Elsevier

Aeolian Research

Volume 19, Part A, December 2015, Pages 55-63
Aeolian Research

Variability of magnetic character of S5-1 paleosol (age ∼470 Ka) along a rainfall transect explains why susceptibility decreased with high rainfall

https://doi.org/10.1016/j.aeolia.2015.09.003Get rights and content

Highlights

  • Both strongly magnetic maghemite and weakly magnetic hematite form simultaneously with development of pedogenesis in warm and wet interglacial times.

  • Parts of the fine-grained oxidized magnetite and maghemite suffer reductive dissolution under periodic heavy rainfall when pH is lowered. Subsequent drying creates additional late generation weakly magnetic hematite and goethite.

  • The variations in composition, size and concentration of both strongly magnetic magnetite and maghemite as well as the weakly magnetic hematite along the rainfall transect led to a more complete picture of iron mobility during pedogenesis.

Abstract

This study was based mainly on environmental magnetic, soil science and geochemical methods to analyze the spatial variation of magnetic characteristics of the S5-1 paleosol (age ∼470 Ka) in a NW–SE transect from Xifeng, Linyou and Baoji sections from the Chinese Loess Plateau. The results show that (a) pedogenesis and chemical weathering of the coeval S5-1 paleosol layers increased with rainfall from Xifeng to Baoji, (b) the proportion of the pedogenic maghemite component decreased, while (c) the concentrations of weakly magnetic antiferromagnetic minerals (hematite and goethite) relatively increased. Maghemite and hematite appear to have formed simultaneously with development of pedogenesis, thus being similar in this aspect to soil in arid regions of SW Spain. However, amount of goethite did not show a consistent relationship as observed from magnetism in the three sections. Maghemite was found to be the main contributor to magnetic enhancement. Iron–manganese cutans, an indicator of diagnesis, increased with the rainfall in the S5-1 paleosol layer from Xifeng to Baoji. As expected, they show that pedogenesis occurred intermittently between wet and dry conditions causing iron mineral transformation. However, we suggest that parts of the fine-grained strongly magnetic minerals (mainly oxidized magnetite and maghemite) were converted into weakly magnetic minerals (mainly hematite and goethite) by pedogenesis, which resulted in a decline in the finest superparamagnetic (SP) and stable single-domain (SSD) ferrimagnetic minerals, thus decreasing the susceptibility of S5-1 paleosol from north to south. This behavior in a high rainfall region is different from the aerobic soil in SW Spain in which part of the initial hydro-maghemite transformed only to hematite rather than goethite and hematite.

Introduction

Over the past 20 years, the relation between magnetic enhancement and rainfall has been systematically investigated over Chinese loess deposits. However, the relationships between soil magnetic susceptibility and pedogenesis in different regions are different. Magnetic susceptibility shows a generally positive correlation with pedogenic intensity in the central Chinese Loess Plateau (CLP), such as Xifeng (Liu et al., 2001), Lingtai (Ding et al., 1998) and Luochuan (Balsam et al., 2004). In the Xinjiang region, however, there is a more complex correlation between magnetic susceptibility and pedogenesis. Furthermore, globally different explanations of the mechanism of susceptibility enhancement have been proposed (Song et al., 2010, Guo et al., 2011, Orgeira et al., 2011). In addition, magnetic susceptibility of the loess profiles in Siberia and Alaska demonstrate a lower susceptibility in paleosols due to aridity. This latter phenomenon is explained by the increased transport of magnetite due to strong activity of “wind vigor” during the glacial periods (Begét et al., 1990, Evans et al., 2003, Chlachula, 2003, Lagroix and Banerjee, 2004, Kravchinsky et al., 2008), or water logging and reducing conditions in permafrost layers which led to the chemical transformation of wind-blown ferromagnetic minerals and the reduction in magnetic susceptibility in paleosols (Bábek et al., 2011, Taylor and Lagroix, 2014).

Previous studies have proposed various explanations about loess–paleosol susceptibility enhancement mechanism. But it is now generally accepted that neoformed fine-grained magnetite and maghemite, generated in situ by inorganic or perhaps microbial pedogenesis, have increased the paleosol susceptibility (Zhou et al., 1990, Banerjee et al., 1993, Hunt et al., 1995, Fine et al., 1995, Maher, 1998, Deng et al., 2001, Porter et al., 2001, Spassov et al., 2003, Qiang et al., 2005, Xie et al., 2009).

