",
#' start.year.var = 2000,
#' start.month.var = 1,
#' end.year.var = 2002,
#' end.month.var = 12,
#' include.month.var = c(1,12))
#' }
#'
#'@import stringr
#'@import RCurl
#'@import ncdf4
#'@import terra
#'@import httr
#'@importFrom utils unzip download.file setTxtProgressBar txtProgressBar
#'
#'
#'@export
Chelsa.CRUts.download <- function(save.location = "./",
parameter = c("prec", "tmax", "tmin"),
start.year.var = 1901,
start.month.var = 1,
end.year.var = 2016,
end.month.var = 12,
include.month.var = c(1:12),
# float.temperature.format = TRUE,
clipping = FALSE,
clip.shapefile = NULL,
buffer = 0,
clip.extent = c(-180, 180, -90, 90),
convert.files.to.asc = FALSE,
stacking.data = FALSE,
combine.raw.zip = FALSE,
delete.raw.data = FALSE,
save.bib.file = TRUE
){
# requireNamespace("stringr")
# requireNamespace("RCurl")
# requireNamespace("ncdf4")
gc()
call.time <- stringr::str_replace_all(stringr::str_replace_all(paste0(Sys.time()), pattern = ":", replacement = "-"), pattern = " ", replacement = "_")
# initial check -----------------------------------------------------------
# normalize Path for easier application later
save.location <- normalizePath(save.location, winslash = "/")
# Check which parameters are put in and if the connected
# month/bio-variables are correctly input
if(is.element("prec", parameter)|
is.element("tmax", parameter)|is.element("tmin", parameter)){
include.month.var <- c(include.month.var)
if(!is.numeric(include.month.var)) {
stop(paste0("CHELSA CRU timeseries is only available for ",
"precipitation and temperature maximum and minimum"))
}
include.month.var <- str_pad(include.month.var, 2, 'left', pad = "0")
# print(include.month.var)
}
if(start.year.var < 1901 | end.year.var > 2016 | end.year.var < 1901 | start.year.var > 2016) {
stop("Timeseries only available from 01.1901 to 12.2016. Please check input!")
}
# check for consistent timeseries
if(end.year.var < start.year.var) {
stop("Endyear is before the startyear. Please correct the input!")
}
if(start.year.var == end.year.var){
if(start.month.var > end.month.var) {
stop("End is before the start. Please correct the input!")
}
}
ts_string <- seq.Date(as.Date(paste(start.year.var,
start.month.var, "01", sep = "-")),
as.Date(paste(end.year.var,
end.month.var, "01", sep = "-")),
by = "month")
ts_string <- format.Date(ts_string, format = "%m_%Y")
if(length(include.month.var)!=12){
ts.string.temp <- c()
for (incl.month in include.month.var) {
# print(incl.month)
ts.string.temp <- c(ts.string.temp,
ts_string[grep(pattern = paste0(incl.month,"_"),
ts_string)]
)
}
ts_string <- ts.string.temp
}
ts_string <- as.Date(paste0(ts_string,"-01"), format = "%d_%Y-%m")
ts_string <- format.Date(ts_string, format = "%e_%Y")
ts_string <- str_remove(ts_string, pattern = " ")
# Parameter and directories -----------------------------------------------
# work through paramerters
for(i in parameter){
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
# create intermediate strings for later use
interm <- switch(i,
"prec" = "prec/",
"tmax" = "tmax/",
"tmin" = "tmin/",
# "temp" = "tmean/",
# "bio" = "bioclim/",
stop())
variable.numbers <- switch(i,
# "bio" = bio.var,
"tmin" = include.month.var,
"tmax" = include.month.var,
# "temp" = month.var,
"prec" = include.month.var,
stop())
# create new directory
dir.create(paste0(save.location, "/", i), showWarnings = FALSE)
temp.save.location <- paste0(save.location, "/", i, "/")
# to go analog to the functions before
temp.temp.save.location <- paste0(temp.save.location,
