Eclipses, Cosmic Clockwork of the Ancients


 

Eclipses have long held a special fascination for humanity. Observing and recording eclipses had a role in early advancement of human knowledge. Eclipses reveal the geometry of the solar system. Eclipse knowledge facilitates contemporary understanding of the cosmos as well as of prehistory and of the history of science. Eclipses function as temporal and spatial references for naked-eye astronomers, presenting an accurate, readily-apparent cosmic clock.

Eclipses are accurately documented in early history and archaeological studies demonstrate knowledge of eclipses in ancient cultures. Today, eclipse records made by ancient astronomers are important to science. The very oldest available eclipse records are used to determine values for temporal changes in the length of day and how fast the earth rotates. Better understandings of and finding additional ancient eclipse records would further assist modern astronomy and archaeology. Eclipse records also aid historians in fixing exact dates of past events.


Contents


 

Pueblo Bonito
 
Introduction

The history and philosophy of science encompasses interest in the discovery of and the role of astronomical knowledge in ancient cultures. Evidence of transmission of mathematics, calendars, and astronomy is a useful indicator of complexity of culture, contact, diffusion, and the relationships of ancient civilizations. Every culture has a cosmovision—understandings and interpretations of the natural universe—and astronomy practices reflect a culture's framework of knowledge. Long before the earliest written records, eclipses played an important role in attaining understandings of earth's cosmology in diverse cultures.

Practical applications of astronomy—function to anthropologists—include calendar keeping, prediction of tides, timing game migrations and other resource cycles, navigation on land and sea, place determination, mathematical geography, surveying, and development of fundamental understandings of reality. In some cases these very activities, especially calendars with cosmic tally keeping, are also a foundation of further advancement in astronomical knowledge. Astronomy is an important force in the development of science and, at the same time, the history of astronomy is considered to be one of the most fragmentary chapters in the history of science.

Envisioning the geometry of the cosmos makes the mechanics of eclipses understandable, and vice-versa. To fully appreciate some of the complexities of eclipses, especially the human ability to predict eclipses, understanding the mathematics and fundamental astronomy is useful. With a naked-eye perspective, the Fundamental Astronomy section presents astronomy basics in the easiest terms possible and introduces terse code terms used in the Excel astronomy calculator accompanying this article. To use the Epoch Calc calculator, you simply input the intuitive code terms for correlated complex numbers, and the applet enters and computes the math for you. When you enter "dy" for example, the precise value for the temporal variable "days per year" is entered for the epoch you choose. Enter a number of moons in the eclipse calculator and the corresponding number of days displays, accurate for the epoch you select.


Eclipse calculator, constants, code: epoch_v2012.xls
Eclipse Calc, an eclipse calculator

The aegeo worksheet in the applet provides convenient drop-down menus to look up terms, code, and the value of variables. Astronomical terminology in this article and in Epoch Calc varies from some common conventions; to better conform with actuality, erroneous, post-Dark Age metaphors are replaced and cycles and periods are distinguished. Only the tropical year is termed a year, only lunar phases are termed a month. Not only are common terms often confusing, but after centuries of knowing we live in a heliocentric solar system it is due time to admit same into our language. In this regard, my own use of terms continues to grow, so expect updated article and applet versions in the future.  Also, we know cosmic bodies are not gods, so I dispense with capitalization.

 

Aztec Great Kiva
 
Fundamental Astronomy
: Earth Motions


Orbits and Rotations, Years and Days

Three astronomical motions referenced to fixed celestial space, earth axial rotation (r), earth solar orbit (o), and lunar orbit of the earth (l), present natural temporal modules. Our point of reference for the day (d) and year (y) is the sun, the one star that apparently—due to our perspective from an orbiting earth—moves in relation to the fixed celestial background. The length of time in the tropical year period is determined by precession motion. The day and year periods each result from combinations of the fundamental motions referenced to fixed heliocentric space.

Earth's solar orbit, one revolution around the sun, is a geometric circumferential motion referenced to fixed celestial space. Because the earth is orbiting the sun at the same time and in the same direction as it rotates on its axis, earthbound observers experience one less day per solar orbit than there are rotations. Simply stated, the sun is the one star that passes overhead one time less per solar orbit because we circle the sun once. After each rotation, the angular position of the sun has changed due to solar orbit, so the day, realigning towards the sun, is longer than rotation, realigning with the other stars.

The current value for rotations per orbit (ro) equals 366.25636. With precisely one less day per orbit than rotations, days per orbit (do) equals 365.25636 (ro - 1 = do). The day is longer than one rotation by the inverse of rotations per orbit (dr + or = 1.0). Table 1 presents the current (year 2000.0) value of these constants and the derived values for rotations per day (rd) and days per rotation (dr).

