Celestial Coordinate Systems

(quick recap)
The Horizontal Coordinates
Reference: the tangent plane to the Earth surface
passing through the observer (i.e., topocentric)
Astronomical Horizon: intersection between the
reference plane and the celestial sphere
Zenith and Nadir: poles corresponding to the horizon
Vertical: any great circle passing through the Zenith
Local Meridian: vertical passing through the North
S
Coordinates:
azimuth: measured clockwise either from North or South
0° < A < 360° or -180° < A < 180°
altitude or elevation: measured from the horizon along the vertical
-90° < a < 90° (a < 0 ° : below the horizon)

zenith angle (or distance): z = 90° - a Nadir

airmass: “amount of air that one is looking through”, X = sec (z) in plane-parallel approximation
The Equatorial Coordinates
Reference: the plane perpendicular to the rotational axis of
the Earth and passing through its center (i.e. geocentric)
Equator of the celestial sphere: projection of the
Equator into the celestial sphere
North and South Celestial Poles

Coordinates:
declination (dec): angular separation from the
equator, -90° (south) < δ < 90° (north)
right ascension (RA): measured counter-clockwise
from the vernal point , 0 < α < 360° or 24h

The vernal point is the intersection between the

equator and the ecliptic. The ecliptic is the path of the Sun on
the sky, i.e., the projection of the orbital plane of the Earth.
The Sun crosses the vernal point during the vernal equinox (in March)
The Ecliptic Coordinates
Reference: the ecliptic (the orbital plane of the Earth)
The vernal point is the direction of the zero axis
The normal to the orbital plane identify the celestial poles
Coordinates:
ecliptic latitude (β): angular distance from the ecliptic
-90° < β < 90°
ecliptic longitude (λ): measured counter-clockwise
from the vernal point, 0< λ < 360°

The angle between the

celestial equator and the ecliptic
is called
obliquity of the ecliptic, ε = 23° 26’
(inclination of Earth rotation axis)
Distances of Stars
Trigonometric parallax (π) is the only direct method to measure the distance of a star

Parsec (pc): at a distance of one parsec,

one astronomical unit subtends an angle of one arc second.

Approximating sin(π) ≈ π, and knowing that 1 radian ≈ 206 265’’

(1 rad × (3600 × 180)/π = 206264.806’’) where π is 3.14…

1 pc ≈ 206 265 AU ≈ 3.086 x 10e16 m ≈ 3.26 light years

The distance in parsec is the inverse of the parallax in arcsec r[pc] = 1 / π[’’]

The nearest star Proxima Centauri is at 1.31 pc (π = 0.762 ’’ )

Alpha Centauri has a distance of 1.34 pc. (π = 0.747 ’’ ) Remember: astronomers
There are less than 50 stars within 15 pc (π = 0.067 ’’ ) use light years only for
Measuring distances is extremely difficult!!! outreach…
Distances of Stars
Hipparcos: High Precision Parallax Collecting Satellite
In Geosynchronous orbit, elliptic due a failure of one of the booster
Baseline ≈ Earth radius
Two catalogues: Hipparcos (120 000 stars with precision of 1 mas),
Tycho (1 million stars with precision of 20-30 mas)

Gaia: formerly Global Astrometric Interferometer for Astrophysics

Lissajous orbit around the Lagrangian point L2
Data Release 2 already provide distances and proper motions
for more than 1.3 billions stars, with a precision of 0.04 mas
Launched in 2013, mission prolonged to 2022
The Gaia Mission
Transformation
of Coordinates
Trasformations through Matrixes Rotation
The coordinates (𝒆𝒙 , 𝒆𝒚 , 𝒆𝒛) of a point P in the Cartesian orthogonal system 𝒙𝒚𝒛 can be
transformed to the coordinates (𝒇𝑿, 𝒇𝒀, 𝒇𝒁 ) of another Cartesian orthogonal system 𝐗𝐘𝐙
through a rigid rotation if the two frames share the origin O:
𝑓' = 𝑒( cos 𝜃(' + 𝑒) cos 𝜃)' + 𝑒* cos 𝜃*'
) 𝑓+ = 𝑒( cos 𝜃(+ + 𝑒) cos 𝜃)+ + 𝑒* cos 𝜃*+
𝑓, = 𝑒* cos 𝜃(, + 𝑒) cos 𝜃), + 𝑒* cos 𝜃*,
This is equivalent to:
𝑓' cos 𝜃(' cos 𝜃)' cos 𝜃*' 𝑒(
𝑓+ = cos 𝜃(+ cos 𝜃)+ cos 𝜃*+ 𝑒)
𝑓, cos 𝜃(, cos 𝜃), cos 𝜃*, 𝑒*

