Magnetic Properties From Soil From Sites With Different Geological and Enviromental Settings - 1-s2.0-S0926985105000935-Main

www.elsevier.com/locate/jappgeo

Journal of Applied Geophys

Magnetic properties of soils from sites with different geological

and environmental settings

Hana Fialova a,*, Gunter Maier b, Eduard Petrovsky a, Ales Kapicka a,

Tetyana Boyko b, Robert Scholger b

MAGPROX Team

a Geophysical Institute ASCR, Bocnı II/1401, 141 31 Prague 4, Czech Republicb Department of Geophysics, University of Leoben, Peter Tunner Str.25, A-8700 Leoben, Austria

Received 21 March 2005; accepted 26 October 2005

Abstract

Measurements of magnetic susceptibility of soils, reflecting magnetic enhancement of topsoils due to atmospherically deposited

magnetic particles of industrial origin, are used recently in studies dealing with outlining polluted areas, as well as with

approximate determination of soil contamination with heavy metals. One of the natural limitations of this method is magnetic

enhancement of soils caused by weathering magnetically rich parent rock material. In this study we compare magnetic properties of

soils from regions with different geological and environmental settings. Four areas in the Czech Republic and Austria were

investigated, representing both magnetically rich and poor geology, as well as point-like and diffuse pollution sources. Topsoil and

subsoil samples were investigated and the effect of geology and pollution was examined. Magnetic data including mass and volume

magnetic susceptibility, frequency-dependent susceptibility, and main magnetic characteristics such as coercivity (Hc and Hcr) and

magnetization (Ms and Mrs) parameters are compared with heavy metal contents. The aim of the paper is to assess the applicability

of soil magnetometry under different geological-environmental conditions in terms of magnetic discrimination of dominant

lithogenic/anthropogenic contributions to soil magnetic signature. Our results suggest that lithology represents the primary effect

on soil magnetic properties. However, in case of significant atmospheric deposition of anthropogenic particles, this contribution can

be clearly recognized, independent of the type of pollution source (point-like or diffuse), and discriminated from the lithogenic one.

Different soil types apparently play no role. Possible effects of climate were not investigated in this study.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Magnetic susceptibility; Heavy metals; Soils; Pollution; Atmospheric deposition; Lithology

1. Introduction

The need for fast and cheap screening and monitor-

ing tools of industrial pollution caused that increased

number of studies deal with magnetic methods as an

0926-9851/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.jappgeo.2005.10.006

* Corresponding author. Tel./fax: +420 267 103 332.

E-mail address: [email protected] (H. Fialova).

approximate tool to detect and characterise environ-

mental pollution (e.g., Dearing et al., 1996; Kapicka

et al., 1999, 2001a,b, 2003; Petrovsky and Ellwood,

1999; Hoffmann et al., 1999a,b; Magiera and Strzyszcz,

2000; Petrovsky et al., 2000; Hanesch and Scholger,

2002; Schibler et al., 2002; Veneva et al., 2004, and

others). Measurements of magnetic susceptibility of

soils proved to be suitable, under certain circumstances,

ics 59 (2006) 273–283

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283274

for spatial delineation of polluted and unpolluted

regions. This method is based on the assumption that

industrial processes, such as combustion of fossil fuel,

produce fly ashes with significant portion of magnetic

minerals (Flanders, 1994, 1999). These are transported

through atmospheric pathways and deposited on the

ground. In soils, such particles penetrate downwards

and accumulate in top layers, and their increased con-

centration can be easily detected using surface magnetic

measurements (e.g., Lecoanet et al., 1999).

In several studies, significant correlation between

magnetic susceptibility and heavy metal content in

soils was found (e.g., Heller et al., 1998; Dearing et

al., 2001; Lecoanet et al., 2001; Hanesch et al., 2003;

Jordanova et al., 2003). Using geochemical analysis,

additional data can be obtained, and in combination

with magnetic data, polluted areas can be well out-

lined and geologic/anthropogenic anomalies identified

(Hanesch and Scholger, 2002). Lecoanet et al. (2003)

used magnetic parameters only in order to discriminate

individual sources of soil contamination. Thus, mag-

netic susceptibility can serve as an indicator of soil

contamination.

However, only few studies attempted to solve the

main limitations of the soil magnetometry. Kapicka et

al. (2000, 2001a,b) studied stability of magnetic prop-

erties of fly-ash particles under different soil conditions.

