1 Introduction

The Karkonosze, a mountain range in the Sudetes (central Europe), was subject to widespread deterioration of forest complexes during the last decades of the twentieth century. The stands of spruce (Picea abies (L) H. Karst) and, to a lesser extent, of mountain dwarf pine (Pinus mugo Turra) were mainly affected. Forests in the Sudety Mountains were referred to as in a state of total disaster and reported as strongly damaged (Mazurski 1986; Grodzińska and Szarek-Łukaszewska 1997; Fabiszewski and Wojtuń 2000; Godek et al. 2008). Such effects were caused by complex interactions of multiple factors, including (1) adverse climatic, geological, and soil conditions; (2) the intensity of anthropogenic impact; (3) the level of air pollution from industrial emissions; and (4) the occurrence of insect pests and parasitic fungi (Sobik and Błaś 2008). Air pollution caused by local emissions in the Sudetes, together with a long-distance atmospheric transport of dust and gases, in particular SO2, from various industrial- and transport-related sources, was considered the most degrading factors (Bochenek et al. 1997; Dore et al. 1999; Błaś and Sobik 2000; Donisa et al. 2000; Stachurski and Zimka 2002; Bytnerowicz et al. 2004, 2008; Małek and Astel 2008). Direct impacts of SO2 deposition on tree needles contributed to their deterioration, while soil acidification, together with other stressors, affected the growth of roots (Ulrich 1986; Markert et al. 1996; Lorenz et al. 2008). Several studies reported increased concentrations of some heavy metals in soils of the Karkonosze (Borkowski et al. 1993; Drozd et al. 1996; Szopka et al. 2013) and supposed that heavy metals should be considered as highly disadvantageous components of pollution that affect biota, including plants, and destroy natural ecosystems (Salemaa et al. 2001; Guala et al. 2010; Chai et al. 2014). Environmental pollution with heavy metals was therefore suggested to be an important factor of forest degradation in the Karkonosze. It should be stressed that although a large-scale dieback of coniferous forests has already been overcome, this problem continues to arise in various locations, indicating that several other factors, other than acid rains and SO2 emissions, remain of considerable importance. Therefore, the effects of potentially toxic concentrations of metals, and particularly their mobile forms, on mountain habitats need close examination.

The total concentrations of trace metals in soils are primarily determined by soil parent rocks and anthropogenic inputs (Fernandez et al. 2000; Kabata-Pendias 2011; Alloway 2013). The fate of airborne metals in soils and their distribution in the soil profile are conditioned by soil properties and usually reflect the course of soil formation (Abollino et al. 2002a; Kabata-Pendias 2011; Alloway 2013; Wang and Xu 2015). The latter are usually particularly well recorded in Podzols. Consequently, the distribution of trace metals in soil profiles, together with data on their speciation in the soil solid phase, provides valuable information on the origin of metals, as well as on the factors and mechanisms responsible for their increased mobility and related bioavailability. Speciation of trace elements in soils was defined by Tack and Verloo (1995) as the identification and quantification of the different, defined species, forms, or phases in which an element occurs.

Although several advanced methods have been recently developed to investigate the behavior of metals in soils, including micromorphological and isotopic studies (D’Amore et al. 2005; Bińczycki et al. 2014), chemical extraction methods are still commonly used. Numerous procedures of chemical extraction have been tested to provide information on the actual and potential solubility of metals in soils in various environmental conditions. Consequently, hundreds of operationally defined sequential extraction methods have been described that recognize the main soil components that bind metals in soils (Tessier et al. 1979; Zeien and Bruemmer 1989; Tack and Verloo 1995; Gleyzes et al. 2002; Pueyo et al. 2003; Rao et al. 2008; Hass and Fine 2010). To quantify the contributions of particular fractions, chemical solutions of varying strength and reactivity are applied to soils in order to release metals bound to different soil components. A commonly used, relatively simple BCR method, developed by the Community Bureau of Reference (Ure et al. 1993; Mossop and Davidson 2003) determines four fractions of metals (acid soluble, reducible, oxidizable, and residual). A method by Zeien and Bruemmer (1989, 1991), which is more laborious, provides greater insight into the mechanisms of metal binding in soils and, therefore, has been widely applied to both non-polluted and variously polluted soils (Karczewska 1996; Wenzel and Jockwer 1999; Kabała and Szerszeń 2002; Marschner et al. 2006; Rao et al. 2008; Kabala et al. 2011). The Zeien and Brümmer method involves the determination of seven operationally defined fractions, believed to correspond with the following species of metals: (1) mobile, (2) exchangeable and specifically bound, (3) bound to manganese oxides, (4) organically bound, (5) occluded in amorphous FeO x , (6) occluded in crystalline FeO x , and (7) residual. Mobile and exchangeable fractions of metals are considered as easily bioavailable and leachable, whereas metals associated with organic matter, as well as those occluded in MnO x and FeO x , might be released from soils under changing conditions and, thus, indicate medium- to long-term availability. The residual fraction is considered entirely non-available and immobile, as it is confined to silicates that remain stable over time.

