Ecological stoichiometry in leaves, roots, litters and soil among different plant communities in a desertified region of Northern China

Catena 166 (2018) 328–338

Contents lists available at ScienceDirect

Catena journal homepage: www.elsevier.com/locate/catena

Ecological stoichiometry in leaves, roots, litters and soil among different plant communities in a desertified region of Northern China Yang Yanga, Bing-Ru Liub, Shao-Shan Ana,c,

T

⁎

a

College of Natural Resource and Environment, Northwest A&F University, Yangling 712100, China Key Lab of Restoration and Reconstruction of Degraded Ecosystem in Northwest China of Ministry of Education, Ningxia University, Yinchuan 750021, China c State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau, Northwest A&F University, Yangling 712100, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: Ecological stoichiometry Leaves Roots Litters Soil Desertified region

Ecological stoichiometry reflects the element content and energy flow, which are important for biogeochemical cycling in ecosystems. However, the ecological stoichiometry in leaves, roots, litter and soil is largely unknown, especially in the desertified region of Northern China. Here, six dominant plant communities (Stipa bungeana, Agropyron mongolicum, Glycyrrhiza uralensis, Cynanchum komarovii, Artemisia ordosica, and Sophora alopecuroides) were collected, and the carbon (C), nitrogen (N) and phosphorus (P) contents of leaves, roots, litters and soil were measured to explore the C:N:P stoichiometry and its driving factors. The C:N:P stoichiometry in leaves, roots, litters, and soil varied widely, and the plant community had a significant effect on the C:N:P stoichiometry in this region. There were high soil C:N, C:P and N:P ratios in non-leguminous plant communities and a high leaf N:P ratio in leguminous plant communities, and the C:N and C:P ratios in leaves were higher than in those in roots in all plant communities (p < 0.05). A correlation analysis showed that the C, N and P contents of leaves, roots, and litter were positively related to the soil C, N and P contents of the 0–5 cm layer, and the correlation coefficients gradually weakened with the soil depth. Additionally, the soil properties (except soil P) led to increased variance of the C:N:P stoichiometry in leaves, roots, and litter, and there were strong links among the C:N:P stoichiometry in leaves, roots, litter and soil, suggesting that the variation in the C:N:P stoichiometry in leaves, roots, and litter was mainly controlled by the soil properties, which was especially true for soil microbial biomass carbon (SMBC) and nitrogen (SMBN) according to redundancy analysis (RDA). Overall, these results demonstrate that the patterns of the C:N:P stoichiometry and element distribution exhibit significant flexibility among these plant communities, providing basic data for improving the parameterization of future ecological models in the desertified region of Northern China.

1. Introduction In terrestrial ecosystems, the balance of nutrient elements in interactions and processes is known as ecological stoichiometry (Li, 2001; Sterner and Elser, 2002; Moe et al., 2005). Ecological stoichiometry represents an organism's demand for natural resources and connects different levels of biogeochemical cycling (Bradshaw et al., 2012; Hu et al., 2018), mainly by scaling up carbon (C), nitrogen (N) and phosphorus (P) (Bai et al., 2012; Yoshihara et al., 2010). Specifically, C provides the structural basis of plants, constituting a relatively stable 50% of the dry plant biomass (Schade et al., 2003; Liu et al., 2011). N is an important constituent of proteins and plays an essential role in plant production, photosynthesis and litter decomposition (Daufresne, 2004; Chen et al., 2016). P is often regarded as the limiting element and is responsible for the cell structure and composition of DNA and RNA, and

⁎

P promotes C/N transpiration and assimilation (Tilman, 1994; Tilman, 1996; Naeem and Li, 1997; Tilman, 2004; Bai et al., 2012). In general, C:N:P stoichiometry can be used to detect nutrient limitations to characterize important ecological processes in terrestrial ecosystems (Koerselman and Meuleman, 1996; Cleveland and Liptzin, 2007; Hättenschwiler and Jørgensen, 2010). There are two important methods of detecting the C:N:P stoichiometry (Frost et al., 2002; Cross et al., 2005). First, the release of key nutrients, such as C, N, and P, occurs in a predictable way depending on the limitation of the C:N:P ratio (Elser et al., 1988; Sterner and Elser, 2002). For example, the leaf N:P ratio (mass ratio) has been suggested to be useful for assessing N or P limitation (Cleveland and Liptzin, 2007). Second, natural resources are in stoichiometric homeostasis (Hessen, 1992; Urabe and Watanabe, 1992; Sterner and Elser, 2002). At the organismal level, life stages characterized by high growth rates are often associated with high RNA

Corresponding author at: Institute of Soil and Water Conservation, Northwest A&F University, 26 Xinong Rd., Yangling, Shaanxi 712100, China. E-mail address: [email protected] (S.-S. An).

https://doi.org/10.1016/j.catena.2018.04.018 Received 27 September 2017; Received in revised form 1 April 2018; Accepted 13 April 2018 Available online 21 April 2018 0341-8162/ © 2018 Published by Elsevier B.V.

Catena 166 (2018) 328–338

Y. Yang et al.

Fig. 1. Location of the study area and layout of the modified Whittaker plots (MWP) for the measurement of plant and soil samples in a desertified region of China.

region is an ideal region for addressing the drawbacks of the exploring C:N:P stoichiometry at a plant community level. To date, attention has been paid to C:N:P stoichiometry in this region (Liu et al., 2010; Zuo et al., 2012; Chen et al., 2016). However, no comprehensive survey of the driving factors of C:N:P stoichiometry in a plant-litter-soil system has been performed. Here, we examine the C:N:P stoichiometry and its driving factors across the desertified region in Ningxia, China. Specifically, this study had the following objectives: (1) explore the relationships among the C:N:P stoichiometry in the plant-litter-soil system; (2) examine the effect of the plant community on the C:N:P stoichiometry in the plant-litter-soil system; and (3) determine the driving factors of the C:N:P stoichiometry in this region.

contents (C:N:P ratios between 18:6:1 and 21:7:1), and therefore, organisms are able to shift their overall stoichiometry (McGroddy et al., 2004). Under this condition, to what extent do plants exhibit stoichiometric homeostasis? Soil C, N and P are affected by organic matter, litter and microbes (Mulder and Elser, 2009; Sinsabaugh et al., 2008), and plants adjust their growth rates by adjusting the ratio of C, N and P (Daufresne and Loreau, 2001; Moe et al., 2005). Litter stores nutrients and plays an important role in element cycling (Melillo et al., 1982; Manzoni et al., 2008); thus, the balance of the C, N, and P ratios is highly complex in a plant-litter-soil system (Moe et al., 2005; Manzoni et al., 2008; Manzoni et al., 2010). Therefore, evaluating the C:N:P stoichiometry in a plant-litter-soil system could improve our understanding of plant nutrient limitations and ecosystem dynamics (Redfield, 1958; Sterner and Elser, 2002). Recently, most studies on C:N:P stoichiometry have focused on the organ level, and only a few studies have been conducted at the plant community or ecosystem levels in grasslands (He et al., 2006; He et al., 2008; Wang and Yu, 2008). A plant community is formed as a result of the adaptation of species to a specific environment and through mutual competition. Thus, the function of C:N:P stoichiometry should be explored at the community/ecosystem level in grasslands. A desertified