Possibilities of the method to be applied in relatively

clean areas were investigated (Kapicka et al., 2003).

Competition of different contributions to soil popula-

tion of magnetic minerals was studied using statistical

analysis of large datasets (Dearing et al., 1996; Hanesch

et al., 2001; Hanesch and Scholger, 2002).

In practical field measurements, lithogenic and an-

thropogenic contributions can be assessed by in-situ

measurements of vertical distribution of magnetic sus-

Fig. 1. Location of the four investigated areas on the basis of MAGPROX

ceptibility in soil columns using new SM400 instru-

ment (Petrovsky et al., 2004). The effect of lithology

and soil type on magnetic susceptibility of soils was

studied by Hanesch and Scholger (2005). On the other

hand, Magiera et al. (in press) analysed some 600

vertical profiles of soil magnetic susceptibility and

distinguished 7 main classes of profiles, independent

of lithology and soil type.

In this study we examine the applicability of magnetic

measurements of soils to discrimination of anthropogen-

ic and lithogenic contributions in areas characterised by

different geological and environmental settings. It is of

great importance to show that the same set of magnetic

parameters and measurements can be applied under

various circumstances, and to provide certain general

guidelines in interpreting magnetic data in terms of

anthropogenic and lithogenic contributions to magnet-

ic-mineral population in soil columns. In this way we

intend to make further step towards standardization of

magnetic measurements of soils in terms of pollution

studies.

2. Methodology

2.1. Area description, field work and soil sampling

Based on large-scale magnetic mapping (with a

mesh of 10 km) within a 5FP EU RTD Project MAG-

PROX, four areas were selected, characterized by (mag-

netically) different underlying geology and by different

environmental settings (Fig. 1). These areas were

mapped in detail with a mesh of about 500 m. In the

Czech Republic, areas close to towns of Prıbram (Cen-

tral Bohemia) and Ostrava (North of Moravia) were

studied (Fig. 1). In the Prıbram area pollution due to

atmospherically deposited dust is relatively low, but this

topsoil magnetic susceptibility map (10�5 SI units, mesh of 10 km).

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283 275

area is well known for uranium and lead ore mining in

the past. Geology is rich on iron oxides, with basaltic

rocks such as granodiorites and gabbros. The Ostrava

region is well known for intense coal mining and

related heavy industry (e.g. power plants, steel works)

and thus belongs to the most polluted areas in the Czech

Republic. From geological point of view this region is

poor in Fe-bearing rocks; there are only sedimentary

rocks like sandstones. The Prıbram area was 10 by 15

km large and 42 soil profiles were investigated, while

the Ostrava area was 6 by 10 km large and 29 soil

profiles were examined (Fialova, 2004). For the pur-

pose of this paper, 6 and 8 most representative vertical

profiles from the Prıbram and Ostrava regions, respec-

tively, were analysed in detail. From pedological point

of view (according worldwide soil classification system

of FAO/UNESCO), the Prıbram region contains mostly

dystric and euric cambisols, dystric planosols, euric

cambisols and euric gleysols. The Ostrava region con-

tains mostly albogleyic luvisols, euric gleysols and

gleyic fluvisols (Nemecek, 2001).

The studied areas in Austria (Fig. 1) are situated

around Linz and around Breitenau (Maier and Schol-

ger, 2004). The city of Linz (capital of Upper Austria)

is a highly industrialized region, which is known for

steel production and processing, chemical industry,

etc. In addition to that, Linz is an important traffic

junction for cars as well as for railway. As the region

is situated in the lowlands, mixed influence of several

pollution sources can be expected. The main geolog-

ical units of the area are metamorphites (migmatites)

and granitoides. The studied area covers 10 by 15 km.

The area of Breitenau is situated in a narrow mountain

valley in Styria with a suspected point emission source

caused by magnesite production and processing which

plays an important role in the ecological situation of

the climatically nearly closed narrow valley. In this

case it could be expected that the area is nearly

separated from pollution influences outside the valley

and that the pollution impact was dominated by a

strong single source. The two main geological units

of the area are schists, limestones, dolomites, gneisses

and amphibolites. The investigation area stretches over

5 by 12 km. Prevailing soil types in the Linz area are

lime-free cambisols and brown podzols, derived from

crystalline rocks, while soils in the Breitenau area

mostly lime-free cambisols, derived from schists,

weathered para-gneiss, or amphibolites. From the

Linz area, 5 out of 17 vertical soil profiles are exam-

ined. From the Breitenau region, previously studied by

Maier and Scholger (2004), 10 profiles out of 27 are

analysed in detail in this study.

