Absolute geomagnetic paleointensity as recorded by ~1.09 Ga Lake Shore Traps (Keweenaw Peninsula, Michigan) EVGENIY V. KULAKOV, ALEKSEY V. SMIRNOV AND JIMMY F. DIEHL Department of Geological and Mining Engineering and Sciences, Michigan Technological University, 630 Dow ESE Building, 1400 Townsend Drive, Houghton, MI 49931, USA (
[email protected],
[email protected],
[email protected]) Received: January 18, 2013; Revised: March 11, 2013; Accepted: April 29, 2013
ABSTRACT Absolute geomagnetic paleointensity measurements were made on 255 samples from 38 lava flows of the ~1.09 Ga Lake Shore Traps exposed on the Keweenaw Peninsula (Michigan, USA). Samples from the lava flows yield a well-defined characteristic remanent magnetization (ChRM) component within a ~375°C590°C unblocking temperature range. Detailed rock magnetic analyses indicate that the ChRM is carried by nearly stoichiometric pseudo-single-domain magnetite and/or low-Ti titanomagnetite. Scanning electron microscopy reveals that the (titano)magnetite is present in the form of fine intergrowths with ilmenite, formed by oxyexsolution during initial cooling. Paleointensity values were determined using the Thellier double-heating method supplemented by low-temperature demagnetization in order to reduce the effect of magnetic remanence carried by large pseudosingle-domain and multidomain grains. One hundred and two samples from twenty independent cooling units meet our paleointensity reliability criteria and yield consistent paleofield values with a mean value of 26.3 ± 4.7 T, which corresponds to a virtual dipole moment of 5.9 ± 1.1 1022 Am2. The mean and range of paleofield values are similar to those of the recent Earth’s magnetic field and incompatible with a “Proterozoic dipole low”. These results are consistent with a stable compositionally-driven geodynamo operating by the end of Mesoproterozoic.
1. INTRODUCTION Data on the long-term evolution and characteristics of Earth’s magnetic field derived from paleomagnetic investigations of Precambrian rocks provide crucial insights into the evolution of the geodynamo and the early history of our planet (e.g., Coe and Glatzmaier, 2006; Evans, 2006; Aubert et al., 2010; Tarduno et al., 2010; Smirnov et al., 2011). However, our knowledge of the Precambrian geomagnetic field remains very limited, especially with regard to the data on field strength (paleointensity) which represents the most challenging aspect of paleomagnetic research (e.g., Tarduno and Smirnov, 2004; Tauxe and Yamazaki, 2007). The Precambrian data comprise less than 5% of the total paleointensity database and are characterized by an uneven temporal and spatial distribution (e.g., Smirnov et al., 2003). Moreover, some of the paleofield values derived
Stud. Geophys. Geod., 57 (2013), 565584, DOI: 10.1007/s11200-013-0606-3 © 2013 Inst. Geophys. AS CR, Prague
565
E. Kulakov et al.
from Precambrian rocks are based on a limited number of cooling units, and some clearly reflect alteration processes (e.g., Yoshihara and Hamano, 2004). If modern reliability criteria are applied, the database becomes even scarcer and does not yet provide a reliable picture of the long-term changes of the Earth’s magnetic field strength in the Precambrian. Accumulation of additional paleointensity data from Precambrian rocks is therefore crucial to improve our understanding of the early Earth processes. The bulk of paleointensity determinations for the Phanerozoic are derived from extrusive rocks (mostly basaltic lavas) using variations of the Thellier double-heating method (Thellier and Thellier, 1959). However, many Precambrian extrusive sequences have been affected by weathering, deformation, and/or metamorphism, hindering the preservation and measurement of paleointensity signal using the Thellier approach. As an alternative, intrusive rocks such as mafic dikes and sills (e.g., Halls et al., 2004; Macouin et al., 2003, 2006; Shcherbakova et al., 2008) or even larger intrusions (e.g., Selkin et al., 2000; Yu and Dunlop, 2001) have been widely used for Precambrian paleointensity studies. Although the paleofield determinations from intrusive rocks may also be adversely affected by the effects of slow cooling (e.g., Selkin et al., 2000) or the presence of thermochemical remanent magnetization (e.g., Smirnov and Tarduno, 2005), these data presently constitute the majority of the Precambrian paleointensity database. When compared to its Phanerozoic counterpart, the Precambrian database is characterized by the abundance of low field intensities with virtual dipole moments (VDMs) typically weaker than 3 1022 Am2 (e.g., Selkin et al., 2000; Yu and Dunlop, 2001; Macouin et al., 2003, 2006; Celino et al., 2007). This observation led some authors to suggest that low field intensity could have been a general characteristic of the early field (e.g., Macouin et al., 2003). The possibility that low values characterized the Archean was discounted by studies of ~2.5 and 3.2 Ga units using single crystal analysis that documented field intensities similar to that of the modern field (Smirnov et al., 2003; Tarduno et al., 2007). However, the possibility of low field strength during the Proterozoic remained; this was called a “Proterozoic dipole low” by Biggin et al. (2009). Notwithstanding a dominance of low paleointensity values from Proterozoic rocks, a few high values exist. In particular, the highest reported VDM values were obtained from normally (11.56 ± 3.63 1022 Am2; N = 12) and reversely (10.92 ± 2.99 1022 Am2; N = 9) magnetized rocks of the ~1.1 Ga Midcontinent Rift (MCR) system exposed around the Lake Superior (Pesonen and Halls, 1983). Taken at face value, these values suggest high geomagnetic field intensities during the 1020 My period of the igneous activity associated with MCR, similar to those observed for the Cretaceous Normal Polarity Superchron (e.g., Tarduno et al., 2001). If confirmed, the high VDM values reported by Pesonen and Halls (1983) would have important implications for our understanding the early geomagnetic field behavior. However, the rock sequences investigated in that study were represented by no more than four independent cooling units and most paleointensity values were derived from only one specimen per cooling unit. In addition, most of determinations by Pesonen and Halls (1983) are based on the temperature intervals that included low temperature steps (sometimes as low as 20°C), which could have resulted in a significant paleointensity bias. Interestingly, low paleointensity values were reported from coeval MCR-related
566
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
dikes and a sill in Ontario (VDM ≈ 1.41 ± 0.33 1022 Am2; McArdle et al., 2004) and from the ~1140 Ma Abitibi dikes (VDM ≈ 1.36 ± 0.35 1022 Am2; Macouin et al., 2003). We feel that the deficiencies and discrepancies described above warrant paleointensity re-investigation of the Midcontinent Rift rocks using modern experimental approaches. In this paper, we report new Thellier paleointensity results obtained from the ~1.09 Ga lava flows of the Lake Shore Traps exposed on the Keweenaw Peninsula of Michigan. The paleointensity study is accompanied by analyses of magnetic mineralogy of the rocks. We also discuss the implications of our results for the Proterozoic geodynamo.