Here, we aimed to assess the contamination of Podzols in a subalpine zone of the Karkonosze Mountains. The Cu, Pb, and Zn concentrations in these soils and their speciation are discussed in relation to the possible influence of industrial emissions and the stage of degradation of dwarf mountain pine stands. Speciation of metals in soil profiles was examined by sequential extraction according to Zeien and Bruemmer (1991) to indicate the most likely sources of the metals, to assess the environmental risk associated with their presence, and to predict potential changes in their mobility under changing environmental conditions.

2 Materials and methods

The study was conducted in the Karkonosze Mountains, a range of the Sudetes, located on the boundary between Poland and Czech Republic (Fig. 1). The profiles of soils derived from granites in the zone occupied by dwarf mountain pine (P. mugo Turra) were examined. The parent rock is typical biotite granite (Weber et al. 2012) and contains the following amounts of analyzed heavy metals: Cu 11–22 mg kg−1, Pb 0.1–0.3 mg kg−1, and Zn 32–65 mg kg−1 (Tyszka et al. 2012). The sites were located at an altitude of 1400 m above sea level (corresponding to upper mountain forest zone). The climatic conditions of the investigated area are characterized by a very high precipitation (1430 mm/year) and low mean annual temperature +1.5 °C (Głowicki 2005). Two sites were selected for investigation that differed in the deterioration stage of dwarf pine stands: area I had no signs of degradation (Fig. 2), whereas area II was affected by dwarf mountain pine dieback. A thicket of dead dwarf pine branches and trunks remained there, without any living shrubs, and the area was entirely colonized by herbaceous plants typical for mountain meadow, dominated by a Festuca ovina L. grass that covered the soil surface underneath dead dwarf pine branches (Fig. 3). Areas I and II were 100 m apart. Within each of those two areas, two soil profiles were analyzed: profiles I/1 and I/2 in the stand with dwarf pine in good condition and profiles II/1 and II/2 in the area with dead dwarf mountain pines.

Fig. 1
figure 1

Location of the investigated site on the map of Europe

Full size image
Fig. 2
figure 2

Area I. Mountain dwarf pine (Pinus mugo Turra) stand without signs of degradation

Full size image
Fig. 3
figure 3

Area II. Strongly degraded mountain dwarf pine (Pinus mugo Turra) stand

Full size image

Soil samples were collected from all of the horizons distinguished in soil profiles. The samples were air-dried, ground, and passed through a stainless steel 2-mm sieve. The basic soil properties were determined. Particle size distribution was analyzed in mineral samples by the sieve and hydrometric method (Pansu and Gautheyrou 2006), following the pretreatment that involved removal of organic matter and chemical dispersion with sodium hexametaphosphate. The content of organic carbon (Corg) was determined using a CS-MAT 5500 analyzer (Ströhlein GmbH & Co., Kaarst, Germany, currently Bruker AXS Inc., Madison, WI, USA). Soil pH was measured potentiometrically in a 1:2.5 suspension of soil and 1 M KCl; soil acidity (Hh) was determined in 1 mol dm−3 KCl (Kappen 1929), and exchangeable base (EB) cations were extracted by 1 mol dm−3 NH4Ac and measured by atomic emission spectroscopy (AES) (K+, Na+, and Ca2+) and atomic absorption spectroscopy (AAS) (Mg2+). Effective cation exchange capacity (CEC) and base saturation (BS) were calculated on the basis of the acidity (Hh) and exchangeable base cations. For analysis of total Cu, Pb, and Zn concentrations in the mineral soil horizons, the samples were wet digested in concentrated perchloric acid, in an open system with reflux, while the samples of organic horizons were ashed in an oven prior to acid dissolution in nitric acid. The concentrations of Cu and Zn in the digests were determined by AAS and the concentrations of Pb by optical emission spectroscopy (OES-ICP). Two certified reference materials, NIST RSM 2711 (Montana Soil) and RSM 2709 (San Joaquin Soil), were used for analytical validation.

