Introduction

Soil–plant transfer is the first step by which radionuclides, stable elements and heavy metals enter the food chain. Compared to green plants, high accumulating ability of many essential and/or toxic trace elements, some radionuclides and heavy metals by mushrooms has been known and numerous works have been published since the 1970s [1–5 (and references therein)]. A number of authors have reported transfer factors of some radionuclides, especially for radiocesium, in different species of mushroom as presented in Table 1, while there still is little data available on the transfer of trace elements and heavy metals to mushrooms [15, 16]. Whereas reliable data on the transfer of these materials in mushrooms are needed for developing and testing models, predicting migrations of these elements and obtaining the associated parameter values appropriate for radioecological and environmental performance assessments [17].

Table 1 Comparison of transfer factors of 137Cs and 40K in mushroom samples of different areas in the world
Full size table

The transfer of radionuclides and other elements of interest between different ecological compartments of an ecosystem is frequently quantified by many ratio types such as transfer factors, concentration ratios, transfer coefficients and concentration factors [2, 12, 18, 19]. These parameters have been recommended by many international agencies IAEA, IUR and ICRP because of their conceptual simplicity and common use in radioecology [20–25]. These coefficients are expressed as the ratio between the contamination of the receptor compartment, and that of the donor. The transfer values can be influenced by some factors such as soil characteristics, climatic conditions, type of plants, part of the plant concerned, physico-chemical form of the radionuclides and the effect of the competitive species [26 and references therein].

In this study, the uptake of 137Cs, 40K, stable elements and heavy metals in 25 mushroom samples, covering 12 biological species such as Agaricus campestris, Lactarius semisanguifluus, Clitocybe bresadoliana, Tricholoma terreum, Lactarius sp., Sarcodon scabrosus, Lactarius deliciosus, Lepista nuda, Suillus bovinus, Tricholoma sp., Russula delica and Macrolepiota excoriate, was evaluated by means of the estimation of the transfer factor.

Materials and methods

Study area and sample collection

A radioecological study was carried out at seven forest sites in and around Izmir, in 2002 during September–December when mushrooms yield reached maximal values. Sampling locations and their coordinates are given in Table 2. At the investigated site, the soils were classified as Dystric Xerorthent (U.S. taxonomy) and characterized by the typical horizons of an undisturbed forest floor with high organic matter content. In this study, soil and mushroom samples were sampled simultaneously, taking each mushroom sample with its corresponding organic soil layers. The different fungi species was identified precisely at Muğla University, Ula Technical High School, Department of Mushroom. The taxonomical classification of the sampled mushrooms is presented in Table 3 according to Solak [27] and Solak et al. [28].

Table 2 Sampling locations, annual precipitation (P), altitudes (A), species of mushroom, sample type, transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K
Full size table
Table 3 Taxonomical classification of the sampled mushrooms in and around Izmir
Full size table

Sample treatments and gamma-spectrometric measurements

The soil samples were dried to a constant weight at 75 °C for 24 h in an electric oven, reweighed and passed through a 2 mm sieve for removing of stones and roots. The loss of weight on drying (d.w. loss) was calculated for each soil samples. Each dried sample (300–1,800 g) was placed in a 1,000 mL Marinelli beaker prior to analysis.

The fruitbodies were carefully cleaned with a plastic brush to eliminate extraneous soil particles and litter tissue, and then they were weighted (“wet” weight). Mushroom samples for each species were dried to a constant weight at 80 °C for 48 h in an electric oven up to the “airdry” weight and weighed again. The loss of weight on drying (d.w. loss) was calculated for each sample. Dried samples (40–90 g) were ground in a small mill and the ground material was transferred to plastic sample containers (45 mm in diameter) prior to analysis.

