igrf use

Koen Hufkens

2025-04-11

Single coordinates

IGRF values can be calculated for a single point using the below call. This will generate a data frame with model values. The routine is relatively fast so looping over a time vector will generate time series fairly quickly. To calculate grids a simple function is provided (see below).

# load the library
library(igrf)

df <- igrf(
  field = "main",
  year = 2000,
  altitude = 2,
  latitude = 0,
  longitude = 0
)

print(df)
#>          X         Y        Z        F         D        H         I
#> 1 27437.31 -3501.466 -14803.5 31372.12 -7.272602 27659.83 -28.15562

Regular grid

You may generate global maps of the IGRF using the igrf_grid() function, setting similar parameters as above while specifying a resolution as decimal degrees of the global grid. Keep in mind that values under 1 (fractions) will take an increasing amount of time and space to store the data.

grid <- igrf::igrf_grid(
  field = "main",
  year = 2000,
  altitude = 2,
  resolution = 5
  )

print(head(grid))
#>    lon lat         X         Y         Z        F        D        H         I
#> 1 -180 -90 -14558.63  7793.387 -53321.91 55820.39 151.8394 16513.34 -72.79269
#> 2 -175 -90 -13823.99  9032.599 -53321.91 55820.39 146.8394 16513.34 -72.79269
#> 3 -170 -90 -12984.14 10203.067 -53321.91 55820.39 141.8394 16513.34 -72.79269
#> 4 -165 -90 -12045.48 11295.884 -53321.91 55820.39 136.8394 16513.34 -72.79269
#> 5 -160 -90 -11015.14 12302.732 -53321.91 55820.39 131.8394 16513.34 -72.79269
#> 6 -155 -90  -9900.97 13215.949 -53321.91 55820.39 126.8394 16513.34 -72.79269
library(ggplot2)
library(metR)
library(rnaturalearth)

coast <- ne_coastline(returnclass = "sf")

ggplot(grid) +
  geom_contour_fill(
    aes(
      lon,
      lat,
      z = D
    ),
    breaks = MakeBreaks(10)
  ) +
  geom_sf(
    data = coast
  ) +
  scale_fill_divergent(
    name = "declination"
  ) +
  labs(
    x = "",
    y = "",
    title = "IGRF magnetic declination"
  ) +
  theme_bw() +
  theme(
    plot.background = element_rect(fill = "#ffffff", color = NA),
    panel.background = element_rect(fill = "#ffffff", color = NA)
  )