All four investigated areas were examined using the

same strategy for topsoil magnetic susceptibility mapping

developed in frame of the MAGPROX-project (Schibler

et al., 2002; Boyko et al., 2004). Field measurements of

topsoil volume magnetic susceptibility were performed

with Bartington-MS2D probe. Each measured point repre-

sents a spot of about 4 m2, where about 15–30 measure-

ments were taken and averaged. In case of large data

scatter, more readings were performed. However, our

experience shows that increasing the number of readings

above 20 does not improve standard deviation signifi-

cantly. Geographic positions of all measured soil profiles

were determined by the GPS Total Station 4700.Maps of

topsoil magnetic susceptibility of investigated areas with

surface measurements and soil profile positions of all

regional studies are shown in Fig. 2. These contour plots

were created by Surfer 8.0 (Golden Software).

Vertical distribution of selected typical profiles was

measured by Bartington MS2F stratigrafic sensor in the

field. In addition, soil cores were collected for further

laboratory measurements using MS2C coil sensor. Then

samples from upper and lower part of the cores were

prepared for detailed laboratory investigation. These

samples were selected in order to represent magnetically

enhanced topsoil (top 5 cm) and the bottom-most part of

the core (depth of at least 25 cm). All collected soil

samples are from forest areas, covered mostly by needle

trees (mainly spruce or pines). In this study, soil cores

were measured in laboratory. At present, sensitive and

fast in-situ measurements are available using new de-

vice, described by Petrovsky et al. (2004).

2.2. Laboratory analyses

The low- and high-frequency magnetic susceptibility

(nlf and nhf) was measured by Bartington MS2B probe,

expressed as mass-normalised susceptibility mlf and

mhf, respectively, and the corresponding frequency-de-

pendent susceptibility was calculated as difference per-

centage jfd =(nlf�nhf)/nlf. This parameter enables

assessment of significance of ultrafine superparamag-

netic magnetite grains (Dearing et al., 1996). Coercive

force (Hc), coercivity of remanence (Hcr), saturation

magnetization (Ms) and saturation remanent magneti-

zation (Mrs) were measured using a Princeton Vibrating

Sample Magnetometer VSM MicroMag 2900. Maxi-

mum applied field was 0.5 T. Heavy metal contents was

analysed using Atomic Absorption Spectrometry

(AAS) after dissolution in 2 M HNO3. Correlation

between magnetic susceptibility and heavy metals (Fe,

Pb, Mn, Zn, Cd, Cu, Ni, Cr) was studied for each

investigated area.

Fig. 2. Contour plots of spatial distribution of topsoil volume magnetic susceptibility (10�5 SI) of the four studied areas. Small dots mark locations of the measured sites, big labeled dots mark

locations where vertical soil profiles were collected (shown in Fig. 4), dashed line delimits the city boundaries.

H.Fialova

etal./JournalofApplied

Geophysics

59(2006)273–283

276

Fig. 3. SEM image of a Fe-rich spherule (top, grain-size of abou

80 Am), found in topsoil layer in the Ostrava region, and cross-sections

of two spherules with elemental analysis (middle: grain-size abou

40 Am, FeO 91%, Al2O3 0.8%, SiO2 0.2%; bottom: about 50 AmFeO 83%, Al2O3 8.4%, SiO2 2,2%, MgO 0.3%).

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283 277

Scanning electron microscopy (SEM) with wavelength

dispersive spectroscopy (WDS) was performed on mag-

netic concentrates obtained from raw soil samples using

hand-magnet separation in ultrasonic bath.

3. Results and discussion

3.1. Magnetic mapping and vertical profiles

Spatial distribution of surface magnetic suscepti-

bility of the four areas in concern (mesh of about 500

m) is outlined as in Fig. 2. Measured data were collected

only from forest soils, mostly in pine woods. Hand-

magnet separation from topsoil samples (top 5 cm)

revealed clearly Fe-rich spherules, typical for particles

of anthropogenic origin, derived from combustion of

fossil fuel (Flanders, 1994, 1999; Maier and Scholger,

2004; Fialova, 2004). Representative spherules, ob-

served in topsoil from the Ostrava region, are shown

in Fig. 3. Contrary to topsoils, bottom soils were lacking

these spherules. In the Breitenau area, lithogenic crystals

and anthropogenic spherules were found frequently in

both top- and subsoils (Maier and Scholger, 2004).

