This article was originally published in the Encyclopedia of Quaternary Science published by Elsevier, and the attached copy is provided by Elsevier for the author's benefit and for the benefit of the author's institution, for non-commercial research and educational use including without limitation use in instruction at your institution, sending it to specific colleagues who you know, and providing a copy to your institution’s administrator.
All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution’s website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier's permissions site at: http://www.elsevier.com/locate/permissionusematerial Hatfield R.G., and Stoner J.S. (2013) Magnetic Proxies and Susceptibility. In: Elias S.A. (ed.) The Encyclopedia of Quaternary Science, vol. 2, pp. 884-898. Amsterdam: Elsevier. © 2013 Elsevier Inc. All rights reserved.
Author's personal copy
Magnetic Proxies and Susceptibility R G Hatfield and J S Stoner, Oregon State University, Corvallis, OR, USA ã 2013 Elsevier B.V. All rights reserved.
Introduction: Magnetic Susceptibility and Environmental Magnetism in Paleoceanography Magnetic susceptibility (MS) is one of the most commonly employed proxies in paleoresearch and is routinely measured on materials collected by the Integrated Ocean Drilling Program (IODP, www.iodp.org) and international piston coring cruises. MS measurements are not restricted to the marine realm and are commonly employed in terrestrial drilling programs like the Antarctic Geological Drilling Program (Andrill, www.andrill.org) and the International Continental Drilling Program (ICDP, www.icdp-online.org), as well as many lake coring, terrestrial stratigraphic, Quaternary geological, and archaeological projects. The widespread use of MS reflects its sensitivity as an indicator of lithologic variation, coupled with the practicality and versatility of measurement. MS measurements can be made on many types of samples, from sediment cores and boreholes to discrete samples of any shape or even directly on an outcrop. Setup costs are low and running costs are negligible, data acquisition is fast, sample preparation is minimal, and as it is both nondestructive and does not interfere with other measurements, it is often the first measurement made in many studies. These factors make MS uniquely applicable to a host of problems, yet MS is still poorly understood by many in the Quaternary and geological communities. Even for those who specialize in magnetism, determining what exactly drives MS variations within any particular record is a nontrivial undertaking. MS is just the simplest of many measurements that make up the tool kit known as environmental magnetism (Thompson and Oldfield, 1986) and often many other magnetic measurements are needed to truly understand what drives MS changes (e.g., Table 1). Here we review and assess both the potential and the opportunities MS and environmental magnetism have for deciphering the Quaternary paleoceanographic record.
What Is MS? All substances respond to a magnetic field, the question is by how much and in what direction. MS is the relationship between an induced magnetization (M) to an applied magnetic field (H), or simply how magnetizable a material becomes when placed in an external magnetic field. Low-field ( 0.1 mT) volumetric MS (k) is the type typically measured in most paleostudies: k ¼ M=H where M is the magnetic moment induced per unit volume (amperes per meter; Am1) and H is the applied field (Am1). As both M and H have the same syste`me international d’unite´s (SI) units, k is dimensionless. Centimeter–gram–second (cgs) units are also dimensionless, where 4p SI ¼ 1 cgs. Unfortunately, it is often not explicitly stated which system is used in
many MS studies, making comparisons difficult. To exclude density-related effects and permit comparison between diverse samples, volumetric susceptibility can be density-corrected to produce mass-normalized MS values (w). w ¼ k=r where r is the sample density. SI units of w are m3 kg1; cgs units are emu Oe1 g1.
Influences on MS Values Different materials, their concentration, and their magnetic grain size and shape, can all affect the MS values of any sample. Resulting from the orbital and spin motions of electrons and their interactions, all materials produce a magnetic response when exposed to a magnetic field, though the nature and strength of this response varies between materials. In the transition metals (Fe, Co, Ni), and their compounds, the magnetization can be particularly strong with parallel (but unequal) coupling of unpaired electrons within a crystal lattice producing a strong magnetization and ferromagnetic or ferrimagnetic behavior (Figure 1; Table 2). True ferromagnetism only exists in the elemental state of these transition metals. As these elements are commonly oxidized (or reduced) into various ferrimagnetic forms in the environment, ferromagnetic behavior is extremely rare in natural samples. Common ferrimagnetic minerals include magnetite, maghemite, and greigite (Table 2). These possess strong positive magnetic susceptibilities and, when present, these minerals will often dominate the MS signal (Figure 1). When atomic arrangements are equal, but opposite, there is no (theoretical) net magnetization as parallel pairings cancel out (antiferromagnetism). However, for a variety of reasons, alignment may not be perfectly parallel in antiferromagnetic minerals and a residual magnetization may result, inducing canted antiferromagnetic behavior (Figure 1). Hematite and goethite are common canted antiferromagnetic minerals possessing weak positive susceptibility (Figure 1; Table 2). Paramagnetic and diamagnetic behavior (Figure 1; Table 2) is induced in samples that have some or no unpaired electrons, respectively. Unlike ferro/ferrimagnetic and canted antiferromagnetic behavior, any induced magnetization in a paramagnetic or diamagnetic material is nonpermanent and lost upon removal of the field. Common examples of paramagnetic magnetic materials are ilmenite, biotite, lepidocrocite, siderite, and pyrite, all exhibiting weakly positive magnetic susceptibilities. Common diamagnetic materials include quartz, calcite, and water with weakly negative magnetic susceptibilities (Figure 1; Table 2). In the absence of ferrimagnetic minerals, paramagnetic or diamagnetic minerals may drive the MS signal. In many environments, the concentration and type of magnetic minerals are often the dominant controls on MS.
