Cretaceous paleomagnetic apparent polar wander path for the Pacific plate calculated from Deep Sea Drilling Project and Ocean Drilling Program basalt cores

Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Cretaceous paleomagnetic apparent polar wander path for the Pacific plate calculated from Deep Sea Drilling Project and Ocean Drilling Program basalt cores William W. Sager Department of Oceanography, Texas A&M University, College Station, TX 77843-3146, United States Received 7 June 2004; received in revised form 26 September 2005; accepted 26 September 2005

Abstract The apparent polar wander path (APWP) of the Pacific plate still has many uncertainties owing to the fact that paleomagnetic data are difficult to obtain for oceanic plates. After more than three decades of coring by the Deep Sea Drilling Project (DSDP) and Ocean Drilling Program (ODP) there are now a large number of reliably dated basalt cores recovered from the Pacific plate and this provides an opportunity to determine paleomagnetic poles based on igneous rock samples, considered by many scientists to be the most reliable data type. Cretaceous Pacific plate basalt core data were compiled, corrected using a standard technique, divided into groups based on age, and combined to calculate five mean paleomagnetic poles with ages of 80, 92, 112, 121, and 123 Ma, the latter two being for two different coeval regions. In all pole analyses, the lack of azimuthal orientation for cored samples leads to large uncertainties in pole locations along a nearly east–west direction. This difficulty was mitigated by using declination data from magnetic anomaly inversions of dated Pacific seamounts for azimuth constraint. The two nearly same-age poles were calculated because paleocolatitudes from Ontong Java Plateau (OJP) are discordant compared to those from other Pacific locations. I interpret the discordant OJP results to indicate that the plateau is on crust that had an early history as an independent plate. The other poles (80, 92, 112, and 123 Ma) fall on a northeast-trending line that suggests slow apparent polar wander during the Early and mid-Cretaceous, followed by rapid polar wander between 92 and 80 Ma. Comparison of the 123 Ma pole with previously published paleomagnetic data of Jurassic age implies southward apparent polar wander followed by a turnaround. Because the 123 Ma pole is the farthest from the geographic pole, it implies the turnaround happened near that time and that the Pacific plate has moved ∼40◦ northward since then. The 80 Ma pole stands ∼17◦ from the geographic pole, indicating that ∼60% of the northward drift occurred prior to that time and ∼40% afterwards. © 2006 Elsevier B.V. All rights reserved. Keywords: Pacific plate; Paleomagnetism; Cretaceous; Mesozoic; Paleomagnetic pole; Basalt; Polar wander; Deep Sea Drilling Project; Ocean Drilling Program

1. Introduction Paleomagnetic APWP have far-reaching implications because they reflect plate motion and are therefore use-

E-mail address: [email protected]. 0031-9201/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.pepi.2005.09.014

ful for understanding tectonics, reassembling continents, and for back-tracking sites used for paleoclimate studies. In addition, they likely contain information about the long-term behavior of the geomagnetic field (e.g., non-dipole geomagnetic field components) and motion of the spin axis relative to the mantle (i.e., true polar wander, TPW) (e.g., Livermore et al., 1984; Besse and

330

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Courtillot, 2002). For most continental plates, a large number of individual paleomagnetic poles exist that can be averaged to produce an APWP, often using a sliding window in age with a width of 10–30 Myr. The result can be a detailed APWP that describes much of the tectonic history of that plate (e.g., Irving and Irving, 1982; Besse and Courtillot, 2002). The situation for oceanic plates is much different because it is difficult to obtain oriented samples from the ocean bottom. As a result, fully oriented sample measurements, the type of paleomagnetic data most often gathered from land, are rare. The Pacific plate is a notorious example of this problem. Despite being the largest plate on Earth, the number of fully oriented outcrop paleomagnetic data are few and restricted mostly to young islands. Instead, most Pacific paleomagnetic data are either derived from geophysical data, such as inversion of seamount magnetic anomalies (Richards et al., 1967) or asymmetry analysis of magnetic lineations (Schouten and McCamy, 1972; Cande, 1976), or they are measured from samples obtained in azimuthally unoriented cores (Cox and Gordon, 1984). As a result of both the scarcity of sample data and uncertainties about errors in some data types, the Pacific APWP is poorly defined and major trends and features of the Pacific APWP are still a source of debate (e.g., Sager and Koppers, 2000; Cottrell and Tarduno, 2000; Riisager et al., 2004). Because basalt flows are usually considered to be accurate recorders of geomagnetic field direction, basalt core data may be the most reliable of all oceanic data sets. The trouble with basalt core data has been their relative scarcity, inaccurate or absent radiometric dates, and the fact that measurements from many igneous units are necessary to average secular variation. During the DSDP, few holes drilled deeply into basalt, so the total number of basalt core paleomagnetic data was small. Nevertheless, Cox and Gordon (1984) summarized Pacific DSDP basalt core paleomagnetic data and calculated two poles, for Early and Late Cretaceous, consistent with other types of paleomagnetic data. They also predicted a hook shape for the Pacific APWP (Fig. 1). After 20 years of additional coring by the ODP, there are significantly more basalt core paleomagnetic data for the Pacific plate. In addition, radiometric dating techniques have greatly improved (e.g., Pringle, 1993; Koppers et al., 2003a,b), so that now many basalt cores have high-precision dates. This situation suggests that Pacific basalt core data can be re-examined profitably. In this study, I analyze Cretaceous Pacific DSDP and ODP basalt core data and calculate paleomagnetic poles for age groups with sufficient numbers of independent magnetic units. This study is focused on Cretaceous cores because few pre- and

post-Cretaceous data are available. The objective was to define the Cretaceous APWP with basalt data and to examine polar wander trends. 1.1. Prior results: Pacific APWP Published poles for the Pacific APWP give an outline of polar motion and provide several working hypotheses that may be examined with new data. The proposed hook shape of the APWP implies southward polar motion during the Late Jurassic and Early Cretaceous (Larson and Lowrie, 1975; Cox and Gordon, 1984; Larson and Sager, 1992), a turnaround and possible stillstand ∼35◦ from the spin axis (Tarduno and Sager, 1995), and subsequent northward motion during the Late Cretaceous and Cenozoic (Cox and Gordon, 1984; Sager and Pringle, 1988). One version of the APWP contains a rapid spurt of northward polar motion at ∼84 Ma (Sager and Koppers, 2000); although, some researchers doubt this event because of uncertainty about the reliability of seamount magnetic anomaly inversion data on which it is based (Cottrell and Tarduno, 2000). One independent, global composite APWP based on continental rock data suggests an earlier rapid shift may have occurred at ∼110 Ma owing to TPW (Pr´evot et al., 2000) and because this is a global event, should be reflected in the Pacific APWP. Poles for defining Cretaceous Pacific APWP are mainly derived from core and remote geophysical data, each with certain limitations and uncertainties. Many of the paleomagnetic poles used to define the APWP are from inversions of magnetic anomalies over seamounts (Richards et al., 1967; Sager and Pringle, 1988; Sager and Koppers, 2000). Two critical assumptions of this method are that a seamount has an overall homogeneous magnetization (or nearly so) and that the magnetic anomaly is caused solely by remanent magnetization. These assumptions may be a good approximation for some seamounts because many Cretaceous Pacific seamount anomalies are simple and consistent with an overall magnetic homogeneity, many seamount poles have consistent locations, and many agree with independent paleomagnetic data (Sager, 1987; Sager and Pringle, 1988). Nevertheless rock magnetic data indicate that induced and/or viscous magnetization may be significant in some seamounts (Gee et al., 1989) and some seamount paleomagnetic poles imply the same (Sager et al., 1993; Sager, 2003). As a result, some researchers question the validity of conclusions based on seamount anomaly data (Parker, 1991; Cottrell and Tarduno, 2000). Another geophysical technique that has been used to obtain Pacific paleomagnetic poles is the analysis of

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

331

Fig. 1. Pacific apparent polar wander path. (Top) mid- to Late Cretaceous poles. Filled circles represent poles calculated from magnetic lineation skewness (Acton and Gordon, 1991; Vasas et al., 1994; Petronotis and Gordon, 1999). Stars are poles determined from seamount paleomagnetic studies (Sager and Pringle, 1988; Sager and Koppers, 2000). Filled squares are composite poles from skewness, seamount, and core data (Petronotis et al., 1992; Sager, 2003). Open star shows pole from Chatham Island, near New Zealand (Grindley et al., 1977). (Bottom) Early Cretaceous to Jurassic poles, calculated from magnetic lineation skewness (Larson and Sager, 1992). Two different pole paths result from solutions computed with (left; filled diamonds) and without (right, open squares) anomalous skewness factor. A 125 Ma combined pole shown for reference. (Inset) Polar wander path of Cox and Gordon (1984). Gray area shows inferred pole path. Pole K is from DSDP basalt cores; pole S from seamount anomaly inversions; KT, from latest Cretaceous cores, M and 307 at either end of the pole path are colatitude arcs from cores from DSDP Site 307 and Midway Atoll. In all plots, thin ellipses and circles show 95% confidence regions.