The measurements of 40K and 137Cs activity concentrations in the soil and mushroom samples were employed using a 184 cc p-type coaxial HPGe detector with a relative efficiency of 25% and a resolution of 1.85 keV at 1.332 MeV (with associated electronics procured from EG&G Ortec). The detector was sealed by 100 mm thick lead bricks internally lined with 1.5 mm copper foil. The spectrum was acquired and analyzed using a PC-based 8 K multichannel analyzer and the associated software. The detector efficiency calibration was performed using the IAEA quality assurance reference materials manufactured by IAEA. The standard materials and samples were put into the containers of the same size and type so that detection geometry should remain the same. The activity of 40K was evaluated from the 1,460.75 keV photopeak while 137Cs was determined by the 661.66 keV photopeak. Counting times were adjusted to achieve reasonable counting statistics.

For mushroom samples, the minimum detectable activity (MDA) based on Currie [29] for 40K and 137Cs of this spectrometer were estimated to be 49 and 0.25 Bq kg−1, respectively for the counting time of 10,000 s and a sample weight of 100 g for 40K and 70 g for 137Cs. For soil samples, the minimum detectable activities (MDA) for 40K and 137Cs in this spectrometer were 4 and 0.01 Bq kg−1, respectively for the counting time of 10,000 s and a sample weight of 1,650 g. The 40K and 137Cs activity concentrations of mushroom samples were expressed in Bq kg−1 on a dry weight basis. Because not all the soil layers are of equal thickness and densities, the 137Cs and 40K activities of soil samples were expressed in both Bq kg−1 and kBq m−2 on a dry weight basis. 137Cs and 40K activity deposition values (kBq m−2) were calculated as the product of the activity per unit mass (Bq kg−1) and the mass depth of each component (kg m−2). The mass depth (kg m−2) for each of the soil layers was estimated such that the soil density (kg m−3) is multiplied by the depth, from the surface down to midpoint of each layer. The bulk density (kg m−3) of all soil samples was determined as the ratio of weight after drying to fresh soil volume as described by McGee et al. [30, 31]. All measured soil and mushroom activities were corrected for radioactive decay to the sampling date for 137Cs.

Agrochemical analyses of the soil samples were conducted according to standard methods as described by Schlichting et al. [32]; Jackson [33] and Black [34] adopted in the Ege University, Faculty of Agriculture, Department of Soil Science. For each sample, nine relevant soil parameter values were determined, i.e., content of organic matter (%), clay content (%), K (%), sand (%), CaCO3 (%), soil acidity (pH), N, silt, salt, texture.

Alkali–alkaline earth element and heavy metal determinations

To determine trace elements in environmental samples, several analytical methods are available. For example, neutron activation analysis (NAA), atomic absorption spectrometry (AAS), X-ray fluorescence (XRF) and inductively coupled plasma-atomic emission spectrometry (ICP-AES) turned out to be a useful tool for the detection of many trace elements in plant and soil samples [35–38]. Recently, inductively coupled plasma mass spectrometry (ICP-MS) has been used for precise determination of trace elements in environmental samples. Due to its low detection limits, analytical speed, relative lack of chemical interferences, and multi-element capability, the method has been applied to more than 50 elements in environmental samples [39]. In the present study, major and trace elements in mushroom samples were measured by ICP-MS technique in ACME laboratory in Canada.

For ICP-MS analysis, mushroom and soil samples were dried at 60 °C. Soil samples were sieved to −80 mesh and mushroom samples were pulverized. Mushroom samples of 1 g were weighed into centrifuge vials, and soil samples of 0.5 g were weighed into test tubes. Mushroom samples were first cold leached with concentrated nitric acid for 1 h then digested in a hot water both for an additional hour. After cooling a modified Aqua Regia Solution of equal parts concentrated ACS grade HCl and HNO3 and de-mineralized H2O was added to each sample (6 mL g−1) to leach in a hot-water bath (~95 °C) for 2 h (for mushroom sample) or for 1 h (for soil sample). After cooling the solution was made up to a final volume with 5% HCl than filtered. Sample weight to solution volume ratio were 1 g per 20 mL for mushroom and 0.5 g per 10 mL for soil. Solutions aspirated into a Perkin Elmer Elan 6000 ICP mass spectrometer were analyzed for the Full package comprising 51 elements. Duplicate sample preparation and measurement were done for each sample. Two standard reference materials were used to validate the analytical procedure and certified against CANMET Reference Materials for analyzed elements. A mixture of soil and rock material (STD DS5) was used for soil as Standard Reference Material. For mushroom analyses STD V6, made from jack pine (Pinus banksiana), mostly twig tissue with a minor component of needles was used.