2. Materials and methods 2.1. Study area This study was conducted in the middle of the desertified region of East Ningxia (37°04′–38°10′N and 106°30′–107°41′E, average elevation 1450 m). This region has a temperate continental semiarid monsoon climate. The mean annual precipitation in this region is 180–300 mm, 329

Catena 166 (2018) 328–338

Y. Yang et al.

Table 1 The basic soil characteristics among different plant communities in a desertified region. Plant communities

pH

S. bungeana A. mongolicum G. uralensis S. alopecuroides A. ordosica C. komarovii

8.13 8.16 8.53 8.69 8.62 8.91

± ± ± ± ± ±

1.06 0.85 1.26 1.05 0.84 1.17

c c b b b a

EC/(μs·cm−2)

SW/(%)

105.63 123.85 121.47 156.98 145.32 162.31

9.25 8.13 8.02 7.15 7.06 7.08

± ± ± ± ± ±

5.62 d 9.54 c 8.13 c 7.41 b 9.06 b 12.37 a

± ± ± ± ± ±

BD/(g·cm−3) 1.56 1.03 1.74 0.98 0.87 1.46

a b b c c c

1.23 1.46 1.58 1.67 1.62 1.59

± ± ± ± ± ±

0.15 0.26 0.24 0.19 0.34 0.27

d c b a ab b

AP/(mg·kg−1)

AN/(mg·kg−1)

1.59 1.23 1.02 1.01 0.98 0.76

35.69 31.02 32.47 26.45 23.14 22.79

± ± ± ± ± ±

0.61 0.25 0.34 0.23 0.19 0.13

a b c c c d

± ± ± ± ± ±

3.06 2.15 4.18 2.87 4.01 2.98

a b b c d d

SMB-C/(mg·kg−1)

SMB-N/(mg·kg−1)

321.47 302.56 265.41 286.72 235.98 279.74

36.02 34.14 32.05 26.23 23.42 22.18

± ± ± ± ± ±

42.36 32.54 46.98 35.02 39.78 41.03

a a c b d bc

± ± ± ± ± ±

3.03 2.98 3.14 1.56 2.87 2.01

a ab b c d d

SW: soil water, BD: bulk density, EC: electric conductivity, pH: pH value, AP: available phosphorus, AN: available nitrogen, SMB-C: soil microbial biomass carbon, SMB-N: soil microbial biomass nitrogen. Values followed by lowercase letters within columns are significantly different at p < 0.05 using the LSD method, n = 15. The same is true below.

dug up all plants of these six plant communities and separated the roots and leaves (1 m2). We also collected the litter from the ground, which was brought to the laboratory to analyze the C, N and P contents of each quadrat. Plant samples were then washed and dried at 60 °C for 48 h. The resulting samples were crushed and sieved through a 0.5–1.0 mm mesh for further analysis. Afterwards, five replicates were randomly chosen for soil collection from three layers (0–5 cm, 5–10 cm, 10–15 cm). Three cores (diameter of 5 cm) from each quadrat were collected along a diagonal line and then thoroughly mixed to form one composite sample. The soil samples were sieved through a 2 mm mesh, and visible roots and organic debris were separated by hand. All samples were ground to a fine powder using a ball mill (MM400 Ball Mill, Retsch, Germany) and an agate mortar grinder (RM200, Retsch, Haan, Germany) for element analysis. We measured total soil C (TC, g·kg−1), total soil N (TN, g·kg−1) and total soil P (TP, g·kg−1) according to the National Standard Methods of the People's Republic of China. Soil cores were oven-dried to determine the soil bulk density (BD, g·cm−3) and water contents (SW, %). The soil pH and soil electrical conductivity (EC, μs cm−2) were determined from a 1:5 mixture of soil and deionized water using a PHS-3C digital pH meter and a DDS-307A conductivity meter (Precision and Scientific Corp., Shanghai, China), respectively. TC and TN were measured with a CNS analyzer (Vario EL, Heraeus); TP and available phosphorus (AP, mg·kg−1) were measured by the molybdenum antimony colorimetric method. Alkali-hydrolyzable nitrogen (AN, mg·kg−1) was measured by NaOH-H3BO3 (Liu et al., 2011). The soil microbial biomass C and N (SMB-C, SMB-N, mg·kg−1) were measured by the fumigation-extraction method (Vance et al., 1987). In addition, changes in the soil properties among different plant communities are presented in Table 1. Plant C (g·kg−1) and N (g·kg−1) were also determined by the CNS analyzer (Vario EL, Heraeus). Before measuring P (g·kg−1), plant samples were acidified with 68% HNO3, while soil samples were acidified with HNO3 and HF for 12 h. Samples were then digested with a microwave digestion system (Mars X press Microwave Digestion system, CEM, Matthews, NC, USA). P in all plant samples was analyzed by an inductively coupled plasma optical emission spectrometer (ICPOES, Optima 5300 DV, Perkin Elmer, Waltham, USA). All of the above measurements were performed in triplicate and averaged.

which largely occurs over 7–9 months, and the mean annual potential evaporation is 1221.9–2086.5 mm at the southwestern fringe of the Mu Us sandy land in Ningxia, China. Seventy percent of the total precipitation occurs between June and September. The mean annual temperature is 7.6 °C, with years of accumulated temperature of 2944.9 °C, and the lowest and highest monthly mean temperatures are 8.7 °C in January and 22.4 °C in July, respectively. The annual frost-free period is approximately 138 days. The mean annual wind velocity is 2.8 m·s−1, and the dominant winds are southwest to south in summer and autumn and northwest in winter and spring. Sand dust blows at velocities of over 5.0 m·s−1 over an average of 323 times per year. Wind erosion often occurs from April to mid-July before the rainy season starts (the climate data are from Yanchi Meteorological Station, 1976–2010). The zonal soil in this region is desertification sierozem according to Chinese Soil Taxonomy (Cooperative Research Group on Chinese Soil Taxonomy, 2001) or Calcaric Cambisol according to the FAO-UNESCO Soil Map of the World (FAO and ISRIC, 1988); the soil in this region developed directly from parent wind-deposited yellow material. In addition, vegetation in this region shows drought-resistant characteristics. The plant community is mostly composed of xerophytic herbaceous species, and the plant leaves generally have obvious xerophytic morphological characteristics. The main plant groups include Gramineae, Asteraceae, Liliaceae, Zygophyllaceae, Leguminosae and Cruciferae. Based on a vegetation investigation, six plant communities, including S. bungeana, A. mongolicum, G. uralensis, C. komarovii, A. ordosica, and S. alopecuroides, were collected to measure the C, N and P contents of the leaves, roots, litter and soil. According to previous studies in this region, these plant communities were fenced after the 1990s, and their coverage is > 60% (Tang et al., 2016). 2.2. Experimental design Plant and soil samples were collected in the summer of 2015. We focused on six plant communities, and each plant community had three sample locations (50 m × 20 m) with similar topography that were at least 1000 m apart from each replicate sample; these locations were chosen to set up a modified Whittaker plot for each plant community (Fig. 1). Nested in each sample was one 200 m2 subplot (10 × 20 m) in the center of the plot and a 15 m2 subplot (5 × 3 m) in opposite corners of the plot. Fourteen quadrats (each quadrat 1 × 1 m) were regularly spaced within each sample (Stohlgren et al., 1999; Zuo et al., 2012), such that two quadrats were along the inner border of each 10 m2 subplot and four quadrats were along the outer border of the central 200 m2 subplot. The number of plant species and coverage were recorded in each quadrat with the letter “S”. In each quadrat, we investigated the morphological characteristics and growth characteristics of the plant community.