884 Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
Table 1
885
Some common magnetic parameters and the discrimination they afford
Parameter
Notes
k, w
Magnetic susceptibility (MS): the ratio of magnetization induced to the intensity of the magnetizing field. Measured within a small (0.1 mT) reversible magnetic field either volumetrically (k) or on a mass-normalized basis (w). Roughly proportional to the concentration of ferrimagnetic minerals within a sample Units: k (dimensionless), w (m3 kg1) Frequency-dependent susceptibility: variation of MS with frequency. Measured at 0.46 and 4.6 kHz on a Bartington MS2B. Indicative of superparamagnetic (SP) grains around the SP/single domain (SD) boundary (0.03 mm). At higher frequencies (whf), grains at the SP/SD boundary respond as SD grains; thus, high frequency MS is proportionally lower than low frequency (wlf) MS Units: expressed as a % of wlf Anhysteretic remanent magnetization: acquired when a sample is placed in a decreasing alternating magnetic field (e.g., 100 mT) in the presence of a small steady biasing field (e.g., 0.1 mT). Sensitive to both ferrimagnetic concentration and ferrimagnetic grain size Units: Am2 kg1 (Saturation) isothermal remanent magnetization (SIRM): magnetic remanence induced in a sample upon removal of the field. Usually attained in increasing field strengths (e.g., IRM 20 mT, IRM 100 mT) up to the saturating field (SIRM), which is usually an arbitrary term for largest field generated in the laboratory (1000 mT). SIRM is an indicator for the concentration of magnetic minerals but also responds to grain size variations Units: Am2 kg1 This ratio can be diagnostic of either mineralogy, or if mineral type and concentrations are similar, magnetic grain size Units: Am1 This ratio is dominantly used to determine magnetic grain size variations, but it is also somewhat sensitive to concentration Units: Am1 Ratio of remanence acquired in a smaller field than SIRM to the SIRM. Different field sizes can reveal different aspects of the magnetic signal independent of magnetic concentration. For example, low fields such as 20 mT are sensitive to coarse-grained magnetite, whereas acquisition at higher fields (e.g., 300 mT) can indicate hematite contributions Units: dimensionless Demagnetization parameters: obtained by applying reversed magnetic fields to a previously saturated sample. The coercivity of remanence (Bo)cr is the reverse field (mT) at which the SIRM returns to zero. Loss in other fields SIRM-x mT/SIRM can be expressed as a ratio. Loss at a field of 100 mT (which discriminates between ferrimagnetic and canted antiferromagnetic minerals) has been termed the S-ratio Units: (Bo)cr (mT), SIRM-x mT/SIRM, and the S-ratio are dimensionless
wfd
ARM
(S)IRM
SIRM/w ARM/w IRM x mT/SIRM
(Bo)cr SIRM-x mT/ SIRM S-ratio
However, magnetic grain size and shape may also influence the MS signal. Ferrimagnetic grains organize their magnetization into different regions, or domains, to achieve minimum energy states; this results in a grain size dependence of MS values. For example, in magnetite ultrafine particles known as superparamagnetic (SP) grains (10 mm) possess higher MS values than smaller single domain (SD) and pseudo SD (PSD) grains in the 0.03 to 10 mm size range (Figure 2). Anisotropy also affects MS values as not all grains are spherical. Magnetic energies favor ordering along the easiest direction, the long axis of elongate grains, with grain shape having as much as a factor of 2 influence on MS. For more information on the factors affecting MS measurement, see Thompson and Oldfield (1986), Dearing (1999), Evans and Heller (2003), and Tauxe (2010).
MS Measurement MS is the ratio of the induced magnetic moment to the applied field. Changes to the strength and frequency of the applied field and the temperature at which it is measured induce nonlinear magnetic responses in a sample. These differences can provide information about the composition of the materials carrying the magnetization, which is important for their interpretation.
Most common measurements of MS are made in low fields (0.1 mT), which do not permanently magnetize the sample, allowing them to return to their original magnetic state on removal of the field. When ferrimagnetic minerals dominate the mineral assemblage, low-field MS is often assumed to be proportional to their concentration. In greater field strengths, a variety of different minerals and magnetic grain sizes can contribute to the MS signal. However, above a certain critical field, the induced magnetization becomes nonreversible and a magnetization, or remanence, is retained by the sample. Measurement of the magnetization in different field sizes permits construction of remanence parameters and ratios (when measured in zero field) (i.e., isothermal remanent magnetization (IRM; Table 1)) and construction of hysteresis loops (when measured infield) that can be diagnostic of magnetic mineralogy and/or magnetic grain size. MS is dependent on the frequency of the applied field, which has some magnetic grain size dependence. Many systems allow measurements to be made at variable frequencies providing information on the proportion of ultrafine SP grains. MS also displays strong temperature dependence, which is a function of intrinsic magnetic mineralogical properties. Regular measurements made through warming or cooling cycles can produce characteristic curves indicative of different magnetic mineral or grain size populations. In summary, MS is an aggregated function of magnetic concentration, mineralogy, and grain size and
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 886
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
Fe
rrim
ag
ne ti
Magnetization (Am-1)
c
Fe rrim ag ne tic
+
etic
agn
ted
Can
-
rom tifer
an
etic
magn
Para
Magnetic field (Am-1) c
0 Diam
+
agne
tic
Ferromagnetic
Ferrimagnetic
Resultants:
Antiferrimagnetic
Canted anti ferrimagnetic
Zero
Figure 1 Schematic diagram illustrating the relative strengths and organizations for a range of magnetic behaviors. Adapted from Dearing J (1999) Magnetic susceptibility. In: Walden J, Oldfield F, and Smith J (eds.) Environmental Magnetism: A Practical Guide, Technical Guide No. 6, pp. 35–62. London: Quaternary Research Association.
shape that is dependent upon the applied field, frequency, and temperature of measurement.
MS Measurement Systems MS (k or w) can be measured on a variety of field and laboratory equipment providing MS measurements of discrete
samples, whole cores, split cores, u-channels, or in situ field samples, and can be made in varying field strengths, frequencies, and at different temperatures. For all systems, the magnetic susceptometer produces a magnetic field that induces and measures an infield magnetic moment in response to these three factors. Due to cost, ease of use, and versatility, the Bartington MS2 is often the most widely available and used MS measurement
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 887
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
Table 2 Magnetic behavior, crystal arrangement and example minerals and mass normalized susceptibilities. Adapted from Dearing J (1999) Magnetic susceptibility. In: Walden J, Oldfield F, and Smith J (eds.) Environmental Magnetism: A Practical Guide, Technical Guide No. 6, pp. 35–62. London: Quaternary Research Association. Magnetic behavior
Crystal arrangement
Minerals
MS
‘True’ ferromagnetism
Parallel coupling of all unpaired electrons to the applied field
Fe – iron Co – cobalt Ni – nickel
276 000 204 000 68 850
Antiferromagnetism
In oxides of Fe, Co, and Ni, electron coupling occurs via an intermediate oxygen atom with alternate crystal layers aligning opposite to each other resulting in no net magnetic moment If the magnetic moments of the alternating iron compounds are unequal, then a magnetic moment is created
Fe3O4 – magnetite gFe2O3 – maghemite Fe3S4 – gregite aFe2O3 – hematite aFeOOH – goethite FeTiO3 – ilmenite Mg,Fe,Al silicate – biotite gFeOOH – lepidocrocite FeCO3 – siderite FeS2 – pyrite SiO2 – quartz CaCO3 – calcite H2O – water
660 440 170 0.7 0.4 1.7 0.5 0.7 1.0 0.3 0.0058 0.0048 0.009
Ferrimagnetism
Canted antiferromagnetism
For various reasons, sublattices in otherwise antiferromagnetic arrangements may not be perfectly aligned and a small residual magnetization exists Some unpaired electrons align to an applied field; this magnetization is lost on removal of the field
Paramagnetism
Diamagnetism
Compounds with no unpaired electrons act to repel the applied field; this magnetization is lost on removal of the field
Units of mass-normalized magnetic susceptibility (MS) – 106 m3 kg1.
