SEARCH

SEARCH BY CITATION

Keywords:

  • red soil sequence;
  • chemical remanent magnetization;
  • lock-in

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

[1] We present a high-resolution paleomagnetic investigation of the subtropical red soil sequence at the Damei section, Bose Basin, southern China. Maghemite with low coercivities and fine-grained hematite with high coercivities but relatively low unblocking temperatures were identified as main carriers of the natural remanent magnetization (NRM). Strong chemical weathering occurring under subtropical climatic conditions in southern China led to a chemical remanent magnetization (CRM) overprint that is sufficiently strong to mask the primary NRM. Analysis of the Bose Basin soil sequence indicates that the CRM has a large lock-in depth (>4 m). This example shows that magnetostratigraphic studies on red soil sequences in subtropical-tropical southern China should be interpreted with caution.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

[2] Paleomagnetostratigraphy is a successful tool for dating both marine and terrestrial sediments [Opdyke and Channell, 1996]. However, chemical remanent magnetization (CRM) often seriously overprints the primary detrital remanent magnetization (DRM) recorded by sediments, resulting in large CRM lock-in depths [e.g., Spassov et al., 2003]. The newly-formed magnetic minerals result either from transformation of existing primary magnetic minerals (for example, low-temperature oxidation), or are directly formed through diagenetic processes [Schwertmann, 1985; Torrent et al., 2006]. During these processes, the newly-formed magnetic mineral assemblages become magnetized in the direction of the predominant geomagnetic field during the crystallization process.

[3] Although the CRM acquisition process and the CRM properties, as carried by different magnetic minerals formed during pedogenesis [Torrent et al., 2006], have been well investigated by laboratory simulation, direct evidence of the CRM is often lacking in natural records. To better understand the CRM overprint effects, we investigated an extreme case from the Bose Basin in subtropical southern China. This sedimentary sequence has experienced moderate to strong chemical weathering during pedogenesis and diagenesis [Yin and Guo, 2006], thus complicating the interpretations of paleomagnetic and paleoclimatic results retrieved from the sequence. Therefore, sediments that have undergone strong pedogenesis might be affected by a strong chemical overprint, as we will demonstrate on the example of the Bose Basin red soils.

2. Sampling and Measurements

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

2.1. Geological Setting and Sampling

[4] The Bose Basin is a northwest-southeast elongated basin in the northwestern part of the Guangxi Zhuang Autonomous Region, South China. The exposed Quaternary deposits of the basin cover an area of about 800 km2, and crop out in seven river terraces (labeled T1 to T7) along the Youjiang River [Hou et al., 2000; Potts et al., 2000]. The investigated red soil sequence contains the Damei Paleolithic site (23°46.664′N, 106°43.720′E, elevation 160 m) and has formed on terrace T4. During the excavation of the site in 2005, 176 stone artifacts (including 2 handaxes) and 155 tektite pieces were unearthed [Wang et al., 2007].

[5] The sedimentary sequence is of fluvial origin [Hou et al., 2000] and consists mainly of latosol (laterite containing ferralsol) and laterite layers and is underlain by Eocene sandstone [Wang et al., 2007]. The sequence can be subdivided into six lithological units from top to bottom: (1) cultivated soil, grey sandy-clay (0.1 m); (2) reddish-brown clay, containing stone artifacts (0.7 m); (3) red clay (0.3 m), displaying few vermiculated features and containing a few stone artifacts; (4) vermiculated red soil (4.15 m), yielding abundant tektites and Paleolithic stone artifacts, such as handaxes and picks; (5) red clay (3.1 m); and (6) well-sorted cobble conglomerate (4 m).

[6] Using a magnetic compass, oriented block samples with dimensions of 10×10×20 cm were cut in the outcrop covering the depth interval between 1.75 and 8.35 m. Then, 164 oriented cubic specimens of 2 cm edge length were cut from the block samples at 3–5-cm intervals.

