Ecological restoration treatments enhanced plant and soil microbial diversity in the degraded alpine steppe in Northern Tibet
Funding information: Jiangsu Collaborative Innovation Center of Technology and Material of Water Treatment; National Key Research & Development Program of China, Grant/Award Number: 2016YFC0502002
Abstract
Plant community structure and microbial ecological functions of alpine steppe in Northern Tibet are susceptible to variations in the natural environment and anthropogenic activities. In recent times, the alpine steppe in Northern Tibet has been severely degraded, that is, plant biodiversity was reduced and plant coverage declined. It is crucial to restoring the degraded alpine steppe in Northern Tibet to maintain the sustainable development of the environment and social economy. This study aimed to evaluate the ecological engineering benefits of four types of ecological restoration treatments: (a) independent treatment of sowing plant seeds; (b) combined treatments with (a) and fertilized microbial inoculum; (c) combined treatments with (a) and spraying hydrophilic polyurethane (which can enhance the ability of moisture retention, fertility, and temperature in the soil subsystem); and (d) the comprehensive treatments with (a) and fertilized microbial inoculum and spraying hydrophilic polyurethane. Noting their effect on plant growth performance, plant taxonomic and functional diversity, soil physicochemical properties, and soil microbial diversity in the degraded alpine steppe. Plots where no seeds were planted and no other amendments were applied were selected as experimental controls. Plant height, leaf size, plant taxonomic and functional diversity, total carbon and nitrogen contents, soil bacterial and fungal diversity were significantly enhanced under the four ecological restoration treatments. Plant growth performance (in particular, competitiveness for sunlight acquisition and leaf photosynthetic area), and soil fertility were significantly enhanced under the four treatments. The improved plant taxonomic and functional diversity and soil microbial diversity could facilitate the biological communities in degraded alpine steppe to develop better ecological functions, especially community stability. Meanwhile, the combination of all three treatments was found to be the best among all treatments applied in this study. This can help restore the degraded alpine steppe in Northern Tibet.
1 INTRODUCTION
As the largest terrestrial ecosystem type, grassland covers about 50% of the Earth's land area and thereby plays a crucial role in the function and stability of global ecosystems (Reeder & Schuman, 2002). Northern Tibet (elevation is often >4,500 m) is occupied by extensive natural grassland (up to 48,000 km2), accounting for approximately 80% of Northern Tibet land area and approximately 60% of the natural steppe area of the Tibet Autonomous Region, respectively (Hu, 2000). The community structure and ecological functions of alpine steppe in Northern Tibet are susceptible to the shifts in the natural environment and anthropogenic activities in Tibet Autonomous Region (Zhao, Yu, Wu, Luo, & Miao, 2013). Furthermore, the alpine steppe possesses unique ecological characteristics, such as low temperature, little rainfall, infertile soil, fewer plant species, low plant productivity, low self-repairing ability, and weak community stability (Gao et al., 2010; Wang, Wei, Wu, Wang, & Jiang, 2019). Moreover, soil structure deterioration and low soil fertility, with less diversity and richness of soil microbial communities, have aggravated the problem in the alpine steppe in Northern Tibet (Rong et al., 2018). Thus, the alpine steppe has been severely degraded, especially in recent decades due to natural environmental and anthropogenic activities (such as climate change, rainfall variation, over-grazed pasture, urban area and road construction, unregulated harvesting of wild plants, and hunting of wild animals). All these acted as strong drivers of the shifts in ecological health and environmental security, as well as socially sustainable development (Gao et al., 2010; Wang, Wei, Wu, Wang, & Jiang, 2019; Zhao et al., 2013). Thus, it is imperative to restore the degraded alpine steppe in Northern Tibet to maintain the sustainability of the ecology and environment and social economy.
Numerous studies have reported that sharp decrease of plant biodiversity (mainly taxonomic diversity) was one of the key reasons for the degradation of alpine steppe (Gao et al., 2010; Han et al., 2008; Wang, Wei, Wu, Wang, & Jiang, 2019; Zhao et al., 2013). Ecosystem function is largely driven by plant functional diversity rather than plant taxonomic diversity (Tobner et al., 2016; Wang, Wu, Jiang, Zhou, & Du, 2019; Wu, Zhang, Jiang, Zhou, & Wang, 2019; Zhang et al., 2017; Zhang, Wang, Kaplan, & Liu, 2015). Furthermore, soil microorganisms, as a key factor in material recycling and energy flows in ecosystems, are essential for the ecological functions and community stability of ecosystems (Harris, 2009; Liang et al., 2014; Zuo et al., 2016). Thus, it was important to assess the effects of ecological restoration treatments on plant taxonomic and functional diversity and soil microbial diversity in the degraded alpine steppe in Northern Tibet.