2. GEOLOGIC SETTINGS AND SAMPLING The Lake Shore Traps (LST) is a sequence of lava flows interbedded within the Copper Harbor Formation which crop out on the Keweenaw Peninsula in Michigan (Lane, 1911). The LST represent the last significant stage of the Midcontinent Rift magmatism (Van Schmus et al., 1982). A single lava flow from this sequence has been U-Pb dated at 1087 ± 1.6 Ma (Davis and Paces, 1990). The LST flows are subaerial basalt and basaltic andesite with thickness from 4.4 to 42.4 m, averaging 18.2 m (Paces and Bornhorst, 1985). The flows have amygdaloidal flow bottoms, massive flow interiors, and vesicular flow tops. Zeolite-grade alteration is common in permeable vesicular flow tops but is negligible or absent in flow interiors (Paces and Bornhorst, 1985). The Lake Shore Traps are divided by thick conglomerate layers into three major groups: the lower, middle, and outer LST. The lower LST succession is not exposed and is observed only from magnetic surveys and drill-hole data (White et al., 1951). The middle LST flows crop out at several locations along the northwest coastline of the Keweenaw Peninsula (Fig. 1). The most extensive outcrop of 31 flows is located at the eastern end of the Peninsula (Fig. 1, Sampling Location 1) where the flows dip to the north-east (22°) at an angle of approximately 24° (Diehl and Haig, 1994). At this location, the middle LST are subdivided into two lava flow packages by a 27 m thick conglomerate layer; these packages are hereafter referred to as the lower and upper middle LST. About ~30 km west from Location 1, ten upper middle LST lava flows are exposed on Silver Island and two lower middle LST lava flows crop out on the mainland, directly across from the island (Fig. 1, Sampling Location 2). The lava flows at this locations dip to the north-north-northwest (356°) at an angle of approximately 38° (Diehl et al., 2009). The number of exposed middle LST lava flows generally decreases westward. The westernmost exposure of the middle LST is represented by only two lava flows separated by a conglomerate layer (Halls and Palmer, 1981). Eight to fourteen lava flows of the outer LST, the youngest portion of the Lake Shore Traps, are exposed southwest of Seven Mile Point where they dip north-west (295°) at 25° dip angle (Fig. 1, Sampling Location 3). Northeast of this location, the outer LST can only be identified from magnetic survey data in the Lake Superior (White et al., 1951). Paleomagnetic analyses of the middle and outer LST (Diehl and Haig, 1994; Kulakov et al., Paleomagnetism of ~1.09 Ga Lake Shore Traps (Keweenaw Peninsula, Michigan): New results and implications, submitted to Can. J. Earth Sci., 2013) have resulted in a mean direction of characteristic remanent magnetization (ChRM), D = 283.3°, I = 34.6° ( 95 = 5.0°, N = 31), consistent with the paleomagnetic directions reported for other
Stud. Geophys. Geod., 57 (2013)
567
E. Kulakov et al.
Fig. 1. A simplified geologic map of the Keweenaw Peninsula showing the sampling locations for this study. SMP - Seven Mile Point.
MCR rocks (e.g., Symons et al., 1994). All studied LST flows yielded ChRM of normal polarity. The primary origin of ChRM has been established by baked contact and fold tests (Palmer et al., 1981; Diehl and Haig, 1994; Kulakov et al., 2013). Most samples manifest a two-component natural remanent magnetization (NRM) with the low temperature component removed by heating to temperatures not exceeding 350375°C. In all samples, the ChRM was well-defined and typically demagnetized by heating to 580590°C. A hematite component parallel to ChRM was also observed in a small number of samples. The only paleointentsity investigation of the Lake Shore Traps was conducted as a part of the broader study by Pesonen and Halls (1983). Four samples from a single middle LST lava flow and adjacent baked conglomerate pebbles were measured using the Thellier method modified by Coe (1967), yielding a mean paleofield intensity of 49 ± 17 T with corresponding VDM of 9.9 ± 2.9 1022 Am2 (Pesonen and Halls, 1983).
568
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
For our paleointensity and rock magnetic analyses, we collected samples of the middle and outer LST at three sampling locations described above (Fig. 1). Nine flows of the lower middle LST and 17 flows of the upper middle LST (LST samples) were sampled at Location 1 (47.43°N, 87.71°W). At Location 2 (47.46°N, 88.07°W), we collected samples from nine flows of the upper middle LST from Silver Island (SI samples) and two flows of the lower middle LST on the mainland (CS samples). Eight flows of the outer LST were sampled at Location 3 (47.34°N, 88.45°W). Five to eight core samples drilled with a water-cooled gasoline drill were taken from each lava flow. Orientation was done with a Brunton magnetic compass and a Pomeroy sun compass.
3. RESULTS 3.1. Rock-magnetism We conducted rock magnetic analyses of the middle and outer LST flows in order to determine their magnetic mineralogy and suitability for paleointensity experiments. Temperature dependences of low-field magnetic susceptibility, (T), were measured upon cycling from room temperature to 600°C (in Argon) using an AGICO MFK1-FA magnetic susceptibility meter equipped with a high-temperature furnace and a cryostat. The (T) curves were also measured during heating from 192°C to room temperature both before and after the high-temperature thermomagnetic runs. For the majority of middle LST samples, the (T) curves are reversible and reveal the presence of a magnetic phase with Curie temperatures in a range of 560°C to 585°C, indicating magnetite to low-Ti titanomagnetite as a magnetic carrier (Fig. 2ac). The presence of a characteristic peak at 153°C, associated with the Verwey transition (Verwey, 1939) further suggests the presence of nearly-stoichiometric magnetite. However, samples from several middle LST flows yield irreversible (T) curves with a bump at ~150400°C (Fig. 2d). On cooling, the bump disappears and Hopkinson peak becomes more expressed. In addition, the post-heating low-temperature run is characterized by a more expressed peak of Verwey transition, suggesting heating-induced formation of magnetite. This behavior can reflect temperature-induced unmixing of homogeneous titanomagnetite grains into a high-Ti and magnetite (low-Ti) phases, although other potential mechanisms have also been proposed (e.g., Kosterov and Prevot, 1998). The samples manifesting the irreversible (T) curves were not used for paleointensity experiments. In contrast to the middle LST, almost all measured samples from the outer LST manifested irreversible thermomagnetic behavior. Consequently, the outer LST group was excluded from further rock magnetic and paleointensity analyses. Magnetic hysteresis parameters (coercivity, Hc; coercivity of remanence, Hcr; saturation remanence, Mr; and saturation magnetization, Ms) were measured on at least two samples from each lava flow using a Model 2900 Princeton Measurement Corporation Alternating Gradient Field Magnetometer. The hysteresis measurements suggest a pseudo-single domain (PSD) magnetic carrier in all samples (Fig. 3ac). Most samples showed no or negligible difference in hysteresis behavior before and after temperature cycling to 600°C, indicating their magnetic stability during laboratory
Stud. Geophys. Geod., 57 (2013)
569
E. Kulakov et al.
Fig. 2. Typical dependences of low-field magnetic susceptibility ( ) versus temperature. Heating and cooling (T) curves are shown by solid and dashed lines, respectively. L1 and L2 show the low temperature runs before and after high-temperature run, respectively.