2.4. Statistical analysis All variables were described by the mean and standard deviation (SD), and statistical analyses were performed using Excel 2010.0 and SPSS 18.0 for Windows (SPSS, Inc., Chicago, IL). Before applying parametric tests, we tested for the normality and homogeneity of the variances. Multiple comparisons and analyses of variance (ANOVAs) were used to determine the differences among the treatments, and significance analysis was performed with least significant difference (LSD) tests. For all tests, statistically significant differences were assigned at p < 0.05 and p < 0.01. The Pearson coefficient test was used to measure the correlation between each variable. Origin7.5 plotting.

2.3. Data collection For sampling and vegetation collection at the community level, we 330

Catena 166 (2018) 328–338

Y. Yang et al.

Ordination techniques were used to determine the relative contribution of the measured soil variables to the community composition (Lepš and Šmilauer, 2003). The taxon abundance data were first analyzed by detrended correspondence analysis (DCA, length of gradient < 4.0), suggesting that redundancy analysis (RDA) was an appropriate approach. A Monte Carlo permutation test was used to determine the conditional effect of the soil properties with the remaining variables as covariables, the effect of soil variables with the remaining variables as covariables, and the effect of plant-soil-litter ecological stoichiometry with the remaining variables as explanatory covariables, and so on. DCA and partial RDA were carried out using Canoco software for Windows 4.5 (Microcomputer Power, Ithaca, NY). Before RDA, the Hellinger transformation was applied to remove the issue of double zeros in the data matrix and to improve analysis. The data and Monte Carlo reduced model tests with 499 unrestricted permutations were used to statistically evaluate significance.

N:P ratio of G. uralensis and S. alopecuroides in leaves and roots were significantly higher than that in leaves and roots of the other plant communities (p < 0.05), and the N:P ratio of A. mongolicum and S. bungeana in leaves and roots were significantly lower than that in leaves and roots of the other plant communities (p < 0.05). In this study, the average C content of leaves was 434.36 g·kg−1, which is under the global average reported by Elser et al. (1988) (464 g·kg−1). The average N content of leaves was 20.1 g·kg−1, which is above the global average (18.6 g·kg−1). Han et al. (2005) reported that the average P contents of leaves were lower than the global average. From our data, the average P content of leaves was 1.27 g·kg−1, which is under the global average (1.77 g·kg−1) and equivalent with the average in China (1.27 g·kg−1). These results are related to the pattern of plant nutrition and growth rhythm and also support the hypothesis that there is higher N in desertified regions. Further, the C, N and P contents in leaves were higher than in roots.

3. Results

3.2. Ecological stoichiometry in litter

3.1. Ecological stoichiometry in leaves and roots

The ecological stoichiometry and C, N, P contents of litter were greatly different (Fig. 2). The C contents of litter varied from 313.32 to 417.37 g·kg−1, with the order A. ordosica > S. bungeana > C. komarovii > S. alopecuroides > A. mongolicum > G. uralensis. The N contents varied from 8.93 to 18.20 g·kg−1 and the P contents varied from 0.46 to 0.80 g·kg−1, which were ranked in the following order: nonLeguminosae plant communities were significantly higher than Leguminosae plant communities (p < 0.05). The C:N ratios varied from 19.20 to 49.23, C:P ratio varied from 464.29 to 721.04, C:P ratio varied from 11.32 to 37.20, and C:N ratio showed the order A. ordosica > S. bungeana > A. mongolicum > C. komarovii > S. alopecuroides > G. uralensis. The C:P ratio and C:N ratio showed the order S. alopecuroides > G. uralensis > C. komarovii > A. ordosica > S. bungeana > A. mongolicum. In total, the C:N ratio was smaller in nonLeguminosae plant communities than in Leguminosae plant communities and the C:P ratio and C:N ratio were greater in non-Leguminosae plant communities than in Leguminosae plant communities.

According to Table 2, the C, N and P contents had the same variation tendency in leaves and roots, with higher C, N and P contents in leaves. The C content of roots ranked in the following order: A. ordosica > G. uralensis > S. alopecuroides > S. bungeana > C. komarovii > A. mongolicum, while the N contents of roots ranked in the following order: S. bungeana > A. mongolicum > G. uralensis > S. alopecuroides > S. bungeana > A. ordosica > C. komarovii. S. alopecuroides and G. uralensis had higher N contents in leaves and roots than the other plant communities (p < 0.05). The N contents of C. komarovii and A. ordosica in leaves and roots were significantly lower than those in leave of the other plant communities. Additionally, the P contents of G. uralensis and S. alopecuroides in roots were significantly higher than those in roots of the other plant communities (p < 0.05), whereas the N contents of A. ordosica in roots were lowest among the plant communities. In total, the C, N and P contents were ordered C > N > P with leaves > roots, which presented obvious leaf enrichment. However, the P contents in leaves were not significantly different among plant communities (p > 0.05). The C:N:P ratios in different plant components were significantly different. As presented in Table 3, the C:N ratios of A. ordosica and C. komarovii in leaves and roots were significantly higher than those in leaves and roots of the other plant communities (p < 0.05). The C:N ratio of G. uralensis and S. alopecuroides in roots and leaves were significantly lower than that in roots and leaves of the other plant communities (p < 0.05). The C:P ratio of A. ordosica and C. komarovii in leaves and roots were significantly higher than that in leaves and roots of the other plant communities (p < 0.05), and the C:P ratio of S. bungeana and A. mongolicum in leaves were significantly lower than that in leaves of the other plant communities (p < 0.05). Additionally, the

3.3. Ecological stoichiometry in soil The soil C contents of S. bungeana varied from 4.32 to 6.93 g·kg−1, soil N contents varied from 0.43 to 0.78 g·kg−1, and soil P contents varied from 0.20 to 0.47 g·kg−1 (Fig. 3). In the vertical direction, the soil C, N and P contents decreased with the soil depth, which showed obvious soil nutrient “surface-aggregation” (higher nutrients in the surface soil layer). Under the surface soil layer, the soil C, N and P contents decreased dramatically, and the C, N, and P contents of the deep soil layer (10–15 cm) were the lowest. Compared within the soil layers, the soil C, N and P contents ordered: S. bungeana > A. mongolicum > C. komarovii > S. alopecuroides > G. uralensis > A. ordosica, with local fluctuation.