According to recent rock magnetic and mineralogical models, fine-grained pedogenic maghemite play a key role in determining increase or decrease of susceptibility (Liu et al., 2005, Liu et al., 2007, Nie et al., 2010). However, Torrent et al., 2007, Torrent et al., 2010 has suggested that under some contains, especially in Spain, hydro-maghemite causes the magnetic enhancement. Pedogenic maghemite has a distribution of grain sizes comprised between 10 and 100 nm and therefore happens to include mainly SP and SD domain states, with very minor addition of larger PSD particles (Liu et al., 2007, Geiss and Zanner, 2006). In limited areas understanding the basis of susceptibility mechanism has greatly promoted quantitative recovery of paleoclimatic indicators (rainfall and temperature) (Liu et al., 1995, Maher et al., 2003). As a valuable climatic proxy from Chinese loess deposits, magnetic susceptibility has been used for quantitative reconstruction of precipitation (Heller et al., 1993, Liu et al., 1995, Han et al., 1996, Hao and Guo, 2005, Balsam et al., 2011), although there is still no agreement about exact precipitation values from the different reconstructions.

Spatial variations of chemical weathering and decreasing trend of paleo-weathering in both paleosol and loess units from southeast to northwest in CLP are consistent with the modern rainfall pattern (Hao and Guo, 2005). The S5 layer is the most prominent and well-developed paleosol unit in CLP. The S5 period represents a climatic optimum during the past 2.5 Ma (An et al., 1987). In general, the paleosol sub-unit S5-1 has the highest or near highest susceptibility value in the central CLP thus representing highest temperature and humidity. However, in the southern CLP, precipitation is more abundant than the central of CLP, yet magnetic susceptibility value of S5-1 layer is not the highest (Kalm et al., 1996, Zhao et al., 2013, Guo et al., 2013). In the work presented here we have attempted a quantitative explanation of this anomaly.

Our work involves paleosol S5-1 (age from 0.448 to 0.479 Ma; Willimas et al., 1993) that has been sampled along a nearly north–south transect from Xifeng to Baoji along which mean annual rainfall (MAR) increases from 550 to 720 mm/y. The parent material is loess or aeolian dust which originated in north-west China and was deposited everywhere in the CLP. Thus, we expected the susceptibility of the S5-1 horizon to show the known positive relationship between MAR and susceptibility. Furthermore, since MAR is highest in our sampling site in the southernmost site, 720 mm/y, we hoped be able to test the models of Orgeira et al., 2011, Maher and Thompson, 1994, and Han et al. (1996) that predict magnetic enhancement declining when rainfall or precipitation–evaporation reaches a critical high value.

Ferrimagnetic magnetite and its low-temperature oxidation product, loosely called ‘maghemite’ are mainly responsible for magnetic enhancement. According to soil effective moisture’s models (Orgeira et al., 2011), the variations in the alternation under strongly oxidizing and reducing conditions should lead to ‘maghemite’ formation and magnetic enhancement initially. Qualitatively, then ‘maghemite’ may gradually decrease under reducing condition (Maher, 1998, Hanesch and Scholger, 2005, Fischer et al., 2008), and finally convert into weakly magnetic goethite etc., which greatly reduce susceptibility in paleosols. The hypothesis we propose to test with a rainfall transect is that under seasonal abundant rainfall, when iron is reduced from ferric to ferrous ions due to lower pH, we should observe a drop in susceptibility after a critical rainfall value is reached in our study transect. Thus we hoped to observe, after the critical value of MAR, susceptibility in this transect is no longer a reliable indicator of past MAR. In addition, we will use geochemical analysis and description of soil properties in addition to magnetic data to elucidate the origin of any observed susceptibility variation. In particular, our conclusion may shed light on the complex unresolved relationship between magnetic susceptibility and pedogenesis especially around the southern edge of the CLP.

Section snippets

Sampling

Three paleosol sections were sampled at 5 cm intervals from the bottom of L5 loess through S5 paleosol layer down to the top of the L6 loess layer of CLP (Fig. 1). Xifeng (XF) (35°46′N, 107°41′E) is located in the central CLP. Baoji (BJ) (34°25′N, 107°07′E) is located about 260 km south from XF on the southern edge of the CLP, at the north ‘foot’ of the Qinling Mountains. Linyou (LY) (34°45′N, 107°49′E) is located between the two sites on the southeast of XF section, and to the northeast of BJ.