Table 1. Orbit, Rotation, and the Day
code
term
value
ro
rotations per orbit
366.25636
do
days per orbit
365.25636
or
orbit per rotation
0.0027303
od
orbits per day
0.0027378
rd
rotations per day
1.0027378
dr
days per rotation
0.9972697

In contrast to an orbit, the tropical year is a more complex formulation referenced to the sun. An orbit is a cycle; orbits begin and end at the same position in spatial reference. The year is a period; it begins and ends at unique points in an orbit. Years are the product of two motions, solar orbit and the earth's precession cycle (p). Precession is the motion of the earth's axis of rotation gradually changing direction in relation to the sun and fixed celestial reference. The seasons and the year result from a.) inclination of and b.) precession motion of the earth's rotation axis in combination with c.) solar orbit motion. The axis inclines the nearest pole away from the sun in winter and, on the opposite side of the sun, towards the sun in summer. Precession very gradually changes which stars are seen in the winter and summer night sky, slowly switching on which side of the sun each pole is nearest.

obliquity of the ecliptic and precession of the equator

Precession (p) motion is a retrograde rotation of axial inclination. The year 2,000 rate of precession was 25,770.17 orbits per precession cycle. The rate is currently slowly decreasing. Retrograde precession realigns the axis inclined fully towards the sun again slightly before completion of an orbit. There is just over one year in a solar orbit, so in one 360° precession of the rotation axis there is one more year per precession (yp) than orbits per precession (op). The year is shorter than one orbit by the inverse of orbits per precession cycle. The formula for years per solar orbit (yo) is yo = 1 - 1/op. The formula for days per year (dy) in reference to fixed heliocentric space demonstrates the appeal of believing the sun simply orbits the earth.

dy = (ro - 1) x (1 - 1/op)

This sidereal (fixed space) equation illustrates how the day and year are complex astronomical formulations combining the three fundamental earth motions: solar orbit, axial rotation, and precession. As such the day and year are not as mathematically practical to or easily quantified by the astronomer as are fundamental orbits and rotation, especially to the pre-computer age astronomer. Table 2 illustrates these formulations and values (Terrestrial Time, year 2000.0).

Table 2. Orbit, Precession, and the Year
code
term
value
do
days per orbit
365.2563605
dy
days per year
365.2421926
od
orbit per day
0.0027378
oy
orbit per year
1.0000388
yo
years per orbit
0.9999612
op
orbits per precession
25,770.17
yp
years per precession
25,771.17

x - 1 = y

x = 366.25636 rotations per orbit
y = 365.25636 days per orbit

x = 1.00273780 rotations per day
y = 0.00273780 orbits per day

x = 1.0000388 years per orbit
y = 0.0000388 precession cycles per year

x = 25,771.17 years per precession cycle
y = 25,770.17 orbits per precession cycle

,

Hovenweerp Castle
 
Fundamental Astronomy: Lunar Motion


Lunar Orbit

Lunar orbits are most easily referenced to the stars. Each time the moon moves past a specific star, it has circled the earth once. Each time a specific star passes overhead, the earth has rotated once. By definition, lunar orbits and rotations each equal one circumference, a 360 degree revolution. Thus with reference to fixed space (the distant stars), this fundamental ratio of two cosmic motions is readily apparent. References to fixed space are termed sidereal—lunar orbits and rotations are sidereal motion cycles.

Lunar orbit and rotation are also the only two fundamental motions presenting both ease of counting and such an easy comparison of their ratio. This fact is important to naked-eye astronomy and is the likely method of attaining accuracy in ancient astronomies. Rotations can be counted and accurately timed using an observation position, a foresight, and a bright star. Each time the star passes the foresight, a rotation has occurred. Using the same stellar reference, lunar orbits and earth rotations can be directly counted and their ratio easily determined.

lunar and solar orbit

Mean lunar orbital motion per earth rotation is the fundamental ratio of the counts underlying a sidereal lunar ephemeris and calendar. Accuracy is easily achieved given sufficiently long counting when both counts begin at the same instant, and the longer the count the greater the accuracy. Simply by thinking and counting, ancient astronomers could precisely determine rotations per lunar orbit with naked eye observations. Table 3 presents the sidereal lunar motion variables, ratios, and several modules.

Table 3. Sidereal Lunar Orbit
code
term
value
degrees
rl
rotations per lunar orbit
27.396462
 
dl
days per lunar orbit
27.3216616
 
lr
lunar orbits per rotation
0.03650106
 
ld
lunar orbits per day
0.03660100
 
s27
solar orbits per lunar orbit
0.07480133
26.9284786
r27
lunar orbits per rotation
  
13.1403824
c27
lunar orbits per day
  
13.1763582

Before clocks, it was possible but not easy to know the time of day using solar angle. On the inclined, rotating earth, solar angle changes with the seasons.  And, the sun is not visible at night. Stellar angles are not the same at the same solar time each night of the year because days and rotations are distinct temporal subdivisions. Night timekeeping is sidereal timekeeping while day timekeeping is solar. During the daylight hours, only a total solar eclipse allows readily observing sidereal time and lunar position.

