Effects of vegetation restoration on distribution characteristics of heavy metals in soil in Karst plateau area of Guizhou

Regional profile of study

The study area is located in the Caohai National Nature Reserve in Weining, Guizhou (26°49: 26°53 N, 104°12: 104°18°E) (Fig. 1). The total area of the Caohai National Nature Reserve is about 120 km², the average elevation is about 2,171.7 m, and the drop between the lake and the surrounding mountains is 100–120 m. It belongs to the subtropical monsoon area with an annual temperature of 10.6 °C and an average annual rainfall of about 1,000 mm. The soil types are mainly yellow soil and limestone soil, and the rock exposure rate is more than 75%. It has well-developed soil surface, stone pits and other niches, which is a typical karst area. The unique geological structure and soil-forming materials of carbonate rocks in karst area are lacking, resulting in shallow and discontinuous soil layer, and soil erosion leads to soil degradation easily (He et al., 2019).

In the past two decades, the two-way evolution of land use types from nature to cultivated land has seriously led to rocky desertification, further increase of soil erosion, aggravation of soil physical clay change and decrease of vegetation coverage and so on. Among them, the sources of heavy metals in the Caohai area are mainly geological weathering and soil forming process and man-made factors (Yang, Xin & Tian, 2022; Xu et al., 2015). Local measures such as “vegetation restoration, returning farmland to forest and grassland” have been taken to control rocky desertification areas in the Caohai area and good progress has been made. In 2016, the “Vegetation Restoration Project in the Rain gathering area on the North Bank of Caohai” was implemented. The project aims at the ecological restoration project of serious rocky desertification, low vegetation coverage and serious soil erosion on the north slope of Caohai.

Sample collection

This study is based on the “Mian Shan vegetation Restoration Project on the North Bank of Caohai”. According to the “National soil Environmental quality Construction Land soil pollution risk Control Standard (trial) (GB36600-2018)” and “National soil Environmental quality Agricultural Land soil pollution risk Control Standard (trial) (GB 15618-2018)”, and combined with the current situation of field investigation of soil heavy metals in Caohai area. Finally, five kinds of heavy metals including chromium, nickel, arsenic, zinc and lead (Cr, Ni, As, Zn and Pb) were selected for quantitative analysis. It was noted that the contents of cadmium (Cd) and mercury (Hg) in this study area too low to detect, and their distribution was not analyzed. Three geomorphological belts including one non-karst belt (Yangguan Mountain in Caohai region), one karst rocky desertification zone (Jiangjiawan) and one vegetation restoration belt (Jiangjiawan) were selected. The two sample zones in karst area have the same habitat, and half of area has vegetation restoration and half without vegetation restoration for comparative study. Karst and vegetation restoration areas are connected by wetland, therefore they owing the same data of the shore and the wetland. Before restoration (karst), the vegetation communities were mainly dwarf shrubs and herbs, and the distribution was relatively uniform. The shrub communities including Cotoneasterhissaricus and Rberiswilsona, and the herb layer communities were mainly Hemisiacarvifol and Perusdulouxi (Wu et al., 2013; Zhang et al., 2013). In particular, the duration time of vegetation restoration transect in vegetation restoration belt is 3–4 years, and mainly artificial vegetation (Yunnan poplar and herbs). The geographical location of the transect and the present situation of vegetation are shown in Table 1. The soil type in Jiangjiawan area is lime soil.

From top to bottom, five sampling points mountaintop (MT), hillside (MO), foothills (MF), shore (SH) and wetland (WL) in each sample belt. A total of 180 samples were collected from 13 sample points through “five-point method”, and the depth of the sampling profile was 0–50 cm with a 10 cm depth interval (i.e., 0, 10, 20, 30, 40 and 50 cm). Subsequently, soil samples collected in the sample site were fully mixed after removing impurities, and the final soil samples were obtained. The soil samples were divided into two parts and sealed in self-sealed bags and marked and stored at low temperature. One soil sample is used for the determination of heavy metals in soil, and the other is used to determine carbon, nitrogen, phosphorus, pH, moisture content, bulk density, electrical conductivity and other indexes.

