Magnetostratigraphy

Once the geomagnetic reversal timescale has been calibrated, oceanic magnetic anomalies may be used to date oceanic lithosphere. The method has been progressively refined so that it is now possible to deduce ages back to mid-Jurassic times with an accuracy of a few million years.

The Vine-Matthews hypothesis explains the sequence of magnetic anomalies away from ocean ridges in terms of normal and reversed magnetizations of the oceanic crust acquired during polarity reversals of the geomagnetic field. Verification of the hypothesis was provided by the consistency of the implied reversal sequence with that observed independently on land. Cox et al. (1967) had measured the remanent magnetization of lavas from a series of land sites. The lavas were dated by a newly refined potassium-argon method, which allowed the construction of a reversal timescale back to 4.5 Ma. The timescale could not be extended to earlier ages, as the errors involved in K-Ar dating become too large. Similarly, polarity events of less than 50,000 years duration could not be resolved. The timescale to 5 Ma before present, as later refined by Cande & Kent (1992), is given in Fig. 4.8. In magnetostratigraphic terminology, polarity chrons are

50 0 50 km

Figure 4.7 Variation of the magnetic anomaly pattern with the direction of the profile at a fixed latitude. Magnetic inclination is 45° in all cases. No vertical exaggeration.

50 0 50 km

Figure 4.7 Variation of the magnetic anomaly pattern with the direction of the profile at a fixed latitude. Magnetic inclination is 45° in all cases. No vertical exaggeration.

Time Epoch (Ma)

Numerical Chrons

Polarity

Named Chrons and Subchrons

Time Epoch (Ma)

Numerical Chrons

Polarity

Named Chrons and Subchrons

Figure 4.8 Geomagnetic polarity timescale for the Plio-Pleistocene (modified from Cande & Kent, 1992, by permission of the American Geophysical Union. Copyright © 1992 American Geophysical Union). Numerical chrons are based on the numbered sequence of marine magnetic anomalies.

defined with durations of the order of 106 years. Chrons may be dominantly of reversed or normal polarity, or contain mixed events.

Further verification of the geomagnetic reversal tim-escale was provided by paleomagnetic investigations of deep sea cores (Opdyke et al., 1966). Unlike lava flows, these provide a continuous record, and permit accurate stratigraphic dating from their microfauna. This method is most conveniently applied to cores obtained in high magnetic latitudes where the geomagnetic inclination is high, because the cores are taken vertically and are not oriented azimuthally. Excellent correlation was found between these results and those from the lava sequences, and confirmed that at least 11 geomagnetic field reversals had occurred over the last 3.5 Ma. Subse quent work on other cores extended the reversal history back to 20 Ma (Opdyke et al., 1974).

Pitman & Heirtzler (1966) and Vine (1966) used the radiometrically dated reversal timescale to compute the magnetic profiles that would be expected close to the crestal regions of mid-ocean ridges. By varying the spreading rate it was possible to obtain very close simulations of all observed anomaly sequences (Fig. 4.9), and consequently to determine the spreading rates. A compilation of such rates is shown in Table 4.1. Extensions of this work show that the same sequence of magnetic anomalies, resulting from spreading and reversals of the Earth's magnetic field, can be observed over many ridge flanks (e.g. Fig. 4.10). Later work has shown that similar linear magnetic anomalies are developed

Reykjanes Ridge

South Atlantic

500 nT

Models

4 km

500 nT

500 nT

Juan De Fuca Ridge Profile reversed 46° N

Juan De Fuca Ridge Profile reversed 46° N

km 100

500 nT

NW Indian Ocean

500 nT

Model 15 mm a 1

r

l 50 -—

' 1

51

East Pacific Rise

Profile reversed 51° S

500 nT

km 100

East Pacific Rise

Profile reversed 51° S

500 nT

Figure 4.9 Magnetic anomaly profiles and models of several spreading centers in terms of the reversal timescale (redrawn from Vine, 1966, Science 154,1405-15, with permission from the AAAS).

Table 4.1 Spreading rates at mid-ocean ridges ("spreading rate"is defined as the accretion rate per ridge flank).

Ridge

Latitude

Observed rate (mm a ')

Predicted rate (mm a ')

Juan de Fuca

46.0°N

29

t

Gulf of California

23.4°N

25

24.7

Cocos -

Pacific

17.2°N

37

39.4

Pacific

3.1°N

67

65.4

Galapagos

2.3°N

22

22.0

Galapagos

3.3°N

34

34.6

Nazca -

Pacific

12.6°S

75

74.2

Chile Rise

43.4°S

31

30.2

Pacific -

Antarctic

35.6°S

50

49.5

Antarctic

51.0°S

44

44.6

Antarctic

65.3°S

26

29.0

North Atlantic

86.5°N

6

5.7

North Atlantic

60.2°N

9.5

9.2

North Atlantic

42.7°N

11.5

11.9

Central Atlantic

35.0°N

10.5

11.0

Central Atlantic

23.0°N

12.5

12.6

Cayman

18.0°N

7.5

5.9

South Atlantic

38.5°S

18

17.6

Antarctic -

South America

55.3°S

10

9.3

Africa -

Antarctic

44.2°S

8

7.4

Northwest Indian Ocean

4.2°N

14

14.6

Northwest Indian Ocean

12.0°S

18.5

17.9

Northwest Indian Ocean

24.5°S

25

24.5

Southeast Indian Ocean

25.8°S

28

28.8

Southeast Indian Ocean

50.0°S

38

37.3

Southeast Indian Ocean

62.4°S

34.5

33.7

Gulf of Aden

12.1°N

8

8.6

Gulf of Aden

14.6°N

12

12.1

Red Sea

18.0°N

10

8.2

Based on data from DeMets et al (1990) and Vine (1966). t Not available because Farallon plate is omitted from the model.