Equating lunar motion with days is not straightforward. However, in due time, sidereal counting reveals the ratio. Using a long duration of counts, days per lunar orbit (dl) can be determined from rotations per lunar orbit (rl). The difference between lunar orbit per rotation (lr) and lunar orbit per day (ld) is 0.00010 lunar orbits (0.035976°). The difference equals 0.00009993 circumference (cir/10,006.7). In 365 lunar orbits there are 9,999.709 rotations, in 366 lunar orbits, there are 9,999.728 days. Rounding to the nearest integer:

In 365 lunar orbits there are 10,000 rotations.
In 366 lunar orbits there are 10,000 days.

lr = 0.0365011
ld = 0.0366010

 Mimbres serpents
 
Lunar Orbit Nodal Periods

Solar eclipses occur at new moon and lunar eclipses at full moon only when their alignments occur in three dimensions. Relative to earth's orbit, the plane of lunar orbit is inclined. Mean inclination of lunar orbit (il) equals 5.1454 degrees. Eclipses only occur near the nodes of lunar orbit intersection with the solar orbital plane. Earth's mean orbital plane is termed the ecliptic (synonymous with eclipse). There are two nodal crossings of the ecliptic per nodal period, the ascending node and the descending node. Half the nodal period is the shortest possible interval between two eclipses. Table 4 presents lunar nodal period variables.

The difference between the nodal and synodic periods determines an eclipse nodal period, the interval between successive alignments of the lunar nodes to the sun.  For this cycle, I substitute the term "eclipse nodal interval" (e) for the common usage "eclipse year." 

Table 4. Lunar Nodal Periods
code
term
value
rn
rotations per lunar nodal period
27.2867225
dn
days per lunar nodal period
27.2122208
nr
lunar nodal periods per rotation
0.03664786
re
rotations per eclipse nodal interval
347.569052
de
days per eclipse nodal interval
346.6200745
il
inclination of lunar orbit, degrees
5.1453964
s22
solar orbit per lunar nodal period, degrees
26.8206128

lunar orbit inclination

 Penasco Blanco pictographs
 
Lunar Anomalistic Period

The lunar anomalistic period is the duration of the elliptical apogee and perigee of lunar orbit. Lunar orbit eccentricity (me) equals 0.0549005. The elliptical shape of lunar orbit effects the speed of lunar motion, the geocentric geometry of lunar position, and the type and duration of eclipses. The size of the proximate moon and the distant sun appear nearly equal from the earth. Whether the moon is near apogee or perigee determines if a solar eclipse is annular or total. Table 5 presents lunar anomalistic period variables.

Table 5. Lunar Anomalistic Period
code
term
value
ma
moon, distance, apogee, meters
406,720,000.0
mp
moon, distance, perigee, meters
356,375,000.0
loe
lunar orbit eccentricity, degrees
0.054900489
mw
moon angular width, geocentric mean, degrees
0.518102946
la
lunar orbits per lunar anomalistic period
1.0085237
ra
rotations per lunar anomalistic period
27.629985
da
days per lunar anomalistic period
27.554547
s25
degrees solar orbit per lunar nodal period
26.820613

Sun dagger at fajada Butte.
 
Fundamental Astronomy: Phases of the Moon


Lunar Synodic Period

The linear alignment of the sun, earth and moon—when the three objects conjoin in the same order to form three nodes on a straight, two-dimensional line—is the lunar synodic period. The lunar illumination phases seen from the earth are called moons (hence months). Eclipses occur during these alignments due to shadow crossings. Lunar synodic period is one fundamental harmonic of eclipses, and the synodic period is the shortest possible interval between successive lunar or solar eclipses. The synodic and nodal periods intercalate as eclipse cycles.

lunar synodic

The orbits of the moon and of the earth are independent motions. The moon orbits the earth in the same direction as the earth orbits the sun, so the earth bound observer sees one less illumination cycle (full moon) than lunar orbits per orbit of the sun (lo - 1 = mo). Simply stated, there is one less lunar synodic period per solar orbit because the moon has also circled the sun once (x - 1 = y).

After each lunar orbit, the sidereal angular position of the sun has changed due to the motion of solar orbit.  The lunar phase period is longer than lunar orbit, just as the day is longer than rotation and solar orbit is longer than the year. One moon is longer than one lunar orbit by the inverse of lunar orbits per solar orbit (ml + ol = 1.0). Table 6 presents synodic lunar period variables and ratios.