1200
Magnetic susceptibility (m3 kg-1)
1000
800
600
400
200 SP 0 0.01
SD 0.1
PSD 1
10 Grain size (mm)
MD 100
1000
Figure 2 Domain-size boundaries and variation in MS as a function of magnetic grain size for magnetite. Domain abbreviations: SP ¼ superparamagnetic, SD ¼ single domain, PSD ¼ pseudo-single domain, MD ¼ multi domain. Data from Maher (1988), Ozdemir and Banerjee (1982) and Dankers (1978).
system. Capable of parts per million (ppm) detection of magnetite, it can be coupled with a variety of sensors to allow initial characterization of the magnetic properties of a sample. These measurements are most commonly made at low field (80 Am1), low frequency (0.47 kHz), and room temperature. For example, MS2C loop sensors coupled with measurements of g-ray porosity, p-wave velocity, and electrical resistivity on multisensor tracks are a standard shipboard measurement on
both coring and drilling operations alike. Standard 1.5-m-long core sections can be characterized in as rapidly as 10 min in some configurations, providing real-time data to aid core correlation, site selection, and for initial sediment interpretation. Such measurements should account for the background MS before and after measurement to help correction for local environment and drift complication. This can be easily corrected for when measuring discrete samples as the background can be recorded after
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 888
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
every measurement. Whole-core- or other long-core-logging measurements at relatively high (e.g., centimeter scale) resolution require many MS measurements between background air measurements. Sensor drift is usually assumed to be linear; however, this may not always be a valid assumption. More accurate measurements of drift can be obtained by measuring background MS mid-logging, though this increases measurement time and introduces the potential for positional errors. Core-logging sensors also integrate data over a stratigraphic interval (e.g., the MS2C loop smoothes over twice its sensor diameter; Dearing, 1999); thus, data are smoothed and edge effects at core ends are introduced. It is theoretically possible to reintroduce resolution and lost edge effect data through deconvolution, though in practice this is almost never employed. However, because these measurements are nondestructive, sections can be relogged, subsampled, and remeasured discretely to provide greater precision, accuracy, and resolution where needed. Different susceptometers have demonstrated precision; however, little work has been done to calibrate between systems to examine data accuracy. Sagnotti et al. (2003) noted that different MS2 systems can underestimate reference samples by up to 15% and that differences between MS2 and Kappabridge systems can range from 2% to 13%. MS2 systems can be susceptible to sample volume and positioning differences between measurements with 1.5% differences between 1 and 7 cc samples of the same material. These factors coupled with simple variances in other factors, for example, changes in room temperature, can introduce measurement error. With these types of measurements becoming increasingly popular, more attention is being focused on these problems and standards and protocols are currently being developed to help address some of these issues, which will permit greater accuracy in comparisons of records across the globe.
Environmental Records of Susceptibility Compounds of iron comprise roughly 5% of the Earth’s crust and are ubiquitous in many environments in a variety of inorganic and bioavailable forms (Table 2); for example, iron oxides can constitute 2–6% of basalts (Thompson and Oldfield, 1986). Iron-bearing minerals are often eroded, transported, and deposited coevally with other minerals, nutrients, or pollutants. Thus, magnetic properties of environmental materials often correlate well with other independently determined properties. In many sedimentary environments, magnetite is the most common ferrimagnetic mineral. Given its strong response to a magnetic field (Table 2), low-field MS is often considered proportional to magnetite concentration (Dearing, 1999). However, although mineral concentration is often considered the principal factor driving MS, this is often not the case and these assumptions are not always valid. Natural samples are rarely unimodally composed and instead are aggregates of varying amounts of different minerals and grain sizes, complicating such a simple interpretation. As magnetic properties are not uniform across magnetic mineralogies or grain sizes, the interpretation of what drives MS variation is often not simply a reflection of ferrimagnetic concentration. Similar values can potentially be driven by different
compositions, for example, the presence of a small amount of magnetite or larger concentrations of hematite. MS can be diluted by diamagnetic minerals (e.g., biogenic carbonates) that are often significant components of ocean sediments. The nature of the magnetic carriers also can affect the MS. For example, authigenic bacterial magnetosomes are composed of SP to SD magnetite, while terrigenous inputs to marine systems are often PSD to MD grain sizes. The interplay of all of these factors can sometimes produce different and unique MS values characteristic of a specific environment or process. However, it is rarely possible to determine the factors driving the MS signal from MS measurements alone; a greater suite of magnetic measurements are normally necessary (e.g., Table 1). Such an evaluation is almost always required, not only to identify the absolute drivers of change, but also to recognize the causes of differences between measurements and to help with the understanding and accounting for the potential processes involved in driving these changes. However, given the abundance of Fe oxides in the environment, their ability to harmonize with natural processes, and the ability of magnetic measurements to detect subtle changes in concentration, mineralogy, and grain size, links are often drawn between measurements of MS and environmental change in many diverse settings. Lacustrine MS records often show strong linkages to other independent climate proxies, and compared to the deep-sea environment, lake records possess relatively high sedimentation rates, enabling highresolution reconstructions of climate. However, many of the most climatically sensitive lakes are found in glacially proximal environments and are thus temporally restricted due to their postglacial formation. Similarly, loess records have shown to possess remarkable MS archives of climate changes; however, their formation is quasi-continuous and restricted to certain areas of the globe. In contrast, relatively slow but continuous sedimentation rates in marine records permit the potential for long climate change reconstructions from large areas of the globe.
MS as a Paleoceanographic Proxy Like all other environmental records of susceptibility, the MS records of marine sediments are an aggregated sum of the interplay of magnetic concentration, magnetic mineralogy, and magnetic grain size. Variations in these properties can be forced by, and often be characteristic of, several different depositional processes and postdepositional (diagenetic) transformations. Depositional processes generally reflect the balance between terrigenous flux, modulated by some combination of surface and bottom currents, eolian transport, icerafted and other glacial marine processes, volcanogenic inputs, tectonic and downslope transport, and biogenic preservation. Postdepositional alterations may result in the creation, destruction, and/or transformation of magnetic minerals. All these factors can influence an MS signal and their relative influence can vary both spatially and temporally. Unraveling these often codependent influences is a nontrivial matter, but with careful selection of both site location and time period, coupled with analysis and understanding of the factors driving the MS signal, MS records of ocean sediments can potentially reveal much
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
about the paleohistory of both oceans and climate. Here we will report some of the important ways MS has been used in the study of paleoceanographic and paleoclimatic records of ocean cores and some of the important factors to consider in their interpretation.