Typical vertical profiles of magnetic susceptibility

are shown in Fig. 4. In the Prıbram area, dominant

lithogenic contribution is assumed to control the in-

crease of magnetic susceptibility with depth. Contrary

to that, Ostrava and Linz profiles show dominant an-

thropogenic contribution, reflected by significant en-

hancement of magnetic susceptibility in the top 10 cm,

followed by rapid decrease of susceptibility with depth.

The Breitenau area represents a mixture of both types.

Anthropogenic contribution is dominant in profiles col-

lected in the narrow valley of this region, while several

profiles from the upper parts of the valley show signif-

icant lithogenic contribution (Maier and Scholger,

2004). Large set of vertical soil profiles of magnetic

susceptibility was recently collected, discussed and clas-

sified by Magiera et al. (in press).

3.2. Laboratory magnetic measurements

In order to analyse magnetic properties, mass-spe-

cific susceptibility mlf, frequency-dependent suscepti-bility jfd, saturation remanence Mrs, saturation

magnetization Ms, coercivity of remanence Hcr and

coercive force Hc were measured in laboratory on soil

samples prepared from the collected soil cores, and the

data were analysed using box-whisker-plots (Fig. 5).

The box-whisker plots summarize the distribution of a

variable by three components. In this study, we used the

mean/SE/SD mode, where SE stands for standard error

t

t

,

(standard deviation of the mean) and SD denotes stan-

dard deviation of the dataset (Tukey, 1977). Thus, dot

in the plot represents the mean (central tendency), large

box represents the meanFSE and whiskers represent

the meanF standard deviation. Note that different scal-

ing for the value axes is intentionally used in Fig. 5. In

this way, we can compare trends in data distribution

from topsoils and subsoils for each respective area

studied. Absolute values could be compared as well,

but in this case they are less important then the general

tendency/behaviour and relative comparison of the

measured data.

Fig. 4. Magnetic susceptibility of vertical soil profiles collected from the investigated areas. Dominant lithogenic contribution in Prıbram and some

Breitenau profiles is reflected by significant enhancement of magnetic susceptibility with depth. Ostrava, Linz and some Breitenau profiles show

dominant anthropogenic influence in the top 10 cm.

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283278

Mass-specific magnetic susceptibility shows the same

tendency for the Ostrava, Linz and Breitenau areas. Large

box-whisker-plots are typical for topsoil samples and

very narrow ones for subsoils. In the Prıbram area, with

the assumed significant lithogenic contribution, dominat-

ing the soil profile, no significant difference between the

top-and subsoil was observed, both showing box-whis-

kers corresponding to large data scatter. There seems to

Fig. 5. Box-whisker plots of measured magnetic parameters of samples prepared from the top and subsoils from the investigated areas. Dot—Mean,

Box—MeanFSE, h–— MeanFSD, dots outside the box-whiskers represent outliers and are not included in the evaluation. Note that different

scales are used for the value axes in order to compare relative trends rather than absolute values. Left-hand sided and right-hand sided box-whiskers

on each plot represent the topsoil and subsoil samples, respectively. Numbers inside each plot are ratios of the topsoil and subsoil mean values of the

corresponding parameters.

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283 279

be a contradiction between relatively narrow box-whisker

plot for subsoils in the Breitenau area (Fig. 5) as com-

pared to the vertical profiles of magnetic susceptibility,

shown in Fig. 4. This is caused by bincompatibilityQ of

volume magnetic susceptibility, measured by Bartington

MS2C, and mass-specific susceptibility. In the former

case, susceptibility values are much more affected by

variations in density, especially considering the fact, that

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283280

in the case of soil cores with the diameter of 3.5 cm,

small effective volume was measured.

