Introduction

Land use and land cover changes are considered as the main components and a primary cause of global environmental changes (Turner et al. 1995). Changes in land use lead to changes in land cover, which affect the provision of ecosystem services and wildlife habitat (Lawler et al. 2014). Comprehensive knowledge of these dynamics may be useful to reconstruct past land cover changes and also predict future changes (Hietel et al. 2004). Specifically, conversion of native vegetation to other ecosystems is widely reported as an important disturbance, which may result in changes in soil fertility and plant growth. Different soil status can alter the structure and dynamics of soil microbial communities (Ding et al. 2013). Furthermore, such changes can lead to a shift of the soil–atmosphere carbon balance leading to further climatic change (Fernandez and Kennedy 2015). Together, land cover change can affect soil physical, chemical and biological properties (Braimoh and Vlek 2004; Ross et al. 1999; Zeng et al. 2009; Zhao et al. 2013).

The Qinghai–Tibetan Plateau (QTP) ecosystem is more fragile and sensitive to global climate change and anthropogenic disturbance than are other ecosystems because of its high elevation (Wang et al. 2007; Yang et al. 2006). Alpine meadows, the representative vegetation of Tibetan Plateau, cover approximately 6.37 × 105 km2 (Ni 2002), comprising 15.3 % of the cold region (Yang et al. 2013). The alpine meadow ecosystem is of special interest, because its soils bear a larger organic carbon pool than grassland soils in other regions of China or tropical savannah soils (Genxu et al. 2008). Colder soils have been shown to carry higher organic C loads than warmer soils (Mcdaniel and Munn 1985). However, there had been dramatic changes of vegetation cover in alpine meadows due to overgrazing, human activities and rodent damage (Zhou et al. 2005). The changes of land cover not only alter ecosystem communities, but also alter the nutrient of the soil and the activities of soil microorganism (Wang et al. 2006a; Zhang et al. 2013). Hence, ecosystem alteration causes changes in C and N cycling by altering plant production, rates of soil organic matter accumulation and decomposition, and the subsequent C and N storage in soils (Wang et al. 2009). Meanwhile, the alpine grassland ecosystem bearing large soil C stocks can became a source of atmospheric CO2 if grassland soils become degraded. Huge amounts of C have been emitted from the grassland soils of the plateau due to grassland degradation, which strengthens the trend of global warming (Wang et al. 2002).

Previous studies have reported differences in SOC or TN stocks among typical vegetation types in the northeastern margin of the QTP (Liu et al. 2012). The northeastern margin of the QTP is mainly influenced by the Asian monsoon system and the geomorphological situation is consistent with mountain topography. Unique alternating mountain basin landforms in the area form a series of relative isolated and delimited land use units. With economic development and population growth, the alpine meadow had suffered severe degradation due to natural and anthropogenic activities (Harris 2010). Wang et al. (2003) found that organic matter content and the total N content of the alpine meadow soil decreased by 14,890 and 5505 kg/hm2, respectively, as the vegetation coverage reduced from 90 % to less than 30 %. Meanwhile, Li et al. (2009) showed that mound making by zokors had negative impact on properties and organic matter content of the topsoil at an alpine grassland site of the QTP. On the other hand, Wang et al. (2006b) found that rehabilitation of degraded grassland ecosystems promoted above-ground biomass, particularly grass biomass, and ground cover, which could increase soil organic C and N stocks. The potential for soil carbon and nitrogen storage in alpine meadow soils after land cover change did not appear to have been widely investigated.

Soil microorganisms control ecosystem processes such as C storage and nutrient recycling (Talbot et al. 2014). They play key roles in ecosystem functioning and in the sustainability of soil health. Microbial-based indicators, such as microbial biomass and community structure, can be useful ecological indicators of stress caused by anthropogenic activities (Wardle 1992). While the fluctuations of the soil microbial biomass pools are thought to be important drivers in affecting soil carbon and nutrient cycling, community-level physiological profiling can effectively distinguish spatial and temporal changes in microbial communities (Garland and Mills 1991). Changes in microbial communities often precede changes in the health and viability of the whole environment. Although the Biolog assay has been used to assess the metabolic characteristics of microbial communities (Zak et al. 1994), it has several limitations (Nannipieri et al. 2003). For example, it cannot reflect the potential metabolic diversity in situ and detects only culturable bacteria. However, it is useful for studying the functional diversity of soil bacterial communities in different habitats (Jiang et al. 2010).

