Arabic Eclipse Records Bring Light to Scientific Analysis of the Earth's Rotation


A total eclipse of the Moon occurs during the night of Wednesday, February 20/21, 2008. The entire event is visible from large parts of our globe. On this occasion, we attract the attention of our readers to the historic contribution of the astronomers that worked in the Muslim lands in the study of such spectacular astronomical events. By analysing the eclipse records let by Arab and Muslim scholars of the past, the scientists today have drawn from them ground breaking results that help them in determining secular changes in the Earth's rotation.

+ Click to read the full article
- Click to close
image alt text  
Figure 1: Eclipse diagram for Greenwich Mean Time, February 21, 2008 (source).  

A total eclipse of the Moon occurs during the night of Wednesday, February 20/21, 2008. The entire event is visible from South America and most of North America as well as Western Europe, Africa, and Western Asia.

From start to finish, the eclipse lasts about three hours and twenty-six minutes. The partial eclipse begins at the Moon's eastern edge, and slowly moves into the Earth's umbral shadow [1]. The colour and brightness of the totally eclipsed Moon can vary considerably from one eclipse to another. Dark eclipses are caused by volcanic gas and dust which filters and blocks much of the Sun's light from reaching the Moon. But since no major volcanic eruptions have taken place recently, the Moon will probably take on a vivid red or orange colour during the total phase. After the total phase ends, it is once again followed by a partial eclipse as the Moon gradually leaves the umbral shadow.

Major sources for the eclipse observations in Islamic astronomy [2]

Ancient observations of solar eclipses from many different cultures and civilizations date back to at least 2500 BCE in the writings that have survived from ancient China and Babylon. Islamic astronomy became the western world's powerhouse of scientific research during the 9th and 10th centuries CE, and stood on the most advanced scientific tradition of its time. Significant contributions were made by scientists such as Al-Khwarizmi (Baghdad, ca 825 CE), the father of algebra, who developed also the first tables of trigonometric functions which remained the standard reference well into the modern era. Al-Khwarazmi's calculations were good to five places, allowing for unprecedented precision in astronomy and other sciences. To the North East of Iraq, in Syria, Muhammad ibn Jabir al-Harrani al-Battani (Albatenius) (b. before 858; d. 929 CE), a Sabian scholar whose work enjoyed a wide diffusion in Muslim lands, began with Ptolemy's works and recalculated the precession of the equinoxes, producing new, more precise astronomical tables. Al-Battani was undoubtedly one of the greatest Islamic medieval astronomers. He made notable contributions in different fields of astronomy. He made precise predictions of eclipses, and worked on the phenomenon of parallax. His greatest work in astronomy is al-Zij as-Sabi', which, due to the monumental work done by Carlo A. Nallino, is the best edited and commented Arabic astronomica table so far [3]. This important text survived in only one Arabic manuscript copy preserved in the Escorial library in Spain. According to Nallino, it was written about 1100 CE. Al-Battani has recorded in his zij two solar and two lunar eclipse observations. These are probably the ones that helped Halley around the year 1693 to discover the tidal acceleration of the Moon.

In Arabic astronomy, simultaneous celestial events – such as the eclipses of the Moon – were used to deduce the difference between local times in different locations and, hence, to build a system of geographical coordinates. A famous example of this simultaneous observation is given by the collaboration between Al-Biruni and Abu 'l-Wafa al-Buzjani. We know indeed that al-Biruni observed at his native town Kath, in Khwarezm, the eclipse of the moon on 24 May 997 CE. The eclipse was an event that was also visible in Baghdad and al-Biruni had arranged with Abu 'l-Wafa to observe it there. Comparing their timings enabled them to calculate the difference in longitude between the two cities [4].

image alt text  
Figure 2: An image of an Islamic observatory, showing an astrolabe in use (source).  

In the 10th century, Ibn Yunus (who died in 1008 CE) observed more than 10,000 entries for the sun's position for many years using a large astrolabe with a diameter of nearly 1.4 metres. His observations on eclipses were still used centuries later in Simon Newcomb's investigations on the motion of the moon, while his other observations inspired Laplace's Obliquity of the Ecliptic and Inequalities of Jupiter and Saturn's. Working in Cairo and following a steady series of advances in Islamic trigonometry, Ibn Yunus preserved in his al-Zij al-Kabir al-Hakimi precious records of lunar and solar eclipses. Albeit its special status in the history of astronomy, this important text has not yet been edited in a critical edition and a serious commentary. It exists in a fragmentary copy in Leiden University Library (MS 1957, Cod. Or. 143). The observations preserved in it, among them the most important compilation of solar and lunar eclipses of the epoch, were used successfully more than two centuries ago to determine the secular tidal acceleration of the Moon. Caussin de Perçeval published in 1804 the Arabic text of these first chapters together with a French translation [5]. The printed Arabic text represents the-only serious preliminary study, but his French translation is not precise enough and sometimes misleading. Although this publication is by no means a critical edition, it was taken by several modem astronomical and geophysical studies in the 19th and 20th centuries.