Frequency-dependent magnetic susceptibility is sup-

posed to reflect the significance of ultrafine SP parti-

cles. Large grains of magnetite (e.g. produced by

combustion processes) are practically insensitive to

change in the frequency of the applied magnetic field

used. Therefore, such samples exhibit small jfd, usu-ally less than 2%. Our data show, that only in the

Ostrava and Linz topsoil, very narrow box-plots are

observed, suggesting single type of grain-size distribu-

tion rather than SP/SD/MD mixture. Low absolute

(mean) value suggests that presence of SP particles,

resulting mostly from pedogenic processes, can be

practically excluded. One single extreme value, ob-

served in Ostrava bottom-soil sample, can be attributed

to measurement error due to very low susceptibility

value, on the sensitivity limit of the Bartington probe.

In such case, apparently high frequency-dependent sus-

ceptibility may be an artefact, resulting from rounding

off by the instrument. This data point was not included

in our evaluation. This instrument effect is most proba-

bly responsible also for high mean value for the Ostrava

subsoil samples, although in these samples one can

expect also presence of higher portion of smaller parti-

cles, which are able to migrate downwards from the

topsoil. However, since susceptibility values are ex-

tremely low, we assume that the instrument error is

more significant. Relatively high jfd for Breitenau top-

soils (almost 5%), calculated with sufficient reliability

(j values of around 50) may suggest the presence of

relatively more SP magnetite of pedogenic origin, com-

pared to other localities. However, threshold values for

jfd, related to significance of SP magnetite, are not

that clear.

Saturation remanence values of the subsoil samples

from both the Ostrava and Linz areas are very low and

show practically no scatter. Saturation magnetization

shows more or less the same tendency like saturation

of remanence, with a small aberration in the Prıbram

samples and much larger irregularity in the case of

Breitenau samples.

Table 1

Coefficient of determination r2 of linear fit between mass-specific magne

Number of samples Fe Pb

PRIBRAM 6 0.00a 0.02a

OSTRAVA 8 0.01a 0.72a

LINZ 12 0.59 0.29

BREITENAU 16 0.69 0.78

b.d.l.—below detection limit.a 16 samples.

Coercivity of remanence and coercive force do not

show any significant differences between the top and

subsoil samples and even any significant differences

between individual areas. Although the reliability of the

two parameters can be discussed due to low maximum

magnetic field applied, our data suggest that, in terms of

discrimination between the anthropogenic and litho-

genic contributions in soil samples, these parameters

alone can be considered as of negligible importance.

Moreover, ratios of magnetic parameters, which are

used as granulometric indicators for the construction

of the Day plot (Day et al., 1977), seem to be quite

similar for top and subsoils from the four regions, and

are, therefore, less significant for the discrimination of

the lithogenic and anthropogenic contributions. How-

ever, as shown by Fialova (2004), samples dominated

by anthropogenic contribution only (topsoils from the

Ostrava region), are clustered more densely within the

pseudo-single domain area of the Day plot, while the

samples with significant lithogenic contribution span

over much larger interval.

3.3. Correlation with heavy metals

Relationship between magnetic susceptibility and

concentration of heavy metals was evaluated using

the coefficient of determination r2 (Table 1) of linear

fit of bi-plots (examples of such bi-plots are shown in

Fig. 6). Due to availability of AAS geochemical analy-

sis, only samples from representative vertical profiles

were selected and analysed. It seems that significant

correlation is observed between susceptibility and con-

centration of Pb in polluted areas (Ostrava, Breitenau

and Linz topsoils). However, it seems that this relation-

ship strongly depends on the leaching method used.

This effect will be subject of our future study. In case of

Fe, apparently contradictory correlation was found.

While it seems to be missing in the Prıbram and

Ostrava regions, reasonably high values of the r2 coef-

ficient were found for the Linz and Breitenau regions.

This is striking especially for the Ostrava region, where

no lithogenic contribution is assumed, and Fe is sup-

tic susceptibility (m3 kg�1) and heavy metal content (ppm) in soils

Mn Zn Cd Cu Ni Cr

0.14 0.02 0.14 0.02 0.21 0.04

0.22 0.69 0.40 0.64 0.41 0.72

0.10 0.55 b.d.l. 0.29 0.23 0.33

0.02 0.01 0.67 0.14 0.05 0.17

Fig. 6. Examples of bi-plot correlation between selected heavy metals and magnetic susceptibility. Data of Mn concentration are divided by a factor

of 10.

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283 281

posed to be solely in the form of industrially produced

Fe-oxides. Obviously, the leaching agent used (2 M

HNO3) is not effective, especially in the case of large

particles with Fe-oxides embedded in Al–Si matrix.