Table 2 C, N and P contents in roots and leaves among different plant communities (mean ± SD). Plant communities

C/(g·kg−1)

N/(g·kg−1)

Roots S. bungeana A. mongolicum G. uralensis S. alopecuroides A. ordosica C. komarovii Mean CV p F

323.11 ± 217.94 ± 436.84 ± 433.45 ± 457.32 ± 387.36 ± 376.00 ± 0.242 < 0.05 89.12

Leaves 42.45 58.76 54.79 43.56 46.58 22.58 91.13

c e b b a d

458.64 ± 446.82 ± 439.37 ± 438.55 ± 435.41 ± 387.36 ± 434.36 ± 0.056 < 0.05 75.36

21.75 19.13 18.37 20.14 17.25 22.58 24.48

a b c c c d

P/(g·kg−1)

Roots

Leaves

10.53 ± 0.78 b 7.53 ± 1.23 c 18.12 ± 2.78 a 18.39 ± 2.54 a 4.94 ± 0.98 d 8.31 ± 1.23 bc 11.30 ± 5.67 0.502 < 0.05 87.14

19.33 ± 16.91 ± 24.59 ± 25.24 ± 15.27 ± 12.32 ± 18.94 ± 0.272 < 0.05 92.26

331

Roots 0.43 0.56 0.37 0.34 0.29 0.41 5.16

b c a a c d

0.82 ± 0.62 ± 1.07 ± 1.03 ± 0.37 ± 0.62 ± 0.76 ± 0.357 < 0.05 73.25

Leaves 0.18 0.13 0.24 0.12 0.08 0.15 0.27

ab ab a a b ab

1.56 ± 1.34 ± 1.53 ± 1.34 ± 1.05 ± 0.82 ± 1.27 ± 0.226 > 0.05 65.47

0.04 0.03 0.01 0.03 0.02 0.02 0.29

a a a a a a

Catena 166 (2018) 328–338

Y. Yang et al.

Table 3 Characteristics of the ecological stoichiometry in roots and leaves among different plant communities (mean ± SD). Plant communities

C:N

C:P

Roots S. bungeana A. mongolicum G. uralensis S. alopecuroides A. ordosica C. komarovii Mean CV p F

30.68 ± 28.94 ± 24.11 ± 23.57 ± 92.57 ± 52.39 ± 42.04 ± 0.640 < 0.05 85.23

Leaves 5.53 c 6.24 c 5.79 d 7.23 d 25.13 a 16.79 b 26.92

23.73 ± 26.42 ± 17.87 ± 17.38 ± 29.69 ± 31.44 ± 24.42 ± 0.242 < 0.05 61.25

1.17 1.54 1.65 1.87 1.58 1.05 5.90

c b d d a a

N:P

Roots

Leaves

394.03 ± 81.32 c 315.52 ± 78.56 d 408.26 ± 93.21 c 420.83 ± 63.79 c 1236.00 ± 105.65 a 702.21 ± 88.74 b 579.48 ± 374.75 0.600 < 0.05 74.89

294.00 ± 333.45 ± 287.17 ± 327.28 ± 431.82 ± 472.39 ± 357.69 ± 0.214 < 0.05 92.26

Roots 28.37 25.12 14.25 13.78 16.54 21.46 76.41

d c d c b a a

12.84 ± 12.15 ± 16.93 ± 17.85 ± 13.35 ± 13.40 ± 14.42 ± 0.164 < 0.05 87.03

Leaves 2.78 3.54 2.87 4.01 3.79 3.15 2.36

b b a a b b

12.39 ± 12.62 ± 16.07 ± 18.84 ± 14.54 ± 15.02 ± 14.91 ± 0.160 < 0.05 76.24

0.23 0.56 0.15 0.75 0.17 0.26 2.39

c c ab a bc b

3.5. Correlations with ecological stoichiometry by RDA

According to Fig. 4, C the contents of the soil varied from 1.49 to 6.01 g·kg−1 and were significantly higher in non-Leguminosae plant communities (p < 0.05). The N contents varied from 0.17 to 0.61 g·kg−1, and the P contents varied from 0.17 to 0.47 g·kg−1. The S. bungeana P contents were significantly higher among plant communities (p < 0.05), while P the contents were not significantly different among plant communities (p > 0.05). The C:N ratio ranged from 5.71 to 14.58, and the C:P ratio ranged from 6.91 to 18.46, with the highest ratio in A. mongolicum. The C:P ratio ranged from 0.81 to 2.28 and was significantly higher in non-Leguminosae plant communities than in Leguminosae plant communities (p < 0.05).

Redundancy analysis (RDA) showed that the soil properties (soil water, bulk density, electric conductivity, pH value, available phosphorus, available nitrogen, soil microbial biomass carbon, and soil microbial biomass nitrogen) explained 97.6% of the total variation in the data, with axes 1 and 2 explaining 80.5% and 9.8% of the total variation, respectively (Table 5). The species-environment relationship for axes 1 and 2 for the soil variables accounted for 90.3% of the total variance. The species-environment correlations for these axes were > 80%, indicating that the species group data were strongly correlated to the soil properties. Further, Monte-Carlo tests revealed that both the first and second axes (p < 0.01) explained a significant amount of the variation within the data. Therefore, the observed distribution of the ecological stoichiometry in leaves, roots, soil and litter could be determine according to the results of the RDA (Fig. 6). The taxonomic group arrow that points in approximately the same direction as the soil properties arrow indicates a strong positive correlation (the longer the species group arrow, the stronger the relationship). The arrows for available phosphorus, available nitrogen, soil microbial biomass carbon, and soil microbial biomass nitrogen (first group of variables) were longer than those for soil water, bulk density, electric conductivity, and pH (second group of variables), indicating that the first group of variables accounted for a greater proportion of the variance of the ecological stoichiometry in leaves, roots, soil and litter than the second group. In addition, the soil microbial biomass carbon and soil microbial biomass nitrogen had a strong influence on the ecological stoichiometry in leaves, roots, soil and litter. Additionally, the canonical coefficients for the soil variables indicate that axis 1 primarily represents a gradient in soil microbial biomass carbon and nitrogen.

3.4. Correlations among ecological stoichiometry The correlations among ecological stoichiometry are shown in Table 4. The C contents of the 0–5 cm soil layer were strongly significantly related to the C and N contents of leaves, roots, and litter (p < 0.01) and significantly related to the P contents in leaves, roots, and litter (p < 0.05). The C contents of the 5–10 cm soil layer were strongly significantly related to the C contents in leaves, litter and roots (p < 0.01) and significantly related to C contents in leaves, litter and roots (p < 0.05). The C contents of the 10–15 cm soil layer were strongly significantly related to the C contents in leaves and litter (p < 0.01) and significantly related to the C contents in roots and leaves (p < 0.05). The N contents of the 0–5 cm soil layer were strongly significantly related to the C, N contents in leaves, roots, and litter (p < 0.01, and significantly related to the leaf P contents in roots and C contents in litter (p < 0.05). The N contents of the 5–10 cm soil layer were strongly significantly related to the N contents in leaves, litter and roots (p < 0.01) and significantly related to the C contents in leaves and roots and P content in litter. The N contents of the 10–15 cm soil layer were significant related to the N contents in leaves, roots, and litters and C content in litter (p < 0.05). The P contents of the 0–5 cm soil layer were significantly related to the P contents of leaves, roots and litter (p < 0.01). However, there were no relationships between the P contents of soil in the 5–10 or 10–15 cm layers and the C:N, C:P, and N:P ratios in leaves, roots and litter. Thus, the results showed that the C, N and P contents of the 0–5 cm soil layer were mainly related to the ecological stoichiometry in the leaves, roots and litters. In addition, we used a simple linear relationship to explore the correlations among the ecological stoichiometry in leaves, roots, soil and litter (Fig. 5). As for the relationships between the C and N contents, N and P contents, C:N and N:P, C:P and N:P, there were significant positive linear relationships in leaves, roots, soil and litter (p < 0.01).