Records of Sun and Moon eclipses in some Arabic historical sources

Arabic historical sources are rich with eclipse records of the Sun and Moon that are given in Hijri dates. Besides the accounts that we mentioned in an earlier article [6], we selected hereafter some of them on the basis of the available literature. It is noteworthy to point out the variety of the sources in which astronomical events are narrated in Muslim heritage, from scientific text books to popular chronicles and general histories. This interest in natural phenomena, especially when they occur on a large scale in the heavens is a good indication of the wide spread of scientific culture in the culture of Islamic lands. Nowadays, this provides a good means to date precisely the events of general history linked to these astronomical events [7].

The following sources are taken as a sample:

IK = Ibn al-Kathir, al-Bidaya Wa al-Nihaya [8],

IA = Ibn al-Athir, al-Kamil Fi al-Tarikh [9],

Z = Al-Zahabi, Tarikh al-Islam [10],

IJ = Ibn al-Jawzi, al-Muntazam [11].

NM = Not mentioned in the four previous Arabic sources.

Source Event Year in CE (S)un / (M)oon Year AH
NM The sun was eclipsed. 733 S 114
NM This year was the moon eclipsed, between cock-crowing and dawn, on the fifth day before the calends of April. 05/04/795 M 10/01/179
NM This year was the moon eclipsed, at eight in the evening, on the seventh day before the calends of February. 07/02/800 M 8/01/184
NM This year was the moon eclipsed, at dawn, on the thirteenth day before the calends of January. 13/01/802 M 05/01/186
NM This year the moon was eclipsed on the first of September. 01/09/806 M 14/10/190
NM This year the moon was eclipsed on the first of September. 01/09/806 M 14/10/190
NM This year was the sun eclipsed, precisely at eleven in the morning, on the seventh day before the calends of August. 07/08/807 S 29/09/191
IA Z IJ In this year in two nights before the end of Dhu al-Hijja the sun was eclipsed so that two thirds of it were covered and its light was dwarfed. 25/06/819 S 203
NM This year was the moon eclipsed, on mid-winter's mass-night. 20/12/827 M 27/09/212
IA This year in Jumada I, the sun was totally eclipsed and the stars appeared and the world got dark and the birds settled. May 867 (S)
14 December 867 (M)
S & M 253
IA In this year the moon was eclipsed completely. 867-868 M 254
IK On the night of the fourteenth of Shawwal most of the moon was eclipsed. 3/09/871 M 257
NM The same year also the sun was eclipsed one hour of the day. 879 S 265-266
Z IA In Muharram of this year the sun and the moon were eclipsed. July 882 S & M 269
IA In Shawwal the moon was eclipsed and the people of Dabil woke up and the world was still dark. 12/893- 01/894 M 280
NM This year the moon was eclipsed. 904 M 291-292
IA This year the moon was totally eclipsed on Friday the 14 left of Rabi' I. It was eclipsed totally again on 14 left of Shawwal. 12 February 936
5 September 936
2M 324
IA This year the moon was totally eclipsed. 959-960 M 348
IA In the middle of Rabi' I, the moon was totally eclipsed. 23 April 962 M 351
IA This year on Saturday the 14 of Safar the moon was totally eclipsed. 19 February 965 M 354
IA The moon was totally eclipsed on Saturday 13 Sha'ban and it set eclipsed. 24 July 966 M 355
IA The night of 14 Rajab the moon was totally eclipsed and it set eclipsed. 2 June 969 M 358
IA IJ In Jumada I of this year the sun was eclipsed completely and the stars appeared and the world was dark and the bird fell. 05/1061 S 453
IA In this year the sun was eclipsed completely. 1087-1088 S 480
IK IA On the fourth of Safar the moon was eclipsed completely. 6/06/1117 M 511
IA On Friday the twenty ninth of Ramadan in the middle of the morning, the sun was eclipsed completely and the world became dark like it was night and the stars appeared. 11/04/1176 S 571
IA On the night of the middle of Rabi II… the moon was eclipsed around the middle of the night and it set eclipsed. 29/09/1178 M M & S 574
IA ... In this year also on the twenty-ninth the sun was eclipsed at evening time and it set eclipsed. 13/10/1178 S S 574
IA The moon was totally eclipsed on the Tuesday 15 Safar. 15 June 1215 M 612
IA This year the moon was eclipsed twice. The first of these was on the night of 14 Safar. 13 February 1226 2M 623
IK ... On the morning of Sunday the twenty-eighth of the month, the sun was eclipsed before noon. 09/11/1341 S 742

A new scientific analysis of Islamic eclipse records

Beyond the traditional work of collecting historical data, a group of scientists examined recently the records of eclipses extracted from Arabic texts of astronomy and drew from them exciting results that shed new light on the secular changes in the Earth's rotation. This trend of work was led by F. Richard Stephenson from Durham University [12]. In order to decipher Arabic texts, Stephenson collaborated with S. S. Said, from the College of Science, King Saud University in Riyadh. A recent significant contribution to this research was made by a German scholar, W. Dalmau from the University of Tubingen (Institute of Astronomy and Astrophysics), who collaborated with a professional historian of Islamic science, Mathias Schram.

image alt text  
Figure 3: Astronomical table for the Sun, the Moon and the planets by Ibn al-Shatir, preserved in Paris, Bibliothèque Nationale, MS Arabe 2522, folio 66vo (Source).  

