rnoaa is an R interface to many NOAA data sources. We don’t cover all of them, but we include many commonly used sources, and add we are always adding new sources. We focus on easy to use interfaces for getting NOAA data, and giving back data in easy to use formats downstream. We currently don’t do much in the way of plots or analysis.

Data sources in rnoaa

Help

Documentation is at https://docs.ropensci.org/rnoaa/, and there are many vignettes in the package itself, available in your R session, or on CRAN. The tutorials:

  • NOAA Buoy vignette
  • NOAA National Climatic Data Center (NCDC) vignette (examples)
  • NOAA NCDC attributes vignette
  • NOAA NCDC workflow vignette
  • Sea ice vignette
  • Severe Weather Data Inventory (SWDI) vignette
  • Historical Observing Metadata Repository (HOMR) vignette
  • Storms (IBTrACS) vignette
  • Complementing air quality data (ropenaq) with weather data using rnoaa

netcdf data

Some functions use netcdf files, including:

  • gefs
  • ersst
  • buoy
  • bsw
  • argo

You’ll need the ncdf4 package for those functions, and those only. ncdf4 is in Suggests in this package, meaning you only need ncdf4 if you are using any of the functions listed above. You’ll get an informative error telling you to install ncdf4 if you don’t have it and you try to use the those functions. Installation of ncdf4 should be straightforward on any system. See https://cran.r-project.org/package=ncdf4

NOAA NCDC Datasets

There are many NOAA NCDC datasets. All data sources work, except NEXRAD2 and NEXRAD3, for an unknown reason. This relates to ncdc_*() functions only.

Dataset Description Start Date End Date Data Coverage
GHCND Daily Summaries 1763-01-01 2019-09-24 1.00
GSOM Global Summary of the Month 1763-01-01 2019-08-01 1.00
GSOY Global Summary of the Year 1763-01-01 2019-01-01 1.00
NEXRAD2 Weather Radar (Level II) 1991-06-05 2019-09-24 0.95
NEXRAD3 Weather Radar (Level III) 1994-05-20 2019-09-22 0.95
NORMAL_ANN Normals Annual/Seasonal 2010-01-01 2010-01-01 1.00
NORMAL_DLY Normals Daily 2010-01-01 2010-12-31 1.00
NORMAL_HLY Normals Hourly 2010-01-01 2010-12-31 1.00
NORMAL_MLY Normals Monthly 2010-01-01 2010-12-01 1.00
PRECIP_15 Precipitation 15 Minute 1970-05-12 2014-01-01 0.25
PRECIP_HLY Precipitation Hourly 1900-01-01 2014-01-01 1.00

NOAA NCDC Attributes

Each NOAA dataset has a different set of attributes that you can potentially get back in your search. See http://www.ncdc.noaa.gov/cdo-web/datasets for detailed info on each dataset. We provide some information on the attributes in this package; see the vignette for attributes to find out more

NCDC Authentication

You’ll need an API key to use the NOAA NCDC functions (those starting with ncdc*()) in this package (essentially a password). Go to https://www.ncdc.noaa.gov/cdo-web/token to get one. You can’t use this package without an API key.

Once you obtain a key, there are two ways to use it.

a) Pass it inline with each function call (somewhat cumbersome)

ncdc(datasetid = 'PRECIP_HLY', locationid = 'ZIP:28801', datatypeid = 'HPCP', limit = 5, token =  "YOUR_TOKEN")

b) Alternatively, you might find it easier to set this as an option, either by adding this line to the top of a script or somewhere in your .rprofile

options(noaakey = "KEY_EMAILED_TO_YOU")

c) You can always store in permamently in your .Rprofile file.

Installation

GDAL

You’ll need GDAL installed first. You may want to use GDAL >= 0.9-1 since that version or later can read TopoJSON format files as well, which aren’t required here, but may be useful. Install GDAL:

Then when you install the R package rgdal (rgeos also requires GDAL), you’ll most likely need to specify where you’re gdal-config file is on your machine, as well as a few other things. I have an OSX Mavericks machine, and this works for me (there’s no binary for Mavericks, so install the source version):

install.packages("https://cran.r-project.org/src/contrib/rgdal_0.9-1.tar.gz", repos = NULL, type="source", configure.args = "--with-gdal-config=/Library/Frameworks/GDAL.framework/Versions/1.10/unix/bin/gdal-config --with-proj-include=/Library/Frameworks/PROJ.framework/unix/include --with-proj-lib=/Library/Frameworks/PROJ.framework/unix/lib")

The rest of the installation should be easy. If not, let us know.