heating. Samples from one flow of the upper middle LST (LST16) with hematite component exhibited wasp-waisted hysteresis loops (Fig. 3d) indicating the presence of magnetic phases with different coercivities (Tauxe et al., 1996). The first-order reversal curve (FORC) diagrams were measured on eight samples representing all four lava flow groups (two samples per group) used for paleointensity analyses (i.e. the lower and upper middle LST sampled at Locations 1 and 2). The FORC diagrams are consistent with small PSD grains (Fig. 4) (e.g., Roberts et al., 2000). All FORC distributions measured from our LST samples meet the acceptance criteria for paleointensity determinations formulated by Carvallo et al. (2006): (1) the full width at half maximum (FWHM) of a FORC distribution must not exceed 29 mT; (2) the full width (FW) along the vertical axis ( H c 0 ) must not exceed 132 mT; and (3) the bulk coercivity Hc must be greater than 5.4 mT.
570
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
Magnetic Moment [a.u.]
a)
CS1-4
Mr/Ms=0.284 Hc=195.5 Oe Hcr=530.3 Oe
-5
Magnetic Moment [a.u.]
d)
Applied Field [kOe] LST16-C
Mr/Ms=0.306 Hc=262.0 Oe Hcr=4.2 kOe
-5
Applied Field [kOe]
b)
LST13-B
Mr/Ms=0.162 Hc=129.1 Oe Hcr=344.2 Oe
5 -5
e)
Applied Field [kOe] SI9-9
Mr/Ms=0.493 Hc=610.7 Oe Hcr=1.1 kOe
5 -5
Applied Field [kOe]
c)
SI7-3
Mr/Ms=0.159 Hc=145.5 Oe Hcr=369.8 Oe
5 -5
f)
Applied Field [kOe]
5
SI6-2
Mr/Ms=0.380 Hc=460.5 Oe Hcr=1.26 kOe
5 -5
Applied Field [kOe]
5
Fig. 3. a)c) Typical magnetic hysteresis loops after paramagnetic slope correction measured at room temperature before (solid line) and after (dashed line) heating to 600C. d)f) Examples of magnetic hysteresis loops for samples that contain hematite. Mr - remanent magnetization, Ms - magnetization of saturation, Hc - coercive force; Hcr - coercivity of remanence.
Fig. 4. First-order reversal curve (FORC) diagrams measured for samples from a) the lower and b) upper middle Lake Shore Traps. The bar legends next to the FORC diagrams show the density of FORC distribution. The smoothing factor is 5. Insets show the parameters used to assess the suitability of these samples for paleointensity determinations (Carvallo et al., 2006). FORC data were analyzed using the FORCinel software (Harrison and Feinberg, 2008).
Stud. Geophys. Geod., 57 (2013)
571
E. Kulakov et al.
We also examined the opaque mineralogy of our samples using a model JEOL JSM6400 scanning electron microscope (SEM) equipped with an energy dispersive spectra (EDS) detector. The SEM analyses were conducted on two representative samples from each of the four lava flow groups used for paleointensity analyses (i.e. the lower and upper middle LST sampled at Locations 1 and 2). Backscattered electron imaging was used to identify oxide grains which varied in size from over a hundred micron to less than five micron (Fig. 5ad). Many grains containing one or several subordinate sets of trellis type lamellae (Haggerty, 1991) (Fig. 5ac). In some cases, no or less regular intergrowths of two mineral phases seen as brighter and darker areas were observed (e.g. Fig. 5d). The composition of the oxide grains was determined by means of energy dispersive spectrometry. The spectra were measured at a 20 kV accelerating voltage, which is optimal for excitation of the Fe K shell. The EDS analyses showed that the bright areas in the oxide grains separated by lamellae represent an iron oxide phase with very low titanium content (Fig. 5e). The EDS measured from the darker phase (lamellae) are consistent with nearly-ilmenite composition (Fig. 5f). We note that an exact measurement of the iron and titanium content was not possible because of the relatively large interaction volume of the electron beam (~2 m). As a result, the EDS for both Fe-rich and Ti-rich phases may include a “contamination signal” from the surrounding and/or underlying regions of the opposite phase. Based on our thermomagnetic analyses, we interpret the Ferich phase as magnetite or low-Ti magnetite. No hematite, rutile, or pseudobrookite was identified by the SEM/EDS analyses. No ilmenite other than in the intergrowths with the Fe-rich phase was observed. Altogether, our SEM and rock magnetic analyses suggest that the middle LST lavas underwent high-temperature oxidation no higher that stage C3 (abundant ilmenite lamellae; equilibrium two-phase intergrowths) of Wilson and Watkins (1967) and Haggerty (1991). The characteristic size of the magnetite regions intergrown with ilmenite within oxide grain mostly falls between 0.5 to 10 m, which is fully consistent with the PSD behavior observed by our hysteresis analyses. However, we cannot rule out that some of these regions may be small enough to behave as single-domain (SD) grains. On the other hand, homogeneous titanomagnetite grains larger than 1020 m which should manifest the hysteresis characteristics of multidomain grains have also been observed. We note that the presence of both SD and MD magnetic regions or grains in our samples is consistent with the observed PSD behavior (e.g., Dunlop and Özdemir, 1997). 3.2. Paleointensity determinations Paleointensity experiments were performed on 255 samples (38 flows) using the stepwise double heating Thellier method (Thellier and Thellier, 1959) modified by Coe (1967). Magnetic remanence was measured with a 2G Enterprises 760-R Superconducting Rock Magnetometer housed in magnetically shielded environment. The samples were heated in an ASC TD-48SC thermal specimen demagnetizer with controlled atmosphere chamber. The specimens were always placed in the furnace at exactly the same location and with the same orientation relative to the applied magnetic field. A laboratory field of 50 T was used for in-field steps to impart partial thermal remanent magnetizations (pTRMs). After measurement of NRM from each sample, the first heating in our Thellier
572
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
Fig. 5. a)d) Backscattered electron (BSE) images of typical magnetic grains in the middle LST lava flows. Gray and dark gray grains are the silicate matrix minerals. The magnetic grains shown in a)c) exhibit the intergrowth of ilmenite lamellae (darker areas) and magnetite or low-Ti titanomagnetite (lighter areas). d) Large magnetite grain with no ilminite lamellae intergroeth. e) Typical energy dispersive spectra from the low-Ti and f) high-Ti phases in Sample CS1 from a).