Table 6. Synodic Lunar Period (moon)
code
term
value
rm
rotations per moon
29.6114378
dm
days per moon
29.5305889
lm
lunar orbits per moon
1.0808489
om
solar orbits per moon
0.0808489
ml
moons per lunar orbit
0.9251987
s29
degrees solar orbit per lunar synodic
29.1056175

x - 1 = y

x = 13.368746 lunar orbits per solar orbit
y = 12.368746 lunar synodic periods per solar orbit

x = 1.08084894 lunar orbits per moon
y = 0.08084894 solar orbit per moon


Yellow Jacket sunrise.
 
Fundamental Astronomy: Lunar Standstill Period

The 18.613 year lunar standstill period (s) is a geometric product, the result of the turning movement of the inclined axis of lunar orbit in combination with the equatorial precession cycle. The cycle of lunar orbit inclination direction turning (t) causes the lunar node to regress around the ecliptic. Due to the inclination of the earth's axis (ob) and inclination of lunar orbit (il), the moon reachs lower or higher declination in relation to earth's orbital plane, the ecliptic.  The moon rises and sets north and south of due east and due west each nodal period just as the sun does each year.

As the moon deviates from the ecliptic plane between nodal crossings, lunar orbit inclination either adds to or subtracts from axial inclination in harmony with earth orbits of the sun. The inclinations of lunar orbit and axis obliquity combine to determine the maximum geocentric angle of the moon from the ecliptic. The terms lunar major and lunar minor describe the variable extrema of nodal period declinations. For example, full moons nearest solstices during standstill add lunar inclination to axis inclination (ob + il), while at mid-cycle 9.3 years later on the opposite side of the sun, lunar orbit inclination is instead subtracted (ob - il).

The plane of lunar orbit turning (t) in fixed space regresses nodal crossings slightly before lunar orbit completions, effecting the shorter nodal period. During each full-circle regression of the node, there is one more nodal period than lunar orbits (lt + 1 = nt). The cycle of one revolution of the node around the ecliptic (1.0 t) is not equal to one standstill period (1.0 s) because earth's rotation axis is also changing direction. Precession altering earth's axis direction complicates the geometry and timing of lunar standstills. The periodicity of lunar standstills is determined by linear alignment to the sun of earth's axial inclination direction and the ecliptic node of lunar orbit. The sidereal cycle of lunar orbit turning effects one revolution of the lunar nodes.

Orbits and revolutions in fixed space are the geometric foundation for equating solar system motions. During one precession cycle there is one less standstill period(s) than revolutions of the regresssing lunar node (t). Lunar orbit inclination rapidly rotates in the same direction as axis inclination slowly precesses. The change in axial inclination direction impacts both year length and duration of lunar standstill. One precession cycle equates both orbits to years (op + 1 = yp) and eclipse nodal periods to standstill cycles (ep - 1 = sp). Eclipse nodal intervals per lunar orbit gyration less one equals solar orbits per lunar orbit gyration. Eclipse nodal periods per lunar standstill period less one equals years per lunar standstill period (es - 1.0 = ys). Table 7 presents lunar standstill period variables and ratios.

Table 7. Lunar Standstill Period
code
term
value
me
moons per eclipse nodal interval
11.7376621
ne
nodal periods per eclipse nodal interval
12.7376621
et
eclipse nodal intervals per lunar orbit turn
19.5992010
ot
orbits per lunar orbit turn
18.5992010
es
eclipse nodal intervals per lunar standstill period
19.613356
os
orbits per lunar standstill period
18.612634
lt
lunar orbits per lunar orbit turn
248.648000
nt
lunar nodal periods per lunar orbit turn
249.648000
ls
lunar orbits per lunar standstill period
248.827581
ns
nodal periods per lunar standstill period
249.828303

x - 1 = y

x = 12.73766 nodal periods per eclipse nodal interval
y = 11.73766 moons per eclipse nodal interval

x = 19.59920 eclipse nodal intervals per lunar orbit turn
y =
18.59920 orbits per lunar orbit turn

x = 248.64800 eclipse nodal intervals per lunar orbit turn
y =
249.64800 eclipse nodal intervals per lunar standstill period

x = 19.613356 eclipse nodal intervals per lunar standstill period
y = 18.613356 years per lunar standstill period

 

Big Horn Medicine Wheel
 

Lunar Eclipses

Solar and lunar eclipses are very distinct. The moon's shadow during a total solar eclipse is only a narrow band on the earth. The earth's conic shadow at the moon's mean distance is over 9,000 km wide, nearly three lunar diameters. Only a small percentage of people experience each solar eclipse while half the world can view each lunar eclipse. Lunar eclipses provide a time standard; occultations begin and end nearly simultaneously for all viewers. The widest angular lunar viewing difference (parallax) for two positions on the earth is under two degrees. While observers see the same eclipse timing, different positions see the moon at very different angles in the sky. During an eclipse, the sidereal position and the local celestial longitude of the moon can be employed to compare longitudes of observers. Observations of sidereal lunar position during eclipses also provides observations inferring values of precession and the solar orbit cycle.