MS as a Proxy for Terrigenous Flux The delivery and distribution of terrigenous sediments to the world’s oceans is subject to a number of complex processes. One of the major factors governing terrigenous sediment concentration in the ocean is the availability and transport of detrital grains. Transport and delivery-related processes can be, and have been, climatically mediated over a range of timescales in response to changes in climatic regimes. In environments with minimal diagenetic transformations, the MS records of ocean cores can simply reflect (and are often assumed to reflect) the composition of the terrigenous material delivered to a site. Given that the production of biogenic materials (e.g., carbonate and silica) can also be climatically dependent, the MS records of many marine sediments often reflect the balance between terrigenous flux and biogenic dilution. Common examples of this type of behavior are found in Pacific, Indian, and Atlantic Oceans. The MS of ODP sites 721 and 722 in the Arabian Sea strongly correlate with the percentage of terrigenous materials (r ¼ 0.98; Bloemendal and deMenocal, 1989). Delivery of this terrigenous fraction is linked to eolian transport and monsoon intensity. Spectral analysis of the Arabian Sea MS records and ODP site 661 from the eastern equatorial Atlantic showed strong influence of the 23 and 19 ka precession-related orbital periodicities prior to 2.4 My with a shift to 41 ka periodicities after 2.4 My, possibly reflecting ice sheet effects on monsoon climate or dust source regions (Bloemendal and deMenocal, 1989; Figure 3). Similarly for ODP site 882 in the North Pacific, Maslin et al. (1996) used MS as a proxy for ice-rafted debris (IRD), and showed synchronous changes between IRD, the d18O record, and the biogenic opal record (Figure 4). Despite the different processes responsible for delivering sediment to these two sites, both MS records identify a major climate reorganization 2.4–2.7 Ma, resulting in global changes in terrestrial flux from the continents. It is thought that these shifts are associated with the initiation of northern hemisphere glaciations and the start of the Quaternary. These long records linked to climatically mediated terrestrial fluxes helped establish the links between orbital and climatic forcings of MS records. Greater sedimentation rates in the North Atlantic, coupled with greater carbonate preservation, have permitted higher resolution studies of individual events, with their chronologies constrained by more accurate dating. Surrounded by a terrestrial regime dominated by the growth and decay of continental-scale ice sheets, bedrock geology composed of both highly magnetic basalts and crystalline basement rocks and weakly magnetic carbonates, and a sediment transport regime dominated by dynamic surface and subsurface circulations, the North Atlantic has proven to be an exceptional archive for MS studies of paleoclimate and paleoceanography. In the mid-latitude IRD belt (40 and 55 N) of the glacial North Atlantic (Ruddiman, 1977), MS has been widely used to
889
correlate IRD events between cores, illustrating the basin-wide coherence of the IRD pattern. Through glacial and interglacial cycles, Robinson (1986) showed that MS peaks were driven by fluxes of IRD in the mid-latitude North Atlantic, linked to increased glacial calving. This signal is also mediated by productivity (Figure 5). During glacial periods, increased delivery of IRD is coupled with a reduction in the production of biogenic carbonates resulting in high MS values. During interglacials, decreased delivery of IRD is coupled with increased biogenic production resulting in increased diamagnetic carbonates resulting in low MS values. This flux/dilution relationship results in strongly contrasting MS values on glacial/ interglacial timescales that are replicable across large spatial scales. The highest glacial susceptibility peaks are associated with sand layers (>150 mm) deposited during Heinrich events (Bond et al., 1992), providing a diagnostic signature of these canonical Laurentide Ice Sheet discharge events that prominently characterize the glacial North Atlantic (e.g., Bond et al., 1992). MS records south of the glacial polar front in the North Atlantic differ greatly from those within the IRD belt, reflecting the strong melting gradient and IRD-forced nature of MS. Mapping of these spatial MS signatures also permitted reconstruction of iceberg trajectory paths, paleosurface currents, and the migration of the glacial polar front. From the Southern Ocean, Pugh et al. (2009) show MS records from three cores from the Scotia Sea to be almost identical to the European Project for Ice Coring in Antarctica (EPICA) dust record (Figure 6). Again, glacial periods are characterized by high susceptibility and interglacials with susceptibility minima. However, these records are apparently not driven by IRD, but by dust export from continental landmasses. The processes involved in driving ice rafting in the North Atlantic and dust flux in the Southern Ocean are very different; however, MS records show similar trends. This similarity is forced by shifts in dominant global climatic patterns over glacial/interglacial timescales, which in turn mediates the terrestrial flux and the attendant MS record. While global climate changes can have a large impact on global sediment flux and the MS record, it is essential to interpret any MS record within the context of its location and processes acting in its formation. For example, Stoner et al. (1995) and Carlson et al. (2008) examined the magnetic and geochemical record of cores from the Eirik Drift in the NE Labrador Sea, southwest of Greenland. Unlike the majority of records from further south in the North Atlantic Basin, glacial terminations and interglacial periods are characterized by higher MS values than glacial periods. During the glacial– interglacial transitions of marine isotope stage (MIS) 2/1 and MIS 6/5, enhanced MS is accompanied by a coarsening of magnetic grain sizes and an increase in Ti/Fe concentrations. Here, minima in IRD (represented by the >125 mm fraction) and increases in the silt fraction are synchronous with increases in MS and magnetic grain sizes and are thought to be associated with sediment-laden melt water pulses associated with increased melting of the Greenland Ice Sheet during deglaciation (Carlson et al., 2008; Stoner et al., 1995). Thus, while these Greenland proximal records, like the North Atlantic IRD and Southern and Indian Ocean eolian records are driven by flux of terrigenous fragments responding to changing climate drivers, their MS signals produce very different temporal signatures.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 890
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
ODP Site 661 MS (10-6 SI units) 0
ODP Site 721 MS (10-6 SI units)
Cycles ka-1 350 0
0.08
0 0
0.5
0.5
1
1
1.5
1.5
2
2
2.5
2.5
3
3
150 0
0.08
Age (My)
0
Cycles ka-1
100 41
3.5
23 19
3.5
100 41
23 19
Figure 3 Magnetic susceptibility time series for ODP Site 661 and ODP site 721, and the variance spectra (inset figures) for time intervals between 3.5 Ma and the present. Adapted by permission from Macmillan Publishers Ltd: [Nature] (Bloemendal J and deMenocal P (1989) Evidence for a change in the periodicity of tropical climate cycles at 2.4 Myr from whole-core magnetic susceptibility measurements. Nature 342: 897–900), copyright 1989.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
5
d 18O 4
3
2
0
Opal (wt%) 50
100
0
Magnetic susceptibility 100
891
200
0
1
Age (Ma)
2
3
4
5
6 Figure 4 Oxygen isotope, opal, and MS records from ODP site 882. Adapted with kind permission from Springer ScienceþBusiness Media: Geologische Rundschau, The progressive intensification of northern hemisphere glaciation as seen from the North Pacific, 85, 1996, 452–465, Maslin MA, Haug GH, Sarnthein M and Tiedemann R Coryright (1996).
