Local Coordinates and the Ellipsoid

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
library(geographiclib)

Contents

Example Locations

# Locations at various latitudes
world_pts <- cbind(
  lon = c(151.21, 0, -74.01, 0, 0),
  lat = c(-33.87, 51.51, 40.71, 0, -90)
)
rownames(world_pts) <- c("Sydney", "London", "New York", "Equator/Prime", "South Pole")

# Antarctic stations
antarctic <- cbind(
  lon = c(166.67, 77.85, -68.13, 39.58, 0),
  lat = c(-77.85, -67.60, -67.57, -69.41, -90)
)
rownames(antarctic) <- c("McMurdo", "Davis", "Palmer", "Progress", "South Pole")

Geocentric (ECEF) Coordinates

Earth-Centered Earth-Fixed (ECEF) coordinates express positions as X, Y, Z relative to the Earth's center:

Basic Conversion

geocentric_fwd(world_pts)

Understanding ECEF

# Points on the equator have Z ≈ 0
equator_pts <- cbind(
  lon = c(0, 90, 180, -90),
  lat = c(0, 0, 0, 0)
)
geocentric_fwd(equator_pts)

# Points at poles have X ≈ 0, Y ≈ 0
pole_pts <- cbind(lon = c(0, 0), lat = c(90, -90))
geocentric_fwd(pole_pts)

Height Above Ellipsoid

ECEF can include height above the WGS84 ellipsoid:

# Ground level vs airplane altitude (10km) vs satellite (400km)
pt <- c(151.21, -33.87)

data.frame(
  location = c("Ground", "Aircraft (10km)", "ISS (~400km)", "GPS satellite (~20200km)"),
  h = c(0, 10000, 400000, 20200000),
  geocentric_fwd(cbind(rep(pt[1], 4), rep(pt[2], 4)), 
                 h = c(0, 10000, 400000, 20200000))
)

Antarctic Stations in ECEF

# Antarctic stations are all near the "bottom" of the coordinate system
geocentric_fwd(antarctic)

# Note the large negative Z values (Southern Hemisphere)
# and relatively small X, Y values (near the axis)

Round-trip Conversion

fwd <- geocentric_fwd(world_pts)
rev <- geocentric_rev(fwd$X, fwd$Y, fwd$Z)

# Verify accuracy
max(abs(rev$lon - world_pts[,1]))
max(abs(rev$lat - world_pts[,2]))
max(abs(rev$h))  # Height should be ~0

GPS Applications

ECEF is the native coordinate system for GPS satellites:

# Convert GPS receiver position to geodetic
# (Example: receiver at Sydney at ~100m altitude)
X <- 4648241   # meters
Y <- -2560342
Z <- -3526276

geocentric_rev(X, Y, Z)

Local Cartesian (ENU) Coordinates

Local Cartesian, also called ENU (East-North-Up), creates a local coordinate system with:

This is ideal for local surveys, robotics, and navigation.

Basic Conversion

# Set up local system centered on Sydney
sydney <- c(151.21, -33.87)

nearby_pts <- cbind(
  lon = c(151.21, 151.31, 151.11, 151.21, 151.21),
  lat = c(-33.87, -33.87, -33.87, -33.77, -33.97)
)
rownames(nearby_pts) <- c("Origin", "East 10km", "West 10km", "North 10km", "South 10km")

localcartesian_fwd(nearby_pts, lon0 = sydney[1], lat0 = sydney[2])

Local Survey Application

# Survey points around McMurdo Station
mcmurdo <- c(166.67, -77.85)

survey_pts <- cbind(
  lon = c(166.67, 166.70, 166.64, 166.67, 166.73),
  lat = c(-77.85, -77.85, -77.85, -77.84, -77.86)
)
rownames(survey_pts) <- c("Base", "Point A", "Point B", "Point C", "Point D")

result <- localcartesian_fwd(survey_pts, lon0 = mcmurdo[1], lat0 = mcmurdo[2])
result

# Distances from base in meters
data.frame(
  point = rownames(survey_pts),
  east_m = round(result$x),
  north_m = round(result$y),
  distance_m = round(sqrt(result$x^2 + result$y^2))
)

Including Height

# Local system with height differences
# Simulating a hill near Sydney
pts_with_height <- cbind(
  lon = c(151.21, 151.22, 151.20),
  lat = c(-33.87, -33.87, -33.86)
)
heights <- c(0, 50, 100)  # meters above ellipsoid

result <- localcartesian_fwd(pts_with_height, 
                              lon0 = 151.21, lat0 = -33.87, h0 = 0,
                              h = heights)
result

Round-trip Conversion

pts <- cbind(
  lon = c(166.67, 166.70, 166.64),
  lat = c(-77.85, -77.84, -77.86)
)

fwd <- localcartesian_fwd(pts, lon0 = 166.67, lat0 = -77.85)
rev <- localcartesian_rev(fwd$x, fwd$y, fwd$z, lon0 = 166.67, lat0 = -77.85)

max(abs(rev$lon - pts[,1]))
max(abs(rev$lat - pts[,2]))

Robotics/Navigation Application

# Robot path planning at Davis Station
davis <- c(77.85, -67.60)

# Waypoints for a robot traverse
waypoints_enu <- data.frame(
  name = c("Start", "WP1", "WP2", "WP3", "End"),
  x = c(0, 100, 200, 250, 300),    # East (meters)
  y = c(0, 50, 100, 100, 150),     # North (meters)
  z = c(0, 0, 0, 0, 0)             # Up (meters)
)

# Convert to geographic coordinates for GPS navigation
result <- localcartesian_rev(
  waypoints_enu$x, waypoints_enu$y, waypoints_enu$z,
  lon0 = davis[1], lat0 = davis[2]
)

data.frame(
  waypoint = waypoints_enu$name,
  lon = round(result$lon, 6),
  lat = round(result$lat, 6)
)

WGS84 Ellipsoid Properties

The WGS84 ellipsoid is the reference surface for GPS and most modern mapping.