Coloration and visibility of the moon varies with each lunar eclipse. Some are a dark gray or brownish, others brightly glow a warm reddish orange due to indirect sunlight refracted by earth's atmosphere. The dark sky at mid-totality of solar eclipses allows observing even faint stars.


Conus Mound, Marietta, ohio
 

Solar Eclipses

From the earth, the moon and the sun coincidentally have nearly the same angular width. Not every solar eclipse is total due to lunar elliptical orbit; moon perigee (mp) is seven-eights of moon apogee (ma). When a distant, smaller moon does not completely obscure the sun, instead of totality an annulus of sunlight surrounds the moon. The inner, umbral shadow blocks all sunlight, the penumbral shadow only partially blocks sunlight. Annular eclipses are seen in the antumbral shadow, when the umbral shadow does not extend to earth's surface. One-third of solar eclipses are annular, even more are only partial. The moon's umbra, the dark inner shadow, misses the earth during only-partial eclipses. Rarely, a total eclipse changes to an annular eclipse or vice versa due to the curvature of the earth. Observers in the penumbral shadow experience total and annular eclipses as partial eclipses.

eclipses of the sun

Lunar eclipses are far easier to predict. Knowing the date of an eclipse facilitates prediction of others. Because solar eclipses inscribe a path of motion on the earth, understanding lunar orbit geometry and speed is required for prediction. Table 8 presents the modern cosmographical data used in calculating eclipse geometry.

Table 8. Cosmographical Data
code
term
value
ma
moon, distance, apogee, meters
406,720,000.0
mp
moon, distance, perigee, meters
356,375,000.0
au
sun, distance, meters
149,597,870,000.0
sun
sun, diameter, meters
1,391,980,000.0
mrm
moon radius, meters
1,737,970.0
rw
earth, semi-major axis, meters, IUGG, WGS84
6,378,137.0
smi
earth, semi-minor axis, meters, derived, WGS84
6,356,752.3141

Solar eclipses provide precise observations of the geometrical configuration of sun, moon and earth as well as the precise timing of new moon and the synodic period. Observers in the path of totality can make sidereal observations of lunar position. A disadvantage of solar eclipse observations is the inability to repeat observations from the same position each eclipse.


Newark Octagon
 

The Metonic Cycle

One short-duration eclipse interval coincides with lunar and earth orbit cycles at a 19-year interval. The intercalations of 19 solar orbits, 254 lunar orbits, 235 full moons and 19 years coincide with 255 lunar nodal months. Four or five Metonic eclipses in series occur on the same date. Table 9 compares these values for epoch 2,000.0, sorted by interval duration.

Table 9. The Metonic Cycle
  Period
Multiple
Degrees
Days
Rotations
Solar Orbit
19
6,840.0
6,939.871
6,958.871
Lunar Orbit
254
6,839.834
6,939.702
6,958.702
Lunar Synodic
235
6,839.820
6,939.688
6,958.688
Tropical Year
19
6,839.735
6,939.601
6,958.602
Lunar Nodal
255
6,839.256
6,939.116
6,958.114

The Metonic cycle accurately intercalates integer lunar sidereal orbits and lunar synodic periods; 254 lunar orbits equal 235 synodic periods. Using modern constants, the difference between these two intervals accumulates 0.0137 days difference each 19 years. The ratio of 235 synodic periods to 254 lunar orbits equals 1 to 1.000 002 in temporal durations.

The tally difference between lunar orbits and lunar synodic equals the number of solar orbits. The geometric principle x - 1 = y applies, with one less synodic period than lunar orbits per solar orbit. In 19 solar orbits there are 235.0062 moons and 254.0062 lunar orbits. In principle, there cannot be 19 more lunar orbits than moons until when 19 solar orbits complete. The geometric principle x - 1 = y also applies to the precession of lunar eclipses; lunar nodal period precessing lunar orbit adds another orbital factor. The precessing lunar node's retrograde orbit (6,793.52 days) is determinative of eclipse timing, and in combination with solar orbit, determinative of the eclipse nodal period (346.62 days). During 18.5993 solar orbits there are 19.5993 eclipse nodal periods. The x - 1 = y theorem sign reverses due to retrograde circularity, hence x + 1 = y.


Monks Mound at Cahokia, Illinois
 

The Saros Interval

Mesopotamian astronomers recorded the Saros cycle of 242 nodal periods and 223 synodic periods. The Saros cycle is also recognized in the Maya astronomical book, the Dresden Codex. Noted for its accuracy, the Saros nodal to synodic ratio is 1.0 to 1.000 005. Metonic eclipses do not repeat for as long a period because their nodal:synodic ratio intercalates less accurately. Table 10 presents the Saros interval data.