332

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

magnetic lineation asymmetry or “skewness” (Schouten and McCamy, 1972; Cande, 1976). This method is based on the assumption that the phase shift needed to deskew anomalies (i.e., reduce them to the pole) is a measure of the remanent magnetization inclination projected into the plane perpendicular to the lineation. Although paleomagnetic declination is indeterminate, a pole location can be determined if results from two or more anomaly sets are combined (e.g., Petronotis et al., 1992). Some limitations are that no such data can be derived for the ∼35 Myr duration Cretaceous Quiet Period and that a factor, termed “anomalous skewness” is sometimes required to derive consistent results (e.g., Cande, 1978). The source of anomalous skewness is uncertain, but it is thought to be significant only for crust created at slow-spreading ridges (e.g., Dyment and ArkaniHamed, 1995). Assuming that skewness is the same for all anomalies of the same age from the same plate, it is possible to independently estimate and correct for anomalous skewness (Petronotis et al., 1992). Nevertheless, some Late Cretaceous skewness poles diverge from poles derived from other data (Fig. 1) and the reason is unclear. DSDP and ODP cores have had surprisingly little impact on determination of the Pacific APWP. This results from two factors. First, sedimentary paleomagnetic data are frequently dismissed because two studies have suggested sediment core paleomagnetic data are biased by inclination error caused by compaction (Gordon, 1990; Tarduno, 1990). Second, despite being more widely accepted, basalt core data have been too few to be used alone. Moreover, there are some questions about the fidelity of basalt core recording of paleoinclination owing to possible tectonic tilting of ocean crustal blocks (Schouten and Denham, 2000). Nevertheless, basalt sample data are considered to be accurate paleofield recorders, so it is likely that such data can be useful for estimating Pacific paleomagnetic poles. 2. Data and methods The primary data for this study are Cretaceous paleocolatitudes (i.e., distance from site to paleomagnetic pole) calculated with inclination data measured from azimuthally unoriented DSDP and ODP basalt cores on the Pacific plate (Table 1; Figs. 2 and 3). Only those data with good age control were used, that being either 40 Ar–39 Ar radiometric dates (except for recent cores from OJP where only osmium–rhenium data are available and are consistent with 40 Ar–39 Ar dates from elsewhere on the plateau) or estimates of crustal age based on mapped magnetic lineations. For example, data with

ages given by the K–Ar method or inferred from the age of sediments overlying basement were not used because both are considered too unreliable. I would have preferred to ignore also some of the data with older older 40 Ar–39 Ar dates, which are likely of lesser reliability (e.g., Koppers et al., 2003a,b), and paleomagnetic analyses done to outdated standards (e.g., less thorough demagnetization experiments), but such data do not exert great control on the calculated paleomagnetic poles because they generally comprise only a few magnetic units each (deep basement coring has been rare, particularly during the early days of ocean drilling). Each datum was treated as a single, mean colatitude for a given site, with the mean and its confidence limits calculated using the methods of Cox and Gordon (1984). This method corrects inclination data for the bias resulting from averaging data without azimuthal orientation and it determines confidence limits based on data scatter, a model of paleosecular variation, and a potential contribution from borehole tilt. Because few other studies have used this method, it is worthwhile summarizing the rationale for using it here. Basalt flows are emplaced at irregular time intervals, the result being that two successive flow units may or may not be separated by a significant time interval. When averaging basalt flows, data from adjacent flows should be combined when erupted close in time but left separate when widely separated in time. Unfortunately, the time interval between flows is rarely known. Many studies of basalt core data simply test adjacent flow mean inclinations for statistical distinctness (usually at the 95% confidence level) and treat those that are distinct as independent units. When within-flow measurement errors are small, this method may imply that flows whose means are not widely separated are independent, when they are likely not. The result is an over estimate of the number of independent units and an underestimate of the true paleocolatitude error. Some studies compare the observed scatter with models of secular variation, but the result has been that secular variation has been judged to be eliminated with as few as 10 units (e.g., Tarduno and Cottrell, 1997). In contrast, Kono’s (1980) benchmark study of cored lava flows from Suiko Guyot (Site 433) suggests that this number should be 20–30. Using the Cox and Gordon (1984) method, within-flow error is determined and combined with an estimate of error caused by secular variation and scaled by the number of independent units to derive error bound estimates that are more realistic. The result is that those colatitudes with few independent units have larger estimated uncertainties. The number of independent units is estimated from the pattern and differences among flow means. Where the

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

333

Table 1 Paleomagnetic colatitude data Name

Location ◦

Flows (n)

Units (N)

Colat

Std. Err.

Pol.

Age

Age type

Pmag ref.

Age ref.

Pole

◦

Lat ( N)

Lon ( E)

Radiometic dates DSDP 433c ODP 871 ODP 1203 ODP 884 ODP 873-877 DSDP 171 DSDP 169 ODP 803 DSDP 315a DSDP 170 ODP 843 DSDP 462 (ext) DSDP 462 (int) ODP 865 ODP 872 ODP 802 ODP 879 ODP 878 DSDP 289 ODP 807 ODP 1183 ODP 1185 ODP 1186 ODP 1187 ODP 878 ODP 800 ODP 866 ODP 1213 ODP 801

44.8 5.6 51.0 51.5 12.0 19.1 10.7 2.4 4.2 11.8 19.3 7.2 7.2 18.5 10.1 12.1 34.2 27.3 −0.5 3.6 −1.2 −0.4 −0.7 0.9 27.3 21.9 21.3 31.6 18.6

170.0 172.3 167.6 168.3 164.9 190.5 173.6 160.5 201.5 177.6 200.9 165.0 165.0 180.4 162.9 153.2 144.3 150.9 158.5 156.6 157.0 161.7 159.8 161.5 150.9 152.3 174.3 157.3 156.4

65 12 16 13 9 1 2 3 6 1 4 6 19 4 11 13 1 12 1 14 8 15 7 14 13 17 12 3 18

20 4 7 5 6 1 2 2 4 1 1 4 9 3 7 6 1 5 1 8 5 7 5 7 3 2 4 1 12

63.2 98.9 55.6 52.7 98.3 101.9 115.2 116.8 107.4 103.5 100.7 112.8 117.3 105.0 120.5 109.3 99.7 103.8 123.3 108.1 119.1 113.7 117.1 114.2 105.9 105.1 103.9 85.3 107.7

3.2 5.2 6.0 6.3 4.5 11.0 8.9 8.2 5.3 9.9 10.0 5.6 4.6 6.6 5.1 5.0 11.0 5.0 11.5 3.5 5.8 4.7 5.8 4.7 5.9 8.5 5.4 19.4 3.7

R N N R R N N N N N N N N N N N N R N N N N N N N M M R M

61.3 ± 0.5 68.3 ± 0.5 75.8 ± 0.6 80.4 ± 0.5 81.5 ± 0.2 88.1 ± 0.4 90 ± 4 90.2 ± 4.0 93.3 ± 2.7 97 ± 3 110 ± 2 .0 110 ± 3 110.8 ± 1 111.3 ± 1.3 113.9 ± 0.9 114.6 ± 3.2 118.4 ± 1.4 119.9 ± 0.8 121.7 ± 2.7 122.3 ± 1.0 122 ± 3 122 ± 3 122 ± 3 122 ± 3 124.0 ± 0.4 126.1 ± 0.6 127.6 ± 2.6 144.7 ± 0.4 167.4 ± 3.4

Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Os Os Os Os Ar Ar Ar Ar Ar

1, 2 3 4 5 3 6 6 6 7, 2 6 8 9, 2 9, 2 6 3 10 3 3 11, 2 12 13 13 13 13 3 6 14 15 16

22 23 24 25 26 27 28 29 27 28 30 31 32 26 26 32 26 26 29 29 33 33 33 33 26 32 26 34 35

– – 80 80 80 92 92 92 92 92 112 112 112 112 112 112 123 123 OJP OJP OJP OJP OJP OJP 123 123 123 – –

Inferred dates ODP 883 DSDP 163 DSDP 165a DSDP 581 DSDP 303a DSDP 304 DSDP 166 ODP 1179 DSDP 307 DSDP 167 DSDP 169

51.2 11.2 8.2 43.9 40.8 39.3 3.8 41.1 28.6 7.1 10.7

167.8 209.7 195.1 159.8 154.5 155.1 184.9 160.0 161.0 183.2 173.6

26 6 2 13 3 1 4 40 5 4 2

5 2 1 5 1 1 2 12 3 3 2

50.7 95.3 96.3 82.9 95.8 101.4 125.7 85.4 82.0 121.5 117.9

6.8 6.6 9.3 4.7 9.2 9.6 9.3 3.6 5.6 7.7 8.3

N R R M N R R R N R N

76 81 81 124 127 128.8 129.5 129.5 147.2 148 148

P, B L P L L L L L L L L

17 18, 2 19, 2 6 20, 2 20, 2 19, 2 21 20, 2 19, 2 6

17 36 37 38 39 39 39 38 38 38 38

80 80 80 123 123 123 123 123 – – –

Table heads: Flows (n) = number of lava flows or igneous units; Units (N) = number of independent magnetic units; Colat = mean, corrected colatitude; Std. Err. = standard error of colatitude; Pol. = polarity (N = normal, R = reversed, M = mixed); Age type = method of determining age (Ar = 40 Ar–39 Ar radiometric; Os = osmium–rhenium isochron; L = from magnetic lineation; P = from position and polarity; B = from biostratigraphy); Poles = used for calculation of pole of noted age; OJP = Ontong Java pole. References: (1) Kono (1980); (2) Cox and Gordon (1984); (3) Nakanishi and Gee (1995); (4) Tarduno et al. (2003); (5) Tarduno and Cottrell (1997); (6) Sager (2005); (7) Cockerham and Hall (1976); (8) Helsley (1993); (9) Steiner (1981); (10) Wallick and Steiner (1992a); (11) Hammond et al. (1975); (12) Mayer and Tarduno (1993); (13) Riisager et al. (2003); (14) Tarduno and Sager (1995); (15) Tominaga et al. (2005); (16) Wallick and Steiner (1992b); (17) Doubrovine and Tarduno (2004); (18) Marshall (1978); (19) Cockerham (1979); (20) Larson and Lowrie (1975); (21) Sager and Horner-Johnson (2005); (22) Sharp and Clague (2002); (23) Koppers (1998); (24) Duncan and Keller (2004); (25) Keller et al. (1995); (26) Pringle and Duncan (1995); (27) Schlanger et al. (1984); (28) Ozima et al. (1983); (29) Mahoney et al. (1993); (30) Waggoner (1993); (31) Ozima et al. (1981); (32) Pringle (1992); (33) Parkinson et al. (2003); (34) Mahoney et al. (2005); (35) Koppers et al. (2003a); (36) Eittreim et al. (1992); (37) Sager (2003); (38) Nakanishi et al. (1999); (39) Nakanishi et al. (1992).

