Environmental Science and Pollution Research

, Volume 19, Issue 8, pp 3362–3370

Heavy metals of the Tibetan top soils

Level, source, spatial distribution, temporal variation and risk assessment


  • Jiujiang Sheng
    • Key laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau ResearchChinese Academy of Sciences
    • Key laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau ResearchChinese Academy of Sciences
  • Ping Gong
    • Key laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau ResearchChinese Academy of Sciences
  • Lide Tian
    • Key laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau ResearchChinese Academy of Sciences
  • Tandong Yao
    • Key laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau ResearchChinese Academy of Sciences
Research Article

DOI: 10.1007/s11356-012-0857-5

Cite this article as:
Sheng, J., Wang, X., Gong, P. et al. Environ Sci Pollut Res (2012) 19: 3362. doi:10.1007/s11356-012-0857-5



Due to its high elevation, rare human activities and proximity to south Asia where industries are highly developed, it is required to investigate the fragile environment of the Tibetan Plateau. We are aiming to obtain the concentration level, source, spatial distribution, temporal variation and potential environmental risk of Tibetan soils.


A total of 128 surf ace soil samples were collected and analyzed f or V, Cr, Mn, Co, Ni, Cu, Zn, As, Cd and Pb, and an additional 111 samples were analyzed f or Hg and total organic carbon. Concentration comparisons coupled with multivariate statistics were used to analysis the sources of elements of soils. We also carried out Risk assessment on the soils.


Concentrations of Hg, Cr, Ni, Cd and Pb are slightly higher than those of the late 1970s. Concentrations of Cr and Ni are higher than averaged world background values. Tibetan soils present a high natural As concentration level.


Anthropogenic sources may partly contribute to the elevated Hg, Cd and Pb concentrations. Cr and Ni are mainly originated from soil parent materials. Soil elements in Anduo and Qamdo regions may threaten the health of local people.


Heavy metal elements of Tibetan Plateau are mainly from the natural source. Arsenic present a high background level. Soil elements in Anduo and Qamdo regions may threaten the health of local people, which should be of concern to scientists and the government.


Trace elementHeavy metalSoilTibetan PlateauMultivariate statisticsRisk assessment

1 Introduction

Soil plays an important role in the environment and acts as a reservoir or a sink for many kinds of pollutants. Pollutants stored in soil can be transferred to other ecosystems, such as groundwater or crops, and consequently affect human health. Among numerous soil pollutants, heavy metals (Cd, Cu, Pb, Zn, Hg and As) are especially dangerous due to their toxicity and persistence in the environment.

Much attention has been paid worldwide to hazardous elements for the purpose of understanding the toxic effect when their concentrations exceed threshold values (Facchinelli et al. 2001). Studies on geochemistry of soil contaminants (particularly heavy metals) have mainly focused on a number of cities, where the sources of heavy metals in soils were mainly anthropogenic contaminations, i.e., traffic emissions (Manta et al. 2002; Lee et al. 2006), industrial wastes (Rawlins et al. 2006), residential activities (Zhang 2006) and the mix of all these sources (Luo et al. 2007; Chen et al. 2008; Shi et al. 2008). Some other researchers conducted their works (Chen et al. 2009; Qishlaqi et al. 2009; Yang et al. 2009) in non-urban areas to assess the possible influence of agricultural activity on the input of hazardous elements into soils.

Anthropogenic inputs are always mixed with lithogenic inputs (elements derived from parent materials and bed-rock), and both of them contribute to the sources of elements in soils. Multivariate analysis [principal component analysis (PCA) and cluster analysis (CA)] are widely used techniques to distinguish between geogenic and anthropogenic sources of elements in soils. Pollution maps, which are produced by interpolating measured concentrations of samples, may also provide information about the spatial distributions of elements.