This effect is subject of our further study and will be

published later on elsewhere. Anthropogenic spherules

could be dissolved by HCl or by total dissolution

(HNO3+HF+H2O2), but with the latter one paramag-

netic minerals can be also dissolved and the data will be

biased.

Our results suggest that, in the areas studied, Pb, Cu

and Cr are elements with significant correlation with

magnetic minerals in anthropogenically affected (pol-

luted) areas, while in the case of Mn and Ni no signifi-

cant correlation was found. Zn is not of typically

lithogenic origin. It can be found in topsoils affected

by specific industrial sources, and is characterised by

high mobility (Karczewska, 1996). In our samples, Zn

correlates well with magnetic susceptibility in the in-

dustrial regions around Ostrava and Linz.

Contrary to that, Ni and Fe can be considered, in the

investigated areas, as elements of typically lithogenic

origin. However, as mentioned above, the anthropo-

genic/lithogenic character of Fe is quite dubious due

to variable efficacy of leaching methods.

Since it is known from linear regression theory that

the ideal number of observations is 20 for one inde-

pendent variable, and the minimum is 5, and that the

number of observations determines the threshold for

the r2 coefficient in terms of significance, we are well

aware of the fact that our data in Table 1 have

different meaning in terms of significance. However,

we do not aim at detailed analysis of relationship

between magnetic properties and geochemistry of the

soils studied. These data are presented only in order to

illustrate basic geochemical meaning of the magnetic

parameters analysed.

4. Conclusions

Top- and bottom-soil samples from four areas, char-

acterised by different geological and environmental

settings, were analysed. The four areas are charac-

terised, in terms of significance of anthropogenic vs.

lithogenic contributions to magnetic-mineral population

in soils, as anthropogenically dominated (Ostrava and

Linz), mixed with dominant lithogenic contribution

(Prıbram) and mixed with sites showing either anthro-

pogenic, or lithogenic prevalence (Breitenau). Although

climatic effects are not a subject of this study, the areas

investigated do not differ practically in terms of climatic

conditions. Regarding possible pedogenic contribution

to population of magnetic minerals in the soils con-

cerned, frequency-dependent magnetic susceptibility

indicates practically insignificant portion of ultra-fine

superparamagnetic magnetite. Our data suggest that if

the lithogenic contribution is dominant, this effect is of

primary significance also for topsoil magnetic suscep-

tibility measurements, and the anthropogenic contribu-

tion cannot be that easily assessed. Contrary to that, in

soils from areas with negligible lithogenic contribution

of strongly magnetic minerals, mass specific magnetic

susceptibility alone is reliable in discriminating mag-

netically enhanced topsoils from the unaffected bottom

soils. Data of magnetic susceptibility are in accordance

with concentration-dependent saturation magnetization,

and thermomagnetic analysis of magnetic phases is not

necessary. In areas, where both anthropogenic and

lithogenic contributions are significant, magnetic sus-

ceptibility and saturation magnetization cannot serve

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283282

for reliable discrimination of the two soil layers, and a

combination of more magnetic parameters have to be

used. In this case, comparison with concentration of

heavy metals of presumably anthropogenic origin (e.g.

Pb) can validate magnetic data and enable discrimina-

tion between anthropogenically affected topsoils and

lithologically controlled bottom soils.

Based on our results, and provided that the signifi-

cance of SP magnetite due to pedogenic effect is small

or negligible, the following guidelines for magnetic

discrimination can be proposed:

! Mass-specific magnetic susceptibility and/or satura-

tion magnetization of topsoil is much higher than

that of subsoil, and the latter values show practically

no scatter: soil profile is dominated by anthropo-

genic effect in topsoil layer.

! Frequency-dependent magnetic susceptibility of top-

soils is lower and less scattered than that of bottom

soils: soil profile is dominated by anthropogenic

effect in topsoil layer.

! Mass-specific magnetic susceptibility and/or satura-

tion magnetization of topsoils is higher than that of

subsoils, but the latter shows significant data scat-

ter: soil is affected by a combination of both an-

thropogenic and lithogenic effects with variable

dominance.

! Mass-specific magnetic susceptibility and/or satura-

tion magnetization of topsoils is comparable to that

of subsoils, or lower, and both top and subsoil data

show significant data scatter: soil is dominated by

strong lithogenic effect.