4. Discussion 4.1. Soil C:N:P stoichiometry among different plant communities This study indicates that the C:N:P stoichiometry in soil varies widely and that the plant community has a significant effect on the soil C:N:P stoichiometry (Tables 2 and 3), which are in agreement with previous studies (Han et al., 2005; Mulder and Elser, 2009; Chen et al., 2016; Tang et al., 2016; Zeng et al., 2016). The large differences in the soil C:N, C:P and N:P ratios can be ascribed to the varied soil properties (Table 4). In addition, the average soil C: N ratio (9.24) is lower than the average level in China (from 10.1 to 12.1) and the global average level (13.33) (McGroddy et al., 2004; Tian et al., 2010). Nevertheless, there was a significant difference in the soil C:N ratios among different plant communities (p < 0.05). The average soil C: P and N: P ratios were 13.76 and 1.21, respectively, which are lower than the average levels in China (C:P ≈ 61.0, N:P ≈ 5.1), suggesting that the soil C and N contents are relatively low in this region. Specifically, the relatively low temperatures and rainfall in this region limit plant production, thereby 332

Catena 166 (2018) 328–338

Y. Yang et al.

Fig. 2. The characteristics of the ecological stoichiometry in litter among different plant communities. Different lowercase letters indicate significant differences (p < 0.05), The same is true below.

(Leroux et al., 2017; Mayor et al., 2017). Additionally, the soil C:N:P ratios exhibit a strong correlation with the soil properties, except soil P (Table 5; Fig. 6). This strong correlation could be explained by a shift in biological processes. It is widely accepted that C and N are linked to biological processes, such as photosynthesis and biological N fixation, which could enhance plant photosynthesis, biological N fixation and the accumulation of C and N (McGroddy et al., 2004; Rivas-Ubach and Peñuelas, 2012; Pausch and Kuzyakov, 2018). Thus, higher C and N accumulation but lower P accumulation could be responsible for the significant variation in the soil C:N, C:P and N:P ratios in this region.

reducing soil nutrients and resulting in the lower soil C, N, and P concentrations that we found (Liu et al., 2010; Zuo et al., 2012). Interestingly, we observed slightly positive correlations between the soil C:N:P ratios and soil properties (Table 5). For example, there was a positive correlation between the soil C:N ratio and soil moisture. In soils with a higher water content, soil microbes prioritize the immobilization of available N (Zhang et al., 2018; Zhou et al., 2018) because available C is relatively more sufficient than available N, which are primarily derived from soil C and N, respectively (Yu et al., 2010; Hu et al., 2018). Consequently, relatively less C will be stored in soil, resulting in a positive correlation between the soil C:N ratio and soil moisture 333

Catena 166 (2018) 328–338

Y. Yang et al.

Fig. 3. The vertical distribution of soil C, N and P among different plant communities.

(Koerselman and Meuleman, 1996; Güsewell, 2004), which means that leguminous plants are not subject to P limitation, whereas non-leguminous plants are restricted by P supply. Under lower P stress, plants form a self-regulating mechanism by strengthening P absorption from soil or re-absorption from senescent organs (Schreeg et al., 2014; Yang et al., 2014; Hu et al., 2018; Pausch and Kuzyakov, 2018), which keeps the element at a relatively stable level to ensure more biochemical reactions in this region.

4.2. Plant C:N:P stoichiometry among different plant communities Plant growth requires photosynthetic products, and ribosomes are required for the synthesis of proteins, which contain large amounts of N and P (Downing and McCauley, 1992; van Duren and Pegtel, 2000). Thus, it is widely reported that the N and P concentrations and N:P ratio can provide important information about nutrient limitation (Kershaw, 1957; Sterner and Elser, 2002; Schreeg et al., 2014; Mayor et al., 2017). For example, Elser et al. (2000) provided an initial ecological stoichiometry overview and reported the range of the C:N:P stoichiometry. For example, in terrestrial plants, the C: N ratio ranged from 5 to 100 and C: P ratio ranged from 250 to 3500; similar findings have also been reported elsewhere. Zhang et al. (2004) found that when the N:P ration was < 21, the plant community was N-limited, whereas an N:P ration > 23 might indicate P limitation. Koerselman and Meuleman (1996) showed that an N:P ratio > 16 indicated P limitation, whereas an N:P ratio < 14 indicated N limitation. With an N:P ratio between 14 and 16, plant growth was co-limited by N and P together. However, Güsewell and Koerselman (2002) found no maximal N for nutrient limitation; when p < 1.0 mg·g−1, there seems to be a P limitation alone (Hector and Bagchi, 2007). In China, Han et al. (2005) reported that an ecosystem was limited by P by measuring > 735 plant species. In this study, the leaf N: P ratio was between 12.39 and 18.84 (Table 3) and the average leaf N: P ratio was 14.91, significantly lower than 16 (Hector and Bagchi, 2007), but higher than the global average level (13.8 and 12.7) (Killingbeck, 1996; Elser et al., 1988; Lockaby and Conner, 1999; Elser et al., 2000). Additionally, leguminous plants had a higher N:P ratio than non-leguminous plants. Based on the theory stating that limited elements are better regulated in homeostasis, our results suggest that P was the most restricted element among these plant communities in this region. The lower N:P ratio that we observed was mainly caused by the higher N content of leguminous plants

4.3. Litter C:N:P stoichiometry among different plant communities Litter plays an important role in nutrient cycling (Sterner and Elser, 2002), which is related to the C:N:P stoichiometry. Previous studies showed that when the content of N in litter was < 0.70% and the content of P as < 0.05%, it can be assumed that the N and P concentrations in litter are absorbed completely in the plant. On the contrary, when the concentration of N in litter was > 1.00% and the P concentration was > 0.08%, it can be assumed that litter N and P were not completely absorbed (Travis et al., 1995; Killingbeck, 1996; Gan and Amasino, 1997; Yu et al., 2010). Indeed, our results indicate that the concentration of N in litter of non-leguminous plants was < 1.00% and of leguminous plants was > 1.00% (Fig. 2). According to the theory of Killingbeck (1996), we can conclude that the litter decomposition rate of leguminous plants was fast and N had not been completely absorbed, while P might have been completely absorbed. In addition, the litter N:P ratios had no significant correlation with the leaf N:P ratios (p > 0.05). These results demonstrate that the litter C:N:P stoichiometry exhibits indirect effects on the nutrient limitations in this region. 4.4. Correlations with the C:N:P stoichiometry Soil properties explained the large variance in the C:N:P 334

Catena 166 (2018) 328–338

Y. Yang et al.

Fig. 4. The characteristics of the ecological stoichiometry in soil among different plant communities.

