Fig. 5.17. The observed arid reversed magnetic profiles across the Pacific-Antarctic ridge over a distance of 1000 km. The reversed sequence beyond 4 Ma is inferred from the magnetic anomalies by assuming a constant spreading rate in the interval 0-4 Ma. Reproduced with permission from Pitman and Heirtzler (1966). © American Association for the Advancement of Science.

model was applied to a profile across the Reykjanes Ridge on the opposite side of the globe (Heirtzler et al., 1966), a good agreement was obtained if a spreading rate of 10 km Myr"1 was assumed.

By 1968 it was apparent that correlatable sequences of magnetic anomalies parallel to and bilaterally symmetric about the mid-ocean ridge system are present over extensive regions of the Pacific (Pitman et al., 1968), Atlantic (Dickson et al., 1968), and Indian (Le Pichon and Heirtzler, 1968) oceans. The system of magnetic anomaly numbers was initiated by assigning numbers from 1 (at the mid-ocean ridges) to 32 (for the oldest anomaly) to the most prominent positive magnetic anomaly peaks in the sequences. In a major review of these results, Heirtzler et al. (1968) assumed uniform spreading rates in the different oceans (the South Atlantic, the North and South Pacific, and the South Indian Oceans.) and extrapolated from the GPTS for the past few million years. Perhaps not surprisingly, this extrapolation led to different estimates of the polarity time scale for each of the four regions involved. Based on the then available information about the age of the sea floor, they took the bold step of resolving the problem by deciding that the South Atlantic Ocean had spread at a constant rate since the Late Cretaceous. Assuming a uniform axial spreading rate of 19 km Myr"1 at mid-latitudes in the South Atlantic, a magnetic reversal

Magnetic anomaly numbers

Magnetic anomaly numbers

Fig. 5.18. Comparison of the paleontological ages of basal sediments in deep-sea drill holes with the basement ages predicted from the magnetic anomalies. The 45° line is that expected for perfect agreement. After Lowrie and Alvarez (1981), with the permission of the Geological Society of America.

Magnetic anomaly age (Ma)

Fig. 5.18. Comparison of the paleontological ages of basal sediments in deep-sea drill holes with the basement ages predicted from the magnetic anomalies. The 45° line is that expected for perfect agreement. After Lowrie and Alvarez (1981), with the permission of the Geological Society of America.

chronology extending back to 80 Ma was then obtained. This chronology was subsequently found to be consistent with ages obtained by paleontological dating of basement sediments obtained by the DSDP (Maxwell et al., 1970; LaBrecque et al., 1977). One comparison of paleontological ages of basal sediments with magnetic anomaly ages is shown in Fig. 5.18.

The first substantial reanalysis of worldwide marine magnetic anomaly data since that of Heirtzler et al. (1968) was undertaken by Cande and Kent (1992a) to produce an updated time scale for the Late Cretaceous and Cenozoic. This was later refined by Cande and Kent (1995), who still used the South Atlantic as a reference ocean but assumed that its spreading rate had smooth variations that could be approximated by fitting a spline through a set of calibration points. Huestis and Acton (1997) argue that this assumption of smooth spreading at a reference ridge forces more erratic spreading rates at other ridges. To eliminate this problem, they proposed a formalism that penalizes nonsmooth spreading behavior equally for all ridges. The method then defines the time scale that has the best agreement with known chron ages and with anomaly-distance data from multiple ridges and allows the spreading rate for each ridge to be as nearly constant as possible. When applied to the data used by Cande and Kent (1992a) to establish their time scale (which singled out one ridge for the preferential assumption of smoothness) only modest changes, of <5%, were needed. A conceptual disadvantage of the Huestis and Acton (1997) approach is that it makes the unrealistic assumption that if any ridge changes spreading rate then all other ridges change their spreading rates at the same time.

