Magnetic Study of Weakly Contaminated Forest Soils

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS

A. KAPICKA1∗, N. JORDANOVA1,2, E. PETROVSKÝ1 and V. PODRÁZSKÝ3

1 Geophysical Institute, Acad. of Sci. of the Czech Republic, Bocní II, 141 00 Prague 4, CzechRepublic; 2 Geophysical Institute BAS, Sofia, Bulgaria; 3 Czech Agricultural University, Prague 6,Czech Republic (∗ author for correspondance, e-mail: [email protected], tel: +420-2-67103341)

(Received 30 October 2001; accepted 3 April 2003)

Abstract. This paper reports on magnetic and magnetomineralogical studies of forest soils fromKrkonoše (Giant Mountains) National Park in the Czech Republic. Low-field magnetic susceptib-ility was measured in 32 soil profiles using a field probe. Thermomagnetic analysis, acquisition ofremanent magnetization, alternating-field demagnetization of saturation remanence and frequency-dependent magnetic susceptibility were measured in laboratory samples from individual soil horizonsas well as on their magnetic extracts. X-ray diffraction and SEM were used to identify ferrimagneticfractions. The uppermost layer, which is dominated by magnetically soft magnetite of presumablyanthropogenic origin, can be reliably identified in soil profiles over the whole region of concern.Subsoil horizons are characterised by significantly different magnetic properties.

Keywords: anthropogenic ferrimagnetics, environmental magnetism, soil pollution

1. Introduction

Present instruments and methods enable very sensitive determination of low con-centrations of strong ferrimagnetic minerals in soils, e.g. primary lithogenic titano-magnetites, and different forms of secondary origin.

Possible mechanisms of magnetic enhancement of soils due to increased con-centrations of secondary ferrimagnetic minerals are discussed in, e.g., Mullins(1977), Maher and Taylor (1988), Stanjek et al. (1994) and Singer et al. (1996). At-mospherically deposited ferrimagnetic particles of anthropogenic origin also con-tribute a great deal to the concentration-dependent magnetic properties of top soils,such as low-field magnetic susceptibility (e.g., Thompson and Oldfield, 1986). Thehighest concentration of anthropogenic ferrimagnetic particles is usually found inhumic or fermentation layers, located right under the litter horizon in forest soils(e.g. Strzyszcz, 1989).

Practically all industrial fly ashes contain a significant fraction of ferrimagneticparticles, the most important sources being fly ashes produced during combustionof fossil fuel (e.g., Flanders, 1994; Kapicka et al., 2000). Other sources, such asiron and steel works, cement works, public boilers and road traffic also contributeto contamination by anthropogenic ferrimagnetics (e.g., Heller et al., 1998; Schol-ger, 1998; Hoffmann et al., 1999). In contrast to particles of pedogenic origin,anthropogenic ferrimagnetics are characterised by specific morphology and dis-

Water, Air, and Soil Pollution 148: 31–44, 2003.© 2003 Kluwer Academic Publishers. Printed in the Netherlands.

32 A. KAPICKA ET AL.

tinct magnetic properties. They are often observed in the form of spherules, withthe magnetic phase frequently sintered on aluminium silicates (e.g., mullite) oramorphous silica. Prevailing ferrimagnetic phases are Fe-oxides, namely magnetiteand maghemite, with Fe ions very often substituted by other cations (Strzyszcz etal., 1996).

Recently, rock-magnetic methods have been applied to modern soils in severalenvironmental studies (for an overview see, e.g., Petrovský and Ellwood, 1999).Application of the comparatively simple technique of measuring magnetic suscept-ibility enables delineation of areas with concentrations of deposited anthropogenicferrimagnetics significantly above background values. Magnetic mapping thus rep-resents a rapid, sensitive and cheap tool for targeting the areas of interest. Meas-urements of low-field magnetic susceptibility of surface soils have been appliedrecently around local pollution sources (e.g., Strzyszcz, 1993; Kapicka et al., 1999;Petrovský et al., 2000; Hoffmann et al., 1999). At a larger, regional scale, areas inPoland and Great Britain have been investigated (Hay et al., 1997; Strzyszcz etal., 1996; Heller et al., 1998). These studies showed that in polluted areas, themagnetic susceptibility of surface soil layers is considerably higher. At the sametime, typical ferrimagnetic particles of anthropogenic origin were identified.

