Introduction

Soil security has direct and indirect effects on human health and ecosystem stability (Zhu et al. 2017). Heavy metal pollution in soils due to rapid urbanisation has become an urgent environmental concern in recent decades (Luo et al. 2012). Identification of the spatial distribution characteristics of heavy metals in soils is the foundation of soil management and ensures the safety of the environment to human beings (Li et al. 2018; Steffan et al. 2018). This is especially important for peri-urban areas where provide ecosystem services, such as food supply, water purification and water conservation, to urban and rural ecosystems but suffer from severe heavy metal pollution due to rapid urbanisation (Kroll et al. 2012). The peri-urban area is the interface of the urban and rural ecosystem and has different types of land use with complex surface structure and high spatial variability. This brings serious challenges to the utilisation, management and protection of soil resources in this area (Fazal 2013). The peri-urban area as a critical area provides the source of urban live and production materials. Meanwhile, the peri-urban area is also the sink of urban and industrial wastes (Zhu et al. 2017). Under the multipressures from the industry, agriculture, transportation and urban lives, the soil quality in the peri-urban area is decreased in recent decades. The accumulation of heavy metals in soils is a typical threat for soil security in peri-urban areas worldwide.

Previous studies have reported that heavy metals in soils have a heterogeneous distribution in vertical and horizontal directions in different ecosystems. For example, because the migration of heavy metals is stable and they accumulate easily in topsoil, many studies have shown that heavy metal concentrations in topsoil are generally higher than their background values and deeper soil layers (Kim et al. 2003; Schulin et al. 2007; Zhao et al. 2007). Furthermore, the heavy metal concentrations in soils usually decrease with increasing distance from the heavy metal sources (Kim and Kim 2008; Massadeh et al. 2016; Poikolainen 1997; Zhang et al. 2015). For example, the areas surrounding factories and roads generally have high heavy metal concentrations, and the concentrations decrease with increasing distance from factories and roads (Gowd et al. 2010; Mahmoudabadi et al. 2015; Wei and Yang 2010; Wu et al. 2011; Velea et al. 2009).

Many factors can affect the spatial distribution of heavy metals in soils: (i) the land use type: significant difference in heavy metal concentrations in different land use types (Li et al. 2015; Wang and Xu 2015; Yu et al. 2014; Zhao et al. 2012). Several studies have reported the effects of land use on heavy metal accumulation in soils of farmlands, vegetable lands, forestlands and orchards (Bai et al. 2010; Guo et al. 2015; Lee et al. 2015; Pandey and Pandey 2009). (ii) The spatial pattern of land use: spatial distribution of land use can significantly influence the spatial variability of heavy metal concentrations in soils. For example, Li et al. (2017) and Yaylalı-Abanuz (2011) found that heavy metal distribution showed spatial characteristics in a rural–urban fringe, where large ecological areas have the lowest heavy metal concentrations and the central region and county residential communities are the main hotspots for heavy metals. In particular, the peri-urban area, where factories, towns, villages, roads and irrigated farmland intersect, shows considerable heavy metal accumulation in soils with greater spatial variability than urban and rural areas (Breward 2003). (iii) Other factors such as altitude, soil properties and human activities also can affect the content and distribution of heavy metals in the soil environment. For example, Dykstra et al. (1995) found that heavy metal concentrations decrease with increasing altitude. Heavy metal distributions and soil properties, such as organic matter and pH, were also significantly correlated. Soil organic matter can absorb high amounts of heavy metals, and soil pH can affect the state, migration and absorption behaviour of heavy metals in soils (Yan et al. 2018; Zhao et al. 2010).

The peri-urban area is the transitional area between urban and rural areas. The peri-urban area generally contains residential land, industrial land, agricultural land, forestland, wetland and other land use types. This area provides essential ecosystem services for urban and rural residents (Huang et al. 2009; Douglas 2006). Therefore, peri-urban areas are highly complex territorial spaces involving interactions between economic, environmental and social factors (Xu et al. 2009). Due to high landscape fragmentation and spatial heterogeneity of land use types occurring during rapid urbanisation in peri-urban areas, this area shows a complex and mosaic distribution pattern of heavy metals in soils. Moreover, due to complex land use composition and spatial pattern in peri-urban areas, higher concentration and more intensive redistribution of heavy metals is observed in the soils of these areas than in those of other areas (Apeagyei et al. 2011; Breward 2003; Meena et al. 2016). The peri-urban area is particularly vulnerable to heavy metal pollution because of high accumulation of heavy metals in soils, leading to decreased soil quality (Huang et al. 2015; Maleki et al. 2015). Heavy metals are a threat to human health as they can enter the human body through the food chain from the polluted peri-urban agricultural lands (Hu et al. 2017). Thus, it is crucial to identify the spatial distribution characteristics of heavy metal concentrations in peri-urban soils to help design effective soil management practices for sustaining environmental and human health (Li et al. 2018; Steffan et al. 2018). Most recent studies on this topic are focused on urban or agricultural areas, and more attention should be paid to peri-urban areas.