2.2. Measurements

[7] High-temperature magnetization measurements (MsT curves) were performed in air using a Magnetic Measurement Variable Field Translation Balance. Magnetic hysteresis measurements were made on a MicroMag 2900 Alternating Gradient Magnetometer. Remanence measurements were carried out using a 2G Enterprises Model 760-R cryogenic magnetometer installed in a magnetically shielded space with rest fields smaller than 300 nT. All the 164 oriented samples were subjected to progressive thermal demagnetization up to 585°C using a Magnetic Measurements thermal demagnetizer with a residual magnetic field less than 10 nT. Some parallel samples were subjected to progressive alternating field (AF) demagnetization up to 100 mT. X-ray diffraction (XRD) analyses were carried out using a DMAX2400 X-ray diffractometer in order to characterize the clay mineral assemblages.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

3.1. Mineral Magnetism and X-Ray Diffraction

[8] The MsT curves for samples representative of the two main lithological units (vermiculated red soils and normal red clays lacking vermiculated features, respectively in the upper and lower part of the investigated section) show a slight kink between ∼300°C and ∼400°C upon heating and the reduced magnetization intensity upon cooling (Figures 1a and 1b), which is interpreted as the conversion of metastable maghemite to hematite. These two samples further display pronounced wasp-waisted hysteresis loops (Figures 1c and 1d), which arise from the interaction of two coexisting magnetic mineral components with strongly contrasting coercivities [Roberts et al., 1995].

image

Figure 1. Mineral magnetic measurements for selected samples of the Damei section. (a, b) MsT curves. Solid and open circles represent heating and cooling runs, respectively. The arrows indicate the signals of maghemite. The applied field is 500 mT. Note that the two selected samples display paramagnetic contribution to the induced magnetization up to 700°C. (c, d) Hysteresis loops after paramagnetic correction. (e) Alternating field demagnetization curves of NRM. (f) Thermal demagnetization curves of NRM.

Download figure to PowerPoint

[9] The AF and thermal demagnetization curves of the natural remanent magnetization (NRM) are shown in Figures 1e and 1f, respectively. Overall, the AF demagnetization curves exhibit at least two distinct components. The magnetically soft components can be easily removed by 30-mT AF demagnetization. In contrast, the magnetically hard components remain relatively stable up to 100-mT AF demagnetization, especially for samples in the vermiculated red soils. Moreover, the contribution of the magnetically hard components to the bulk NRM (defined by the residual remanence after 30-mT AF demagnetization) is much higher for samples in the vermiculated red soils than in the underlying normal red clays.

[10] Compared to the AF demagnetization curves, the thermal demagnetization curves show similar patterns among different samples (Figure 1f). A broad distribution of unblocking temperatures characterizes the entire soil sequence. Maximum unblocking temperatures are 500–585°C. Nevertheless, the NRMs for samples from the vermiculated red soils are slightly softer than those from the underlying normal red clays upon heating.

[11] The XRD spectra of bulk samples reveal that a mixture of quartz and clay minerals with small amounts of hematite and goethite is present in the Damei section (Figure 2). Quartz grains predominate the coarse fraction of the sediments. The clay mineral assemblages are dominated by kaolinite and a small amount of and illite.

image

Figure 2. XRD spectra of bulk samples from the Damei section. H, hematite; G, goethite; K, kaolinite; I, illite; Q, quartz.

Download figure to PowerPoint

3.2. Paleomagnetic Results

[12] The intensity of the NRM of the samples varies between 10−4 − 10−2 A/m. Progressive demagnetization successfully isolated characteristic remanent magnetization (ChRM) components for most of the samples. Thermal demagnetization generally yields better results than AF demagnetization. Demagnetization results were evaluated by orthogonal diagrams [Zijderveld, 1967] and the principal components direction was computed by a least-squares fitting technique [Kirschvink, 1980]. The ChRM component was defined between 200°C and 450°C for most samples. Thermal demagnetization results of representative samples are shown in Figures 3a, 3b, 3c, 3d, 3e, and 3f. The maximum angular deviations (MAD) were usually smaller than 10°. Total 135 (82%) samples gave ChRM directions by thermal demagnetization. Virtual geomagnetic pole (VGP) latitudes were determined from the ChRM vector directions. For thermal demagnetization curves, only one component can be confidently defined above 150°C. In contrast, the AF demagnetization curves exhibit that the NRMs consist of at least two components with distinct remanence coercivity spectra (Figures 3g, 3h, and 3i).

image

Figure 3. Orthogonal projections of representative progressive demagnetization of the Damei section. (a, b, c, d, e, f) Thermal demagnetization. (g, h, i) Alternating field demagnetization. The solid (open) circles represent the horizontal (vertical) planes. The numbers refer to the temperatures in °C (Figures 3a–3f) or the fields in mT (Figures 3g–3i). NRM is the natural remanent magnetization. Note the variable NRM intensities for parallel samples due to the nonuniformity of the red soils (see text).