Up to now, magnetic methods have been primarily used in areas with relativelyhigh concentrations of anthropogenic particles in soils. For example, in areas witha high concentration of industry, the annual amount of atmospherically depositeddust reaches several thousands of tons. A single coal-burning power plant can pro-duce hundreds of tons of fly ash per year (e.g., Heller et al., 1998). Since such flyash contains some 9% of ferrimagnetic particles (Kapicka et al., 1999), it is obviousthat these particles can significantly influence magnetic properties of soils in thesurrounding areas. However, the problem of the reliability of magnetic mappingremains more problematic in relatively unpolluted areas where the contribution ofanthropogenic ferrimagnetic particles to total magnetic properties can be relativelylow.

The Krkonoše National Park, located in northeast Bohemia, represents an areawith a relatively low pollution impact. There are no major sources of pollution inits near neighbourhood. The soil is potentially contaminated due to long-distanceatmospheric transport of dust from power plants in southern Poland and northeastBohemia, or from smaller local sources on the margin of the park area. In our study,we present results of detailed magnetic and magnetomineralogic investigations ofa set of soil profiles in this region as well as of soil magnetic extracts.

2. Methods

Detailed magnetic investigations were carried out at a permanent net of 32 open soilpits, used for soil monitoring, distributed evenly over the territory of the KrkonošeNational Park. The depth of individual pits varies from 40 to 60 cm and in most

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS 33

cases includes a layer of basement rock, or soil dominated by the basement rock(C horizon). The pits were available for both in situ measurements and samplingfor subsequent laboratory measurements. Our investigations were aimed at mag-netic classification of individual soil (sub)horizons and at elucidating the prob-lem of magnetic discrimination of the potentially contaminated surface layer and,eventually, at resolving the depth distribution of anthropogenic magnetic particles.

At each pit, low-field magnetic susceptibility was measured in situ, using a Bart-ington MS2 susceptibility meter with MS2F stratigraphic probe, resulting in an al-most continuous vertical profile of magnetic susceptibility. Some 160 samples werealso collected for further laboratory examination; sampled soil subhorizons wereclassified in cooperation with pedologists. The samples were air-dried and sievedat 2 mm in order to remove rock fragments. Laboratrory measurements of low-field magnetic susceptibility were performed using an Agico KLY-3 kappabridge.Frequency-dependent magnetic susceptibility, defined as kFD(%)= (klf - khf)/klf,where klf and khf represent susceptibility values measured at 0.47 and 4.7 kHz,respectively, was measured using a Bartington MS2 dual-frequency meter. In or-der to determine magnetic mineral phases, low and high temperature dependenceof magnetic susceptibility was measured using an Agico KLY-3 kappabridge andCS-3 furnace. Alternating-field demagnetization was performed using a SchonstedGSD-1 demagnetizer and isothermal remanent magnetization (IRM) was measuredusing a JR-5 spinner magnetometer. Magnetic extracts were obtained from soilssuspended in isopropanol by using a permanent hand magnet. To avoid coagu-lation of particles, magnetic separation was carried out in an ultrasound device.Mineralogy and chemical composition of the extracts were determined by X-raydiffraction, SEM and microprobe analysis.

3. Results and Discussion

The soils sampled represent typical forest soils, most being podzols, developedon granite/biotite granite or mica schist/phyllite. Other soil profiles are Cambisolsdeveloped on different parent rocks (biotitic granite at sites 1, 8, 9; metadiabaziteat sites 30, 31, 32; and gneiss/phyllite at sites 6, 7, 28, 29). All the soils are stronglyacidic (pH between 4–5 in the upper soil horizons).