Especially in the latest case, detailed analysis of

vertical profiles of magnetic susceptibility, accompa-

nied by correlation with selected heavy metals (of

presumably anthropogenic origin) has to be carried

out in order to identify soil layer affected by anthropo-

genic contribution.

Acknowledgments

This study was carried out within the framework of

MAGPROX project (5FP EU Project EVK2-CT-1999-

00019), and the Acad. Sci. Czech Rep. Project No.

S3012354. Financial support from the Styrian Govern-

ment is gratefully acknowledged. Our thanks are due to

the CEREGE (University Aix-Marseille III, France)

laboratory staff (Prof. P. Rochette and Ms. F. Vade-

boine) and Ms. Z. Korbelova from the Geological

Inst., Acad Sci. Czech Republic, for their assistance

with the SEM observations, Dr. T. Grygar from the Inst.

Anorg. Chemistry, Acad. Sci. Czech Rep., for fruitful

discussion and help with chemical analysis, and to Ms.

T. Doleckova from the Geophysical Inst., Acad, Sci.

Czech Republic, for great help with laboratory mea-

surements. The authors also thank Dr. W. Krainer from

the Styrian Agricultural Laboratory for his help and

advice.

References

Boyko, T., Scholger, R., Stanjek, H., 2004. Topsoil magnetic suscep-

tibility mapping as a tool for pollution monitoring: repeatability of

in situ measurements. J. Appl. Geophys. 55, 249–259.

Day, R., Fuller, M.D., Schmidt, V.A., 1977. Hysteresi properties of

titanomagnetites: grain size and composition dependence. Phys.

Earth Planet. Inter. 13, 260–267.

Dearing, J.A., Hay, K.L., Baban, S.M.J., Huddleston, A.S., Welling-

ton, E.M.H., Loveland, P.J., 1996. Magnetic susceptibility of soil:

an evaluation of conflicting theories using a national data set.

Geophys. J. Int. 127, 728–734.

Dearing, J.A., Hannam, J.A., Anderson, A.S., Wellington, E.M.H.,

2001. Magnetic, geochemical and DNA properties of highly-

magnetic soils in England. Geophys. J. Int. 144, 183–196.

Fialova, H. 2004. Magnetic discrimination of lithogenic and an-

thropogenic minerals in soils. PhD Thesis, Czech Technical

University, Faculty of Civil Engineering, Prague, Czech Repub-

lic (in Czech).

Flanders, P.J., 1994. Collection, measurement and analysis of airborne

magnetic particulates from pollution in the environment. J. Appl.

Phys. 75, 5931–5936.

Flanders, P.J., 1999. Identifying fly ash at a distance from fossil fuel

power stations. Environ. Sci. Technol. 33, 528–532.

Hanesch, M., Scholger, R., 2002. Mapping of heavy metal loadings in

soils by means of magnetic susceptibility measurements. Environ.

Geol. 42, 857–870.

Hanesch, M., Scholger, R., 2005. The influence of soil type on the

magnetic susceptibility measured throughout soil profiles. Geo-

phys. J. Int. 161, 50–56.

Hanesch, M., Scholger, R., Dekkers, M., 2001. The application of

fuzzy c-means cluster analysis and non-linear mapping to a soil

data set for the detection of polluted sites. Phys. Chem. Earth 26,

885–891.

Hanesch, M., Maier, G., Scholger, R., 2003. Mapping heavy metal

distribution by measuring the magnetic susceptibility of soils.

J. Geophys. IV 107, 605–608 (Part 1.).

Heller, F., Strzyszcz, Z., Magiera, T., 1998. Magnetic record of

industrial pollution in forest soils of Upper Silesia, Poland.

J. Geophys. Res. 103/B8, 767–774.

Hoffmann, V., Knab, M., Appel, E., 1999a. Magnetic susceptibility

mapping of roadside pollution. J. Geochem. Explor. 66, 313–326.

Hoffmann, G., Knab, M., Appel, E., 1999b. Magnetic susceptibility

mapping of road side pollution. L. Geochem. Expl. 66, 313–326.

Jordanova, N.V., Jordanova, D.V., Veneva, L., Yorova, K., Pet-

rovsky, E., 2003. Magnetic response of soils and vegetation to

heavy metal pollution—A case study. Environ. Sci. Technol. 37,

4417–4424.