Pretreatment and determination of samples

Soil organic carbon (SOC), total nitrogen (TN), and total phosphorus (TP) were measured as described previously (Walkley & Black, 1934). Nitrate nitrogen (NO₃⁻-N), ammonium nitrogen (NH₄⁺-N), soil organic carbon (SOC), readily oxidizable carbon (ROC), dissolved organic carbon (DOC), pH, soil bulk density (BD), electrical conductivity (EC) and soil moisture (SWC) were measured as described previously (Wu et al., 2022a), Chemicals, including sulfuric acid, hydrochloric acid, sodium hydroxide, were purchased from Sinopharm Chemical Reagent Co., Ltd with reagent grade or higher purity.

Concentrations of Cr, Ni, As, Zn, Pb, iron (Fe), manganese (Mn) and calcium (Ca) were determined by DELTA Professional-Handheld XRF analyzers (Olympus Corporation, Shinjuku City, Tokyo, Japan). Soil aggregates were determined by wet sieve method, and the classified through the National Forestry and Grassland Administration of China (1987). The physical and chemical properties of the soil are listed in Table 2.

Data analysis and statistics

The data were analyzed by Excel 2010 and SPSS 26. Redundancy analysis (RDA) of the heavy metals content and physicochemical properties of the soil was performed by Canoco software (version 5.0; http://www.canoco5.com/). All graphics were plotted in Origin 9.1 (https://www.originlab.com/index.aspx?go=Support&PID=2131).

Distribution characteristics of soil aggregates under different landforms

As shown in Fig. 2, soil aggregates shows different distribution characteristics in three landforms. In non-karst geomorphology, the proportion of soil aggregates larger than 2 mm increased gradually from MT to MF and gradually occupied the dominant position, while the proportion of 0.25–2 mm soil aggregates decreased slowly. In karst landform without vegetation restoration, although the proportion of soil aggregates larger than 2 mm decreases from MT to MF, and the proportion of 0.25–2 mm soil aggregates shows the opposite trend with non-karst geomorphology, indicating that soil aggregates and microaggregates are easy to be lost under karst landforms. In addition, more than 50% of soil aggregates in SH and WL sample sites are mainly larger than 2 mm soil aggregates, accounting for more than 50%, and the proportion of soil aggregates of each particle size changes little with the increase of depth.

However, the distribution of soil aggregates after vegetation restoration is different from non-karst geomorphology and karst landform without vegetation restoration. The proportion of surface soil 0.25–2 mm aggregates in vegetation restoration belt increased to more than 40% especially larger than 2 mm (13–17%). The proportion of surface soil ≤2 mm aggregates was also increased to 83% and 88%, respectively. The distribution proportion of soil aggregates in each sample site of vegetation restoration was similar and the change of depth is not obvious. The stable proportion of surface soil 0.25–5 mm aggregates accounted for more than 90%. This result indicated that vegetation restoration significantly improved the soil structure. Previous studies has shown the similar phenomenon, that is, the content of organic carbon in soil aggregates is higher in the range of 5–0.25 mm particle size, and the increase of organic carbon content is helpful to the formation of soil water-stable aggregates as well as change the soil structure and properties (Xu et al., 2017; Cheng et al., 2013). In addition, the changed soil structure could also attributed to the increased vegetation coverage, which increased plant roots and further improved the ability to fix soil.

Furthermore, this study showed that the content of soil organic carbon is not always in a steady upward trend in the early stage of vegetation restoration, and the frequent export and input of soil organic carbon leads to great fluctuation (Table 2), which is consistent with the results of previous studies (Xie et al., 2015; Zhang et al., 2020b). The increase of soil organic matter content is beneficial to the oxidizable transformation and enrichment of heavy metals (Zhang & Han, 2022).