334

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Fig. 2. Site locations of basalt core and seamount data analyzed in this study. Different core symbols denote age. “X” symbols show locations of seamounts used for declination data. Gray lines show representative magnetic lineations (labeled by anomaly number and age). Gray area shows Ontong Java Plateau.

difference between flow means is small (less than ∼10◦ ), means are considered correlated and averaged. Other evidence of time breaks is taken into account including changes and discontinuities in paleocolatitude versus depth trend as well as geologic indicators of time passage (such as soil or clay layers). The end result of this method is a paleocolatitude estimate that usually agrees closely (within a fraction of a degree) of that determined by other methods (e.g., McFadden and Reid, 1982) but with a slightly larger and therefore more conservative estimate of error bounds. The estimated error bounds are important when averaging paleocolatitudes because the error for each datum is used as a weight in determining the mean pole position and is used to estimate the confidence region of the mean pole (Gordon and Cox, 1980; Cox and Gordon, 1984). Age data were used to estimate mean pole ages using a weighted averaging scheme. The mean age and standard deviation of a paleomagnetic pole was calculated using a weighted average with the weighting factor being the importance of the datum in determining the mean pole position (Gordon and Cox, 1980; Appendix). This method takes into account the fact that paleocolatitudes have greater or lesser weight on the mean pole position depending on the estimated error bounds, with data

having smaller uncertainties generally having greater weight. Because there is some inconsistency and debate about the presentation of age errors (e.g., Koppers et al., 2003a,b), I simply used the reported ages for cores without any weighting for uncertainty of the age determination. When colatitudes were used to calculate mean paleomagnetic poles, it became evident that the lack of azimuthal control was a serious problem. Pacific colatitude arcs meet obliquely (Fig. 3), resulting in large confidence limits in a generally east–west direction. To obtain better azimuthal constraint, I added declinations from seamount magnetic anomaly inversions (and one declination estimated from basalt core sample overprint magnetization) to the mean pole calculations (Table 2). Although seamount anomaly data may be biased by overprinting, this effect is mostly in calculated paleoinclination and should be small for paleodeclination. This dichotomy results because the Pacific plate has not rotated greatly and thus overprints in the present field direction are nearly parallel to (or antiparallel) the original seamount magnetization (Sager, 2003). As a result, a present-day overprint (representing induced or viscous magnetization) typically changes calculated declination by only a few degrees or less (Sager, 2003).

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

335

Fig. 3. Colatitude arcs determined from Pacific plate basalt cores. Each arc is a segment of the small circle upon which a paleocolatitude constrains the paleomagnetic pole to lie. Heavier lines show data from N ≥ 5 independent units. Solid arcs show data from seamounts; dashed arcs, from crustal basalt; and long-dash arcs from intrusives.

Because a standard error for the declination data is needed for the mean pole calculations, I simply used the average observed standard deviation of declinations in different age groups as an approximation for all (7.2◦ ). In a previous work (Sager, 2003), I made overprint corrections for seamount declinations by assuming that the K¨oenigsberger ratio of a seamount could be determined assuming that the deviation of its pole from a mean pole position results solely from the overprint. Further experimentation suggested that the result of this calculation is highly dependent on the angle between the seamount magnetization and the ambient geomagnetic field direction and some combinations of vectors implied unrealistically large overprint magnetizations or declination corrections. Consequently, in this study I made no overprint corrections for seamount magnetization declination estimates.

3. Analysis and results Although the Mesozoic Pacific basalt paleocolatitude data set contains 180 independent magnetic units from 38 sites, the number of data in many intervals of time is too small for calculating a mean pole using only basalt paleocolatitudes (Figs. 3 and 4). Large numbers of paleocolatitudes and independent units are available only for the time span 110–130 Ma (Fig. 4), with a lesser cluster between 80 and 95 Ma. Because of this uneven distribution, use of a standard windowing routine gives too few data in most windows to allow determination of a reliable pole position unless the window is wide (i.e., 20 Myr or more). However, use of large width windows smooths out important details of the APWP. For this reason, I calculated mean paleomagnetic poles for largely independent clusters of data. Data are either too few for reliable pole calculation or absent in the 60–80, 95–110,

336

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Table 2 Declination and age data Name

ID

Location Lat

Chatauqua Show Kona 5S Hadyn Mendelssohn-E Mendelssohn-W Site 884 Wilde Mahler Makarov Miami Winterer Mij Lep Lo En Taikuyo-Daisan Taikuyo-Daini Daichi-Kashima MIT

CHA SHO KON HAY MENE MENW 884 WIL MAH MAK MIA WIN MIJ LOE DAIS DAIN DAIK MIT

(◦ N)

22.2 17.9 17.1 26.6 25.1 25.1 51.5 21.2 31.8 29.5 21.7 32.8 8.8 10.1 34.2 34.3 35.8 27.3

Lon

Dec.

Age (Ma)

Age type

Pmag ref.

Age ref.

Pole(s)

189.3 197.0 196.4 7.3 356.2 9.7 1.9 4.9 17.5 −6.5 −12.2 8.0 1.7 −9.8 −2.1 2.2 −12.0 183.3

79 81 81 75.1 ± 1.2 78.5 ± 0.9 82.4 ± 1.3 80.4 ± 0.5 90.6 ± 0.3 91.0 ± 0.7 93.9 ± 2.6 96.8 ± 1.2 108.3 ± 2.0 109.5 ± 1.0 113.9 ± 0.9 117.5 ± 1.1 118.1 ± 2.0 120.2 ± 2.6 120.1 ± 0.9

P P P Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar

1 1 2 3 2 2 4 5 3 5 6 5 7 8 5 5 9 5

10 11 11 12 12 12 13 14 12 14 14 14 14 14 14 14 15 14

80 80 80 80 80 80 80 92 92 92 92 112 112 112 112, 123, OJP 112, 123, OJP 123, OJP 123, OJP

(◦ E)

197.4 207.3 205.8 198.7 198.3 197.2 168.3 163.3 195.0 153.6 161.9 148.4 163.2 162.8 144.3 143.9 142.7 151.9

Table headings: Dec. = declination; Age type = method of determining age (P = inferred from polarity (Chron 33r); Ar = 40 Ar–39 Ar radiometric); Pmag, Age ref. = paleomagnetic and age references; Poles = used for calculation of pole of noted age; OJP = Ontong Java pole. References: (1) Sager (1992); (2) Francheteau et al. (1970); (3) Sager and Pringle (1987); (4) Cottrell and Tarduno (2003); (5) Sager et al. (1993); (6) Hildebrand and Parker (1987); (7) Sager (1983); (8) Bryan et al. (1993); (9) Ueda (1985); (10) Sager and Pringle (1988); (11) Gordon (1983); (12) Pringle (1993); (13) Keller et al. (1995); (14) Koppers et al. (2003b); (15) Takigami et al. (1989).

and 140–170 Ma intervals, so I did not attempt to calculate poles for those periods. Although there are six paleocolatitudes in the 140–170 Ma period, these data represent only a few independent units, with the exception of Site 801, which is probably affected by crustal tilt (Pockalny and Larson, 2002).

insignificantly different in location, 72.9◦ N, 352.6◦ E, but with a much larger 95% confidence ellipse major semiaxis of 18.0◦ . The mean age for the pole is 79.9 ± 2.8 Ma (1σ uncertainty).

3.1. 80 Ma pole

There are few data in the interval between the clusters around 75–82 and 110–130 Ma, but I calculated a mean pole for this period because it represents a large gap and because it may contain rapid apparent polar wander (Sager and Koppers, 2000). Paleocolatitudes from Sites 169, 170, 171, 315, and 803 intersect in a small area and have ages from 88 to 97 Ma. By themselves, these data give a pole with a large east–west uncertainty (30◦ major semi-axis), so I combined these data with seamount declinations from four seamounts (Wilde, Mahler, Makarov, and Miami) ranging in age from 91 to 97 Ma (Table 2). The result is a pole located at 59.7◦ N, 345.4◦ E (Fig. 5; Table 3) with an average age of 92.4 ± 2.3 Ma. This pole is in good agreement with a pole determined from oriented sediment core samples from Site 869, dated at 97 Ma (Fig. 6) (Sager et al., 1995). Although the error ellipse for the 92 Ma pole is large (major and minor semi-axes of 10.3◦ and 7.6◦ ), this pole

The 80 Ma mean pole is a revision of the Chron 33r pole determined previously (Sager, 2003), adding other nearly contemporaneous data with normal polarity. I used the same reversed polarity paleocolatitudes that represent Chron 33r (Sites 163, 165, 873–877, and 884) in addition to the two sediment core paleocolatitudes used in the prior calculations (Sites 462 and 869). To these I added two paleocolatitudes of slightly younger age (Sites 883, 1203), seamount declinations from five seamounts (Chatauqua, Show, Kona 5S, Mendelssohn-E, and Mendelssohn-W), and the declination estimate from Site 884 basalt core samples (Tables 1 and 2). The 80 Ma pole position was calculated to be at 73.2◦ N, 349.2◦ E with a 95% confidence ellipse having major and minor semi-axis lengths of 7.9◦ and 4.8◦ (Fig. 5; Table 3). A mean pole calculated without seamount declinations is

3.2. 92 Ma pole

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

337

Table 3 Mean paleomagnetic poles Age (Ma)

Pole type

Location

95% Confidence

Lat (N)

Lon (E)

Maj.

N Min.

Azim.

79.9 ± 2.8

A B C

72.2 72.9 73.2

347.3 352.6 349.2

8.0 18.0 7.9

5.3 6.4 4.8

117 88 106

33 27 67#

92.4 ± 2.3

A B

59.7 58.4

345.4 337.1

10.3 29.5

7.6 9.0

122 60

14 10

112.2 ± 3.6

A B

55.6 55.3

334.9 350.3

7.7 29.6

5.5 5.7

67 96

35 30

120.5 ± 1.8 (OJP)

A B

65.3 64.6

331.0 355.2

9.0 124.9

4.9 5.0

75 108

37 32

122.7 ± 4.4

A B

50.0 50.1

329.1 329.9

8.6 28.2

4.6 4.7

77 82

40 35

Table headings: Pole type: A = including seamount declinations; B = excluding seamount declinations; C = basalt, seamount declination, and sediment data. Maj., Min., Azim. = major semi-axis, minor semi-axis, and azimuth of major semi-axis of 95% confidence ellipse; N = number of independent magnetic units, with seamount declinations having a value of one. # Assumes N = 23 and 11 for Site 462 and 869 sediment data.

suggests a large change in pole position from the 80 Ma pole. Without the seamount declinations, the mean pole position is 58.4◦ N, 337.1◦ E (Table 2). 3.3. 112 Ma pole

Fig. 4. Histogram of paleomagnetic data vs. age, in 5 Myr bins. Black bars show the number of site colatitudes, whereas gray bars show the estimated number of independent magnetic units. Three gray bars are truncated and numbers atop these bars give the number of independent units. Small open circles show seamount declination data.