The present study aimed to evaluate the ecological engineering benefits of four types of ecological restoration treatments on plant growth performance, plant taxonomic and functional diversity, soil physicochemical properties, and soil microbial diversity in the degraded alpine steppe.
Based on the ecological restoration treatments applied in this study, the following hypotheses were proposed: (a) Plant growth performance improves significantly; (b) plant taxonomic and functional diversity increases substantially; and (c) soil microbial diversity improves significantly under these treatments.
2 MATERIALS AND METHODS
2.1 Study design
In early-June 2018, 134,400 m2 of the degraded alpine steppe were randomly selected in Northern Tibet in Shenza, Tibet Autonomous Region (30°56′46″ N, 88°37′54″ E), characterized by a sampling area with the semiarid climate of the plateau subfrigid zone. It is approximately 4,700 m above sea level. Figure S1 shows the approximate geographic location of the selected degraded alpine grassland in Northern Tibet (Plateau). The annual mean temperature of the area is approximately 0.2°C, with maximum and minimum temperatures of 25.1 and −30.1°C, respectively. The annual precipitation is approximately 299 mm. These data were collected from local climate records (Wang, 2015). There were no shrubs or trees in the selected grassland due to harsh ecological conditions. The selected degraded alpine steppe was treated with four types of ecological restoration, that is: (1) independent treatment with sowing plant seeds only (P); (2) combined treatments with sowing plant seeds and fertilized microbial inoculum (PM); (3) combined treatments with sowing plant seeds and spraying hydrophilic polyurethane [HP, which can enhance moisture retention, fertility, and temperature in the soil subsystem; Chemical formula: (C10H8N2O2·C6H14O3)n; Manufacturer: Toho Chemical Industry Co., Ltd., Tokyo, Japan] (PHP); (4) comprehensive treatment with sowing plant seeds and fertilize microbial inoculum, as well as spraying HP (PMHP). Plots where no seeds were planted and no other amendments were applied were considered to be experimental controls (C). Specifically, the sown seeds of plant species, including Elymus nutans Griseb., Festuca elata Keng ex E. Alexeev, Elymus sibiricus Yajiang, Poa annua L., Lolium perenne L., and Medicago sativa L. with a 5:1:1:1:1:1 ratio. E. nutans Griseb. is the most important grass species among these species in the alpine steppe in Northern Tibet. The amount of microbial inoculum was set at approximately 1 kg/666.667 m2, and the content of HP was approximately 3% (Liang & Wu, 2016; Liang, Wu, Noori, & Deng, 2017; Liang, Wu, Noori, Yang, & Yao, 2017).
In late August 2018, 10 quadrat replicates (2 m × 2 m) per ecological restoration treatment were evaluated. Three replicates of plant samples of the same species from the same quadrat were randomly selected to determine plant functional traits. Furthermore, all quadrats were analyzed to record the number of individuals of all plant species, the total number of individuals per plant species, and the number of plant species. Three soil cores (top 10 cm) per quadrat were randomly harvested and homogenized to form soil samples (three soil samples per quadrat). All soil samples were stored in sealed bags and kept at −20°C before further analysis.
2.2 Determination of plant growth performance
Four functional traits related to plant growth performance were measured that included plant height (indicating competitiveness for sunlight acquisition) (Wang, Jiang, Liu, Zhou, & Wu, 2018; Wang, Jiang, Zhou, Xiao, & Wang, 2018; Wang, Wu, Jiang, & Zhou, 2018; Wang, Wu, Jiang, Zhou, & Du, 2019; Wu et al., 2019), leaf size (estimated as leaf length and leaf width, indicating leaf photosynthetic area) (Wang, Jiang, Liu, Zhou, & Wu, 2018; Wang, Wu, Jiang, & Zhou, 2018; Wang, Wu, Jiang, Zhou, & Du, 2019; Wu et al., 2019), and green leaf area (indicating leaf photosynthetic area) (Huang et al., 2018; Xia et al., 2016). The value of each functional trait of plant species in one quadrat was calculated using the mean value of identical functional traits of all plant species in the same quadrat.
