This R package provides current and historic weather data for around 28,000 weather stations around the world. The data is provided by the National Oceanic and Atmospheric Administration's (NOAA) Integrated Surface Data (ISD). The package provides a function to download the data and to load it into R. The data is provided as a data frame with the following columns: station_id, time, wind direction in compass degrees wd, wind speed in miles per hour ws, cloud ceiling in hundreds of feet ceil_hgt, visibility in statute miles to nearest tenth visibility, temperature in Celsius temp, dew point in Celsius dew_point, sea level pressure in Millibars atmos_pres, and relative humidity rh. The data is provided in the ISD-Hourly format.
This package is largly inspired by the stationaRy packagy by Richard Iannone. Currently, the package is not updated since 2020 and can't download recent years. This package is an attempt to provide a similar package that is updated and maintained. If Richard updates his package, go over to stationaRy as it provides more functionality.
Map of all stations:
To install and load the package, run the following code:
devtools::install_github("hendrikmahlkow/weatheR")
library(weatheR)To download all available weather stations, run the following code:
stations = available_weather_stations(save_file = FALSE)
stations
#> station_id name country state icao lat lon elev begin_date end_date
#> 1 007018-99999 WXPOD 7018 <NA> NA <NA> 0.000 0.000 7018.0 2011-03-09 2013-07-30
#> 2 007026-99999 WXPOD 7026 AF NA <NA> 0.000 0.000 7026.0 2012-07-13 2017-08-22
#> 3 007070-99999 WXPOD 7070 AF NA <NA> 0.000 0.000 7070.0 2014-09-23 2015-09-26
#> 4 008260-99999 WXPOD8270 <NA> NA <NA> 0.000 0.000 0.0 2005-01-01 2012-07-31
#> 5 008268-99999 WXPOD8278 AF NA <NA> 32.950 65.567 1156.7 2010-05-19 2012-03-23
#> # ... with 28,442 more rowsNOAA uses FIPS 2-digit country codes. For example, in FIPS 'ES' is El Salvador, while the ISO-2 code for El Salvador is 'SV' and Spain has 'ES'. The package come with a country concordance table that matches FIPS country code to country names and ISO-3 codes. Call country_concordance to retrieve the data frame.
country_concordance
#> NOAA_ccode country_name country_code
#> 1 AA Aruba ABW
#> 2 AC Antigua and Barbuda ATG
#> 3 AE Abu Dhabi - United Arab Emirates ARE
#> 4 AF Afghanistan AFG
#> 5 AG Algeria DZA
#> # ... with 245 more rowsLet's get the NOAA country code for Germany and filter all weather stations in Germany.
library(data.table)
# transform data.frame into data.table
country_concordance = as.data.table(country_concordance)
# filter country code
german_stations = country_concordance[country_code == "DEU", NOAA_ccode]
german_stations
#> [1] "GM" "GQ"
stations = as.data.table(stations)
# filter all German stations
station[country %in% german_stations]
#> station_id name country state icao lat lon elev begin_date end_date
#> 1: 090910-99999 ARKONA (CAPE) & GM NA <NA> 54.683 13.433 42.0 1975-07-01 2002-07-29
#> 2: 091610-99999 BOLTENHAGEN & GM NA <NA> 54.000 11.200 15.0 1975-07-01 1991-10-31
#> 3: 091620-99999 SCHWERIN & GM NA <NA> 53.633 11.417 59.0 1975-07-01 1992-05-17
#> 4: 091700-99999 WARNEMUENDE GM NA <NA> 54.183 12.083 4.0 1975-07-01 2001-11-21
#> 5: 091770-99999 TETEROW & GM NA <NA> 53.767 12.617 46.0 1975-07-01 1991-10-31
#> ---
#> 504: 999999-41406 GUAM WFO GQ NA PGUM 13.483 144.800 77.4 1945-08-31 1995-03-31
#> 505: 999999-41407 GUAM MARSHALL ISLAND GQ NA <NA> 13.467 144.250 54.9 1946-05-31 1946-12-31
#> 506: 999999-41411 GUAM OROTE GQ NA <NA> 13.433 144.633 24.1 1945-05-31 1948-05-29
#> 507: 999999-41414 GUAM NORTH AFB GQ NA <NA> 13.583 144.917 186.5 2004-07-01 2004-12-31
#> 508: 999999-41415 GUAM WFO GQ NA PGUM 13.483 144.800 76.2 1967-12-31 1972-12-31The last rows indicate that the NOAA country code GQ is not explicit. It is assigned to a few German stations, e.g. station_id = "104700-99999", name = "LEIPZIG /MOCKAU" but also to stations in Guam.
To find the closest weather station to any given location us the closest_weather_station() function. Let's find the closest weather station to Kiel, Germany.
# Five closest weather stations to Kiel
closest_weather_station(latitude = 54.323293, longitude = 10.122765, stations, n = 5)
#> distance station_id name country elev begin_date end_date
#> 1 1831 100450-99999 KIEL/KRONSHAGEN GM 19.0 1934-01-01 1990-03-01
#> 2 6475 100465-99999 KIEL HOLTENAU GM 30.8 1995-08-27 2023-05-21
#> 3 6877 100460-99999 KIEL/HOLTENAU(GNVY) GM 31.0 1929-08-01 2003-08-16
#> 4 7710 100480-99999 FRIEDRICHSORT GM 11.0 1975-05-23 2001-09-25
#> 5 22242 100440-99999 KIEL (LGT-H/AUT) GM 21.0 1937-08-15 2003-08-16To download the data for Kiel's weather in 2022, run the following code:
download_weather_data(
station_id = "100465-99999",
year = 2022,
dir = "temp"
)To load the data, run the following code:
read_weather_data(file = list.files("temp", full.names = TRUE))
#> # A tibble: 17,114 x 10
#> station_id time wd ws ceil_hgt visibility temp dew_point atmos_pres rh
#> <chr> <dttm> <int> <dbl> <int> <int> <dbl> <dbl> <dbl> <dbl>
#> 1 100465-99999 2022-01-01 00:20:00 270 6.2 152 6000 9 9 NA 100
#> 2 100465-99999 2022-01-01 00:50:00 280 6.2 122 3300 9 9 NA 100
#> 3 100465-99999 2022-01-01 01:20:00 270 5.1 122 9999 9 9 NA 100
#> 4 100465-99999 2022-01-01 01:50:00 260 5.1 183 9999 9 9 NA 100
#> 5 100465-99999 2022-01-01 02:20:00 260 5.7 183 9999 9 9 NA 100
#> 6 100465-99999 2022-01-01 02:50:00 260 5.1 183 9999 9 9 NA 100
#> 7 100465-99999 2022-01-01 03:20:00 260 5.1 244 9999 9 9 NA 100
#> 8 100465-99999 2022-01-01 03:50:00 270 6.2 244 9999 9 9 NA 100
#> 9 100465-99999 2022-01-01 04:20:00 270 5.1 244 9999 9 8 NA 93.5
#> 10 100465-99999 2022-01-01 04:50:00 260 4.6 518 9999 9 8 NA 93.5
# … with 17,104 more rows