In this study, land cover mosaics of alpine meadow, artificial grassland and denuded land were characteristic of the alpine areas on the Tibetan Plateau in Qinghai Province, northwest China. Soil characteristics and microbial properties were compared among three land cover types in an alpine meadow on the Tibetan Plateau. The specific objectives within these three land cover types were: (1) to compare soil carbon and nitrogen stocks of topsoil and (2) to evaluate the variations of soil microbial properties in an alpine meadow. The results provided key data for the biogeochemical characteristics in the protection of alpine grassland ecosystems of the Qinghai–Tibetan Plateau, China.

Materials and methods

Site characteristics

The study was conducted at the Qilian Alpine Ecology and Hydrology Research Station, Chinese Academy of Sciences (Fig. 1). The study area was located at latitude 38°15′N and longitude 99°52′E in the northeast part of the Qinghai–Tibetan Plateau. The average altitude was about 3000 m above sea level. The annual air temperature was approximately 0.2 °C, with the highest mean monthly temperature reaching 19.0 °C in July and the lowest reaching −18.4 °C in January. The mean annual precipitation was 495.1 mm, with 80–93 % of the precipitation occurring from June to October. Native vegetation was mainly alpine Kobresia meadow. The soil type in the study area was mainly classified into mattic cryic cambisoils (Chinese soil taxonomy: alpine meadow soil; FAO: cambisoils). The most significant characteristic was that there were mattic epipedon in alpine meadow soil.

Fig. 1
figure 1

Map of the location of the study area and the set of sampling sites

Full size image

Experimental design, soil sampling and laboratory analysis

Three land cover types, alpine meadow, artificial grassland and denuded land, were selected for this study. Three plots (30 × 40 m) were selected for each land cover type. The plots with different land cover types were located as close as possible to each other, in plates with comparable topography and soil and vegetable types (which appeared to be present before land cover change). The topography of the study area was undulating with flat to almost flat zones. Due to alteration of original thick sod layer of depth 0–20 cm during land cover conversion, the study about the effect of soil properties and soil microbial activity was carried out at 0–20 cm depth.

  1. 1.

    The native alpine Kobresia meadow was undisturbed as control and contained approximately 20 species that together made up 80 % of the total vegetation cover. The dominant species were Kobresia capillifolia and Kobresia humilis, and the main accompanying species were Festuca ovina, Stipa capillata, Potentilla fragarioides, Elymus nutans, Poa annua, Leontopodium nanum, Gentiana lawrencei, Achnatherum splendens, Saussurea superba, and Carex sp. The soil surface was well developed with a mixture of root mass and the root systems were densely distributed with rich organic matter.

  2. 2.

    The mound-shaped denuded land was heavily degraded grassland with little or no vegetation. This was representative of secondary bare land or heavily damaged turf landscape that occurs after degradation of the original vegetation (Zhou et al. 2005). The severely degraded grassland was characterized by low vegetation cover and the loss of the upper soil horizons. The mounds were mainly made of Myospalax fontanierii and Marmota himalayana. They made mounds by digging, wriggling and mixing soil while foraging the plants. All selected mounds were intact since their origin and had a size of about 0.16 m2. On each replicate, five soil slices of mound surface (10 × 5 cm in area and 20 cm in depth) were randomly collected and mixed.

  3. 3.

    The artificial grassland was a field of oat (Avena sativa), which had been converted from alpine meadow 6 years before. The oat stubble coverage was about 25 %. Crops were routinely planted in mid-May and harvested in late September; thus land was utilized 4 months per year. In this area, oats were cultivated as forage and hay crop without mechanization or pesticide treatments. To establish artificial grassland, farmers sow oat seeded with a manual dibble and did not apply fertilizers. After harvest, the oat stubble was left in the field and plowed into the soil as manure.

Soil samples were collected from the three land cover types in January 2013, when the oat crop had been harvested and rodents had the slowest digging activities. In each plot, the litter layer was removed and then five soil cores (3.8 cm diameter × 20 cm long) were randomly collected from the 0–20 cm layer and mixed to obtain a homogenous sample. A portion of each soil sample was air dried for analyses of soil properties. The remaining portion of each sample was stored at 4 °C until soil microbial biomass and Biolog analyses.