2.3 Determination of plant taxonomic and functional diversity
Plant taxonomic diversity was calculated by using the number of plant species (S), Shannon's diversity index (H′; Shannon & Weaver, 1949), Simpson's dominance index (D; Simpson, 1949), and Margalef's richness index (F; Margalef, 1951).
Plant functional diversity was calculated based on community-weighted mean functional traits (CWM), Rao's quadratic entropy (FDQ), and Mason α functional diversity (Fα). Rao's quadratic entropy was assessed as FDQ = ∑dijPiPj (Botta-Dukat, 2005; Ricotta & Moretti, 2011), where Pi and Pj represent the relative abundances of plant species, i and j, in one quadrat, respectively, and dij represents the interspecific distance. The interspecific distance was estimated as dij = 1/n∑(Xik − Xjk)2, where Xik and Xjk represent the functional trait k of plant species, i and j, in one quadrat, respectively, and n represents the number of the determined functional traits. Mason α functional diversity index was assessed as Fα = ∑Pi(Xi − X) (Mason, MacGillivray, Steel, & Wilson, 2003), where Pi represents the relative abundance of a plant species, i, in one quadrat (Pi was calculated as Pi = ni/N, where ni represents the number of individuals of plant species i and N represent the total number of individuals of all plant species in one quadrat), Xi represents CWM (calculated as the mean functional trait value of the community, weighted by the relative abundance of each plant species) of plant species, i, in one quadrat, and X represents CWM of all plant species in the same quadrat (Lavorel et al., 2008; Ricotta & Moretti, 2011).
2.4 Determination of soil physicochemical properties
Soil pH was determined in situ using a digital soil acidity meter (ZD-06; ZD Instrument Co., Ltd., Taizhou, China) (Wang et al., 2018; Wang, Jiang, Zhou, & Wu, 2018; Wang, Jiang, Zhou, Xiao, & Wang, 2018; Wang, Wu, Jiang, Wei, & Wang, 2019; Wang, Zhou, Liu, & Du, 2017). Dry matter content, total carbon (C), total nitrogen (N), total phosphorus (P), and total potassium (K) contents were estimated by Genepioneer Biotechnologies Co., Ltd., Nanjing, China according to the equivalent industry standards of the People's Republic of China (NY/T 1121.6–2006, LYT 1228–2015, LYT 1232–2015, and LYT 1234–2015, respectively).
2.5 Determination of soil microbial diversity
Soil bacterial and fungal communities were determined by high-throughput sequencing with Hiseq PE250 at Genepioneer Biotechnologies Co., Ltd., Nanjing, China. This technique can identify the variations in soil bacterial and fungal communities in response to different types of ecological restoration treatments. The amplification of the V4-V5 region of bacterial rRNA genes was performed using universal bacterial primers: 515F (5′-GTGCCAGCMGCCGCGG-3′) and 907R (5′-CCGTCAATTCMTTTRAGTTT-3′) (Liu et al., 2015; Ren, Ren, Teng, Li, & Li, 2015; Zhang, Chen, Zhang, & Lin, 2016). The amplification of ITS region of fungal rRNA genes was performed by using the universal fungal primers: ITS1F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2R (5′-GCTGCGTTCTTCATCGATGC-3′) (Dorn-In et al., 2013; Kataoka, Taniguchi, Ooshima, & Futai, 2008; Liu et al., 2015).
Sequences with similarities of greater than or equal to 0.97 were grouped into operational taxonomic units (OTUs) using UCLUST (version 1.2.22) (Bokulich et al., 2013; Edgar, Haas, Clemente, Quince, & Knight, 2011), and the taxonomic classification (at the kingdom, phylum, class, order, family, genus, and species levels, respectively) was performed based on ribosomal DNA sequences using the Ribosomal Database Project (RDP; Version: 2.2; based on Bergey's taxonomy) with a classification threshold of 0.8 (Cole et al., 2005; Wang, Garrity, Tiedje, & Cole, 2007). The relative abundance of OTUs at the phylum level was then compared across all samples.
The alpha diversity of soil bacterial and fungal communities was assessed using the following indices: Observed species (indicative of the number of species), PD whole tree (indicative of the phylogenetic diversity), Shannon's diversity index (indicative of species diversity) (Rodrigues, Torres, & Ottoboni, 2014; Shannon & Weaver, 1949), Simpson's index (indicative of species dominance) (Simpson, 1949), and Chao1 index (indicative of species richness) (Chao, Chazdon, Colwell, & Shen, 2005). Good's coverage was estimated as an indicator of the level of coverage of the sample library (Rodrigues et al., 2014).