The Tibetan Plateau is one of the most remote and isolated regions in the world. Due to its high altitude and cold climate, the Tibetan Plateau is often called “the third pole”. The huge Tibetan Plateau, occupying an area of almost 2.5 million km2, is adjacent to Eastern China, India and Nepal, where industries are developed and population is booming. Air masses over the Tibetan Plateau are mainly dominated by continental air from central Asia and maritime air from the Indian Ocean (Wang et al. 2010). Although there are limited industrial and agricultural activities in Tibet, model predictions clearly showed that the Tibetan anticyclone could “trap” anthropogenic emissions lifted from South Asia (Fu et al. 2006). Recently, the environmental change of the third pole area is being a great concern among the scientists (Qiu 2010). Findings and data in this region would be useful in achieving a better understanding of the environmental quality there, thereby being helpful to administrators for adopting appropriate management strategies to keep the pristine environment of the third pole.

To date, only a few papers have reported the concentrations of heavy metals in top soils of the Tibetan Plateau. Li et al. (2009) collected soil samples in western Tibet and found that concentrations of some elements, such as B, As, Cs and Bi in soils are much higher than those of upper continental crust. The study area of that research only covers a small part of the plateau; thus, it cannot represent the overall situation of the Tibetan soils. Zhang et al. (2002) conducted extensive samplings and collected more than 200 samples across the Plateau. However, these samples were obtained several decades ago (1979, 1982, 1983, 1984, 1985 and 1988) when a series of scientific expeditions to the Tibetan Plateau were implemented. Although local human influence is very limited, the increasing expansions of cities and tourism-related activities have occurred in the Plateau as well. Mercury is able to undergo long range atmospheric transport and can be transported from India to south Tibet and further deposits at the Earth’s surface, i.e., glacier, surface soil and forest vegetation (Wang et al. 2008). Thus, both the atmospheric input of pollutants from neighboring regions and input from local activities may lead to the contamination of the Tibetan soil environment. However, few studies have been conducted to evaluate the current concentration levels of heavy metals on plateau scale and to provide information on temporal variation of elements during ~30 years’ (from late 1970s to late 2000s) development.

The objectives of the this study were (1) to determine the concentrations of V, Cr, Mn, Co, Ni, Cu, Zn, As, Cd, Hg and Pb in top soils of Tibetan Plateau, (2) to delineate their main sources using GIS, multivariate statistics and spatial interpolation, (3) to characterize their spatial and temporal variations and (4) to assess the environmental risk regarding heavy metals.

2 Materials and methods

2.1 Sampling and chemical analysis

Between the year 2007 and 2009, 128 surface soil samples (0–15 cm) were collected across the Tibetan Plateau and analyzed for ten elements: V, Cr, Mn, Co, Ni, Cu, Zn, As, Cd and Pb. These samples and an additional 111 samples (239 samples in total) were measured for Hg and total organic carbon (TOC) to better understand the spatial difference that may be caused by soil TOC and/or potential anthropogenic sources of Hg. Sampling locations (Fig. 1) were identified using a global positioning system (GPS). Tibetan Plateau is so mountainous and has a poor road connection. Thus, sampling sites are somewhat unevenly distributed. In addition, we try to collect samples from the same or similar position where old samples had been collected in the late 1970s. This gives us opportunities to make comparison. In the field, about 1 kg of soil material was collected at each site. Fresh samples were contained in plastic bags and taken to a laboratory for further analysis. All samples were air-dried at the temperature of ~ 20 °C and removed of visible stones or other debris. Soil was then sieved to < 2 mm, and the < 2-mm fraction was used for all analyses. A 10-g split of the retained soils was transferred to an agate grinder for milling until all soil powder could go through the 0.074-mm mesh. Finally, the milled soils were kept in plastic bags for chemical analysis.
Fig. 1

Map of Tibetan soil sampling sites(dark blue points represent samples analyzed for all 11 elements, while yellow points represent extra samples only analyzed for mercury and TOC

For Hg determination, 0.1 g of the grinded sample was extracted using 10 mL of mixed solution (2 mol L−1 HNO3 and 4 mol L−1 HCl) at 100 °C for 1 h in rigorously acid-leached silicic borate beakers. Mercury concentrations were measured by cold vapor-atomic absorption spectrometry (CV-AAS) following reduction with SnCl2. The accuracy of the method was assessed by analysis of the national standard reference material (soil, GSS-2), and values within 81 to 95 % of the certified values were obtained (relative standard deviation is between 6 and 10 %). During digestion and measurement, GSS-2 and sample blanks were analyzed every 15 samples.

