Orbital mechanics · checked against NASA/JPL
Analemmas
A sundial agrees with mean solar time on only four days a year. The rest of the time the Sun runs up to 16 minutes ahead or 14 behind, and its noon position draws a figure eight. This page computes that curve with a small ephemeris engine and follows the same mechanics to the retrograde loops of the planets and the pentagram of Venus.
What an analemma is
An analemma is the closed curve traced by the apparent position of a celestial body when it is observed from a planet's surface at the same mean solar time over a full orbital cycle. The word comes from the Greek ἀνάλημμα ("support", "pedestal"); in antiquity it named a geometric construction for sundials described by Vitruvius, and its modern astronomical sense —the figure-eight curve— only became common in the 18th century (Sawyer, 1994). The physical basis of the curve, the equation of time, was first tabulated rigorously by Flamsteed (1672); its history and modern computation are reviewed in Hughes, Yallop and Hohenkerk (1989).
The resulting shape is the sum of two physical effects acting at once. If both were zero, the analemma would shrink to a point.
Effect I — Orbital eccentricity
Orbits are ellipses with the Sun at one focus (Kepler's 1st law). By the 2nd law (the law of areas), a planet moves faster at perihelion and slower at aphelion. This uneven speed makes the Sun "arrive" late or early compared with a clock that keeps uniform time (Meeus, 1998, pp. 183–185).
Effect II — Axial obliquity
The rotation axis is tilted with respect to the orbital plane (obliquity ε). The tilt makes the projection of the Sun's motion onto the celestial equator uneven through the year and produces the north–south component of the analemma. The contribution Eobl(λ) vanishes at the equinoxes and solstices and reaches its four extremes near λ ≈ 45°, 135°, 225° and 315°, halfway through each season (Müller, 1995; Meeus, 1998, chap. 28).
Analemmas appear on 18th- and 19th-century globes as small figure eights in the Pacific Ocean: they helped navigators and clockmakers convert between solar time and mean time (Sawyer, 1994; Waugh, 1973).
Mathematical formalism
The analemma is computed from the equation of time E(t), the difference between the hour angle of the true Sun and that of the fictitious mean Sun. It splits into two physically independent terms (Duffett-Smith, 1990; Hughes et al., 1989; Meeus, 1998, chap. 28):
The simulation does not stop at the series. It computes E from its definition, the Greenwich hour angle of the apparent Sun minus mean solar time, with the Earth rotation angle and the IAU 2006 sidereal-time polynomial (Petit & Luzum, 2010); the series is kept to show the two effects apart, and both agree to within one second.
An asymmetric figure eight
When Eobl dominates Eexc, the curve forms a figure eight. Its lobes are unequal because the Sun moves faster near perihelion: on Earth, perihelion (January) falls close to the December solstice, so the southern lobe is the larger one (Müller, 1995).
Other shapes
The shape depends on the relative weight of Eexc (∝ 2e) and Eobl (∝ tan²(ε/2)). On Jupiter (ε ≈ 3.1°, e ≈ 0.048) eccentricity dominates and the curve is an oval with no crossing. On Mars the figure is a teardrop, photographed by the Opportunity rover between 2006 and 2008 (Lakdawalla, 2014). On Mercury the classic analemma cannot even be built: its solar day (~176 Earth days) lasts two of its years (Urschel, n.d.).
How large Earth's equation of time gets
In 2024 the extremes of E(t) were −14 min 12 s (11 February), +3 min 39 s (13 May), −6 min 33 s (25 July) and +16 min 27 s (2 November); they shift by a few seconds from year to year. The components behave differently: Eexc vanishes at perihelion and aphelion and reaches ±7.7 min in early April and early October; Eobl vanishes at the equinoxes and solstices and reaches ±9.9 min around the middle of each season (Hughes et al., 1989; Meeus, 1998, chap. 28; Müller, 1995).
Earth's solar analemma
Drag the slider or use ← → on the chart to go day by day (Shift for ten days); hover the curve to read any date. Any year from 1800 to 2050.
The first photograph of the analemma was taken by Dennis di Cicco in 1978–1979 with multiple exposures on a single negative, showing 44 positions of the Sun in a figure eight (di Cicco, 1979). Many astrophotographers have repeated it since, now digitally (Ayiomamitis, 2004).
Retrograde loops
Seen from Earth, a planet usually drifts eastwards among the stars. Around opposition (the outer planets) or inferior conjunction (Mercury and Venus) it stops, backs westwards for weeks or months and then carries on: a retrograde loop. Nothing happens to the planet; it is the Earth overtaking it, or being overtaken, on the inside lane (Meeus, 1998, chap. 36).
The animation shows both views at once. On the left, the orbits seen from above, with the line of sight from the Earth through the planet; on the right, where that line points on the sky, with the retrograde stretch in red between the two stationary points S1 and S2. Every event is a real one between 1800 and 2050, and its dates match the JPL ephemeris to within a few hours. The pattern repeats with the synodic period S:
◀ ▶ jump to the previous or next event. On the chart, ← → move day by day; hover the sky path to read any date. East is on the left, as on a sky chart facing south.