SNB features in each group were characterized using the linear discriminant analysis (LDA) effect size (LEfSe) method for biomarker discovery, which emphasizes the statistical significance and biological relevance (Segata et al., 2011; Zhang et al., 2013; Zhong, Yan, & Shangguan, 2015). With the normalized relative abundance matrix, the LEfSe method uses the Kruskal–Wallis rank-sum test to unearth the features with significantly different abundances between the assigned taxa and performs LDA to estimate the effect size of each feature. Only taxa with average abundances of >1% were considered significant. A significance level of .05 and an effect-size threshold of 3 were used for all the biomarkers evaluated in this study (Segata et al., 2011; Zhong et al., 2015).
2.6 Statistical analysis
Differences in the values of plant growth performance, plant taxonomic and functional diversity, soil physicochemical properties, and soil microbial diversity were analyzed by ANOVA followed by the Tukey's honestly significant difference post hoc test for multiple comparisons. Correlation patterns were measured between soil physicochemical properties and soil microbial diversity by correlation analysis (using the Pearson product–moment correlation coefficient, r). Statistical significance was determined at p value equal to or less than .05. All statistical analyses were performed using IBM SPSS Statistics (version 25.0; IBM Corp., Armonk, NY).
3 RESULTS
3.1 Differences in plant growth performance
The differences in plant growth performance were observed in all four types of ecological restoration treatments (Figure 1). Specifically, plant height, leaf size, and green leaf area under PMHP were significantly greater than those under C (p < .05; Figure 1). Furthermore, these characteristics under PMHP were considerably greater than all other ecological restoration treatments (p < .05; Figure 1). However, leaf width was found to be greater under P compared to C (p < .05; Figure 1). Similarly, leaf width under PMHP was distinctively greater than that under PM (p < .05; Figure 1).
3.2 Differences in plant taxonomic and functional diversity
Differences in plant taxonomic diversity were detected across all types of ecological restoration treatments (Figure 2). Specifically, S was significantly higher under PM and PHP than that under C (p < .05; Figure 2). Furthermore, H′ was remarkably lower under PHP and PMHP than C (p < .05; Figure 2). Overall, H′ under P was found to be higher than those under all other ecological restoration treatments (p < .05; Figure 2). Similarly, D was significantly greater under PM, PHP, and PMHP compared to that of C and P (p < .05; Figure 2).
Furthermore, differences in plant functional diversity were observed across all types of ecological restoration treatments (Figures 3, 4, & S2). Specifically, CWM of plant height was significantly higher under PM, PHP, and PMHP compared to that of C and P (p < .05; Figure 3). Furthermore, CWM of leaf length under PM, PHP, and PMHP was noticeably greater than that under C (p < .05; Figure 3). Overall, PMHP showed significantly higher CWM of leaf length compared to all remaining ecological restoration treatments (p < .05; Figure 3). Similarly, CWM of leaf width was significantly higher under P, PHP, and PMHP than that under C (p < .05; Figure 3). Furthermore, CWM of leaf width under PMHP was considerably higher than that under PM and PHP (p < .05; Figure 3). CWM of green leaf area under PMHP was predominantly higher than that under C and other remaining ecological restoration treatments (p < .05; Figure 3). FDQ of plant height was greater under PM, PHP, and PMHP than that under C (p < .05; Figure 4). Similarly, FDQ of plant height under PM was also higher than that under P (p < .05; Figure 4). FDQ of leaf length under PMHP was found to be significantly higher than that under C, PM, and PHP (p < .05; Figure 4). FDQ of green leaf area under PMHP was noticeably higher than that under C (p < .05; Figure 4). Furthermore, PM, PHP, and PMHP exhibited greater Fα compared to C and P (p < .05; Figure S2).
3.3 Differences in soil physicochemical properties
Differences in soil physicochemical properties were detected across all ecological restoration treatments (Figure 5). Low soil pH was determined under PM, PHP, and PMHP compared to C (p < .05; Figure 5). Total C content under all ecological restoration treatments was apparently higher than that under C (p < .05; Figure 5). Moreover, total C content under PM, PHP, and PMHP was higher than that under P (p < .05; Figure 5). Similarly, total N content was significantly decreased in the following order: PMHP > PHP > PM > P > C (p < .0001; Figure 5). Furthermore, total P content under P was significantly higher than that under C but contrary for PM and PHP (p < .05; Figure 5). Total K content under P was significantly higher compared to C but contrary for PM, PHP, and PMHP (p < .05; Figure 5).