For other elements, approximately 0.1 g of milled soil was digested in 2 mL concentrated HNO3 and 1 mL HClO4 at a temperature of 180 °C for 16 h. The residue was then dissolved in 2 mL of 4 mol L−1 HCl and diluted to 10 mL with deionised water. The concentrations of V, Cr, Mn, Co, Ni, Cu, Zn, As, Cd and Pb were measured by inductively coupled plasma mass spectrometer (ICP-MS; X-7, Thermo-elemental, USA). 115In was used as an internal standard, and the reference samples (GSS-1 and GSR-1) were used for quality control. The measured values for reference samples are within the range of 84–103 % of the certified values for all ten elements, and the corresponding relative standard deviation values (analytical precision) of them are less than 5 %. TOC content in soil (percent) was measured using a TOC analyzer (Vcph, SHIMADZU, Japan).

2.2 Statistics

Supporting information (Table S-X1) provides the detailed elements concentrations. Descriptive statistics including arithmetic mean and median, maximum, minimum and standard deviation were calculated and listed in Table S-1. The temporal variations of element concentrations for all samples were checked through a general consideration of the skewness, minimum, maximum, mean and median values. The distributions of the data were tested for normality by Kolmogorov–Smirnov (K–S) test. When concentrations were not normally distributed, they were transformed by applying logarithms to obtain a normal distribution.

Principle component analysis is a technique widely used for reducing the dimension of multivariate task by explaining the correlation among a large number of variables through a smaller number of underlying factors (principal components). The factors can be rotated to make them more explicable to interpret (Luo et al. 2007).

To obtain the overall patterns of the elements, the spatial interpolation method of IDW (Inverse Distance Weighted) was applied with 12 neighboring sampling points used for estimation of each grid point (pixel of map). The power of two was chosen to acquire some degree of smoothing effect.

2.3 Risk assessment

Which elements in Tibetan top soils may result in harmful effects? Where may be the risk areas in the Plateau? In order to answer these questions, three indexes, namely geoaccumulation index (Igeo), enrichment factor (EF) and Nemero Synthesis Index (PN) were employed to evaluate the possible environmental risks. Igeo enables the assessment of soil contamination (Gowd et al. 2010; Iqbal and Shah 2011).
$$ {I_{\rm{geo}}} = {\log_2}\left( {\frac{{{C_{\rm{n}}}}}{{1.5{B_{\rm{n}}}}}} \right) $$
where Cn is the concentration of a given element in the top soil, while Bn is the concentrations of elements in the upper continental crust (Taylor and Mclennan 1995; Wedepohl 1995). The constant 1.5 allows us to analyze natural fluctuations in the content of a given substance in the environment (Müller 1969). Seven classes of Igeo represent the increasing soil contamination levels (Table S-2).

EF is a useful tool to differentiate between the anthropogenic and natural sources of the metals (Loska et al. 2004; Reimann and de Caritat 2005; Iqbal and Shah 2011). EF can be calculated as [Me(sample)/Fe(sample)]/[Me(earth crust)/Fe(earth crust)]. Five contamination categories are shown in Table S-2.