stringr::str_replace_all(interm,
pattern = "/",
"_"),
"CRU_timeseries","/")
if(!dir.exists(temp.temp.save.location)){
dir.create(temp.temp.save.location)
}
# temp.temp.save.location <- normalizePath(temp.temp.save.location,
# winslash = "/")
# Download ----------------------------------------------------------------
for (year_month in ts_string){
output.year_month <- str_split(year_month, pattern = "_")
output.year_month <- unlist(output.year_month)
zero <- switch (as.character(str_length(paste0(output.year_month[2],
"_", output.year_month[1]))),
"7" = "",
"6" = "0",
"")
output.year_month <- paste0(output.year_month[2],
"_", zero,
output.year_month[1])
rm(zero)
URL.temp <-
paste0("https://os.zhdk.cloud.switch.ch/envicloud/chelsa/chelsa_V1/chelsa_cruts/",
interm, "CHELSAcruts_", i,"_",year_month,
"_V.1.0.tif")
dest.temp <- paste0(temp.temp.save.location, "CHELSA_CRUts_", i,
"_",output.year_month, "_V_1_0.tif")
# check if URL is available
if(!http_error(URL.temp)){
gc()
# clear up the temporary directory
unlink(list.files(tempdir(), recursive = T, full.names=T))
if(!file.exists(dest.temp)){
# download file to save location
download.file(url = URL.temp,
destfile = dest.temp,
overwrite = TRUE,
mode = 'wb',
quiet = FALSE)
gc()
if(i != "prec"){
# print(Sys.time())
raster.temp.file.dest <- paste0(normalizePath(tempdir(),
winslash = "/",
mustWork = T),
"/temp_gdalwarped.tif")
# gdalUtils::gdalwarp(dest.temp, raster.temp.file.dest)
# raster.temp <- terra::rast(raster.temp.file.dest)
raster.temp <- terra::rast(dest.temp)
raster.temp <- terra::clamp(raster.temp, lower = -1000)
raster.temp <- process.raster.int.doub(raster.temp)
terra::writeRaster(x = raster.temp,
filename = dest.temp,
overwrite = TRUE)
rm(raster.temp)
gc()
# print(Sys.time())
}
}
}else{
# Warning message
warning(paste0("File does not exist. Did not download: \n", URL.temp, "\n\n"),
call. = TRUE, immediate. = FALSE)
}
if(year_month == ts_string[length(ts_string)] &
length(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE)) != 0){
if(clipping == TRUE){
clipping.tif(clip.save.location = temp.temp.save.location,
clip.shapefile = clip.shapefile,
clip.extent = clip.extent, buffer = buffer,
convert.files.to.asc = convert.files.to.asc,
time.stamp.var = call.time)
}
if(convert.files.to.asc == TRUE){
convert.to.asc(temp.temp.save.location,
time.stamp.var = call.time)
}
if(stacking.data == TRUE){
if(clipping==TRUE){
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.clipped = TRUE,
stack.time.series = TRUE,
time.series = ts_string,
time.stamp.var = call.time)
}else{
stacking.downloaded.data(stack.save.location = temp.temp.save.location,
parameter.var = i,
variable.numbers = variable.numbers,
stack.time.series = TRUE,
time.series = ts_string,
time.stamp.var = call.time)
}
}
if(combine.raw.zip == TRUE){
combine.raw.in.zip(save.location = temp.temp.save.location,
zip.name = paste0("CHELSACRUts_", i, "_V.1.0"),
time.stamp.var = call.time)
}
if(delete.raw.data == TRUE){
unlink(list.files(temp.temp.save.location,
pattern = ".tif",
include.dirs = FALSE, full.names = T), force = TRUE)
}
}
}
# Clean up, if no data was downloaded. ------------------------------------
if(length(list.files(temp.temp.save.location,
include.dirs = TRUE)) == 0){
unlink(str_sub(temp.temp.save.location, 1, end = str_length(temp.temp.save.location)-1),
force = T, recursive = TRUE)
}
}
# Saving BIB File
if(save.bib.file == TRUE) save.citation(save.location = save.location, dataSetName = "CHELSA")
}