GSODR

Adam H Sparks

Introduction

The GSOD or Global Surface Summary of the Day (GSOD) data provided by the US National Centers for Environmental Information (NCEI) are a valuable source of weather data with global coverage. However, the data files are cumbersome and difficult to work with. GSODR aims to make it easy to find, transfer and format the data you need for use in analysis and provides four main functions for facilitating this:

When reformatting data either with get_GSOD() or reformat_GSOD(), all units are converted to International System of Units (SI), e.g., inches to millimetres and Fahrenheit to Celsius. File output can be used in an R session as a tibble(), saved as a Comma Separated Value (CSV) file or in a spatial GeoPackage (GPKG) file, implemented by most major GIS software, summarising each year by station, which also includes vapour pressure and relative humidity elements calculated from existing data in GSOD.

For more information see the description of the data provided by NCEI, http://www7.ncdc.noaa.gov/CDO/GSOD_DESC.txt.

Using get_GSOD()

Find Stations in Australia

GSODR provides lists of weather station locations and elevation values. Using dplyr, we can find all the stations in Australia.

## 
## GSOD is distributed free by the U.S. NCEI with the
## following conditions.
## 'The following data and products may have conditions placed
## their international commercial use. They can be used within
## the U.S. or for non-commercial international activities
## without restriction. The non-U.S. data cannot be
## redistributed for commercial purposes. Re-distribution of
## these data by others must provide this same notification.
## WMO Resolution 40. NOAA Policy'
## 
## GSODR does not redistribute any weather data itself. It 
## only provides an interface for R users to download these
## data, however it does redistribute station metadata in the
## package.
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
## # A tibble: 1,412 x 16
##      USAF  WBAN                      STN_NAME  CTRY STATE  CALL     LAT
##     <chr> <chr>                         <chr> <chr> <chr> <chr>   <dbl>
##  1 695023 99999           HORN ISLAND   (HID)    AS  <NA>  KQXC -10.583
##  2 749430 99999            AIDELAIDE RIVER SE    AS  <NA>  <NA> -13.300
##  3 749432 99999     BATCHELOR FIELD AUSTRALIA    AS  <NA>  <NA> -13.049
##  4 749438 99999          IRON RANGE AUSTRALIA    AS  <NA>  <NA> -12.700
##  5 749439 99999      MAREEBA AS/HOEVETT FIELD    AS  <NA>  <NA> -17.050
##  6 749440 99999                     REID EAST    AS  <NA>  <NA> -19.767
##  7 749441 99999  TOWNSVILLE AUSTRALIA/GARBUTT    AS  <NA>  ABTL -19.249
##  8 749442 99999                     WOODSTOCK    AS  <NA>  <NA> -19.600
##  9 749443 99999 JACKY JACKY AUSTRALIA/HIGGINS    AS  <NA>  <NA> -10.933
## 10 749455 99999            LAKE BUCHANAN WEST    AS  <NA>  <NA> -21.417
## # ... with 1,402 more rows, and 9 more variables: LON <dbl>, ELEV_M <dbl>,
## #   BEGIN <dbl>, END <dbl>, STNID <chr>, ELEV_M_SRTM_90m <dbl>,
## #   COUNTRY_NAME <chr>, iso2c <chr>, iso3c <chr>
## # A tibble: 2 x 16
##     USAF  WBAN          STN_NAME  CTRY STATE  CALL     LAT     LON ELEV_M
##    <chr> <chr>             <chr> <chr> <chr> <chr>   <dbl>   <dbl>  <dbl>
## 1 945510 99999         TOOWOOMBA    AS  <NA>  <NA> -27.583 151.933    676
## 2 955510 99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.550 151.917    642
## # ... with 7 more variables: BEGIN <dbl>, END <dbl>, STNID <chr>,
## #   ELEV_M_SRTM_90m <dbl>, COUNTRY_NAME <chr>, iso2c <chr>, iso3c <chr>

Download a Single Station and Year

Now that we’ve seen where the reporting stations are located, we can download weather data from the station Toowoomba, Queensland, Australia for 2010 by using the STNID in the station parameter of get_GSOD().

## # A tibble: 365 x 48
##      USAF  WBAN        STNID          STN_NAME  CTRY STATE  CALL    LAT
##     <chr> <chr>        <chr>             <chr> <chr> <chr> <chr>  <dbl>
##  1 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  2 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  3 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  4 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  5 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  6 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  7 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  8 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
##  9 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
## 10 955510 99999 955510-99999 TOOWOOMBA AIRPORT    AS  <NA>  <NA> -27.55
## # ... with 355 more rows, and 40 more variables: LON <dbl>, ELEV_M <dbl>,
## #   ELEV_M_SRTM_90m <dbl>, BEGIN <dbl>, END <dbl>, YEARMODA <date>,
## #   YEAR <chr>, MONTH <chr>, DAY <chr>, YDAY <dbl>, TEMP <dbl>,
## #   TEMP_CNT <int>, DEWP <dbl>, DEWP_CNT <int>, SLP <dbl>, SLP_CNT <int>,
## #   STP <dbl>, STP_CNT <int>, VISIB <dbl>, VISIB_CNT <int>, WDSP <dbl>,
## #   WDSP_CNT <int>, MXSPD <dbl>, GUST <dbl>, MAX <dbl>, MAX_FLAG <chr>,
## #   MIN <dbl>, MIN_FLAG <chr>, PRCP <dbl>, PRCP_FLAG <chr>, SNDP <dbl>,
## #   I_FOG <int>, I_RAIN_DRIZZLE <int>, I_SNOW_ICE <int>, I_HAIL <int>,
## #   I_THUNDER <int>, I_TORNADO_FUNNEL <int>, EA <dbl>, ES <dbl>, RH <dbl>

Using nearest_stations()

Using the nearest_stations() function, you can find stations closest to a given point specified by latitude and longitude in decimal degrees. This can be used to generate a vector to pass along to get_GSOD() and download the stations of interest.