The PN method was applied to obtain the potential contaminated area (Chen et al. 2008; Lia and Feng 2010). PN can be calculated by Eq. (2):
$$ {\hbox{PN}} = \sqrt {{\frac{{\left( {\frac{{{C_i}}}{{{S_i}}}} \right)_{\max }^2 + \left( {\frac{{{C_i}}}{{{S_i}}}} \right)_{\rm{ave}}^2}}{2}}} $$
where PN is the synthesis evaluation score corresponding to each sample, Ci is the measured concentration value of the ith kind of element for each sample, Si is evaluation criterion of the ith kind of element, and n is the total number of elements analyzed for each sample. In this study, the concentrations of the elements obtained in the late 1970s (Cheng and Tian 1993) were taken as criterion values. Five grades were defined based on Chinese Soil Environmental Quality Assessment Standard for Green-Food Production area and are listed in Table S-2 (State Development Center for Green-Food of China 2000). PN represents the integrated risk loads of all studied elements, and each sample site has a PN value of its own. After interpolating, a map of spatial distribution of PN value can be obtained. The areas with higher PN values have possible higher environment risks.

2.4 Data analyses using computer software

Prior to multivariate analysis, the tests for normality on the raw and log-transformed data were performed using SPSS software. To reduce the complexity of the data and classify elements into independent groups, multivariate analyses of PCA and CA were carried out using SPSS as well. All element concentration maps were produced by ArcGIS Version 9.2 software.

3 Results and discussion

3.1 Basic statistics and concentration comparisons

The basic statistics for the raw data are shown in Table S-1. Roughly judging the parameter values, most elements show very high variation, except for Mn. The Tibetan Plateau consists of several blocks (Aitchison et al. 2000), including Himalaya Block, Lhasa Block and Qiangtang Block separated by east–west trending sutures. The Himalaya consists mainly of Precambrian clastic sediments and metasedimentary rocks in the south (Brookfield 1993), late Proterozoic to early Cambrian metasedimentary rocks in the middle (Parrish and Hodges 1996), late Precambrian to early Paleozoic sedimentary and metasedimentary rocks (Yin et al. 1988) and thick Permian to Cretaceous continental margin sequences in the north (Brookfield 1993). In Lhasa Block, the exposed rocks mainly comprise Jurassic–Tertiary Gangdese intermediate-felsic magmatic rocks and some other marine–terrestrial sedimentary rocks. The exposed rocks in Qiangtang Block mainly include metamorphic rocks and Late Paleozonic shallow marine strata in the west and Triassic–Jurassic shallow marine carbonate rocks interbedded with terrestrial clastic and volcaniclastic strata in the east (Sun et al. 2007). Figure 2 is modified from the previous study (Yin and Harrison 2000) and presents the major plutonic rocks on the Tibetan Plateau. The complexity of the geology possibly induced the big variations of element concentrations in top soils developed from different parent materials (Bowen 1979). In addition, local human activities, for example, tourist visiting, traffic emission and use of fertilizers or pesticides containing trace metals may also lead to the variations of element concentrations (Wang et al. 2001; Yang et al. 2010).
Fig. 2

Simplified geology map of the Tibetan Plateau area after Yin and Harrison (2000)

Compared with the mean element concentrations of the upper continental crust (Table 1), Cr, Mn, As, Cu, Pb, Zn, V and Ni show enriched values, while Co, Cd and Hg are at similar levels. When compared with the mean values of world background soils, slightly higher values of Cr, Ni and As are observed (Table 1). There is no big difference between element concentrations obtained in our study and concentrations obtained in previous studies in the Nam Co Basin and west Tibet (Table 1). All these comparisons indicate that weathering products of the underlying bedrocks may be the main origins of heavy metals in Tibetan soils. Table 1 also shows element concentrations of soil samples that were collected nearly 30 years ago (Cheng and Tian 1993). One way to evaluate human influence is to compare the concentration values with those archived by others. Only slight increases of Hg, Cr, Ni, Cd and Pb concentrations of our study were observed.
Table 1

Element concentrations from various studies (in milligrams per kilogram)