The pentagram of Venus
"Cycles" overlays the stars of the following 8-year cycles in blue, to show how the figure turns. ◀ ▶ move a whole cycle between 1800 and 2050.
Instead of a figure eight, the five inferior conjunctions of Venus in a ~8-year cycle draw an approximate pentagram on the zodiac (Aveni, 2001). The figure is neither perfect nor closed: it comes from a near orbital resonance (periods: Bretagnon & Simon, 1986; Williams, 2025):
Each inferior conjunction happens on average ≈ 215.5° further on in ecliptic longitude (≡ 144.5° back): joining them in chronological order gives the {5/2} star, with its five vertices ≈ 72° apart. But the sixth conjunction does not land on the first: it falls ≈ 2.3° behind, so the pentagram does not close and turns a full circle in ≈ 1 250 years. When the cycle completes, the red marker shows the sixth conjunction. Seen from above, the same five dates mark five points of the Earth's orbit, and the Earth–Venus line sweeps out a five-petalled rose.
Venus in ancient astronomy
The oldest surviving systematic observations of Venus come from Mesopotamia: the "Venus tablet of Ammisaduqa" records its heliacal risings and settings and is dated to the 17th century BCE (Hunger & Pingree, 1999). The Venus table of the Maya Dresden Codex (c. 11th–13th century) spans 65 synodic periods —about 104 years— with a canonical value of 584 days and ways of correcting the accumulated error (Aveni, 2001; Bricker & Bricker, 2011). The relation 2920 = 8 × 365 = 5 × 584 days was built into the Maya calendar.
Solar system comparison
| Body | e | ε | Sidereal T | Synodic S* | Retrograde† | Note |
|---|---|---|---|---|---|---|
| ● Sun / Earth | 0.01671 | 23.44° | 365.256 d | — | — | Solar analemma: asymmetric figure eight, E(t) ∈ [−14, +16.5] min |
| ● Moon | 0.05490 | 6.68° | 27.32 d | 29.53 d | never | Orbital inclination 5.14°; always moves east among the stars |
| ● Mercury | 0.20560 | 0.034° | 87.97 d | 115.9 d | 20–24 d · 8.5–16° | Greatest elongation 18°–28° |
| ● Venus | 0.00677 | 177.4° | 224.7 d | 583.9 d | 41–43 d · 15–16.5° | 8:13:5 near resonance: the pentagram |
| ● Mars | 0.09340 | 25.19° | 686.98 d | 779.9 d | 60–81 d · 10–19.5° | Longest loops at aphelic oppositions, shortest at perihelic ones |
| ● Jupiter | 0.04890 | 3.13° | 4 332 d | 398.9 d | 118–124 d · 10° | One loop a year |
| ● Saturn | 0.05650 | 26.73° | 10 759 d | 378.1 d | 134–141 d · 7° | Loops nearly identical year after year |
| ● Uranus | 0.04720 | 97.77° | 30 685 d | 369.7 d | 149–154 d · 4° | Retrograde rotation (ε > 90°) |
| ● Neptune | 0.00860 | 28.32° | 60 189 d | 367.5 d | 156–160 d · 2.8° | Barely moves between cycles |
* Synodic period relative to Earth. † Duration between the stationary points and arc travelled backwards, minimum–maximum over 1900–2050, computed with this site's engine. e, ε and T: NASA fact sheets (Williams, 2025); the simulation uses the mean elements of Standish and Williams (1992), which differ slightly (e.g. eJupiter = 0.0484, eSaturn = 0.0539).
Minor bodies and exoplanets
Pluto (e = 0.249, ε ≈ 122.5°; Williams, 2025) would produce one of the most complex analemmas. On exoplanets with e > 0.3 and arbitrary obliquities, insolation patterns and their analemmas depart radically from the terrestrial case (Dobrovolskis, 2013; Heller et al., 2011).
Method and references
Every figure on this page comes from one file, astro.js, about 350 lines with no
dependencies. It turns a date into apparent positions the way an ephemeris does, only with Keplerian
elements instead of numerical integration:
The engine is checked on every push by validacion.mjs against reference data downloaded from
the NASA/JPL Horizons system (DE441 ephemeris; Park et al., 2021) by tools/fetch-horizons.mjs,
and against the test values of the IAU's own libraries. What remains is the error of the Keplerian elements
themselves, which ignore mutual perturbations between planets; refraction and the observer's parallax are
also left out.
| Body | RMS error | Max. error | Body | RMS error | Max. error |
|---|---|---|---|---|---|
| Sun | 9″ | 27″ | Jupiter | 211″ | 614″ |
| Mercury | 10″ | 35″ | Saturn | 360″ | 822″ |
| Venus | 13″ | 62″ | Uranus | 51″ | 126″ |
| Mars | 34″ | 206″ | Neptune | 29″ | 61″ |
| Quantity | Cases | Largest difference |
|---|---|---|
| Equation of time, 1805–2050 (six years, every 5 days) | 440 | 1.1 s |
| Its four extremes in 2024 | 4 | 0.25 s |
| Series Eexc + Eobl against the exact E | 440 | 0.95 s |
| Inferior conjunctions of Venus, 1995–2035 | 25 | 14 min |
| Inferior conjunctions of Mercury, 1995–2035 | 126 | 4 min |
| Oppositions and stationary points of Mars | 56 | 1.2 h |
| Oppositions and stationary points of Jupiter and Saturn | 226 | 4.2 h |
| Transits of Venus 1995–2035 | 2 | both found: 2004-06-08 and 2012-06-06 |
Nutation reproduces the ERFA test value to 10⁻¹⁹ rad and Meeus's example 22.a to 0.001″. Times are TT; ΔT = TT − UT1 is neglected, which moves the equation of time by less than 0.2 s.