There are missing stations in this query. Not all that are listed and queried actually have files on the server.

If you wished to drop the stations, 949999-00170 and 949999-00183 from the query, you could do this.

Plot Maximum and Minimum Temperature Values

Using the first data downloaded for a single station, 955510-99999, plot the temperature for 2010 using read_csv() from Hadley’s readr package.

## 
## Attaching package: 'lubridate'
## The following object is masked from 'package:base':
## 
##     date

Creating Spatial Files

Because the stations provide geospatial location information, it is possible to create a spatial file. GeoPackage files are a open, standards-based, platform-independent, portable, self-describing compact format for transferring geospatial information, which handle vector files much like shapefiles do, but eliminate many of the issues that shapefiles have with field names and the number of files. The get_GSOD() function can create a GeoPackage file, which can be used with a GIS for further analysis and mapping with other spatial objects.

After getting weather stations for Australia and creating a GeoPackage file, rgdal can import the data into R again in a spatial format.

Importing the GeoPackage file can be a bit tricky. The dsn will be the full path along with the file name. The layer to be specified is “GSOD”, this is specified in the get_GSOD() function and will not change. The file name, specified in the dsn will, but the layer name will not.

Since GeoPackage files are formatted as SQLite databases you can use the existing R tools for SQLite files (J. Stachelek 2016). One easy way is using dplyr, which we’ve already used to filter the stations.

This option is much faster to load since it does not load the geometry.

Using reformat_GSOD()

You may have already downloaded GSOD data or may just wish to use an FTP client to download the files from the server to you local disk and not use the capabilities of get_GSOD(). In that case the reformat_GSOD() function is useful.

There are two ways, you can either provide reformat_GSOD() with a list of specified station files or you can supply it with a directory containing all of the “WBAN-WMO-YYYY.op.gz” station files that you wish to reformat.

Reformat a list of local files

Reformat all local files found in directory

Using update_station_list()

GSODR uses internal databases of station data from the NCEI to provide location and other metadata, e.g. elevation, station names, WMO codes, etc. to make the process of querying for weather data faster. This database is created and packaged with GSODR for distribution and is updated with new releases. Users have the option of updating these databases after installing GSODR. While this option gives the users the ability to keep the database up-to-date and gives GSODR’s authors flexibility in maintaining it, this also means that reproducibility may be affected since the same version of GSODR may have different databases on different machines. If reproducibility is necessary, care should be taken to ensure that the version of the databases is the same across different machines.

The database file isd_history.rda can be located on your local system by using the following command, paste0(.libPaths(), "/GSODR/extdata")[1], unless you have specified another location for library installations and installed GSODR there, in which case it would still be in GSODR/extdata.

To update GSODR’s internal database of station locations simply use update_station_list(), which will update the internal station database according to the latest data available from the NCEI.

Using get_inventory()

GSODR provides a function, get_inventory() to retrieve an inventory of the number of weather observations by station-year-month for the beginning of record through to current.

Following is an example of how to retrieve the inventory and check a station in Toowoomba, Queensland, Australia, which was used in an earlier example.

inventory <- get_inventory()

inventory

subset(inventory, STNID == "955510-99999")

Additional Climate Data Availability

Additional climate data, GSODRdata, formatted for use with GSOD data provided by GSODR are available as an R package, which can only be installed through GitHub due to the package size, 5.1Mb, being too large for CRAN.

#install.packages("devtools")
devtools::install_github("adamhsparks/GSODRdata")
library("GSODRdata")

Notes

Elevation Values

90 metre (90m) hole-filled SRTM digital elevation (Jarvis et al. 2008) was used to identify and correct/remove elevation errors in data for station locations between -60˚ and 60˚ latitude. This applies to cases here where elevation was missing in the reported values as well. In case the station reported an elevation and the DEM does not, the station reported is taken. For stations beyond -60˚ and 60˚ latitude, the values are station reported values in every instance. See https://github.com/ropensci/GSODR/blob/master/data-raw/fetch_isd-history.md for more detail on the correction methods.

WMO Resolution 40. NOAA Policy

Users of these data should take into account the following (from the NCEI website):

“The following data and products may have conditions placed on their international commercial use. They can be used within the U.S. or for non-commercial international activities without restriction. The non-U.S. data cannot be redistributed for commercial purposes. Re-distribution of these data by others must provide this same notification.” WMO Resolution 40. NOAA Policy

References

Stachelek, J. (2016) Using the Geopackage Format with R. URL: https://jsta.github.io/2016/07/14/geopackage-r.html

Appendices

Appendix 1: GSODR Final Data Format, Contents and Units

GSODR formatted data include the following fields and units:

Appendix 2: Map of Current GSOD Station Locations