MS records driven by the flux of terrestrial fragments can yield high-resolution information about changes in climate and its effect on the terrestrial landscape. The glacial/interglacial, flux/dilution relationship in the ocean remains a major driver behind many MS records and permits comparisons over local, regional, and to some degree even global scales. However, these records must first be interpreted within the context of the processes driving these records and other factors which may potentially modify this relationship must also be assessed.
MS as a Proxy for Sediment Redistribution In the deep ocean, MS records often correlate well with IRD percentage and reflect surface currents and iceberg melt routes. However, Rasmussen et al. (1997) showed a remarkable similarity between the MS record of ENAM93-21 from the Faeroe–
Shetland Channel in the North Atlantic and the d18O of the Greenland Ice Core Project (GRIP) record (Figure 7). This MS record captured all 15 Dansgaard–Oeschger (D–O) cycles within the MIS 3 and MIS 2 intervals. In contrast to the North Atlantic IRD pattern, abrupt warming is characterized by sharp increases in MS, with gradual cooling associated with decreasing MS values, showing little relation to IRD inputs (Figure 7(c)) either in the model proposed by Robinson (1986) or regionally. MS lows are associated with colder stadial periods and Heinrich event intervals. Subsequent studies have found similar patterns along the path of dense overflow waters in the subarctic basins of the North Atlantic that contribute to North Atlantic Deep Water (NADW). Because these overflow waters have been implicated as drivers of the D–O cycles observed in the Greenland ice cores, these MS observations have implications for examination of the processes involved. MS patterns were shown to be strongly controlled by variations
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 892
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
d 18O
50
k (dimensionless SI units) 50
100
0
50
100
0
50
100
50
0
100 150
0
50
100 150
0
50
100 150
0
50
100 150
0
50
100
150
I/G 1
1
2
2
Glacial
0
0 Stages
3
3
4
Depth (cm)
100
Interglacial
4
5 5
T
150 Tephra
5
Glacial
T
200
S L U M P
6
6 T Tephra
T T
T
S8-79-8
S8-79-3
S8-79-6
I/G 7
S8-79-1
S8-79-2
S8-79-5
S8-79-7
S8-79-4
Figure 5 Correlation of cores using MS through seven glacial/interglacial stages. Reprinted from Physics of the Earth and Planetary Interiors, 42, Robinson SG, The late Pleistocene paleoclimatic record of North Atlantic deep-sea sediments revealed by mineral-magnetic measurements, 22–47, Copyright (1986), with permission from Elsevier.
MIS 1
40
2-4
5
6
7
8
9
10
30 20 10
1750
0
1500 1250 1000 750 500 250
TPC290
100 80 60 40 PS2319-1
20
k MSbulk (⫻10-5 SI)
k MSbulk (⫻10-5 SI)
0
140 120 100 80 60 40 20 0
EPICA dust conc.(mg-1)
kMSbulk (⫻10-5 SI)
50
0 0
100
200
300
Age (ka)
Figure 6 Correlation between EPICA dust concentration and three Scotia Sea cores. Note the high susceptibility in glacial phases and low MS during interglacials, similar to the IRD record. Reprinted from Earth and Planetary Science Letters, 284, Pugh R, McCave I, Hillenbrand C, and Kuhn G, Circum-Antarctic age modelling of Quaternary marine cores under the Antarctic Circumpolar Current: Ice-core dust-magnetic correlation, 113–123, Copyright 2009, with permission from Elsevier.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 893
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
Core ODP609
Core ENAM 93-21
GRIP Ice core
Depth (m) 1600
Interstadial no.
10
ASH I
YD 1 1
H1
1700
LGM
1800
2
20 H2
3 4
Years (14C ka)
H3
1900
5 6
30
2
7
2000
8
3 4
H4 9 5
10
2100
6
40
11
7 8
12
2200
9 10
H5
11
50
13 12 14 ASH II
15
5 (a)
14 15
Interstadial no.
60 4
3
d 18O (‰)
2 5 (b)
4
3
d 18O (‰)
2
0
100
200 0
IRD no./6 cm3 (c) >500 mm
2300
13
200
400
Magnetic (d) susceptibility
-43 (e)
-41
-39
-37
2400 -35
d 18O (‰)
Figure 7 Comparison of (a) d18O from ODP site 609, (b) d18O, (c) IRD grains >500 mm, (d) MS from ENAM 93-21, and (e) d18O from the GRIP ice core record. Reprinted from Quaternary Science Reviews, 16, Rasmussen TL, van Weering TCE, and Labeyrie L, Climatic instability, ice sheets and ocean dynamics at high northern latitudes during the last glacial period (58-10 ka BP), 71–80, Copyright (1997), with permission from Elsevier.
in titanomagnetite concentration (Kissel et al., 1997; Snowball and Moros, 2003) and similar patterns were shown to be spatially replicable over a transect of several cores (e.g., Kissel, et al., 2009). It is suggested that during colder phases when NADW formation was sluggish, transport and delivery of ferrimagnetic magnetite to drift sites was restricted resulting in lower MS values. Conversely, during warmer phases, NADW strength increased, as did basaltic fragment transport, increasing the MS values at these locations. Peaks and minima of MS in oceanic sediments can often be interpreted in a paleoclimatic context. However, resuspension during downslope turbidite flows and subsequent densityrelated settling of relatively heavy magnetic minerals within an otherwise magnetically weak matrix can also produce characteristic signatures. These MS peaks can be important for identifying turbidite layers and for distinguishing them from paleosignals, as these events can have severe implications for the paleointerpretation of sedimentary records (e.g. core S8-79-6 in Figure 5). MS peaks generated by these events can often be process and event diagnostic. Goldfinger et al. (2007, 2008) used MS to study paleoseismology, correlating MS peaks between earthquake-event-triggered turbidite flows over several hundred kilometers along the Cascadia margin. Earthquakegenerated turbidite deposits have not only been recorded by magnetic measurements on the Pacific margins, but also in Quebec and off the Portuguese Coast, illustrating the potential of MS for identifying these deposits.