Saros eclipse series visibility shifts one third circumference, so a triple Saros—the Exeligmos Interval with a near integer number of days—presents eclipses visible from the same location. In 669 moons there are 19,755.96 days, almost precisely an integer number of days. Also, the intercalation of the anomalistic period with the Saros eclipse interval produces similar eclipses in each Saros sequence.

Table 10.  The Saros Eclipse Interval
Period
Multiple
Degrees
Days
Rotations
Lunar Synodic Period
223
6490.553
6585.321
6603.351
Lunar Nodal Period
242
 6490.588
6585.357
6603.387
Anomalistic Period
239
6490.766
6585.537
6603.567

 

Eclipses in Archaeological and Early Historical Contexts

This section is intended to convey a general idea of the current history of eclipse knowledge. Astronomical data concerned with the cosmic objects which move along the ecliptic, the sun, moon, and the visible planets, is known from prehistoric cultures. The most remote eclipse record is likely a Rg-Veda description of a solar eclipse observed by Atri about 3,928 B.C. Before inventing paper, the Chinese kept records on bones and shells. Li Shu wrote about astronomy around 2,650 B.C. and observatory buildings are known by 2,300 B.C. Inscribed bones and tortoise shells from the Shang dynasty reference solar eclipses. From the Chou dynasty and the Warring States period, over 40 solar eclipse observations are recorded from 720 BC onward, the earliest recorded series.

"... the Spring and Autumn Annals records as many as 36 eclipses of the Sun.  This series of observations, which commences with the event of Feb 22 in 720 BC, is the earliest from any part of the world ... prove to be in exact accord with those of eclipses listed in modern tables." Stephenson (1997:221-223)

In Anyang, five recorded solar eclipses between 1,161 BCE and 1,226 BCE are known. Based on a Chinese inscription, astronomers from NASA's Jet Propulsion Laboratory (JPL) fixed the exact date and path of a solar eclipse in the year 1,302 B.C. to determine delta T, a measure of the slowing of the earth’s rotation. They concluded the length of each day was 0.0047 seconds shorter in 1,302 B.C.

An early Mesopotamian record of a total solar eclipse at Ugarit is from May 3, 1,375 B.C. Clay tablets with astronomical observations survive from Mesopotamian civilizations. Babylonian astronomical records on tablets dating from 1,700 B.C. report the motions of Mercury, Venus, and the Moon. Later records include a total solar eclipse on July 31, 1,063 B.C. and the well-documented Nineveh eclipse of June 15, 763 BC, recorded by the Assyrians. Diodorus of Sicily suggests ziggurats were the observation platforms. Continuous usage and layered rebuilding of ziggurats dates to the dawn of civilization.

uruk, sumeria

The Babylonian civil calendar was regulated by a Metonic cycle. The oldest record of the 223 moon Saros interval is Mesopotamian. The historical astronomy of Mesopotamia evidences, around 400 B.C., the celestial zodiac of 12 signs of 30 degrees each, a lunisolar calendar, a lunar synodic (full moon cycle) value of 29.530592644 days (as a fraction), knowledge of the Metonic and Saros eclipse cycles, eclipse record keeping, and use of arithmetical progressions for accurate eclipse prediction.

The Mesopotamian achievements evidence a long history of observations and the most advanced astronomy documented for the time. A great number of Mesopotamian astronomical texts from the last three centuries B.C. include evidence of sexagesimal place value notation, eclipse records, and the use of zero.

"... the eclipse observations made by medieval Arab astronomers are among the most accurate and reliable data from the whole of the pre-telescopic period." Stephenson (1997:456)

Astronomy also flourished in the Middle East and India during Europe's Dark Age. The 500 A.D. Indian book on astronomy, The Àryabhatiya of Àryabhata, evidences accurate astronomical knowledge in South Asia, in particular the precise sidereal ratio of the readily-observed earth rotations and lunar orbits. Àryabhata provided numerical and geometrical methods for calculating eclipses.

Eclipse calculations are evidenced by the Anikythera mechanism, a sophisticated mechanism dating from the 2nd century B.C and discovered near Crete in 1,900 by sponge divers. The mechanism's dials include a Metonic cycle calendar and a Saros eclipse-prediction dial with prediction glyphs. A Greek text by Archimedes on astronomical mechanisms has not survived. Homer's The Odyssey reports the 1,178 B.C. solar eclipse, "... the Sun has perished out of heaven, and an evil mist hovers over all."

The May 28, 585 B.C. total eclipse predicted by Thales in Asia Minor is the beginning of the Greek Classical Period. Greek historical documents evidence the role of Egyptian and Mesopotamian astronomy in Greek science. Ptolemy's Almagest included Babylonian observations of six solar eclipses, the earliest eclipse dating to 721 B.C. The Almagest also includes Egyptian sidereal observations of the moon and employs a Callippic calendar to date them.

