Effects of vegetation restoration on soil nutrients, plant diversity, and its spatiotemporal heterogeneity in a desert–oasis ecotone
Funding information: the Central Non-profit Research Institution of CAF(CAFBB2017ZB004), Grant/Award Number: CAFBB2017ZB004
Abstract
Vegetation restoration has been proposed as an effective measure for rehabilitating degraded land and slowing desertification in arid regions. However, the spatial variation in soil quality and plant diversity following vegetation restoration remains unclear. This study was designed to explore soil nutrient dynamics and how soil nutrients affect plant diversity and spatial heterogeneity after shrub restoration. We assessed the effect of Haloxylon ammodendron (C.A.Mey.) Bunge (which has been planted over 30 years) on the soil nutrients and plant diversity in a desert–oasis ecotone in Minqin County, Gansu, China, using geostatistics, beta diversity and rarefaction analyses, and Hill number extrapolation. Soil nutrients, including soil organic matter, total nitrogen, and alkali nitrogen, increased significantly after H. ammodendron planting. Species richness gradually increased from 1–5 years to 10–20 years after H. ammodendron was planted but then decreased at 20–30 years. The largest differences in plant composition were observed at 15 and 20 years. Plant diversity increased in the whole 30 years after shrub planting, increasing in the first 25 years and then decreasing at 26–30 year stage. The maximum coefficient of determination for the spatial heterogeneity model fit was 0.84 (25 years). The spatial heterogeneity in vegetation decreased with increasing soil available K content at 1–10 years. Our results suggest that planting shrubs can improve soil conditions and plant species diversity in desert–oasis ecotones and soil nutrients have a strong influence on plant diversity patterns and spatial heterogeneity following vegetation restoration.
1 INTRODUCTION
Ecosystem restoration plays an important role in the habitat management of degraded ecosystems in semiarid and arid regions (Parra, Zornoza, Conesa, Faz, & Gómez-López, 2017) and is the first step in biome reconstruction (Haverkamp et al., 2017). It is necessary to ensure the maintenance of biological diversity and forest ecosystem functions and sustainable economic development. Human-related activities such as excessive reclamation and overgrazing cause the loss of vegetation cover, leading to land degradation in many arid and semiarid regions worldwide (Tietjen et al., 2017). The land of such extremely degraded ecosystems is extremely barren and has severe soil erosion, which exacerbates habitat degradation (Deng, Wang, Tang, & Shangguan, 2016). Pioneer species plantations, which are suitable for the environment during early periods, play an important role in the habitat recovery of degraded ecosystems (Zhao et al., 2017). Vegetation restoration is an effective measure to restore degraded ecosystems, enrich soil nutrients, increase species diversity and plant spatial heterogeneity, and accelerate succession (Deng et al., 2016; Li, Yang, Shi, Li, & Guo, 2018; Manhães et al., 2018; Tietjen et al., 2017; Zhao et al., 2017). Vegetation restoration can slow desertification in arid regions, help regulate community composition, and change plant emergence, survival, growth, and density (Keesstra et al., 2018). Shrubs may be degraded and grasses become a dominant plant owing to reduced soil moisture (Chen, Yin, Gebhardt, & Yang, 2018). However, the restoration of degraded habitats is becoming increasingly urgent and difficult with increasing human impacts on global ecosystems (Deng, Han, Zhang, Tang, & Shangguan, 2017; Deng et al., 2017; Mao et al., 2018).
Previous studies have shown that afforestation not only improves the soil environment (Zhao et al., 2017), including soil nutrients, moisture, and structure (Deng, Kim, Peng, & Shangguan, 2018; Xiao et al., 2017; Zhang, An, Cai, Guo, & Xiao, 2017; Zhu, Deng, Zhang, & Shangguan, 2016) of plants, but also significantly influences plant diversity (Abella, 2010; Díaz-García, Pineda, López-Barrera, & Moreno, 2017; Laughlin et al., 2017; Xiao et al., 2017) and ecosystem productivity (Uthappa et al., 2018). Many studies have also reported that soil nutrients such as nitrogen and phosphorus are important factors affecting vegetation coverage, plant community composition, and plant diversity (Chen et al., 2017; Liu, Xing, & Yang, 2017; Sanaei, Ali, & Chahouki, 2018; Seabloom et al., 2017; Song, Liu, Wu, Gao, & Hao, 2018; Wei et al., 2017; Zhu, Tang, Shangguan, & Deng, 2018). Competition for soil resources, especially nitrogen, leads to spatial heterogeneity in communities and there is a relationship between competition and coexistence within plant communities (Tilman, 1982, 1985). The effect of plant functional components on soil carbon and nitrogen accumulation is obvious (Fornara & Tilman, 2008). Soil fertility increases with an increase in plant species diversity (Dybzinski, Fargione, Zak, Fornara, & Tilman, 2008). Nitrogen and phosphorus fertilizers that have been applied to soil have significant effects on the survival, regeneration, succession, richness, and diversity of dominant tree species in tropical dryland forests (Ceccon, Huante, & Campo, 2003; Ceccon, Sánchez, & Campo, 2004).