This study


Background values

Nam Co Basind

West Tibete

Former Tibetan soilsf




































































































































UCC upper continental crust

*No data

aElement concentrations in the upper continental crust (Wedepohl 1995)

bBackground values of the world soils (Bowen 1979)

cBackground values of the world soils (Vinogradov 1959)

dElement concentrations in the Nam Co Basin (Li et al. 2008)

eElement concentrations in west Tibet (Li et al. 2009)

fElement concentrations of Tibetan Plateau soils (Cheng and Tian 1993)

3.2 Probability distribution

Prior to performing multivariate analysis, it is necessary to check the probability distribution features of the variables. The shape parameters and results of the Kolmogorov–Smirnov (K–S) test for normality (Table S-1) show that the skewness and kurtosis coefficients are large for Hg, V, Cr, Co, Ni and As, showing quite inhomogenous concentration distributions of these elements. This is probably due to extremely high concentrations of several samples. Only one element, Mn, can pass the K–S test with the K–S p value greater than 0.05. We therefore performed log-transformations of the remaining elements, but Cr and Ni still did not pass the K–S test. If mean value plus three times of standard deviations of Cr or Ni concentration was treated as threshold, we found three outliers (samples). After subtracting these three samples, log-transformed concentrations of all elements can pass the normality test (Table S-1). These eliminated samples were not further used in the multivariate analysis and risk assessment.

3.3 Multivariate statistical analysis

Based on the results of probability distribution, the log-transformed data sets were subject to multivariate analysis by PCA and CA (Table S-3 and Fig. S-1). From the rotated component matrix for the PCA (Table S-3), five PCs were extracted, accounting for 84.85 % of the total variance. Based on PC loadings, 11 elements and TOC can be grouped into five PCs (F1–F5).
  1. F1:

    Vanadium, Mn, Cu and Co are associated, displaying high loadings of 0.788, 0.836, 0.867 and 0.864, respectively. Nickel and Zn also show an affinity for F1, with loadings of 0.329 and 0.642, respectively. Cu can be “specifically” adsorbed by oxides of Mn and Fe. V can substitute Fe and be sorbed onto Fe oxides (Adriano 1986). Cobalt always occurs in copper minerals (Young 1957) and is mainly obtained by human as a by-product of copper mining activities (Shedd 2006). These four elements, as well as Ni and Zn, may exist together in soil parent materials, as the weathering products of the exposed bedrocks.

  2. F2:

    Ni and Cr are strongly associated in the second factor (F2), with high loadings of 0.903 and 0.960, respectively. Co also shows relatively high affiliation with these two elements, with the value of 0.385 in F2. Soils developed from ultramafic rocks are usually enriched in Ni, Cr and Co. The collision between the Indian and Eurasian plates led to the uplift of the Tibetan Plateau. The thrust of Lhasa terrane from south to north probably gave the widely distributed ultramafic rock there (Yin and Harrison 2000). Therefore, Tibetan soils developed on ultramafic rocks likely have high concentrations of Cr, Ni and Co.

  3. F3:

    The third factor (F3) consists of Zn, Cd and Pb. The loadings are 0.615, 0.782 and 0.766, respectively. All three elements are commonly found together in various types of ore deposits (Bowen 1979). We deem that these elements primarily share a natural source feature. The third factor is then taken as a natural lithogenic factor.

  4. F4:

    The fourth factor includes Hg and TOC, with values of 0.608 and 0.882, respectively. Pb also shows a positive high value (0.472) in F4. This indicates that the TOC concentration may impact the concentrations of Hg and Pb of soils, probably in wet or nutrition-enriched areas. According to this feature, we conclude that partial Hg (perhaps also Pb) released by human activities may be chelated by soil organic matter.

  5. F5:

    Element As, with a value of 0.842, is somewhat associated with Hg (value is 0.609). The higher As mean concentration compared with the averaged world background value indicates a high background level of As in Tibetan soils. This is probably due to the widely distributed arsenic-enriched rock, for example, shales and arc-related rocks (Li et al. 2011).


From Fig. S-1, a tree diagram (dendrogram, CA analysis) was generated, and it shows five element clusters. These clusters have strong relations with the above discussed five PCA factors. This further confirms the PCA results.