Analyses of soil properties

Soil moisture content (SMC) was determined by drying the soils at 105 °C for 24 h. Soil pH was measured in a soil:water mixture (1:2.5 w/v) using a PB-21 pH detector (Sartorius, Göttingen, Germany). Soil electrical conductivity (EC) was measured in a soil:water mixture (1:5 w/v) using a DDSJ-308A electrical conductivity analyzer (Shanghai Rex Instrument Co. Ltd., Shanghai, China). Soil bulk density (BD) was measured for the soil cores using a cutting ring (volume, 100 cm3); this allowed us to estimate the density of soil organic carbon and total nitrogen at each site. Soil total organic C (SOC) was determined by the dichromate oxidation method. Total nitrogen (TN) was determined by the semimicro Kjeldahl method. To measure total phosphorus (TP), the soil was digested with HF–HClO4 and then TP was determined by the molybdenum blue colorimetric method. Available phosphorus (AP) was measured after extraction by the molybdenum antimony resistance colorimetric method (Lu 1999). To measure NH4+-N and NO3−-N, 10 g dry weight of soil samples was suspended in 50 ml of 2 mol/L KCL solution. After shaking at room temperature for 1 h and subsequent standing for 30 min, the supernatant was filtered through a filter paper. NH4+-N and NO3−-N were analyzed by a continuous flow analyzer (SKALAR, Dutch).

Determinations of SOC and TN density

The gravel content (>2 mm) of alpine meadow was approximately 10 % (C AM = 0.10), of artificial grassland approximately 5 % (C AR = 0.05) and of mound-shaped denuded land approximately 15 % (C DL = 0.15). The topsoil thickness was 20 cm. SOCD and TND were calculated density for topsoil profile (T = 20) using the following equations (Yang et al. 2008):

$$\;{\text{SOCD }} = \, T \times {\text{BD }} \times {\text{SOC }} \times \frac{{(1 {-}\,C)}}{100},$$
(1)
$${\text{TND }} = \,\,T \times {\text{BD }} \times {\text{TN }} \times \frac{{(1 {-}\,C)}}{100} ,$$
(2)

where SOCD and TND are SOC density (kg/m2) and TN density (kg/m2), respectively; and T, BD, SOC and C are soil thickness (cm), bulk density (g/cm3), soil organic carbon (g/kg) and volume percentage of the >2 mm fraction, respectively. TN density of topsoil was described analogously.

Analysis of microbial biomass C and N

Soil microbial biomass C and N (SMBC/SMBN) were determined using the chloroform fumigation extraction method (Brookes et al. 1985; Vance et al. 1987). A K EC of 0.45 (Vance et al. 1987) and a K EN of 0.54 (Brookes et al. 1985) were used to account for the extraction efficiency of C and N, respectively. The values were calculated using the following equations:

$${\text{SMBC }} = \frac{{E_{\text{C}} }}{{K_{{{\text{EC}} }} }},$$
(3)
$${\text{SMBN }} = \frac{{E_{\text{N}} }}{{K_{{{\text{EN}} }} }} ,$$
(4)

where E C and E N are the differences between the C and N concentrations in fumigated and non-fumigated extracts and \(K_{{{\text{EC}} }} = 0. 4 5 {\text{ and }}K_{{\text{EN}} }= 0. 5 4.\)

Soil microbial metabolic diversity

Biolog-ECO microplates (BIOLOG Inc., Hayward, CA, USA) were used to determine the metabolic diversity of microbial communities following a procedure adapted from Garland and Mills (1991). The 31 substrates represented six classes of organic compounds: amides/amines, amino acids, carbohydrates, carboxylic acids, polymers and miscellaneous substrates (Insam 1997).

Briefly, fresh soil samples (10 g fresh weight) were serially diluted to a 10−3 suspension in buffer (0.85 % NaCl; w/v) before inoculation onto Biolog plates (150 μL soil suspension per well). The plates were incubated at 25 °C in the dark and scanned every 12 h at 590 nm for 8 days using an Emax precision microplate reader (BIOLOG Inc.). The data were analyzed by Microlog Release 4.2 software (ML3420, Microlog, USA). The absorbance values recorded at 96 h were used to calculate the diversity indices and in the principal component analysis (PCA). In this study, the average well color development (AWCD) was calculated as follows:

$${\text{AWCD }} = \,\;\frac{{\mathop \sum \nolimits (C_{\text{i}} - R )}}{31},$$
(5)

where C i refers to the 31 absorbance values of the different C sources and R is the absorbance value of the control well. The absorbance values recorded at 96 h were used to calculate the Shannon–Wiener diversity index (Zak et al. 1994) as follows:

$$\,H\, = \, - \mathop \sum \nolimits P_{\text{i}} \times \,\ln P_{\text{i}} ,$$
(6)
$$P_{\text{i}} \, = \,\frac{{n_{\text{i}} }}{N}.$$
(7)