We call with R the rotation matrix:

𝑓' 𝑒( cos 𝜃(' cos 𝜃)' cos 𝜃*'
𝑓+ = 𝑹 𝑒) 𝑹 = cos 𝜃(+ cos 𝜃)+ cos 𝜃*+
𝑓, 𝑒* cos 𝜃(, cos 𝜃), cos 𝜃*,
Polar coordinates
We express the position of the point P=(𝒆𝒙 , 𝒆𝒚 , 𝒆𝒛) using the polar coordinates system (r, 𝝀, 𝜷)
The rotated system (always around O) is identified by the coordinates (r, 𝚲, 𝚩)
𝒆𝒛
𝑒( = 𝑟 cos 𝛽 cos 𝜆 𝑓' = 𝑟 cos Β cos Λ
P=(𝒆𝒙 , 𝒆𝒚 , 𝒆𝒛 ) ) 𝑒) = 𝑟 cos 𝛽 sin 𝜆 > 𝑓+ = 𝑟 cos Β sin Λ
𝑒* = 𝑟 sin 𝛽 𝑓, = 𝑟 sin Β
Be careful on the definition of
𝒆𝒚 cos Β cos Λ cos 𝛽 cos 𝜆 𝛽/ θ! Physics & Maths may
cos Β sin Λ = 𝑹 cos 𝛽 sin 𝜆 use a different definition

sin Β sin 𝛽
𝒆𝒙 The rotation is independent of r, so it holds for a unit sphere
The inverse rotation is obtained changing the role of (𝒆𝒙 , 𝒆𝒚 , 𝒆𝒛) and
(𝒇𝑿, 𝒇𝒀, 𝒇𝒁 ) paying attention in maintaining the positive verse of the angles.
This implies that the matrix of the inverse rotation is the transpose of R.
𝑹./ = 0𝑹 0𝑅
12 = 𝑅21 𝑹./(𝜃) = 𝑹(−𝜃)
Sequences of rotations
In general, every rotation can be considered as the result of three different successive rotations.
R1 by angle ϕ1 around e1 axis, R2 by angle ϕ2 around e2 axis, and R3 by angle ϕ3 around e3 axis.
Let’s consider the first rotation:
𝜃(' = 0, 𝜃)' = 𝜋/2, 𝜃*' = 𝜋/2
4
𝜃(+ = 90°, 𝜃)+ = 𝜑/, 𝜃*+ = 5 𝜋 + 𝜑/
𝜃(, = 90°, 𝜃), = 𝜋/2 + 𝜑/, 𝜃*, = 𝜑/
Remember the trigonometric relationships: 𝒆𝒛
𝒇𝒁
cos 𝜋/2 = 0
3 𝒇𝒀
cos 𝜋 + 𝜑/ = sin 𝜑/
2
cos 𝜋/2 + 𝜑/ = − sin 𝜑/ 𝒆𝒚

1 0 0
𝑹𝟏 𝜑/ = 0 cos 𝜑/ sin 𝜑/
0 −sin 𝜑/ cos 𝜑/ 𝒆𝒙 = 𝒇𝑿
Sequences of rotations
In general, every rotation can be considered as the result of three different successive rotations.
R1 by angle ϕ1 around e1 axis, R2 by angle ϕ2 around e2 axis, and R3 by angle ϕ3 around e3 axis.