The sequential extraction procedure according to the method by Zeien and Bruemmer (1989) was applied to determine seven operationally defined fractions of metals: mobile (F1), exchangeable and specifically sorbed (F2), occluded in Mn oxides (F3), organically bound (F4), occluded in amorphous Fe oxides (F5), occluded in crystalline Fe oxides (F6), and residual (F7). A brief summary of the procedure is given in Table 1. Concentrations of Cu and Zn in soil extracts were determined by AAS and concentrations of Pb by OES-ICP. In the samples of ectohumus, built mainly of organic matter, the procedure was simplified so that only the F1 and F2 fractions were isolated to determine the most available species of metals, without further steps of fractionation. All extractions were carried out in duplicates, of which the mean values are reported in tables and diagrams.

Table 1 Procedure of sequential extraction according to Zeien and Bruemmer (1989)
Full size table

3 Results and discussion

All soils examined in this study represented shallow mountain soils derived from granite, with properties typical for Haplic Podzols (Fig. 4) (WRB IUSS Working Group 2014). They contained considerable amounts of gravel (over 50% in the mineral part of the profile), which tended to increase as progressing downward through the soil profile; up to 90% is in a parent rock (C horizon). Soil earthy fractions indicated a sandy loam texture and differed slightly among horizons. All of the soil properties were found to be strongly influenced by a podzolization process (Table 2). The soils indicated strongly acidic reaction through all horizons, and the lowest pH values (pH 2.0–2.7) were found in the Oe horizons, which is typical for humus under conifers, due to the transformation of organic compounds. Strong soil acidification should be considered as an important factor of increased bioavailability and toxicity of heavy metals (Huchinson and Whilby 1977; Han et al. 2002; Kabata-Pendias 2011; Alloway 2013). BS of soil material was low (Table 2), which is typical for Podzols, particularly in mountain soils (Kabala et al. 2011). However, the effective cation sorption capacity (CEC) in the organic horizons was considerably high, much higher than in mineral horizons (Table 2), thus indicating a potential for the accumulation of heavy metals.

Fig. 4
figure 4

Profile of Podzol derived under mountain dwarf pine stand (object I/2)

Full size image
Table 2 Basic properties of the tested soils
Full size table

The total concentrations of Cu in the soils varied between 5.54 and 32.0 mg kg−1 and did not differ from natural concentrations of Cu in European forest soils developed from granitic rocks (Kabata-Pendias and Szteke 2015). The Cu distribution indicated small variations through the soil profiles (Table 3). The highest concentrations of Cu were found in organic horizons, as well as in the illuvial Bh horizon, especially in area II. By sequential fractionation of Cu in soils (Fig. 5), we found that the mobile (F1) and exchangeable (F2) fractions constituted roughly about 10% of the total concentrations of that element, while the residual (F7) fraction of Cu dominated in all mineral horizons. This fact might be considered as a proof of a predominantly natural origin of the Cu in these soils (Karczewska 1996). However, a profile distribution of Cu in soils indicated significant enrichment of surface horizons, which can only partly be attributed to natural bioaccumulation (Kabata-Pendias 2011; Alloway 2013). The high share of residual Cu fraction (F7), which is similar between areas I and II, indicates that neither the total nor soluble Cu concentrations affected the status and condition of dwarf pine. Relatively high concentrations (27–30 mg kg−1) of Cu in the Bh horizons of the area II soils should be considered as evidence of past conditions that most likely facilitated Cu mobilization and leaching, such as particularly low pH (Tyler 1978) or accelerated decomposition of organic matter associated with a release of chelating compounds (Alloway 2013; Karczewska et al. 2013).

