noaanclimgrid

Overview

R tools to fetch and visualize NOAA’s Monthly U.S. Climate Gridded Dataset (NClimGrid):

The NOAA Monthly U.S. Climate Gridded Dataset (NClimGrid) consists of four climate variables derived from the GHCN-D dataset: maximum temperature, minimum temperature, average temperature and precipitation. Each file provides monthly values in a 5x5 lat/lon grid for the Continental United States. Data is available from 1895 to the present.

Basic functions include:

• get_nclimgrid_monthly() Downloads and formats nClimGrid monthly data for recent years
• get_nclimgrid_normals() Downloads and formats nClimGrid normals data for ranges of period of record
• compute_anomaly() Computes the difference between two datasets of the same measurement type

Some plotting functions are included to visualize measurements and anomalies:

• plot_nclimgrid() Creates a monthly faceted plot of measurement (either monthly or normals) or anomaly
• plot_nclimgrid_histogram() Creates a histogram (if evaluating one dataset) or mirrored histogram (if comparing two datasets)

This package is not created or maintained by NOAA.

Installation

# Development version from GitHub:
devtools::install_github("toozler/noaa_nclimgrid_r")

Usage

Fetching and manipulating data

NOAA stores the nClimGrid dataset in yearly files containing all months (up to previous month for the current year). The get_nclimgrid_*() functions fetch this data and perform some basic parsing:

nclim_monthly_data <- get_nclimgrid_monthly(year = 2023, 
                                            measurement = "tave", 
                                            region = "us", 
                                            wide = FALSE, 
                                            verbose = FALSE)
nclim_monthly_data %>% str
#> tibble [5,637,096 × 4] (S3: tbl_df/tbl/data.frame)
#>  $ lat  : num [1:5637096] 24.6 24.6 24.6 24.6 24.6 ...
#>  $ long : num [1:5637096] -81.8 -81.8 -81.8 -81.8 -81.8 ...
#>  $ month: Factor w/ 12 levels "1","2","3","4",..: 1 2 3 4 5 6 7 8 9 10 ...
#>  $ value: num [1:5637096] 73 77.1 77.9 80.1 81.9 ...
#>  - attr(*, "year")= num 2023
#>  - attr(*, "measurement")= chr "tave"
#>  - attr(*, "region")= chr "us"
#>  - attr(*, "unit")= chr "Fahrenheit"
#>  - attr(*, "wide")= logi FALSE
#>  - attr(*, "anomaly_df")= logi FALSE

The example above pulls average temperature (tave) for the Continental US (us) in 2023. Available datasets include also tmin (minimum temperatures), tmax (maximum temperatures) and prcp (precipitation). More details about the nClimGrid dataset are available here. As default, the data frame is converted to a long format for easier integration with tidyverse packages. Metadata is stored in each data frame as attributes.

nclim_normals_data <- get_nclimgrid_normals(period = "1901-2000", 
                                            measurement = "tave", 
                                            region = "us")

The above example will pull the normals (typical values) for a given set of available periods of record. By default, “1901-2000” (20th Century) is used. Other available periods can be listed with validate_normals_period()

Same measurement type datasets (i.e. temperature or precipitation) can be compared by using the compute_anomaly() function, which returns the difference between the two datasets. Months not present on the LHS are not included (i.e. your monthly measurements only cover a few months of the year).

tave_anomaly <- compute_anomaly(nclimgrid_data_a = nclim_monthly_data, 
                                nclimgrid_data_b = nclim_normals_data)

tave_anomaly %>% str
#> tibble [5,637,096 × 4] (S3: tbl_df/tbl/data.frame)
#>  $ lat  : num [1:5637096] 24.6 24.6 24.6 24.6 24.6 ...
#>  $ long : num [1:5637096] -81.8 -81.8 -81.8 -81.8 -81.8 ...
#>  $ month: Factor w/ 12 levels "1","2","3","4",..: 1 2 3 4 5 6 7 8 9 10 ...
#>  $ value: num [1:5637096] 3.62 7.11 5.28 3.79 2.29 ...
#>  - attr(*, "year")= num 2023
#>  - attr(*, "measurement")= chr "tave"
#>  - attr(*, "region")= chr "us"
#>  - attr(*, "unit")= chr "Fahrenheit"
#>  - attr(*, "wide")= logi FALSE
#>  - attr(*, "anomaly_df")= logi TRUE

Visualization

You can visualize nClimGrid data with plot_nclimgrid(). The month column is automatically faceted as only one month can be displayed per each map. Values are automatically labelled with the month’s name. A default title and subtitle (where applicable) is generated based on the data being plotted.

nclim_monthly_data %>% 
  filter(month %in% 2:3) %>% 
  plot_nclimgrid(facet_cols = 1, 
                 show_credit = T)

Anomalies can be visualized by plotting the output of compute_anomaly().

temperature_anomaly <- compute_anomaly(nclim_monthly_data, 
                                       nclim_normals_data)

temperature_anomaly %>% 
  filter(month %in% 1:4) %>% 
  plot_nclimgrid(facet_col = 2)

The output of the plotting function is a ggplot2 object, which allows you tack on additional geoms (i.e. to add features like points and labels), to remove features (i.e. remove titles and facet labels) or to modify features (i.e. theme options):

require(maps)
#> Loading required package: maps
require(ggrepel)
#> Loading required package: ggrepel

nclim_monthly_prcp_data <- get_nclimgrid_monthly(year = 2023, 
                                                 measurement = "prcp", 
                                                 region = "us")
large_us_capitals <- maps::us.cities %>% 
  filter(capital != 0, pop >= 400000) %>% 
  mutate(name = gsub(" ", ", ", name))

nclim_monthly_prcp_data %>%
  filter(month %in% 7) %>%
  plot_nclimgrid(show_credit = F, show_states = F) +
  geom_point(data = large_us_capitals, aes(x=long, y=lat)) +
  geom_label_repel(data = large_us_capitals, aes(x=long, y=lat, label=name),
                    box.padding = 1, segment.alpha = 0.5, size = 2.5) +
  labs(title = NULL, 
       subtitle = NULL) +
  theme_void() +
  theme(legend.position = 'none', 
        strip.text = element_blank())

You can visualize and compare the distribution of values from monthly, normals or anomaly datasets with plot_nclimgrid_histogram().

Future features

• Subset data by state or region
• Fetch data from nClimGrid’s archive of monthly data from 1895 onwards.
• Option to compute anomaly also as a percentage
• Function to compute ranks based on normal percentiles - this dataset is not readily available at NCIS.