Skip to contents

NOAA North American Regional Reanalysis (NARR)

Mitchell Manware

2025-08-14

Source: vignettes/narr_workflow.Rmd
narr_workflow.Rmd

This vignette demonstrates how to download, process, and calculate covariates from the NOAA North American Regional Reanalysis (NARR) dataset using amadeus functions. Details are provided for each function’s parameters and outputs. The examples utilize daily air temperature at 2m height (“air.2m”) data. The messages returned by amadeus functions have been omitted for brevity.

Download

Start by downloading the netCDF data files with download_data.

  • dataset_name = "narr": NARR dataset acronym.
  • variable = "air.2m": air temperature at 2m height variable code.
  • year = c(2021, 2022): years of interest.
  • directory_to_save = dir: directory to save the downloaded files.
  • acknowledgement = TRUE: acknowledge that the raw data files are large and may consume lots of local storage.
  • download = TRUE: download the data files.
  • remove_command = TRUE: remove the temporary command file used to download the data.
  • hash = TRUE: generate unique SHA-1 hash for the downloaded files.
dir <- tempdir()
amadeus::download_data(
  dataset_name = "narr",
  variable = "air.2m",
  year = c(2021, 2022),
  directory_to_save = dir,
  acknowledgement = TRUE,
  download = TRUE,
  remove_command = TRUE,
  hash = TRUE
)
[1] "3a382ac1c383c1d048f4044214cb450f"

Check the downloaded netCDF files.

list.files(dir, recursive = TRUE, pattern = "air.2m")
[1] "air.2m/air.2m.2021.nc" "air.2m/air.2m.2022.nc"

Process

Import and process the downloaded netCDF files with process_covariates.

  • covariate = "narr": NARR dataset acronym.
  • variable = "air.2m": air temperature at 2m height variable code.
  • date = c("2021-12-28", "2022-01-03"): date range of interest.
  • path = paste0(dir, "/air.2m"): directory containing the downloaded files.
air2m_process <- amadeus::process_covariates(
  covariate = "narr",
  variable = "air.2m",
  date = c("2021-12-28", "2022-01-03"),
  path = file.path(dir, "/air.2m")
)

Check the processed SpatRaster object.

air2m_process
class       : SpatRaster
dimensions  : 277, 349, 7  (nrow, ncol, nlyr)
resolution  : 32462.99, 32463  (x, y)
extent      : -16231.49, 11313351, -16231.5, 8976020  (xmin, xmax, ymin, ymax)
coord. ref. : +proj=lcc +lat_0=50 +lon_0=-107 +lat_1=50 +lat_2=50 +x_0=5632642.22547 +y_0=4612545.65137 +datum=WGS84 +units=m +no_defs
sources     : air.2m.2021.nc:air  (4 layers)
              air.2m.2022.nc:air  (3 layers)
varnames    : air (Daily Air Temperature at 2 m)
              air (Daily Air Temperature at 2 m)
names       : air.2~11228, air.2~11229, air.2~11230, air.2~11231, air.2~20101, air.2~20102, ...
unit        :           K,           K,           K,           K,           K,           K, ...
time        : 2021-12-28 to 2022-01-03 UTC
terra::plot(air2m_process[[1]])

Calculate covariates

Calculate covariates for North Carolina county boundaries with calculate_covariates. County boundaries are accessed with the tigris::counties function.

  • covariate = "narr": NARR dataset acronym.
  • from = air2m_process: processed SpatRaster object.
  • locs = tigris::counties("NC", year = 2021): North Carolina county boundaries.
  • locs_id = "NAME": county name identifier.
  • radius = 0: size of buffer radius around each county.
  • geom = "terra": return covariates as a SpatVector object.
library(tigris)
air2m_covar <- amadeus::calculate_covariates(
  covariate = "narr",
  from = air2m_process,
  locs = tigris::counties("NC", year = 2021),
  locs_id = "NAME",
  radius = 0,
  geom = "terra"
)

Check the calculated covariates SpatVector object.

air2m_covar
class       : SpatVector
geometry    : polygons
dimensions  : 700, 3  (geometries, attributes)
extent      : 7731783, 8506154, 3248490, 3694532  (xmin, xmax, ymin, ymax)
coord. ref. : +proj=lcc +lat_0=50 +lon_0=-107 +lat_1=50 +lat_2=50 +x_0=5632642.22547 +y_0=4612545.65137 +datum=WGS84 +units=m +no_defs
names       :     NAME       time air.2m_0
type        :    <chr>   <POSIXt>    <num>
values      :  Chatham 2021-12-28    289.3
              Alamance 2021-12-28    288.8
              Davidson 2021-12-28    289.1

Temporal summaries

The aggregate function can be used to calculate a summary statistic for each unique spatial point or polygon. In the following example, average air.2m_0 is calculated for each county for the time period December 28, 2021 to January 3, 2022. air.2m_0 ~ NAME directs the function to summarize air.2m_0 values per unique NAME. The FUN = mean directs the function to take the mean value. The head() function is applied to show only the first few entries, as the entire data.frame is 100 rows long.

head(aggregate(air.2m_0 ~ NAME, data = air2m_covar, FUN = mean))
       NAME air.2m_0
1  Alamance 289.5930
2 Alexander 289.1961
3 Alleghany 286.9486
4     Anson 290.5306
5      Ashe 285.5771
6     Avery 285.2288