Afforestation has a significant effect on ecosystem restoration in arid and semiarid regions (Lu, Zhao, Shi, & Cao, 2018; Tateno et al., 2017), as well as on soil nutrients, plant diversity, and spatial heterogeneity in soil properties (Mcnicol et al., 2017) and plant diversity. Spatial heterogeneity in plant diversity describes the degree of aggregation and spatial distribution of the species in a community concerning ecological processes (Levine, Bascompte, Adler, & Allesina, 2017). Soil nutrients, plant diversity patterns, and spatial heterogeneity in plant communities are key parameters that reflect the structure and function of ecosystems (Wiegand et al., 2017). Many studies have determined that spatial heterogeneity in plant communities is related to soil nutrients (Schweiger et al., 2017) and plant diversity is most strongly related to plant-induced heterogeneity in ecosystems (Díaz-García et al., 2017).
Desert areas are broadly distributed in arid and semiarid regions of northwest China (Sun, Chai, Liu, & Xue, 2017). Survival-adapted shrub afforestation is an effective measure for restoring desertified land throughout the region (Sandercock & Hooke, 2017). Shrubs are an extremely important component of drylands and desert ecosystems (Paz-Kagan, Zaady, Shachak, & Karnieli, 2016). Their underground root systems are grown in deep soil and have strong drought resistance and sand fixation, windbreak, and soil and water conservation capacities (Wang et al., 2015; Wang, Zhao, Liu, Zhang, & Li, 2015).
Haloxylon ammodendron is a precious plant resource in the desert area of northwest China (Ma et al., 2010). Previous studies of H. ammodendron have focused on the effects of spatial heterogeneity in soil nutrients, soil water, and other factors. Many studies have been conducted on the relationship between plant species and soil nutrients (Sanaei et al., 2018). However, previous studies have examined the effect of afforestation on plants in arid and semiarid regions at only a single time point (Tateno et al., 2017). Therefore, determining how soil nutrients affect plant diversity and their spatial heterogeneity after afforestation for the long-term afforestation sequence of H. ammodendron planting is important for the restoration and management of ecosystems in arid and semiarid regions. It is also necessary for vegetation restoration and biodiversity conservation. The objectives of the present study were to (a) identify the function of H. ammodendron that has been introduced into desert and semiarid areas to determine whether they increase the soil nitrogen over a 30-year chronosequence; (b) clarify the characteristics of plant species composition, diversity patterns, and spatial heterogeneity during the growth process of shrubs at different ages; and (c) describe the relationships among soil nutrients, species diversity patterns, and spatial heterogeneity in vegetation.
2 MATERIALS AND METHODS
2.1 Study area
Minqin County (101° 49′ 41″–104° 12′ 10″ E, 38° 3′ 45″–39° 27′ 37″ N) in Gansu Province was chosen as the study site (Figure 1). It is located between the Tengger and Badanjilin Deserts. A large area of H. ammodendron has been planted in the peripheral areas of the Minqin Oasis and on the edges of the Tengger and Badain Jaran Deserts. This area has a temperate desert climate. The multiyear average temperature and precipitation are 7.6°C and 116 mm, respectively (Ma et al., 2010). The main types of sites are mobile dunes, semifixed dunes, fixed dunes, and flat land between hills, wind-eroded residual hillsides, and piedmont alluvial fans. There are many shuttles on fixed and semifixed dunes. In terms of vegetation types, Nitraria tangutorum is distributed in one place outside the Minqin Lake basin oasis, which mainly includes Nitraria shrub stands, other fixed and semifixed dunes, and clay-dominated lowlands. The natural dominant plant is Nitraria. Spinosa, with a small number of Tamarix laxa, Artemisia arenaria, and Calligonum mongolicum, and some H. ammodendron and Hedysarum scoparium that were planted in the 1960s. The soil is characterized by a rough texture, small stones, thin consistency, and poor water retention capacity. The distribution pattern of the H. ammodendron population at the edge of the desert oasis in Minqin County is generally aggregated (Ma et al., 2010; Zhang, Liu, Zhang, & Sun, 2016). H. ammodendron forms large deciduous shrubs or small trees and is a Second-Class National Protected Plant (Yu et al., 2012). H. ammodendron is distributed in Central Asia, western Inner Mongolia, Gansu, and the Qinghai and Xinjiang Desert areas in China (Ma et al., 2010; Xu, Li, & Xu, 2011; Yu et al., 2012).
