mapproject {mapproj} | R Documentation |
Apply a Map Projection
Description
Converts latitude and longitude into projected coordinates.
Usage
mapproject(x, y, projection="", parameters=NULL, orientation=NULL)
Arguments
x , y |
two vectors giving longitude and latitude coordinates
of points on the earth's surface to be projected.
A list containing components named |
projection |
optional character string that names a map projection to
use. If the string is |
parameters |
optional numeric vector of parameters for use with the
|
orientation |
An optional vector |
Details
Each standard projection is displayed with the Prime
Meridian (longitude 0) being a straight vertical line, along which North
is up.
The orientation of nonstandard projections is specified by
the three parameters=c(lat,lon,rot)
.
Imagine a transparent gridded sphere around the globe.
First turn the overlay about the North Pole
so that the Prime Meridian (longitude 0)
of the overlay coincides with meridian lon
on the globe.
Then tilt the North Pole of the
overlay along its Prime Meridian to latitude lat
on the globe.
Finally again turn the
overlay about its "North Pole" so
that its Prime Meridian coincides with the previous position
of (the overlay) meridian rot
.
Project the desired map in
the standard form appropriate to the overlay, but presenting
information from the underlying globe.
In the descriptions that follow each projection is shown as a function call; if it requires parameters, these are shown as arguments to the function. The descriptions are grouped into families.
Equatorial projections centered on the Prime Meridian (longitude 0). Parallels are straight horizontal lines.
- mercator()
equally spaced straight meridians, conformal, straight compass courses
- sinusoidal()
equally spaced parallels, equal-area, same as
bonne(0)
- cylequalarea(lat0)
equally spaced straight meridians, equal-area, true scale on
lat0
- cylindrical()
central projection on tangent cylinder
- rectangular(lat0)
equally spaced parallels, equally spaced straight meridians, true scale on
lat0
- gall(lat0)
parallels spaced stereographically on prime meridian, equally spaced straight meridians, true scale on
lat0
- mollweide()
(homalographic) equal-area, hemisphere is a circle
- gilbert()
sphere conformally mapped on hemisphere and viewed orthographically
Azimuthal projections centered on the North Pole. Parallels are concentric circles. Meridians are equally spaced radial lines.
- azequidistant()
equally spaced parallels, true distances from pole
- azequalarea()
equal-area
- gnomonic()
central projection on tangent plane, straight great circles
- perspective(dist)
viewed along earth's axis
dist
earth radii from center of earth- orthographic()
viewed from infinity
- stereographic()
conformal, projected from opposite pole
- laue()
radius = tan(2 * colatitude)
used in xray crystallography- fisheye(n)
stereographic seen through medium with refractive index
n
- newyorker(r)
radius = log(colatitude/r)
map from viewing pedestal of radiusr
degrees
Polar conic projections symmetric about the Prime Meridian. Parallels are segments of concentric circles. Except in the Bonne projection, meridians are equally spaced radial lines orthogonal to the parallels.
- conic(lat0)
central projection on cone tangent at
lat0
- simpleconic(lat0,lat1)
equally spaced parallels, true scale on
lat0
andlat1
- lambert(lat0,lat1)
conformal, true scale on
lat0
andlat1
- albers(lat0,lat1)
equal-area, true scale on
lat0
andlat1
- bonne(lat0)
equally spaced parallels, equal-area, parallel
lat0
developed from tangent cone
Projections with bilateral symmetry about the Prime Meridian and the equator.
- polyconic()
parallels developed from tangent cones, equally spaced along Prime Meridian
- aitoff()
equal-area projection of globe onto 2-to-1 ellipse, based on
azequalarea
- lagrange()
conformal, maps whole sphere into a circle
- bicentric(lon0)
points plotted at true azimuth from two centers on the equator at longitudes
+lon0
and-lon0
, great circles are straight lines (a stretched gnomonic projection)- elliptic(lon0)
points are plotted at true distance from two centers on the equator at longitudes
+lon0
and-lon0
- globular()
hemisphere is circle, circular arc meridians equally spaced on equator, circular arc parallels equally spaced on 0- and 90-degree meridians
- vandergrinten()
sphere is circle, meridians as in
globular
, circular arc parallels resemblemercator
- eisenlohr()
conformal with no singularities, shaped like polyconic
Doubly periodic conformal projections.