3.4 Differences in soil microbial community structure
Differences in soil microbial diversity were observed across all ecological restoration treatments (Figure 6). Specifically, observed species and PD whole tree of soil bacteria under P and PM were higher than those under C (p < .05; Figure 6). Similarly, observed species and PD whole tree of soil fungi under PM were evidently higher than those under C and PHP (p < .05; Figure 6). Furthermore, Chao1 index of soil fungi under PM was significantly higher than that under C, PHP, and PMHP (p < .05; Figure 6).
The Good's coverage for soil bacteria and fungi across all of the samples was approximately 99.7240 and 99.8977%, respectively. At the phylum level, four types of ecological restoration treatments significantly decreased the relative abundance of Proteobacteria for soil bacteria but considerably increased the relative abundance of Ascomycota for soil fungi compared to that of CK (Figure S3).
LEfSe analyses (LDA values are shown in Figure 7) were performed to determine statistically significant differences in taxon abundance and to verify the biological relevance of the species under four types of ecological restoration treatments (Figure 8). Specifically, p_Cyanobacteria, g_Pseudonocardia, g_Singulisphaera, and o_Rhodospirillales were found to be the four dominant biomarkers of soil bacteria with maximum LDA values for samples under C (Figure 7). f_Xanthomonadaceae was the dominant biomarker of soil bacteria with maximum LDA values for the sample under P (Figure 7). Similarly, f_0319_6M6, c_Acidimicrobiia, o_Acidimicrobiales, and g_Phreatobacter were four dominant biomarkers of soil bacteria with maximum LDA values for samples under PHP (Figure 7). Further, c_Thermoleophilia, o_Rhizobiales, o_Gaiellales, and p_Nitrospirae were the four dominant biomarkers of soil bacteria with maximum LDA values for the sample under PMHP (Figure 7). Furthermore, o_Geastrales, s_Geastrum_hungaricum, f_Geastraceae, and g_Geastrum were four dominant biomarkers of soil fungi with maximum LDA values for the sample under C (Figure 7). f_unidentified, s_unidentified, and g_unidentified were three dominant biomarkers of soil fungi with maximum LDA values for the sample under P (Figure 7). g_Ramularia, g_Isaria, g_Dioszegia, and o_Sordariomycetes_ord_Incertae_sedis were found to be the four dominant biomarkers of soil fungi with maximum LDA values for the sample under PM (Figure 7). Similarly, c_Orbiliomycetes, o_Dothideomycetes_ord_Incertae_sedis, f_Dothideomycetes_fam_Incertae_sedis, and g_Minutisphaera were four dominant biomarkers of soil fungi with maximum LDA values for the sample under PMHP (Figure 7). Additionally, f_Xanthomonadaceae and f_0319_6M6 were primarily changed for soil bacteria under P and PHP, respectively (Figure 8). Moreover, f_Micrococcaceae, f_Streptomycetaceae, f_Gaiellaceae, f_288_2, f_Elev_16S_1332, f_FFCH13075, f_Gsoil_1167, f_Hyphomicrobiaceae, f_JG34_KF_361, f_Rhizobiaceae, f_RhizobialesIncertaeSedis, f_Xanthobacteraceae, and f_Archangiaceae were primarily changed for soil bacteria under PMHP (Figure 8). In the same way, f_Geastraceae and f_unidentified were primarily changed for soil fungi under C and P, respectively (Figure 8). Further, f_Tubeufiaceae and f_Sordariomycetes_fam_Incertae_sedis were primarily changed for soil fungi under PM (Figure 8). Similarly, f_Dothideomycetes_fam_Incertae_sedis and f_Hygrophoraceae were primarily changed for soil fungi under PMHP (Figure 8).