3.4 Spatial distributions

The interpolated concentration maps of elements are presented in Fig. 3. V, Cu and Mn have similar distribution patterns. High levels of these elements with red and orange colors on the maps were only found in south central Tibet. Clastic sediments and metasedimentary rocks are well-preserved in southern Tibet (Yin and Harrison 2000). The black shales from the rocks possibly contain relatively high concentrations of Mn, Cu and V (Vine and Tourtelo 1970). These three elements in soils should originate from the weathering of the parent rocks. With regard to other huge areas, the interpolated concentrations are all below the averaged world background values.
Fig. 3

Spatial distribution of trace metals and TOC across the Tibetan Plateau. Classification of concentration was based on Cheng and Tian (1993)

Similar spatial distribution patterns were also found for Ni, Co and Cr. The high concentration sites with dark red colors are shown as two belts on the maps. One is located at north Tibet, extending from Anduo to Qamdo, and the other one is from the southern part, from Mt. Everest to the north edge of Lhasa Block. There is a chromium mining at Anduo, which may result in high level of Cr. Chromium, Ni and Co are always associated with each other during mineralization process. Thus, relatively high levels of Ni and Co were observed in the same belt as well. Volcanic rocks, clastic sandstones, shales and marls are the dominant rocks in Lhasa and Qiangtang blocks. Soils developed on these rocks can have high concentrations of Cr, Ni and Co, indicating the inheritance from parent rocks.

For Zn, Cd, Pb and Hg, the hot spots are mainly located at northern and eastern Tibet. Generally, these elements could be found together in several kinds of rocks in nature. The sources of them may be from the bedrock. Meanwhile, high levels of TOC were observed in some of these regions. Elements chelated by soil organic matter should be an important principle in explaining the storage and distribution of elements (Obrist et al. 2011). Soil may accumulate elements coming from polluted areas far away by long distance atmosphere transport. In addition, local contamination of these elements is possible. The highest concentrations of Zn, Cd and Hg were found at Qamdo, close to western China. Qamdo has relatively developed economy, and industrial and agricultural activities there may contribute to these high concentrations. The traditional cooking with wood, dung and crop residues and butter burning for religious purposes might be sources of Hg in Tibet (Yang et al. 2010). The forest fire could emit Hg, Cd and Pb into the environment (Huang et al. 2011). Similarly, Wang et al. (2010) observed higher levels of volatile pollutants (i.e. hexachlorobenzene) in a Tibean forest site where forest fires occurred during the sampling period. Therefore, local sources could also contribute to the concentration loads of the surface soil.

Spatial distribution of As is quite different from other elements. Most parts of the Tibetan Plateau have a relatively high As concentrations (red areas on the map) in soils, which exceed the averaged world background value. The interpolated concentration map indicates that both the western and northeastern regions of Tibet have high concentrations of As, while southeastern Tibet has a relatively low level. This is probably due to the difference of exposed rocks between different regions.

3.5 Environmental risk assessment

In order to evaluate which elements show relatively higher risks in Tibetan soils, Igeo and EF were calculated. Based on the equation introduced earlier in the article, the Igeo of Hg, V, Cr, Mn, Co, Ni, Cu, Zn, Cd and Pb (Fig. S-2a) are all smaller than 1, indicating minor contaminated levels of these elements. However, the Igeo of As (2.4) is much bigger than values of the other elements, falling into the third category of the index (Table S-2). This indicated a moderate or even heavy contamination in soil. According to the PCA analyses, As in Tibetan soils is mainly from the underlying bedrocks. The high value of Igeo for As demonstrated that the Tibetan soils may have higher As risks which is caused by its higher natural origin. In addition, the EF value of As is the largest among all elements, showing the significant enrichment of this element in soils (Fig. S-2b). Although three samples (outliers) were excluded for EF calculations, relatively higher EF values of Cr, Ni and Pb (falling into class 2) indicated moderate enrichment in soil. Taking all these results (both Igeo and EF calculation) into account, As displayed a distinct feature of environmental risk as compared with other elements. Apart from the element risk assessment, we also carried out the PN calculation which can provide the information about the possible risk regions in the plateau (Fig. 4).
Fig. 4