The Yangtze River Delta has experienced the fastest urbanisation in China. This has led to an increase in population density, intense industrialisation and excessive exploitation of natural resources, with detrimental effects on the structures and functions of ecosystems (Cai et al. 2017; Tao et al. 2018; Tian et al. 2011). Zhang et al. (2009) demonstrated high heavy metal concentrations in the soils of this area. Thus, identifying the spatial distribution of heavy metals in the soils of peri-urban areas undergoing rapid urbanisation is important for enhancing soil security and sustaining human health. Such information will help in designing effective soil management practices for sustaining multiple ecosystem services in peri-urban areas. Thus, the objectives of this study were to (1) analyse the effects of land use types on heavy metal concentrations in peri-urban soils, (2) identify the spatial distribution characteristics of heavy metals in these soils and (3) explore the main factors affecting the spatial variability of heavy metals in these soils.

Materials and methods

Study area

The Zhangxi Watershed, located in the peri-urban area of Ningbo City in the Yangtze River Delta of eastern China (Fig. 1), was selected as the study area. This watershed has a subtropical monsoon climate with mean annual temperature of 17.4 °C and mean annual precipitation of 1463 mm. The land use types in this watershed are farmland, orchard, forestland, river, reservoirs, villages and towns. Because this study area is located in a peri-urban area, it has a complex land use pattern, especially in the floodplain. The main crops cultivated in the farmland are fritillaries and vegetables. The farmland, villages and towns are mainly distributed in the floodplain along the Zhangxi River. The forestland is covered with a mid-subtropical evergreen broad-leaved forest. The frequent intensive human activities in recent decades have resulted in the destruction of the original vegetation; therefore, the forest now includes plantations and secondary forests. The soil depth in the forestland and orchard of this watershed is 0–50 cm. The soil types include paddy soil, yellow soil, red soil and red yellow soil.

Fig. 1
figure 1

Location of the study area and experimental sites

Full size image

Experimental site selection and soil sampling

Eighty-two experimental sites were selected in this study based on land use types and their spatial locations, and 222 soil samples were collected from different soil depths: topsoil (0–10 cm), second layer (10–20 cm) and subsoil (20–40 cm). Furthermore, three transects (transect 1, transect 2 and transect 3) covering 35 experimental sites were selected to evaluate the correlations between heavy metal concentrations in the soil and land use pattern (Fig. 1). Transect 1 was along the mainstream of the Zhangxi River from upstream to downstream. This transect was from the agricultural ecosystem to the urban ecosystem and comprised 16 experimental sites. Transect 2 and transect 3 were perpendicular to the mainstream river from top mountains to the farmland and comprised 11 and 8 experimental sites, respectively. At each experimental site, soils from five random points were sampled and combined into one composite sample as a representative sample (Zhang et al. 2009). Altitude and distances from roads, the river, towns and villages were measured using Google Earth. To explore the relationships between soil heavy metal concentrations and land use composition, the floodplain was divided into several parts by using the Tyson polygon method. This method is widely used in spatial analysis. Based on the Tyson polygon theory, a straight line is drawn between two adjacent points. Then, a vertical line will be drawn in the midpoint of this straight line. Based on this method, the floodplain was divided into several parts.