MS and Source Variation IRD, bottom current, and slope process records of MS often assume that magnetic concentration linked to flux/dilution relationships are the principal drivers behind the MS record. However, Pirrung et al. (2002a,b) showed that MS varied over three orders of magnitude in the surface sediments of both the Nordic Seas (Pirrung et al., 2002a) and the Atlantic sector of the Southern Ocean (Pirrung et al., 2002b), linked to the source of minerals from different provinces (Figure 8). Highest values were found around Scoresby Sund and along the Iceland–Faeroe–Shetland Ridge, linked to basaltic sources in the Nordic Seas; similarly, in the Southern Ocean, highest values are proximal to mafic sources. It is often assumed that all ferrimagnetic minerals are terrigenously derived, yet all terrigenous materials are not ferrimagnetic. In the absence of a strong ferrimagnetic source component, MS values may be influenced by weaker magnetic components; such areas are evident, for example, in the Norweigan and western Barents Sea (Figure 8). Variations are evident in other regions of the North Atlantic, with Andrew and Hardardottir (2009) magnetically discriminating between Greenlandic and Icelandic sources on either side of the Denmark Strait. Stoner and Andrews (1999) also showed that sediments sourced from the Hudson Bay ice streams were carbonate-rich and this was reflected in the sediments of the Labrador Sea that possess relatively low MS compared to the rest of the North Atlantic.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 894
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
90⬚ E
90⬚ W c If (10-9 m3 kg-1)
Mafic magmatites
Surface sediment Glacier surface
80⬚ W
Palaeozoic red beds
80⬚ E
Dirty sea ice < 200 200–500 500–1000 1000–2000 > 2000
60⬚ E
70 ⬚
30 ⬚
70⬚ W
70⬚ E
50⬚ E
25 ⬚
60⬚ W 72⬚
40⬚ E
70
⬚
20⬚
50⬚ W
Sc or es by Su nd 40⬚ W Geikie Plateau
30⬚ E
trait
ark S
Denm
Ice
lan
dS
ea
Deep water and bottom currents ISDW Iceland sea D.W. GSDW Greenland sea deep water ISOW Iceland Scotland overflow water Formation of deep water
30⬚ W ISOW
Source: GEBCO (IOC, 1994) Coastline 500 m 2000 m
20⬚ W 40⬚ W
60⬚
10⬚ W 30⬚ W
20⬚ W
0⬚ W 10⬚ W
20⬚ E
10⬚ E 0⬚ W
10⬚ E
55⬚ S
20⬚ E 60⬚ S 50⬚ W 65⬚ S 30⬚ E
WF 60⬚ W
75⬚ S
40⬚ E
70⬚ W 80⬚ S c If (10-1 m-1 kg-1) Surface sediment
< 100 100–500 500–1000 1000–2000 > 2000
CF WF
50⬚ E
Crazy fan Weddell fan
Figure 8 MS variation in surface sediments from the Nordic Seas and the Atlantic sector of the Southern Ocean. Reproduced with kind permission from Springer ScienceþBusiness Media: Geo-Marine Letters, Magnetic susceptibility and ice-rafted debris in surface sediments of the Nordic Seas: Implications for Isotope Stage 3 oscillations, 22, 2002a, 1–11, Pirrung M, Fu¨tterer D, Grobe H, Matthiessen J, and Niessen F, Figure 1; Geo-Marine Letters, Magnetic susceptibility and ice-rafted debris in surface sediments of the Atlantic sector of the Southern Ocean, 22, 2002b, 170–180, Pirrung M, Hillenbrand C, Diekmann B, Fu¨tterer D, Grobe H, and Kuhn G, Figure 2. Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 895
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
Heinrich layer MS values are also low in this region, illustrating MS dilution due to diamagnetic carbonate background terrigenous concentrations. Instantaneous tephra deposits can also possess distinct magnetic signatures and can be used as stratigraphic markers (e.g., Peters et al., 2010). Watkins and Maher (2003) and Watkins et al. (2007) extended the work of Pirrung et al. (2002a) to include both potential sources and core sediments surrounding the whole North Atlantic Basin (Figure 9), both at the present day and during the Last Glacial Maximum (LGM). Their results neatly illustrated that even as sediment from the IRD belt (poleward of 40 N) during the LGM possessed higher susceptibility than present (Watkins et al., 2007), there exists significant regional variation in MS, both for the present day and at the LGM. For example, IRD originating from Icelandic basalts and the North American Shield possess characteristically high susceptibilities, and are reflected in the sedimentary properties of the Western North Atlantic. In contrast, the lower susceptibility areas identified by Pirrung et al. (2002a) are proximal to the gray and red sandstones from Spitsbergen which outcrop and surround the Norwegian and Barents Seas, and eolian dust sources are responsible for the relatively low susceptibility of the central Atlantic. With this variation defined, MS can potentially be used to map the timing and intensity of IRD events,
and under optimal conditions can provide information on their sources through the last glacial–interglacial transition (cf. Watkins and Maher, 2003; Watkins et al., 2007). However, isolating the relative influences of sediment flux and sediment source is an almost impossible process using MS alone. Instead, a wider range of magnetic measurements are often needed (see Table 1), with combinations of these parameters permitting assessment independent of magnetic concentration. Source variation and the roles of IRD and bottom water currents also rarely act in isolation of each other or indeed several other processes which may deliver or restrict transport of terrigenous fractions. Site selection and magnetic measurement which consider these processes, and an interpretation that accounts for them, are often key to successfully relate the MS signal to a certain aspect of the climate system.
Postdepositional Records of MS Magnetic minerals accumulating in sedimentary environments are often assumed by the paleocommunity to be conservative, maintaining properties that reflect their original source materials and mechanisms of deposition. In areas with high organic matter accumulation and/or in low oxygen environments, organic material may be buried before fully oxidizing, promoting
80⬚ N
70⬚ N
50⬚ N
10⬚ N
0⬚ 80⬚ W
70⬚ W
60⬚ W
50⬚ W
Core locations Sieved samples 6
3
40⬚ W
30⬚ W
20⬚ W
10⬚ W
0⬚
Basalt (Iceland)
Ash (Iceland)
Soil (Morocco) Soil (Niger)
Loess (Russia)
Loess (US)
Soil (Tunisia)
Loess (UK)
Pink sandstone (St. Lawrence)
Loess (Belgium)
20⬚ N
Grey sandstone (Spitsbergen)
Bacterial magnetite (90%) Red sandstone (N.Bylot Is.) Granite–granodiorite (Baffin Is.) Red sandstone (E.Greenland) Bacterial magnetite (75%) Loess (France) Rhyolite (Iceland) Soil (Turkey)
30⬚ N
Red sandstone (Spitsbergen)
40⬚ N
Granitic–gneiss (N.Bylot Is.) Granite (Devon Is.)