Greek astronomer Meton and his associate Euctemon instituted a 19-year-eclipse calendar in Athens on the summer solstice of 432 B.C. Callippus' calendar cycle began astronomically in 330 B.C. on a start date when summer solstice and the lunar month beginning nearly coincided. In 330 B.C., new moon on June 28 was at 1:44 A.M. U.T. while solstice was close to midnight, less than two hours earlier. Geminus reports Callippus corrected the 19-year cycle, deducting one day every four Metonic periods. Apparently, Callippus knew of the one day difference between days per orbit and days per year every 4 Metonic cycles. This infers Callippus observed the Metonic cycle against sidereal reference. Aristarchus (Aristarchos) of Samos proposed a heliocentric solar system in 297 B.C. In the second century B.C. Hipparchos understood the elliptic orbit of the moon and precession of the equator.

"... the god visits the island every nineteen years, the period in which the return of the stars to the same place in the heavens is accomplished." Diodorus Siculus "the Greeks [who] use the nineteen-year cycle ...
are not cheated of the truth." Diodorus Siculus

76 * 365.25636 days per orbit = 27,759.48 days
76 * 365.24248 days per year = 27,758.43 days

Greek records indicate an Egyptian practice of sidereal astronomy. Diodorus noted ancient Egyptian astronomers predicted solar eclipses. Thales, who brought Egyptian land surveying rules to the Greeks—the basis of Euclidian geometry, predicted the total solar eclipse on May 28, 585 B.C. According to historical documents, the Greeks had not predicted a solar eclipse prior to Thales predicting the eclipse near Miletus in Asia Minor. Thales studied mathematical and astronomical knowledge in Egypt before predicting the Miletus eclipse. Unfortunately, burning the Great Library in Alexandria destroyed untold Egyptian history.

"Timocharis, who observed at Alexandria, records that in year 36 of the first period according to Callippus, on Elaphebolion 15, which is Tybi 5, as the third hour was beginning, the moon overtook Spica with the middle of the part of its rim that points towards the equinoctial rising, and Spica traversed it, cutting off exactly one third of its diameter on the north side."

The libraries in the Americas were also burned. Surviving Maya writing provides the best evidence of astronomy in pre-Hispanic America. Mayan symbols of the sun and the moon are unequivocally known. The 405 lunation eclipse cycle of 11,960 days is known from the Dresden Codex (810 nodal periods = 879 synodic periods or 1.0 : 1.000 010). From Copan Stelae 3 and A, the Metonic eclipse cycle is reported as 6,940 days, equaling 19 solar years and 235 moons. A sequence of bar and dot numerators in the Fejérváry Codex, pages 15 to 22, totaling 6,940 days, one Metonic cycle, was the first notice of this cycle in iconographic texts. The Metonic cycle also appears in the histories of Tilantongo and the Mixteca.

Knowledge of the eclipse cycle is evidenced in the House of the Sun at Palenque, dedicated in A.D. 692. The Palenque ratio is 81 moons to 2,392 days. Maya mathematics employed large integers instead of decimal notation. The 81 to 2392 ratio expresses a value of 29.530864 days per lunar cycle, 0.00028 days less than the present cosmic period. Longer Maya astronomical count intervals accurately equating solar and lunar orbits indicate early astronomical understandings. Evidence of lunar dates begin at A.D. 357 in an inscription from Uaxactun, bearing a Long Count date of 8.16.0.0.0. A vase from Uaxactun places the origin of a lunar calendar prior to A.D. 42.

In the Dresden Codex the Saros cycle was recognized. The most important codex is the Dresden with its eclipse series table. Eclipse prediction infers a certain level of knowledge, including the length of the lunar month, the interval between lunar nodes and the ecliptic limit for solar eclipses. Evidence of eclipse prediction infers this knowledge existed. About 200 lunar observations are known in Maya inscriptions.

dresden codex


Intihuatana at Machu Picchu
 

Research Results

2009.12.06 - While examining the approximate 345-year eclipse interval known to the Ancient Greeks and Babylonians, one equating 4,267 synodic lunar periods with 4,573 anomalistic periods, I noticed the interval also has accurate integer equation to days and sidereal rotations. Using Epoch v2009, I readily noticed the near solar orbit integer equates 126,352 rotations to 126,007 days, 345 fewer or one day less per orbit. The accurate integer equation of lunar synodic with the apogee-perigee period ensures greater interval constancy of eclipse observations. Each lunar orbit eccentricity period imparts a correlated orbital motion speed change. Comparing eclipse intervals having whole multiples of the apogee-perigee cycle ameliorates the impact of this orbit speed inconstancy. Ancient astronomers utilized this knowledge to determine the appropriate eclipse spans for more accurate mean full moon period determinations. Historically, Hipparchos compared eclipses with Babylonian records from 345 years earlier.