Statistical analysis

All statistical analyses were carried out with SPSS 11.5 software for Windows. All data were shown as mean values ± standard error. One-way analysis of variance (one-way ANOVA) was used to calculate the statistical significance. The least significant difference (LSD) test was performed for post hoc multiple comparisons if the data were homogeneous, or otherwise Tamhane’s T2 test was performed. A Pearson’s correlation analysis was carried out among variables using the bivariate correlations procedure. Principal component analyses (PCA) were conducted on the BIOLOG data to investigate the differences in soil microbial metabolic diversity of three land cover types. Differences were considered to be significant if P was <0.05.

Results and discussion

Soil characteristics

The physical and chemical properties of the tested soils in the top 20 cm were listed in Table 1. Artificial grassland had much lower soil pH (P < 0.05), while denuded land had higher values compared to the reference alpine meadow. However, no significant differences in soil pH were found between denuded land and alpine meadow. Soil pH values of >7 indicated alkaline conditions at most sampling sites in the study area. Lower soil moisture content was found for denuded land than for alpine meadow and artificial grassland (P < 0.01) due to less coverage and higher evaporation. Soil electrical conductivities were greater in artificial grassland than in alpine meadow and denuded land, whereas they did not differ between artificial grassland and denuded land (P ≥ 0.05). Generally, the average soil electrical conductivities were much greater in artificial grassland and denuded land than in alpine meadow as a result of greater evaporation rates from more bare soil surfaces, which facilitated the upward flow of soluble salts from deep layers with groundwater, thus increasing soil electrical conductivity. Average soil bulk density was greater in alpine meadow than in artificial grassland and denuded land (P < 0.05). SOC in different land cover types also followed the order: alpine meadow > artificial grassland > denuded land, even though there were no significant differences in SOC between alpine meadow and artificial grassland.

Table 1 Soil characteristics of the topsoil under different land cover types
Full size table

Soil characteristics provided a medium for root development and moisture and nutrient for plant growth. Alteration of original alpine meadow to the other land cover types significantly changed vegetation composition and soil status. Especially in the cold season, it was not covered with vegetation, which left the soil bare and prone to erosion (Wu et al. 2010). Conservation of the original alpine grassland brought about loss in SOC in Northern China (Wu and Tiessen 2002). In this study, the average SOC in artificial grassland and denuded land decreased 9.54 and 30.20 %, respectively, compared to alpine meadow plots. In general, perennial plants (Kobresia capillifolia and Kobresia humilis) stored more C in soil than annuals (Avena sativa), because the annual cycle of cultivation did little to maintain storage because perennial crops were in the ground longer and intercepted more of the sun’s radiation than annuals (Whitmore et al. 2014). On the other hand, the constant additions of below-ground biomass by perennials roots was also attributed to the SOC accumulation in alpine meadow plots. Comparatively, foraging and mounding creating by rodent animals significantly reduced the below-ground plant biomass (Li et al. 2009). Therefore, more carbon would be released into the atmosphere in denuded land plots than the other land cover types (Wang et al. 2010).

Abundance of nitrogen and phosphorus

For three land cover types, soil total nitrogen in the alpine meadow was higher than in the artificial grassland and denuded land. Soil total nitrogen contents in alpine meadow and denuded land soil showed significant differences (P < 0.05) (Table 2). Only subtle changes in soil total phosphorus contents were observed among the three land cover types. Soil NH4+ showed similar tendencies. However, artificial grassland soil contained higher soil NO3− and available phosphorus than the other land cover types. This could be explained by the fact that soil inorganic nutrients were released back into soil due to soil microbial decomposition of oat residues.