Data in the 110–120 Ma window fall mainly into two groups, most having dates between 110 and 115 Ma and two having dates that are indistinguishable from 120 Ma (Site 879 and the upper unit at Site 878). Consequently, I used the latter two paleocolatitudes in the calculation of the 123 Ma pole, described below. The remaining data in this period consist of six paleocolatitudes, representing 30 independent units (Sites 462 intrusives and extrusives, 802, 843, 865, and 872) and spanning the age range 110–115 Ma (Table 1). Using only basalt paleocolatitudes, the calculated mean pole location is 55.3◦ N, 350.3◦ E; however, the 95% confidence ellipse reflects a large uncertainty in the pole longitude and has a major axis length of 29.6◦ . When seamount declination data from five seamounts (Winterer, Mij Lep, Lo En, Takuyo-Daisan, Takuyo-Daini) are used, the pole is nearly the same, 55.6◦ N, 334.9◦ E, but has a much smaller confidence ellipse with a major and minor semi-axes of 7.7◦ and 5.5◦ (Fig. 7; Table 3). I used declinations from Takuyo-Daini and TakuyoDaisan seamounts for this pole calculation even though the ages are close to 120 Ma (Table 2) because the rate of declination change during this part of the Cretaceous is apparently small. The average age for this pole is 112.2 ± 3.6 Ma.

338

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Fig. 5. Mean paleomagnetic pole for 80 Ma (age range 75–82). Solid and dashed arcs show paleocolatitudes from basalt core data. Solid lines represent paleocolatitudes from seamounts; short dashed lines denote ocean crust; and long dashed line show intrusives. Gray, dash–dot lines show loci of the paleomagnetic pole constrained by seamount declination data. Numeric identifiers correspond to DSDP and ODP site numbers (see Table 1). Letter identifiers abbreviate a seamount name (Table 2). Ellipse shows 95% confidence region for mean pole, whoses location is shown by the open circle. Filled circle symbol shows the location of the pole determined without using seamount declination data.

3.4. 123 Ma pole and Ontong Java Plateau pole Data for the 120–130 Ma period are many in number (69 units from 16 sites, including the aforementioned 118.4 and 119.9 Ma data from Sites 878 and 879) and show a large amount of scatter (Figs. 3 and 8). If all of these paleocolatitudes are used to determine a mean pole, its location is at 25.2◦ N, 43.1◦ E. This location is not near any other Pacific paleomagnetic poles and appears to be the result of averaging data from a limited region with significant scatter. A trivial example of the geometric problem is two paleocolatitudes determined from the same site, but with different values owing to measurement uncertainties. The two colatitude circles would intersect ∼90◦ from the actual paleomagnetic pole because of spherical geometry. Examining the constituent data, it appears that much of the problem is a result of discordant paleocolatitutdes from OJP. After analyzing basalt core data from Site 807 on the plateau, Mayer and Tarduno (1993) concluded the section was tectonically tilted because the paleolatitude did not agree with expectations. Data from ODP Leg 192 yielded similar paleocolatitudes from four additional sites (Riisager et al., 2003, 2004), implying that Site 807 and other OJP sites are both consistent and unique. This

conclusion is strengthened by the fact that the non-OJP paleocolatitude data set contains results from oceanic crust, seamounts, and intrusives and these data are in reasonably good agreement (Fig. 8). Paleocolatitude arcs from OJP are clearly different. Because of this difference, I calculated two different mean poles, one for non-OJP data (Sites 166, 303, 304, 581, 800, 866, 878 (two units), 879, and 1179) and one for sites on the plateau (Sites 289, 807, 1183, 1185, 1186, and 1187) (Fig. 8). For both, I used the same four seamount declinations for azimuthal control (Takyo-Daisan, Takuyo-Daini, MIT, and DaichiKashima). This duplication was used because there are few paleodeclination data for this period and none for the plateau. In doing so, I have tacitly assumed that the plateau has not rotated relative to the surrounding Pacific plate, an assumption of uncertain validity. The result of the calculations is two distinct poles about 15◦ apart (Fig. 8; Table 3), with that for OJP located at 65.3◦ N, 331.0◦ E (95% confidence ellipse major and minor semi-axes of 9.0◦ and 4.9◦ ) and that for the rest of the Pacific plate located at 50.0◦ N, 329.1◦ E (95% confidence ellipse major and minor semi-axes of 8.6◦ and 4.6◦ ). Using no seamount paleodeclination data for these pole calculations gives poles at 64.6◦ N, 355.2◦ E

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

339

Fig. 6. Mean paleomagnetic pole for 92 Ma (age range 90–94 Ma). Conventions as in Fig. 5. Filled square shows pole determined from oriented sediment cores recovered from ODP Site 869 (Sager et al., 1995).

for the plateau and 50.1◦ N, 329.9◦ E for the rest of the Pacific. The uncertainty for the no-declination OJP pole is huge with a 124.9◦ major semi-axis for the 95% confidence ellipse resulting from the geographic proximity of the OJP drill sites. Mean ages for the two poles are 120.5 ± 1.8 Ma for OJP and 122.7 ± 4.4 Ma for the rest. The OJP average age is highly dependent on the assumption that all of the sites record contemporaneous volcanism over a very short time period, as suggested by Re–Os data (Parkinson et al., 2003). Given the estimated standard deviations, the two mean ages appear indistinguishable. 4. Discussion At the end of DSDP it was only possible to calculate Early and Late Cretaceous poles from drill core basalt samples because of the small numbers of independent units, sites, and precise dates. At the end of ODP, the situation has improved and five different Cretaceous poles can be calculated. Although it is likely that future research will refine or change some of the datum ages,

with the result that some data will be found to be in an incorrect age group, the mean poles are robust and change little with the removal of one to several data, especially the poles determined with larger numbers of independent data (i.e., 80, 112, 123, and OJP). The lack of azimuthal orientation still poses a serious problem that limits certainty in calculated pole locations in the direction parallel to the average trend of paleocolatitude arcs (approximately east–west). When numbers of paleocolatitude data are small and samples are from a limited geographical area, this problem seriously limits the possible resolution for pole longitude. A solution is to combine basalt paleocolatitude data with other data, but with the added risk of bias owing to errors inherent other data types. In this study, I used declinations derived from seamount magnetic anomaly inversions because these are almost the only data available from the Cretaceous Pacific plate that provide azimuthal control. Control calculations performed without the seamount declinations show, however, that the declinations have little effect on mean pole position, while they greatly decrease the major semi-axis of the confidence ellipses.

340

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Fig. 7. Mean paleomagnetic pole for 112 Ma (data range 110–115 Ma). Conventions as in Fig. 5. Filled square is 125 Ma pole of Petronotis et al. (1992), which is based mainly on the skewness of anomaly M0, augmented with seamount magnetic anomaly inversion poles and core paleocolatitudes. Heavier lines show paleocolatitudes determined from N ≥ 5 independent units.

4.1. APWP trends In this study, I calculated five different paleomagnetic poles, of which one is apparently discordant and may not apply to the Pacific plate as a whole. The four concordant poles, with ages of 80, 92, 112, and 123 Ma, agree with most other Pacific paleomagnetic data and show slow drift, covering ∼14◦ in 31 Myr (123–92 Ma) with a northeast trend (Fig. 9). Because of large uncertainties in pole locations, this rate could vary by a factor of ∼2.5. Indeed, owing to the significant overlap of confidence ellipses, polar motion could have even been small to nonexistent (i.e., a stillstand) as suggested by Tarduno and Sager (1995). Nevertheless, the APWP trend implies that the poles did indeed drift northeast, albeit at a slow rate. The slow rate of polar motion during this interval is inconsistent with the interpretation of paleomagnetic data from the continents that imply ∼20◦ of rapid polar motion around 110 Ma resulting from true polar wander

(TPW) (Pr´evot et al., 2000). If such an event occurred during this period, it should have been apparent in Pacific paleomagnetic data. Following the period of slow motion, however, there may have been a spurt of rapid polar motion. The 92 and 80 Ma poles are separated by 13.6 ± 11.3◦ . With an age difference of 12 ± 3.6 Ma, this implies a polar drift rate of ∼1◦ /Ma. Although not as high as the rate of rapid drift suggested by Sager and Koppers (2000), these data are largely independent confirmation (because the polar shift is mostly in latitude, controlled by the basalt core data) of a significant increase in polar drift speed. This analysis suggests that the rate increased by about a factor of two. Because the basalt data are few, pole and age uncertainties are large and the rate could be higher or lower, but the appearance of this feature in different data types implies that it is real. Motion prior to the Early to mid-Cretaceous period is uncertain because too few basalt data exist to calculate

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

341

Fig. 8. Mean paleomagnetic poles for 123 Ma (left; age range 118–130 Ma) and 121 Ma (right; age range 118–122 Ma) from cores recovered from Ontong Java Plateau. Conventions as in Figs. 5 and 6.

a reliable Jurassic pole. Paleocolatitudes from Sites 307 and 1213 are consistent with Jurassic-age skewness analysis poles (Larson and Sager, 1992) and imply southward motion of the pole as proposed by Cox and Gordon (1984). Because these basalt colatitudes only sample 4

independent units, they provide only a weak constraint on the pole position. The difference between the 142 Ma skewness pole and 123 Ma basalt core pole (11.3 ± 4.9◦ ) implies a slow drift rate (∼0.5◦ /Ma) similar to that among the 123 to 92 Ma poles. Given existing data, the

Fig. 9. Apparent polar wander path summary. Published paleomagnetic poles shown as in Fig. 1 (filled circles, stars, and squares) with the addition of poles from this study (open circles). Ellipses show 95% confidence regions. Heavy gray arrows represent polar wander. Dashed Jurassic section shows probable southward polar motion, which is poorly defined by existing data. Solid arrow from 123 to 92 Ma shows slow Early to mid-Cretaceous motion. Dashed line between 92 and 81 Ma poles shows section of possible rapid motion. Arrow at northern end of polar path shows northward drift in Late Cretaceous and afterwards.