Table 3 Cu fractions sequentially extracted according to the method by Zeien and Brümmer
Full size table
Fig. 5
figure 5

Percentage of Cu extracted from soils in fractions 1–7 according to the Zeien and Bruemmer (1989) method. The soils were collected either under a dwarf mountain pine without signs of degradation (I/1, I/2) or under dead dwarf pine (II/1, II/2)

Full size image

The total concentrations of Pb in the organic horizons reached a maximum of 315 mg kg−1 and were clearly higher than in similar horizons of European lowland forest or meadow soils (Kabata-Pendias 2011; Kabata-Pendias and Szteke 2015), in which they rarely exceed 50 mg kg−1. Particularly high Pb concentrations, over 265 mg kg−1, were found in Oe horizons of all soil profiles (Table 4). A similar enrichment of organic top horizons in mountain soils, including the soils of Karkonosze Mountains, has been reported by various authors (Kabała and Szerszeń 2002; Szopka et al. 2013). The Pb concentrations in the ectohumus (Oi, Oe, and Oa) were higher than those in lower, mineral parts of the soil profiles but with a secondary accumulation of Pb in the Bh horizons (Table 4). This was apparently caused by leaching of organically bound Pb complexes from the topsoil. The very high concentrations of Pb in organic horizons of all soils examined are undoubtedly from atmospheric pollution and absorption of airborne particles rich in Pb, which has already been widely reported and proven (Donisa et al. 2000; Kaste et al. 2006; Steinnes and Friedland 2006; Shotyk 2008; Nygard et al. 2012; Tyszka et al. 2012). The Karkonosze mountain range was under the influence of long-distance pollution, mainly from lignite-based power plants and metal smelters (Grodzińska and Szarek-Łukaszewska 1997; Dore et al. 1999; Godek et al. 2008; Tyszka et al. 2012). The distribution of Pb in the soil was clearly affected by the podzolization process, so that the eluvial E horizons were apparently impoverished in Pb (14.8–38.1 mg kg−1), whereas the upper illuvial horizons Bh were considerably enriched (67.2–94.2 mg kg−1). The sequential fractionation of Pb showed that a predominating part of Pb accumulated in the soils was in the F4 fraction, defined as bound with organic matter (Fig. 6). The organically bound fraction (F4) made up 13.2–44.7% of the total Pb in the soil, and its contributions were particularly high in illuvial Bh horizons. This effect demonstrates a high affinity of Pb to organic matter and is in line with the commonly accepted theory that it is chelation by low molecular weight compounds rather than acidification that determines the high solubility of anthropogenic Pb in acid forest soils (Tyler 1978; Steinnes and Friedland 2006). High shares of the mobile (F1) and potentially soluble (F2) fractions of Pb in ectohumus, which reach up to 50% in the Oi horizon of the I/1 profile, confirm an anthropogenic origin of the Pb in these soils (Abollino et al. 2002b; Lu et al. 2003; Neel et al. 2007; Agbenin et al. 2010; Szolnoki and Farsang 2013).

Table 4 Pb fractions sequentially extracted according to the method by Zeien and Brümmer
Full size table
Fig. 6
figure 6

Percentage of Pb extracted from soils in fractions 1–7 according to the Zeien and Bruemmer (1989) method. The soils were collected either under a dwarf mountain pine without signs of degradation (I/1, I/2) or under dead dwarf pine (II/1, II/2)

Full size image

Comparison of the results obtained from the areas I and II, with different degradation of dwarf pine stands, indicated that there were no apparent differences in total concentrations of Pb or in their mobile (F1) and easily mobilizable (F2) fractions between these two areas. This result indicates that a forest dieback should not be attributed to enhanced concentrations of Pb in the soil. Then again, possible adverse effects of soluble Pb on plants or other biota in ecosystems cannot be definitely excluded, particularly in relation to the past, because several ecotoxicological studies set the threshold of Pb ecotoxicity at the level of 70–150 mg kg−1 (Szopka et al. 2013). Unfortunately, the properties of the soils in the sites subject to dwarf pine dieback have apparently changed compared to past decades, so that present soil properties do not provide a basis for an explanation of past phenomena (Fabiszewski and Wojtuń 2000).