Spatial and depth variation of magnetic properties of S5-1 paleosol

Magnetic susceptibility is controlled mainly by magnetic mineral composition, concentration, and grain size. In general, the maximum χ value of paleosol S5-1 is seen to decrease from XF to LY to BJ (Fig. 2a and Table 1). It can be seen that the χ value of BJ S5-1 paleosol unit (highest rainfall) is the lowest in all three sections, and also lower by about a factor of two than the relatively weak pedogenic S3 horizon (maximum value of 36.75 × 10−7 m3 kg−1) higher up in the same section (Guo et al.,

Disscussion and conclusion

As is well-known, χ is a useful palaeoclimatic indicator in the central CLP. At Xifeng (Liu et al., 2001), Lingtai (Ding et al., 1998) and Linyou profiles, the S5-1 layer has the highest χ value in the whole profile. In combination with pedogenic study (An et al., 1987), this indicates that S5-1 is the most developed paleosol. However, the χ value of Baoji S5-1 horizon is low within the profile, and also lower than S5-1 horizon at XF and LY profiles (Fig. 2 and Table 1). The MAP values (Fig. 1)

Acknowledgements

We thank Prof. Subir K. Banerjee of the IRM, who checked the English and revised the ideas expressed in the whole manuscript carefully. The MPMS and VSM measurements were made at the Institute for Rock Magnetism (IRM), University of Minnesota. We thank Mike Jackson, Peat Sølheid and Dario Bilardello for their help with the experiments, and thank Prof. W.G. Zhang and Y. Dong Ph.D., State Key Laboratory of Estuarine and Coastal Research, East China Normal University, for their help with the DRS

References (70)

  • X.F. Hu et al.

    Magnetic susceptibility of the quaternary red clay in subtropical China and its paleoenvironmental implications

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2009)
  • P.X. Hu et al.

    Characterizing and quantifying iron oxides in Chinese loess/paleosols: implications for pedogenesis

    Earth Planet. Sci. Lett.

    (2013)
  • L. Huang et al.

    Characteristics of micromorphology and element distribution of iron–manganese argillans in typical soils of subtropical China

    Geoderma

    (2008)
  • J.F. Ji et al.

    Mineralogic and climatic interpretations of the Luochuan loess section (China) based on diffuse reflectance spectrophotometry

    Q. Res.

    (2001)
  • V.E. Kalm et al.

    Clay minerals and their paleoenvironmental interpretation in the Baoji loess section, Southern Loess Plateau, China

    Catena

    (1996)
  • J. King et al.

    A comparison of different magnetic methods for determining the relative grain size of magnetite in natural materials: some results from lake sediments

    Earth Planet. Sci. Lett.

    (1982)
  • V.A. Kravchinsky et al.

    Magnetic indicators of global paleoclimate cycles in Siberian loess-paleosol sequences

    Earth Planet. Sci. Lett.

    (2008)
  • F. Lagroix et al.

    The regional and temporal significance of primary aeolian magnetic fabrics preserved in Alaskan loess

    Earth Planet. Sci. Lett.

    (2004)
  • X.M. Liu et al.

    Quantitative estimates of paleoprecipitation at Xifeng in the loess plateau of China

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1995)
  • Q.S. Liu et al.

    Review of recent developments in mineral magnetism of the Chinese loess

    Q. Sci. Rev.

    (2007)
  • C.C. Liu et al.

    Mineral magnetic studies of the vermiculated red soils in southeast China and their paleoclimatic significance

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2012)
  • B.A. Maher

    Magnetic properties of modern soils and quaternary loessic paleosols: paleo-climatic implications

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1998)
  • J.S. Nie et al.

    HIRM variations in the Chinese red-clay sequence: insights into pedogenesis in the dust source area

    J. Asian Earth Sci.

    (2010)
  • S.C. Porter et al.

    Dependence of near surface magnetic susceptibility on dust accumulation rate and precipitation on the Chinese Loess Plateau

    Q. Res.

    (2001)
  • Z.X. Shen et al.

    Spectroscopic analysis of iron-oxide minerals in aerosol particles from northern China

    Sci. Total Environ.

    (2006)
  • S. Spassov et al.

    Detrital and pedogenic magnetic mineral phases in the loess/palaeosol sequence at Lingtai (central Chinese Loess Plateau)

    Phys. Earth Planet. Inter.

    (2003)
  • Y.B. Sun et al.

    Changing color of Chinese loess: geochemical constraint and paleoclimatic significance

    J. Asian Earth Sci.

    (2011)
  • J. Torrent et al.