Based on data from DeMets et al (1990) and Vine (1966). t Not available because Farallon plate is omitted from the model.

1500_1000_500_0 km

1500_1000_500_0 km

Figure 4.10 Magnetic anomaly profile and model over the southern Mid-Atlantic Ridge (redrawn from Heirtzler et al., 1968, by permission of the American Geophysical Union. Copyright © 1968 American Geophysical Union).

over oceanic crust dating back to the Jurassic. Although there is no oceanic crust older than this, paleomagnetic investigations on land have shown that geomagnetic reversals have occurred at least back to 2.1 Ga.

That spreading rates have varied with time is apparent from an examination of magnetic profiles from different oceans. Examples are given in Fig. 4.11 in which the spreading rate in the South Atlantic is assumed to be constant and the distances to various magnetic anomalies from ridge crests in other oceans are plotted against the distance to the same anomaly in the South Atlantic. Inflection points in the curves for the other oceans indicate when the spreading rates changed there ifthe implicit assumption that the spreading rate has remained constant in the South Atlantic is correct. However, spreading rates may have changed with time in all oceans.

The first long-term geomagnetic timescale was constructed by Heirtzler et al. (1968). Again they made the assumption that spreading in the South Atlantic had remained constant at the same rate as had been deduced for the last 4 Ma. A model of normal and reversely magnetized blocks was constructed which simulated the observed anomaly pattern, and the distance axis transformed into a geomagnetic timescale of reversals extending back in time nearly 80 Ma. Prominent anomalies corresponding to periods of normal polarity were numbered from 1 to 32 with increasing time (Fig. 4.10).

Leg 3 of the Deep Sea Drilling Program (DSDP), in 1968, was specifically designed to test the hypothesis of sea floor spreading and the assumption of a constant rate of spreading in the South Atlantic (Maxwell et al., 1970). A series of holes was drilled in the South Atlantic along a traverse at right angles to the Mid-Atlantic Ridge (Fig. 4.12a). The age of the oceanic crust would ideally have been determined by radiometric dating of the layer 2 basalts that were penetrated in each hole. However the basalts were too weathered for this to be possible, and so their ages were determined, albeit slightly underestimated, by paleontologic dating of the basal sediments of layer 1. In Fig. 4.12b oldest sediment age is plotted against distance from the ridge axis, and it is readily apparent that there is a remarkable linear relationship, with crustal age increasing with distance from the ridge. The predicted ages imply a half spreading rate in this region of 20 mm a-1, as predicted, and hence agree well with the age of the ocean floor and the reversal times-cale proposed by Heirtzler et al. (1968) (Fig. 4.10).

A thorough review of the calibration of this polarity timescale was carried out by Cande & Kent (1992, 1995). It drew on oceanic magnetic anomaly data, magnetostratigraphic studies of sedimentary sequences on land and at sea, and radiometric dating of nine specific stratigraphic horizons. From this they concluded that sea floor spreading in the South Atlantic had been continuous, with some variation about an essentially constant rate, and that it was still appropriate to use the South Atlantic as a standard against which the spreading history in the other ocean basins could be compared. The revised timescale for the past 80 Ma suggested by Cande & Kent (1995) is illustrated in Fig. 4.13.

Distance from ridge x 10~2 (km)

Figure 4.11 Relationship between the distance to a given anomaly in the South Atlantic and the distance to the same anomaly in the South Indian, North Pacific and South Pacific Oceans. Numbers on the right refer to magnetic anomaly numbers (redrawn from Heirtzler et al., 1968, by permission of the American Geophysical Union. Copyright © 1968 American Geophysical Union).

Distance from ridge x 10~2 (km)

Figure 4.11 Relationship between the distance to a given anomaly in the South Atlantic and the distance to the same anomaly in the South Indian, North Pacific and South Pacific Oceans. Numbers on the right refer to magnetic anomaly numbers (redrawn from Heirtzler et al., 1968, by permission of the American Geophysical Union. Copyright © 1968 American Geophysical Union).

100° 80° 60° 40° 20° 0° 20° 40°

0 400 800 1200 1600

Distance from ridge crest (km)

Figure 4.12 (a) Location map of drilling sites on Leg 3 of the DSDP in the South Atlantic. (b) Relationship between greatest sediment age and distance from the Mid-Atlantic Ridge crest (after Maxwell et al., 1970, Science 168,1047-59, with permission from the AAAS).

0 400 800 1200 1600

Distance from ridge crest (km)

Figure 4.12 (a) Location map of drilling sites on Leg 3 of the DSDP in the South Atlantic. (b) Relationship between greatest sediment age and distance from the Mid-Atlantic Ridge crest (after Maxwell et al., 1970, Science 168,1047-59, with permission from the AAAS).

The discovery of Larson & Pitman (1972) of older magnetic anomalies in three regions of the western Pacific allowed the Heirtzler geomagnetic timescale to be extended back to 160 Ma. Lineations ofsimilar pattern were also found in the Atlantic. The timescale was extended by assuming a constant spreading rate in the Pacific, calibrated by DSDP sites in the Pacific and Atlantic. The longer periods of reversed polarity in this sequence are numbered M0 to M28 (M representing Mesozoic). It appears that spreading in the major ocean basins has been continuous as all polarity events are present, although the rate of spreading has varied.

The version of the reversal timescale to 160 Ma shown in Fig. 4.13 combines the timescale of Cande & Kent (1995), for the Late Cretaceous and Cenozoic (anomalies 1-34), with that of Kent & Gradstein (1986) for the Early Cretaceous and Late Jurassic (anomalies M0-M28).

0 0

Post a comment