1 0 0
𝑹𝟏 𝜑/ = 0 cos 𝜑/ sin 𝜑/
0 −sin 𝜑/ cos 𝜑/
cos 𝜑5 0 −sin 𝜑5
𝑹 𝟐 𝜑5 = 0 1 0
sin 𝜑5 0 cos 𝜑5
cos 𝜑4 sin 𝜑4 0
𝑹 𝟑 𝜑4 = −sin 𝜑4 cos 𝜑4 0
0 0 1
From Equatorial to Ecliptic
Equatorial (𝜶, 𝜹) and ecliptic (𝝀, 𝜷) share the same origin and x axis (identified with the 𝛾 point)
A rotation equal to the obliquity of the ecliptic 𝜀 around the x axis is sufficient
𝜀 ≈ 23° 268 = 23.43° = 0.40899 rad
1 0 0 1 0 0
𝑹 𝜀 = 0 cos 𝜀 sin 𝜀 = 0 0.91741 0.39073
0 −sin 𝜀 cos 𝜀 0 −0.39073 0.91741
From equatorial to ecliptic
cos 𝛽 cos 𝜆 = cos 𝛿 cos 𝛼
)cos 𝛽 sin 𝜆 = cos 𝛿 sin 𝛼 cos 𝜀 + sin 𝛿 sin 𝜀
sin 𝛽 = − cos 𝛿 sin 𝛼 sin 𝜀 + sin 𝛿 cos 𝜀
From ecliptic to equatorial
cos 𝛿 cos 𝛼 = cos 𝛽 cos 𝜆
)cos 𝛿 sin 𝛼 = cos 𝛽 sin 𝜆 cos 𝜀 − sin 𝛽 sin 𝜀
Three equations are needed to determine two
sin 𝛿 = cos 𝛽 sin 𝜆 sin 𝜀 + sin 𝛽 cos 𝜀
angles and their signs (quadrants).
From Alt-Az to Equatorial
Equatorial (𝜶, 𝜹) and horizontal (𝑨, 𝒉) have the East-West axis in common between the two
reference planes, although they have different reference axis on those planes.
𝝋: the height of the Celestial North Pole = latitude of observer
HA (hour angle): distance of the star from the local meridian
(I.e. south direction) along the equator, measured clockwise HA

Using HA, the two systems have their yz plane in common:

Rotation around the West direction (𝑨, 𝒉) → (𝑯𝑨, 𝜹)
sin 𝐻𝐴 cos 𝛿 = sin 𝐴 cos ℎ
>cos 𝐻𝐴 cos 𝛿 = cos 𝐴 cos ℎ sin 𝜑 + sin ℎ cos 𝜑
sin 𝛿 = − cos 𝐴 co𝑠 ℎ cos 𝜑 + sin ℎ sin 𝜑
Inverse transformation (𝑯𝑨, 𝜹) → (𝑨, 𝒉)
cos ℎ sin 𝐴 = cos 𝛿 sin 𝐻𝐴
>cos ℎ cos 𝐴 = cos 𝛿 cos 𝐻𝐴 sin 𝜑 − sin 𝛿 cos 𝜑
sin ℎ = cos 𝛿 co𝑠 𝐻𝐴 cos 𝜑 + sin 𝛿 sin 𝜑
The Sidereal Time
right ascension (α) of a star: as before, counter-clockwise
hour angle (HA or h) of a star: the distance between its projection
on the equator and the south meridian, measured clockwise
sidereal time (ST or Θ): the hour angle of 𝛾
ST = HA + α → Θ=h+α
Celestial Pole
Sidereal Time

South
Vernal
equinox

A star crossing the meridian has h=0 and α=Θ → h<0 before crossing the meridian
Sinodic and Sidereal Time
Sinodic day: time between two consecutive
passages of the Sun across the meridian
Sidereal day (ST): time between two consecutive
passages of a reference star across the meridian.

Sidereal time runs faster than the Solar time

Sinodic and Sidereal Time
The Sun and
Betelgeuse
are aligned
on the Local
Meridian
Sinodic and Sidereal Time
After one week,
Betelgeuse has
alredy transited
the Local
Meridian when
the Sun is
crossing it
The movement of the stars in the Sky
The altitude of the North Celestial Pole depends on the geographical latitude φ of the observer
• Stars moves parallel to the celestial equator
• A star culminates (it reaches the maximum altitude)
when crossing the meridian (on the South side)
• If the declination of the star is higher than the
zenith angle of the pole north, the star is always
visible ( circumpolar star) (NE = PZ)