Table 2 Abundance of nitrogen and phosphorus
Full size table

Inorganic N (NO3− + NH4+) concentration ranged from 19.94 to 45.11 mg N kg/soil. There was a trend toward higher inorganic nutrient concentrations in plots with more existent dead plant roots, though no significant differences were observed in soil TN between alpine meadow and artificial grassland plots. Thus, short-term relatively small amendment applications could alter soil biological processes and nutrient cycling (Ninh et al. 2015).

Carbon and nitrogen stocks

Variations in C and N stocks exhibited similar tendencies with C and N contents (Fig. 2). As values of bulk density were distinct at 0–20 cm depth across land cover types (Table 1), gaps of C and N contents in topsoil were narrowed as compared to C and N stocks. SOC and TN stocks presented declines when original alpine meadow changes. The SOCD and TND in artificial grassland soils were characterized by a slight drop within a 14.5 and a 10.9 % decrease, respectively, as compared to original alpine meadow soils. Meanwhile, a great decrement in SOCD and TND was observed in denuded land soils as compared to alpine meadow soils, within a 52.9 and a 51.7 % decrease, respectively. The data indicated a greater decline in SOCD and TND of denuded land spots.

Fig. 2
figure 2

Carbon and nitrogen stocks of the three land covers

Full size image

The amount of C retained by soils was influenced greatly by the rate of input of organic matter and its decomposition by biological or chemical means (Whitmore et al. 2014). Losses of SOC and TN storage were subjected to land cover conversion in topsoil, decreasing 14.5 and 10.9 % of alpine meadow to artificial grassland and 52.9 and 51.7 % to denuded land. Guo and Gifford (2002) showed that a change from perennial grassland to arable land was likely to lead to a significant loss in soil C. The development of artificial grasslands in place of current perennial plants might lead to loss in soil C, which was consistent with the results of Qin et al. (2014). The activities of rodents accelerated erosion and the degradation rate by loosening the Kobresia sod and killing its roots (Zhou et al. 2005). This result coincided with the finding that highlighted mound making had negative impact on soil properties and organic matter content of the topsoil (Li et al. 2009).

Soil microbial biomass C and N

Although microbial biomass represented <5 % of SOM, the contribution of microbial-derived compounds to SOM was potentially much greater (Grandy et al. 2009). SMBC and SMBN differed among land cover types (Fig. 3). SMBC values in the alpine meadow were higher than in denuded land (P < 0.05), whereas they did not differ between alpine meadow and artificial grassland (P ≥ 0.05). SMBN in the alpine meadow was significantly higher than in the artificial grassland and denuded land (P < 0.05). A series of bivariate correlations analyses showed that soil MBC and MBN were strongly related to soil bulk density and soil organic carbon (Table 3). Microbial biomass was a sensitive indicator of soil disturbances caused by land cover changes (Melo et al. 2012). Many factors were considered to elucidate the effects of vegetation types on soil microbial biomass (Wardle 1992). The quantity and quality of substrates via leaf litter inputs and root exudations influenced soil carbon and nutrient availability. Soil carbon and nutrient availability were important drivers that affected soil microbial biomass (Wardle 1992), as SMBC was proved to have positive correlations with soil organic carbon and total nitrogen (Table 3). Correspondingly, higher SMBC and SMBN in the original alpine meadow than in the artificial grassland and denuded land were mainly attributed to higher soil organic carbon and total nitrogen. Furthermore, soil moisture might play an important role in influencing microbial biomass, as soil moisture was positively correlated with microbial biomass C. Less soil moisture in the denuded land than the other land cover types could suppress the growth and activity of soil microbial communities. Soil microorganisms could adopt a survival strategy for coping with harsh environmental conditions, such as sporulation and cell shrinkage (Jones and Lennon 2010). This result provided evidence for a shift of microbial biomass among three land cover types.

Fig. 3
figure 3

Soil microbial biomass C and N of the three land covers

Full size image
Table 3 Correlation between soil microbial characteristics and selected soil physicochemical properties
Full size table

Community-level physiological profiles (CLPP) analysis

The AWCD values obtained from Biolog Ecoplates could be used to reflect the microbial metabolic activity (Zheng et al. 2005). The highest AWCD values were obtained from alpine meadow soils and the lowest from denuded land soils. All of the AWCD values were low after 24 h incubation, indicating that the carbon sources were not fully used during that period. However, the AWCD values for alpine meadow soil increased rapidly after 24 h, while those for artificial grassland and denuded land soils increased more slowly (Fig. 4). The three land cover types could be ranked in terms of their Shannon–Wiener diversity index (H) values as follows: denuded land < artificial grassland < alpine meadow.