Calculation of transfer parameters

The soil-mushroom transfer factors or concentration ratios for a radionuclide are frequently defined as the ratio between the activity of this radionuclide in the fruitbodies and that in the soil. The transfer of a radionuclide from soil to fungal fruitbodies of selected species has been estimated by calculating both the transfer ratio “TF” (the traditional transfer factor) and the aggregated transfer parameter “T agg”, where;

$$ {\text{TF}} = {\frac{{{\text{Radionuclide}}\,{\text{concentration}}\,{\text{in}}\,{\text{fruitbodies}}\,{\text{Bq}}\,{\text{kg}}^{ - 1} \, ( {\text{dry}}\,{\text{wt}} . )}}{{{\text{Radionuclide}}\,{\text{concentration}}\,{\text{in}}\,{\text{soil}}\,{\text{Bq}}\,{\text{kg}}^{ - 1} \, ( {\text{dry}}\,{\text{wt}} . )}}} $$
(1)
$$ T_{\text{agg}} \, ( {\text{m}}^{ 2} \,{\text{kg}}^{ - 1} ) ={\frac{{{\text{Radionuclide}}\,{\text{concentration}}\,{\text{in}}\,{\text{fruitbodies}}\,{\text{Bq}}\,{\text{kg}}^{ - 1} \, ( {\text{dry}}\,{\text{wt}} . )}}{{{\text{Total}}\,{\text{radionuclide}}\,{\text{deposition}}\,{\text{in}}\,{\text{soil}}\,{\text{Bq}}\,{\text{m}}^{ - 2} \, ( {\text{dry}}\,{\text{wt}} . )}}} $$
(2)

For agricultural soils, which are characterized by a more or less homogeneous distribution in soil of the investigated radionuclide, the transfer factor, where the specific activities of the mushroom and the soil are both expressed in Bq kg−1 air dry soil, is frequently used. The situation, however, is different in the case of semi-natural environments, such as forests, meadows or peatlands, where soil is characterized by layers or so-called horizons with different fallout radionuclide activity levels. Here, because the multi-layered character of forest soils in which bulk density, chemical properties and the distribution of root systems (including mycelia) are highly heterogeneous, the vertical distribution of a radionuclide in the root zone is, even many years after its deposition, usually quite inhomogeneous and the dry weight density of these soils can vary greatly, especially if organic soils are considered [40, 41 and references therein]. For these ecosystems the aggregated transfer factor T agg, defined as the ratio of the radionuclide activity concentration in a mushroom sample (Bq kg−1, air-dry weight) to the radionuclide soil deposition at the forest site where mushrooms were collected (in Bq m−2), is usually proposed. T agg value defines the fraction of radionuclide activity deposited on a square meter transferred to 1 kg of mushroom fruit body dry matter [6].

According to what is taken to be the activity of the soil, such as the activity deposited on the surface layer, the total specific activity, or the activity of a particular layer (depth-specific measurements), a major dispersion of the reported transfer factors can be found in the literature (Table 1) and this situation makes it difficult to compare the transfer parameters among different studies. Therefore, although it is more valid to express the T agg values in the studies on forest ecosystem, the traditional transfer factors were also presented in this study.

Statistical analysis

All statistical evaluations were carried out with SPSS 8.0 version. Statistical analyses for possible significant correlations between the parameters were performed with non-parametric Spearman rank correlation analysis. The frequency distribution of data sets was tested against a normal or lognormal distribution by the Kolmogorov–Smirnov test (significance level p > 0.05).

