Spherical trigonometry studies the relationships between the sides and angles of spherical triangles.
Let a spherical triangle be drawn on the surface of a sphere of radius , centered at a point
, with vertices
,
, and
. The vectors from the center of
the sphere to the vertices
are therefore given by
,
, and
. The angular lengths of its sides (in radians)
are
,
,
and
,
and the actual arc lengths of the sides are
,
,
and
.
Explicitly,
|
(1)
| |||
|
(2)
| |||
|
(3)
|
Now make use of ,
,
and
to denote both the vertices themselves and the
angles of the spherical
triangle at these vertices, so that the dihedral angle between planes
and
is written
,
the dihedral angle between planes
and
is written
,
and the dihedral angle between planes
and
is written
.
(These angles are sometimes instead denoted
,
,
; e.g., Gellert et al. 1989.)
For the remainder, take , so the angular side lengths and arc
lengths agree, and use
,
, and
for the angular side lengths. The dihedral
angle
between planes
and
can be calculated using the dot
product of their normal vectors. These normal
vectors are given by cross products of the vectors to the vertices,
so
|
(4)
| |||
|
(5)
|
However, using the vector triple product identity gives
|
(6)
| |||
|
(7)
| |||
|
(8)
| |||
|
(9)
|
Since these two expressions must be equal, we obtain the identity (and its two analogous formulas)
|
(10)
| |||
|
(11)
| |||
|
(12)
|
known as the cosine rules for sides (Smart 1960, pp. 7-8; Gellert et al. 1989, p. 264; Zwillinger 1995, p. 469).
The identity
|
(13)
| |||
|
(14)
| |||
|
(15)
|
where
is the scalar triple product, gives
|
(16)
|
so the spherical analog of the law of sines can be written
|
(17)
|
(Smart 1960, pp. 9-10; Gellert et al. 1989, p. 265; Zwillinger 1995, p. 469), where is the volume of the tetrahedron.
The analogs of the law of cosines for the angles of a spherical triangle are given by
|
(18)
| |||
|
(19)
| |||
|
(20)
|
(Gellert et al. 1989, p. 265; Zwillinger 1995, p. 470).
Finally, there are spherical analogs of the law of tangents,
|
(21)
| |||
|
(22)
| |||
|
(23)
|
(Beyer 1987; Gellert et al. 1989; Zwillinger 1995, p. 470).
Additional important identities are given by
|
(24)
|
(Smart 1960, p. 8),
|
(25)
|
(Smart 1960, p. 10), and
|
(26)
|
(Smart 1960, p. 12).
Let
|
(27)
|
be the semiperimeter, then half-angle formulas for sines can be written as
|
(28)
| |||
|
(29)
| |||
|
(30)
|
for cosines can be written as
|
(31)
| |||
|
(32)
| |||
|
(33)
|
and tangents can be written as
|
(34)
| |||
|
(35)
| |||
|
(36)
|
where
|
(37)
|
(Smart 1960, pp. 8-9; Gellert et al. 1989, p. 265; Zwillinger 1995, p. 470).
Let
|
(38)
|
be the sum of half-angles, then the half-side formulas are
|
(39)
| |||
|
(40)
| |||
|
(41)
|
where
|
(42)
|
(Gellert et al. 1989, p. 265; Zwillinger 1995, p. 470).
The haversine formula for sides, where
|
(43)
|
is given by
|
(44)
|
(Smart 1960, pp. 18-19; Zwillinger 1995, p. 471), and the haversine formula for angles is given by
|
(45)
| |||
|
(46)
| |||
|
(47)
|
(Zwillinger 1995, p. 471).
Gauss's formulas (also called Delambre's analogies) are
|
(48)
| |||
|
(49)
| |||
|
(50)
| |||
|
(51)
|
(Smart 1960, p. 22; Zwillinger 1995, p. 470).
|
(52)
| |||
|
(53)
| |||
|
(54)
| |||
|
(55)
|
(Beyer 1987; Gellert et al. 1989, p. 266; Zwillinger 1995, p. 471).