In Table 11, the 345-year eclipse interval compares number of days in integer multiples of six motion periods using 432 B.C. astronomical values (Terrestrial Time). Two accurate integer ratios are with earth sidereal rotations. The most accurate integers interval ratio is moons to days. Use of an eclipse interval which equates to accurate integer rotations and days infers earth motion and time of day and rotation had a role in the eclipse observation interval, likely augmenting the accuracy capability. Accuracy makes this interval a favorable observation choice for those who understand lunar elliptical geometry.

Table 11.  The 345-Year Eclipse Interval
Period
Code
Value
Multiple
Days TT
Lunar Nodal Period
dn
27.2122108
4,630.5
126,006.142
Lunar Synodic Period
dm
29.53058247
4,267
126,006.9954
Days
d
1.0
126,007
126,007.0
Anomalistic Period
da
27.5545727
4,573
126,007.061
Rotations
dr
0.99726967
126,352
126,007.0176
Years
dy
365.2423351
345
126,008.6056

126,007 days : 4,267 lunar synodic = 1.0 : 1.000 000 036
126,352 rotations : 126,007 days = 1.0 : 1.000 000 140
126,352 rotations : 4,267 lunar synodic = 1.0 : 1.000 000 176
4,573 anomalistic : 4,267 lunar synodic = 1.0 : 1.000 000 520

2012.07.25 - Some refinements and corrections to the temporal terms of my astronomy 'constants' formulas has altered the deep decimal expressions and ratio accuracies on this page. Formula details are available in Epoch Calc applet. Conversion to Universal Time will alter the day values.


Astronomy PowerPoint
Watching Eclipses, Counting Orbits
PowerPoint
 

References, Links, and Readings

More References and Literature: Archaeoastronomy Page


 


Some Quotations

" Tarrutius ... pronounced that Romulus was conceived in his mother's womb the first year of the second Olympiad, the twenty-third day of the month the Aegyptians call Choeac, and the third hour after sunset, at which time there was a total eclipse of the sun; that he was born the twenty-first day of the month Thoth, about sunrising; and that the first stone of Rome was laid by him the ninth day of the month Pharmuthi, between the second and third hour. For the fortunes of cities as well as of men, they think, have their certain periods of time prefixed, which may be collected and foreknown from the position of the stars at their first foundation." Plutarch, Romulus "Sulpicius Gallus ... was the first Roman to make public the explanation of each eclipse when, on the day before King Perses was defeated by Paulus, he was brought before the assembly of troops by the commander-in-chief in order to explain an eclipse and freed the army from anxiety, and a little later when he wrote a book. Among the Greeks, Thales of Miletus, who explained the eclipse of the Sun which occured in the 4th year of the 48th Olympiad when Alyattes was king, that is, in the 170th year from the founding of Rome, was the very first to make inquiry about eclipses. After them, Hipparchus proclaimed the daily progress of each star for 600 years, who understood the months and days of the nations, the longest daytimes and geographical locations of places ..." Pliny, Natural History

"... (information) from Anaxagoras, ... this phenomenon occurs at fixed periods and by inevitable law, whenever the moon passes entirely beneath the orb of the sun, and that therefore, though it does not happen at every new moon, it cannot happen except at certain periods of the new moon. ... it was a strange and unfamiliar idea that the sun was regularly eclipsed by the interposition of the moon - a fact which Thales of Miletus is said to have been the first to observe. But late even our own Ennius was not ignorant of it, for he wrote that, in about the three hundred and fiftieth year after Rome was founded: In the month of June - the day was then the fifth - The moon and night obscured the shininq sun. And now so much exact knowledge in regard to this matter has been gained that, by the use of the date recorded by Ennius and in the Great Annals, the dates of previous eclipses of the sun have been reckoned, all the way back to that which occured on July fifth in the reign of Romulus. For even though, during the darkness of that eclipse, Nature carried Romulus away to man's inevitable end, yet the story is that it was his merit that caused his translation to heaven." Cicero, De Republica


 
 
FOR MORE ASTRONOMIC AND GEODETIC DATA CHECK OUT:

Astronomical Constants Index - - Astronomy Formulas Pages - - Geodesy Page
Did ancient astronomers survey the Earth? - - Read the Archaeogeodesy Pages


WORLD WIDE WEB HUBS BY THE AUTHOR:
Home  |  Anthropology  |  Archaeoastronomy  |  Photo Stock  |  Web Design  |  Art

©2011 by James Q. Jacobs.  All rights reserved.
Your comments, etc. are appreciated: Contact.
Published July 13, 2009. Updated Jan. 13, 2011.
Cite as: http://www.jqjacobs.net/astro/eclipse.html