Basic Parameters

ellipsoid_params()

Key parameters: - a: Semi-major axis (equatorial radius) ≈ 6,378,137 m - b: Semi-minor axis (polar radius) ≈ 6,356,752 m - f: Flattening ≈ 1/298.257 - e2: First eccentricity squared - area: Total surface area - volume: Total volume

Earth's Shape

params <- ellipsoid_params()

# Equatorial vs polar radius difference
equatorial_radius <- params$a
polar_radius <- params$b

cat("Equatorial radius:", equatorial_radius, "m\n")
cat("Polar radius:", polar_radius, "m\n")
cat("Difference:", equatorial_radius - polar_radius, "m\n")
cat("Flattening:", 1/params$f, "(1/f)\n")

Radii of Curvature

The Earth's curvature varies with latitude:

# Curvature at various latitudes
lats <- c(0, -33.87, -42.88, -67.60, -77.85, -90)
names_lat <- c("Equator", "Sydney", "Hobart", "Davis", "McMurdo", "South Pole")

result <- ellipsoid_curvature(lats)

data.frame(
  location = names_lat,
  latitude = lats,
  meridional_km = round(result$meridional / 1000, 2),
  transverse_km = round(result$transverse / 1000, 2)
)

Circle of Latitude Properties

# Properties of circles at different latitudes
lats <- c(0, -30, -45, -60, -75, -90)

result <- ellipsoid_circle(lats)

data.frame(
  latitude = lats,
  circle_radius_km = round(result$radius / 1000, 2),
  meridian_dist_km = round(result$meridian_distance / 1000, 2)
)

Auxiliary Latitudes

For advanced geodetic calculations, various auxiliary latitudes are used:

# Compare different latitude types at 45°S
lats <- c(0, -30, -45, -60, -90)

result <- ellipsoid_latitudes(lats)
result

# The differences are small but matter for precise calculations

Combining Coordinate Systems

GNSS Data Processing

# Typical GNSS workflow:
# 1. Receive ECEF coordinates from GPS
# 2. Convert to geodetic (lat/lon/height)
# 3. Convert to local for navigation

# Example GPS receiver data (ECEF, meters)
gps_ecef <- data.frame(
  time = 1:5,
  X = c(4648241, 4648242, 4648240, 4648243, 4648241),
  Y = c(-2560342, -2560340, -2560343, -2560341, -2560342),
  Z = c(-3526276, -3526275, -3526277, -3526274, -3526276)
)

# Convert to geodetic
geodetic <- geocentric_rev(gps_ecef$X, gps_ecef$Y, gps_ecef$Z)

# Convert to local (relative to first point)
local <- localcartesian_fwd(
  cbind(geodetic$lon, geodetic$lat),
  lon0 = geodetic$lon[1], lat0 = geodetic$lat[1], h0 = geodetic$h[1],
  h = geodetic$h
)

data.frame(
  time = gps_ecef$time,
  east_m = round(local$x, 2),
  north_m = round(local$y, 2),
  up_m = round(local$z, 2)
)

Antarctic Field Survey

# Simulated survey data at McMurdo
mcmurdo <- c(166.67, -77.85, 10)  # lon, lat, height

# Survey points in local coordinates
survey_local <- data.frame(
  point = c("Control", "P1", "P2", "P3", "P4"),
  east = c(0, 500, -300, 200, -100),
  north = c(0, 200, 400, -150, -300),
  up = c(0, 5, -2, 8, -5)
)

# Convert to geodetic for GPS upload
geodetic <- localcartesian_rev(
  survey_local$east, survey_local$north, survey_local$up,
  lon0 = mcmurdo[1], lat0 = mcmurdo[2], h0 = mcmurdo[3]
)

# Convert to ECEF for satellite positioning
ecef <- geocentric_fwd(
  cbind(geodetic$lon, geodetic$lat),
  h = geodetic$h
)

data.frame(
  point = survey_local$point,
  lon = round(geodetic$lon, 5),
  lat = round(geodetic$lat, 5),
  X = round(ecef$X),
  Y = round(ecef$Y),
  Z = round(ecef$Z)
)

Coordinate System Summary

| System | Description | Use Case | |--------|-------------|----------| | Geodetic (lon/lat/h) | Geographic coordinates | Maps, GIS, human communication | | ECEF (X/Y/Z) | Earth-centered Cartesian | GPS satellites, orbit calculations | | ENU (E/N/U) | Local tangent plane | Robotics, local surveys, navigation |

See Also



Try the geographiclib package in your browser

Any scripts or data that you put into this service are public.

geographiclib documentation built on March 4, 2026, 9:07 a.m.