Download figure to PowerPoint

[13] The magnetostratigraphy is illustrated in Figure 4. These latitudes of VGP show that only one magnetozone with normal polarity is recognized in the Damei sedimentary sequence. Moreover, the paleomagnetic inclination is slightly shallower for samples in the normal red clays of the lower part of the section than in the overlying vermiculated red soils.

image

Figure 4. (a) Lithostratigraphy and magnetic polarity stratigraphy of the Damei red soil sequence. (b) Planktonic δ18O records [Clemens and Prell, 2003], microtektite counts [Li et al., 2004], and the Matuyama/Brunhes (M/B) geomagnetic reversal identified at 115 mcd (meter composite depth) [Kissel et al., 2003] for ODP site 1146 in the South China Sea. The arrow arbitrarily indicates the presumed position of the M/B boundary in the vermiculated red soil. The shade shows the correlation between the terrestrial Damei red soil sequence and the marine sediments of ODP site 1146. Dec., declination; Inc., inclination; MAD, maximum angular deviation; VGP Lat., latitude of virtual geomagnetic pole; MIS, marine oxygen isotope stage.

Download figure to PowerPoint

4. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

4.1. Magnetic Carriers of NRM

[14] The Bose Basin, dominated by a subtropical climate [Teilhard de Chardin et al., 1935], has a mean annual precipitation of ∼1100 mm and a mean annual temperature of ∼22°C. The dominance of kaolinite in the clay mineral assemblages in the Damei red soil sequence indicates a strong chemical weathering of the parent materials in this basin. The 0.270 nm and 0.416 nm XRD lines respectively indicate that hematite and goethite are present (Figure 2). The characteristics are typical for tropical-subtropical soils [Yin and Guo, 2006]. In this environment, maghemite and hematite are often formed through the same pedogenic process in red soils [Torrent et al., 2006]. The AF demagnetization curves of the red soils are consistent with this model that the magnetically soft/hard components of the NRM are interpreted to be carried by maghemite and hematite, respectively. However, the goethite in terrestrial environments is usually Al-substituted. Al-substitution may sufficiently suppress its Néel temperature (TN) below room temperature. Around TN, Al-substituted goethite loses its ability to carry a remanence [Liu et al., 2004]. Thus, the contribution of goethite to the NRMs of the red soils is possibly negligible. Moreover, thermal demagnetization above 150°C can efficiently remove effects from goethite, and thus avoid complexities of interpretation of ChRM.

[15] It should be noted that the red soils in subtropical southern China are very nonuniform. They usually consists of red and/or white veins [Yin and Guo, 2006]. The red veins contain a significantly higher amount of antiferromagnetic minerals (dominantly hematite) than the white veins. Thus parallel samples at the same depth usually show variable magnetic properties. Yin and Guo [2006] have systematically investigated the soil properties of the vermiculated red soils as well as the underlying red clays. They put forward that the formation of the vermiculated features within the soil profile results from iron-depletion due to enhanced amount of precipitation. Therefore, hematite contributes more to the NRM for samples in the vermiculated red soils than in the underlying normal red clay because maghemite is preferentially consumed in a reductive environment. The newly formed hematite might be of superparamagnetic origin and will hence not contribute to the NRM. However, the vermiculated red soils have hard NRM components higher than the normal red clays (Figure 1e). This strongly indicates that the newly formed hematite significantly contributes to the bulk NRM.

[16] The thermal demagnetization curves of representative samples show that the major unblocking temperatures for hematite are much lower than its Néel temperature (Figure 1f). This behavior may be explained as follows. First, unblocking temperatures decrease generally with decreasing grain size [Dunlop and Bina, 1977], which would indicate the presence of very fine grained hematite in our case. Second, the Curie temperature of the ferromagnetic defect moment of hematite is lowered because of aluminium substitution into the crystal lattice, which is expected to occur in hematites from lateritic soils [Herbillon and Nahon, 1988]. However, the observed strong rubification strongly suggests that hematite was formed by pedogenic and diagenetic processes and hence carries a CRM.

4.2. Large CRM Lock-In Depth

[17] The handaxes-bearing sedimentary layer associated with the Australasian tektites occurring in the Bose Basin has been well dated to be ∼803 ka old by 40Ar/39Ar analyses of the tektites [Hou et al., 2000]. Therefore, the tektite-bearing sedimentary layer, which is observed at about 3.2–3.7 m in the Damei section (Figure 4), serves as a reliable tie point to link this profile to coeval marine sediments.