342

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

APWP turnaround appears to have occurred at about 123 Ma as this is the pole that is farthest south. Because it is ∼40◦ from the spin axis, this pole implies that amount of northward drift of the Pacific plate since Early Cretaceous time. 4.2. Ontong Java Plateau pole and Pacific plate integrity The Ontong Java pole, distinct from the coeval 123 Ma pole from other Pacific sites, is an intriguing and somewhat unexpected result because it implies a significant anomaly. Riisager et al. (2004) noted the discrepancy and concluded that OJP data indicate the true Pacific pole for this age, a pole closer to the spin axis than those determined in most prior studies. They attributed the difference to several factors including the use of seamount paleomagnetic poles, anomaly skewness poles, and colatitudes with misinterpreted magnetic polarity in previous pole calculations. The trouble with this conclusion is that one must ignore a large number of basalt paleocolatitudes as well as other data types (Fig. 8). I considered the possibility that Riisager et al. (2004) are correct and that there is some problem that causes a systematic error in other Pacific data. For example, tectonic rotation of ocean crust (e.g., Schouten and Denham, 2000) could bias paleocolatitudes from ocean crust sites. If ocean crust rotates away from the spreading axis significantly, crust emplaced at the Pacific-Izanagi ridge (Japanese magnetic lineations) could give larger paleocolatitude values that would give a paleomagnetic pole farther south in the Atlantic. Although this tilting could account for the southward offset of paleocolatitudes from Sites 303, 304, 581, and 1179, it would not have the same affect on data from Site 166 (Phoenix magnetic lineations), nor should it affect data from Sites 800 or 802 (Cretaceous intrusives) or Sites 872, 878, and 879 (seamount basalts), all of which are consistent with similar age data. Furthermore, the agreement of tilt-biased paleocolatitudes would be fortuitous in that the tectonic rotations would all have to be nearly identical in magnitude. A simpler conclusion is to accept that these data are in agreement because none are significantly biased. I also considered whether some paleocolatitudes were misinterpreted owing to incorrect assignment of polarity, as suggested for Site 878 by Riisager et al. (2004). In the case of Site 878, the paleocolatitude would be ∼30◦ less (for the oldest unit) if the igneous units were formed north of the equator, rather than south, as originally interpreted (Nakanishi and Gee, 1995). Although the polarity for samples from this site were based on the authors’

assumption that the basalts were magnetized in the southern hemisphere, this interpretation gives reversed polarity basement with a normal-polarity volcanic unit in the sedimentary cap. This interpretation is consistent with a magnetic inversion of the seamount anomaly that gives an overall reversed polarity (Sager et al., 1993). It is also consistent with the interpreted stratigraphy, which gives the formation of the seamount basement during chron M0, followed by post-shield building stage volcanism during the Cretaceous Quiet Period (Nakanishi and Gee, 1995). Moreover, to change polarity interpretations for other sites requires the assumption that similar misinterpretations have been repeated in many other studies. Such a mass misinterpretation seems unlikely. If the 123 Ma pole is representative of the Pacific plate as a whole, then what could cause Ontong Java paleocolatitudes to be discrepant? Could the OJP paleomagnetic data be biased by systematic error? Given the number of paleocolatitudes, their self-consistency, and their areal distribution across the plateau, this seems unlikely. One possibility is tectonic tilting, as suggested for Site 807 by Mayer and Tarduno (1993). This explanation was plausible when applied to one apparently discrepant site, but not with current data from many sites because it implies coherent tilting of virtually the entire plateau. Another explanation is that these sites do not properly sample secular variation and that the OJP pole represents an extreme field direction. This is also implausible because it implies that a huge area of basalt was emplaced in a few thousand years or less (i.e., most of the upper surface of the plateau is a single, virtual cooling unit). Besides, Riisager et al. (2003) examined inclination data scatter in the four Leg 192 basalt sections and concluded that the observed variance was consistent with that expected if a full range of secular variation was sampled. Furthermore, discarding selected non-OJP data to lessen the discrepancy (Riisager et al., 2004) does not solve the problem because it requires too many data to be ignored. In many paleomagnetic studies, finding an area within a plate that gives a anomalous paleomagnetic pole leads to the conclusion that the area had a different drift history. Given the lack of an alternate explanation for the anomalous OJP data, I conclude that the plateau underwent differential tectonic motion relative to the northern Pacific plate or that the OJP data are affected by some obscure bias. The former is a radical departure from accepted plate tectonic history, but the latter is scientifically unpalatable. Taken at face value, the ∼15◦ distance between the 123-Ma and OJP poles implies that the OJP has undergone less northward motion by about the same amount. Fig. 10 shows that the Jurassic Quiet Zone, located north

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

343

Fig. 10. Plot showing areas with anomalous paleomagnetic data in the south Pacific. Filled circles show sites with data used to determine the Pacific APWP. Note that no data are available from south of the equator, except for OJP and Chatham Island. Open stars show sites with anomalous paleolatitudes on OJP. Light gray areas show magnetic lineations that give anomalous effective inclinations from skewness studies (Larson and Sager, 1992; Petronotis and Gordon, 1999). Box shows location of anomalous Jurassic ocean crust on island of Malaita (Ishikawa et al., 2005). Star shows Chatham Island, a site that gives a paleomagnetic pole that is apparently concordant with north Pacific plate. KQZ and JQZ stand for Cretaceous Quiet Zone and Jurassic Quiet Zone, respectively. Straight lines show magnetic lineations and fracture zones. Heavy isochrons are from M¨uller et al. (1997) and thin lines are M-anomalies from Nakanishi et al. (1992).

of OJP, contains barely enough room to fit 15◦ of seafloor spreading. The lack of space argues that a simple spreading ridge boundary between the OJP and rest of the Pacific is unlikely and that if there was an unknown plate boundary, it may have been more complex. For example, if the Euler pole describing OJP versus Pacific motion was close to the plateau, it could have resulted in little differential translation (i.e., mostly rotation) of the plateau whereas the motion of the paleomagnetic pole, nearly 90◦ distant, could have been significant. Recall

also that the constraint on the OJP pole paleolongitude is poor and that the actual OJP pole may be located significantly east or west of the calculated location. Without better azimuthal constraint, an attempt to determine the relative rotation between the OJP and rest of the Pacific is difficult and potentially misleading. Unfortunately, existing tectonic knowledge of the OJP and surrounding Cretaceous Quiet Zone is poor, so it is difficult test tectonic hypotheses. Furthermore, the Cretaceous Quiet Zone in the south Pacific is much

344

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

wider than that in the north Pacific, leading some authors to postulate large ridge jumps (e.g., Winterer et al., 1974; Joseph et al., 1993), while others disagree (Larson et al., 2002). Although such a southward ridge jump would not have produced the observed lesser northward drift of OJP, it is clear that the Cretaceous Quiet Zone tectonics were complex and are poorly understood. As seen in Fig. 10, there is only one paleomagnetic datum from the Pacific plate south of OJP and thus the integrity of the plate as a whole is unsubstantiated. Moreover, some authors (Acton and Gordon, 1994) have suggested that problems reconciling paleomagnetic data from the Pacific plate to surrounding areas through plate circuits result from a fossil plate boundary in the south Pacific that caused relative motion between the north and south Pacific plate. Several pieces of evidence may support Cretaceous Quiet Zone tectonic complexities or missing plate boundary hypotheses. In a study of magnetic anomaly skewness, Larson and Sager (1992) noted that paleomagnetic pole estimates from the Japanese, Hawaiian, and Phoenix lineations did not agree unless significant anomalous skewness was postulated or it was assumed that an unknown plate (the “Stealth plate”) once separated the lineation sets, causing southward motion of the Phoenix anomalies (Fig. 10). The Stealth plate offset is ∼15◦ , similar to the OJP offset. Furthermore, Jurassic ocean crust found in Malaita (Solomon Islands), associated with OJP, is ∼15◦ farther south than other Pacific crust of the same age (Fig. 10) (Ishikawa et al., 2005). Magnetic anomaly data from the south Pacific are also anomalous and show a similar offset to that of the Phoenix lineations. Paleomagnetic pole loci estimated from magnetic anomalies formed along the Pacific–Antarctic ridge are significantly displaced from loci estimated from Pacific-Farallon anomalies in the north Pacific (see Fig. 14 of Petronotis and Gordon, 1999). Petronotis and Gordon (1999) ascribe the difference to anomalous skewness, even though the amount is large for fast spreading crust (Dyment and ArkaniHamed, 1995). An alternative explanation is a tectonic offset of approximately the sense and magnitude suggested by the OJP pole offset. Tectonic displacement could also explain why paleomagnetic poles determined from magnetic anomaly skewness for the Late Cretaceous are offset to the east of the poles determined in this study and others from other data (Fig. 1). When published skewness data from chron 32 (Petronotis and Gordon, 1999), for example, are averaged without the data from the south Pacific, the resulting pole is closer to the prime meridian and other Pacific paleomagnetic data of the same age (e.g., Sager and Pringle, 1988).

