CoordRefSystems.jl provides conversions between Coordinate Reference Systems (CRS) in native Julia. It was designed to work with units from Unitful.jl, respects projection bounds catalogued in https://epsg.io, and is very fast thanks to advanced parametrizations at compile-time.
This package addresses various design issues encountered in previous attempts such as Geodesy.jl and MapMaths.jl. Our benchmarks show that CoordRefSystems.jl is sometimes faster than PROJ, which is the most widely used software library for cartography in the world (written in C/C++).
Get the latest stable release with Julia's package manager:
] add CoordRefSystems
Consider the following conversions between Cartesian
, Spherical
,
Cylindrical
and Polar
coordinates to get started:
julia> cartesian = Cartesian(1, 1, 1)
Cartesian{NoDatum} coordinates
├─ x: 1.0 m
├─ y: 1.0 m
└─ z: 1.0 m
julia> spherical = convert(Spherical, cartesian)
Spherical{NoDatum} coordinates
├─ r: 1.7320508075688772 m
├─ θ: 0.9553166181245093 rad
└─ ϕ: 0.7853981633974483 rad
julia> convert(Cartesian, spherical)
Cartesian{NoDatum} coordinates
├─ x: 1.0 m
├─ y: 0.9999999999999998 m
└─ z: 0.9999999999999999 m
julia> cartesian = Cartesian(1, 1, 1)
Cartesian{NoDatum} coordinates
├─ x: 1.0 m
├─ y: 1.0 m
└─ z: 1.0 m
julia> cylindrical = convert(Cylindrical, cartesian)
Cylindrical{NoDatum} coordinates
├─ ρ: 1.4142135623730951 m
├─ ϕ: 0.7853981633974483 rad
└─ z: 1.0 m
julia> convert(Cartesian, cylindrical)
Cartesian{NoDatum} coordinates
├─ x: 1.0000000000000002 m
├─ y: 1.0 m
└─ z: 1.0 m
julia> cartesian = Cartesian(1, 1)
Cartesian{NoDatum} coordinates
├─ x: 1.0 m
└─ y: 1.0 m
julia> polar = convert(Polar, cartesian)
Polar{NoDatum} coordinates
├─ ρ: 1.4142135623730951 m
└─ ϕ: 0.7853981633974483 rad
julia> convert(Cartesian, polar)
Cartesian{NoDatum} coordinates
├─ x: 1.0000000000000002 m
└─ y: 1.0 m
Julia's convert
methods can be triggered with special
syntax assuming that a list of coordinates is available:
julia> Mercator[LatLon(0, 0), LatLon(30, 30), LatLon(20,30)]
3-element Vector{Mercator}:
Mercator{WGS84Latest}(x: 0.0 m, y: 0.0 m)
Mercator{WGS84Latest}(x: 3.33958e6 m, y: 3.48219e6 m)
Mercator{WGS84Latest}(x: 3.33958e6 m, y: 2.25842e6 m)
The example above is equivalent to running
convert(Mercator, latlon)
for all latlon
coordinates in the list.
CRS are most useful to locate objets in the physical world. Given an ellipsoid of revolution and a standardized origin (a.k.a. datum), it is possible assign coordinates to points without ambiguity.
Below is an example converting geodetic LatLon
coordinates
on the WGS84Latest
datum to Mercator
, WebMercator
, and
Robinson
projected coordinates on the same datum:
julia> latlon = LatLon(30, 60)
GeodeticLatLon{WGS84Latest} coordinates
├─ lat: 30.0°
└─ lon: 60.0°
julia> convert(Mercator, latlon)
Mercator{WGS84Latest} coordinates
├─ x: 6.679169447596414e6 m
└─ y: 3.482189085408618e6 m
julia> convert(WebMercator, latlon)
WebMercator{WGS84Latest} coordinates
├─ x: 6.679169447596414e6 m
└─ y: 3.5035498435043753e6 m
julia> convert(Robinson, latlon)
Robinson{WGS84Latest} coordinates
├─ x: 5.441866544132874e6 m
└─ y: 3.2085576115038935e6 m
julia> latlon = LatLon(30, 60)
GeodeticLatLon{WGS84Latest} coordinates
├─ lat: 30.0°
└─ lon: 60.0°
julia> mercator = convert(Mercator, latlon)
Mercator{WGS84Latest} coordinates
├─ x: 6.679169447596414e6 m
└─ y: 3.482189085408618e6 m
julia> convert(LatLon, mercator)
GeodeticLatLon{WGS84Latest} coordinates
├─ lat: 29.999999999999996°
└─ lon: 59.99999999999999°
It is also possible to convert between different datum, transparently.
In the following examples, we convert coordinates between the WGS84Latest
datum, currently an alias to WGS84{1762}
, and the ITRF{2008}
datum:
julia> latlon = LatLon{WGS84Latest}(30, 60)
GeodeticLatLon{WGS84Latest} coordinates
├─ lat: 30.0°
└─ lon: 60.0°
julia> convert(LatLon{ITRF{2008}}, latlon)
GeodeticLatLon{ITRF{2008}} coordinates
├─ lat: 30.00000000081754°
└─ lon: 59.99999999999999°
julia> mercator = convert(Mercator{WGS84Latest}, latlon)
Mercator{WGS84Latest} coordinates
├─ x: 6.679169447596414e6 m
└─ y: 3.482189085408618e6 m
julia> convert(WebMercator{WGS84Latest}, mercator)
WebMercator{WGS84Latest} coordinates
├─ x: 6.679169447596414e6 m
└─ y: 3.5035498435043753e6 m
julia> convert(WebMercator{ITRF{2008}}, mercator)
WebMercator{ITRF{2008}} coordinates
├─ x: 6.679169447596414e6 m
└─ y: 3.5035498436094625e6 m
CRS are catalogued with numerical codes in https://epsg.io.
The package provides EPSG{code}
and ERSI{code}
, and the
utility CoordRefSystems.get
function to query the database:
julia> CRS1 = CoordRefSystems.get(EPSG{4326})
GeodeticLatLon{WGS84Latest}
julia> CRS2 = CoordRefSystems.get(EPSG{3395})
Mercator{WGS84Latest}
julia> CRS1(0, 90)
GeodeticLatLon{WGS84Latest} coordinates
├─ lat: 0.0°
└─ lon: 90.0°
julia> convert(CRS2, CRS1(0, 90))
Mercator{WGS84Latest} coordinates
├─ x: 1.0018754171394622e7 m
└─ y: 0.0 m
Most implementations in this package are adaptations from PROJ - Cartographic Projections and Coordinate Transformations Library and its list of references. Our tests were designed to match their results to the last digit via the Proj.jl wrapper.