Since the first major summary and interpretation by Heirtzler et al. (1968) in terms of the Vine-Matthews model and sea-floor spreading, detailed magnetic anomaly surveys and analyses have been made over most of the world's oceans. In the North Atlantic, analyses in terms of sea-floor spreading include those of Matthews and Williams (1968), Vogt et al. (1970, 1971), Rona et al. (1970), Vogt and Johnson (1971), Williams and McKenzie (1971), Pitman et al. (1971), Pitman and Talwani (1972), and Srivastava (1978). Sea-floor spreading in the Indian Ocean since the Late Cretaceous has been analyzed by McKenzie and Sclater (1971), Weissel and Hayes (1972), Sclater and Fisher (1974), and Norton and Sclater (1979) and reviewed by Schlich (1982). Recent analyses of sea-floor spreading between Australia and Antarctica include those by Cande and Mutter (1982), Veevers et al. (1990), and Tikku and Cande (1999). Sea-floor spreading between Madagascar and Africa has been analyzed by Rabinowitz et al. (1983). The Pacific Ocean has been investigated in most detail and significant analyses include those of Hayes and Heirtzler (1968), Pitman and Hayes (1968), Atwater

Fig. 5.19. (a) Isochron map of the ocean floor on a color-shaded relief map illuminated from the northwest. After Müller et al. (1997).

(1970), Atwater and Menard (1970), Hayes and Pitman (1970), Larson and Chase (1972), Larson et al. (1972), Christoffel and Falconer (1972), Molnar et al. (1975), Larson and Hilde (1975), Hilde et al. (1976), and Weissel et al. (1977). A summary of the coverage of the Pacific Ocean from shipboard magnetic surveys is given by Isezaki (1988). Cande et al. (1989) provide a valuable summary in their map of magnetic lineations of the world's ocean basins, together with a comprehensive list of references.

The oldest magnetic anomalies occur in the western Pacific, the western North Atlantic, and off northwest Africa and are referred to as the M sequence anomalies (Larson and Chase, 1972; see §5.3.2). Larson and Pitman (1972) were the first to make a detailed worldwide correlation of these anomalies and establish their age. They realized that negative anomalies in the Pacific sequence correlated with positive anomalies in the North Atlantic because the Pacific anomalies were formed at or below the equator. Their reversal time scale for the Late Jurassic and Early Cretaceous involved anomalies M1-M22. In later studies of the western Pacific anomalies, Larson and Hilde (1975) added MO and M23-M25 to the time scale and Hilde et al. (1976) recognized M26. However, Cande et al. (1978) resolved this into M26-M28 and added M29 to the sequence. The oldest anomalies observed in the oceans are M30-M41; small-amplitude anomalies in the western Pacific Jurassic quiet zone delineated by Handschumacher et al. (1988) initially to M38 and later by Sager et al. (1998) to M41. The Jurassic quiet zone anomalies are discussed further in §5.3.3.

Using the reversal time scale established from analyzing marine magnetic anomalies, it is possible to construct detailed maps of the age of the ocean floor worldwide. One of the most significant developments in our understanding of the structure of the ocean floor has come from the use of satellite altimetry data. The surface of the ocean bulges outwards and inwards according to variations in the topography of the ocean floor. These variations can be measured using satellite radar altimetry with closely spaced profiles and can then be converted to grids of vertical gravity gradient and gravity anomalies (Haxby, 1987; Smith and Sandwell, 1994; Sandwell and Smith, 1997). Maps produced from these measurements show many tectonic features in the oceans with more detail than was previously known, especially in areas not well covered by ship surveys. By combining these data with the latest magnetic anomaly identifications Muller et al. (1997) compiled a digital age grid of the ocean floor using a self-consistent set of global isochrons and associated plate reconstruction poles.

Figure 5.19a shows the isochron map of Muller et al. (1997) of the ocean floor and Fig. 5.19b shows the corresponding spreading rates at each age. The

Fig. 5.19. (b) Spreading rate map (kindly provided by D. Müller) corresponding to the isochron map ofFig. 19a.

Was this article helpful?

0 0

Post a comment