3.5 Relationships between soil physicochemical properties and soil microbial diversity
Soil pH was negatively correlated with the observed species, PD whole tree, and Shannon's diversity index of soil bacteria (p < .05; Table 1). Furthermore, it was observed that dry matter content was positively correlated with observed species, PD whole tree, and Chao1 index of soil bacteria (p < 0.05; Table 1). However, soil physicochemical properties did not show strong relationships with soil fungal diversity (p > .05; Table 1).
| Soil bacteria | Soil fungi | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Observed species | PD whole tree | Shannon's diversity index | Simpson's index | Chao1 index | Observed species | PD whole tree | Shannon's diversity index | Simpson's index | Chao1 index | ||
| Soil pH | r | −.523* | −.535* | −.544* | −.265 | −.492 | −.272 | −.250 | −.141 | −.173 | −.325 |
| p | .046 | .040 | .036 | .339 | .063 | .326 | .370 | .617 | .538 | .237 | |
| Dry matter content | r | .633* | .621* | .239 | −.005 | .572* | .074 | .087 | −.088 | −.083 | .134 |
| p | .011 | .013 | .392 | .985 | .026 | .794 | .757 | .755 | .769 | .635 | |
| Total C content | r | .336 | .377 | .378 | .115 | .343 | .316 | .283 | .028 | .133 | .341 |
| p | .221 | .166 | .164 | .684 | .210 | .251 | .306 | .922 | .638 | .214 | |
| Total N content | r | .326 | .363 | .538* | .301 | .335 | .256 | .217 | .168 | .306 | .251 |
| p | .235 | .184 | .038 | .276 | .222 | .358 | .437 | .549 | .267 | .366 | |
| Total P content | r | .268 | .258 | .148 | .157 | .236 | .132 | .153 | .388 | .252 | .090 |
| p | .335 | .354 | .599 | .576 | .397 | .638 | .586 | .153 | .364 | .749 | |
| Total K content | r | .094 | .075 | −.042 | .058 | .061 | .032 | .055 | .249 | .107 | −.001 |
| p | .739 | .791 | .882 | .837 | .830 | .911 | .845 | .370 | .706 | .998 | |
- Note: p values equal to or less than .05 are shown in bold.
- * Indicates a significant difference at the .05 probability level.
4 DISCUSSION
The results of this study demonstrated that PMHP significantly increased plant height, leaf size, and green leaf area of plant species. However, P showed a significant increase in the leaf width of plant species. This showed that the competitiveness for sunlight acquisition and leaf photosynthetic area was apparently enhanced under PMHP. Thus, the increase in plant height, leaf size, and green leaf area of plant species under PMHP can confer them high competitiveness for resource acquisition and utilization, especially sunlight, which may be one of the major environmental factors affecting the growth and development of plant (Liu et al., 2010; Wang, Jiang, Liu, Zhou, & Wu, 2018; Wang, Wu, Jiang, Zhou, & Du, 2019). Moreover, the result implied that the maximum growth performance of plant species was observed under PMHP. The possible reason for best plant growth performance under PMHP can be attributed to the positive effects of microbial inoculum and HP on the plant. Specifically, microbial taxa in the inoculum could have several ecological functions to promote the plant growth, such as auxin secretion (mainly indoleacetic acid), iron ionophores production, and the ability to dissolve phosphorus (data not shown). Furthermore, HP can enhance the ability of moisture retention, fertility, and temperature in the soil subsystem (Liang & Wu, 2016; Liang, Wu, Noori, & Deng, 2017; Liang, Wu, Noori, Yang, & Yao, 2017; Wu, Gao, Wu, Iwashita, & Yang, 2011). Based on plant growth performance, PMHP was found to be the best among all four ecological restoration treatments in the degraded alpine steppe in Northern Tibet. These results validated the first hypothesis of this study.