Changes in low-field magnetic susceptibility down typical soil depth profilesare shown in Figure 1. Both in-situ measurements, (in most cases carried out ontwo parallel profiles within one soil pit), and laboratory measurements on samplescollected from individual soil horizons, are shown. Typical susceptibility beha-viour is that shown for profiles No 13, 16, 20, 25, 28 and 29, considered to berepresentative. A significant increase of magnetic susceptibility in the uppermost,organic L-F and Ah horizons was commonly observed. Although in some profiles,individual soil layers can be less developed or some of them can be missing,increased susceptibility was only observed in depths between 4–6 cm below the

34 A. KAPICKA ET AL.

surface. In deeper layers, in particular in B and C horizons, magnetic susceptibilitywas considerably less. Maximum values of volume magnetic susceptibility in theKrkonoše National Park varied from 30 to 60 × 10−5 SI units, while those in deeperlayers were between 10 and 20 × 10−5 SI units. A very significant and well-expressed maximum of magnetic susceptibility was observed in several profilesfrom ombrotrophic peat bogs (Figure 1, No 4). A maximum of about 40 × 10−5

SI units was found at a depth of 4 cm, while the deeper layers showed a muchlower and practically constant value of 0–3 × 10−5 SI units. Only four Cambisolprofiles exhibited susceptibility behaviour, as shown for profile No 31 in Figure 1,with a comparatively weak maximum in the subsurface layer and a systematicallyincreasing susceptibility with increasing depth.

In order to interpret correctly magnetic properties of soils, careful evaluation ofthe contribution due to natural pedogenic/lithogenic processes and anthropogenicinput is required. Two main pathways of pedogenic formation of strongly mag-netic iron oxides in soils are considered (e.g. Dearing et al., 1996, Maher, 1998)– induced hydrolysis and biologically induced mineralisation (BIM). For the firstmechanism, at room temperature and near-neutral pH (Taylor et al., 1987), finemagnetite grains are produced while at low pH, lepidocrocite and goethite form.Since all the soils studied are strongly acid (pH < 5), this mechanism of formationof a strongly magnetic phase does not seem to be likely. The second pathway forpedogenic formation of magnetite is described by Dearing et al. (1996). In thiscase, a high Fe-II supply, organic matter and a locally anoxic microenvironment arenecessary, and this transformation is most effective at pH 6. Cornell and Schwert-mann (1996) point out that in the case of podzolic soils the only Fe forms foundare ferrihydrite and Fe-humic complexes and, when organic matter is present inextremely high amounts, all the Fe is organically complexed, so that no oxides canform. In accordance with these results, magnetic susceptibility measurements onpodzol profiles (Maher, 1998) show strong magnetic depletion in the uppermosthorizons and higher concentrations of secondary, re-precipitated iron oxides insubsoil horizons. For the podzolic soils studied here, the two main conditions forpedogenic magnetite formation – high Fe supply and organic matter – are met.However, as pointed out above, low pH is not favourable for magnetite formation,although it cannot be entirely excluded. Low magnetic susceptibility values, meas-ured also in the B and C horizons of podzols (Figure 1) are most probably due toadvanced acidification and destruction of the primary minerals, products of whichare transported outside the soil profile. Cambisols are characterized by slight ormoderate weathering of the parent material, therefore for some of these profiles(Figure 1) we obtained increasing magnetic susceptibility with depth, reflectingthe influence of parent material. In contrast, the specific pattern of susceptibility indepth profiles of peat bogs (Figure 1) results from negligible lithogenic contribu-tion and these sites are obviously particularly suitable for the purpose of magneticmonitoring of contamination due to deposition of industrial dust.

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS 35

Fig

ure

1.C

hang

esin

low

-fiel

dm

agne

tic

susc

epti

bili

tyw

ith

dept

hfo

rty

pica

lsoi

lpro

file

s.