and Meuleman, 1996; Güsewell, 2004; Zuo et al., 2012). In addition, we used RDA to disentangle the effect of the soil properties on the C:N:P stoichiometry (Table 5, Fig. 6). Subsequent comparisons using correlation coefficients indicated that axis 1 primarily represented a gradient in soil microbial biomass carbon and nitrogen. By contrast, P had no significant correlation with the C:N:P stoichiometry. These RDA results also highlight how the effects of the C:N:P stoichiometry are mediated by soil properties. For example, soil pH is an integrated index of soil nutrient availability and is correlated with various processes (soil microbial activities); a lower soil pH generally promotes higher nutrients (Yang et al., 2016; Hu et al., 2018). Thus, there is a negative correlation

stoichiometry (Table 4). Correlation analysis indicated that the C, N and P contents of leaves, roots, and litter were positively related to the soil C, N and P in the 0–5 cm layer and that the correlation coefficient gradually weakened with the soil depth, suggesting that the concentrations of C, N, and P aboveground mainly depended on the contents of C, N, and P in surface soil layer (0–5 cm). This observation is mainly related to the accumulation of organic matter and nutrients on the surface soil layer (Hooper et al., 2000; Bertilsson et al., 2003; Yang et al., 2016; Zhou et al., 2018). Moreover, the concentrations of C and N and N and P presented a significantly positive linear relationship (p < 0.05), which is in agreement with previous studies (Koerselman 335

Catena 166 (2018) 328–338

Y. Yang et al.

Table 4 Pearson correlations between the soil ecological stoichiometry and ecological stoichiometry in roots and leaves. Item

Soil C 0–5 cm

Leaves C Leaves N Leaves P Roots C Roots N Roots P Litters C Litters N Litters P ⁎⁎ ⁎

⁎⁎

0.914 0.825⁎⁎ 0.546⁎ 0.876⁎⁎ 0.663⁎⁎ 0.510⁎ 0.813⁎⁎ 0.619⁎⁎ 0.638⁎

Soil N 5–10 cm ⁎⁎

0.703 0.504⁎ 0.109 0.732⁎⁎ 0.546⁎ 0.063 0.724⁎⁎ 0.578⁎ 0.039

10–15 cm ⁎

0.516 0.128 0.234 0.578⁎ 0.109 0.289 0.639⁎⁎ 0.156 0.254

Soil P

0–5 cm

5–10 cm

⁎⁎

⁎

0.756 0.904⁎⁎ 0.536⁎ 0.723⁎⁎ 0.903⁎⁎ 0.562⁎ 0.569⁎ 0.869⁎⁎ 0.158

0.423 0.783⁎⁎ 0.302 0.538⁎ 0.854⁎⁎ 0.354 0.236 0.723⁎⁎ 0.486⁎

10–15 cm 0.158 0.553⁎ 0.217 0.324 0.528⁎ 0.303 0.498⁎ 0.524⁎ 0.327

0–5 cm ⁎

0.508 0.369 0.622⁎⁎ 0.389 0.259 0.723⁎⁎ 0.468⁎ 0.523⁎⁎ 0.622⁎⁎

5–10 cm

10–15 cm

0.157 0.152 0.485⁎ 0.495⁎ 0.147 0.518⁎ 0.039 0.254 0.486⁎

0.246 0.203 0.179 0.087 0.368 0.253 0.157 0.377 0.536⁎

Correlation is significant at the 0.01 level (2-tailed). Correlation is significant at the 0.05 level (2-tailed).

5. Conclusions

between the soil pH and soil properties. Additionally, we found a strong relationship between the C:N:P stoichiometry and soil properties, which is consistent with previous studies (Han et al., 2005; Chen et al., 2016; Tang et al., 2016; Zeng et al., 2016), indicating that the C:N:P stoichiometry in plant exhibits a strong relationship with soil properties. However, this strong relationship cannot provide a clear explanation for the interactions among the C:N:P stoichiometry in the plant-litter-soil system. Therefore, future research should focus on the C:N:P stoichiometry in the plant-litter-soil system in the desertified region of China.

This study was the first comprehensive field investigation that examined the driving factors of the C:N:P stoichiometry in leaves, roots, litter and soil at the plant community level in a desertified region of China. The findings have the following implications. First, our results reveal significant differences in the C:N:P stoichiometry among different plant communities, suggesting that the C:N:P stoichiometry may be inherently or biologically flexible in this region. Second, our results

Fig. 5. Correlations between the leaves, roots, litter and soil ecological stoichiometry. 336

Catena 166 (2018) 328–338

Y. Yang et al.

Table 5 Redundancy analysis (RDA) of the ecological stoichiometry and soil properties using forward selection with a Monte Carlo permutation test. Variables SMB-C SMB-N AP AN pH EC SW BD Total

Note:

⁎

p < 0.05;

Initial conditional effects 0.426 0.316 0.154 0.132 0.098 0.056 0.021 0.000 – ⁎⁎

p < 0.01;

MCR (%) 35.6 24.2 16.9 14.3 3.2 1.5 1.0 0.0 97.6

⁎⁎⁎

F-ratio

p-Value ⁎⁎

9.365 8.123 6.021 5.154 1.236 0.356 0.269 0.058 –

0.002 0.005⁎⁎ 0.016⁎ 0.029⁎ 0.256 0.389 0.639 0.987 –

Axis

1

2

Eigenvalues Cumulative percentage variance Species data Species-environment relation Summary of Monte Carlo test F-ratio p-Value Species-environment correlations –

80.5 – 19.2 63.9 For all axes 6.35 0.016⁎ 0.863 –

9.8 – 24.7 78.4 8.24 0.008⁎⁎ 0.652 –

p < 0.001.