Distribution characteristics of soil heavy metal contents in non-karst and karst landforms

The distribution of heavy metals (Cr, Ni, Zn, As and Pb) in non-karst soil profile (0–50 cm) is shown in Fig. 3. The contents of Cr, As and Pb in the soil of SH and WL increased with the increase of soil depth, while the contents of Ni and Zn in the surface and bottom layer were higher than those in the middle layer, showing a “V” type. Except for As, other heavy metals in WL were enriched in the surface layer, and the contents of Zn and Pb were significantly different from those in other soil profiles (p < 0.05). As a whole, the content of heavy metals in soil of MO was higher than that of MT and MF as a whole, and the content of heavy metals increased gradually along the soil profile, in which Cr showed significant difference in profile (p < 0.05). The content of heavy metals in MF soil showed a tipping point in 20–30 cm profile in MF was the lowest, and it decreased in 0–20 cm while increased in 30–50 cm profile. Each profile showed significant difference (p < 0.05). There is no clear trend for the content of heavy metals in each sample site and profile did not follow the obvious rule.

The profile distribution of heavy metals in karst soil without vegetation restoration is shown in Fig. 4. Four kinds of heavy metals (except As) in the sample belt show a decreasing trend from MT to WL. MT, MO and MF have different degrees of enrichment of heavy metals in the surface layer of soil, and the content of heavy metals reaches the lowest in 10–20 or 20–30 cm layer, and then increases, showing different degrees of “V” type. The content of heavy metal as increased with the increase of soil depth at MT and MO, but there was no significant difference. The content of five heavy metals at SH was higher than that at WL. There was no significant difference in the content of heavy metals in soil profiles among different sample sites (p > 0.05), and there was also no significant difference in numerical range.

Distribution characteristics of soil heavy metal content in vegetation restoration sample zone

The distribution of heavy metals in soil profile of vegetation restoration was shown in Fig. 5. Heavy metals Cr, Ni and Pb gradually decreased gradually from MT to MF. The lowest value reached at 10–20 or 20–30 cm in soil profile, and accumulated in surface and deep soil (20–50 cm). In sample sites MT, MO and MF, the distribution trend of heavy metals in soil profile was basically the same, and showed different degrees of surface enrichment. The soil heavy metals in each profile were the same as those in the karst zone without vegetation restoration, and there was no significant difference (p > 0.05).

Correlation analysis between heavy metal content and physicochemical factors in topsoil

Principal coordinate analysis (PCA) of soil physical and chemical properties in different geomorphological landforms

The principal component analysis (PCA) of the soil physical and chemical properties under three sample bands is shown in Fig. 6. In every zone, soil properties were mainly controlled by two principal coordinate (PC) components. There was a cumulative variance of 97.36% among them with PC1 explaining 95.06% and PC2 explaining 2.30%. The overlap area of vegetation restoration landform and karst landform was large, which showed that the soil physical and chemical properties of the two were similar, but the overlap area of non-karst landform was smaller, indicating that there was a great difference between non-karst landform soil physical and chemical properties and the two.

Influence of distribution characteristics of heavy metals in karst geomorphology soil

The content of heavy metals in soil is closely related to soil parent material, soil physical and chemical properties and microorganisms, and is also affected by human activities (Li et al., 2022). Based on the results obtained in Figs. 3 and 4, it was appeared that the content and distribution of heavy metals in the soil profile between the non-karst zone and the karst zone are markedly different. In non-karst zone, heavy metals enriched in the surface soil and then decreased in 0–10 cm, while increased in the range of 10–50 cm. In contrast, only Ni and Pb showed this tendency in the karst zone. These phenomena can be attributed to the following reasons. On the other hand, the rich organic complexes in Arbor forest soil have strong adsorption and chelation, thus inhibiting the migration of heavy metals (Zhang et al., 2019). Moreover, the forms of heavy metals are mainly composed of stable residues, which can be stable in sediments for a long time. Therefore, the heavy metals remain in the middle and upper layers of the soil, and finally enriched in the surface layer (Jitendra, Rachna & Naveen, 2017). On the other hand, a large number of microorganisms could grow in the forest ecosystem with the developed roots and abundant litter, and heavy metals may be adsorbed on extracellular polymers or cell wall components produced by microorganisms, which can regulate the concentration of heavy metals in the soil (Ma et al., 2014; Zhao et al., 2015).