Results and discussion

Soil-to-mushroom relationship of 137Cs and 40K

Because of quantification of the radionuclide uptake of a plant from the soil by transfer factors is only meaningful if a linear relation between the radionuclide concentration in the soil and the plant exists [40] and previous studies show that 137Cs and 40K activity concentration in soil can be an influencing factor on 137Cs levels in plants [19], the relationship between the 137Cs and 40K activities in soil and in mushrooms was examined. In order to investigate the presence of a possible correlation, the Spearman rank correlation coefficients (r sp) were calculated for both activity concentration and deposition in soil. The four correlation analyses performed and the results are given below as follows: (1) 137Cs activity of mushrooms versus 137Cs activity concentration of soil; (2) 137Cs activity of mushroom versus 137Cs activity deposition of soil; (3) 40K activity of mushrooms versus 40K activity concentration of soil; (4) 40K activity of mushrooms versus 40K activity deposition of soil. Using the complete data set, significant relationship was found (r sp = 0.756, p < 0.01) between the 137Cs activity of mushrooms and 137Cs activity concentration of soil. There is not a statistically significant relationship between the 137Cs activity of mushroom and 137Cs activity deposition of soil; between the 40K activity of mushrooms and 40K activity concentration of soil; between the 40K activity of mushroom and 40K activity deposition of soil at the 90% or higher confidence level. Nevertheless it must be pointed out that failure to detect a significant correlation between two quantities does not necessarily imply that such a correlation does not exist as mentioned recent publications [19 and references therein].

In addition, the relationships between the 137Cs activity of mushrooms versus 40K activity concentration of soil and 137Cs activity of mushrooms versus 40K activity deposition of soil were also investigated. Although there is not a correlation between 40K and 137Cs activity in mushroom species, statistically significant negative correlations were found both 137Cs activity of mushrooms versus 40K activity concentration of soil (r sp = −0.767) at the p = 0.01 level, and 137Cs activity of mushrooms versus 40K activity deposition of soil (r sp = −0.854) at the p = 0.01 level.

Transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K

Based on the evidence that mushrooms take up radiocesium and stable cesium from organic layers of forest soil [42, 43] and on the results of our studies which showed that most of radiocesium in Turkish forest soil is located in organic layers [44], transfer factors for 137Cs and 40K were calculated according to Eqs. 1 and 2 for the investigated species, according to the total organic layer.

Transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K, both the mass concentration averaged over the soil column (Bq kg−1 d.w.) and the area concentration (or deposition density, Bq m−2) for some mushroom species, collected in areas with different soil properties are presented in Table 2. The results given in Table 2 shows significant difference in the transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K not only for mushrooms with different nutritional status, but also for mushrooms, belonging to different genera and families. The results for 137Cs and 40K were within the range of variation reported in previous studies (Table 1) and the most efficiently transferred radionuclide was 40K. While the species of mushroom that presented the highest TF for 137Cs was S. scabrosus, for 40K was L. nuda. Because all the fungus species have TF and T agg for 40K greater than unity, it seems to indicate that potassium is essential for fungi.

Table 4 lists the statistical data (median, mean value, standard error of mean, standard deviation, range, skewness and kurtosis) and the type of frequency distributions of transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K. Figure 1 shows the fits made to the empirical frequency distributions, taking into account the values obtained for skewness and kurtosis coefficient. Given the approximate null value of the skewness coefficient obtained for transfer factors (TF) and aggregated transfer factors (T agg) of 40K, this distribution is practically symmetrical as can be seen in Fig. 1b–d, while the positive values obtained in the statistics of the transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs indicate that the distribution is asymmetric with the right tail being longer than the left, as can be seen in Fig. 1a–c. The fact that the kurtosis coefficient values of transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs is positive indicates that the distribution is higher and narrower than normal.

