Galactic Astronomy
Galactic Coordinates
A celestial reference frame anchored to the Milky Way, not to Earth
Galactic longitude ℓ counts eastward in the disk plane from the Galactic Center (Sgr A*, ℓ=0); latitude b measures angle above/below midplane. Sun sits at R0 = 8.178 kpc; north pole in Coma Berenices.
- OriginSgr A* (Galactic Center, ℓ=0, b=0)
- Galactic-center distanceR0 = 8.178 ± 0.013 kpc (GRAVITY 2019)
- North galactic poleRA 12h 51m 26.3s, Dec +27° 07' 41″ (Coma Berenices)
- Sun's height above plane~20-25 pc north
- Defined by IAU1958 (HI 21-cm fit to midplane)
- Sgr A* mass4.297 × 106 M☉ (GRAVITY 2022)
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why a separate coordinate system for the galaxy
Astronomers have three main spherical celestial coordinate systems in routine use: equatorial (right ascension and declination, anchored to Earth's rotation axis), ecliptic (anchored to Earth's orbital plane around the Sun), and galactic (anchored to the Milky Way disk). The first two are about Earth's geometry. Only the third is about the cosmos we actually want to map.
Consider an HI 21-cm survey of the Milky Way. The neutral hydrogen sits in a thin disk a few hundred parsecs thick. Plotted on an equatorial all-sky map, that disk traces a sinusoidal curve that meanders 60 degrees across the sky — the Milky Way wraps from Cassiopeia through Cygnus, down to Sagittarius, and back through Vela and Carina. Plotted in galactic coordinates, the same gas is a clean horizontal stripe at b = 0 with deviations of a few degrees. The bias of the chart no longer fights the bias of the data. Every Milky Way structural feature is easier to find, to compare, and to interpret in galactic coordinates than in equatorial coordinates.
The exact frame definition
The IAU 1958 system fixes the galactic frame by specifying three Euler angles (or equivalently, the equatorial position of one point in the galactic frame plus the position angle of the system at that point). The canonical numbers are:
North Galactic Pole (NGP):
α_NGP = 192.85948° (RA 12h 51m 26.28s, J2000)
δ_NGP = 27.12825° (Dec +27° 7' 41.7″)
Position in galactic frame: ℓ = 0, b = +90°
Longitude of the North Celestial Pole, expressed in galactic coordinates:
ℓ_NCP = 122.93192° (this fixes the rotation about NGP)
Galactic Center (ℓ=0, b=0):
α_GC = 266.40500° (RA 17h 45m 37.20s, J2000)
δ_GC = -28.93617° (Dec -28° 56' 10.2″)
Note: Sgr A* is at δ = -29° 0' 28″, ~4' from formal ℓ=0,b=0
because the IAU 1958 origin was best-fit to HI 21-cm, not to Sgr A*.The 4-arcminute offset between the formal galactic center direction and the Sgr A* position is a known artifact: when the IAU defined the galactic frame in 1958, the existence of a supermassive black hole at the centre of the Milky Way was not yet known. The frame was anchored to the centroid of the HI 21-cm emission map. The mismatch has not been re-defined since — standard practice is to use the IAU 1958 frame and to quote Sgr A* at its small offset from (ℓ, b) = (0, 0).
Worked example — converting Sgr A* from equatorial to galactic
The Sgr A* black hole sits at equatorial J2000 (α, δ) = (266.41684°, -29.00781°). Apply the IAU 1958 transformation:
Constants (J2000, IAU 1958):
α_NGP = 192.85948°, δ_NGP = 27.12825°, ℓ_NCP = 122.93192°
Step 1 — sin b:
sin b = sin(δ)·cos(δ_NGP)·cos(α - α_NGP) ... [full IAU formula]
Numerical: sin b = -0.001142 ⇒ b = -0.0654° = -3.93'
Step 2 — ℓ:
Solving simultaneously gives:
ℓ = 359.944° = -0.056° = -3.36'
Result: Sgr A* at galactic (ℓ, b) ≈ (-0.056°, -0.065°)
i.e., ~3.4 arcmin from formal ℓ=0 and ~3.9 arcmin from b=0.
Total offset from origin ≈ 5.1 arcmin = ~12 pc at R₀ = 8.2 kpc.That 5-arcmin offset is a residual of the 1958 best-fit and is reproduced by any standard library (astropy.coordinates.Galactic, IAU SLALIB, IRAF). The result is consistent across implementations to milliarcsecond precision when applied correctly.