Laboratory analysis

All soil samples were air-dried before laboratory analysis and sieved through a 125-μm polyethylene sieve. The copper (Cu), zinc (Zn), cadmium (Cd), nickel (Ni), arsenic (As), chromium (Cr), lead (Pb) and mercury (Hg) contents of each soil sample were measured. The soil samples were weighed to 0.2000 g, placed in polytetrafluoroethylene (PTFE) tubes and digested with a mixed acid containing 2 ml HF and 8 ml HNO3. The PTEF tube contents were then digested in a Microwave Digestion System (MARS, CEM, UK) until the soil was completely dissolved. The contents were then diluted with deionised water to a final volume of 50 ml. Cr, Cu, Zn and Ni concentrations were determined by inductively coupled plasma optical emission spectrometry (ICP-OES, OPTIMA 8300, USA), and Cd and As concentrations were determined by inductively coupled plasma mass spectrometry (ICP-MS, NexION 300X, USA). Pb and Hg concentrations were determined by X-ray fluorescence spectrometry (XRF; ARL Perform’X 4200 XRF spectrophotometer). To prepare pellets for XRF analysis, the contents were pressed on a boric acid base in a hydraulic press at a pressure of 25 tons.

The basic physical and chemical properties of the soil samples were also evaluated. Soil texture was measured using a laser particle size analyser (Mastersizer 2000). To measure soil pH, the soil samples were dissolved in deionised water (soil to water volume ratio of 1:2.5) and equilibrated overnight (Esser 1996). The amount of available nitrogen (AN) in the soil samples was measured by the micro-Kjeldahl method after extraction with 1 N KCl. The amount of available phosphorus (AP) was measured colorimetrically after extraction with 0.5 M NaHCO3 (pH 8.5) (Page et al. 1982). The amount of available potassium (AK) was measured using a flame photometer after extraction with 50 ml 1 mol L-1 NH4OAc (pH 7.0), and the amount of soil organic carbon (SOC) was determined by the dichromate oxidation method (Walkley and Black 1934).

Statistical analysis

Statistical analysis was performed using Microsoft Office Excel 2013 and SPSS 22.0. One-way analysis of variance was used to determine the differences in heavy metal concentrations in soils between different experimental sites. p values of ≤ 0.05 were considered statistically significant. Pearson correlation analysis was used to identify the correlations between heavy metal concentrations in soils and soil properties.

Results

Heavy metal concentrations in soils of different land use types

Comparison of heavy metal concentrations in soils of different land use types is shown in Fig. 2. The background values reported by Dong et al. (2007) were used. The results showed that concentrations of all heavy metals except As were higher in the farmland than in the orchard and forestland. Furthermore, most heavy metal concentrations in the farmland were higher than the background values, and those in the orchard and forestland were lower than the background values. The concentrations of heavy metals other than Cr and Ni were higher than the background values in one or more land use types (Fig. 2). The vertical distribution of each heavy metal is shown in Table 1. Because Pb and Hg concentrations were determined only in the top soil, vertical distributions of the other six heavy metals were analysed. Table 1 shows that heavy metal concentrations in soils decreased with increasing soil depth.

Fig. 2
figure 2

Mean heavy metal concentrations in soils of different land use types

Full size image
Table 1 Vertical distribution of heavy metals in soils of different land use types (mean ± standard deviation, unit: mg kg−1)
Full size table

Spatial distribution of heavy metals in soils at different transects

Transect 1 was used to evaluate the spatial distribution of heavy metals from the agricultural ecosystem to the urban ecosystem in the farmland soils. When transect 1 passed through the Miyan village in the upstream (Fig. 1), the concentrations of all heavy metals in the farmland gradually decreased, except that of Ni, which gradually increased (Fig. 3). When transect 1 passed through Zhangshui town, the heavy metal concentrations in the farmland rapidly increased initially and then gradually decreased with increasing distance from Zhangshui town towards downstream. The results also showed that the heavy metal concentrations in soils varied with land use pattern. For example, Fig. 3 shows that maximum heavy metal concentrations in soils occurred in the midstream area comprising towns and villages. The concentrations then decreased in the downstream area and increased at sampling sites close to Yinjiang town in the downstream. Figure 3 also indicates that high heavy metal concentrations in soils were mostly found in town locations. Interestingly, the concentrations of most heavy metals, except Ni and Pb, were significantly higher in the midstream (peri-urban area) than in the upstream and downstream (Table 2).