Fig. 4
figure 4

Average well color development (AWCD) of soil microbial communities with incubation times

Full size image

The results of the PCA of Biolog data are shown in Table 4 and Fig. 5. The first principal component (PC 1) explained 33.68 % of the variation and the second (PC 2) explained 20.19 % of the variation (eigenvalues 10.44 and 6.26, respectively). Axis 1 and 2 explained 53.87 % of the variation in soil microbial community metabolic diversity. For three land cover types, the PC 2 could not clearly separate the denuded land soil samples from the alpine meadow soil samples. However, PC 1 clearly divided the substrate utilization patterns of the microbial community between alpine meadow and artificial grassland. The variability in PC1 was explained by the different use of carbohydrates, carboxylic acids, amino acids, miscellaneous substrates and polymers (β-methyl-d-glucoside, d-xylose, 2-hydroxy benzoic acid, α-cyclodextrin, N-acetyl-d-glucosamine, γ-hydroxybutyric acid, l-threonine, glycogen, glycyl-l-glutamic acid, d-cellobiose, glucose-1-phosphate, α-ketobutyric acid, α-d-lactose, d-malic acid). The variability in PC 2 was also explained by different use of carbohydrates, carboxylic acids, amino acids, polymers and miscellaneous substrates (d-galactonic acid γ-lactone, d-galacturonic acid, l-asparagine, Tween 80, d-mannitol, 4-hydroxy benzoic acid, l-serine and d-Glucosaminic acid) (loading values of principal components >0.600) (Table 4). The use of carbohydrates by microorganisms was significantly greater in the alpine meadow than in artificial grassland and denuded land (P < 0.05) (Table 5).

Table 4 Loading values of principal components of 31 sole-carbon sources
Full size table
Table 5 Categorized substrate utilization pattern by microbial communities of the three land covers after 96 h of incubation
Full size table

The differences in the carbon source metabolism activities of soil microorganisms under different land cover types were mainly caused by soil properties (Chakraborty et al. 2011). As denuded land soils had a limited nutrient content due to less organic matter input, an increase in soil fertility could promote the soil microbial community metabolism activities (Islam et al. 2011). A large amount of SOC could provide C resources to microbes and promoted C utilization (Zhao et al. 2013). Alpine meadow soil had dense root systems, while artificial grassland and denuded land soil had loose and had sparse root systems. The root distribution and its activities might accelerate soil nutrient translation and promoted the soil metabolic activity of soil microbial communities. For example, retaining crop residues in the field and changed in soil organic matter could also affect the metabolic diversity of soil microbial communities (Bending et al. 2002). The spatial heterogeneity of soil factors such as C/N ratio and pH could affect soil microbial functional diversity (Zhang et al. 2013). In the present study, soil electrical conductivity might play an important role in influencing microbial metabolic activity, as soil electrical conductivity was positively correlated with microbial metabolic activity. Salinity (soil electrical conductivity) inhibited microbial activities by affecting the availability of water or the physiological and metabolic processes of the microbial cells (Jannike et al. 2006). Alteration of the original alpine meadow to the other land cover types could reduce the resistance of microbial communities to environmental stress or disturbances (Degens et al. 2001). These results demonstrated that soil microbial community activities in soil exhibited obvious differences among the three land cover types. However, the CLPP values represented only a subpopulation of culturable aerobic microorganisms (Chakraborty et al. 2011). Future research is required to evaluate the structure of microbial communities using genetic analyses.

Fig. 5
figure 5

Principal component analysis (PCA) of Eco-Biolog plate profiles from microbial communities of soil samples (96 h)

Full size image

Conclusions

This paired study for adjacent fields with three land cover types revealed specific differences in soil properties and soil microbial activity. Land cover conversion from native alpine meadow to artificial grassland and denuded land resulted in visible SOC and TN losses in topsoil. Alpine meadow soil had the highest soil microbial biomass and microbial metabolic activity of the three land cover types. The destruction of soil properties, losses of nutrients and rooting layer, owing rodents damaging, could cause long-term damage of land productivity. This study provided implications for assessing effects of land cover conversion on soil C and N stocks and soil microbial properties, and highlighted the importance of protecting alpine meadow ecosystem.