References
- Aveni, A. F. (2001). Skywatchers: A revised and updated version of Skywatchers of ancient Mexico. University of Texas Press.
- Ayiomamitis, A. (2004, January). Afternoon analemma for Athens [Photograph]. Earth Science Picture of the Day, Universities Space Research Association. https://epod.usra.edu/blog/2004/01/afternoon-analemma-for-athens.html
- Bretagnon, P., & Simon, J.-L. (1986). Planetary programs and tables from −4000 to +2800. Willmann-Bell.
- Bricker, V. R., & Bricker, H. M. (2011). Astronomy in the Maya codices (Memoirs of the American Philosophical Society, Vol. 265). American Philosophical Society.
- Capitaine, N., Wallace, P. T., & Chapront, J. (2003). Expressions for IAU 2000 precession quantities. Astronomy & Astrophysics, 412(2), 567–586. https://doi.org/10.1051/0004-6361:20031539
- di Cicco, D. (1979, June). Exposing the analemma. Sky & Telescope, 57(6), 536–540.
- Dobrovolskis, A. R. (2013). Insolation on exoplanets with eccentricity and obliquity. Icarus, 226(1), 760–776. https://doi.org/10.1016/j.icarus.2013.06.026
- Duffett-Smith, P. (1990). Astronomy with your personal computer (2nd ed.). Cambridge University Press.
- Flamsteed, J. (1672). De inaequalitate dierum solarium dissertatio astronomica. London.
- Giorgini, J. D., Yeomans, D. K., Chamberlin, A. B., Chodas, P. W., Jacobson, R. A., Keesey, M. S., Lieske, J. H., Ostro, S. J., Standish, E. M., & Wimberly, R. N. (1996). JPL's on-line solar system data service. Bulletin of the American Astronomical Society, 28(3), 1158.
- Heller, R., Leconte, J., & Barnes, R. (2011). Tidal obliquity evolution of potentially habitable planets. Astronomy & Astrophysics, 528, A27. https://doi.org/10.1051/0004-6361/201015809
- Hughes, D. W., Yallop, B. D., & Hohenkerk, C. Y. (1989). The equation of time. Monthly Notices of the Royal Astronomical Society, 238(4), 1529–1535. https://doi.org/10.1093/mnras/238.4.1529
- Hunger, H., & Pingree, D. (1999). Astral sciences in Mesopotamia. Brill.
- IAU SOFA Board. (n.d.). IAU SOFA software collection. International Astronomical Union. https://www.iausofa.org/
- Lakdawalla, E. (2014, May 6). A Martian analemma. The Planetary Society. https://www.planetary.org/articles/a-martian-analemma
- Meeus, J. (1998). Astronomical algorithms (2nd ed.). Willmann-Bell.
- Müller, M. (1995). Equation of time — Problem in astronomy. Acta Physica Polonica A, 88(Suppl.), S-49.
- Park, R. S., Folkner, W. M., Williams, J. G., & Boggs, D. H. (2021). The JPL planetary and lunar ephemerides DE440 and DE441. The Astronomical Journal, 161(3), 105. https://doi.org/10.3847/1538-3881/abd414
- Petit, G., & Luzum, B. (Eds.). (2010). IERS Conventions (2010) (IERS Technical Note No. 36). Verlag des Bundesamts für Kartographie und Geodäsie.
- Sawyer, F. W., III. (1994, June). Of analemmas, mean time and the analemmatic sundial — Part I. Bulletin of the British Sundial Society, 6(2), 2–6.
- Standish, E. M., Newhall, X. X., Williams, J. G., & Yeomans, D. K. (1992). Orbital ephemerides of the Sun, Moon, and planets. In P. K. Seidelmann (Ed.), Explanatory supplement to the astronomical almanac (Rev. ed., chap. 5, pp. 279–323). University Science Books.
- Standish, E. M., & Williams, J. G. (1992). Approximate positions of the planets. NASA Jet Propulsion Laboratory, Solar System Dynamics. https://ssd.jpl.nasa.gov/planets/approx_pos.html
- Urschel, B. (n.d.). Other analemmas. Analemma.com. https://www.analemma.com/other-analemmas.html
- Waugh, A. E. (1973). Sundials: Their theory and construction. Dover Publications.
- Williams, D. R. (2025, March 18). Planetary fact sheet — Metric. NASA Goddard Space Flight Center. https://nssdc.gsfc.nasa.gov/planetary/factsheet/