We summarize below the main results of these fascinating researches which personify how the history of science may be a testbed for contemporary science, and in the same time provide a concrete basis to the general slogan of "bringing life to Muslim heritage", by showing that this slogan is adapted not only for the general learned public but also for the scientific community as well.

As expressed by W. Dalmau, the thesis for this new research program is as follows. The use of some Arabic medieval lunar eclipse records [13] for the determination of secular changes in the Earth's rotation opens a critical area of research in which several scientists are engaged. The published results derived from these data suggest a non-uniform decrease in the Earth's rotation rate over the last millennium. There is, however, up to this day no sound physical explanation for the deduced "non-tidal oscillations", with an apparent period of about 1500 years, which overlaid to a constant secular tidal change in the Earth's rotation rate and produce a non-uniform deceleration of the Earth's rotation. Relying on the appropriate analysis of a set of observations which were executed by professional Arabic astronomers, the scientists show the way in which the non-uniform deceleration of the Earth's rotation was constructed. A correct reading of the Arabic medieval observations shows that they do not contradict a secular constant decrease in the Earth's rotation rate of nearly -4.6×10-22 rad s-2. This value is in accordance with other similar ones derived from ancient eclipse records and from satellite tracking data [14].

Relying on the numerous solar and lunar eclipse timings measured by Arab astronomers between about 830 and 1020 CE, Stephenson and his colleagues scrutinized in particular eclipse records exposed by Ibn Yunus in al-Zij al-Kabir al-Hakimi.

Remarking that the times of the various eclipse phases were usually obtained indirectly: by determining the altitude of the Sun, Moon or a selected bright star – usually to the nearest degree – with an astrolabe and afterwards reducing the measurement to local time, they found that these observations were remarkably accurate. As an example attesting the precision of these observations, the following timed observation was recorded by Ibn Yunus at Cairo on a date equivalent to November 6-7 979 CE:

"This lunar eclipse was in the month of Rabi ‘al-Akhir in the year 369 [H], on the night when the morning was Friday the 13th of the month… A group of scholars gathered to observe this eclipse… The altitude of the Moon when they perceived the eclipse was 64 1/2 deg in the east. The altitude when its clearance completed was 65 deg in the west."

Analysing these data, Stephenson concluded that it may be calculated that at Cairo the Moon would reach an altitude of 64.5° in the east at a UT of 20.08 h; this was the observed time of first contact. Similarly, fourth contact was observed at a UT of 23.30 h. Comparing these times with the computed values, he inferred that these two results are in adequate accord [15].

The data from solar and lunar eclipse observations obtained by Islamic astronomers during the period 829-1019 CE were compiled in tables and analyzed to estimate non-tidal variations in the earth rotation rate. Describing the data set and computational procedures, the results were presented in graphs. It turns out that the average length of the day in 950 CE is found to be 11.6 + or - 0.6 msec shorter than the present standard value (86,400 sec), consistent with an increase of 1.37 + or - 0.07 msec/century over the intervening period. The latter value is less than that predicted for tidal friction alone or that estimated on the basis of ancient Babylonian records [16]. On the basis of the above, significant long-term non-tidal variations are inferred [17].

image alt text  
Figure 4: Astronomical calculations in the Arabic translation of Ptolemy's Almagest, translated by Ishâq b. Hunayn (830-910), revised by Thâbit b. Qurra (836-901). Copied in the Maghrib or al-Andalus in 1221. Paris, Bibliothèque Nationale, MS Arabe 2482 folio. 36v-37r (Source).  

Historical observations of eclipses have been the best tools for determining secular changes in the Earth's rotation rate, until today. Relying on the analysis of eclipse records produced by Arab astronomers, mainly al-Battani, al-Biruni and Ibn Yunus, F. R. Stephenson, L. V. Morrison, S. S. Said, R. R. Newton and W. Dulmau set to study the Earth's rotation rate. These scientists do not agree in their interpretation of the conclusions they infer from the analysis of the data. The line of demarcation between them is the following: while the British group led by Stephenson concludes that the Arabic records of eclipses provides the ground for a decrease in the rotation rate of the Earth, Dalmau rejects this conclusion and states: "In our opinion, the authors listed above have not evaluated the Arabic records with the sufficient care. They have taken the numerical data from the sources without considering the context. The statistical method used by them meets by no means the requirements of the subject. The theoretical models and astronomical tables used by the Arabic astronomers are essential for their work and their observations, but they have been completely neglected. For the present, it can be stated that the Arabic medieval observations do not contradict a constant secular change in the rate of the Earth's rotation, its numerical value being nearly -4.6×10-22 rad s-2 "[18].

Dalmau concludes his article by calling to historians of science to provide the documentary basis to decide in this debate. He considers indeed that the exact determination of the secular change in the Earth's rotation will depend on a critical edition with a serious commentary of the zij of Ibn Yunus, because it yields the best source of information on the relevant observations.


1) Publications on astronomy in

Rate this article: 
No votes yet
See full gallery