2.2 Experimental design and sampling
In the present study, a chronosequence was examined to investigate the vegetation and soil dynamics of different aged H. ammodendron plantations in the study area (Li, Gao, & Tang, 2016). A series of H. ammodendron plantations established 5, 10, 15, 20, 25, or 30 years ago with similar soil and climatic conditions, land use histories, and historical record databases were selected as research samples, providing a time-series of vegetation restoration. A mobile dune without H. ammodendron was selected as the reference 'control' condition (0 years). We established sample plots on land that had relatively homogeneous surface coverage. A total of 65 typical sites (300 m × 300 m), 10 for each year and five controls, of H. ammodendron were established. Three shrub plots (20 × 20 m), separated by at least 50 m, were randomly selected at each site.
Three herb quadrats (1 m × 1 m) separated by at least 5 m along the diagonal of each shrub plot were randomly selected. A total of 90 (10 × 3 × 3) quadrats corresponding to each year were obtained from each site. The control included 45 quadrats. Basic information for all the shrubs and herbs was obtained by measuring their height, basal diameter, canopy density (individual area−1), and vegetation coverage. To obtain the aboveground plant biomass, four H. ammodendron (the largest and smallest and two moderately sized individuals) were cut down in each shrub plot, the fresh weight of each tree was obtained, and all the samples were placed in an oven at 80°C for 48 hr. The water content and dry weight of the sample biomass were then calculated (Ye, Fang, Shi, Deng, & Tan, 2019).
Soil samples were collected from under shrubs and inter shrub areas in each selected shrub plot with three different ages (5, 20, and 30 years). Soil samples at two depths (0–10 and 10–20 cm) were randomly collected from three random points using a stainless-steel auger at each plot. Each soil sample was modified to a weight of 200 g and transported to the laboratory before the removal of residual material. The samples were stored under dry, well-ventilated conditions to allow them to dry naturally before they were sieved through a 200-μm mesh and then they were sent to the laboratory for physical and chemical analyses (Pingree & DeLuca, 2018). According to previous studies (Olego, Visconti, Quiroga, Paz, & Garzón-Jimeno, 2016), the soil nutrient grading standards for China's second national soil survey were used as the evaluation criteria (Table S1). Soil organic matter (SOM), total N (TN), alkali-hydrolyzable N (AN), available P (AP), and rapidly available K (AK) were used as evaluation indicators. To obtain the five soil nutrient indicators, the soil samples were chemically analyzed. We measured the SOM using the K dichromate volumetric 'heating' method, the TN using the semi-micro-Kjeldahl method, and the AN content using the Kang Hui dish method. Additionally, we used the 0.5 mol L−1 sodium bicarbonate extraction molybdenum-antimony resistance colorimetric method to determine the AP and ammonium acetate atomic absorption spectrometry to determine the AK content. The gravimetric analysis was used to determine the soil water content (Pingree & DeLuca, 2018).
2.3 Data calculation
2.3.1 Beta diversity analysis
Beta diversity measures change in species composition between locations or communities (Baselg & Orme, 2012). Beta diversity analysis was performed using only herb data in the present study (Figure S1). Baselga (2012) developed a unified framework for assessing beta diversity that applied the Sørensen and Jaccard indices and their turnover and nesting components to calculate the total differences (Sun et al., 2017). We used the Sørensen index as follows:
((1))
((2))
((3))2.3.2 Rarefaction and extrapolation curves with Hill numbers
Rarefaction and Hill number extrapolation are used for sampling and estimation in species diversity studies (Chao, Chiu, & Jost, 2014). These criteria quantify and assess changes in the biodiversity, allowing for the differential weighting of rare and abundant species, which is similar to a diversity index but with a more easily understood meaning. Chao et al. (2014) applied a uniform approach to sample- and individual-based data to estimate the first three Hill numbers to characterize the diversity of a species assemblage: species richness (q = 0), the Shannon entropy index (Shannon diversity, q = 1), and the anti-Simpson concentration (Simpson diversity, q = 2). These proposed estimators are accurate for both sparse and short-range extrapolations (Sun et al., 2017).
Sparse and extrapolated curves and Hill numbers were used to compare the plant diversity patterns (Sun et al., 2017). We estimated plant diversity (species richness and Shannon and Simpson diversities) as the mean of 200 replicates with a 95% confidence interval (Chao, Chiu, & Jost, 2014).
2.3.3 Spatial pattern analysis
((4))where γ(h) is the variation function, Z(x) is the value of property Z in spatial position x, Z(x + h) is the value of Z at (x + h), and E[Z(x) − Z(x + h)]2 is the mathematical expectation of the sample value variance for distance h.