[18] Previous studies on ODP cores from marginal seas of the Indonesian archipelago and northern South China Sea indicated that the microtektites preceded the Matuyama/Brunhes (M/B) geomagnetic polarity reversal by ∼12 kyr [Schneider et al., 1992] and ∼10.7 kyr [Li et al., 2004], respectively. Constrained by the age of the tektite layer, the adjacent overlying and underlying sediments of the Damei profile should be situated in the Matuyama reverse chron. On the contrary, Figure 4 shows that these layers record a normal polarity. This indicates that the primary DRM of these layers has been replaced by the secondary CRM.

[19] Note that the red clays in the lower part of the Damei section recorded a systematically shallower inclination than the overlying vermiculated red soils (Figure 4). This could be either due to compaction or due to the mixing of two paleomagnetic vectors. The first mechanism cannot explain why inclination shallowing occurs only within the red clays. In fact, vermiculated features in the upper red soils indicate a very strong degree of pedogenesis. During this process, the primary geomagnetic signals will be masked by a much stronger CRM and thus the NRM of the sediments will be consistently reset to the normal polarity direction. In contrast, the underlying red clays underwent relatively weaker alteration from pedogenesis, thus it is possible that the primary DRM with a negative polarity could be partially preserved. For an ideal case, if both polarities aligned perfectly antiparallel, the recorded paleomagnetic inclination will not be shallowed. However, hematite and maghemite may not record consistent directions of remanence because they have distinct coercivities. Thus, the vector summation of primary DRM with a negative polarity and the CRM with a normal polarity carried by the pedogenically-produced magnetic particles could shallow the overall paleomagnetic inclination. The pedogenic maghemite and the primary magnetic minerals (possibly magnetite) have similar AF and thermal demagnetization spectra, and therefore, both demagnetization methods cannot isolate them. Nevertheless, for our case, the shallowed paleomagnetic inclination could be a useful tool to identify the negative polarity zone.

[20] Spassov et al. [2003] put forward a CRM lock-in model, which assumed that the lock-in of CRM could occur well below the termination of the lock-in of DRM up to 2–3 m for Chinese weathered loess/palaeosol sequences. Our results show a case that strongly demonstrates that the CRM can indeed yield a very large lock-in depth (e.g., >4 m) in the subtropical red soil sequence in the Bose Basin, southern China (Figure 4). However, reliable magnetostratigraphy results have been obtained from red clays in the middle-lower reaches of the Yangtze River [Qiao et al., 2003] and the Chinese Loess Plateau [Ding et al., 2001], where chemical weathering intensity is much weaker than in the subtropical Bose Basin. Further studies on the relationship between the degree of pedogenesis and the CRM overprinting will be useful to decode the exact overprinting process, and will be addressed elsewhere.

5. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

[21] A high-resolution paleomagnetic investigation, coupled with mineral magnetic measurements and XRD analyses, has been carried on the red soil sequence at the Damei section, Bose Basin, southern China. Mineral magnetic measurements suggest that maghemite with low coercivities and fine-grained hematite with high coercivities and low unblocking temperatures are main carriers of the NRMs. XRD analyses show that kaolinite and a small amount of and illite dominate the clay mineral assemblages. Strong chemical weathering occurring under subtropical climatic conditions in the Bose Basin, southern China led to a CRM overprint that is sufficiently strong to mask the primary NRM, yielding a large lock-in depth (e.g., >4 m). In summary, regardless of the complexities of the CRM overprinting processes, by directly correlating the tektite layer of the Bose red soil sequence with the microtektite layer of ODP site 1146, we demonstrate that the subtropical Bose sequence was strongly overprinted by a CRM. Therefore, magnetostratigraphic studies on red soil sequences in subtropical-tropical southern China should be evaluated with great caution.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References