- guyou
W and E hemispheres are square
- square
world is square with Poles at diagonally opposite corners
- tetra
map on tetrahedron with edge tangent to Prime Meridian at S Pole, unfolded into equilateral triangle
- hex
world is hexagon centered on N Pole, N and S hemispheres are equilateral triangles
Miscellaneous projections.
- harrison(dist,angle)
oblique perspective from above the North Pole,
dist
earth radii from center of earth, looking along the Date Lineangle
degrees off vertical- trapezoidal(lat0,lat1)
equally spaced parallels, straight meridians equally spaced along parallels, true scale at
lat0
andlat1
on Prime Meridian- lune(lat,angle)
conformal, polar cap above latitude
lat
maps to convex lune with givenangle
at 90E and 90W
Retroazimuthal projections. At every point the angle between vertical
and a straight line to "Mecca", latitude lat0
on the prime meridian,
is the true bearing of Mecca.
- mecca(lat0)
equally spaced vertical meridians
- homing(lat0)
distances to Mecca are true
Maps based on the spheroid. Of geodetic quality, these projections do not make sense for tilted orientations.
- sp_mercator()
Mercator on the spheroid.
- sp_albers(lat0,lat1)
Albers on the spheroid.
Value
list with components
named x
and y
, containing the projected coordinates.
NA
s project to NA
s.
Points deemed unprojectable (such as north of 80 degrees
latitude in the Mercator projection) are returned as NA
.
Because of the ambiguity of the first two arguments, the other
arguments must be given by name.
Each time mapproject
is called, it leaves on frame 0 the
dataset .Last.projection
, which is a list with components projection
,
parameters
, and orientation
giving the arguments from the
call to mapproject
or as constructed (for orientation
).
Subsequent calls to mapproject
will get missing information
from .Last.projection
.
Since map
uses mapproject
to do its projections, calls to
mapproject
after a call to map
need not supply any arguments
other than the data.
References
Richard A. Becker, and Allan R. Wilks, "Maps in S", AT&T Bell Laboratories Statistics Research Report, 1991. https://web.archive.org/web/20070824013345/http://www.research.att.com/areas/stat/doc/93.2.ps
M. D. McIlroy, Documentation from the Tenth Edition UNIX Manual, Volume 1, Saunders College Publishing, 1990.
Examples
library(maps)
# Bonne equal-area projection with state abbreviations
map("state",proj='bonne', param=45)
data(state)
text(mapproject(state.center), state.abb)
# this does not work because the default orientations are different:
map("state",proj='bonne', param=45)
text(mapproject(state.center,proj='bonne',param=45),state.abb)
map("state",proj="albers",par=c(30,40))
map("state",par=c(20,50)) # another Albers projection
map("world",proj="gnomonic",orient=c(0,-100,0)) # example of orient
# see map.grid for more examples
# tests of projections added RSB 091101
projlist <- c("aitoff", "albers", "azequalarea", "azequidist", "bicentric",
"bonne", "conic", "cylequalarea", "cylindrical", "eisenlohr", "elliptic",
"fisheye", "gall", "gilbert", "guyou", "harrison", "hex", "homing",
"lagrange", "lambert", "laue", "lune", "mercator", "mollweide", "newyorker",
"orthographic", "perspective", "polyconic", "rectangular", "simpleconic",
"sinusoidal", "tetra", "trapezoidal")
x <- seq(-100, 0, 10)
y <- seq(-45, 45, 10)
xy <- expand.grid(x=x, y=y)
pf <- c(0, 2, 0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 1, 0, 0, 2, 0, 1, 0, 2, 0, 2,
0, 0, 1, 0, 1, 0, 1, 2, 0, 0, 2)
res <- vector(mode="list", length=length(projlist))
for (i in seq(along=projlist)) {
if (pf[i] == 0) res[[i]] <- mapproject(xy$x, xy$y, projlist[i])
else if (pf[i] == 1) res[[i]] <- mapproject(xy$x, xy$y, projlist[i], 0)
else res[[i]] <- mapproject(xy$x, xy$y, projlist[i], c(0,0))
}
names(res) <- projlist
lapply(res, function(p) rbind(p$x, p$y))