Table 4 Summary statistics for the transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K
Full size table
Fig. 1
figure 1

Frequency distributions of a TF (137Cs), b TF (40K), c T agg (10−3 m2/kg) (137Cs) and d T agg (10−3 m2/kg) (40K) for different species of mushrooms analyzed. Also shown are fits of transfer factors (TF) and aggregated transfer factors (T agg) for 137Cs and 40K to a log-normal distribution or a normal distribution

Full size image

Effect of soil characteristics on the transfer factors

It has been suggested that various factors influence the transfer of radiocaesium to fungal fruitbodies, including: mycelium depth; habitat and or sampling strategy; soil clay content; and soil moisture and or microclimate [45 and references therein]. Table 5 presents the following agrochemical soil characteristics of the investigated areas: organic matter (%), clay (%), K (%), pH, sand (%), CaCO3 (%), silt (%) and salt (%). The statistical significance of the association between T agg, TF values of 137Cs and eight soil agrochemical characteristics was explored for mushroom samples and presented in Table 6. Relationships between 137Cs transfer factors of mushrooms and most soil parameters are negative as indicated by previous studies [6]. The significant correlations are observed with silt (%) and K (%) in soil.

Table 5 Agrochemical soil properties of the investigated sites, mushrooms were sampled
Full size table
Table 6 Pair correlation coefficients between the transfer parameters (TF, T agg) and agrochemical soil properties for 137Cs
Full size table

Transfer factors for alkali–alkaline earth elements and heavy metals

If the averaged mushroom and soil concentrations for each element are compared (Fig. 2), it is seen that the linear relationship required for the TF approach is valid. In order to estimate the accumulation of each element by mushrooms, transfer factors (TFs) from soil to mushrooms were calculated by using the element concentrations in the surface soil (total organic layer) collected in the Çinardibi and Kurudere forest sites. The calculated TFs for 8 alkali–alkaline earth elements and 11 heavy metals are summarized in Tables 7 and 8. The results are within the range of variation reported in previous studies except for Zn [15, 16, 39]. According to the concentration ratios that higher than 1, the most efficiently transferred elements seems to have been K, followed by Rb, Zn, Cu, Cd, S, Cs and Hg. While the values for these elements are above the best-fit line in Fig. 2, the ratios for the others were lie below the line. It is obvious that K is an essential element that must be taken up for plants to complete their life cycles and mushrooms can be characterized as Rb, Zn, Cu, Cd and Cs accumulators as mentioned previous studies [5, 39].

Fig. 2
figure 2

Relationship of mean mushroom concentration to mean soil concentration for studied elements in Çinardibi forest site. Also shown is best-fit line (r = 0.760, significance level: p < 0.01)

Full size image
Table 7 Transfer factor (TF) values of alkali–alkaline earth elements for different mushroom species at sampling areas
Full size table
Table 8 Transfer factor (TF) values of heavy metals for different mushroom species at sampling areas
Full size table

Unlike other publications, the transfer factor of 137Cs was lower than that of stable Cs in this study [35, 46]. Although TF for stable Cs had the higher variability depending on the mushroom species (range 0.2–10.6), the high TFs for stable Cs indicate that mushrooms are important Cs accumulators also. In addition, a good correlation between the transfer factors for 137Cs and stable Cs was shown in our previous study [12]. This finding shows that about 16 years after the accident bioavailability of 137Cs from Chernobyl and stable Cs is similar, at least for species taking up radiocesium from the organic layers.

Conclusion

The present study was initiated aiming at evaluating the soil to mushroom transfer parameters of 137Cs, 40K, some stable elements and heavy metals in 25 mushroom samples, covering 12 biological species. The following conclusions can be derived;

  • Significant positive relationship was found between the 137Cs activity of mushrooms and 137Cs activity concentration of soil, while negative correlations were found both 137Cs activity of mushrooms versus 40K activity concentration of soil and 137Cs activity of mushrooms versus 40K activity deposition of soil.

  • Transfer factors for 137Cs and 40K were within the range of variation reported in previous studies and the most efficiently transferred radionuclide was 40K.

  • Statistics of the transfer parameters (TF and T agg) indicate that the distribution is asymmetric for 137Cs and symmetrical for 40K.

  • The influence of agrochemical properties of forest soils on 137Cs transfer factors were revealed, so that the significant correlations are observed between silt and K in soil and 137Cs transfer factors.

  • The most efficiently transferred elements seems to have been K, followed by Rb, Zn, Cu, Cd, S, Cs and Hg.