GitHub

Geometry

Global Geometry

ClimaCore.Geometry.CartesianGlobalGeometry — Type
CartesianGlobalGeometry()

Specifies that the local coordinates align with the Cartesian coordinates, e.g. XYZPoint aligns with Cartesian123Point, and UVWVector aligns with Cartesian123Vector.

source

LocalGeometry

Internals

ClimaCore.Geometry.:⊗ — Function
x ⊗ y

Shorthand for the outer product x * y'.

# vector ⊗ scalar = vector
julia> [1.0,2.0] ⊗ 2.0
2-element Vector{Float64}:
 2.0
 4.0

# vector ⊗ vector = matrix
julia> [1.0,2.0] ⊗ [1.0,3.0]
2×2 Matrix{Float64}:
 1.0  3.0
 2.0  6.0
source

Coordinates

ClimaCore.Geometry.AbstractPoint — Type
AbstractPoint

Represents a point in space.

The following types are supported:

  • XPoint(x)
  • YPoint(y)
  • ZPoint(z)
  • XYPoint(x, y)
  • XZPoint(x, z)
  • XYZPoint(x, y, z)
  • LatPoint(lat)
  • LongPoint(long)
  • LatLongPoint(lat, long)
  • LatLongZPoint(lat, long, z)
  • Cartesian1Point(x1)
  • Cartesian2Point(x2)
  • Cartesian3Point(x3)
  • Cartesian12Point(x1, x2)
  • Cartesian13Point(x1, x3)
  • Cartesian123Point(x1, x2, x3)
source

Points represent locations in space, specified by coordinates in a given coordinate system (Cartesian, spherical, etc), whereas vectors, on the other hand, represent displacements in space.

An analogy with time works well: times (also called instants or datetimes) are locations in time, while, durations are displacements in time.

Note 1: Latitude and longitude are specified via angles (and, therefore, trigonometric functions: cosd, sind, acosd, asind, tand,...) in degrees, not in radians. Moreover, lat (usually denoted by $\theta$) $\in [-90.0, 90.0]$, and long (usually denoted by $\lambda$) $\in [-180.0, 180.0]$.

Note 2:: In a Geometry.LatLongZPoint(lat, long, z), z represents the elevation above the surface of the sphere with radius R (implicitly accounted for in the geoemtry).

Note 3: There are also a set of specific Cartesian points (Cartesian1Point(x1), Cartesian2Point(x2), etc). These are occasionally useful for converting everything to a full Cartesian domain (e.g. for visualization purposes). These are distinct from XYZPoint as ZPoint can mean different things in different domains.