THE EARTHS PALAEOROTATION, POSTGLACIAL REBOUND AND LOWER MANTLE VISCOSITY FROM ANALYSIS OF ANCIENT CHINESE ECLIPSE RECORDS

Citation
Kd. Pang et al., THE EARTHS PALAEOROTATION, POSTGLACIAL REBOUND AND LOWER MANTLE VISCOSITY FROM ANALYSIS OF ANCIENT CHINESE ECLIPSE RECORDS, Pure and Applied Geophysics, 145(3-4), 1995, pp. 459-485
Citations number
82
Categorie Soggetti
Geochemitry & Geophysics
Journal title
ISSN journal
00334553
Volume
145
Issue
3-4
Year of publication
1995
Pages
459 - 485
Database
ISI
SICI code
0033-4553(1995)145:3-4<459:TEPPRA>2.0.ZU;2-X
Abstract
Of the forces changing the Earth's rotation, tidal braking and postgla cial rebound predominate at a timescale greater than or equal to 10(2) yr (HIDE and DICKEY, 1991; DICKEY, 1992). Analysis of ancient eclipse records has given values for the clock error BT and the time derivativ e of the Earth's dynamic oblateness J(2) for the past 3,300 yr. Since Delta T = AT - UT = ct(2), where AT is Atomic (cesium clock) Time, UT is Universal (Earth rotation) Time, and t is the number of centuries b efore 1800, the oldest data have the most weight. Sunrise and sunset e clipses are especially valuable, as they can be retrospectively timed. The Bamboo Annals, entombed in 299 B.C. and unearthed in A.D. 281, st ates that ''in the first year of King Yi of the Western Zhou dynasty. the day dawned twice at Zheng (34.5 degrees N, 109.8 degrees E).'' Kai yuan zhanjing (Siddhartha, A.D. 724) cites this passage and adds that ''in the 2nd (actually 12th) year of Sheng Ping reign period of King S hang (actually King Xi) the day began twice at Zheng.'' Matching these records with the April 21, 899 B.C. and April 4, A.D. 368 sunrise ecl ipses (Oppolzer eclipse Nos. 732 and 3747) gave Delta T values of 5.8 +/- 0.15 and 1.7 +/- 0.1 hr, respectively. The recurrence of a central solar eclipse at the same site under almost identical circumstances a ccurately links up an ancient Delta T value with a more precise mediev al one, and makes the statistics of such early data more robust. The b rightness changes for the magnitudes 0.95-0.97 and 0.991-0.998 eclipse s were greater than those for the January 4, 1992 ''double sunset'' ov er Southern California, U.S.A. (magnitude 0.91-0.92). David H. Levy no ted that ''...as annularity ended. Sunset had come and gone, but the s ky began to brighten not darken. For almost 15 minutes it continued to brighten until the onrushing shadow of Earth took over and darkness f ell again...'' (Sky Telesc. 83, 694). We have analyzed even earlier re cords from the Shang dynasty. Six solar eclipse records have been iden tified among 160,000 oracle bones unearthed from the Shang dynasty cap ital Anyang (36.1 degrees N, 114.3 degrees E). Four of the 12th-centur y-B.C. inscriptions have cyclic days of 18, 42, 17 and 25. The Chinese 60-day cycle is like our week in design, and has been in continuous u se from time immemorial. These records have been uniquely matched to t he sunrise eclipses of June 7, 1172 B.C. and October 31, 1161 B.C., an d sunset eclipses of October 21, 1198 B.C. and June 27, 1163 B.C., res pectively. Using visibility constraints on the rising and setting ecli psed Sun from Anyang we have derived upper or lower limits on Delta T. Three of them duster around 7 hr 10 min, consistent with a Delta T of 7 hr 20 min, from the analysis of a record of the June 5, 1302 B.C. t otal solar eclipse, which states that ''three flames ate the Sun, big stars were seen.'' Analysis of our data gave an equation of best-fit o f Delta T=(30 +/- 2.5)t(2), for the secular lunar acceleration rate n( moon) = -26 ''/cen(2) (WILLIAMS ei al., 1992). From this we derived an omega/omega of -(19 +/- 1.6) x 10(-11)/yr, where omega is the angular velocity of the Earth's rotation. Subtracting a tidal omega/omega of -27.8 x 10(-11)/yr(LAMBECK, 1980) gave a nontidal omega/omega of (9 +/ - 1.6) x 10(-11)/yr, which is equivalent to a J(2) of -(4.5 +/- 0.8) x 10(-11)/yr. The average J(2) for the past 3,300 yr is larger than the present J(2) from satellite laser ranging, -3 x 10(-11)/yr (CHENG et al., 1989), as expected. Both J(2) values are consistent with postglac ial rebound from an upper mantle of viscosity 10(21) Pa s, and a lower mantle of viscosity (2-4) x 10(21) Pa s, deformed by Pleistocene ice sheet loading (PELTIER, 1985). Our mantle viscosity values are consist ent with those from the analyses of free air gravity anomalies and rel ative sea-level variations (MITROVICA and PELTIER, 1991, 1992). Accura te values of the mantle viscosity are critical to our understanding of thermal convection patterns, that are responsible for plate tectonics (PELTIER, 1986). Finally, the bounceback to its less oblate interglac ial shape makes the Earth spin faster, overcoming a third of the tidal braking by the Moon and Sun. The net effect has been lengthening the day by 1.64 +/- 0.14 msec/cen.