Besides plant growth performance, plant diversity is one of the factors that need to be evaluated to reap the ecological engineering benefits of these restoration treatments. Previous studies have shown that high plant diversity can protect the ecosystem productivity from negative perturbations caused by external disturbances, leading to greater ecological functions and community stability (Hautier et al., 2014, 2015; Isbell et al., 2011; Wagg et al., 2017; Wang et al., 2019; Wang, Wei, Wu, Wang, & Jiang, 2019). In this study, it was observed that PM and PHP significantly increased the S value compared with C mainly due to the positive effects of these treatments on plant growth performance. In contrast, PHP and PMHP showed decreased H′ compared with C that can be attributed to the increasing interspecific competition with improved plant growth performance. Furthermore, D has prominently improved under PM, PHP, and PMHP treatments compared with C. Moreover, plant functional diversity (including CWM, FDQ, and Fα) was significantly enhanced under all four types of ecological restoration treatments, especially PMHP. This may be due to the enhanced plant growth performance under all treatments. The increased plant functional diversity can be ascribed to the significant increase in plant communities due to thriving dominant species (species with maximum relative abundance), especially under PMHP. Previous studies showed that increased plant diversity was closely associated with higher stability (Hautier et al., 2014, 2015; Isbell et al., 2011; Wagg et al., 2017; Wang, Wei, Wang, Wu, & Cheng, 2020; Wang, Wei, Wu, Wang, & Jiang, 2019; Wang, Wu, Jiang, Zhou, Liu, & Lv, 2019). The results supported the second hypothesis of this study. Increased plant functional diversity under all treatments could increase the resource utilization efficiency (Gazol & Camarero, 2016; Wang, Wu, Jiang, Zhou, & Du, 2019) depending mostly on niche complementary effects (Schleicher, Peppler-Lisbach, & Kleyer, 2011; Spasojevic, Copeland, & Suding, 2014) and/or the plasticity in resource use of a few species (Ashton, Miller, Bowman, & Suding, 2010; Grassein, Till-Bottraud, & Lavorel, 2010). Various studies reported that plant functional diversity played a significant role in ecosystem processes than plant taxonomic diversity (Tobner et al., 2016; Wang et al., 2020; Wang, Wu, Jiang, Zhou, & Du, 2019; Wu et al., 2019; Zhang et al., 2015, 2017). Therefore, the highest values of most plant functional diversity under PMHP could enable the plant communities to possess greater ecological functions, especially community stability through intensified ecosystem services (Isbell et al., 2011; Tilman, Isbell, & Cowles, 2014; Tilman, Reich, & Knops, 2006; Wang, Wei, Wu, Wang, & Jiang, 2019; Wang, Wu, Jiang, Zhou, Liu, & Lv, 2019). It was also found that high plant diversity was strongly related to greater ecological functions, including community stability (Isbell et al., 2011; Tilman et al., 2006, 2014; Wang, Wei, Wu, Wang, & Jiang, 2019; Wang, Wu, Jiang, Zhou, Liu, & Lv, 2019). The possible reasons for this may be the following four mechanisms: (a) increased complementary effects (efficient complementary resource use was increased when more plant species coexisted together; Isbell, Polley, & Wilsey, 2009; Loreau & De Mazancourt, 2013); (b) efficient asynchrony effects (decrease in the abundance of several plant species can be balanced by the increase in the abundance of another plant species in one particular community; Yang et al., 2012; Loreau & De Mazancourt, 2013; Hautier et al., 2014); (c) synergetic covariance effects (negative covariance in the abundance of plant species in plant communities with greater diversity was higher than that with lesser diversity; Hector et al., 2010; Yang et al., 2012); and (d) secure portfolio effects (loss of certain plant species can be supplemented by the occurrence of similar plant species to maintain corresponding ecological functions; Isbell et al., 2009; Yang et al., 2012; Gherardi & Sala, 2015).
Total C and N contents significantly increased under the four types of ecological restoration treatments. This might be mainly attributed to the enhanced plant growth performance, which can release more C and N into the soil subsystem. However, total P and K contents increased under P but reduced under all other ecological restoration treatments. This implied that there might be P and K scarcity in the degraded alpine steppe in Northern Tibet for plant growth. Meanwhile, the ecological restoration treatments, especially PM, PHP, and PMHP resulted in soil acidification because plant species secreted more acidic substances into the soil subsystem. The enhanced acidification may also be due to the increased nitrification (Xu, Yu, Ma, & Zhou, 2012).