References Bai, Y., Wu, J., Clark, C.M., Pan, Q., Zhang, L., Chen, S., Han, X., 2012. Grazing alters ecosystem functioning and C:N:P stoichiometry of grasslands along a regional precipitation gradient. J. Appl. Ecol. 49, 1204–1215. Bertilsson, S., Berglund, O., Karl, D.M., Chisholm, S.W., 2003. Elemental composition of marine Prochlorococcus and Synechococcus: implications for the ecological stoichiometry of the sea. Limnol. Oceanogr. 48, 1721–1731. Bradshaw, C., Kautsky, U., Kumblad, L., 2012. Ecological stoichiometry and multi-element transfer in a coastal ecosystem. Ecosystems 15, 591–603. Chen, Y., Chen, L., Peng, Y., Ding, J., Li, F., Yang, G., 2016. Linking microbial C:N:P stoichiometry to microbial community and abiotic factors along a 3500-km grassland transect on the Tibetan plateau. Glob. Ecol. Biogeogr. 25, 1416–1427. Cleveland, C.C., Liptzin, D., 2007. C:N:P stoichiometry in soil: is there a “Redfield ratio” for the microbial biomass? Biogeochemistry 85, 235–252. Cooperative Research Group on Chinese Soil Taxonomy, 2001. Chinese Soil Taxonomy. Science Press. Cross, W.F., Benstead, J.P., Frost, P.C., Thomas, S.A., 2005. Ecological stoichiometry in freshwater benthic systems: recent progress and perspectives. Freshw. Biol. 50, 1895–1912. Daufresne, T., 2004. Optimal nitrogen-to-phosphorus stoichiometry of phytoplankton. Nature 429, 171–174. Daufresne, T., Loreau, M., 2001. Ecological stoichiometry, primary producer-decomposer interactions, and ecosystem persistence. Ecology 82, 3069–3082. Downing, J.A., McCauley, E., 1992. The nitrogen: phosphorus relationship in lakes. Limnol. Oceanogr. 37, 936–945. Elser, J.J., Elser, M.M., MacKay, N.A., Carpenter, S.R., 1988. Zooplankton-mediated transitions between N-and P-limited algal growth. Limnol. Oceanogr. 33, 1–14. Elser, J.J., Sterner, R.W., Gorokhova, E., Fagan, W.F., Markow, T.A., Cotner, J.B., Weider, L.W., 2000. Biological stoichiometry from genes to ecosystems. Ecol. Lett. 3, 540–550. FAO, ISRIC, 1988. Revised Legend of the FAO-UNESCO SoilMap of the World. ISRIC Report. ISRIC-World Soil Information. Frost, P.C., Stelzer, R.S., Lamberti, G.A., Elser, J.J., 2002. Ecological stoichiometry of trophic interactions in the benthos: understanding the role of C:N:P ratios in lentic and lotic habitats. J. N. Am. Benthol. Soc. 21, 515–528. Gan, S., Amasino, R.M., 1997. Making sense of senescence (molecular genetic regulation and manipulation of leaf senescence). Plant Physiol. 113, 313. Güsewell, S., 2004. N:P ratios in terrestrial plant: variation and functional significance. New Phytol. 164, 243–266. Güsewell, S., Koerselman, W., 2002. Variation in nitrogen and phosphorus concentrations of wetland plants. Perspect. Plant Ecol. 5, 37–61. Han, W., Fang, J., Guo, D., Zhang, Y., 2005. Leaves nitrogen and phosphorus stoichiometry across 753 terrestrial plant species in China. New Phytol. 168, 377–385. Hättenschwiler, S., Jørgensen, H.B., 2010. Carbon quality rather than stoichiometry controlslitters decomposition in a tropical rain forest. J. Ecol. 98, 754–763. He, J.S., Fang, J., Wang, Z., Guo, D., Flynn, D.F., Geng, Z., 2006. Stoichiometry and largescale patterns of leaves carbon and nitrogen in the grassland biomes of China. Oecologia 149, 115–122. He, J.S., Wang, L., Flynn, D.F., Wang, X., Ma, W., Fang, J., 2008. Leaves nitrogen: phosphorus stoichiometry across Chinese grassland biomes. Oecologia 155, 301–310. Hector, A., Bagchi, R., 2007. Biodiversity and ecosystem multifunctionality. Nature 448, 188. Hessen, D.O., 1992. Nutrients element limitation of zooplankton production. Am. Nat. 799–814. Hooper, D.U., Bignell, D.E., Brown, V.K., Brussard, L., Dangerfield, J.M., Wall, D.H., Van Der Putten, W.H., 2000. Interactions between aboveground and belowground biodiversity in terrestrial ecosystems: patterns, mechanisms, and feedbacks we assess the evidence for correlation between aboveground and belowground diversity and conclude that a variety of mechanisms could lead to positive, negative, or no relationship—depending on the strength and type of interactions among species. Bioscience 50, 1049–1061. Hu, Y.F., Shu, X.Y., He, J., Zhang, Y.L., Xiao, H.H., Tang, X.Y., 2018. Storage of C, N and P affected by afforestation with salix cupularis in an alpine semi-arid desert ecosystem. Land Degrad. Dev. 2, 188–198. Kershaw, K.A., 1957. The use of cover and frequency in the detection of pattern in plant communities. Ecology 38, 291–299.

Fig. 6. Redundancy analysis (RDA) showing the relationship between the ecological stoichiometry and soil variables. In the biplot, plants are represented by numbers. Six plants are represented as follows: 1) S. bungeana, 2) A. mongolicum, 3) G. uralensis, 4) S. alopecuroides, 5) A. ordosica, and 6) C. komarovii. SW: soil water, BD: bulk density, EC: electric conductivity, pH: pH value, AP: available phosphorus, AN: available nitrogen, SMB-C: soil microbial biomass carbon, SMB-N: soil microbial biomass nitrogen Le-C: leaves C, Le-N: leaves N, Le-P: leaves P T-C: soil C, T-N: soil N, T-P: soil P Li-C: litter C, Li-N: litter N, Li-P: litter P R-C: roots C, R-N: roots N, R-P: roots P.

demonstrate that variations in the C:N:P stoichiometry among different plant communities are related to soil properties (except for soil P) and that there are strong links among the C:N:P stoichiometry in leaves, roots, litter and soil, suggesting that the variations in the C:N:P stoichiometry may be induced by shifts in soil properties, especially soil microbial biomass carbon and nitrogen (by RDA). In particular, our study supports the homeostasis hypothesis, with this unique mechanism allowing plants to adapt to their environment, which highlights the importance of soil properties in shaping the C:N:P stoichiometry and strengthens our understanding of the C:N:P stoichiometry in desertified regions in China. Acknowledgements This study was funded by the National Natural Science Foundation of China (31660168); The Integrative Assessment of Eco-environmental Impact of Three Northern Regions Shelter Forest System Project in North Shaanxi Province from the Shaanxi Shelterbelt Construction Workstation; The Key Projects in the National Science & Technology Pillar Program during the Twelfth Five-year Plan Period (2015BAC01B01). 337