    Magnetic minerals in Calcic Luvisols (Chromic) developed in a warm Mediterranean region of Spain: origin and paleoenvironmental significance

    Geoderma

    (2010)
  • S.Y. Wang et al.

    Rock magnetism, iron oxide mineralogy and geochemistry of Quaternary red earth in central China and their paleopedogenic implication

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2013)
  • W.G. Zhang et al.

    East Asian summer monsoon intensity inferred from iron oxide mineralogy in the Xiashu Loess in southern China

    Q. Sci. Rev.

    (2009)
  • Z.S. An et al.

    The paleosol complex S5 in the China Loess Plateau – a record of climatic optimum during the last 1.2 Ma

    Geojournal

    (1987)
  • S.K. Banerjee et al.

    A rapid method for magnetic granulometry with applications to environmental studies

    Geophys. Res. Lett.

    (1981)
  • S.K. Banerjee et al.

    Separation of local signals from the regional paleomonsoon record of the Chinese Loess Plateau: a rock-magnetic approach

    Geophys. Res. Lett.

    (1993)
  • V. Barrón et al.

    Evidence for a simple pathway to maghemite in Earth and Mars soils

    Geochim. Cosmochim. Acta

    (2002)
  • V. Barrón et al.

    Critical evaluation of the revised akdalaite model for ferrihydrite – discussion

    Am. Mineral.

    (2012)
  • Cited by (5)

    • Decoupled paleosol-based proxies in Chinese loess deposits: Role of leaching and illuviation processes

      2022, Quaternary Science Reviews
      Citation Excerpt :

      Previous studies have extensively investigated paleosol-based proxies in loess deposits. Earlier observations from limited sites show consistent spatiotemporal trends between paleosol-based proxies (Heller and Liu, 1982; Liu, 1985; Liu and Ding, 1998; Maher, 2016), but growing evidence indicates that those proxy systems have complex behaviors and sometimes display inconsistent spatiotemporal patterns (Balsam et al., 2004; Bloemendal et al., 2008; Ding et al., 2001b; Guo et al., 2015; Ji et al., 2001; Meng et al., 2015, 2018; Sun et al., 2011; Wang et al., 2016; Yang et al., 2006; Zheng et al., 2022). For example, based on the chemical weathering proxies of carbonate content (Meng et al., 2015), chemical index of alteration (CIA), and elemental ratios (Bloemendal et al., 2008; Xiong et al., 2010; Yang et al., 2006), the top of loess layer L2 has an unexpectedly higher CWI on the southeastern (SE) CLP than the CWI of L1 at the same sites but a low CWI on the northwestern (NW) CLP, similar to the CWI of L1 at the same sites.

    • Palaeoclimatic records of the loess/palaeosol sequences of the Chinese Loess Plateau

      2016, Quaternary Science Reviews
      Citation Excerpt :

      In some areas, however, including the southern Plateau, its pedogenic magnetic susceptibility is lower than the immediately younger or older palaeosols. This indicates that in these sites the S5 soil microclimate and geochemistry was less favourable to pedogenic magnetite formation and/or preservation, and more favourable to haematite accumulation (Guo et al., 2013; Guo et al., 2015 and Boxes 2–4). Pedogenic alteration of loessic soils across the Chinese Loess Plateau can be rapid; incorporation of some organic matter, dissolution and leaching of detrital carbonate, translocation of clay minerals (Hallberg et al., 1978), and formation of haematite, goethite and a trace amount of magnetite can all occur within a few 100 s of years (Maher and Hu, 2006; Nie et al., 2010).

    • Climatic thresholds for pedogenic iron oxides under aerobic conditions: Processes and their significance in paleoclimate reconstruction

      2016, Quaternary Science Reviews
      Citation Excerpt :

      According to previous studies on the possible paleoclimate range of the CLP reconstructed by multiple parameters (Wang et al., 2012; Nie et al., 2014), the loess on the CLP was deposited under low temperature and rainfall conditions, within climate regimes ranging mostly to the left of the inflection lines (Fig. 6b). This validates the reliability of χ and its consistent correlation with chemical weathering in paleorainfall reconstructions of the Quaternary, except for some stages like S5 and S8 (Guo et al., 2001, 2015; Balsam et al., 2004). However, red clay was deposited under higher temperature and rainfall conditions, with climate regimes ranging mostly across the inflection lines (Fig. 6b).

    View full text