((5))where Z(xi) and Z(xi + h) are the numbers of plant species at point xi and point (xi + h), respectively; N(h) is the log of the sample number; and h is the spatial distance.
The main parameters of the function were nugget values, partial sills, ranges, and fractal dimension or spatial structure ratios, which were used to analyze the spatial patterns of the plant species. Under the second-order stationary hypothesis or intrinsic hypothesis, γ(h) is the mean variance of the measured values of points with all spatial distances (lag distances and step lengths) as H, that is, the correlation or average difference between the two regionalized variables Z(xi) and Z(xi + h). With distance identification, H can reflect the spatial variation characteristics of regionalized variables, especially via stochastic responses to the structural nature of the regionalized variables. Before fitting, the semivariogram replaces the original data logarithmically to form a normal distribution and the model is fitted under isotropic conditions. The basic principles and theoretical explanations of the semivariance function can be found in the reference material. The spatial patterns of the plants were analyzed using an age chronosequence (Table S3) and an exponential model was chosen for all the sites for comparison with other models in the present study.
2.3.4 Relative importance value
2.3.5 Data analysis
One-way analysis of variance was used to examine the soil nutrient characteristics in the different aged shrub plots at each site. Linear regression analysis was used to study the relationships between the shrub age and response variables, including the density, coverage, and aboveground biomass, of the shrubs. All statistical analyses were performed using R version 3.5.1 (R Development Core Team, 2017) and GS+ 7.0. Beta diversity was analyzed using the Betapart package, and the sparse and extrapolated curves were compiled using the iNEXT package. The numbers were calculated, the figures were drawn, and the data were processed using the Ggplot2 package (Hadley, 2007; Sun et al., 2017).
3 RESULTS
3.1 Species composition of vegetation
We identified 20 species of vegetation belonging to 16 genera from 10 families in the 65 H. ammodendron plots (Table S2). The species richness gradually increased from 1–5 years to 10–20 years but then decreased at 20–30 years after H. ammodendron planting. The RIVs of the vegetation species showed that the dominant species were Phragmites communis, Scirpus wallichii, Limonium aureum, Agriophyllum squarrosum, Nanophyton erinaceum, and Caragana korshinskii in years 1–30 (Table 1).
| Period | Phragmites communis | Scirpus wallichii | Limonium aureum | Agriophyllum squarrosum | Nanophyton erinaceum | Caragana korshinskii |
|---|---|---|---|---|---|---|
| 1–5 year | 11.14 | 15.14 | 10.78 | 14.75 | 13.08 | 17.12 |
| 5–10 year | 12.16 | 11.85 | 12.54 | 16.24 | 11.07 | 14.15 |
| 10–15 year | 10.74 | 10.41 | 9.66 | 11.05 | 10.06 | 12.11 |
| 15–20 year | 10.11 | 6.11 | 9.15 | 16.12 | 6.38 | 5.75 |
| 20–25 year | 8.15 | 7.15 | 5.61 | 5.08 | 6.05 | 6.94 |
| 25–30 year | 8.76 | 7.94 | 7.09 | 5.26 | 5.12 | 9.76 |
The total and turnover differences had similar change trends during the 30-year growth period: the largest differences were observed at 15 and 20 years and the smallest differences were observed at 5 and 10 years. The nesting difference was different among the six ages, being the largest in the 20-year-old plants and the smallest in the 5- and 25-year-old plants.
3.2 Diversity pattern of vegetation
The results showed that the sample size (i.e. the number of individual plants) at 5, 10, 15, 20, 25, and 30 years was 985, 992, 890, 1,006, 1,027, and 1,014, respectively, with Hill numbers = 0, 1, and 2 (Figure 2). The sample size had a base coverage rate of 1.0; therefore, it was close to the completion of the recovery phase. The reference sample number was extrapolated to 1913. The corresponding observed species richness (q = 0), Shannon diversity (q = 1), and Simpson diversity (q = 2) showed consistent trends, that is, richness gradually increased from 5 to 10 years but then remained stable with diversity reaching a peak at 25 years. The diversity value was the highest at 15 and 25 years and then decreased during the final stage (30 years).
