Precise ages of the Réunion event and Huckleberry Ridge excursion: Episodic clustering of geomagnetic instabilities and the dynamics of flow within the outer core
Introduction
Poorly understood physical processes within earth's core cause the geomagnetic field to undergo reversals of polarity, excursions, or events. The timing of these phenomena, particularly the number and sequence of short-lived excursions or events relative to complete polarity reversals, has long been central to debate about conditions and processes originating either in deep or shallow regions of the core that may modulate geomagnetic field stability on time scales of (e.g., McFadden and Merrill, 1986, Gubbins, 1999, Glatzmaier et al., 1999, Hoffman and Singer, 2008, Davies and Gubbins, 2011). The dramatic improvement during the last two decades in the radioisotopic dating of many short-lived excursions and events has led to a nascent geomagnetic instability time scale (GITS) for the Quaternary (Singer et al., 1999, Singer, 2014). Notably, this chronology has revealed a temporal clustering of excursions during the early (730–530 ka) and late (212–17 ka) portions of the Brunhes chron (Singer et al., 2008a, Singer et al., 2008b; Singer, 2014). Here, we present new geochronologic data from volcanic rocks that, together with published data from marine sediments, record a sequence of excursions, reversals, and events during 125 kyr of the early portion of the Matuyama chron. The chronology strongly suggests that dynamo instability is temporally modulated by a process that operates on a 100–200 kyr time scale. We offer one possible explanation for directional instability in the axial dipole component of the geomagnetic field for such periods of time that is based on thermodynamic analysis and modeling the effects of crystallization of the mush zone atop the outer core (Braginsky, 1963, Moffatt and Loper, 1994).
Section snippets
Previous work on the Réunion event
There is a half-century-long history related to paleomagnetic investigations on Réunion Island. The first paleomagnetic and K–Ar dating experiments on Réunion lavas revealed a normal-to-reversed polarity transition at (Chamalaun and McDougall, 1966; K–Ar ages are calculated using the updated decay constants of Steiger and Jaeger (1977). Although initially considered to be part of the Olduvai subchron (Grommé and Hay, 1971), this distinction was later changed to the “Réunion normal
Magnetic measurements
In parallel to new geochronologic experiments, new paleomagnetic directions were determined through alternating field demagnetization of standard specimens from the more than 100 lava flows previously studied from three distinct sections on Réunion Island (RSD, GC, and LM) and combined with the original dataset of Baksi and Hoffman (2000). All measurements, new and old, were conducted at California Polytechnic State University, San Luis Obispo. The primary remanence directions were identified
Results
New 40Ar/39Ar ages were obtained for eight of the lava flows in the three sections previously analyzed on Réunion Island by Baksi et al. (1993) and Baksi and Hoffman (2000) (Table 1; Fig. 2; Complete argon data in Table A.3). Lavas of the RSD, GC, and LM sections record R–N–T, N–T–N, and N–T–R field behavior, respectively (Fig. 3). For most lavas, demagnetization behavior was typical for Pleistocene basalts, the process having removed the only significant secondary component of magnetization,
Untangling geomagnetic field behavior between 2.2 and 2.0 Ma
The duration of the field behavior recorded in the RSD/Réunion lavas is difficult to evaluate as lavas with normal, reversed and transitional polarity are bounded by statistically indistinguishable ages. Nevertheless, the level of uncertainty indicates a maximum duration of . Ohno et al. (2012) observed a number of lows in relative paleointensity between 2.8 and 2.1 Ma in a sedimentary record from the North Atlantic. These lows were accompanied by directional changes that lasted between
New vs. published ages for the Huckleberry Ridge Tuff
Our new 40Ar/39Ar age for the HRT member B is within error of the 40Ar/39Ar data of Gansecki et al. (1998), Rivera et al. (2014), and Lanphere et al. (2002). However, the Lanphere et al. (2002) age is based on sanidine total fusions and is older than the weighted mean of our plateau ages obtained using the Noblesse five collector mass spectrometer. This is not surprising given the presence of antecrysts in the HRT member B, coupled with the fact that Lanphere et al. (2002) performed
Implications for the geodynamo and beyond
Singer et al., 2008a, Singer et al., 2008b proposed that there are two periods during the Brunhes chron in which the dynamo is relatively weak, each lasting about 200 kyr and encompassing at least five excursions. These two periods are separated by during which there is little compelling evidence for excursional behavior (Singer, 2014). Singer et al., 2008a, Singer et al., 2008b hypothesized that the pattern of fluid flow within the outer core may be such that there is a 200–300 kyr
Conclusions
Based on new 40Ar/39Ar age determinations from Réunion Island, Cerro Fraile in Argentina, and Gamarri in Ethiopia, plus equivalent U–Pb zircon and 40Ar/39Ar sanidine ages for the Huckleberry Ridge Tuff, we conclude that at least two field excursions—one designated as an event having recorded full normal polarity on Réunion Island—and two polarity reversals bounding a normal subchron, occurred between 2.20 and 2.07 Ma. Specifically, these include: (1) the Réunion event at 2.200 Ma, (2) the much
Acknowledgments
We are grateful for financial support from U.S. NSF grants EAR-1250446, EAR-0943584, and EAR-0943770. Condon was supported by the European Community's Seventh Framework Programme (FP7/2007–2013) grant agreement 215458 (GTSnext) and NIGFSC award IP/1011/1107. The Shell Corporation sponsored the Senior Honors thesis research of Macho at UW–Madison. Brown was partially supported by German DFG grant SPP 1488.
References (62)
Comment on “40Ar/39Ar dating of the tephra layer in the Pliocene Seneze maar lacustrine sequence (French Massif Central): constraint on the age of the Réunion–Matuyama transition and implications on paleoenvironmental archives” by Roger et al.
Earth Planet. Sci. Lett.
(2001)- et al.
Testing the accuracy of the geomagnetic polarity time-scale (GPTS) at 2–5 Ma, utilizing 40Ar/39Ar incremental heating data on whole-rock basalts
Earth Planet. Sci. Lett.
(1993) - et al.
The Réunion Subchronozone at ODP site 981 (Feni Drift, North Atlantic)
Earth Planet. Sci. Lett.
(2003) - et al.
Temporal dissection of the Huckleberry Ridge Tuff using the 40Ar/39Ar dating technique
Quat. Geochronol.
(2012) - et al.
Chronology of late Pliocene and early Pleistocene glacial and magnetic events in southern Argentina
Earth Planet. Sci. Lett.
(1972) - et al.
Geomagnetic polarity epochs: age and duration of the Olduvai normal polarity event
Earth Planet. Sci. Lett.
(1971) - et al.
Palaeomagnetic analysis of the Sterkfontein palaeocave deposits; age implications for the hominin fossils and stone tool industries
J. Hum. Evol.
(2011) - et al.
The upper Olduvai geomagnetic field reversal from Death Valley, California: a fold test of transitional directions
Earth Planet. Sci. Lett.
(1995) - et al.
Geomagnetic excursions
Treatise Geophys.
(2007) Zircon U–Pb chemical abrasion (“CA-TIMS”) method: combined annealing and multi-step partial dissolution analysis for improved precision and accuracy of zircon ages
Chem. Geol.
(2005)