36 A. KAPICKA ET AL.

As a result of pedogenic precipitation of strongly magnetic iron oxides, a mix-ture of superparamagnetic (SP) and single-domain (SD) magnetite grains (e.g.,Oldfield et al., 1985) is produced, showing significantly high values of the frequency-dependent magnetic susceptibility (Thompson and Oldfield, 1986; Eyre, 1997).Values of kFD%, obtained for our samples of organic soil horizons, are relativelylow, below 4% (Figure 1). Our results are consistent with the data for podzolicsoils, described in Maher (1998), also showing negligible frequency dependenceof susceptibility. Relatively acid conditions in the soils studied and generally lowvalues of frequency-dependent susceptibility indicate that increased magnetic sus-ceptibility in the surface layers is most probably due to a higher concentration ofcoarse-grained ferrimagnetics, deposited from atmospheric dust. Coarse-grainedparticles of anthropogenic origin are not subjected to rapid dissolution becausetheir surface/volume ratio is not as high as for pedogenic ultra-fine grains, whichare easily dissolved under reducing conditions (Cornell and Schwertmann, 1996).

In all soil profiles, we observed the maximum of susceptibility in F, H or Ah soilhorizons, but not in an L horizon. This typical distribution of susceptibility can beexplained by considering different binding conditions of atmospheric dust in theselayers. The litter (L) horizon contains undecayed organic material, so that depositedatmospheric particulates are only physically held on the surface of organic remains(needles, leaves) and could be mechanically washed out and transported down theprofile. Deeper F, H or Ah horizons are characterised by high geochemical activity,so that iron oxide particles are strongly bound to the organic-mineral matter. Thiscould explain the fact that we usually observed maximum magnetic susceptibilityat 4–6 cm below the soil surface.

In order to verify the above conclusions derived from the magnetic susceptib-ility measured in the field, experiments were carried out to identify and describein more detail the ferrimagnetic phases present in individual soil layers. Specialattention was devoted to the top soil layer characterised by increased susceptibilityvalues. Thermomagnetic curves (temperature- dependence of magnetic susceptib-ility) of individual Ah, Ae, B1 and B2 horizons from soil pit No 29 (Cambisol) areshown in Figure 2a. Very similar K(T) curves were obtained also for the podzolicsoil sites. The upper soil layer shows the presence of a magnetite-like phase withTc of about 580 ◦C. The local maximum of susceptibility below this temperaturecan be interpreted either as a Hopkinson peak, or as the creation of a new magnetitephase during heating. Deeper pedogenic horizons (Ae, B1, B2) show very differentand more complex behaviour of magnetic susceptibility with increasing temperat-ure. It could be supposed that the high-temperature susceptibility behaviour in thesesubsoil horizons (and especially B horizons, where the secondary transformationproducts have accumulated) reflects thermal transformations of ferrihydrite, sinceits preservation in soils is promoted by low pH and the presence of Si and Al cations(Cornell and Schwertmann, 1996). The other mineral which possibly forms in suchacid media is goethite. The concentration of the initial ferrimagnetic phase in thesesubsoil horizons is very low and the heating curve is dominated by the creation of

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS 37

secondary ferrimagnetics. The maximum above 500 ◦C could reflect neoformationof magnetite as a result of thermal alteration of Fe-rich clay minerals. Moreover,one could suggest creation of another, yet undetermined, magnetic phase at about260 ◦C in the B-horizon. Neoformation of secondary ferrimagnetics was also con-firmed by cooling curves, ending at room temperature at values much higher thanthe initial ones. In Figure 2b, heating curves related to potentially contaminatedtop layers of several soil pits are compared. Despite the fact that these samples arefrom different podzols and Cambisols from the whole of the Krkonoše territory,their behaviour is very similar, suggesting the presence of magnetite with Tc of580 ◦C and also, probably maghemite, taking into account the significant drop inK after ∼ 250 ◦C, usually interpreted in soils as transformation of the maghemiteto hematite (e.g. Dunlop and Özdemir, 1997).

IRM acquisition curves of samples of individual horizons from typical depthprofiles are shown in Figure 3. The curve corresponding to the top layer is clearlydistinct, reaching saturation of remanence at a magnetic field of 200 mT, indicatingthe prevalence of a magnetically soft, magnetite-like phase. On the other hand,samples from deeper horizons contain significant portions of a magnetically hardphase and saturate at a much higher field, above 400 mT. Figure 4 compares IRMacquisition curves for topsoil samples from different soil pits and peat samplesscattered over the whole territory of concern. As in the case of thermal behaviour ofmagnetic susceptibility (Figure 2b), these curves are also very similar to each other.A low saturation magnetic field indicates predominance of a soft phase, likely tobe a multi-domain magnetite. The concentration of this phase differs at differentsites. Curves of alternating-field demagnetization of SIRM, shown also in Figure4, can be interpreted in the same way as IRM acquisition curves. Basically, theyconfirm a very soft magnetic phase present in the top layers, which is similar in allsoils across the area studied.