The concentrations of Zn in the soil profiles were in a broad range 24.8–202.4 mg kg−1, with the highest values in the deepest parts of soil profiles, namely in C and Bs horizons (Table 5). The distribution of Zn was apparently influenced by podzolization, and minimum Zn concentrations were found in the eluvial horizons E or in overlying organic horizon Oa. Such a distribution clearly indicates an enrichment of the top soil layers, most likely from atmospheric depositions (Steinnes and Friedland 2006), and confirms a high susceptibility of Zn to leaching (Tyler 1978; Kabata-Pendias 2011; Alloway 2013). There were no substantial differences between the profiles located in the areas with different stages of dwarf pine degradation. However, the total concentrations of Zn in the bottom horizons of soils, that were clearly higher in area II, may have resulted from more intensive leaching in the past, possibly associated with a higher solubility of Zn. Such a hypothesis cannot be now checked, as at the time of soil sampling for the present study, soil pH values in II/1 and II/2 profiles were considerably higher compared to area I. This effect must have been caused by a recent succession of grasses that replaced dead coniferous dwarf pine shrubs (Drozd et al. 1996). Therefore, it should be stressed that the present soil concentrations of Zn, a particularly easily mobilized element, cannot be used to illustrate Zn solubility and leaching in the past. The sequential extraction of Zn in the samples of organic horizons revealed that fractions F1 and F2 of this element made up a predominating part of total Zn, up to 66% (Fig. 7). No simple correlations were, however, found between the contributions of soluble or potentially soluble Zn and soil pH (or related concentrations of H+ ions). The total concentrations of Zn were low in the top soil horizons, and the residual fraction (F7) of Zn and Cu was a predominating one in all the mineral parts of soil profiles. Moreover, the absolute concentrations of Zn in the F7 fraction (expressed in mg kg−1) tended to increase progressively downward through the soil profile, being much higher in the C horizon than in the geochemical background in granites. Similar patterns were also reported by Kabała and Szerszeń (2002) from Haplic Cambisols. This effect requires closer investigation but might be explained by a mechanical transport of tiny Zn-bearing particles of airborne origin and simultaneous leaching of mobile Zn fractions out of the soil profiles.

Fig. 7
figure 7

Percentage of Zn extracted from soils in fractions 1–7 according to the Zeien and Bruemmer (1989) method. The soils were collected either under a dwarf mountain pine without signs of degradation (I/1, I/2) or under dead dwarf pine (III/1, II/2)

Full size image
Table 5 Zn fractions sequentially extracted according to the method by Zeien and Brümmer
Full size table

4 Conclusions

  1. 1.

    Very high Pb concentrations in the ectohumus of the Karkonosze soils tested here have most likely been caused by a long-distance transport of anthropogenic emissions.

  2. 2.

    The total concentrations of Cu, Pb, and Zn found in soils cannot be considered as essential factors of the ecological disaster that affected Karkonosze Mountains.

  3. 3.

    This study did not reveal any remarkable differences in the patterns of profile distribution or in the fractionation of Cu, Pb, and Zn between the areas differing in the deterioration stage of dwarf pine stands.

  4. 4.

    High concentrations of Pb in the ectohumus horizons, as well as very high contributions of its organically bound (F4) fraction in mineral horizons, confirm a crucial role of organic matter in the processes of Pb accumulation.

  5. 5.

    Profile distributions of all metals examined have been influenced by soil properties that developed during the podzolization processes. It was particularly well expressed in the case of Pb, a metal that strongly accumulated in Bh horizons, predominantly in organically bound (F4) fraction that made up about 40% of total Pb.

  6. 6.

    Low concentrations of Cu throughout the soil profiles and the predominance of residual (F7) fraction present in soil mineral horizons support a lithogenic origin of Cu in these soils.

  7. 7.

    High contributions of easily mobilizable (F1 and F2) fractions of Zn in ectohumus, as well as the prevalence of its residual (F7) form in deeper mineral horizons, indicate that a considerable part of the Zn pool is subject to leaching out of the soil profile.