Notable positions on the galactic sphere
| Object / feature | (ℓ, b) | Distance / nature | Notes |
|---|---|---|---|
| Galactic Center (Sgr A*) | (-0.056°, -0.065°) | 8.178 kpc | SMBH, 4.3 × 106 M☉ |
| Galactic Anticenter | (180°, 0°) | ~12 kpc behind Sun | Toward Taurus / Auriga |
| Galactic Bar (axis) | major axis at ℓ ~ 25° | length ~5 kpc | Tilted ~25° from Sun-GC line |
| Orion Arm (local) | ~(0-90°, ±5°) | 0-2 kpc | Solar neighbourhood |
| Cygnus X complex | (78°, +0.5°) | 1.4 kpc | Massive star-forming region |
| Carina Nebula | (287.5°, -0.6°) | 2.3 kpc | HII region, η Car |
| LMC (Large Magellanic Cloud) | (280.5°, -32.9°) | 49.6 kpc | Satellite dwarf spiral |
| SMC | (302.8°, -44.3°) | 62.4 kpc | Satellite irregular |
| M31 (Andromeda) | (121.2°, -21.6°) | 765 kpc | Spiral galaxy |
| North Galactic Pole | (0°, +90°) | — | Coma Berenices direction |
| South Galactic Pole | (0°, -90°) | — | Sculptor direction |
| CMB dipole maximum | (264.0°, +48.3°) | — | Sun's motion against CMB |
How the frame got fixed
- 1785. William Herschel publishes the first "galactic" star-count map of the heavens, identifying the Milky Way as a flattened disk system with the Sun roughly central. Anticipates a galactic plane but no formal coordinate system.
- 1845-1900. Lord Rosse, Otto Struve, and others map nebulae and the band of the Milky Way. The term "galactic equator" enters use, anchored to a visually estimated bright band.
- 1920. Harlow Shapley uses globular cluster distances to place the Sun far from the Galactic Center (~16 kpc in his estimate — about right within a factor of 2).
- 1932. Karl Jansky discovers radio noise from the Milky Way at 20.5 MHz, peaking in the direction of Sagittarius. The "Galactic Center" gains a directional definition independent of optical extinction.
- 1958. Working from extensive HI 21-cm data, the IAU defines the modern galactic coordinate system by specifying the position of the north galactic pole and the longitude of the galactic center. Replaces older systems based on B1900 equatorial.
- 1971. Sgr A West is identified as the compact radio source at the dynamical center of the galaxy by Balick & Brown. The black hole label "Sgr A*" follows in 1982.
- 2002. The Galactic Center distance R0 is refined to ~7.6 kpc using S-star orbits (Gerhard).
- 2018-2019. The GRAVITY interferometer measures R0 = 8.178 ± 0.013 kpc by direct astrometric monitoring of the S2 orbit, settling a long-running debate.
- 2022. EHT image of Sgr A* fixes the central BH mass at 4.297 × 106 M☉.
- 2024. Gaia DR4 publishes galactic-coordinate positions for > 1.8 billion sources to micro-arcsecond precision.
Where galactic coordinates show up
- HI 21-cm gas surveys. HI4PI (2016) measures atomic hydrogen brightness temperature across the whole sky on a (ℓ, b) grid — the cleanest existing map of disk gas.
- CO molecular surveys. Dame, Hartmann, Thaddeus (2001) traced ~106 CO clouds in (ℓ, b), revealing the molecular ring at R = 5 kpc.
- Pulsar catalogues. The ATNF Pulsar Catalogue (~3500 pulsars in 2024) is primarily plotted in galactic coordinates; pulsar populations cluster in &|b| < 5°.
- Star-formation regions. GLIMPSE (Spitzer 3.6 μm) covered (ℓ, b) in -65° < ℓ < +65°, |b| < 1° to identify all major MW HII regions.
- X-ray binaries and supernova remnants. The Galactic Center hosts ~1000 SNRs and ~600 X-ray binaries; the (ℓ, b) plot is the natural projection for population studies.
- CMB anisotropy maps. Planck and WMAP publish galactic-coordinate maps with the galactic stripe at b=0 masked out; the residual cosmic background is what's left.
- Extinction maps. Schlegel-Finkbeiner-Davis 1998 reddening E(B-V) is published in galactic coordinates because dust correlates with the disk plane.
- Gaia and astrometry. Gaia coordinates are formally in ICRS (equatorial) but immediately mapped to (ℓ, b) for any analysis of disk kinematics or substructure.
Common misconceptions
- "Galactic coordinates rotate with the galaxy." No — they are anchored to a fixed direction (the IAU 1958 specification). The galaxy rotates within the frame, the frame does not rotate.
- "ℓ=0 is exactly the supermassive black hole." The formal ℓ=0 is ~3 arcmin from Sgr A*'s position. Modern catalogues sometimes use a "Galactic Center frame" centred on Sgr A* itself, but this is not the IAU 1958 standard.
- "Galactic and ecliptic coordinates differ by a tilt of ~62 degrees." Roughly true — the angle between the Milky Way disk and Earth's orbit is about 60.2°. But the frames have offset origins as well.