Stud. Geophys. Geod., 57 (2013)
573
E. Kulakov et al.
experiments was done to 375°C to eliminate the low-temperature component of NRM in a single step. The temperature increments were chosen to cover the ChRM component carried by magnetite and low-Ti titanomagnetite (Kulakov et al., 2013) with the sufficient number of heating steps (375525C range, 25C increment; 525540; 540590C range, 10C increment). In order to reduce the effect of magnetic remanence carried by large PSD and multidomain grains, samples were subjected to three low-temperature demagnetizations (LTD) in liquid nitrogen after each heating (e.g., Schmidt, 1993; Celino et al., 2007). The magnetic remanence was measured both before and after LTD. After the experiment, the measured data were plotted on an Arai (NRM-lost versus pTRM-gained) plot (Nagata et al., 1963) (Fig. 6). To monitor possible alteration during the paleointensity experiment, pTRM checks were performed for temperatures above 425°C; after every second zero-field temperature step, an in-field step at a lower temperature was measured. The pTRM checks which fell within 10% of the original TRM value were judged successful. In addition, we used the following reliability criteria: (1) The linear segment on the Arai plot used to calculate the paleofield is based on at least 5 data points and represents at least 40% of natural remanent magnetization (fraction of NRM, f); (2) A paleointensity value for a site is based on successful determinations from at least two samples; (3) The quality factor q (Coe at al., 1978) is five or greater; (4) The directional data of the zero-field steps must have a maximum angle of deviation (MAD) less than 10°. Overall, paleointensity determinations from 102 samples representing 20 independent cooling units were accepted based on our reliability criteria. The paleofield directions defined by demagnetization of NRM after zero-field heatings are characterized by a linear decay to origin on the vector end-point diagrams and are consistent with the directions obtained in preceding paleomagnetic studies of the middle LST (Diehl and Haig, 1994; Kulakov et al., 2013) (Fig. 6). The majority of accepted samples manifest similar behavior on Arai plots with concave-up shape of the low-temperature segment of the plot (375°C to 475500°C) and linear (~475500°C to 590°C) high-temperature segment (Fig. 6). The low-temperature concave-up segment predominantly reflects the signal from large PSD or even MD grains, whereas the high-temperature segment reflects the contributions from the entire ensemble including small PSD and nearly SD grains (e.g., Xu and Dunlop, 2004). The low temperature demagnetization aims to remove the part of the remanence carried by large PSD and MD grains that may bias paleointensity determinations towards lower values. On cycling through the Verwey transition, magnetite experiences sudden changes in magnetocrystalline anisotropy (Bickford, 1950; Bickford et al., 1957). These changes result in removal of part of remanence associated with MD grains due to destruction of ‘soft’ domains structures (Muxworthy and McClelland, 2000). The remaining magnetization is carried by SD, small PSD, and “hard” domain structures. The remanence carried by such grains is reversible or nearly reversible upon cycling through the Verwey transition (Carter-Stiglitz et al., 2004). In our samples, low temperature treatments result in removal of part of NRM at low temperature range (375°C to 475500°C) (Fig. 6), indicating that at least some portion of ChRM is carried by MD grains after the secondary NRM component is eliminated. At higher temperatures, this effect on NRM disappears and
574
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
Fig. 6. Examples of Thellier paleointensity determinations on the lava flows from the middle Lake Shore Traps for samples CS2-1 (a,b), SI3-7 (c,d), and LST20-C (e,f). (a,c,e) Natural remanence magnetization (NRM) lost versus partial thermoremanent magnetization (pTRM) gained. Open and closed circles show the data measured before and after low temperature demagnetization (see text), respectively. Triangles are pTRM checks. (b,d,f) Orthogonal vector plots of field-off steps (vertical projection of NRM, open circles; horizontal projection of NRM, closed circles).
Stud. Geophys. Geod., 57 (2013)
575
E. Kulakov et al.
LTDs change only pTRMs. For many samples, the fraction of pTRM removed by LTDs increases with increasing temperature (Fig. 6bd) suggesting that the large magnetic particles acquire magnetization throughout the entire blocking temperature spectrum. For some samples, low-temperature treatment influence is less significant and nearly equal for all temperatures (Fig. 6a). Magnetization in these samples is blocked by MD grains at approximately 475500°C. At higher temperatures, SD and small PSD grains are responsible for TRM and the MD contribution is negligible. For our paleointensity determinations, we used high-temperature part of the Arai plot that corresponds to NRMpTRM pairs of SD and SD-like magnetic grains. Samples from nine lava flows of the lower middle LST resulted in paleointensity values ranging from 17.2 T to 36.8 T with the mean of 25.2 ± 4.0 T (N = 9; VDM = 5.7 ± 0.9 1022 Am2) (Table 1 in the Appendix). However, the paleomagnetic directions from six flows (LST1 to LST 6) are serially correlated likely recording the same field value (Kulakov et al., 2013). Combining the paleofield values obtained from these flows results in a slightly higher paleointensity mean of 28.5 ± 4.8 T (N = 4; VDM = 6.5 ± 1.1 1022 Am2) (Table 1). Successful paleointensity determinations were obtained from 21 lava flows of the upper middle LST with the paleofield values ranging from 16.7 to 33.1 T with the mean of 24.9 ± 4.3 T (N = 21; VDM = 5.6 ± ± 1.0 1022 Am2) (Table 1). When the serial correlation of paleomagnetic directions is taken into account, the overall mean becomes 25.0 ± 4.5 T (N = 16, VDM = 5.7 ± ± 1.0 1022 Am2). The total paleofield value calculated from both LST sequences is 26.3 ± 4.7 T (N = 20, VDM = 5.9 ± 1.1 1022 Am2) (Table 1). To estimate whether the mean paleointensity value represents the time-averaged geomagnetic field, we compared the angular dispersion (S) of virtual geomagnetic poles (VGP) calculated from all independent paleomagnetic mean directions of the middle LST (Sall) with that calculated from the subset of flows that yielded accepted paleointensities (SPI). The S values were calculated using
S2
1 N 2 i , N 1 i 1
where N is the number of individual VGPs and Δi is the angle between the i-th VGP and the mean pole. Confidence intervals on the dispersion values were estimated using the N 1 jackknife method (Efron, 1982). The values of Sall = 10.5 ± 1.0° (N = 25) and SPI = 10.1 ± 1.3° (N = 20) are statistically indistinguishable indicating that the mean paleointensity faithfully represents the field recorded by the entire middle LST sequence. The calculated S values are lower than the values of 1416° suggested for ~20° latitude by the data for the last 5 My (Johnson et al., 2008). We note, however, that low VGP dispersions were also found in other rocks associated with the MCR (e.g., Tauxe and Kodama, 2009), indicating that the geomagnetic field could have been characterized by a lower rate of secular variation during the Proterozoic (e.g., Smirnov et al., 2011). A difference in cooling rate between the natural remanence and laboratory pTRM acquisitions may result in an overestimation (e.g., Dodson and McClelland-Brown, 1980; Halgedahl et al., 1980) or underestimation (McClelland-Brown, 1984) of paleointensity
576
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
for rocks containing SD or MD magnetic carriers, respectively. We used a conductive cooling model of Jaeger (1968) to calculate the time required for lava flows with thickness of 4 and 42 m (the thickness range of LST flows) emplaced over a surface with ambient temperature to cool through the blocking temperature range (590°C to 500°C). The calculated cooling times for the flow centers are about 1 month and 2.8 years, respectively. If the paleointensity signal in our samples was carried by elongate SD grains, our Thellier data could overestimate the true field by ~20% to 30% (Halgedahl et al., 1980). However, recent theoretical and experimental results (Winklhofer et al., 1997; Yu, 2011; Ferk et al., 2012) strongly suggest that the cooling rate correction for PSD grains is negligible. Accordingly, we did not apply cooling rate correction to our paleointensity results.