3.3 Spatial heterogeneity in plants
The maximum coefficient of determination of model fitting was 0.84 (25 years). The plant abutment (C0 + C) at 10 years was high (15.204), and the spatial heterogeneity degree of the plants between 5 and 10 years was greater than that of the plants at the other ages. Additionally, the range in variation at the age of 10 years was the smallest (21.6); therefore, the spatial autocorrelation scale at this age was smaller than that at the other, older ages (Table 2, Figure S2).
| Time after revegetation (years) | Theoretical model | Nugget (C0) | Sill (C0 + C) | Spatial variability [C/(C0 + C)] | Range (A/m) | (R2) | (RSS) × 10−3 |
|---|---|---|---|---|---|---|---|
| 5 | Exponential | 1.704 | 8.985 | 0.810 | 28.3 | 0.82 | 4.193 |
| 10 | Exponential | 2.297 | 15.204 | 0.728 | 21.6 | 0.91 | 4.671 |
| 15 | Exponential | 2.521 | 15.638 | 0.839 | 30.7 | 0.64 | 1.578 |
| 20 | Exponential | 2.506 | 14.824 | 0.831 | 29.8 | 0.72 | 2.865 |
| 25 | Exponential | 0.913 | 5.094 | 0.821 | 29.5 | 0.73 | 2.976 |
| 30 | Exponential | 3.54 | 13.02 | 0.849 | 31.2 | 0.58 | 3.642 |
3.4 Characteristics of the soil nutrients
The results showed that soil depth (two levels), plant age (years), and the interaction between soil depth and plant age (years) had a significant effect on soil nutrients (SOM, TN, AN, AP, and AK) (p < .05) (Table 3). The concentrations of SOM, TN, AN, AP, and AK increased significantly over time. The SOM concentration (15.6 g kg−1) increased 5.6-times at a depth of 0–10 cm and 1.9-times at a depth of 10–20 cm under the shrub canopy (Figure 3a,b) after 30 years of vegetation restoration. The concentration of TN increased 5.8-times in the 0–10-cm soil and 1.6-times in the 10–20-cm soil, accumulating mainly under the shrub canopy and following the same pattern as that of the SOM concentration (Figure 3c,d). The AN concentration showed the same increasing trend in the canopy and intershrub areas at 0–10 cm and 10–20 cm (Figure 3e,f). The AP concentration increased (38.5 times) under the shrub canopy, with the same trend as that of the TN concentration at the 0–10-cm depth (Figure 3g,h). The increasing trend of the AK concentration was similar to that of the AP concentration at the 0–10-cm soil depth; however, it was not evident at a depth of 10–20 cm (Figure 3i,j). Therefore, the number of soil nutrients at a soil depth of 0–10 cm was greater than that at a depth of 10–20 cm and the content of the soil nutrients under the shrub canopy was higher than that in the intershrub area (Figure 3).
| Property | p | |||
|---|---|---|---|---|
| T | D | T × D (under shrub) | T × D (intershrub) | |
| SOM | <.01 | <.05 | .017 | .025 |
| TN | <.05 | <.05 | .023 | .032 |
| AN | <.01 | <.05 | .025 | .016 |
| AP | <.05 | <.05 | .013 | .015 |
| AK | <.01 | <.05 | .011 | .037 |
- Note: Soil organic matter (SOM, g kg−1), total nitrogen (TN, g kg−1), alkali nitrogen (AN, mg kg−1), available phosphorus (AP, mg kg−1), and available potassium (AK, mg kg−1) at a soil depth (D) of 0–10 cm and 10–20 cm with seventime-series (T).
3.5 Relationship between spatial heterogeneity in plants and the soil nutrient variables
The results showed that AK, AN, and SOM had a significant effect on vegetation distribution at 1–10 years and TN and AK had a significant effect on vegetation distribution at 11–20 years, whereas AP and TN had significant effects on vegetation distribution at 21–30 years (Table 4). SOM and AP were positively correlated with plant distribution, and TN, AN, and AK were negatively correlated at 1–10 years. The absolute value of the AK correlation coefficient was greater than 0.5; therefore, there was more AK content in the soil at 1–10 years and the heterogeneity in the spatial distribution of vegetation decreased. In the 10–30-year period, SOM, TN, AN, AP, and AK were positively correlated with the plant distribution; therefore, the increased content of SOM, TN, AN, AP, and AK in the soil resulted in less heterogeneity in the spatial distribution of the vegetation. The spatial distribution of vegetation and AN exhibited the strongest correlations, followed by SOM, and the correlation with TN was the weakest during the 10–30-year period. The soil nutrient variables had strong explanatory power for the 1–10-year vegetation spatial distribution and weaker explanatory power for the 20–30-year vegetation spatial distribution.