Interpolated PN value of Nemero Synthesis Index

PN calculation considered the elements with the extremely high loadings, which is represented by a maximum value of Ci/Si. At the same time, PN contained the sum contribution of all kinds of elements, which is finally expressed as the mean value of Ci/Si of all elements. Since the averaged element concentrations for samples collected in the late 1970s were selected as criterions, the spatial distribution map of PN can indicate the areas (1) with element concentrations that increased during the recent 30 years, probably suggesting the anthropogenic contributions and (2) that are strongly influenced by their high parent rock concentrations which are much higher than the averaged level of the plateau. Regarding each hot area distinguished as grade of PN > 3, the reason for the possible higher risks should be carefully investigated.

Although the interpolated PN map (Fig. 4) looks similar to the spatial distribution maps of some single elements in certain areas, it has some differences, showing new insights. Area 1 from the map is a NW–SE strip situated along the Tibetan north boundary. High PN value of this area indicates an integrated element loading driven by Cr and Ni, which were shown by the ratios of their concentrations to the averaged values (Ci/Si in PN calculation, Table S-X1). The existence of chromium mining at the northern Tibet near Anduo could be a source. Erosion or weathering of the Cr- and Ni-enriched bedrock may also release elements to soils in these areas. Area 2 occupied the northeast of Tibet (centered at Qamdo) and some places along Yarlung Zangbo River. The high PN value of this area is dominated by concentrations of Cd, Pb and Hg (Table S-X1). In addition to the mining and the weathering effect of the possible Cd-, Pb- and Hg-enriched bedrocks at Qamdo district, the possible use of fertilizers and/or pesticides containing Cd, Pb and Hg (Wang et al. 2001) at the relatively populous Yarlung Zangbo riverside can also result in the high element loading. In terms of the whole Tibetan Plateau, the areas from Anduo to Qamdo pose potential health risk. If agricultural activities are to be conducted there, the government should do detailed and high-density soil surveys to ensure the safety of the local people.

Although the mean As concentration of our study (19.27 mg kg−1) is only slightly higher than the defined first class (represents natural background of arsenic, 15 mg kg−1) of Chinese Soil Environmental Quality Standard (State Environmental Protection Administration of China 1995), the As concentration can range from 1.83 to 154.5 mg kg−1. High levels of As are found at most eastern, some central south and southwestern parts of the plateau. The high As concentrations may act as a threat to local people through direct plant uptake. Also, the arsenic in soils and bedrock suggests the possibility of high As in groundwater, which would pose a threat to drinking water, especially in the downstream areas (Smith et al. 2000; Chakraborti et al. 2004). In general, the high As background level should be paid great attention to by Tibetan administrators in the future.

4 Conclusions

Heavy metals in Tibetan soils mostly originated from the weathering of underlying bedrocks. However, contamination of Hg, Cd and Pb possibly occurred in some areas of the Plateau that may have a human-influenced source. Tibetan soils present a high natural arsenic concentration level, which should be of concern to scientists and the government. To better evaluate environmental risks, the bioavailable portions of metals in soil should be extracted and measured. In prospective researches, forest areas of the plateau should be further considered due to a high level of carbon content and chelating abilities for several toxic metals.


This study was supported by the Innovation Program of the Chinese Academy of Sciences (KZCX2-YW-146) and the National Natural Science Foundation of China (40871233 and 41071321).

Supplementary material

11356_2012_857_MOESM1_ESM.xls (89 kb)
ESM 1(XLS 89 kb)
11356_2012_857_MOESM2_ESM.doc (306 kb)
ESM 2(DOC 305 kb)

Copyright information

© Springer-Verlag 2012