Fig. 3
figure 3

Variations in heavy metal concentrations in the soils of transect 1. Note: W*** represents the experimental site name in the X axis

Full size image
Table 2 Statistics of heavy metal concentrations in the soils of different parts of transect 1 (mg kg−1)
Full size table

Figure 4a demonstrates that the heavy metal concentrations in forestlands increased from top hill to foothill in transect 2. Human activities on forestland 1 were limited; therefore, the heavy metal concentrations in soils were not significantly different between various experimental sites on this hillslope. Further, the heavy metal concentrations in forestland 2 were higher than those in forestland 1. Pb, Cr and Cu concentrations were slightly higher on the top hill than on the foothill in forestland 2. The heavy metal concentrations were significantly higher in farmland soils than in forestland soils in transect 2 (Table 3). In particular, Hg was observed only in the farmland in transect 2. The land use pattern in transect 3 was the same as that in transect 2 (Fig. 4b). Thus, the spatial distribution of heavy metals in soils in transect 3 also showed a pattern similar to that in transect 2 (Table 3). Interestingly, in transect 3, some heavy metal concentrations were higher on the top hill than on the foothill. This difference was because of the presence of an ancient road running from the foothill to the top hill, which explains heavy metal accumulation in top hill soils due to human activities and transportation.

Fig. 4
figure 4

Heavy metal distribution from the natural ecosystem to the agricultural ecosystem of a transect 2 and b transect 3

Full size image
Table 3 Statistics of heavy metal concentrations in the soils of different land use types in transect 2 and transect 3 (mg kg−1)
Full size table

Factors affecting the spatial variability of heavy metal concentrations in peri-urban soils

To identify the effect of land use composition on heavy metal concentrations in soils, the floodplain of the watershed was divided into 16 parts to calculate the percentage of town area in each part based on the spatial distribution of experimental sites (Fig. 5). The results showed that the concentrations of all heavy metals except As in soils increased with the increasing percentage of town area (Fig. 6). In particular, Cu, Zn and Cd concentrations showed significant positive correlations with the percentage of the town area. These results also proved that the land use type has a determinate role on heavy metal concentrations in peri-urban soils. The results of this study showed that the average concentration of Cr in farmland soil was lower than that of background value (Fig. 2), but Cr concentration in parts of the experimental sites in farmland were higher than the background value (Fig. 4). These experimental sites were usually located in an area with relatively higher percentage of town area. This was the reason that Cr concentration was significantly positive correlated to the ratio of town area to farmland area in the floodplain (Fig. 6). This also can prove that the spatial distribution of Cr in floodplain farmland was influenced by human activities.

Fig. 5
figure 5

Division map of the floodplain of the study area

Full size image
Fig. 6
figure 6

Correlations between heavy metal concentrations in soils and ratio of town area to farmland area

Full size image

Significant correlations between different heavy metals reflect the homologous relationship or compound pollution between these heavy metals; not significant correlations reflect that heavy metals may come from more than one source (Micó et al. 2006). Pearson’s correlations between various heavy metals and soil properties are shown in Table 4. Most of the heavy metals evaluated in this study were significantly correlated with each other, except for As, Pb and Hg. A significant correlation was observed between Cu, Zn and Cd, suggesting that they were introduced in the peri-urban soils from the same sources. Significant positive correlations of Cu, Zn and Cd with SOC, AP and AK were also found, indicating that Cu, Zn and Cd may have been introduced in the soils due to fertilisation practices. Further, similar correlations of Cr and Ni with SOC and AN were observed. The correlation analysis showed that the amounts of fertilisers used in the farmland affected the heavy metal concentrations in farmland soils; however, because random amounts of fertilisers were used, it was difficult to quantify the correlation between heavy metal concentrations in soils and fertilisation.

Table 4 Correlation between heavy metal concentrations in soils and soil properties in transect 1
Full size table

Correlations between heavy metal concentrations in soils and environmental factors, namely distance from roads, distance from towns, altitude and soil clay content, are shown in Fig. 7. The concentrations of most heavy metals in soils decreased with the increase in all four factors. No significant negative correlations were found between soil clay content and Cr and Ni concentrations. Furthermore, Ni and As concentrations showed no significant correlation with any of these four factors (Fig. 7).

Fig. 7
figure 7

Correlations between heavy metal concentrations in soils and a distance from roads, b distance from towns, c altitude and d soil clay content

Full size image

Discussion

Effects of land use types on the heavy metal concentrations in soils

Previous studies have revealed that land use types, especially those involving intensive human activities, significantly affect heavy metal concentrations in soils (Simon et al. 2013). In the present study, the concentrations of most heavy metals in the farmland and orchard were higher than those in the forestland and the background values (Fig. 2). This finding was consistent with those of Bai et al. (2010) and Wang and Xu (2015), who found that the heavy metal concentrations in the soils of vegetable and maize fields were higher than those in the forestland. They concluded that farmlands have higher heavy metal concentrations than other land use types because of fertilisation. The results of this study showed that Cu, Zn, Cd, As, Pb and Hg concentrations in the farmland; Cd, As, Pb and Hg concentrations in the orchard; and Cd, Pb and Hg concentrations in the forestland were higher than their background values (Table 1). The results of heavy metal concentrations in farmland and orchard soils were consistent with those reported by Huang et al. (2018) who showed that Cd, As, Pb and Hg were highly accumulated in the soil of Ningbo due to increased anthropogenic emissions.