4. DISCUSSION AND CONCLUSIONS Our rock magnetic analyses show that the lava flows of the middle Lake Shore Traps possess rock magnetic characteristics which make them suitable for paleointensity experiments using the Thellier method and are likely to retain a pristine record of the strength of Earth’s magnetic field that existed at ~1.09 Ga. One hundred two samples out of 255 measured have resulted in paleointensity determinations of high technical quality. The experimental success rate of 40% is relatively high even when compared with the paleointensity studies of much younger rocks. The incorporation of low temperature demagnetization into the Thellier protocol (Schmidt, 1993; Celino et al., 2007) have resulted in a better linearity of the high-temperature segments on Arai plots and increased the quality factor of paleointensity determinations. For the majority of samples, the LTD resulted in steepening of the high-temperature linear segment on Arai plot (Fig. 6). On average, the paleointensity values based on pre-LTD data are 16% weaker than the values calculated from the post-LTD data. We feel, however, that the post-LTD values provide truer estimates of the paleofield intensity because the LTD removes the adverse effects associated with magnetizations of PSD and MD magnetite grains. Our SEM analyses indicate that some samples may contain homogeneous titanomagnetite grains that are unaffected by LTD. The presence of such grains may result in some shallowing of the high temperature slope used for paleointensity calculations. Hence, we cannot rule out a possibility that some of our paleointensity values may slightly underestimate the field strength. The mean VDM value that we have obtained from the middle Lake Shore Traps is almost two times lower than the VDM values for Midcontinent Rift rocks reported by Pesonen and Halls (1983) (Fig. 7). We note, however, that the latter are likely to overestimate the paleofield intensity because they were calculated by using the data from low-temperature (150550°C) segments of Arai plots. On the other hand, the paleofield intensity recorded by the Lake Shore Traps is two to three times stronger than the other Thellier data reported for the Neo- and Mesoproterozoic (Yu and Dunlop, 2001, 2002; Macoiun et al., 2003, 2006; McArdle et al., 2004; Celino et al., 2007) (Fig. 7). Taken at face value, these prior results suggest that the field intensity was low during this time period, and this was interpreted as supporting a “Proterozoic dipole low” (Biggin et al., 2009). We note, that most of these paleofield
Stud. Geophys. Geod., 57 (2013)
577
E. Kulakov et al.
Fig. 7. Neo- and Mesoproterozoic virtual dipole moment (VDM) Thellier data (with 1 uncertainties). PH83: the mean VDM values for Keweenawan rocks of normal (closed circle) and reversed (open circle) polarities (Pesonen and Halls, 1983). ON: the microwave paleointensity from Keweenawan intrusions in Ontario (McArdle et al., 2004). CGA and CGB: Cordova Gabbro, components A and B (Yu and Dunlop, 2002); AD: Abitibi dikes (Macouin et al., 2003); NF: Nova Floresta Gabbro (Celino et al., 2007); TD: Tudor Gabbro (Yu and Dunlop, 2001) with age interpretation from Dunlop and Yu (2004); MD: Mackenzie dikes (Macouin et al., 2006); LST: Lake Shore Traps (present study). Dashed line shows the mean of Thellier paleointensity data for the last 10 My with 95% uncertainty interval (gray box).
values are lower than the critical field value of approximately 4 1022 Am2 which is characteristic for transitional or excursional geomagnetic field recorded in 0.0310 Ma volcanic rocks (Tanaka et al., 1995) and may be associated with the emergence of large non-dipole fields at many locations (Guyodo and Valet, 1999; Tarduno and Smirnov, 2004). However, the paleodirectional data indicate that the paleointensity results shown in Fig. 7 represent a non-transitional geomagnetic field. Furthermore, large long-term nondipole components of the geomagnetic field have not been identified in searches of data from the late Archean to Proterozoic interval (e.g. Smirnov and Tarduno, 2004; Evans, 2006; Smirnov et al., 2011). Therefore, if true, the long-term low field intensities would imply a Proterozoic geomagnetic field that is significantly different from its modern counterpart. We note, however, that the studies resulting in low paleofield values are almost entirely based on intrusive rocks. Recently, it was suggested that the low paleointensities recorded by intrusive rocks are likely to be an artifact due to the presence of a thermochemical remanent magnetization (TCRM) (Smirnov and Tarduno, 2005). A TCRM is imparted if the oxyexsolution process during initial lava cooling continues at temperatures below the Curie point of magnetite. Estimates of the TCRM/TRM ratio show
578
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw
that the Thellier data could underestimate the true field value by a factor of 4 without violating experimental selection criteria. However, this is more likely if the cooling rates are slow, as is expected in large dikes and sills, and other large intrusions that currently dominate the Mesoproterozoic paleointensity data base. These are unlike the relatively thin, fast cooled lavas that we have studied. In light of these effects, we feel it is premature to conclude that the Precambrian field was weak based on paleointensity data from mafic dikes and other intrusive rocks. The paleointensities that we obtained from samples of the middle Lake Shore Traps are similar to those of the recent Earth’s magnetic field and are hence incompatible with a Proterozoic dipole low. Instead, they are consistent with the stable compositionally-driven geodynamo operating by 1.1 Ga. Such a scenario is consistent with the young age of the inner core preferred by many models of the Earth’s thermal history (e.g., Labrosse et al., 2001; Aubert et al., 2010).