Confounding the situation, the only paleomagnetic datum from the south Pacific plate, from Chatham Island (Figs. 1 and 10), does not apparently support a tectonic offset of north and south Pacific plate. This pole, which is within the confidence ellipse of the 80 Ma north Pacific pole, was determined from 66 to 78 Ma basalts erupted on the island (Grindley et al., 1977). Although the agreement of the Chatham data cast doubt on a split between north and south Pacific, these data are poorly dated and the paleomagnetic analyses are not up to modern standards. If the age of the Chatham basalts is underestimated by 10–15 Myr, the Chatham pole corresponds to the 92 Ma pole, in which case there is a significant offset of the correct sign and magnitude to agree with the OJP offset. In summary, the discrepancy between the OJP pole with other north Pacific poles suggests a tectonic disconnect between OJP (and perhaps the entire south Pacific plate) and the rest of the Pacific. Unfortunately, uncertainties in paleomagnetic data and tectonic reconstructions do not allow unequivocal independent confirmation of this hypothesis. Because of this situation, the interpretation that OJP moved relative to north Pacific must be treated with caution and tested with new data in future studies. 5. Conclusions Basalt paleocolatitude data from 38 sites on the Pacific plate, cored by DSDP and ODP, were used to calculate five Cretaceous paleomagnetic poles with ages of 80, 92, 112, 121 (OJP), and 123 Ma. Despite >30 years of coring in the Pacific, most time intervals have too few basalt paleocolatitude data to reliably define a paleomagnetic pole. Data are abundant only for the intervals between 76–82 and 110–130 Ma, although poles can be calculated for other time intervals (i.e., 88–97 Ma), albeit with larger uncertainties. Because of poor control of the pole location along the trend of paleocolatitude arcs, I found it advantageous to use declination data from dated seamount anomaly inversions to reduce uncertainty in that direction. The 92, 112, and 123 Ma poles are consistent with other Pacific paleomagnetic data and delineate a period of slow polar motion during the Early to midCretaceous. The 123 Ma pole is ∼40◦ from the spin axis and appears to represent the maximum northward motion of the Pacific plate. A gap of 13.6◦ between the 92 and 80 Ma poles suggests rapid polar motion in the Late Cretaceous at ∼1◦ /Myr. Polar motion prior to the 123 Ma pole is poorly defined by basalt paleocolatitude data but appears consistent with anomaly skewness poles that imply southward polar motion at a similar rate to that

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

between 92 and 123 Ma. The timing of the change in direction of polar motion from north to south is uncertain, but must be around 123 Ma because this pole is farthest south and apparently forms the apex of the polar curve. Data from Ontong Java Plateau are discordant and give a 121 Ma pole position ∼15◦ north of the pole calculated from other Pacific sites of the same age. A possible explanation for this discrepancy is that the plateau resided on an independent plate that was subsumed into the Pacific plate, probably during the Cretaceous Quiet Period. Acknowledgments I thank Richard Gordon for his pole calculation program and discussions about basalt core data. Reviews by Peter Riisager, Katerina Petronotis, Gary Acton, and an anonymous reviewer were a great help in clarifying and polishing the manuscript. Funding for some

345

of the study came from grants awarded by JOI/USSAC to work on ODP Legs 191 and 198. Additional funding was provided by the Jane and R. Ken Williams ’45 Chair for Ocean Drilling Science, Education, and Technology. And, of course, this study would not have been possible without the Deep Sea Drilling Program, the Ocean Drilling Program, as well as the technical staff, scientists, and crew on board the D/V Glomar Challenger and D/V JOIDES Resolution. This research used samples and/or data provided by the Ocean Drilling Program (ODP). ODP is sponsored by the U.S. National Science Foundation (NSF) and participating countries under management of Joint Oceanographic Institutions (JOI), Inc. Appendix A See Table A.1.

Table A.1 Pole data groups, residuals, and importances Data type

Location

Datum

Std. Err.

Model

Resid.

Import.

ID

55.6 50.7 52.7 98.3 95.3 96.3 96.4 104.1 91.9 99.3 109.8 106.5 97.3 86.2 99.7

6.0 6.8 6.3 4.5 6.6 9.3 3.9 9.4 13.8 7.2 7.2 7.2 7.2 7.2 7.2

55.8 54.6 55.3 94.8 91.7 96.9 99.6 95.8 89.7 97.9 100.3 99.9 98.4 98.2 97.9

−0.2 −4.9 −2.6 3.5 3.6 −0.6 −3.2 8.3 2.2 1.4 6.7 6.5 −1.1 −12.0 1.8

0.12 0.09 0.11 0.20 0.19 0.08 0.27 0.05 0.08 0.14 0.12 0.12 0.15 0.15 0.15

ODP 1203 ODP 883 ODP 884 ODP 873 DSDP 163 DSDP 165 DSDP 462 (sed) ODP 869 (sed) ODP 884 Dec Chatauqua Show Kona 5S Haydn Mendelsshon-E Mendelssohn-W

92 Ma; 59.7◦ N, 345.4◦ E; χ2 = 3.96; d.f. = 7 CO 10.7 173.6 CO 11.8 177.6 CO 19.1 190.5 CO 4.2 201.5 CO 2.4 160.5 DEC 21.2 163.3 DEC 29.5 153.6 DEC 31.8 195.0 DEC 21.7 161.9

115.2 103.5 101.9 107.4 116.8 94.9 83.5 107.5 77.8

8.9 9.9 11.0 5.3 8.2 7.2 7.2 7.2 7.2

109.3 107.8 98.6 110.1 117.8 88.9 84.1 104.5 −88.2

5.9 −4.3 3.3 −2.7 −1.0 6.0 −0.6 3.0 −10.4

0.18 0.14 0.10 0.38 0.24 0.23 0.21 0.30 0.23

DSDP 169 DSDP 170 DSDP 171 DSDP 315 ODP 803 Wilde Makarov Mahler Miami

112 Ma; 55.6◦ N, 334.9◦ E; χ2 = 11.52; d.f. = 9 CO 7.2 165.0 CO 7.2 165.0 CO 18.5 180.4 CO 12.1 153.2

117.3 112.8 105.0 109.3

4.6 5.6 6.6 5.0

116.7 116.7 102.8 112.3

0.6 −3.9 2.2 −3.0

0.25 0.17 0.12 0.24

DSDP 462(i) DSDP 462(e) ODP 865 ODP 802

Lat 80 Ma; CO CO CO CO CO CO CO CO DEC DEC DEC DEC DEC DEC DEC

73.3◦ N,

(◦ N)

351.3◦ E; 51.0 51.2 51.5 12.0 11.2 8.2 7.2 11.0 51.5 22.2 17.9 17.1 26.6 25.1 25.1

χ2

Lon

(◦ E)

= 3.96; d.f. = 11 167.6 167.8 168.3 164.9 209.7 195.1 165.0 164.8 168.3 197.4 207.3 205.8 198.7 198.3 197.2

346

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Table A.1 (Continued ) Data type

Datum

Std. Err.

Model

Resid.

Import.

ID

100.7 120.5 98.0 91.7 80.2 87.9 92.2

10.0 5.1 7.2 7.2 7.2 7.2 7.2

95.6 114.0 86.3 95.2 94.9 84.0 83.8

5.1 6.5 11.7 −3.5 −14.7 3.9 8.4

0.06 0.21 0.17 0.23 0.22 0.17 0.17

ODP 843 ODP 872 Winterer Mij Lep Lo En Takuyo-Daini Takuyo-Daisan

121 Ma; 65.3◦ N, 331.0◦ E; χ2 = 3.81; d.f. = 8 CO 3.6 156.6 CO −1.2 157.0 CO −0.4 161.7 CO −0.7 159.8 CO 0.9 161.5 CO −0.5 158.5 DEC 27.3 151.9 DEC 35.8 142.7 DEC 34.2 144.3 DEC 34.3 144.9

108.1 119.1 113.7 117.1 114.2 123.3 93.3 78.0 87.9 92.2

3.5 5.8 4.7 5.8 4.7 11.5 7.2 7.2 7.2 7.2

110.9 115.7 114.6 115.1 113.3 114.9 91.8 85.9 84.0 83.8

−2.8 3.4 −0.9 2.0 0.9 8.4 2.9 −8.5 0.7 5.2

0.35 0.13 0.19 0.12 0.19 0.03 0.25 0.25 0.25 0.25

DSDP 462(i) DSDP 462(e) ODP 865 ODP 802 ODP 843 ODP 872 MIT Daichi-Kashima Takuyo-Daisan Takuyo-Daini

123 Ma; 50.0◦ N, 329.1◦ E; χ2 = 6.05; d.f. = 12 CO 27.3 150.9 CO 27.3 150.9 CO 34.2 144.3 CO 21.9 152.3 CO 21.3 174.3 CO 43.9 159.8 CO 3.8 184.9 CO 40.8 154.5 CO 39.3 155.1 CO 41.1 160.0 DEC 27.3 151.9 DEC 35.8 142.7 DEC 34.2 144.3 DEC 34.3 144.9

103.8 105.9 99.7 105.1 103.9 82.9 125.7 95.8 101.4 85.4 93.3 78.0 87.9 92.2

5.0 5.9 11.0 8.5 5.4 4.7 9.3 9.2 9.6 3.6 7.2 7.2 7.2 7.2

102.7 102.7 95.7 108.1 105.3 85.6 118.0 89.1 90.6 88.4 91.8 85.9 84.0 83.8

1.1 3.2 4.0 −3.0 −1.4 −2.7 7.7 6.7 10.8 −3.0 1.5 −7.9 1.0 5.6

0.16 0.11 0.04 0.05 0.13 0.17 0.06 0.04 0.04 0.28 0.24 0.22 0.22 0.22

ODP 878(y) ODP878(o) ODP 879 ODP 800 ODP 866 DSDP 581 DSDP 166 DSDP 303 DSDP 304 ODP 1179 MIT Daichi-Kashima Takuyo-Daisan Takuyo-Daini

CO CO DEC DEC DEC DEC DEC

Location Lat (◦ N)

Lon (◦ E)

19.3 10.1 32.8 8.8 10.1 34.2 34.3

200.9 162.9 148.4 163.2 162.8 144.3 143.9

Table headings: Data type: CO = colatitude; DEC = declination; Std. Err. = standard error; Model = datum calculated from mean pole; Resid. = observed minus calculated datum (residual); Import. = datum importance; χ2 = chi-squared statistic; d.f. = degrees of freedom.

