Response of Soil Respiration to Grazing in an Alpine Meadow at Three Elevations in Tibet

Alpine meadows are one major type of pastureland on the Tibetan Plateau. However, few studies have evaluated the response of soil respiration (R s) to grazing along an elevation gradient in an alpine meadow on the Tibetan Plateau. Here three fenced enclosures were established in an alpine meadow at three elevations (i.e., 4313 m, 4513 m, and 4693 m) in July 2008. We measured R s inside and outside the three fenced enclosures in July–September, 2010-2011. Topsoil (0–20 cm) samples were gathered in July, August, and September, 2011. There were no significant differences for R s, dissolved organic C (DOC), and belowground root biomass (BGB) between the grazed and ungrazed soils. Soil respiration was positively correlated with soil organic C (SOC), microbial biomass (MBC), DOC, and BGB. In addition, both R s and BGB increased with total N (TN), the ratio of SOC to TN, ammonium N (NH4 +-N), and the ratio of NH4 +-N to nitrate N. Our findings suggested that the negligible response of R s to grazing could be directly attributed to that of respiration substrate and that soil N may indirectly affect R s by its effect on BGB.


Introduction
Soil respiration (R s ) is an important flux in the C cycle [1][2][3]. Raich and Schlesinger [1] indicated that R s is the second in magnitude to gross primary production but equivalent to or even greater than net primary production. Root and microbial respiration are the two most important components of R s ; thus the factors which affect root growth and microbial activity could all influence R s [4][5][6]. Water availability and temperature are the two most important abiotic factors controlling R s at various spatial and temporal scales [7,8].
The positive relationship between R s and temperature could be weakened or masked by other factors (e.g., respiration substrate) [9,10]. Previous studies have shown that R s increases with respiration substrate, including labile C (e.g., microbial biomass C, MBC; dissolved organic C, DOC) and belowground root biomass (BGB) [5,[10][11][12]. Soil N also affect R s by influencing plant growth and microbial activity [6,13].
Grazing is a major type of land use in grasslands and previous studies have shown inconsistent results on the response of R s to grazing [14,15]. Many studies have indicated that grazing significantly decreased R s [14,16,17], whereas other studies have shown quite the contrary result [18,19]. The responses of soil C and N (e.g., MBC, DOC, microbial biomass N, and dissolved organic N) to grazing differ among previous studies [20][21][22][23]. There are also inconsistent results on the response of BGB to grazing [15,24]. The response of R s to grazing is complex and may be dependent on the responses of respiration substrate [25] and soil N.
Alpine meadows are a major type of pastureland on the Tibetan Plateau [14,22] and store 4.68 Pg soil organic C (SOC) with density of 9.05 kg m −2 at depth of 0-100 cm [26]. Few studies have evaluated the response of R s to grazing in an alpine meadow along an elevation gradient, although pasture for domestic sheep and yak is a common land use type on the Tibetan Plateau [22]. Here, we investigated the grazing effect 2 The Scientific World Journal on R s in an alpine meadow at three elevations (i.e., 4313 m, 4513 m, and 4693 m) on the Northern Tibetan Plateau.
The main objectives of this study were to examine (1) the effect of grazing on R s , soil C, and N and (2) the relationships between R s and respiration substrate (soil organic C, MBC, DOC, and BGB) and soil N (total N, microbial biomass N, dissolved organic N, ammonium N, and nitrate N) along an elevation gradient in an alpine meadow in Tibet.  [12]. Annual average precipitation was around 476.8 mm and annual potential evapotranspiration was about 1725.7 mm [27]. Annual mean air temperature was 1.3 ∘ C [12]. The soil was classified as a shallow sandy loam (∼ 0.5-0.7 m), with organic matter of 0.3-11.2%, total N of 0.03-0.49%, and pH of 6.0-6.7 [22]. The vegetation surrounding the study site was Kobresia-dominated alpine meadow [28]. Roots are mainly concentrated in the topsoil layer (0-20 cm) [29].

Materials and Methods
Based on meteorological observations from 1963 to 2012 at the Damxung Station (4288 m, approximately 4 km from our study site), there was no significant change for annual precipitation, while annual mean air temperature increased at a rate of 0.04 ∘ C a −1 [12].

Experimental Design.
Three sites (about 20 m × 20 m for each) were fenced in an alpine meadow on a southfacing slope on the Nyainqentanglha Mountains along an elevation gradient (i.e., 4313 m, 4513 m, and 4693 m) in July 2008. Before enclosure, the site at elevation 4313 m was winter pasture, while the other two sites were summer pasture [22]. A more detailed description of the experimental design can be found in Fu et al. [22].
Soil temperature (T s ) at a depth of 5 cm, soil water content (SWC) at a depth of 10 cm, and air temperature and relative humidity at a height of 15 cm were continuously monitored using data loggers (HOBO weather station, Onset Computer Corporation, USA) at each elevation [22]. Both air temperature and T s increased with decreasing elevation [22].