APPENDIX Table 1. Summary of paleointensity data. VDM dVDM H dH [T] [T] [1022 Am2] [1022 Am2]
Site
n/N
LST1* LST2* LST3* LST4* LST5* LST6* LST(16)2 LST7 LST8 LST9 CS1 CS2 Mall (N=9) Mindep (N=4)
3/6 2/6 5/5 6/10 5/7 4/10 25/44 0/6 3/6 0/9 5/8 3/6 36/79 36/79
24.8 1.3 21.4 10.4 25.3 3.6 22.6 3.1 17.2 1.5 23.9 1.1 22.5 3.5 ----29.3 9.3 ----25.3 1.9 36.8 4.4 25.2 4.0 28.5 4.8
LST12 LST13 LST14* LST15* LST16* LST17* LST18* LST19* LST(1419)2
2/6 4/10 6/7 3/6 3/6 3/6 3/7 0/9 18/41
32.7 24.2 28.0 31.7 28.8 27.2 26.6 --28.5
f df
q dq
Lower Middle LST 5.6 4.8 5.7 5.1 3.9 5.4 5.1 --6.6 --5.7 8.3 5.7 6.5
0.3 2.4 0.8 0.7 0.3 0.3 0.8 --2.1 --0.4 1.0 0.9 1.1
0.67 0.23 0.77 0.10 0.55 0.14 0.52 0.12 0.54 0.11 0.51 0.09 0.59 0.13 --0.59 0.08 --0.67 0.15 0.69 0.19 0.61 ± 0.12 0.64 ± 0.14
9.3 3.9 7.7 1.7 8.2 3.0 9.4 4.3 12.7 3.91 7.6 1.2 9.2 3.0 --10.1 2.7 --16.8 4.7 22.0 5.9 11.5 ± 3.5 14.5 ± 4.1
0.3 0.7 0.7 1.3 0.8 1.2 0.5 --0.9
0.62 ± 0.05 0.61 ± 0.09 0.64 ± 0.11 0.67 ± 0.10 0.56 ± 0.03 0.61 ± 0.06 0.59 ± 0.08 --0.61 ± 0.08
10.8 ± 1.3 11.5 ± 1.4 14.5 ± 3.7 18.3 ± 13.1 10.9 ± 2.0 11.3 ± 1.6 13.2 ± 2.6 --13.6 ± 4.6
Upper Middle LST
Stud. Geophys. Geod., 57 (2013)
1.5 3.1 2.9 5.7 3.3 5.2 2.4 --3.8
7.4 5.5 6.3 7.2 6.5 6.2 6.0 --6.4
579
E. Kulakov et al.
Site
n/N
LST20 4/10 LST22 5/6 LST24 0/6 LST25 0/9 LST26 3/6 0/6 LST281* LST29* 2/6 2 2/12 LST(2829) LST30 3/10 LST31 2/6 SI1 3/6 SI2 3/7 SI3 3/6 SI4* 2/6 SI5* 5/6 7/12 SI(45)2 0/5 SI61 SI7 2/6 SI8 2/6 1 0/6 SI9 * SI10* 3/6 3/12 SI(910)2 Mall (N=21) 66/176 Mindep (N=16) 66/176 MLST (N=20) 102/176
dVDM VDM H dH [T] [T] [1022 Am2] [1022 Am2] 25.1 26.9 ----24.3 --20.5 20.5 16.7 23.8 29.1 20.5 27.4 27.2 20.9 24.1 --33.1 32.8 --22.2 22.2 24.9 25.0 26.3
4.0 6.9 ----2.9 --3.7 3.7 2.4 1.2 8.4 5.5 7.6 4.5 5.5 5.0 --4.2 9.5 --4.1 4.1 4.3 4.5 4.7
5.7 6.1 ----5.5 --4.6 4.6 3.8 5.4 6.6 4.6 6.2 6.2 4.7 5.4 --7.5 7.4 --5.0 5.0 5.6 5.7 5.9
0.9 1.6 ----0.7 --0.8 0.8 0.5 0.3 1.9 1.2 1.7 1.0 1.2 1.1 --1.0 2.2 --0.9 0.9 1.0 1.0 1.1
f df
q dq
0.63 ± 0.12 0.54 ± 0.04 ----0.60 ± 0.09 --0.51 ± 0.09 0.51 ± 0.09 0.59 ± 0.07 0.67 ± 0.12 0.64 ± 0.09 0.66 ± 0.11 0.68 ± 0.14 0.65 ± 0.09 0.60 ± 0.05 0.63 ± 0.07 --0.61 ± 0.02 0.67 ± 0.04 --0.62 ± 0.06 0.62 ± 0.06 0.62 ± 0.08 0.62 ± 0.08 0.63 ± 0.09
10.9 ± 1.9 5.6 ± 0.3 ----13.1 ± 2.8 --6.5 ± 0.9 6.5 ± 0.9 10.1 ± 1.8 13.9 ± 4.9 26.4 ± 19.3 16.0 ± 11.0 17.6 ± 7.9 9.7 ± 3.1 14.9 ± 6.3 12.3 ± 4.7 --15.7 ± 2.7 12.6 ± 1.5 --10.5 ± 3.1 10.5 ± 3.1 13.0 ± 6.3 12.9 ± 5.0 13.3 ± 4.8
* Sites that yielded serially correlated paleomagnetic directions (Kulakov et al, 2013); 1 - sites for which all paleointensity determinations were rejected according to acceptance criteria; 2 - combined paleointensities for the sites with serially correlated paleomagnetic directions; n - number of accepted determinations; N - number of samples used in experiment; H - site-mean paleofield intensity and its standard deviation (dH); VDM - site-mean virtual dipole moment with standard deviation (dVDM); f - the mean fraction of initial remanence removed during paleointensity experiment and its standard deviation (df); q - site-mean quality factor of Coe (1978); dq - quality factor’s standard deviation; Mall indicates mean values calculated for all sites including those with serially correlated directions. Mindep indicates mean values calculated after combining serially correlated sites; MLST - the mean values for all independent site-mean values (N = 20). All paleointensities listed in the table are calculated after LTDs. Acknowledgements: We thank William Rose for the permission to conduct paleomagnetic sampling on Silver Island. We thank Danford Moore, Ashley Kern, Daniel Nevins, and Matthew Laird for their participation in the field work and help with the rock magnetic and paleointensity experiments. This paper benefited from the thoughtful reviews by Andrei Kosterov and an anonymous reviewer.