References Acton, G.D., Gordon, R.G., 1991. A 65 Ma palaeomagnetic pole for the Pacific plate from the skewness of magnetic anomalies 27r-31. Geophys. J. Int. 106, 407–420. Acton, G.D., Gordon, R.G., 1994. Paleomagnetic tests of Pacific plate reconstructions and implications for motion between hotspots. Science 263, 1246–1254. Besse, J., Courtillot, V., 2002. Apparent and true polar wander and the geometry of the geomagnetic field over the last 200 Myr. J. Geophys. Res. 107, 2300, doi:10.1029/2000JB000050. Bryan, P.C., Shoberg, T., Gordon, R.G., Petronotis, K.E., Bergersen, D.D., 1993. A paleomagnetic pole and estimated age for Lo-En Guyot, Republic of the Marshall Islands. In: Pringle, M.S., Sager, W.W., Sliter, W.V., Stein, S. (Eds.), The Mesozoic Pacific: Geology, Tectonics, and Volcanism, Geophysical Monograph Series, vol. 77. American Geophysical Union, Washington, DC, pp. 387–400.

Cande, S.C., 1976. A palaeomagnetic pole from Late Cretaceous marine magnetic anomalies in the Pacific. Geophys. J. R. Astron. Soc. 44, 547–566. Cande, S.C., 1978. Anomalous behavior of the paleomagnetic field inferred from the skewness of anomalies 33 and 34. Earth Planet. Sci. Lett. 40, 275–286. Cockerham, R.S., 1979. A paleomagnetic and magnetic property study of DSDP basalts and sediments of Cretaceous age (abstract). EOS Trans. Am. Geophys. Union 60, 239. Cockerham, R.S., Hall, J.M., 1976. Magnetic properties and paleomagnetism of some DSDP Leg 33 basalts and sediments and their tectonic implications. J. Geophys. Res. 81, 4207–4222. Cottrell, R.D., Tarduno, J.A., 2000. Late Cretaceous true polar wander: not so fast. Science 288, 2283a (Science Online). Cottrell, R.D., Tarduno, J.A., 2003. A Late Cretaceous pole for the Pacific plate: implications for apparent and true polar wander and the drift of hotspots. Tectonophysics 362, 321–333.

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349 Cox, A., Gordon, R.G., 1984. Paleolatitudes determined from paleomagnetic data from vertical cores. Rev. Geophys. Space Phys. 22, 47–72. Doubrovine, P.V., Tarduno, J.A., 2004. Late Cretaceous paleolatitude of the Hawaiian hot spot: new paleomagnetic data from Detroit Seamount (ODP Site 883). Geochem. Geophys. Geosys. 5, Q11L04, doi:10.1029/2004GC000745. Duncan, R.A., Keller, R.A., 2004. Radiometric ages for basement rocks from the Emperor Seamounts, ODP Leg 197. Geochem. Geophys. Geosys. 5, Q08L03, doi:10.1029/2004GC000704. Dyment, J., Arkani-Hamed, J., 1995. Spreading-rate-dependent magnetization of the oceanic lithosphere inferred from the anomalous skewness of marine magnetic anomalies. Geophys. J. Int. 121, 789–804. Eittreim, S.L., Ragozin, N., Gnibidenko, H., Helsley, C.E., 1992. Crustal age between the Clipperton and Clarion fracture zones. Geophys. Res. Lett. 24, 2365–2368. Francheteau, J., Harrison, C.G.A., Sclater, J.G., Richards, M.L., 1970. Magnetization of Pacific seamounts: a preliminary polar curve for the northeastern Pacific. J. Geophys. Res. 75, 2035– 2061. Gee, J., Staudigel, H., Tauxe, L., 1989. Contriubution of induced magnetization to magnetization of seamounts. Nature 342, 170– 173. Gordon, R.G., 1983. Late Cretaceous apparent polar wander of the Pacific plate: evidence for a rapid shift of the Pacific hotspots with respect to the spin axis. Geophys. Res. Lett. 10, 709– 712. Gordon, R.G., 1990. Test for bias in paleomagnetically determined paleolatitudes from Pacific plate Deep Sea Drilling Project sediments. J. Geophys. Res. 95, 8397–8404. Gordon, R.G., Cox, A., 1980. Calculating palaeomagnetic poles for oceanic plates. Geophys. J. R. Astron. Soc. 63, 619– 640. Grindley, G.W., Adams, C.J.D., Lumb, J.T., Watters, W.A., 1977. Paleomagnetism, K–Ar dating, and tectonic interpretation of Upper Cretaceous and Cenozoic volcanic rocks of the Chatham Islands, New Zealand. N. Z. J. Geol. Geophys. 20, 425–467. Hammond, S.R., Kroenke, L.W., Theyer, F., 1975. Northward motion of the Ontong Java Plateau between −110 and −30 m.y.: a paelomagnetic investigation of DSDP Site 289. Init. Repts. DSDP 20, 415–418. Helsley, C.E., 1993. Paleomagnetic results from Leg 136. Proc. ODP Sci. Res. 136, 45–50. Hildebrand, J.A., Parker, R.L., 1987. Paleomagnetism of Cretaceous Pacific seamounts revisited. J. Geophys. Res. 92, 12695–12712. Irving, E., Irving, G.A., 1982. Apparent polar wander paths Carboniferous through Cenozoic and the assembly of Gondwana. Geophys. Surv. 5, 141–188. Ishikawa, A., Nakamura, E., Mahoney, J.J., 2005. Jurassic oceanic lithosphere beneath the southern Ontong Java Plateau: Evidence from xenoliths in aln¨oite, Malaita, Solomon Islands. Geology 33, 393–396. Joseph, D., Taylor, B., Shor, A.N., Yamazaki, T., 1993. The NovaCanton Trough and Late Cretaceous evolution of the Central Pacific. In: Pringle, M.S., Sager, W.W., Sliter, W.V., Stein, S. (Eds.), The Mesozoic Pacific: Geology, Tectonics, and Volcanism. American Geophysical Union, Washington, DC, pp. 171– 185. Keller, R.A., Duncan, R.A., Fisk, M.R., 1995. Geochemistry and 40 Ar/39 Ar geochronology of basalts from ODP Leg 145. Proc. ODP Sci. Res. 145, 333–344.

347

Kono, M., 1980. Paleomagnetism of DSDP Leg 55 basalts and implications for the tectonics of the Pacific plate. Init. Repts. DSDP 55, 737–752. Koppers, A.A.P., 1998. 40 Ar/39 Ar Geochronology and Isotope Geochemistry of the West Pacific Seamount Province. Ph.D. Dissertation, Vrije Universiteit, Amsterdam, 262 p. Koppers, A.A.P., Staudigel, H., Duncan, R.A., 2003a. High resolution 40 Ar/39 Ar dating of the oldest oceanic basalts in the western Pacific basin. Geochem. Geophys. Geosys. 4, 8914, doi:101029/2003GC000574. Koppers, A.A.P., Staudigel, H., Pringle, M.S., Wijbrans, J.R., 2003b. Short-lived and discontinuous intraplate volcanism in the South Pacific: hot spots or extensional volcanism? Geochem. Geophys. Geosys. 4, 1089, doi:101029/2003GC000533. Larson, R.L., Lowrie, W., 1975. Paleomagnetic evidence for motion of the Pacific plate from Leg 32 basalts and magnetic anomalies. Init. Repts. DSDP 32, 571–577. Larson, R.L., Pockalny, R.A., Viso, R.F., Erba, E., Abrams, L.J., Luyendyk, B.P., Stock, J.M., Clayton, R.W., 2002. Mid-Cretaceous tectonic evolution of the Tongareva triple junction in the southwestern Pacific. Geology 30, 67–70. Larson, R.L., Sager, W.W., 1992. Skewness of magnetic anomalies M0 to M29 in the northwestern Pacific. Proc. ODP Sci. Res. 129, 471–481. Livermore, R.A., Vine, F.J., Smith, A.G., 1984. Plate motions and the geomagnetic field, II, Jurassic to Tertiary. Geophys. J. R. Astron. Soc. 79, 939–961. Mahoney, J.J., Duncan, R.A., Tejada, M.L.G., Sager, W.W., 2005. Jurassic-Cretaceous boundary age and mid-ocean-ridge-type mantle source for Shatsky Rise. Geology 33, 185–188. Mahoney, J.J., Storey, M., Duncan, R.A., Spencer, K.J., Pringle, M., 1993. Geochemistry and age of the Ontong Java Plateau. In: Pringle, M.S., Sager, W.W., Sliter, W.V., Stein, S. (Eds.), The Mesozoic Pacific: Geology, Tectonics, and Volcanism, Geophysical Monograph Series, vol. 77. American Geophysical Union, Washington, DC, pp. 233–261. Marshall, M., 1978. The magnetic properties of some DSDP basalts from the North Pacific and inferences for Pacific plate tectonics. J. Geophys. Res. 83, 289–308. Mayer, H., Tarduno, J.A., 1993. Paleomagnetic investigation of the igneous sequence, Site 807, Ontong Java Plateau, and a discussion of Pacific true polar wander. Proc. ODP Sci. Res. 130, 51–59. McFadden, P.L., Reid, A., 1982. Analysis of paleomagnetic inclination data. Geophys. J. R. Astron. Soc. 69, 307–319. M¨uller, R.D., Roest, W.R., Royer, J.-Y., Gahagan, L.M., Sclater, J.G., 1997. Digital isochrons of the world’s ocean floor. J. Geophys. Res. 102, 3211–3214. Nakanishi, M., Gee, J., 1995. Paleomagnetic investigations of volcanic rocks: Paleolatitudes of the Northwestern Pacific Guyots. Proc. ODP. Sci. Res. 144, 585–604. Nakanishi, M., Sager, W.W., Klaus, A., 1999. Magnetic lineations within Shatsky Rise, northwest Pacific Ocean: implications for hot spot-triple junction interaction and oceanic plateau formation. J. Geophys. Res. 104, 7539–7556. Nakanishi, M., Tamaki, K., Kobayashi, K., 1992. Magnetic anomaly lineations from late Jurassic to Early Cretaceous in the west–central Pacific ocean. Geophys. J. Int. 109, 701–719. Ozima, M., Kaneoka, I., Saito, K., Honda, H., Yanagisawa, M., Takigami, Y., 1983. Summary of geochronological studies of submarine rocks from the western Pacific Ocean. In: Hilde, T.W.C., Uyeda, S. (Eds.), Geodynamics of the Western Pacific-