Measurement of R .
Soil respiration was measured using a soil CO 2 flux system (LI-8100, LI-COR Biosciences, Lincoln, NE, USA) [6,30] during the period from July to September in 2010 and 2011 ( Figure 1). Soil respirationat 9:00-11:00 am was close to daily average R s [6,31]; thus R s was measured between 9:00 and 11:00 (local time) in this study. Four polyvinyl chloride (PVC) collars (20 cm in diameter and 5 cm in height) were inserted into the soil to depths of about 2-3 cm on each measuring date. All the PVC collars were installed and the aboveground biomass was removed at least 12 h before R s measurement in order to reduce disturbance [6,12]. The opaque survey chamber was manually mounted  [30]. One cycle was performed on each measuring date.

Soil Sampling and Analysis.
Topsoil samples (0-20 cm depth) inside and outside the three fenced enclosures were collected (using a soil auger of 3.0 cm in diameter) on July 7, August 9, and September 10, 2011 [22]. Five soil subsamples were randomly sampled and composited into one soil sample for each of the four replicates. The composited soil samples were stored in an icebox and transferred to laboratory. We sieved soil samples (with a sieve of 1 mm diameter) and picked up any visible roots from the sieved soil. Subsamples of the sieved soil were used to measure NO 3 − -N, NH 4 + -N, DOC, and DON. All the roots in the soil samples were washed, dried at 65 ∘ C for 48 h, and weighed.
We extracted 20 g fresh soil samples using 100 mL K 2 SO 4 . The K 2 SO 4 extracts were filtered through 0.45 m filter membrane and then soil available N (SAN, NO 3 − -N, and NH 4 + -N) in the extracts were analyzed on a LACHAT Quickchem Automated Ion Analyzer.
The methods of Jones and Willett [32] were used to determine DOC and dissolved total N (DTN). Briefly, we extracted 20 g fresh soil samples using 100 mL ultrapure water and filtered the extracts through 0.45 m filter membrane. We analyzed the extractable soil organic C and total N in the ultrapure water extracts using a Liqui TOC II elementar analyzer (Elementar Liqui TOC, Elementar Co., Hanau, Germany) and a UV-1700 PharmaSpec visible spectrophotometer (220 nm and 275 nm), respectively. Dissolved inorganic N (DIN) concentrations in the ultrapure water extracts were also determined on a LACHAT Quickchem Automated Ion Analyzer. Then, DON was calculated as the difference between DTN and DIN. Soil organic C, TN, MBC, and MBN data were obtained from Fu et al. [22].

Statistical Analysis.
Repeated-measures analysis of variance (ANOVA) was used to estimate the main and interactive effects of measuring date and grazing on R s for each site (

Results
At elevation 4313 m, NO  Table 2). In contrast, there were no significant differences of NO 3 − -N, NH 4 + -N, and SAN between ungrazed and grazed soils at elevation 4693 m ( Figure 2 and Table 2).
In addition, grazing had no significant effects on NH 4 + -N/NO 3 − -N ratio, DOC, DON, DOC/DON ratio, and BGB for the three alpine meadow sites ( Figure 2 and Table 2 Average DON at elevations 4313 m and 4513 m across all the three sampling dates was 50.78% and 33.84% lower than that of elevation 4693 m under grazing, respectively ( < 0.05), whereas there was no significant difference between elevations 4313 m and 4513 m. There were no significant differences of average DON among the three sites when grazing was absent.
Average SAN at elevations 4313 m and 4513 m across all the three sampling dates was 38.69% and 38.88% lower than that of elevation 4693 m when grazing was absent, respectively ( < 0.05), while no significant difference between the two lower elevations was found. By contrast, average SAN increased with increasing elevation under grazing (F = 375.30, < 0.001).
No significant differences of DOC/DON ratio and NO 3 − -N were found among the three sites whether or not grazing was present.
The main effect of grazing and its interactive effect with measuring date on R s were not significant for each alpine meadow site ( Figure 1 and Table 1). Grazing only tended to decrease the average R s across all the measuring dates by 14.02% (0.25 mol CO 2 m −2 s −1 ), 4.70% (0.11 mol CO 2 m −2 s −1 ), and −4.07% (−0.15 mol CO 2 m −2 s −1 ) at elevations 4313 m, 4513 m, and 4693 m, respectively. In contrast, there was significant seasonal variation for R s (Figure 1 and Table 1). Regardless of grazing, R s showed similar seasonal dynamics among the three elevations ( Figure 1).