580
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw References Aubert J., Tarduno J.A. and Johnson C.L., 2010. Observations and models of the long-term evolution of Earth's magnetic field. Space Sci. Rev., 155, 337370. Bickford L.R. Jr., 1950. Ferromagnetic resonance absorption in magnetite single crystals. Phys. Rev., 78, 449457. Bickford L.R. Jr., Brownlow J.M. and Penoyer R.F., 1957. Magnetocrystalline anisotropy in cobaltsubstituted magnetite single crystals. IEE Proc. B, 104, 238244. Biggin A.J., Geert H.M., Strick A. and Langeris C.G., 2009. The intensity of the geomagnetic field in the late-Archaean: new measurements and an analysis of the upgraded IAGA paleointensity database. Earth Planets Space, 61, 922. Carvallo C., Roberts A., Leonhardt R., Laj C., Kissel C., Perrin M. and Camps P., 2006. Increasing the efficiency of paleointensity analyses by selection of samples using first-order reversal curve (FORC) diagrams. J. Geophys. Res., 111, B12103, DOI: 10.1029/2005JB004126. Carter-Stiglitz B., Moskowitz B. and Jackson M., 2004. More on the low temperature magnetism of stable single domain magnetite: Reversibility and non-stoichiometry. Geophys. Res. Lett., 31, L06606, DOI: 10.1029/2003GL019155. Celino K.R., Trindade I.F. and Tohver E., 2007. LTD-Thellier paleointensity of 1.2 Ga Nova Floresta mafic rocks (Amazon craton). Geophys. Res. Lett., 34, L12306, DOI: 10.1029 /2007GL029550. Coe R.S., 1967. Paleointensities of Earth’s magnetic field determined from Tertiary and Quaternary Rocks. J. Geophys. Res., 72, 32473262. Coe R.S., Gromme S. and Mankinen E.A., 1978.Geomagnetic Paleointensities from radiocarbondated lava flows on Hawaii and question of the Pacific nondipole low. J. Geophys. Res., 83, 17401756. Coe R.S. and Glatzmaier G.A., 2006. Symmetry and stability of the geomagnetic field. Geophys. Res. Lett., 33, L21311, DOI: 10.1029/2006GL027903. Davis D.W. and Paces J.B., 1990. Time resolution of geologic events on the Keweenaw Peninsula and implications for development of the Midcontinent rift system. Earth Planet. Sci. Lett., 97, 5464. Diehl J.F. and Haig T.D., 1994. A paleomagnetic study of the lava flows within the Copper Harbor Conglomerate, Michigan: new results and implications. Can. J. Earth Sci., 31, 369380. Diehl J.F., Durant A.J. and Schepke C., 2009. Paleomagnetism of Silver Island, Keweenaw Peninsula, Michigan: Additional Support (?) for the Primary Curvature of the MCR. American Geophysical Union, Spring Meeting 2009, Abstract #GP11E-01. Dodson M.H. and McClelland-Brown E., 1980. Magnetic blocking temperatures of single-domain grains during slow cooling. J. Geophys. Res., 85, 26252637. Efron B., 1982. The Jackknife, the Bootstrap, and Other Resampling Plans. CBMS-NSF Regional Conference Series in Applied Mathematics 38. Society for Industrial and Applied Mathematics, Philadelphia, PA. Evans D.A.D., 2006. Proterozoic low orbital obliquity and axial-dipolar geomagnetic field from evaporite palaeolatitudes. Nature, 444, 5155.
Stud. Geophys. Geod., 57 (2013)
581
E. Kulakov et al. Ferk A., Leonhardt R., Hess K.U., Koch S., Krasa D., Egli R., Winklhofer M. and Dingwell D.B., 2012. Cooling rate dependence of the TRM of SD, PSD and MD particles. Geophys. Res. Abs., 14, 5244. Guyodo Y. and Valet J.P., 1999. Global changes in intensity of the Earth’s magnetic field during the past 800 kyr. Nature, 399, 249252. Haggerty S.E., 1991. Oxide textures - a mini-atlas. Rev. Mineral., 25, 129219. Halgedahl S.L., Day R. and Fuller M., 1980. The effect of cooling rate on the intensity of weak-field TRM in single-domain magnetite. J. Geophys. Res., 85, 36903698. Halls H.C. and Palmer H.C., 1981. Remagnetization in Keweenawan rocks. Part II: lava flows within the Copper Harbor Conglomerate, Michigan. Can. J. Earth Sci., 18, 13951408. Halls H.C., McArdle N.J., Gratton M.N., Hill M.J. and Shaw J., 2004. Microwave Paleointensities from dyke chilled margins: a way to obtain long-term variations in geodynamo intensity for the last three billion years. Phys. Earth Planet. Inter., 147, 183195. Harrison R.J. and Feinberg J.M., 2008. FORCinel: An improved algorithm for calculating first-order reversal curve distributions using locally weighted regression smoothing. Geochem. Geophys. Geosyst., 9, Q05016, DOI: 10.1029/2008GC001987. Hollerbach R. and Jones C.A., 1995. On the magnetically stabilizing role of the Earth’s inner core. Phys. Earth Planet. Inter., 87, 171181. Jaeger J.C., 1968. Cooling and solidification of igneous rocks. In: Hess H.H. and Poldervaart A. (Eds.), Basalts: The Poldervaart Treatise on Rocks of Basaltic Composition. Interscience Publishers, New York, London, Sydney, 503536. Johnson C.L., Constable C., Tauxe L., Barendregt R., Brown L., Coe R., Layer P., Mejia V., Opdyke N., Singer B., Staudigel H. and Stone D., 2008. Recent investigations of the 05 Ma geomagnetic field recorded by lava flows. Geochem. Geophys. Geosyst., 9, Q04032. DOI: 10.1029/2007GC001696 Kosterov A.A. and Prévot M., 1998. Possible mechanisms causing failure of Thellier palaeointensity experiments in some basalts. Geophys. J. Int., 134, 554572. Labrosse S., Poirier J.P. and Le Mouel J.L., 2001. The age of the inner core. Earth Planet. Sci. Lett., 190, 111123. Lane A.C., 1911. The Keweenaw Series of Michigan. Wynkoop Hallenbeck Crawford Co., Lansing, Michigan, Canada. Macouin M., Valet J.P., Besse J., Buchan K., Ernst R., LeGoff M. and Scharer U., 2003. Low paleointensities recorded in 1 to 2.4 Ga Proterozoic dykes, Superior Province, Canada. Earth Planet. Sci. Lett., 213, 7995. Macouin M., Valet J.P., Besse J. and Ernst R., 2006. Absolute paleointensity at 1.27 Ga from Mackenzie dyke swarm (Canada). Geochem. Geophys. Geosyst., 7, DOI: 10.1029 /2005GC000960. McArdle N.J., Halls H.C. and Shaw J., 2004. Microwave palaeointensities from 1.1 Ga Keweenawan dykes, Ontario. Geophys. Res. Abs., 6, 06433. McClelland-Brown E., 1984. Experiments on TRM intensity dependence on cooling rate. Geophys. Res. Lett., 11, 205208.