Stable version from CRAN

or development version from GitHub

remotes::install_github("ropensci/rnoaa")

Load rnoaa

library('rnoaa')

NCDC v2 API data

Fetch list of city locations in descending order

ncdc_locs(locationcategoryid='CITY', sortfield='name', sortorder='desc')
#> $meta
#> $meta$totalCount
#> [1] 1987
#> 
#> $meta$pageCount
#> [1] 25
#> 
#> $meta$offset
#> [1] 1
#> 
#> 
#> $data
#>       mindate    maxdate                  name datacoverage            id
#> 1  1892-08-01 2019-07-31            Zwolle, NL       1.0000 CITY:NL000012
#> 2  1901-01-01 2019-09-22            Zurich, SZ       1.0000 CITY:SZ000007
#> 3  1957-07-01 2019-09-22         Zonguldak, TU       1.0000 CITY:TU000057
#> 4  1906-01-01 2019-09-22            Zinder, NG       0.9025 CITY:NG000004
#> 5  1973-01-01 2019-09-22        Ziguinchor, SG       1.0000 CITY:SG000004
#> 6  1938-01-01 2019-09-22         Zhytomyra, UP       0.9723 CITY:UP000025
#> 7  1948-03-01 2019-09-22        Zhezkazgan, KZ       0.9302 CITY:KZ000017
#> 8  1951-01-01 2019-09-22         Zhengzhou, CH       1.0000 CITY:CH000045
#> 9  1941-01-01 2019-06-30          Zaragoza, SP       1.0000 CITY:SP000021
#> 10 1936-01-01 2009-06-17      Zaporiyhzhya, UP       1.0000 CITY:UP000024
#> 11 1957-01-01 2019-09-22          Zanzibar, TZ       0.8016 CITY:TZ000019
#> 12 1973-01-01 2019-09-22            Zanjan, IR       0.9105 CITY:IR000020
#> 13 1893-01-01 2019-09-24     Zanesville, OH US       1.0000 CITY:US390029
#> 14 1912-01-01 2019-09-22             Zahle, LE       0.9819 CITY:LE000004
#> 15 1951-01-01 2019-09-22           Zahedan, IR       0.9975 CITY:IR000019
#> 16 1860-12-01 2019-09-22            Zagreb, HR       1.0000 CITY:HR000002
#> 17 1929-07-01 2019-09-22         Zacatecas, MX       1.0000 CITY:MX000036
#> 18 1947-01-01 2019-09-22 Yuzhno-Sakhalinsk, RS       1.0000 CITY:RS000081
#> 19 1893-01-01 2019-09-24           Yuma, AZ US       1.0000 CITY:US040015
#> 20 1942-02-01 2019-09-24   Yucca Valley, CA US       1.0000 CITY:US060048
#> 21 1885-01-01 2019-09-24      Yuba City, CA US       1.0000 CITY:US060047
#> 22 1998-02-01 2019-09-22            Yozgat, TU       0.9993 CITY:TU000056
#> 23 1893-01-01 2019-09-24     Youngstown, OH US       1.0000 CITY:US390028
#> 24 1894-01-01 2019-09-24           York, PA US       1.0000 CITY:US420024
#> 25 1869-01-01 2019-09-24        Yonkers, NY US       1.0000 CITY:US360031
#> 
#> attr(,"class")
#> [1] "ncdc_locs"

Search for data

out <- ncdc(datasetid='NORMAL_DLY', stationid='GHCND:USW00014895', datatypeid='dly-tmax-normal', startdate = '2010-05-01', enddate = '2010-05-10')

See a data.frame

Note that the value column has strangely large numbers for temperature measurements. By convention, rnoaa doesn’t do any conversion of values from the APIs and some APIs use seemingly odd units.

You have two options here:

  1. Use the add_units parameter on ncdc to have rnoaa attempt to look up the units. This is a good idea to try first.

  2. Consult the documentation for whiechever dataset you’re accessing. In this case, GHCND has a README which indicates TMAX is measured in tenths of degrees Celcius.

See a data.frame with units

As mentioned above, you can use the add_units parameter with ncdc() to ask rnoaa to attempt to look up units for whatever data you ask it to return. Let’s ask rnoaa to add units to some precipitation (PRCP) data:

From the above output, we can see that the units for PRCP values are “mm_tenths” which means tenths of a millimeter. You won’t always be so lucky and sometimes you will have to look up the documentation on your own.

Plot data, super simple, but it’s a start

out <- ncdc(datasetid='GHCND', stationid='GHCND:USW00014895', datatypeid='PRCP', startdate = '2010-05-01', enddate = '2010-10-31', limit=500)
ncdc_plot(out, breaks="1 month", dateformat="%d/%m")

plot of chunk unnamed-chunk-14

Note that PRCP values are in units of tenths of a millimeter, as we found out above.

More plotting

You can pass many outputs from calls to the noaa function in to the ncdc_plot function.

out1 <- ncdc(datasetid='GHCND', stationid='GHCND:USW00014895', datatypeid='PRCP', startdate = '2010-03-01', enddate = '2010-05-31', limit=500)
out2 <- ncdc(datasetid='GHCND', stationid='GHCND:USW00014895', datatypeid='PRCP', startdate = '2010-09-01', enddate = '2010-10-31', limit=500)
ncdc_plot(out1, out2, breaks="45 days")

plot of chunk unnamed-chunk-15

HOMR metadata

In this example, search for metadata for a single station ID

homr(qid = 'COOP:046742')

Storm data

Get storm data for the year 2010

storm_data(year = 2010)

GEFS data

Get forecast for a certain variable.

Argo buoys data

There are a suite of functions for Argo data, a few egs:

# Spatial search - by bounding box
argo_search("coord", box = c(-40, 35, 3, 2))

# Time based search
argo_search("coord", yearmin = 2007, yearmax = 2009)

# Data quality based search
argo_search("coord", pres_qc = "A", temp_qc = "A")

# Search on partial float id number
argo_qwmo(qwmo = 49)

# Get data
argo(dac = "meds", id = 4900881, cycle = 127, dtype = "D")

Meta

rofooter