Heavy metal sources can be natural or anthropogenic. The natural sources include underlying bedrock and geological parent materials, and anthropogenic sources are mainly fertilisers, pesticides, irrigation practices and atmospheric deposition (Cai et al. 2012; Imperato et al. 2003; Naveedullah et al. 2013; Zhao et al. 2007). The high accumulation of heavy metals in urban and peri-urban soils mainly results from fertilisation, traffic and industrial activities (Wei and Yang 2010). Fertilisers contain a large amount of heavy metals, and their concentrations are higher in organic fertilisers than in chemical fertilisers (Atafar et al. 2010; Wang and Li 2014). Long-term fertilisation can lead to the accumulation of heavy metals, especially Cr, Ni, Cu, Zn, Pb and Cd (Guo et al. 2018; Zhou et al. 2015). In this study, field investigation revealed that approximately 15,000 kg hm-2 of organic fertilisers are applied on farmland, with regular spraying of pesticides and almost no irrigation. Most of fertiliser was organic fertiliser, such as pig and chicken manure. The contents of heavy metal in organic fertilisers were relatively high, especially for Cu, Zn, As, Pb and Cr (Qin et al. 2015). Thus, long-term fertilisation led the heavy metal accumulation in soils, especially in surface soils. For example, Brock et al. (2006) found that the content of Zn and Cu in farmland soil increased after 40 years of fertilisation experiment. Previous studies also found Cd, As, Pb, Zn and Hg accumulation in surface soils after long-term fertilisation (He et al. 2017; Wu et al. 2012). Table 4 indicated that Cd, Cu and Zn concentrations are significantly associated with SOC, AP and AK and that Cr and Ni are significantly associated with SOC and AN. These correlations indicate that the main source of Cr, Ni, Cu, Zn and Cd is fertilisation. Long-term fertilisation with organic fertiliser is the main source of heavy metals in soils, and the specific types of fertiliser and amount used in agricultural land affect the spatial variability of heavy metal concentrations in soils.

Traffic is also considered as a source of heavy metals, especially Cd, Pb, Zn and Pb (Imperato et al. 2003; Wei and Yang 2010). Studies have shown that the soils along roads are highly polluted with Ni, Cr, Pb, Zn and Cd (Li et al. 2015; Luo et al. 2011; Wei and Yang 2010). This explains the effect of the distance from roads on heavy metal distribution in soils. The present study results also showed that the heavy metal concentrations in soils decreased with increasing distance from roads (Fig. 7). Further, the comparison of heavy metal concentrations in soils between different land use types indicated that the land use type and human activities in peri-urban areas significantly affect the heavy metal concentrations.

Human activities and land use pattern affect the spatial distribution of heavy metals in peri-urban soils

Comparison of heavy metal concentrations of soils in the three transects showed that the spatial distribution of land use caused spatial variation in the heavy metal concentrations. In transect 1, heavy metal concentrations were higher in the midstream than in the upstream and downstream (Fig. 3) because the midstream comprised towns and villages. This result confirmed that the location of land use type affects the spatial variability in heavy metal concentrations. Furthermore, the ratio of town and village area to farmland area was found to affect heavy metal concentrations (Fig. 6), indicating that the land use structure was one of the main influencing factors causing spatial variability in heavy metal concentrations in peri-urban areas.

In transect 2 and transect 3, the concentrations of most heavy metals gradually increased from the top hill to the foothill in the forestland (Fig. 4), probably because human activities increased along the hillslope towards the foothill. This result of spatial distribution was consistent with that reported by Ding et al. (2013). Further, higher heavy metal concentrations were found in farmland soils due to human activities than in orchard and forestland soils (Fig. 4, Table 3). This result was consistent with that reported by Li et al. (2007) who concluded that human activities are important factors influencing the spatial distribution of heavy metals in soils.