- "Galactic latitudes don't precess." They don't precess with Earth's axis (the whole point), but Gaia-quality astrometry does need to account for proper motions of stars relative to the galactic frame.
- "All Milky Way objects have b ~ 0." Disk objects do (HI gas, young stars). But globular clusters, halo stars, dwarf satellites span all latitudes. Sagittarius A* itself is at b = -0.065° (just below the formal plane).
- "The North Galactic Pole is fixed in space." The Milky Way as a galaxy rotates, but the NGP — the direction perpendicular to the disk midplane — is fixed (to good approximation, given the < 109 yr timescales involved).
Frequently asked questions
What are galactic coordinates?
Galactic coordinates are a spherical celestial coordinate system whose equator is the midplane of the Milky Way disk and whose origin of longitude points toward the Galactic Center (Sgr A*). Galactic longitude ℓ is measured eastward in the disk plane from 0 to 360 degrees; galactic latitude b is the angular distance above (positive) or below (negative) the midplane, from -90 to +90 degrees. The system was defined by the IAU in 1958. Compared to equatorial (RA, Dec) coordinates which are anchored to Earth's rotation axis (and therefore precess with it), galactic coordinates are anchored to the galaxy itself.
Where is the Galactic Center?
The Galactic Center sits in the constellation Sagittarius at equatorial coordinates RA 17h 45m 40.04s, Dec -29° 0' 28.1″ (J2000), and at galactic ℓ=0, b=0. The supermassive black hole Sgr A* (mass 4.297 ± 0.013 × 106 solar masses, GRAVITY 2022) marks the exact center. Our distance to it was measured by the GRAVITY interferometer at 8.178 ± 0.013 kpc — the most precise distance ever measured to a galactic-scale object. Visible behind ~30 magnitudes of optical extinction; infrared, radio, and X-ray bands see through the dust.
Where is the north galactic pole?
The north galactic pole (NGP) is in the constellation Coma Berenices at J2000 equatorial coordinates RA 12h 51m 26.282s, Dec +27° 7' 41.704″. The orientation was chosen by the IAU to make the galactic equator best-fit the HI 21-cm intensity midplane. The south galactic pole sits in Sculptor at RA 0h 51m 26s, Dec -27° 7' 42″. The position angle between galactic and equatorial systems is determined by the longitude of the celestial pole (l = 122.93192°), b = 27.12825°.
How are galactic coordinates used in practice?
Galactic coordinates are the standard frame for any all-sky survey concerned with Milky Way structure: HI 21-cm gas surveys (HI4PI), CO molecular surveys (Dame et al.), pulsar catalogues (ATNF), star-formation regions (GLIMPSE, MIPSGAL), X-ray binaries, and the Wilkinson microwave background residual sky maps (where the Galactic plane stripe is the dominant feature requiring masking). Coordinates are typically quoted with ℓ first, then b — e.g. the Cygnus X complex is at (ℓ, b) ~ (78°, +0.5°), and the Magellanic Clouds at LMC (280°, -33°), SMC (303°, -44°).
How do you convert from equatorial to galactic?
The transformation is a fixed three-rotation matrix defined by the IAU 1958 / NGP orientation. Given equatorial (alpha, delta) in J2000, galactic (ℓ, b) is computed via: sin b = sin delta cos delta_NGP cos(alpha - alpha_NGP) - cos delta sin delta_NGP; the longitude ℓ is then recovered from the remaining trigonometric relations with alpha_NGP = 192.85948°, delta_NGP = 27.12825°, and the longitude of the celestial pole l_NCP = 122.93192°. Standard astronomy libraries (astropy, IAU SLALIB) provide these conversions to milliarcsecond precision.
Why not just always use equatorial coordinates?
Two reasons. (1) Equatorial coordinates precess with Earth's axis — a Gaia source at 2000 January 1 has slightly different RA/Dec from the same source at 2026 January 1 (~0.4 arcsec drift). Galactic coordinates are stable across centuries because they are tied to the galaxy itself. (2) For Milky Way science, the relevant geometry is the disk plane, not Earth's equator. Plotting HI distribution on an equatorial all-sky map shows the Milky Way as a sinusoidal wave; plotting it on galactic coordinates makes it a horizontal stripe at b=0 — much easier to interpret.
Why is the Sun above the plane?
The Sun lies approximately 20-25 pc north of the formal galactic midplane (b=0 line as defined by IAU 1958). Why north rather than south is not deeply meaningful — the Sun is part of an oscillation typical of disk stars, which bounce up and down through the plane every ~80 Myr with maximum excursion ~80 pc. The Sun is currently above the plane on its way back down. The 25-pc offset adds a small zero-point shift to the galactic latitudes of nearby objects: a source at b=0° formally is actually 25 pc / d below the Sun, where d is the distance. For sources at d ≫ 1 kpc this is negligible.