E
Rising and Setting Stars
1) Given the equatorial (𝜶, 𝜹) coordinates of a star and the Sidereal Time, we get HA = ST – α
2) Visibility is given by the third relationship: sin ℎ = cos 𝛿 co𝑠 𝐻𝐴 cos 𝜑 + sin 𝛿 sin 𝜑
• Limit of visibility: h = 0 → sin ℎ = 0, cos ℎ = 1
• Hour Angle of rising and setting: cos 𝐻𝐴 = − tan 𝛿 tan 𝜑
• Azimuth or rising/setting: cos 𝐴 = − sin 𝛿 /cos 𝜑
Only equatorial stars (𝛿 = 0) rise and set exactly
on E and W, respectively
In the North Emisphere (𝝋 > 𝟎), a star with given 𝛿 is
• Circumpolar if 𝛿 ≥ 90° − 𝜑
• Setting star (occidua) −(90° − 𝜑) ≤ 𝛿 < 90° − 𝜑
• Never observable if 𝛿 < −(90° − 𝜑)

A star crosses the meridian twice during a day (S and N side)

• ℎ9:; = 90° − 𝜑 − 𝛿 , ℎ9<= = 𝜑 + 𝛿 − 90°
Circumpolar: ℎ9<= > ℎ9:; > 0°, rising: ℎ9<= > 0° > ℎ9:;,
Rising and Setting Stars
1) Given the equatorial (𝜶, 𝜹) coordinates of a star and the Sidereal Time, we get HA = ST – α
2) Visibility is given by the third relationship: sin ℎ = cos 𝛿 co𝑠 𝐻𝐴 cos 𝜑 + sin 𝛿 sin 𝜑
• Limit of visibility: h = 0 → sin ℎ = 0, cos ℎ = 1
• Hour Angle of rising and setting: cos 𝐻𝐴 = − tan 𝛿 tan 𝜑
• Azimuth or rising/setting: cos 𝐴 = − sin 𝛿 /cos 𝜑
Only equatorial stars (𝛿 = 0) rise and set exactly
on E and W, respectively
In the South Emisphere (𝝋 < 𝟎), a star with given 𝛿 is
• Circumpolar if 𝛿 ≤ −90° − 𝜑
• Setting star (occidua) −90° − 𝜑 < 𝛿 ≤ 90° + 𝜑
• Never observable if 𝛿 > 90° + 𝜑

Above relationship are the opposite of the NE case

(remeber the negative sign of 𝜑)
Altitude relationships are the same (thanks to the absolute function)
The motion of the Sun on the Sky
During the year the Sun moves around on the ecliptic (i.e., it
will be below or above the equator depending on the
season), while rotating around the Earth rotation axis
Every day the Sun
rises and set in a
different direction
and at a different
time, depending
on the season

Using the ecliptic to

equatorial
transformation and
imposing 𝛽⨀ = 0,
sin 𝛼⨀ = tan 𝛿⨀/ tan 𝜀
Angular Velocities
It is useful to derive angular velocities using the Sidereal Time as time variable.

Note: we neglect all the effects causing coordinate perturbations → α = const

HA = ST – α 𝑑𝐻𝐴 𝑑𝛿
We take the first derivative using ST as variable: → =1 =0
𝛿 = const 𝑑𝑡 𝑑𝑡

Not much interesting so far… how about the HORIZONTAL coordinates of a target?
+ previous
sin ℎ = cos 𝛿 co𝑠 𝐻𝐴 cos 𝜑 + sin 𝛿 sin 𝜑 → ℎ̇ cos ℎ = − cos 𝛿 sin 𝐻𝐴 cos 𝜑
relationships

ℎ̇ = −sin 𝐴 cos 𝜑 The velocity in height is always confined between ±1

𝐴̇ = sin 𝜑 + co𝑠 𝐴 tan ℎ cos 𝜑 (i.e. ±15° /(sideral hour).
• It is 0 for a telescope at the Poles (𝜑 = 90°)
• it is maximum for a telescope at the equator (𝜑 = 0°).
Azimuthal Velocities
𝐴̇ = sin 𝜑 + co𝑠 𝐴 tan ℎ cos 𝜑 𝐴̇ can be re-written as: 𝐴̇ = cos 𝜑 cos 𝐴
cos ℎ
At the horizon (h = 0°), i.e. at rising and
setting, the Azimuth speed 𝐴̇ is sin 𝜑
𝐴̇ → ∞
• positive in the North Hemisphere,
at zenith
• negative in the South one
(h = 90°)
• stationary if the observer is on the Equator.