Catena 166 (2018) 328–338

Y. Yang et al.

Molecules to the Biosphere. Princeton University Press. Stohlgren, T.J., Binkley, D., Chong, G.W., Kalkhan, M.A., Schell, L.D., Bull, K.A., Otsuki, Y., Newman, G., Bashkin, M., Son, Y., 1999. Exotic plant species invade hot spots of native plant diversity. Ecol. Monogr. 69, 25–46. Tang, Z., An, H., Deng, L., Wang, Y., Zhu, G., Shangguan, Z., 2016. Effect of desertification on productivity in a desert steppe. Sci. Rep. 6, 27839. Tian, H., Chen, G., Zhang, C., Melillo, J.M., Hall, C.A., 2010. Pattern and variation of C:N:P ratios in China's soil: a synthesis of observational data. Biogeochemistry 98, 139–151. Tilman, D., 1994. Competition and biodiversity in spatially structured habitats. Ecology 75 (1), 2–16. Tilman, D., 1996. Biodiversity: population versus ecosystem stability. Ecology 77, 350–363. Tilman, D., 2004. Niche trade-offs, neutrality, and community structure: a stochastic theory of resource competition, invasion, and community assembly. PNAS 101, 10854–10861. Travis, S.E., Slobodchikoff, C.N., Keim, P., 1995. Ecological and demographic effects on intraspecific variation in the social system of prairie dogs. Ecology 12, 1794–1803. Urabe, J., Watanabe, Y., 1992. Possibility of N or P limitation for planktonic cladocerans: an experimental test. Limnol. Oceanogr. 37, 244–251. Van Duren, I.C., Pegtel, D.M., 2000. Nutrients limitations in wet, drained and rewetted fen meadows: evaluation of methods and results. Plant Soil 220, 35–47. Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method form measuring soil microbial biomass C. Soil Biol. Biochem. 19, 703–707. Wang, S.Q., Yu, G.R., 2008. Ecological stoichiometry characteristics of ecosystem carbon, nitrogen and phosphorus elements. Acta Ecol. Sin. 8, 3937–3947. Yang, X., Tang, Z., Ji, C., Liu, H., Ma, W., Mohhamot, A., 2014. Scaling of nitrogen and phosphorus across plant organs in shrubland biomes across northern China. Sci. Rep. 4, 5448. Yang, X., Chi, X., Ji, C., Liu, H., Ma, W., Mohhammat, A., 2016. Variations of leaf N, P concentrations in shrubland biomes across northern China: phylogeny, climate and soil. Biogeosciences 12, 18973–18998. Yoshihara, Y., Sasaki, T., Okuro, T., Undarmaa, J., Takeuchi, K., 2010. Cross-spatial-scale patterns in the facilitative effect of shrubs and potential for restoration of desertified region. Ecol. Eng. 36, 1719–1724. Yu, Q., Chen, Q., Elser, J.J., He, N., Wu, H., Zhang, G., Han, X., 2010. Linking stoichiometric homoeostasis with ecosystem structure, functioning and stability. Ecol. lett. 13, 1390–1399. Zeng, Q., Li, X., Dong, Y., An, S., Darboux, F., 2016. Soil and plant components ecological stoichiometry in four steppe communities in the Loess Plateau of China. Catena 147, 481–488. Zhang, L.X., Bai, Y.F., Han, X.G., 2004. Differential responses of N: P stoichiometry of Leymus chinensis and Carex korshinskyi to N additions in a steppe ecosystem in Nei Mongol. Acta Bot. Sin. 46, 259–270. Zhang, J., Zhao, N., Liu, C., Yang, H., Li, M., Yu, G., 2018. C:N:P stoichiometry in China's forests: from organs to ecosystems. Funct. Ecol. 32, 50–60. Zhou, X.B., Matthew, A., Tao, Y., Wu, L., Zhang, M.Y., 2018. Chronic nitrogen addition induces a cascade of plant community responses with both seasonal and progressive dynamics. Sci. Total Environ. 626, 99–108. Zuo, X., Zhao, X., Wang, S., Li, Y., Lian, J., Zhou, X., 2012. Influence of dune stabilization on relationship between plant diversity and productivity in Horqin Sand Land, Northern China. Environ. Earth Sci. 67, 1547–1556.

Killingbeck, K.T., 1996. Nutrients in senesced leaves: keys to the search for potential resorption and resorption proficiency. Ecology 77, 1716–1727. Koerselman, W., Meuleman, A.F.M., 1996. The vegetation N: P ratio: a new tool to detect the nature of nutrients limitation. J. Appl. Ecol. 1441–1450. Lepš, J., Šmilauer, P., 2003. Multivariate Analysis of Ecological Data Using CANOCO. Cambridge University Press, Cambridge, UK. Leroux, S.J., Wal, E.V., Wiersma, Y.F., Charron, L., Ebel, J.D., Ellis, N.M., 2017. Stoichiometric distribution models: ecological stoichiometry at the landscape extent. Ecol. Lett. 20, 1495–1506. Li, X.R., 2001. Study on shrub community diversity of Ordos Plateau, Inner Mongolia, northern China. J. Arid Environ. 47, 271–279. Liu, X.Z., Zhou, G.Y., Zhang, D.Q., Liu, S.Z., Chu, G.W., Yan, J.H., 2010. N and P stoichiometry of plant and soil in lower subtropical forest successional series in southern China. J. Plant Ecol. 34, 64–71. Liu, R., Zhao, H., Zhao, X., Drake, S., 2011. Facilitative effects of shrubs in shifting sand on soil macro-faunal community in Horqin Sand Land of Inner Mongolia, Northern China. Eur. J. Soil Biol. 47, 316–321. Lockaby, B.G., Conner, W.H., 1999. N: P balance in wetland forests: productivity across a biogeochemical continuum. Bot. Rev. 65, 171–185. Manzoni, S., Jackson, R.B., Trofymow, J.A., Porporato, A., 2008. The global stoichiometry oflitters nitrogen mineralization. Science 321, 684–686. Manzoni, S., Trofymow, J.A., Jackson, R.B., Porporato, A., 2010. Stoichiometric controls on carbon, nitrogen, and phosphorus dynamics in decomposing litter. Ecol. Monogr. 80, 89–106. Mayor, J.R., Sanders, N.J., Classen, A.T., Bardgett, R.D., Clément, J.C., Fajardo, A., 2017. Elevation alters ecosystem properties across temperate treelines globally. Nature 542, 91–96. McGroddy, M.E., Daufresne, T., Hedin, L.O., 2004. Scaling of C:N:P stoichiometry in forests worldwide: implications of terrestrial Redfield-type ratios. Ecology 85, 2390–2401. Melillo, J.M., Aber, J.D., Muratore, J.F., 1982. Nitrogen and lignin control of hardwood leaveslitters decomposition dynamics. Ecology 63, 621–626. Moe, S.J., Stelzer, R.S., Forman, M.R., Harpole, W.S., Daufresne, T., Yoshida, T., 2005. Recent advances in ecological stoichiometry: insights for population and community ecology. Oikos 109, 29–39. Mulder, C., Elser, J.J., 2009. Soil acidity, ecological stoichiometry and allometric scaling in grassland food webs. Glob. Chang. Biol. 15, 2730–2738. Naeem, S., Li, S., 1997. Biodiversity enhances ecosystem reliability. Nature 390, 507–509. Pausch, J., Kuzyakov, Y., 2018. Carbon input by roots into the soil: quantification of rhizodeposition from roots to ecosystem scale. Glob. Chang. Biol. 24, 1–12. Redfield, A.C., 1958. The biological control of chemical factors in the environment. Am. Sci. 46, 221–230. Rivas-Ubach, A., Peñuelas, J., 2012. Strong relationship between elemental stoichiometry and metabolome in plants. PNAS 109, 4181–4186. Schade, J.D., Kyle, M., Hobbie, S.E., Fagan, W.F., Elser, J.J., 2003. Stoichiometric tracking of soil nutrients by a desert insect herbivore. Ecol. Lett. 6, 96–101. Schreeg, L.A., Santiago, L.S., Wright, S.J., Turner, B.L., 2014. Stem, root, and older leaf N: P ratios are more responsive indicators of soil nutrient availability than new foliage. Ecology 95, 2062–2068. Sinsabaugh, R.L., Lauber, C.L., Weintraub, M.N., Ahmed, B., Allison, S.D., Crenshaw, C., Gartner, T.B., 2008. Stoichiometry of soil enzyme activity at global scale. Ecol. Lett. 11, 1252–1264. Sterner, R.W., Elser, J.J., 2002. Ecological Stoichiometry: The Biology of Elements from

338