348

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349

Indonesian Region. American Geophysical Union, Washington, DC, pp. 137–142. Ozima, M., Saito, K., Takigami, Y., 1981. 49 Ar–39 Ar geochronological studies on rocks drilled at Holes 462 and 462A, Deep Sea Drilling Project Leg 61. Init. Repts. DSDP 61, 701–703. Parker, R.L., 1991. A theory of ideal bodies for seamount magnetism. J. Geophys. Res. 96, 16101–16112. Parkinson, I.J., Schaefer, B.F., Arculus, R.J., 2003. Residue-melt relationships; Os isotopes from the Ontong Java Plateau (abstract). Geophys. Res. Abstr. (Eur. Geophys. Soc.) 5, 14351. Petronotis, K.E., Gordon, R.G., Acton, G.D., 1992. Determining palaeomagnetic poles and anomalous skewness from marine magnetic anomaly skewness data from a single plate. Geophys. J. Int. 109, 209–224. Petronotis, K.E., Gordon, R.G., 1999. A Maastrichtian paleomagnetic pole for the Pacific plate from a skewness analysis of marine anomaly 32. Geophys. J. Int. 139, 227–247. Pockalny, R.A., Larson, R.L., 2003. Implications for crustal accretion at fast spreading ridges from observations in Jurassic oceanic crust in the western Pacific. Geochem. Geophys. Geosys. 4 (1), 8903, doi:10.1029/2001GC000274. Pr´evot, M., Mattern, E., Camps, P., Dagnieres, M., 2000. Evidence for a 20◦ tilting of the Earth’s rotation axis 110 million years ago. Earth Planet. Sci. Lett. 179, 517–528. Pringle, M.S., 1992. Radiometric ages of basaltic basement recovered at Sites 800, 801 and 802, Leg 129, western Pacific Ocean. Proc. ODP Sci. Res. 129, 389–404. Pringle, M.S., 1993. Age progressive volcanism in the Musicians Seamounts: a test of the hot spot hypothesis for the Late Cretaceous Pacific. In: Pringle, M.S., Sager, W.W., Sliter, W.V., Stein, S. (Eds.), The Mesozoic Pacific: Geology, Tectonics, and Volcanism, Geophysical Monograph Series, vol. 77. American Geophysical Union, Washington, DC, pp. 187–215. Pringle, M.S., Duncan, R.A., 1995. Radiometric ages of basement lavas recovered at Loen, Wodejebato, MIT, and Takuyo-Dasan guyots, northwestern Pacific Ocean. Proc. ODP Sci. Res. 144, 547–557. Richards, M.L., Vacquier, V., Van Voorhis, G.D., 1967. Calculations of the magnetization of uplifts from combining topographic and magnetic surveys. Geophysics 32, 678–707. Riisager, P., Hall, S., Antretter, M., Zhao, X., 2003. Paleomagnetic paleolatitude of Early Cretaceous Ontong Java Plateau basalts: Implications for Pacific apparent and true polar wander. Earth Planet. Sci. Lett. 208, 235–252. Riisager, P., Hall, S., Antretter, M., Zhao, X., 2004. Early Cretaceous Pacific palaeomagnetic pole from Ontong Java Plateau basement rocks. In: Fitton, J.G., Mahoney, J.J., Wallace, P.J., Saunders, A.D. (Eds.), Origin and Evolution of the Ontong Java Plateau. Geol. Soc. Lond., Spec. Pub. 229, pp. 31–44. Sager, W.W., 1983. Seamount Paleomagnetism and Pacific Plate Tectonics. Ph.D. Dissertation, University of Hawaii, Honolulu, 492 p. Sager, W.W., 1987. Late Eocene and Maastrichtian paleomagnetic poles for the Pacific plate: implications for the validity of seamount paleomagnetic poles. Tectonophysics 144, 301– 314. Sager, W.W., 1992. Seamount age estimates from paleomagnetism and their implications for the history of volcanism on the Pacific plate. In: Keating, B., Bolton, B. (Eds.), Geology and Offshore Mineral Resources of the Central Pacific Basin, CircumPacific Council for Energy and Mineral Resources, Earth Science Series, vol. 14. Springer-Verlag, New York, pp. 21– 37.

Sager, W.W., 2003. Chron 33r paleomagnetic pole for the Pacific plate. Geophys. Res. Lett. 30 (16), 1873, doi:10.1029/2003GL017964. Sager, W.W., 2005. Data report: paleomagnetism of basaltic rocks cored from Western Pacific DSDP and ODP boreholes. Proc. ODP Sci. Res. 191, 1–27 (CD-ROM). Sager, W.W., Duncan, R.A., Handschumacher, D.W., 1993. Paleomagnetism of the Japanese and Marcus-Wake seamounts, western Pacific Ocean. In: Pringle, M.S., Sager, W.W., Sliter, W.V., Stein, S. (Eds.), The Mesozoic Pacific: Geology, Tectonics, and Volcanism, Geophysical Monograph Series, vol. 77. American Geophysical Union, Washington, DC, pp. 401– 435. Sager, W.W., Horner-Johnson, B., 2005. Paleomagnetism of basaltic rocks cored at ODP Site 1179: Implications for paleolatitude and the Mid-Cretaceous Pacific Paleomagnetic pole. Proc. ODP, Sci. Res. 191, 1–10 (CD-ROM). Sager, W.W., Koppers, A.A.P., 2000. Late Cretaceous Polar Wander of the Pacific plate: evidence of a rapid true polar wander event. Science 287, 455–459. Sager, W.W., Pringle, M.S., 1987. Paleomagnetic constraints on the origin and evolution of the Musicians and South Hawaiian seamounts, central Pacific Ocean. In: Keating, B., Fryer, P., Batiza, R., Boehlert, G. (Eds.), Seamounts, Islands, and Atolls, AGU Geophysical Monograph Series, vol. 43. American Geophysical Union, Washington, DC, pp. 133–162. Sager, W.W., Pringle, M.S., 1988. Mid-Cretaceous to early tertiary apparent polar wander path of the Pacific plate. J. Geophys. Res. 93, 11753–11771. Sager, W.W., Tarduno, J.A., MacLeod, C.J., 1995. Paleomagnetism of Cretaceous sediments, ODP Hole 869B, western Pacific Ocean: magnetic polarity, paleolatitude, and a paleomagnetic pole. Proc. ODP Sci. Res. 143, 405–418. Schouten, H., Denham, C.R., 2000. Comparison of volcanic construction in the Troodos ophiolite and ocean crust using paleomagnetic inclinations from Cyprus Crustal Study Project (CCSP) CY-1 and CY-1A and Ocean Drilling Program (ODP) 504B drill cores. Geol Soc. Am. Spec. Publ. 346, 181–194. Schlanger, S.O., Garcia, M.O., Keating, B.H., Naughton, J.J., Duncan, R.A., Sager, W.W., Haggerty, J.A., Philpotts, J.A., 1984. Geology and geochronology of the Line Islands. J. Geophys. Res. 89, 11261–11272. Schouten, H., McCamy, K., 1972. Filtering marine magnetic anomalies. J. Geophys. Res. 77, 7089–7099. Sharp, W.D., Clague, D.A., 2002. An older, slower Hawaii-Emperor Bend. Eos Trans. AGU, 83 (47), Fall Meet. Suppl., Abstract T61C04. Steiner, M.B., 1981. Paleomagnetism of the igneous complex, Site 462. Init. Repts. DSDP 61, 717–729. Takigami, Y., Kaneoka, I., Ishii, T., Nakamura, Y., 1989. 40 Ar–39 Ar ages of igneous rocks recovered from Daiichi-Kashima and Erimo seamounts during the Kaiko project. Palaeogeog. Palaeoclim. Palaeoecol. 71, 71–81. Tarduno, J.A., 1990. Absolute inclination values from deep sea sediments: a reexamination of the Cretaceous Pacific record. Geophys. Res. Lett. 17, 101–104. Tarduno, J.A., Cottrell, R.D., 1997. Paleomagnetic evidence for motion of the Hawaiian hotspot during formation of the Emperor Seamounts. Earth Planet. Sci. Lett. 153, 171– 180. Tarduno, J.A., Duncan, R.A., Scholl, D.W., Cottrell, R.D., Steinberger, B., Thordarson, T., Kerr, B.C., Neal, C.R., Frey, F.A., Torii, M., Carvallo, C., 2003. The emperor seamounts: southward motion of

W.W. Sager / Physics of the Earth and Planetary Interiors 156 (2006) 329–349 the Hawaiian hotspot plume in the Earth’s mantle. Science 301, 1064–1069. Tarduno, J.A., Sager, W.W., 1995. Polar standstill of the MidCretaceous Pacific plate and its geodynamic implications. Science 269, 956–959. Tominaga, M., Sager, W.W., Channell, J.E.T., 2005. Paleomagnetism of the Igneous Section, Hole 1213B, Shatsky Rise. Proc. ODP. Sci. Res., 198, [online]. Available from World Wide Web: http://www-odp.tamu.edu/publications/198 SR/VOLUME/CHAPTERS/113.PDF. Ueda, Y., 1985. Geomagnetic study on seamounts Daiiti-kasima and Katori with special reference to a subduction process of Daiitikasima. J. Geomag. Geoelectr. 37, 601–625. Vasas, S.M., Gordon, R.G., Petronotis, K.E., 1994. New paleomagnetic poles for the Pacific plate from analysis of the shapes of anomalies

349

33n and 33r (abstract). EOS, Trans. Am. Geophys. Union Fall Mtg. Suppl. 75, 203. Waggoner, D.G., 1993. The age and alternation of central Pacific oceanic crust near Hawaii, Site 843. Proc. ODP Sci. Res. 136, 119–132. Wallick, B.P., Steiner, M.B., 1992a. Paleomagnetism of Cretaceous basalts from the East Mariana Basin, western Pacific Ocean. Proc. ODP Sci. Res. 129, 447–454. Wallick, B.P., Steiner, M.B., 1992b. Paleomagnetic and rock magnetic properties of Jurassic Quiet Zone basalts, Hole 801C. Proc. ODP Sci. Res. 129, 455–470. Winterer, E.L., Lonsdale, P.F., Matthews, J.L., Rosendahl, B.R., 1974. Structure and acoustic stratigraphy of the Manihiki Plateau. DeepSea Res. 21, 793–814.