Higher heavy metal concentrations were found in forestlands accessible to humans than those inaccessible to humans. For example, forestland 1 in transect 2 was difficult for humans to access and therefore remained unperturbed, but forestland 2 in transect 2 was close to towns and villages and thus more prone to being affected by human activities (Fig. 4a). In forestland 2, Cu and Cd concentrations were higher in top hill soils than in foothill soils, probably because the forestland on the top hill was converted from a farmland that would have left residual heavy metals from fertilisation practices at this experimental site (Fig. 4b). The heavy metal concentrations were higher in forestland 3 than in forestland 4 because the experimental sites in forestland 3 were at a lower altitude than those in forestland 4 and were located close to villages. These results indicate that human activities are important factors affecting the spatial distribution of heavy metals in peri-urban soils. This finding is consistent with that of a similar study conducted along an urban–rural ecological transect in New York (Pouyat and Mcdonnell 1991).

The heavy metal concentrations in peri-urban soils increased with the increasing percentage of towns and villages (Fig. 6). This was especially true for Cu, Zn and Cd concentrations, which were significantly associated with the ratio of urban area to farmland area (p < 0.05). This result indicated that greater urbanisation leads to higher accumulation of heavy metals in peri-urban soils, which was consistent with the results reported by Škrbić and Đurišić-Mladenović (2013) and Fang et al. (2011). In their studies, they found that along the urban–peri-urban–rural gradient, the highest heavy metal concentrations often occurred in the peri-urban soils. Urbanisation involves diverse land use types leading to their complex spatial distribution in peri-urban areas; this consequently causes complex spatial variability in heavy metal distribution in soils. Furthermore, correlations between the heavy metal concentrations in soils and the percentage of town area can reflect the potential spatial variability in heavy metal concentrations in peri-urban soils.

Other factors affecting the spatial distribution of heavy metals in soils

The correlation analysis of heavy metal concentrations in soils and environmental factors showed that Cu, Zn and Pb concentrations in peri-urban soils were significantly associated with altitude, soil clay content, distance from towns and distance from roads (Fig. 7). In particular, the Cd concentration was significantly associated with altitude and distance from towns, and the Hg concentration was significantly associated with distance from towns. This result is consistent with that reported by Li et al. (2014) who showed that Cu, Zn, Pb, Cd and Hg concentrations in soils tend to decrease with increasing distance from built-up areas. As, Cu, Zn, Pb and Hg are the major anthropogenic heavy metals (Abdallah 2011; Han et al. 2006). Based on the results of this study and previous studies, it can be concluded that human activity is one of the dominant factors affecting the introduction and distribution of heavy metals in peri-urban soils. Huang et al. (2018) also found that human activities have significantly higher contribution to heavy metal accumulation in peri-urban areas than natural sources. Compared with urban and rural areas, the human activities’ influences on the spatial distribution of heavy metals in soils are significant in peri-urban areas, because of complex land use pattern associated with intensive human activities. No significant correlations of Cr, As and Ni concentrations with the environmental factors were found in this study (Fig. 7), indicating that their distributions were not affected by these factors. The distribution of these heavy metals is probably affected by natural factors, such as geological parent materials and bedrock, but not human activities or land use pattern. Based on the above discussion, it can be concluded that land use pattern and human activities have a combined effect on the concentrations and distribution of heavy metals in peri-urban soils. Thus, the urbanisation process should incorporate a reasonable land use pattern in peri-urban areas. Land use management practices incorporating reasonable farming activities, reasonable land use structure and spatial pattern optimisation are suggested for sustaining soil security in peri-urban areas.

Conclusion

In this study, concentrations and spatial distribution of eight heavy metals (Cu, Zn, Cd, Ni, As, Cr, Pb and Hg) were analysed in a typical peri-urban watershed. The result showed that the concentrations of most heavy metals were higher in farmland soils than in orchard and forestland soils. The heavy metal concentrations in soils decreased with increasing soil depth and increased from top hill to foothill. The heavy metal concentrations in different transects varied with the spatial distribution of different land use types. These results indicate that the land use type and its spatial locations significantly affect the heavy metal concentrations in peri-urban soils. Furthermore, the concentrations increased with the increasing percentage of town area, indicating a significant influence of the land use structure on heavy metal concentrations in peri-urban soils. The correlation analysis of heavy metal concentrations and environmental factors demonstrated that human activities, including fertilisation, are important factors affecting heavy metal concentrations. The urbanisation process should optimise land use patterns to sustain and enhance soil security in peri-urban areas.