| Soil nutrient variables | SOM | TN | AN | AP | AK | ||
|---|---|---|---|---|---|---|---|
| Time after revegetation (years) | 5 | Correlation coefficients | 0.1467 | −0.1971 | −0.3675 | 0.0129 | −0.5370 |
| p | .024 | .114 | .015 | .463 | .001 | ||
| Explains% | 5.1 | 7.3 | 11.6 | 3.3 | 20.4 | ||
| 10 | Correlation coefficients | 0.2103 | −0.4855 | −1.9121 | 0.6887 | −3.1451 | |
| p | .015 | .406 | .023 | .286 | .011 | ||
| Explains% | 1.3 | 7.3 | 2.9 | 2.4 | 0.63 | ||
| 15 | Correlation coefficients | 0.1674 | 0.0798 | 0.2218 | 0.1799 | 0.0851 | |
| p | .051 | .004 | .152 | .071 | .028 | ||
| Explains% | 5.6 | 3.7 | 2.81 | 4.3 | 12.5 | ||
| 20 | Correlation coefficients | 0.1952 | 0.0032 | 0.1819 | 0.0424 | 0.0112 | |
| p | .589 | .006 | .101 | .069 | .031 | ||
| Explains% | 1.7 | 12.7 | 4.3 | 1.4 | 2.3 | ||
| 25 | Correlation coefficients | 0.1703 | 0.0182 | 0.2365 | 0.0352 | 0.0743 | |
| p | .467 | .031 | .189 | .024 | .612 | ||
| Explains% | 2.1 | 14.1 | 4.3 | 1.6 | 3.2 | ||
| 30 | Correlation coefficients | 0.1430 | 0.0163 | 0.1892 | 0.0278 | 0.0654 | |
| p | .039 | .025 | .170 | .018 | .539 | ||
| Explains% | 0.6 | 5.4 | 2.1 | 1.4 | 0.03 |
- Note: p value is a significant test of correlation coefficient between soil nutrient variables and herbage spatial distribution.
3.6 Changes in the biomass of H. ammodendron
The results showed that shrub density decreased significantly in the 30-year-old H. ammodendron plantation. There was a decrease of 0.36 ha−1 individual−1 for the 5-year-old plants and a decrease of 0.16 ha−1 individual−1 for the 30-year-old plants (Figure 4a). The aboveground biomass of the shrubs increased significantly from 5 to 25 years. The leaf biomass was 5.34 × 10−5 kg m−2, and the branch biomass was 6.22 × 10−5 kg m−2 at 25 years. However, the leaf biomass decreased to 4.45 × 10−5 kg m−2, and the shoot biomass decreased to 4.00 × 10−5 kg m−2 at 30 years (Figure 4b). The vegetation coverage increased significantly from 0 to 25 years but had a decreasing trend after 25 years, which was consistent with the pattern of the vegetation cover (Figure 4c). The proportion of dead leaves and branches increased after 30 years (Figure 4d).
4 DISCUSSION
4.1 Effects of vegetation restoration on vegetation and soil nutrients
Community characteristics, such as species diversity, vegetation coverage, and soil nutrients, are changed due to vegetation restoration with increasing succession time (Lu et al., 2018; Tateno et al., 2017). The results from the present study showed that the species richness and vegetation coverage were increased after H. ammodendron planting, and thus restored the vegetation in the desert–oasis transition zone. Previous studies have revealed a similar pattern whereby afforestation was shown to accelerate the understory community dynamics (Bourgeois, Vanasse, González, Andersen, & Poulin, 2016). Species engineering effects are similar among ecosystems; however, they can be different depending on the environmental situation (Bourgeois et al., 2016). Vegetation growth also affects the availability of site resources such as soil nutrients and soil condition restricts the growth of vegetation (Pardon et al., 2017). The results from the present study showed that SOM, TN, AN, AP, and AK concentrations significantly increased over time. The study revealed a pattern similar to that found by a previous study in which soil and vegetation were two interdependent factors (Tilman, 1982). Soil nutrient content plays a decisive role in species diversity (Li, Zhao, Zhu, Li, & Wang, 2007). Increased plant diversity results in the full utilization of limited nutrients in the soil and reduces the leaching loss of N in ecosystems (Wang et al., 2010). Ecosystems with higher diversity can maintain higher ratios of nutrients and increase the total reserves of soil nutrients, thus increasing productivity (Tilman, 1982).
The results from the present study showed that the number of soil nutrients in the 0–10 cm depth was greater than that in the 10–20 cm depth and the soil nutrient content under the shrub canopy was higher than that in the intershrub area. This pattern was related to the changes in climatic conditions and soil moisture after vegetation restoration. The soil moisture content of the deep soil can degrade shrubs, causing grasses to become the dominant species (Chen et al., 2018). Organic matter and plant roots play an important role in water conservation (Garcia, Amann, & Hartmann, 2018). High soil fertility can increase SOM and nitrogen content and promote root development, allowing roots to absorb water from the deep soil and thereby improving drought resistance (Grime, 1977). Organic matter can loosen the soil and enhance water retention (Grime, 1977). A previous study found that herb productivity increased rapidly during the initial growth stage because of the increased light saturation in intershrub areas (Dresp-Langley & Reeves, 2014). Therefore, afforestation and the restoration of arid shrubs have a positive effect on the availability of soil nitrogen and cycling in desert areas.