The alpha diversity of soil microorganism was increased under all four types of ecological restoration treatments, especially PM, which might be attributed to the fertilized microbial inoculum. Moreover, the relative abundance of soil microbial proportions changed significantly under all treatments. Specifically, the relative abundance of Proteobacteria for soil bacteria was significantly reduced; however, the relative abundance of Ascomycota for soil fungi was increased at the phylum level under all ecological restoration treatments. Furthermore, LEfSe analysis showed that all four types of ecological restoration treatments triggered significant variations in numerous microbial taxa (i.e., f_Geastraceae for soil fungi under C; f_Xanthomonadaceae for soil bacteria and f_unidentified for soil fungi under P; f_Tubeufiaceae and f_Sordariomycetes_fam_Incertae_sedis for soil fungi under PM; f_0319_6M6 for soil bacteria under PHP; f_Micrococcaceae, f_Streptomycetaceae, f_Gaiellaceae, f_288_2, f_Elev_16S_1332, f_FFCH13075, f_Gsoil_1167, f_Hyphomicrobiaceae, f_JG34_KF_361, f_Rhizobiaceae, f_RhizobialesIncertaeSedis, f_Xanthobacteraceae, and f_Archangiaceae for soil bacteria, as well as f_Dothideomycetes_fam_Incertae_sedis and f_Hygrophoraceae for soil fungi under PMHP). This indicated that these ecological restoration treatments trigger a substantial effect on certain microbial taxa. In this study, these four treatments exhibited significant impacts on soil microbial community structure, including alpha diversity and soil microbial proportions. The main reason for the altered soil microbial community structure might be attributed to the significant differences in plant diversity under these four treatments, which can release various types of litters and/or root exudates. Previous studies have reported that the secondary metabolites in plant litters and/or root exudates can influence soil microbial communities (Eisenhauer et al., 2010; Li, Yang, Zhou, & Shao, 2020; Wang, Zhou, Liu, Jiang, et al., 2018; Zuo et al., 2016). Moreover, the quantity and quality of organic matters released into the soil subsystem may be altered due to the mixing of plant litters and/or root exudates as a result of modified plant diversity and community composition. This altered organic matter and corresponding patterns of resource availability and utilization can significantly impact the soil microbial community structure (Chapman, Newman, Hart, Schweitzer, & Koch, 2013; Eisenhauer et al., 2010; Li et al., 2020; Wang, Zhou, Liu, Jiang, et al., 2018). High plant diversity could result in greater biochemical diversity of organic matters that can be colonized by highly diverse microbial communities (Chapman et al., 2013; Eisenhauer et al., 2010; Li et al., 2020; Zuo et al., 2016). Previous studies showed that soil microbial diversity was positively associated with plant diversity (Li et al., 2020; McGuire, Fierer, Bateman, Treseder, & Turner, 2012; Wang, Zhou, Liu, Jiang, et al., 2018; Zuo et al., 2016). Besides, an increase in organic matters has been associated with an increase in microbial growth (De Vries et al., 2012; Wardle, 2006). These results demonstrated that plant growth performance and most of plant taxonomic and functional diversity were significantly enhanced under all four ecological restoration treatments. This supported the third hypothesis of this study.
The correlation analysis showed that dry matter content was the strongest controlling factor for soil bacterial diversity, indicating that dry matter could have limited the growth of soil bacteria in the degraded alpine steppe in Northern Tibet. Plant productivity is very low, with few plant species in the degraded alpine steppe. Thus, there is very little for organic matter input in the barren area. In this study, soil pH was the main environmental variables of soil physicochemical properties, affecting the community structure of soil bacterial diversity. This was in agreement with previous studies (Shen et al., 2013; Wang et al., 2017; Wang, Jiang, Zhou, Xiao, & Wang, 2018). The negative association between soil pH and soil bacterial diversity might be due to reduced soil pH under four types of ecological restoration treatments. Previous studies showed that microbial communities (bacterial dominant) occurred in acidic environments, whereas fungal dominated microbial communities were found in neutral or alkaline environments (Högberg, Högberg, & Myrold, 2007; Rousk et al., 2010; Rousk, Brookes, & Bååth, 2009). Thus, it was confirmed that the community structure of soil bacterial diversity was mainly affected by soil pH and dry matter content.
In conclusion, plant growth performance, plant taxonomic and functional diversity, total C and N contents, and soil microbial diversity were significantly improved under the four types of ecological restoration treatments. Thus, these treatments were found to be feasible to restore the degraded alpine steppe. The findings of this study will help to understand the mechanisms of ecological restoration treatments on plant growth performance, plant taxonomic and functional diversity, and soil microbial diversity. Moreover, this study will provide a theoretical platform and practical basis for the ecological restoration of the degraded alpine steppe in Northern Tibet.
ACKNOWLEDGMENTS
This study was funded by the National Key Research & Development Program of China (2016YFC0502002) and Jiangsu Collaborative Innovation Center of Technology and Material of Water Treatment. We greatly appreciate the anonymous reviewers for the insightful comments that improved this manuscript greatly.
CONFLICTS OF INTEREST
The authors declare they have no competing financial interests.
Open Research
DATA AVAILABILITY STATEMENT
All data generated or analyzed during this study are included in this article.