Kapicka, A., Petrovsky, E., Ustjak, U., Machackova, K., 1999. Proxy

mapping of fly ash pollution of soils around a coal-burning power

plant, a case study in the Czech Republic. J. Geochem. Int. 66,

291–297.

H. Fialova et al. / Journal of Applied Geophysics 59 (2006) 273–283 283

Kapicka, A., Jordanova, N., Petrovsky, E., Ustjak, S., 2000. Magnetic

stability of power-plant fly ash in different soil solutions. Phys.

Chem. Earth 25, 431–436.

Kapicka, A., Petrovsky, E., Jordanova, N., Podrazsky, V., 2001.

Magnetic parameters of forest top soils in Krkonose Mountains,

Czech Republic. Phys. Chem. Earth 26, 917–922.

Kapicka, A., Jordanova, N., Petrovsky, E., Podrazsky, V., 2001. Effect

of different soil conditions on magnetic parameters of power-plant

fly ashes. J. Appl. Geophys. 48, 93–102.

Kapicka, A., Jordanova, N., Petrovsky, E., Podrazsky, V., 2003.

Magnetic study of weakly contaminated forest soils. Water Air

Soil Pollut. 148, 31–44.

Karczewska, A., 1996. Metal species distribution in top-and sub-soil

in an area affected by copper smelter emissions. Appl. Geochem.

11, 35–42.

Lecoanet, H., Leveque, F., Segura, S., 1999. Magnetic susceptibility

in environmental applications: comparison of field probes. Phys.

Earth Planet. Inter. 115, 191–204.

Lecoanet, H., Leveque, F., Ambrosi, J.P., 2001. Magnetic properties

of salt-marsh soils contaminated by iron industry emissions

(southeast France). J. Appl. Geophys. 48, 67–81.

Lecoanet, H., Leveque, F., Ambrosi, J.P., 2003. Combination

of magnetic parameters: an efficient way to discriminate soil-

contamination sources (south France). Environ. Pollut. 122,

229–234.

Magiera, T., Strzyszcz, Z., 2000. Ferrimagnetic minerals of anthro-

pogenic origin in soils of some polish national parks. Water Air

Soil Pollut. 124, 37–48.

Magiera, T., Strzyszcz, Z., Kapicka, A., Petrovsky, E., in press.

Discrimination of lithogenic and anthropogenic influences on

topsoil magnetic susceptibility in Central Europe. Geoderma.

Maier, G., Scholger, R., 2004. Demonstration of connection between

pollutant dispersal and atmospheric boundary layers by use of

magnetic susceptibility mapping, St. Jacob (Austria). Phys. Chem.

Earth 29, 997–1009.

Nemecek J., 2001. Digital map of soils of Czech Republic 1 :200000.

Czech University of Agriculture, Prague, Czech Republic.

Petrovsky, E., Ellwood, B.B., 1999. Magnetic monitoring of pollution

of air, land and waters. In: Maher, B.A., Thompson, R. (Eds.),

Quaternary Climates, Environments and Magnetism. Cambridge

Univ. Press, UK, pp. 279–322.

Petrovsky, E., Kapicka, A., Jordanova, N., Knab, M., Hoffmann, V.,

2000. Low-field magnetic susceptibility: a proxy method of esti-

mating increased pollution of different environmental systems.

Environ. Geol. 39, 312–318.

Petrovsky, E., Hllka, Z., MAGPROX Team, 2004. A new tool for in-

situ measurements of the vertical distribution of magnetic suscep-

tibility in soils as basis for mapping deposited dust. Environ.

Technol. 25, 1021–1029.

Schibler, L., Boyko, T., Ferdyn, M., Gajda, B., Holl, S., Jordanova,

N., Magprox Team, 2002. Topsoil magnetic susceptibility map-

ping: data reproducibility and compatibility, measurement strate-

gy. Stud. Geophys. Geod. 46, 43–57.

Tukey, J., 1977. Exploratory Data Analysis. Addison-Wesley, Boston,

USA.

Veneva, L., Hoffmann, V., Jordanova, D., Jordanova, N., Fehr, T.,

2004. Rock magnetic, mineralogical and microstructural charac-

terization of fly ashes from Bulgarian power plants and the nearby

anthropogenic soils. Phys. Chem. Earth 29, 1011–1023.