60⬚ N
100
10
1
0.1
0.01
10⬚ E
20⬚ E
30⬚ E
40⬚ E
Measured sources 0–0.2
0.2–0.5
0.5–1.0
1
1.0–6.6
Figure 9 MS values (10 m kg ) of potential sources (bar chart inset) and LGM values of MS for cores covering the North Atlantic (shaded ranges). Reprinted from Paleoceanography, 22, Watkins SJ, Maher BA and Bigg GR, Ocean circulation at the Last Glacial Maximum: A combined modeling and magnetic proxy-based study, 1–20, Copyright (2007), with permission from Elsevier.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 896
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
bacterially mediated reduction diagenesis (Froelich et al., 1979). Fine ferrimagnetic grains (either of biogenic or terrestrial origin) are most susceptible to reduction in a profile undergoing diagenetic changes. This can result in partial loss of the deposited magnetic signal and is therefore a significant issue for interpretation of MS records. Deeper in these records, soluble iron can be reprecipitated as other magnetic phases, including pyrite and greigite, and these minerals have shown to be significant components of oceanic records. These minerals can nucleate around larger surviving ferrimagnetic grains protecting their core (and magnetic signal) from further reduction. Oceanic areas producing strong diagenetic change are generally restricted to continental margins and/or where sedimentation and organic accumulation rates are high and in areas of upwelling and in partially landlocked basins (e.g., the Mediterranean and the Sea of Japan). Such areas are generally well known and while assessment of diagenetic change cannot be conclusive using MS alone, simple additional diagnostic magnetic measurements can help document evidence for diagenetic change. Distinctive signatures include steep decreases in concentration indicators (e.g., MS) with a parallel coarsening of parameters sensitive to magnetic grain size pointing to the selective loss of finer ferrimagnetic grains. As hematite and
6
Sulfur (%)
5
1
3
5
7
8
9
10
11
12
7
8
9
10
11
12
13
15
16 17 18
15
16
d 18O stage
7
goethite are less affected by suboxic diagenesis, an increase in coercivity may also accompany selective loss of finer grains. Diagenetic processes usually result in degradation of the depositional signal and the paleoclimatic information it contains. However, reduction-driven diagenetic transformations may not always result from steady-state conditions, but from changing boundary conditions providing additional opportunities for the generation of paleoinformation. For example, in an analysis of four sediment records from the Sea of Japan, light/dark rhythms in the sediment record suggested cyclic switching between times of oxic and anoxic conditions, related to glacio-eustatic sea level changes (Vigliotti, 1997). Here MS correlates with the d18O record, showing sea level-lowered glacial periods are more favorable for production of anoxic basin conditions resulting in an increase in sulfuric compounds and a reduction in magnetic concentration and coarsening of the magnetic grain size. Interglacial periods contain deeper waters, greater basin oxygenation, and greater retention of the fine ferrimagnetic fraction (Figure 10). Postdepositional transformations act not only to remove ferrimagnetic minerals from a profile, but in situ production of fine-grained authigenically derived bacterial magnetite can also potentially significantly alter the depositional record. Bacterial
4 3 2 1 0
100 d 18O stage
k (10-6 SI units)
1000
1
3
5
13
17 18
10 Time Figure 10 Sulfur (%) and MS for ODP sites 794 (open symbols) and 795 (full symbols) from the Sea of Japan through 18 marine isotope stages. Reprinted from Quaternary Science Reviews, 16, Vigliotti L, Magnetic properties of light and dark sediment layers from the Japan sea: Diagenetic and paleoclimatic implications, 1093–1114, Copyright (1997), with permission from Elsevier.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
magnetite chains have been identified as significant remanence carriers in both oxic and anoxic oceanic records (e.g., Kirschvink, 1982; Petersen et al., 1986; Stolz et al., 1990). Mechanisms controlling their formation can also possess certain boundary conditions. Lean and McCave (1998) examined a deep-sea core east of New Zealand, identifying bacterial magnetite phases and proposing climatic mediation of conditions favoring bacterial magnetite formation. However, this record is probably not reflective of all authigenic contributions, especially in areas susceptible to diagenesis. As with fine-grained ferrimagnetic constituents within terrigenous sediment, these phases are rapidly lost as they pass through the upwardly migrating Fe redox boundary by iron-reducing bacteria (Karlin et al., 1987; Robinson et al., 2000). Thus, in general, bacterial magnetosomes typically are only found in the upper few meters of sediment. These organisms produce extracellular SP magnetite, which can be readily distinguished from coarsegrained IRD or high-coercivity eolian dusts by additional magnetic measurements. However, if they go undetected in a sediment record, they have the potential to overprint the depositional magnetic signal which may result in an erroneous interpretation.
Conclusions The MS record at any one site is the sum of the influences of many factors including flux and subsequent redistribution (by numerous processes, the source and mineralogy of the terrigenous material, and dilution by diamagnetic materials). This signal can then be a modified transformation of the magnetic fraction postdepositionally. Correlations between MS and other climate proxies result from the ubiquitous nature of iron oxides in marine sedimentary environments and their sensitivity to climatically driven processes. The key is to derive which aspect of the climate system they are recording. Similar records can be produced by very different processes, which can lead to erroneous interpretations. Therefore, independent constraints are required when interpreting MS record beyond the local scale. As similar delivery processes can often produce different MS records or as postdepositional processes can affect records to different degrees, it is often necessary to couple MS measurements with other simple magnetic analyses to understand the drivers of the MS changes before attempts are made to extract any paleoinformation. In summary, MS measurements have proven to be a highly effective tool for the interpretation of paleoenvironmental change, possessing sensitivity to aspects often unresolvable by other methods. With good understanding of the reasons driving any MS changes, MS records have become part of a two-way iterative feedback process for the understanding of paleoclimate. We have to understand the climatic drivers before we can interpret the MS signal, but the MS record can provide excellent new insights into paleoclimate that may not be available using other proxies. Ongoing work to address issues of interlab calibration and recognition of the potential for increased resolution through particle-sized specific measurement will only increase the robustness and understanding of these techniques.
897
See also: Paleoceanography, Physical and Chemical Proxies: Terrigenous Sediments. Paleosols and Wind-Blown Sediments: Mineral Magnetic Analysis.