In addition, the results showed that the number of families, genera, and species increased after H. ammodendron planting and species richness gradually increased from 1–5 years to 10–20 years but then decreased at 20–30 years. Therefore, the species diversity index of the desert plant community increased with time, whereas the trend of ecological dominance has shown an opposite pattern (Garcia et al., 2018). Because the plots in the present study were not next to each other and there was space between them, the number, richness, and productivity of plant community species might have been directly affected by the variation in precipitation (Li et al., 2018).
4.2 Effect of soil nutrients on species diversity and spatiotemporal heterogeneity
The results showed that the number of species increased after H. ammodendron planting and species richness gradually increased from 1–5 years to 10–20 years but then decreased at 20–30 years. The spatial heterogeneity of the vegetation decreased with increasing AK content in the soil at 1–10 years. This result is similar to that found in previous studies in which the species diversity of plants decreased due to low levels of soil nutrients (Deng et al., 2018; Xiao et al., 2017). Any study on changes in species diversity should consider soil nutrient dynamics after vegetation restoration at a regional scale. Previous studies have indicated that changes in species diversity and the spatial distributions of many tree species after vegetation restoration are strongly associated with a combination of factors, including soil nutrients (Deng et al., 2016; Pingree & DeLuca, 2018). Soil nutrient resource availability plays an important role in the assembly of tropical plant communities at the local scale. The soil nutrient supply ratio influences changes in plant species diversity in grassland communities; however, the functional diversity of nutritional strategies can result in high plant species diversity in infertile soils, and species distributions in tropical tree communities can respond to individual soil nutrients and seasonal drought (Tilman, 1982).
Climatic conditions and soil moisture content can create changes in vegetation recovery increases over time. The vegetation restoration method is the main influencing factor of soil moisture change in the region. Soil moisture is the main factor restricting plant growth and the primary limiting factor for vegetation community structure and ecosystem function. Precipitation affects soil moisture content and its spatial variability at a large scale. Precipitation changes with an increase in the vegetation restoration time series, which affects species diversity and spatiotemporal heterogeneity (Ye et al., 2019). Soil nutrients are the primary factor limiting land vegetation coverage and greatly influence the distribution of vegetation (Liu, Jiang, Zhu, Li, & Jin, 2018). Several studies have suggested that planted shrubs degrade and shift to the dominance of grasses because of a decrease in the moisture content in the deep soil (Chen et al., 2018). Our study demonstrated that soil nutrients, in response to vegetation activity, were mainly dependent on vegetation type, species diversity, and restoration activity. Therefore, the spatial and temporal patterns of soil nutrients strongly affect important facets of plant community structure, and soil nutrients have a significant effect on plant diversity and spatiotemporal heterogeneity.
4.3 Effects of the landscape on species diversity and spatial heterogeneity
The present study showed that the dominant plant species in shrub plantations changed significantly over a 30-year period with a clear pattern that first increased and then decreased. This result was the same as that found in the previous studies in which field conditions and landscape background influenced the successional process of vegetation (Johnson & Miyanishi, 2008), especially during the early stage of succession (van Hall, Cammeraat, Keesstra, & Zorn, 2016). As plant species coverage changes, the importance of the landscape environment increased gradually (Liu et al., 2017; van Hall et al., 2016). Changes in the landscape pattern have significant effects on the evolution of soil properties and habitat conditions (Kovář, Štefánek, & Mrázek, 2011; Tulloch et al., 2018). The distribution of vegetation is affected by habitat conditions (Kovář et al., 2011). The species richness of herbaceous plant communities may be affected by forest fragmentation, habitat loss, and management practices (Kovář et al., 2011). The negative effects of habitat fragmentation on species richness and total herbaceous species in ancient forests and for Red List species are mainly attributed to the edge effect itself or the combined effects of the forest edge and forest continuity (Kovář et al., 2011; Tulloch et al., 2018).
In the present study, the number of dominant species increased, then the plant density and coverage increased, after which the biomass and diversity increased, reaching the highest level at 20 years, stabilizing at 25 years, and exhibiting a slight decrease at 30 years. Species diversity indicators of plant communities have been shown to vary because the relationship between environmental factors and photosynthesis is not constant and vary greatly with different climatic conditions, soil moisture conditions, soil nutrients, plant species, and growth stages, and because of the demand by individual plants for light, nutrients, and water increase with age (Tilman, 1982). In addition, spatial heterogeneity is greatly affected by spatial autocorrelation factors (Liu et al., 2018). Because light is sufficient for increasing the soil temperature in locations that have small canopies, increased soil temperatures promote seed germination and seedling growth. In the present study, the variation in soil moisture in different H. ammodendron plantations was related to the distribution patterns of the vegetation and roots, as well as the slope and depth. However, the effects of climate change were not considered.