[22] Financial assistance was provided by NSFC (grants 40221402, 40163001 and 40325011) and CAS. Q. Liu acknowledges further support from the European Marie-Curie Fellowship, proposal 7555.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Sampling and Measurements
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  • Clemens, S. C., and W. L. Prell (2003), Data report: Oxygen and carbon isotopes from Site 1146, northern South China Sea, Proc. Ocean Drill. Program Sci. Results, 184, 18.
  • Ding, Z., S. Yang, S. Hou, X. Wang, Z. Chen, and T. Liu (2001), Magnetostratigraphy and sedimentology of the Jingchuan red clay section and correlation of the Tertiary eolian red clay sediments of the Chinese Loess Plateau, J. Geophys. Res., 106, 63996407.
  • Dunlop, D. J., and M.-M. Bina (1977), The coercive force spectrum of magnetite at high temperatures: Evidence for thermal activation below the blocking temperature, Geophys. J. R. Astron. Soc., 51, 121147.
  • Herbillon, A. J., and D. Nahon (1988), Laterites and lateritization processes, in Iron in Soils and Clay Minerals, edited by J. W. Stucki, B. A. Goodman, and U. Schwertmann, pp. 779796, D. Reidel, Dordrecht, Netherlands.
  • Hou, Y., R. Potts, B. Yuan, Z. Guo, A. Deino, W. Wang, J. Clark, G. Xie, and W. Huang (2000), Mid-Pleistocene Acheulean-like stone technology of the Bose Basin, South China, Science, 287, 16221626.
  • Kirschvink, J. L. (1980), The least-squares line and plane and the analysis of palaeomagnetic data, Geophys. J. R. Astron. Soc., 62, 699718.
  • Kissel, C., C. Laj, S. Clemens, and P. Solheid (2003), Magnetic signature of environmental changes in the last 1.2 Myr at ODP Site 1146, South China Sea, Mar. Geol., 201, 119132.
  • Li, X., Q. Zhao, B. Huang, and X. Su (2004), High-resolution age estimation of the Mid-Pleistocene impact event (in Chinese with English abstract), Mar. Geol. Quat. Geol., 24(2), 7377.
  • Liu, Q., J. Torrent, Y. Yu, and C. Deng (2004), Mechanism of the parasitic remanence of aluminous goethite [α-(Fe, Al)OOH], J. Geophys. Res., 110, B12106, doi:10.1029/2004JB003352.
  • Opdyke, N. D., and J. E. T. Channell (1996), Magnetic Stratigraphy, 346 pp., Academic, San Diego, Calif.
  • Potts, R., W. Huang, Y. Hou, A. Deino, B. Yuan, Z. Guo, and J. Clark (2000), Tektites and the age paradox in Mid-Pleistocene China, Science, 289, 507.
  • Qiao, Y., Z. Guo, Q. Hao, W. Wu, W. Jiang, B. Yuan, Z. Zhang, J. Wei, and H. Zhao (2003), Loess-soil sequence in southern Anhui Province: Magnetostratigraphy and paleoclimatic significance, Chin. Sci. Bull., 48, 20882093.
  • Roberts, A. P., Y. Cui, and K. L. Verosub (1995), Wasp-waisted hysteresis loops: Mineral magnetic characteristics and discrimination of components in mixed magnetic systems, J. Geophys. Res., 100, 17,90917,924.
  • Schneider, D. A., D. V. Kent, and G. A. Mello (1992), A detailed chronology of the Australasian impact event, the Brunhes-Matuyama geomagnetic polarity reversal, and global climate change Pages, Earth Planet. Sci. Lett., 111, 395405.
  • Schwertmann, U. (1985), The effect of pedogenic environments on iron oxide minerals, Adv. Soil Sci., 1, 171200.
  • Spassov, S., F. Heller, M. E. Evans, L. P. Yue, and T. von Dobeneck (2003), A lock-in model for the complex Matuyama-Brunhes boundary record of the loess/paleosol sequence at Lingtai (Central Chinese Loess Plateau), Geophys. J. Int., 155, 350366.
  • Teilhard de Chardin, P., C. C. Young, W. C. Pei, and H. C. Chang (1935), On the Cenozoic formations of Kwangsi and Kwangtung, Bull. Geol. Soc. China, 14, 179205.
  • Torrent, J., V. Barrón, and Q. Liu (2006), Magnetic enhancement is linked to and precedes hematite formation in aerobic soil, Geophys. Res. Lett., 33, L02401, doi:10.1029/2005GL024818.
  • Wang, W., J. Mo, and Z. Huang (2007), Recent recovery of handaxes associated with tektites in Nanbanshan locality of Damei site, Bose basin, Guangxi, south China, Chin. Sci. Bull., in press.
  • Yin, Q., and Z. Guo (2006), Mid-Pleistocene vermiculated red soils in southern China as an indication of unusually strengthened East Asian monsoon, Chin. Sci. Bull., 51, 213220.
  • Zijderveld, J. D. A. (1967), A. C. demagnetization of rocks: Analysis of results, in Methods in Paleomagnetism, edited by D. W. Collinson, K. M. Creer, and S. K. Runcorn, pp. 254286, Elsevier, New York.