References Andrew JT and Hardardottir J (2009) A comparison of Holocene sediment- and paleo-magnetic characteristics from the margins of Iceland and East Greenland. Jokull 59: 51–66. Bloemendal J and deMenocal P (1989) Evidence for a change in the periodicity of tropical climate cycles at 2.4 Myr from whole-core magnetic susceptibility measurements. Nature 342: 897–900. Bond G, Heinrich H, Broecker W, et al. (1992) Evidence for massive discharges of icebergs into the North Atlantic Ocean during the last glacial period. Nature 360: 245–249. Carlson AE, Stoner JS, Donnelly JP, and Hillaire-Marcel C (2008) Response of the southern Greenland Ice Sheet during the last two deglaciations. Geology 36: 359–362. Dankers PH (1978) Magnetic Properties of Dispersed Natural Iron Oxides of Known Grain Size. PhD Thesis, University of Utrecht. Dearing J (1999) Magnetic susceptibility. In: Walden J, Oldfield F, and Smith J (eds.) Environmental Magnetism: A Practical Guide, pp. 35–62. London: Quaternary Research Association Technical Guide No. 6. Evans ME and Heller F (2003) Environmental Magnetism: Principals and Applications of Enviromagnetics. San Diego, CA: Academic Press. Froelich P, Klinkhammer G, Bender M, et al. (1979) Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: Suboxic diagenesis. Geochimica et Cosmochimica Acta 43: 1075–1090. Goldfinger C, Grijalva K, Burgmann R, et al. (2008) Late Holocene rupture of the northern San Andreas Fault and possible stress linkage to the Cascadia subduction zone. Bulletin of the Seismological Society of America 98: 861–889. Goldfinger C, Morey AE, Nelson CH, et al. (2007) Rupture lengths and temporal history of significant earthquakes on the offshore and north coast segments of the Northern San Andreas Fault based on turbidite stratigraphy. Earth and Planetary Science Letters 254: 9–27. Hassold N, Rea D, van der Pluijm B, Pare´s J, Gleason J, and Ravelo A (2006) Late Miocene to Pleistocene paleoceanographic records from the Feni and Gardar Drifts: Pliocene reduction in abyssal flow. Palaeogeography, Palaeoclimatology, Palaeoecology 236: 290–301. Karlin R, Lyle M, and Heath GR (1987) Authigenic magnetite formation in suboxic marine sediments. Nature 326: 490–493. Kirschvink JL (1982) Paleomagnetic evidence for fossil biogenic magnetite in western Crete. Earth and Planetary Science Letters 59: 388–392. Kissel C, Laj C, Lehman B, Labyrie L, and Bout-Roumazeilles V (1997) Changes in the strength of the Iceland-Scotland Overflow Water in the last 200,000 years: Evidence from magnetic anisotropy analysis of core SU90-33. Earth and Planetary Science Letters 152: 25–36. Kissel C, Laj C, Mulder T, Wandres C, and Cremer M (2005) The magnetic fraction: A tracer of deep water circulation in the North Atlantic. Earth and Planetary Science Letters 288: 444–454. Lean C and McCave I (1998) Glacial to interglacial mineral magnetic and palaeoceanographic changes at Chatham Rise, SW Pacific Ocean. Earth and Planetary Science Letters 163: 247–260. Maher BA (1988) Magnetic properties of synthetic magnetites. Geophysical Journal 94: 83–96. Maslin MA, Haug GH, Sarnthein M, and Tiedemann R (1996) The progressive intensification of northern hemisphere glaciation as seen from the North Pacific. Geologische Rundschau 85: 452–465. McCave IN (2007) Deep-sea sediment deposits and properties controlled by currents. In: Hillaire-Marcel C and De Vernal A (eds.) Developments in Marine Geology, vol. 1, pp. 19–62. New York: Elsevier. Ozdemir O and Banerjee SK (1982) A preliminary magnetic study of soil samples from west-central Minnesota. Earth and Planetary Science Letters 59: 393–403. Peters C, Austin WEN, Walden J, and Hibbert FD (2010) Magnetic characterization and correlation of a Younger Dryas tephra in North Atlantic marine sediments. Journal of Quaternary Science 25: 339–347. Petersen N, von Dobeneck T, and Vali H (1986) Fossil bacterial magnetite in deep-sea sediments from the South Atlantic Ocean. Nature 320: 611–615.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
Author's personal copy 898
PALEOCEANOGRAPHY, PHYSICAL AND CHEMICAL PROXIES | Magnetic Proxies and Susceptibility
Pirrung M, Fu¨tterer D, Grobe H, Matthiessen J, and Niessen F (2002a) Magnetic susceptibility and ice-rafted debris in surface sediments of the Nordic Seas: Implications for Isotope Stage 3 oscillations. Geo-Marine Letters 22: 1–11. Pirrung M, Hillenbrand C, Diekmann B, Fu¨tterer D, Grobe H, and Kuhn G (2002b) Magnetic susceptibility and ice-rafted debris in surface sediments of the Atlantic sector of the Southern Ocean. Geo-Marine Letters 22: 170–180. Pugh R, McCave I, Hillenbrand C, and Kuhn G (2009) Circum-Antarctic age modelling of Quaternary marine cores under the Antarctic Circumpolar Current: Ice-core dust–magnetic correlation. Earth and Planetary Science Letters 284: 113–123. Rasmussen TL, van Weering TCE, and Labeyrie L (1997) Climatic instability, ice sheets and ocean dynamics at high northern latitudes during the last glacial period (58-10 ka BP). Quaternary Science Reviews 16: 71–80. Robinson SG (1986) The late Pleistocene palaeoclimatic record of North Atlantic deep-sea sediments revealed by mineral-magnetic measurements. Physics of the Earth and Planetary Interiors 42: 22–47. Robinson SG, Sahota JTS, and Oldfield F (2000) Early diagenesis in North Atlantic abyssal plain sediments characterized by rock-magnetic and geochemical indices. Marine Geology 163: 77–107. Ruddiman WF (1977) North Atlantic ice-rafting: A major change at 75,000 years before the present. Science 196: 1208–1211. Sagnotti L, Rochette P, Jackson M, Vadeboin F, Dinare`s-Turell J, and Winkler A (2003) Inter-laboratory calibration of low-field magnetic and anhysteretic susceptibility measurements. Physics of the Earth and Planetary Interiors 138: 25–38. Snowball I and Moros M (2003) Saw-tooth pattern of North Atlantic current speed during Dansgaard–Oeschger cycles revealed by the magnetic grain size of Reykjanes Ridge sediments at 59 N. Paleoceanography 18: 1026–1037. Stolz JF, Lovley DR, and Haggerty SE (1990) Biogenic magnetite and the magnetization of sediments. Journal of Geophysical Research 95: 4355–4361.
Stoner JS and Andrews JT (1999) The North Atlantic as a Quaternary magnetic archive. In: Maher BA and Thompson R (eds.) Quaternary Climates, Environments and Magnetism, pp. 49–80. Cambridge: Cambridge University Press. Stoner JS, Channell JET, and Andresen C (1995) Magnetic properties of deep-sea sediments off southwest Greenland: Evidence for major differences between the last two deglaciations. Geology 23: 241–244. Tauxe L (2010) Essentials of Paleomagnetism. Berkeley, CA: University of California Press. Thompson R and Oldfield F (1986) Environmental Magnetism. London: Allen & Unwin. Vigliotti L (1997) Magnetic properties of light and dark sediment layers from the Japan sea: Diagenetic and paleoclimatic implications. Quaternary Science Reviews 16: 1093–1114. Watkins S and Maher BA (2003) Magnetic characterization of present-day deep-sea sediments and sources in the North Atlantic. Earth and Planetary Science Letters 214: 379–394. Watkins SJ, Maher BA, and Bigg GR (2007) Ocean circulation at the Last Glacial Maximum: A combined modeling and magnetic proxy-based study. Paleoceanography 22: 1–20.
Relevant Websites http://www.gmw.com/magnetic_properties/pdf/Om0409%20J_Dearing_Handbook_ iss7.pdf – Environmental Magnetic Susceptibility: Using the Bartington MS2 system. www.icdp-online.org – International Continental Drilling Program (ICDP). www.iodp.org – International Ocean Drilling Program (IODP). http://magician.ucsd.edu/Essentials/index.html – Essentials of Paleomagnetism.
Encyclopedia of Quaternary Science, (2013), vol. 2, pp. 884-898