582
Stud. Geophys. Geod., 57 (2013)
Absolute paleointensity from Lake Shore Traps, Keweenaw Muxworthy A.R. and McClelland E., 2000. The cause of low-temperature demagnetization of remanence in multidomain magnetite. Geophys. J. Int., 140, 115131. Nagata T., Arai Y. and Momose K., 1963. Secular variation of the geomagnetic total force during the last 5000 years. J. Geophys. Res., 68, 52775281. Paces J.B. and Bornhorst T.J., 1985. Geology and geochemistry of lava flows within the Copper Harbor Conglomerate, Keweenaw Peninsula, Michigan. In: Blackburn C.E. (Ed.), Abstracts, Institute on Lake Superior Geology 31st Annual Meeting. Department of Geology, Lakehead University, Thunder Bay, Ontario, Canada, 7172 (http://www.d.umn.edu/prc /lakesuperiorgeology/Volumes/ILSG_31_1985_pt1_Kenora.cv.pdf) Palmer H.C., Halls H.C. and Pesonen L.J., 1981. Remagnetization in Keweenawan rocks. Part I: Conglomerates. Can. J. Earth Sci., 18, 599618. Pesonen L.J. and Halls H.C., 1983. Geomagnetic field intensity and reversal asymmetry in late Precambrian Keweenawan rocks. Geophys. J. Roy. Astr. Soc., 73, 241270. Roberts A.P., Pike C.R. and Verosub K.L., 2000. FORC diagrams: A new tool for characterizing the magnetic properties of natural samples. J. Geophys. Res., 105, 2846128475. Schmidt P.W., 1993. Paleomagnetic cleaning strategies. Phys. Earth Planet. Inter., 76, 169178. Shcherbakova V.V., Lubnina N.V., Shcherbakov V.P., Mertanen S., Zhidkov G.V., Vasilieva T.I. and Tselmovich V.A., 2008. Paleointensity and paleodirectional studies of early Riphaean dyke complexes in the Lake Ladoga region (Northwestern Russia). Geophys. J. Int., 175, 433448. Selkin P.A., Gee J.S., Tauxe L., Meurer W.P. and Newell A.J., 2000.The effect of remanence anisotropy on paleointensity estimates: a case study from the Archean Stillwater Complex. Earth Planet. Sci. Lett., 183, 403416. Smirnov A.V. and Tarduno J.A., 2003. Paleointensity of the early geodynamo (2.45 Ga) as recorded in Karelia: a single crystal approach. Geology, 31, 415418. Smirnov A.V. and Tarduno J.A. 2004. Secular variation of the Late Archean Early Proterozoic geodynamo. Geophys. Res. Lett., 31, L16607, DOI: 10.1029/ 2004GL020333. Smirnov A.V. and Tarduno J.A., 2005. Thermochemical remanent magnetization in Precambrian rocks: Are we sure the geomagnetic field was weak? J. Geophys. Res., 110, B06103, DOI: 10.1029/2004JB003445. Smirnov A.V., Tarduno J.A. and Evans D.A.D., 2011. Evolving core conditions ca. 2 billion years ago detected by paleosecular variation. Phys. Earth Planet. Inter., 187, 225231. Symons D.T.A., Lewchuk M.T., Dunlop D.J., Costanzo-Alvarez V., Halls H.C., Bates M.P., Palmer H.C. and Vandall T.A.,1994. Synopsis of paleomagnetic studies in the Kapuskasing structural zone. Can. J. Earth Sci., 31, 12061217. Tanaka H., Kono M. and Uchimara H., 1995. Some global features of paleointensity in geologic time. Geophys. J. Int., 120, 97102. Tarduno J.A., Cottrell R.D. and Smirnov A.V., 2001. High geomagnetic field intensity during the mid-Cretaceous from Thellier analyses of single plagioclase crystals. Science, 291, 17791783. Tarduno J.A. and Smirnov A.V., 2004. The paradox of low field values and the long-term history of the geodynamo. In: Channell J.E.T., Kent D.V., Lowrie W. and Meert J.G. (Eds.), Timescale of the Paleomagnetic Field. AGU Geophysical Monograph 145, 7584.
Stud. Geophys. Geod., 57 (2013)
583
E. Kulakov et al. Tarduno J.A., Cottrell R.D., Watkeys M.K. and Bauch D., 2007. Geomagnetic field strength 3.2 billion years ago recorded by single silicate crystals. Nature, 446, 657660. Tarduno J.A., Cottrell R.D., Watkeys M.K., Hofmann A., Doubrovine P.V., Mamajek E., Liu D., Sibeck D.G., Neukirch L.P. and Usui Y., 2010. Geodynamo, solar wind and magnetopause 3.45 billion years ago. Science, 327, 12381240. Tauxe L., Mullender T.A.T. and Pick T., 1996. Potbellies, wasp-waists and superparamagnetism in magnetic hysteresis. J. Geophys. Res., 101, 571583. Tauxe L. and Yamazaki T., 2007. Paleointensities. In: Kono M. (Ed.), Schubert G. (Ed.), Treatise on Geophysics 5, Geomagnetism (Editor-in-Chief Schubert G.). Elsevier Ltd., Oxford, U.K., 509564. Tauxe L. and Kodama K.P. 2009. Paleosecular variation models for ancient times: clues from Keweenawan lava flows. Phys. Earth Planet. Inter., 177, 3145. Thellier E. and Thellier O., 1959. Sur l’intensité du champ magnétique terrestre dans le passé historique et géologique. Annales Géophysique, 15, 285378. Verwey E.J.W., 1939. Electronic conduction of magnetite (Fe3O4) and its transition point at lowtemperature. Nature, 144, 327328. White W.S., Cornwall H.R. and Swanson R.W., 1951. Bedrock geology of the Ahmeek Quadrangle, Michigan. Geologic Quadrangle Map GQ-27, U.S. Geological Survey. Wilson R.L and Watkins N.D., 1967. Correlation of magnetic polarity and petrological properties in Columbia Plateau Basalts. Geophys. J. Roy. Astr. Soc., 12, 405429. Winklhofer M., Fabian K. and Heider F., 1997. Magnetic blocking temperatures of magnetite calculated with a three-dimensional micromagnetic model. J. Geophys. Res., 102, 2269522709. Winklhofer M. and Zimanyi G.T., 2006. Extracting the intrinsic switching field distribution in perpendicular media: A comparative approach. J. Appl. Phys., 99, 08E710, DOI: 10.1063 /1.2176598. Xu S. and Dunlop D.J., 2004. Thellier paleointensity theory and experiments for multidomain grains. J. Geophys. Res., 109, B07103, DOI: 10.1029/2004JB003024. Yoshihara A. and Hamano Y., 2004. Paleomagnetic constraints on the Archean geomagnetic field intensity obtained from komatiites of the Barberton and Belingwe greenstone belts, South Africa and Zimbabwe. Precambrian Res., 131, 111142. Yu Y. and Dunlop D.J., 2001. Paleointensity determination on the late Precambrian Tudor Gabbro, Ontario. J. Geophys. Res., 106, 2633126343. Yu Y. and Dunlop D.J., 2002. Multivectorial paleointensity determination from the Cordova Gabbro, southern Ontario. Earth Planet. Sci. Lett., 203, 983998. Yu Y. and Dunlop D.J., 2004. Intensity and polarity of the geomagnetic field during Precambrian time. In: Channell J.E.T., Kent D.V., Lowrie W. and Meert J.G. (Eds.), Timescale of the Paleomagnetic Field. AGU Geophysical Monograph 145, 85100. Yu Y., 2011. Importance of cooling rate dependence of thermoremanence in paleointensity determination. J. Geophys. Res., 116, B09101, DOI: 10.1029/2011JB008388
584
Stud. Geophys. Geod., 57 (2013)
