berryFunctions (version 1.11.0)

dataDWD: download data from DWD

Description

Get climate data from the German Weather Service (DWD) FTP-server. The desired .zip dataset is downloaded into dir, unpacked, read, processed and returned as a data.frame

Usage

dataDWD(file, base1 = "ftp://ftp-cdc.dwd.de/pub/CDC/observations_germany/climate", base2 = "hourly/precipitation/recent", dir = "DWDdata", browse = 0:2, meta = 0:2, read = TRUE, format = NA, quiet = FALSE, ...)

Arguments

file

Filename (must be available at the location given by base1 and base2)

base1

Main directory of DWD ftp server (can probably always be left unchanged)

base2

Characterstring with subdirectory. DEFAULT: "hourly/precipitation/recent"

dir

Writeable directory on your computer. Created if not existent. DEFAULT: "DWDdata" at your current getwd()

browse

Integer specifying whether and how to open repository via browseURL. 0 for regular file download. 1 to open base1 (no download). 2 to open base1/base2). If 1 or 2, no dir is created. DEFAULT: 0

meta

Integer specifying whether to get metadata instead of actual data. 0 for regular file. 1 for meta data of all stations (meta is automatically set to 1 if file ends in ".txt". Column widths for read.fwf are computed internally). 2 for a list of the available files (requires RCurl to be installed. If meta=2, file="" is possible, as it is ignored anyways). DEFAULT: 0

read

Read the file with readDWD? If FALSE, only download is performed. DEFAULT: TRUE

format

Format used in strptime to convert date/time column, see readDWD. DEFAULT: NA

quiet

Suppress message about directory? DEFAULT: FALSE

...

Further arguments currently ignored

Value

data.frame of the desired dataset (returned by readDWD if meta=0), presuming downloading and processing were successfull. Alternatively, links that were opened if browse!=0.

See Also

readDWD, download.file, monthAxis, climateGraph

Examples

## Not run:  ## Not run in CRAN checks because of downloading, writing files, etc
# # Basic usage:
# prec <- dataDWD(file="stundenwerte_RR_02787_akt.zip")
# plot(prec$MESS_DATUM, prec$NIEDERSCHLAGSHOEHE, main="DWD hourly rain Kupferzell", col="blue",
#      xaxt="n", las=1, type="l", xlab="Date", ylab="Hourly rainfall  [mm]")
# monthAxis(1, ym=T)
# 
# prec2 <- dataDWD("stundenwerte_RR_03987_akt.zip") # writes into the same folder
# 
# clim <- dataDWD(base2="monthly/kl/recent", file="monatswerte_03987_akt.zip")
# # Potsdam monthly averages/mins/maxs of: wind, clouds, rainfall, sunshine, temperature
# 
# # For several stations (do this at your own risk of getting kicked off the FTP)
# # metadata for existing stations:
# stats <- dataDWD("RR_Stundenwerte_Beschreibung_Stationen.txt")
# str(stats)  # data.frame with 8 columns (4 int, 2 num, 2 factor), 1292 rows (July 2016)
# head(stats)
# 
# # List of actually available files (needs RCurl):
# # install.packages("RCurl")
# files <- dataDWD("", meta=2)
# headtail(sort(files),6)
# 
# # Apply the function to several files, create a list of data.frames:
# # Exclude the pdf and txt files, or dataDWD will break
# prec <- lapply(files[1:2], function(f) {Sys.sleep(runif(1,0,5)); dataDWD(f)})
# names(prec) <- substr(files[1:2], 14, 21)
# str(prec, max.level=1)
# 
# # Test Metadata part of function:
# files <- read.table(as.is=TRUE, text="
# #ftp://ftp-cdc.dwd.de/pub/CDC/observations_germany/climate/
# daily/kl/historical                KL_Tageswerte_Beschreibung_Stationen.txt
# daily/kl/recent                    KL_Tageswerte_Beschreibung_Stationen.txt
# daily/more_precip/historical       RR_Tageswerte_Beschreibung_Stationen.txt
# daily/more_precip/recent           RR_Tageswerte_Beschreibung_Stationen.txt
# daily/soil_temperature/historical  EB_Tageswerte_Beschreibung_Stationen.txt
# daily/soil_temperature/recent      EB_Tageswerte_Beschreibung_Stationen.txt
# daily/solar                        ST_Beschreibung_Stationen.txt
# hourly/air_temperature/historical  TU_Stundenwerte_Beschreibung_Stationen.txt
# hourly/air_temperature/recent      TU_Stundenwerte_Beschreibung_Stationen.txt
# hourly/cloudiness/historical       N_Stundenwerte_Beschreibung_Stationen.txt
# hourly/cloudiness/recent           N_Stundenwerte_Beschreibung_Stationen.txt
# hourly/precipitation/historical    RR_Stundenwerte_Beschreibung_Stationen.txt
# hourly/precipitation/recent        RR_Stundenwerte_Beschreibung_Stationen.txt
# hourly/pressure/historical         P0_Stundenwerte_Beschreibung_Stationen.txt
# hourly/pressure/recent             P0_Stundenwerte_Beschreibung_Stationen.txt
# hourly/soil_temperature/historical EB_Stundenwerte_Beschreibung_Stationen.txt
# hourly/soil_temperature/recent     EB_Stundenwerte_Beschreibung_Stationen.txt
# hourly/solar                       ST_Beschreibung_Stationen.txt
# hourly/sun/historical              SD_Stundenwerte_Beschreibung_Stationen.txt
# hourly/sun/recent                  SD_Stundenwerte_Beschreibung_Stationen.txt
# hourly/wind/historical             FF_Stundenwerte_Beschreibung_Stationen.txt
# hourly/wind/recent                 FF_Stundenwerte_Beschreibung_Stationen.txt
# monthly/kl/historical              KL_Monatswerte_Beschreibung_Stationen.txt
# monthly/kl/recent                  KL_Monatswerte_Beschreibung_Stationen.txt
# monthly/more_precip/historical     RR_Monatswerte_Beschreibung_Stationen.txt
# monthly/more_precip/recent         RR_Monatswerte_Beschreibung_Stationen.txt")
# i=1
# meta <- dataDWD(file=files[i,2], base2=files[i,1])
# colPoints(geoLaenge, geoBreite, Stations_id, data=meta, add=F, asp=1.5)
# colPoints(geoLaenge, geoBreite, Stationshoehe, data=meta, add=F, asp=1.5)
# meta$von_jahr <- meta$von_datum/1e4
# meta$bis_jahr <- meta$bis_datum/1e4
# meta$dauer <- meta$bis_jahr - meta$von_jahr
# colPoints(geoLaenge, geoBreite, von_jahr, data=meta, add=F, asp=1.5)
# colPoints(geoLaenge, geoBreite, bis_jahr, data=meta, add=F, asp=1.5)
# colPoints(geoLaenge, geoBreite, dauer, data=meta, add=F, asp=1.5) 
# hist(meta$bis_jahr, breaks=50, col="purple")
# hist(meta$dauer, breaks=50, col="purple")
# sum(meta$dauer>50); mean(meta$dauer>50) 
# # 356 (32.7%) stations with more than 50 years of data (according to metadata)
# ## End(Not run)