Magnetic parameters in paleolimnological studies in Antarctica

Abstract

Environmental magnetism and paleomagnetism methods are useful tools for climate-related and environmental changes studies, as well as for age dating. In particular, they may be applied in magnetic monitoring and to investigate the behavior, response, and history of environments like rivers, lakes, soils, etc. Although the usefulness of magnetic susceptibility in such applications is well known, in this contribution the potential and the need of using additional parameters obtained from measurements of remanent magnetization (natural, anhysteretic, isothermal), magnetic hysteresis and thermomagnetic studies are discussed. Magnetic signals from sediments and soils are often dominated by ferromagnetic minerals sensu lato: ferrimagnetic (titano)-magnetite and/ or antiferromagnetic hematite. These minerals are often not detected with standard analytical techniques due to their low concentrations (<1% wt). However, rock magnetic measurements are sensitive enough to detect and identify such minerals, with concentrations that are several orders of magnitude smaller. One important phenomenon to be taken into account is “magnetic enhancement” (Le Borgne 1955, Tite and Linington, 1975), which is caused by the in situ conversion of paramagnetic to ferrimagnetic materials. New formation of ferrimagnetic materials may occur by fermentation, heating, fire, biological activity and chemical processes (Mullins, 1977; Maher, 1986), and thus also be the result of pollutants and magnetized particles of anthropogenic origin (Chaparro et al., 2006, 2013a,b; Bu?ko et al., 2010; Marié et al., 2010; Zhang et al., 2011).

Natural systems are complex and their behavior depends on the interaction of several independent forcing factors like climate variations and human activity affecting the environment. Parameters describing indirectly a system are called “proxies” and may be derived from physical, chemical or biological properties. Magnetic proxies have the advantage to be determined with high sensitivity combined with fast laboratory processing; sample preparation is easy, laboratory instruments are of relatively low cost, and most measurements are non- destructive. Magnetic properties are well suited as proxies of: a) magnetic mineral concentration; b) magnetic grain size; c) correlation of sediment cores and soil profiles; d) geomagnetic field records; e) climatic changes records; f) pollution records; and g) environmental processes records (Maher et al., 1999; Kravchinsky et al., 2003; Chaparro et al. 2006, 2007, 2013a,b, 2015; Irurzun et al., 2006, 2014a; Blundell et al., 2009; Bidegain et al., 2009; Torrent et al., 2010; Lascu and Plank 2013; Oldfield et al. 2014; Quijano et al., 2014).

In this work, preliminary results from James Ross Archipelago lakes (NE Antarctic Peninsula) are presented. Sediment cores and water samples were taken from lake systems located on James Ross, Vega and Marambio islands (JRI, VI, MI, respectively, Fig. 1). Lake water’s pH and total dissolved solids exhibit a wide range of variation (<7 to 10 and <50 to 5000 mg/L, respectively), differentiating MI from JRI and VI in these aspects, due to the evaporation processes and a deficit of biological activity on MI. Organic and inorganic carbon contents in sediments are low and points to the dominance of siliciclastic materials. Thermomagnetic studies reveals in both JRI and VI the dominance of magnetite (Tc= 580°C) and subordinated titanomagnetite and/or iron sulfur (Tc= 290°C), goethite and hematite (Tc= 110°C and 680°C, Fig. 2). Lake Esmeralda (Fig. 1) shows a strong increase of magnetization after heating (sample ESM7-24, Fig. 2), maybe due to the initial presence of hematite or goethite and paramagnetic minerals, and their conversion to ferrimagnetic minerals.

Magnetic measurements of both JRI and VI sediments, suggest the presence of very fine ferrimagnetic minerals (magnetite, <0.1 ?m), while Esmeralda is characterized by larger particles (1-5 ?m); and MI by fine ferrimagnetic and paramagnetic minerals (0.2-1 ?m, Fig. 3). Magnetic hysteresis shows that magnetites tend to be of single domain size (SD) for JRI and Anónima (in Vega island) and of pseudo-single domain size (PSD) for Esmeralda (Fig. 4). The relation between a magnetic concentration dependent parameter (?) and a magnetic grain size dependent parameter (?ARM/?) is shown in figure 5. MI and Esmeralda exhibits a trend of increasing ? and decreasing ?ARM/?, that is, higher magnetic particle concentration is accompanied by coarser magnetic grain size. In contrast, JRI and Anónima show a linear relationship between parameters ? and ?ARM/?, indicating a correlation of magnetic mineral concentration with decreasing grain size. A multivariate statistical analysis of magnetic concentration, magnetic grain size, physic- chemical, and geochemical characteristics shows an inverse relationship between magnetic and physic-chemical variables. Moreover, statistically significant differences between lakes from each studied island are observed. These differences may be due to individual basin and lake environment characteristics and the processes (diagenetic, biological, geochemical, etc.) occurring in them.

Finally, Lake Anónima was selected to deter- mine a curve of relative paleointensity (RPI) of high quality, and to calibrate these Antarctic lake sediments through its correlation with a dated RPI curve from Lake Esmeralda (Fig. 9). Five cores were correlated using ? and ?ARM/? variations and they were converted to a common sedimentary depth, using one core as the master core. For RPI normalization, sediments must satisfy requirements proposed e.g. by Tauxe (1998): NRM/k, NRM/ARM and NRM/SIRM ratios should only depend on the intensity of the geomagnetic field. Those ratios were normalized to the unity for comparison with records from other lakes. The NRM15mT/k results (NRM15mT is the remanence left after demagnetization with 15 mT) for the Anónima cores are shown in Figure 9 and compared to the corresponding Esmeralda mean curve (Irurzun et al., 2013). Both curves show a similar behavior, with at least three possible lines of correlation. The minimum ratio NRM15mT/k at 68 cm for Anónima is clearly correlated with Esmeralda minimum ratio at the age of 7,700 cal. years. This suggests that sediments recovered from Anónima may have recorded changes of the Earth magnetic field during the last 6,000-7,000 years.

We conclude that the magnetic proxies, besides being useful environmental indicators, allow the discrimination between study areas and also to select target region for future studies (e.g. paleomag- netism, environmental changes, etc.). The RPI tech- nique is a very promising tool for dating sediments in Antarctic lakes. This study contributed to a better understanding of Antarctic lake systems from a multidisciplinary approach.