Mineralogical and chemical analyses were carried out on magnetic extractsfrom topsoil horizons showing the maximum magnetic susceptibility. Samples withlow magnetic susceptibility from deeper soil horizons were also analysed for com-parison. X-ray diffraction and SEM analyses were carried out on samples 16F +H, 4F, 12F, 29Ah, 4H, 16 Ae, 12H, and 16Bs. Magnetic extracts from topsoillayers contain predominantly magnetite/maghemite, although hematite was alsoidentified (Figure 5). The diffraction data also suggest the presence of traces ofthe Fe-oxyhydoxide, akaganeite, though its origin is not clear. High-temperatureand low-temperature behavior of magnetic susceptibility for the magnetic extractof a fermentation horizon is shown in Figure 6. A clear Verwey transition (∼ –150◦C) on the low-temperature K(T) curve and a Tc ∼ 580 ◦C on the high-temperaturecurve (Figure 6) show unambiguously that magnetite is the dominant ferrimagneticphase in magnetic extracts from the topsoil horizons.

SEM analysis showed that anthropogenic Fe-oxide particles of typical sphericalshape are abundant in the uppermost soil horizons. The concentration of anthro-pogenic ferrimagnetics in the uppermost soil layers is relatively high. They were

38 A. KAPICKA ET AL.

Figure 2. a) Temperature dependence of magnetic susceptibility (heating only) for samples fromdepth soil profile No.29, and, b) the same for top-soil samples from different test sites.

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS 39

Figure 3. IRM acquisition curves for samples of individual horizons from a typical depth soil profile.

Figure 4. IRM acquisition curves and AF demagnetization of SIRM for top-soil samples fromdifferent test sites.

40 A. KAPICKA ET AL.

Figure 5. X-ray diffractogram of magnetic extract from a top soil layer.

easily identified in all the samples studied (five individual subsamples were ana-lysed from each layer) and comprise a dominant part of the ferrimagnetic Fe-oxides(Figure 7, Table I). The situation is different in the deeper soil horizons. Despitethe fact that they are immediately below the contaminated uppermost soil horizons,it was practically impossible to identify anthropogenic ferrimagnetics by means ofSEM observations. In cases where some approximately spherical particles wereobserved in extracts from these horizons, phase analyses showed that they do notcontain Fe-oxides (Table I).

Magnetic and thermomagnetic analyses (Figures 3, 4 and 6) and X-ray diffrac-tion confirmed that the dominant ferrimagnetic component in the upper organicsoil horizons in Krkonoše National Park is magnetically soft coarse-grained mag-netite. Our results are consistent with recently published results by Magiera andStrzyszcz (2000) dealing with weakly contaminated soils in Polish national parks.SEM analysis of magnetic extracts clearly showed that this magnetite is of an-thropogenic origin and causes the enhanced magnetic susceptibility. Therefore,the surface magnetic susceptibility of forest soils in the Krkonoše region reflectsmainly the concentration of anthropogenic ferrimagnetics.

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS 41

Figure 6. Low- and high-temperature dependence of magnetic susceptibility for a soil magneticextract.

42 A. KAPICKA ET AL.

Figure 7. Typical concentration of anthropogenic Fe-oxides in upper soil layer 4 F (numberingcorresponds to Table I).

