The Saros cycle is an eclipse cycle. It is a period of about 6585 + 1/3 days (approximately 18 years 11 days) which can be used to predict eclipses of the Sun and the Moon. One Saros period after an eclipse, the Sun, Earth, and the Moon return to approximately the same places and therefore almost identical solar and lunar eclipses repeat one Saros apart. For example, there was a total eclipse of the Sun on 11 August 1999 centered over Europe, and there will be a similar eclipse of the Sun on 21 August 2017. Due to the one third of a day fraction, Earth will be one third of a turn (120 degrees) further through its daily rotation so the eclipse will happen 120 degrees west, i.e. over North America.

The Saros was discovered by Babylonian astronomers several centuries BC. It is very useful since the calculations involved are simple. The only problem is that the next eclipse of the same Saros cycle occurs about 8 hours later in the day. In the case of an eclipse of the Sun this means the region of visibility shifts west one third of the way around the world, and most places from which the first eclipse was visible do not see any of the second one. In the case of an eclipse of the Moon the next eclipse might still be visible from the same location as long as the Moon is above the horizon. Therefore a longer cycle of three Saroses (54 years and a month, almost 19756 full days), known as a Triple Saros or exeligmos (Greek: "turn of the wheel"), has been used. After an exeligmos, an eclipse will again be visible at or near the original location.

In astronomical terms the Saros is due to several lunar and solar cycles repeating after about the same period of time:

• 223 synodic months (period from one New Moon to the next),
• approximately 242 draconic months (the period of the Moon to return to the ascending node of its orbit, i.e. cross the plane of Earth's orbit twice)
• 239 anomalistic months (period of the Moon to return to its perigee, i.e. the period of its elliptic orbit) and
• 18 anomalistic years.

Therefore the circumstances of an eclipse are also very similar to an eclipse one Saros earlier, and an eclipse (which happens when a conjunction or opposition of the Sun and Moon occurs in one of the nodes, that is, crossing the plane of the orbit) occurs again one Saros later.

So in principle one Saros after an eclipse, there will be another eclipse: however this does not repeat indefinitely because the match of the underlying periods (223 synodic = 242 draconic = 239 anomalistic months) is not perfect. In practice there is a long series of eclipses separated by one saros, that lasts many centuries but has a definite first and last eclipse.

At any one time theoretically there could be at most 223 possible Saros series of solar eclipses running simultaneously, because there are only 223 New Moons in the time span of a Saros. Similar for lunar eclipses at Full Moons.

• For solar eclipses, these Saros series have been numbered by van den Bergh (1955). Currently (2003) the 39 series numbered 117 to 155 are active, i.e. a solar eclipse occurs at a New Moon that belongs to one of these series. Solar Saros series last for 69 to 86 eclipses (1226 to 1532 years), but on average 77 eclipses (1370 yr). They start and end with partial eclipses, but have about 48 central (total or annular) eclipses around the middle of the series.

• For lunar eclipses, there are now 41 Saros series active. They last from 71 to 87 eclipses (1262 to 1551 years), but on average are not as long lived as for solar eclipses: 72 eclipses (1280 years), of which 40 to 58 are total.

The Saros cycle was probably known to the Chaldeans (ancient Babylonian astronomers), and later to Hipparchus, Pliny (Naturalis Historia II.10^*) and Ptolemy (Almagest IV.2), but not under this name. The Babylonian "Saros" appears to have been a name for a period of 3600 years. The name "Saros" was first given to the eclipse cycle by Edmund Halley in 1691, who took it from the Suda, a Byzantine lexicon of the 11th century. Halley's naming error was pointed out by Guillaume Le Gentil in 1756, but the name stuck.

An example: Lunar Eclipses of Saros 131

---

A single eclipse cycle can be considered a a representative example of the life cycle every Saros cycle. Considering only the lunar eclipses for simplicity because they are more visible, we can consider Saros cycle 131 as beginning in the year AD 1427 as a partial southern penumbral lunar eclipse and will end in the year AD 2707 as a partial northern penumbral lunar eclipse. The total cycle period was 1280 years.

Within this long period, there's a smaller period from 1950 to 2184 where total lunar eclipses will occur. We can consider 1950 as the birthday for the total lunar eclipse cycle:

• The first total lunar eclipse in the cycle began in 1950 when the moon crosses fully into the southern edge of the earth's shadow.
• The darkest and longest (central shadow) total lunar eclipse will occur in 2076 (126 years later)
• The final total lunar eclipse in the cycle ends in 2202 when the moon makes its final pass across the northern edge of the earth's shadow.

This total lunar eclipse cycle lifetime is 252 years.

Sets of three

The Saros cycle repeats approximately every 18 years, 10+1/3 days. The 1/3 day shift means the projected longitude on the Earth's for the maximum eclipse is shifted by about 120 degrees in sequential eclipses and it return again to approximately the same longitude after 3 cycles.

This is demonstrated below for the total eclipses from the perspective of the United States and western hemisphere, the eclipses are labeled as:

1. rising - seen east after sunset
2. setting - seen west before sunrise
3. down - invisible (occurs during the day)

In sequence, the Saros 131 lunar eclipse dates are: Partial lunar eclipses (Southern edge of shadow)

• May 10, 1427 (First partial penumbral south)
• ...
• July 25, 1553 (First partial umbral south)
• ...

1. March 22, 1932 (Final partial umbral south)

Total lunar eclipses

1. April 2, 1950 (0 years) – down - First total
2. Apr 13, 1968 (18 years) – rising
3. Apr 24, 1986 (36 years) - setting
4. May 4, 2004 (54 years) – down
5. May 16, 2022 (72 years) - rising
6. May 26, 2040 (90 years) - setting
7. Jun 6, 2058 (108 years) – down
8. Jun 17, 2076 (126 years) – rising - Central
9. Jun 28, 2094 (144 years) - setting
10. Jul 8, 2112 (162 years) – down
11. Jul 21, 2130 (180 years) - rising
12. Jul 31, 2148 (198 years) - setting
13. Aug 11, 2166 (216 years) – down
14. Aug 21, 2184 (234 years) – rising
15. Sep 3, 2202 (252 years) – setting – Last total

Partial lunar eclipses (Northern edge of shadow)

• Sep 13, 2220 (First new umbral partial north)
• ...
• Apr 9, 2563 (Last partial umbral north)
• ...
• Jul 7, 2707 (Last partial penumbral north)

Saros 131 reference

• Saros series 131 table

External links

---

• Eclipse Search -- here one can search 5,000 years of eclipse data by type, magnitude, Saros number or simply by year.

Reference

---

• G. van den Bergh, Periodicity and Variation of Solar (and Lunar) Eclipses, 2 vols. H.D. Tjeenk Willink & Zoon N.V., Haarlem, 1955

Eclipses

Saros | Sarosperiode | Saros | Ciclo di Saros | サロス周期 | Saros-cyclus | Сарос | 沙罗周期