4.4 Implications of ecological restoration
The present study showed that H. ammodendron reached its highest coverage and biomass at 25 years. This result is similar to that of a previous study in which the development of a shrub canopy reduced intraforest gaps (Wang, Zha, et al., 2015; Wang, Zhao, et al., 2015). H. ammodendron richness, vegetation coverage, and plant biomass decreased and the proportion of dead plant biomass increased significantly with increasing age. Many studies have shown that the number of species, vegetation coverage, and biomass do not increase substantially at the beginning growth stage, but they do increase sharply to their maximum levels during the later growth stages (Frouz et al., 2015; Liu et al., 2015). H. ammodendron has strong adaptability and it can form a good shrub–herb community because of its stable growth characteristics. It can fix sand, form windbreaks, and rebuild and stabilize ecosystems by increasing biodiversity and vegetation growth and improving landscape patterns. The cultivation of H. ammodendron has a moderate influence on vegetation coverage, plant number, species diversity, and community biomass. It also affects plant photosynthetic capacity and leaf-specific apparent hydraulic conductivity (Xu et al., 2011). Shrub planting increased the quality of the soil nutrients, vegetation diversity, and vegetation spatial heterogeneity in the early years; however, these trends began to decline after 25 years. Therefore, measures could be initiated, such as replanting shrubs in the forest, after 25 years of growth to promote ecosystem restoration and maintain vegetation diversity.
The surface soil nutrient status and content increased slightly after H. ammodendron planting. AN and SOM contents in the various soil layers showed a decreasing trend from the surface to the deep layers. In addition, the soil nutrients were enriched by the cultivation of H. ammodendron and the soil nutritional quality was improved. H. ammodendron is a shrub that tolerates arid conditions and is extremely environmentally adaptable. The distribution pattern of the H. ammodendron population in the desert–oasis edge of Minqin County was generally aggregated. The species diversity patterns and spatial heterogeneity of the vegetation changed in the shrub planting areas, increasing the ecological forest area in Minqin County and forming a green gallery, which has become an important barrier that protects the Minqin Oasis, with a length of more than 300 km. During the 30 years of shrub growth, soil nutrition first increased, then decreased, and finally began to decrease slightly after 25 years. Therefore, management measures after 25 years could improve the species diversity patterns and spatial heterogeneity of the forest.
Afforestation changes the composition and structure of vegetation and the distribution pattern of soil nutrients in ecosystems. It also contributes to the ecosystem stability and ecological restoration in desert–oasis ecotones. The reconstruction of shrub and herb species is an effective method for ecosystem restoration and improvement of land productivity, and to combat desertification in desert–oasis ecotones. Exploring the influence of shrub planting on the spatial pattern and species diversity of vegetation and soil nutrients is theoretically significant for understanding the role and efficiency of shrubs in sand fixing and as windbreaks for ecosystem restoration in desert–oasis ecotones. However, only the relationships between soil nutrients and spatial heterogeneity in vegetation were used to analyze the effects of shrubs on the ecosystem in the present study. This resulted in certain limitations when studying the effects of shrub planting on vegetation and soil nutrients.
5 CONCLUSIONS
Overall, the findings of our study indicate that shrub planting has a significant and extensive effect on soil nutrients, plant species composition, diversity patterns, and spatial heterogeneity. Our results showed that the soil nutrient content increased and soil quality was improved by shrub planting and changed with time since vegetation restoration. We demonstrated that plant diversity increased, stabilized, and then decreased, with the greatest diversity at 15–25 years. Therefore, our findings highlight the importance of spatiotemporal heterogeneity analyses when investigating the influence of shrub planting on soil nutrients, plant diversity, and spatiotemporal heterogeneity. Our study suggests that management strategies, such as shrub planting and aerial herb seeding, should be used for revegetation in arid areas to improve soil quality and plant diversity over a long period, resulting in sustainable ecosystem restoration. Due to different water use strategies, drought tolerances, and ecosystem service capacities, the shrubs or grasses used in ecosystem restoration should be chosen based on the environmental conditions, restoration aims, and management strategies.
ACKNOWLEDGMENTS
The work was financially supported by the Central Non-profit Research Institution of CAF (CAFBB2017ZB004), a China Scholarship Council (CSC) scholarship at the University of Quebec at Montreal, Canada, and a National Science and Engineering Research Council of Canada (NSERC) Discover Grant.
CONFLICT OF INTEREST
The authors declare no conflicts of interest.