4. Conclusions

Magnetic mapping has been recently successfully used in pollution studies primar-ily in areas with intensive industrial activity and major pollution sources. In con-trast, this paper examines a comparatively unpolluted area of Krkonoše Nationalpark. Our results confirm that over the whole area in concern the top-soil layer ismagnetically enhanced in terms of magnetic susceptibility. This layer is limitedto depths of 4–6 cm below the soil surface, usually in F, H or top of Ah soilhorizons. Magnetic properties of the top layers are consistent over the whole territ-ory of the Park, independently of the particular soil type (podzol, Cambisol, peatbog). Magnetomineralogical analysis supported by X-ray diffraction, microprobeanalysis and direct microscope observation confirm that the top soil horizons aredominated by coarse-grained magnetically soft magnetite of anthropogenic origin.This anthropogenic phase is responsible for strong enhancement of magnetic sus-ceptibility in the uppermost organic horizons. Magnetomineralogy of deeper soillayers is more complex and different from site to site.

Because the magnetic method of soil- pollution monitoring is sensitive, fast andrelatively cheap, it can be applied over large regions. Our results, based on detailed

MAGNETIC STUDY OF WEAKLY CONTAMINATED FOREST SOILS 43

TABLE I

Microprobe analysis of spherical particles in topsoil F (0–5 cm) and subsoil H (5–13 cm)layers from profile No 4

Sample 4F Sample 4H

No 1 No 2 No 3 No 4 No 5 No 1 No 2

Elem: Wt% Wt% Wt% Wt% Wt% Oxide: Wt% Wt%

O K 32.53 34.55 38.9 36.14 42.88 Na2O 33.99 3.31

MgK 0.19 0 0.61 0.17 MgO 4.82 2.88

AlK 0.79 0 8.23 0.13 0.36 Al2O3 3.02 4.16

SiK 0.55 0.44 17.44 0.36 0.63 SiO2 7.26 9.85

S K 0.2 0.1 0.26 0.13 0.29 SO3 6.59 3.92

ClK 0.17 0.13 0.11 0.1 0.08 Cl2O 20.06 3.76

K K 0.2 0.16 0.58 0.07 0.1 K2O 11.94 2.59

CaK 5.5 0.82 1.49 0.35 0.82 CaO 11.31 67.1

TiK 0.27 0.32 0.78 0.14 0.15 TiO2 0.33 0.82

MnK 0.74 0 0 0.53 0

FeK 58.87 63.49 31.61 61.89 54.69 Fe2O3 0.68 1.61

Total 100 100 100 100 100 Total 100 100

laboratory investigation of soils from the Krkonoše National Park indicate that thismethod can be used also in areas with relatively lower pollution levels.

Acknowledgements

The help of Dr. S. Vacek, and Mr. J. Soucek (Research Institute of Forestry inOpocno) with sampling and pedological determination of individual pedozones ishighly acknowledged. Our thanks are due to Dr. K. Melka and Dr. A. Langrova(Geological Inst. ASCR in Prague) for X-ray diffractometry and microprobe meas-urement. This study was supported by Grant Agency of the Academy of Sciencesof the Czech Republic through grant No.A3012905/1999 and Grant Agency of theCzech Republic through grant No.205/00/1349.

References

Cornell, R. and Schwertmann, U.: 1996, The Iron Oxides. Structure, Properties, Reactions, Occur-rence and Uses, VCH Verlagsgesellschaft, 338–347.

Dearing, J. A., Hay, K. L., Baban, S. M. J., Huddleston, A. S., Wellington, E. M. H. and Loveland,P. J.: 1996, ‘Magnetic susceptibility of soil: an evaluation of conflicting theories using a nationaldata set’, Geophys. J. Int. 127, 728–734.

44 A. KAPICKA ET AL.

Dunlop, D. J., Özdemir, Ö.: 1997, Rock Magnetism: Fundamentals and Frontiers, CambridgeUniversity Press.

Eyre, J. K.: 1997, ‘Frequency dependence of magnetic susceptibility for populations of single domaingrains’, Geophys. J. Int. 129, 209–211.

Flanders, P. J.: 1994, ‘Collection, measurements and analysis of airborne magnetic particulates frompollution in the environment’, J. Appl. Phys. 75, 5931–5936.

Hay, K. L., Dearing, J. A., Baban, S. M. J., Loveland, P. J.: 1997, ‘A preliminary attempt to identifyatmospherically-derived pollution particles in English topsoils from magnetic susceptibilitymeasurements’, Phys. Chem. Earth 22, 207–210.

