Info

FIGURE 4.3 Paleomagnetic polarity timescale for the last 6 million years. Normal polarity periods in black. Dates are based on K/Ar dates on lava flows (Cande and Kent, 1995; Berggren et al., 1995).

orbital forcing is now so well-established that it has been used to refine age estimates on the timing of paleomagnetic reversals (Shackleton et al., 1990; Bassinot et al., 1994; Tauxe et al., 1996). Indeed, astrochronological estimates for the ages of magnetic reversals are now considered by some to provide the best framework for the late Quaternary geomagnetic timescale (Renne et al., 1994; Cande and Kent, 1995; Berggren et al., 1995).

4.1.4 Geomagnetic Excursions

In addition to polarity epochs and events (chrons and subchrons) there have been many reports of short-term geomagnetic fluctuations known as polarity excursions

(or "cryptochrons"; Cande and Kent, 1992). Excursions are considered to last for a few thousand years (at most) and differ from events in that a fully reversed field is generally not observed, perhaps because the partial reversal is due to variations in the non-dipole component of the field. It is not yet clear whether excursions are "abortive reversals" or whether they represent normal geomagnetic behavior during a polarity epoch. In this regard it is interesting that Cox (1969) in a study of the spectrum of known reversal frequency formulated a statistical model of geomagnetic field behavior from which he inferred the existence of short events or excursions during the last 10 million yr, which were at that time undiscovered. Nevertheless, a great deal of controversy surrounds geomagnetic excursions and their usefulness in paleoclimatic studies is quite limited. Because of their short duration, they are only likely to be recorded in areas with high sedimentation rates and/or very frequent lava flows. Hence, it is often the case that an excursion apparently found in one location may not be registered nearby, where sedimentation rates at the critical time may not have been sufficient to record it. Such occurrences are common and lead to skepticism about the reality and significance of reported excursions (Verosub and Banerjee 1977; Lund and Banerjee, 1979). For an excursion to warrant recognition it should be based on observations of synchronous changes in both declination and inclination in several geographically separate cores, and the analysis should be restricted to fine-grained, homogeneous sediment. However, even if the individual excursions reported are correct, their absence from other sites in the same region mean that they are rarely of value in constructing regional chronologies or in magnetostratigraphic correlation.

In a recent review, Opdyke and Channell (1996) conclude that there were five times in the late Quaternary when there is credible evidence for excursions. These are named after the region where the record was first recognized: Mono Lake (27,00028,000 yr B.P.); Laschamp (-42,000 yr B.P.); Blake (108,000-112,000 yr B.P.); Pringle Falls (218 ±10 ka B.P.); and Big Lost (-565 ka B.P. ). Of these, the Blake excursion is the one most widely recorded, having been recognized in Chinese loess, as well as in marine sediments from the Caribbean, Atlantic, and Mediterranean. The Laschamp excursion has been recorded in Icelandic lava flows as well as in the Massif Central, France, where it was originally recognized (Condomes et al., 1982; Levi et al., 1990). The other excursions are all from the western United States where they have been seen in some (but not all!) lake sediments.

4.1.5 Secular Variations of the Earth's Magnetic Field

A number of studies of well-dated lake sediments from different parts of the world indicate that quasi-periodic changes in declination (and to a lesser extent inclination) have occurred during the Holocene (Mackereth, 1971; Verosub, 1988). These changes are of a smaller magnitude than those described as excursions and appear to be regional in extent (over distances of 1000-3000 km) presumably because they result from changes in the non-dipole component of the Earth's magnetic field.

If the changes observed can be accurately dated in one or more cores, it should be possible to construct a "master chronology" that would enable peaks and troughs in the declination record to be used as a chronostratigraphic template. Such a template would be valuable in estimating the age of highly inorganic sediments, which do not yield enough carbon for conventional 14C dating, but which do contain a clear record of declination and inclination variations. However, it is first necessary to establish a reliable, well-dated magnetostratigraphic record for a "type-section" and to determine over what area this record can be considered to provide a master chronology. So far, this has been demonstrated for western Europe (Thompson, 1977) and the western United States (Verosub, 1988; Negrini and Davis, 1992). Secular paleomagnetic chronologies have also been proposed for other areas (Gillen and Evans, 1989; Ridge et al., 1990). Undated sediments should have approximately the same sedimentation rate as that of the master chronology and not have experienced any significant disturbances, but even these problems may not be critical. Figure 4.4 shows the master declination and inclination chronology from Pleistocene Lake Russell (now Mono Lake, east-central California) for the period 12.5-30 ka B.P. Of particular note is the large excursion around 29-27 ka B.P. (named the Mono Lake geomagnetic excursion) which has been seen in other records from the region (Levi and Karlin, 1989). The chronology of the Lake Russell record was determined by 17 14C dates, but other

Lake Russell Inclination (Wilson Creek Locality)

Lake Russell Declination [Wilson Creek Locality)

Lake ChewaucanDeclination (Summer Lake E Locality)

Lake Russell Inclination (Wilson Creek Locality)

Lake ChewaucanDeclination (Summer Lake E Locality)

FIGURE 4.4 The master secular geomagnetic chronology from Pleistocene Lake Russell sediments (extreme left and right columns) and the correlated geomagnetic record from Lake Chewaucan, Oregon. Apparent episodes of nondeposition are shaded.The HC-dated master chronology provides a chronological template within which the undated sediments from Lake Chewaucan can be placed (Negrini and Davis, 1992).

Lake Russell Declination [Wilson Creek Locality)

—--

-\——

HE

—I— \r

_____

FIGURE 4.4 The master secular geomagnetic chronology from Pleistocene Lake Russell sediments (extreme left and right columns) and the correlated geomagnetic record from Lake Chewaucan, Oregon. Apparent episodes of nondeposition are shaded.The HC-dated master chronology provides a chronological template within which the undated sediments from Lake Chewaucan can be placed (Negrini and Davis, 1992).

lakes in the region have far less organic material; by using the Lake Russell secular geomagnetic record as a regional signal, other sedimentary records can be dated by correlating sequences of characteristic features. In this way, the secular paleomagnetic record from Pleistocene Lake Chewaucan (500 km north of Lake Russell) was mapped onto the master chronology, to fix the sequence in time (Fig. 4.4). In doing so, a number of periods of non-deposition were assumed to have occurred (with good supporting sedimentological evidence bracketing these apparent gaps). The chronological model for Lake Chewaucan, so defined, can be tested further because tephras embedded within the sediments have been found elsewhere in more organic-rich environments; preliminary results from 14C dates associated with these tephras generally support the proposed geomagnetic timescale applied to the Lake Chewaucan sediments (Negrini and Davis, 1992).

Clearly, using secular variations from one record to date another can only be as accurate as the initial chronological control on the master record. Providing correlations between the master record and the undated target record are statistically significant, and changes in both inclination and declination support the match, it should be possible to build up regional templates that will be invaluable in many areas.

0 0

Post a comment