In addition, the contents of Cr, Ni, Zn, As and Pb in Caohai area detected in present study were lower than or close to the background contents of soil heavy metals in Guizhou Province (Wang et al., 2021). Some studies have reported that when the lower levels of heavy metals in soil had no or little effect on the normal metabolic activities of plants, plants could increase the number of rhizosphere soil microorganisms and enzyme activities, secrete H⁺ to reduce rhizosphere soil pH value (Zeng et al., 2008), increase the bioavailability of mineral nutrients and heavy metals in the root zone by regulated root secretions, and thus promote the absorption and utilization of heavy metals by plants (Gai et al., 2020). At the same time, the migration ability of heavy metals in the acidic non-karst soil environment (pH < 5) will also enhance under the action of leaching, resulting in heavy metals were precipitated in bottom and enrichment (Tang et al., 2021).

In summary, the common distribution characteristic of heavy metals in non-karst soils is surface and bottom precipitation enrichment, while that of karst soils are affected by surface vegetation, root exudates, microorganisms and leaching, and leading to the different distribution characteristic.

Effect of soil heavy metal profile distribution characteristics in vegetation restoration sample zone

Compared with karst soil, the contents of heavy metals Cr, Ni and Pb in vegetation restoration sample zone were decreased in a lesser extent. This may be ascribed to heavy metals in the soil were absorbed by plants. In previous study (Qi et al., 2022), it was found that the dominant herbaceous plants in native Compositae and Gramineae had strong enrichment and transport ability to heavy metals such as Cd, Cr and Ni. Moreover, the contents of heavy metals in the surface layer of soil were higher than those in 10–20 cm layer. The vegetation in the vegetation restoration sample zone is mainly covered by dwarf trees, shrubs and herbaceous vegetation communities, and the herbaceous communities take Chrysanthemum morifolium and Artemisia annua as the dominant species in the vegetation restoration area, and plants absorb heavy metals in the soil. Reduce the mass fraction of heavy metals (Li et al., 2005), which is also an important factor that the heavy metals in the surface soil are higher than those in the middle layer.

In this study, it was found that ROC, DOC and inorganic nitrogen in vegetation remediation soil were significantly higher than those in karst soil. The research of Xie et al. have clarified the vegetation restoration could not only strengthen the water retention ability of soil, but also change the dynamic balance of various elements, which reducing soil organic carbon mineralization and soil erosion to a certain extent. Therefore, soil carbon sequestration capacity gradually enhanced, and then the carbon content in vegetation remediation soil were significantly increased. Except the change of carbon content in soil, pH was also decreased (0.2–0.4). This phenomenon maybe caused by the H⁺ secreted from plants, and the root exudates produced by plant further promoting the growth of soil microorganisms. The reduced pH demonstrated that vegetation restoration can enhance soil microbial activity significantly.

In addition, heavy metal content is enriched on the soil surface and deep layers, which will be attributed to the following points. Firstly, there is competition between plant adsorption and microbial biosorption for heavy metals (Ma et al., 2014; Zhao et al., 2015; Ma et al., 2010). Some research have proved that soil pH and DOC are important factors affecting the bioavailability of heavy metals (Sizmur & Hodson, 2009), which can transform residual heavy metals into highly mobile forms in soil, and further increase bioavailability (Zhang et al., 2019; Xie, Ren & Li, 2007, Zhang et al., 2020a). The absorption process of heavy metals by plants would be affected by root exudates, which have the effect on the pH value, occurrence and solubility of heavy metals in rhizosphere soil (Wang et al., 2022a). For example, organic acids and metal complexes play an important role in the mechanism of heavy metal absorption in roots by increasing the concentration of free metal ions (Lin et al., 2022). Secondly, although vegetation restoration increases vegetation coverage and soil water retention capacity, it is still in the early stage of vegetation restoration. Study has reported that in many kinds of heavy metals are still in residual form different degrees of rocky desertification soil, (Xie, Ren & Li, 2007), so the content of heavy metals in soil is further leached by surface runoff. Therefore, the content of heavy metals shows a deep aggregation (soil depth ≥ 30 cm).