Discussion
Previous studies indicated that grazer urine and dung stimulated soil microbial activity and accelerated nutrient cycling 4 The Scientific World Journal  in grasslands [33,34]. However, this effect may be often weakened at the three alpine meadow sites because the dung of yak and goat was removed by local residents.
Generally, grazing did not alter the distributions of R s , DOC, DON, DOC/DON ratio, NO 3 − -N, NH 4 + -N, and BGB along the elevation gradient, which was in line with previous studies [22,35,36].
Soil microbial biomass N at elevations 4313 m and 4693 m and SAN (including NO 3 − -N and NH 4 + -N) at elevations 4313 m and 4513 m under grazing were significantly lower compared with ungrazed soils, while there were no significant differences of SOC, TN, DOC, DON, BGB, and R s between grazed and ungrazed soils (Tables 1 and 2, [22]). This suggests that soil microbial biomass and available N may respond more rapidly to grazing than SOC, TN, DOC, DON, BGB, and R s . The negligible response of R s to grazing was consistent with some previous studies conducted on the Tibetan Plateau (e.g., [15]).
Previous studies showed that BGB increased with increasing TN in alpine grasslands on the Tibetan Plateau  Figure 3: Relationships between belowground root biomass (BGB, kg m −2 ) and total N (TN, g kg −1 ), ammonium N (NH 4 + -N, mg kg −1 ), soil available N (SAN, mg kg −1 ), the ratio of SOC and TN (SOC/TN ratio), and the ratio of NH 4 + -N and nitrate N (NH 4 + -N/NO 3 − -N ratio), respectively. SOC and TN data were obtained from Fu et al. [22]. [36,37]. Our study confirmed this finding ( Figure 3). Besides, BGB increased with increasing NH 4 + -N, SAN, and NH 4 + -N/NO 3 − -N ratio, but not with NO 3 − -N. Therefore, the positive relationship between BGB and SAN may be mainly attributed to that between BGB and NH 4 + -N. In addition, the ratio of different soil available N forms could affect BGB.
Similar to BGB, R s increased with TN, SOC/TN ratio, NH 4 + -N, SAN, and NH 4 + -N/NO 3 − -N ratio, respectively ( Figure 4). Meanwhile, R s was positively correlated with BGB ( Figure 4). Therefore, the effect of soil N availability and form on R s was probably associated with the effect of soil N availability and form on BGB. In addition, the negligible response of R s to grazing may be directly attributed to that of SOC, MBC, DOC, and BGB.
The positive relationships between DOC and SOC, DON and TN, DOC and MBC, DON and MBN, and SAN and MBN (data not shown) were in accordance with previous studies which were made in alpine meadows on the Tibetan Plateau [22,38] and an upland grassland of northern England [39]. Previous studies found that DOC was a good index in reflecting C availability of soil microorganisms [40,41]. Therefore, the variation of soil microbial biomass along the elevation gradient may be not only associated with that of SOC and TN [22] but also with that of DOC and DON [6,11,42]. In other words, soil microbial activity may regulate the balances of soil inorganic and organic C and N pools in this alpine meadow.
Many studies have found the positive relationship between R s and T s in various ecosystems [3,15]. In contrast, R s increased significantly with increasing elevation, while both soil and air temperatures declined in the current study. In other words, the relationship between R s and T s was negative along the elevation gradient. This implied that other factors (e.g., respiration substrate) probably regulated or confounded the positive relationship between respiration and temperature [10,25]. This viewpoint was as confirmed by the positive relationships between R s and BGB, SOC, MBC, and DOC ( Figure 4).

Conclusions
In this study, we measured soil respiration under grazed and ungrazed conditions in an alpine meadow along an elevation gradient (4313-4693 m with approximate 200 m interval) in Tibet during the period from July to September in 2010-2011. We found that grazing did not significantly affect soil respiration, which was probably attributed to the insignificant response of respiration substrate (e.g., soil organic C and belowground root biomass) to grazing. Soil N availability and the ratio of ammonium to nitrate N might also influence soil respiration by affecting belowground root growth.  (k) Figure 4: Relationships between soil respiration (R s , mol CO 2 m −2 s −1 ) measured early in August and September, 2011, and soil organic C (SOC, g kg −1 ), total N (TN, g kg −1 ), the ratio of SOC and TN (SOC/TN ratio), microbial biomass C (MBC, mg kg −1 ), microbial biomass N (MBN, mg kg −1 ), dissolved organic C (DOC, mg kg −1 ), dissolved organic N (DON, mg kg −1 ), ammonium N (NH 4 + -N, mg kg −1 ), soil available N (SAN, mg kg −1 ), the ratio of NH 4 + -N and nitrate N (NH 4 + -N/NO 3 − -N ratio), and belowground root biomass (BGB, kg m −2 ), respectively. SOC, TN, MBC, and MBN data were obtained from Fu et al. [22].