Heller, F., Strzyszcz, Z. and Magiera, T.: 1998, ‘Magnetic record of industrial pollution in forest soilsof Upper Silesia, Poland’, J. Geophys.Res. 103, 17,767–17,774.

Hoffmann, V., Knab, M. and Appel, E.: 1999, ‘Magnetic susceptibility mapping of roadsidepollution’, Journ. of Geochem. Explor. 66, 313–326.

Kapicka, A., Petrovský, E., Ustjak, S. and Machácková, K.: 1999, ‘Proxy mapping of fly-ash pollu-tion of soils around a coal-burning power plant: a case study in the Czech republic’, J. Geochem.Explor. 66, 291–297.

Kapicka, A., Jordanova, N., Petrovský, E. and Ustjak, S.: 2000, ‘Magnetic stability of power-plantfly ash in different soil solutions’, Phys. Chem. Earth (A) 25(5), 431–436.

Magiera, T. and Strzyszcz, Z.: 2000, ‘Ferrimagnetic minerals of anthropogenic origin in soils of somePolish national parks’, Water, Air, Soil Pollut. 124, 37–48.

Maher, B. A. and Taylor, R. M.: 1988, ‘Formation of ultrafine – grained magnetite in soils’, Nature336, 368–370.

Maher, B. A.: 1998, ‘Magnetic properties of modern soils and Quaternary loessic paleosols:palaeoclimatic implications’, Palaeogeogr. Palaeoclim. Palaeoecol. 137, 25–54.

Mullins, C. E.: 1977, ‘Magnetic susceptibility of the soil and its significance in soil science: Areview’, J. Soil. Sci. 28, 223–246.

Oldfield, F., Hunt, A., Jones, M. D. H., Chester, R., Dearing, J. A., Olsson, L. and Prospero, J. M.:1985, ‘Magnetic differentation of atmospheric dusts’, Nature 317, 516–518.

Petrovský, E., Kapicka, A., Jordanova, N., Knab, M. and Hoffmann, V.: 2000, ‘Low-field mag-netic susceptibility: a proxy method of estimating increased pollution of different environmentalsystems’, Environ. Geol. 39, 312–318.

Petrovský, E. and Ellwood, B. B.: 1999, ‘Magnetic Monitoring of Air-land and Water-Pollution’,in B. A. Maher and R. Thompson (eds), Quaternary Climates, Environments and Magnetism,Cambridge Univ. Press.

Scholger, R.: 1998, ‘Heavy metal pollution monitoring by magnetic susceptibility measurementsapplies to sediments of the river Mur (Styria, Austria)’, Eur. J. Environ. Eng. Geophys. 3, 25–37.

Singer, M. J., Verosub, K. L. and Fine, P.: 1996, ‘A conceptual model for enhancement of magneticsusceptibility of soils’, Quatern. Int. 34–36, 243–248.

Stanjek, H., Fassbinder, J. W. E., Vali, H., Wagele, H. and Graf, W.: 1994, ‘Evidence of biogenicgreigite (ferrimagnetic Fe3S4) in soil’, Eur. J. Soil. Sci. 445, 97–104.

Strzyszcz, Z.: 1989, ‘Ferromagnetic Properties of Forest Soils Being under Influence of Indus-trial Pollution’, in J. B. Bucher and I. Bucher-Wallin (eds), Air Pollution and Forest Decline,Birmensdorf, 201–207.

Strzyszcz, Z.: 1993, ‘Magnetic Susceptibility of Soils in the Areas Influenced by IndustrialEmissions’, in Soil Monitoring, Monte Verita, Birkhääuser Verlag, Basel, 225–269.

Strzyszcz, Z., Magiera, T. and Heller, F.: 1996, ‘The influence of industrial immisions on themagnetic susceptibility of soils in Upper Silesia’, Studia Geoph. Geod. 40, 276–286.

Taylor, R., Maher, B. and Self, P.: 1987, ‘Magnetite in soils: I. The synthesis of single-domain andsuperparamagnetic magnetite’, Clay Minerals 22, 411–422.

Thompson, R. and Oldfield, F.: 1986, Environmental Magnetism, Allen and Unwin, London.