Response of Soil C and N, Dissolved Organic C and N, and Inorganic N to Short-Term Experimental Warming in an Alpine Meadow on the Tibetan Plateau

Although alpine meadows of Tibet are expected to be strongly affected by climatic warming, it remains unclear how soil organic C (SOC), total N (TN), ammonium N (NH4 +-N) , nitrate N (NO3 +-N), and dissolved organic C (DOC) and N (DON) respond to warming. This study aims to investigate the responses of these C and N pools to short-term experimental warming in an alpine meadow of Tibet. A warming experiment using open top chambers was conducted in an alpine meadow at three elevations (i.e., a low (4313 m), mid-(4513 m), and high (4693 m) elevation) in May 2010. Topsoil (0–20 cm depth) samples were collected in July–September 2011. Experimental warming increased soil temperature by ~1–1.4°C but decreased soil moisture by ~0.04 m3 m−3. Experimental warming had little effects on SOC, TN, DOC, and DON, which may be related to lower warming magnitude, the short period of warming treatment, and experimental warming-induced soil drying by decreasing soil microbial activity. Experimental warming decreased significantly inorganic N at the two lower elevations,but had negligible effect at the high elevation. Our findings suggested that the effects of short-term experimental warming on SOC, TN and dissolved organic matter were insignificant, only affecting inorganic forms.


Introduction
Soil organic C (SOC) and total N (TN) are very important C and N pools in the terrestrial ecosystems [1,2]. As the components of labile C and N pools in soils, dissolved organic C (DOC) and N (DON) and soil ammonium and nitrate N (NH 4 + -N and NO 3 − -N) play crucial roles in the biogeochemistry of C and N and in the nutrient transformation [3][4][5]. With the context of climatic warming, how SOC, TN, DOC, DON, NH 4 + -N, and NO 3 − -N respond is vital to global C and N cycling [1,2]. However, inconsistent results on the responses of these C and N pools to climatic warming have been observed with respect to vegetation types and initial soil characteristics [2,3,[6][7][8][9][10][11][12][13][14]. For example, He et al. [2] demonstrated that six-year warming (∼1.4 ∘ C increase of 10 cm soil temperature) significantly decreased soil C by 129.3 g m −2 in a temperate steppe of Inner Mongolia. In contrast, Li et al. [7] found that two-year warming significantly increased SOC in an alpine meadow (∼2.1 ∘ C increase of air temperature) but significantly reduced TN in an alpine swamp meadow (∼2.3 ∘ C increase of air temperature) on the Tibetan Plateau. Hagedorn et al. [13] indicated that one-growing-season warming (∼4 ∘ C increase of 5 cm soil temperature) did not significantly influence DOC. Song et al. [1] pointed out that six-year warming (∼1.2 ∘ C increase of 10 cm soil temperature) significantly reduced DOC in a temperate steppe in Inner Mongolia. Biasi et al. [15] indicated that two-year warming (∼ 0.9 ∘ C increase of 5 cm soil temperature) did not have obvious effects on DON, NH 4 + -N, NO 3 − -N, and N min in a lichenrich dwarf shrub tundra in Siberia. Bai et al. [14] stated that experimental warming (∼0.6-6.7 ∘ C in soil temperature) had a significant positive effect on N min but not on TN across all biomes. Therefore, how climatic warming acts on C and N cycling still remains unclear.
More than 70% of the Tibetan Plateau is covered with grasslands [16]. The alpine grasslands of this Plateau are one of the systems most sensitive to global change [17,18]. In alpine grasslands, understanding the responses of SOC, DOC, TN, DON, NH 4 + -N, and NO 3 − -N to climatic warming are crucial for predicting future changes in soil fertility and C sequestration. The alpine meadow is one of the most typical grasslands types on the Tibetan Plateau being subjected to climatic warming [19]. Information on how these C and N pools along an elevation gradient respond to climatic warming is scarce on the Tibetan Plateau. Here we set up a warming experiment in an alpine meadow at three elevations (i.e., 4313 m, 4513 m, and 4693 m) on the Northern Tibetan Plateau.
The main objective was to investigate the effects of short-term experimental warming on SOC, TN, DOC, DON, NH 4 + -N, and NO 3 − -N. Our previous study indicated that short-term experimental warming could not affect soil microbial biomass [20] and soil microbial activity regulated the balances of soil C and N pools in the alpine meadow [21]. We hypothesized that experimental warming may not affect these C and N pools in this study.

Study Area, Experimental Design, and Soil Sampling.
A detailed description of the study area, the warming experimental design, the measurements of microclimate factors 4 The Scientific World Journal (including soil temperature and soil moisture), and the soil sampling are given in Fu et al. [20,22]. Briefly, three alpine meadow sites were established at three elevations (i.e., a low (30 ∘  Annual mean air temperature and precipitation is 1.3 ∘ C and ∼476.8 mm, respectively [20,21]. The vegetation is Kobresia-dominated alpine meadow and roots are mainly concentrated in the topsoil layer (0-20 cm) [21,22]. The soil is classified as sandy loam, with pH of 6.0-6.7, organic matter of 0.3-11.2%, and total N of 0.03-0.49% [20,22].
Open top chambers (OTCs, 3 mm thick polycarbonate) were used to enhance temperature [22,23]. The bottom and top diameters and the height of OTCs were 1.45 m and 1.00 m and 0.40 m, respectively [20,22]. For each site, four OTCs and their paired control plots (1 m × 1 m) were randomly established in May 2010. There was ∼3 m distance between plots.
Daily mean soil temperature ( ) during the study period of July-September in 2011 inside the OTCs increased by 1.26 ∘ C, 0.98 ∘ C, and 1.37 ∘ C at the low, mid-, and high elevation, respectively, compared to control plots [20]. In contrast, experimental warming decreased daily mean soil moisture (SM) by 0.04 m 3 m −3 in all sites [20]. Daily mean decreased with increasing elevation from the low to high elevation [20].
We collected topsoil samples (0-20 cm depth) inside each plot using a probe 3.0 cm in diameter on July 7, August 9, and September 10, 2011 [20]. Five soil subsamples were randomly sampled and composited into one soil sample for each plot [20]. Subsamples of the fresh soil were used to measure DOC, DON, NH 4 + -N, and NO 3 − -N and other subsamples of the fresh soil were air-dried for the measurements of SOC and TN.

Soil Analysis.
A more detailed description of measurements of soil inorganic N (N min , i.e., sum of NH 4 + -N and NO 3 − -N), DON, and DOC can be found in Fu et al. [21]. Briefly, soil inorganic N in 20 g fresh soil sample was extracted with 100 mL K 2 SO 4 , filtered through 0.45 m membrane, and analyzed on a LACHAT Quikchem Automated Ion Analyzer. Dissolved organic C and TN (DTN) in another 20 g fresh soil sample was extracted with 100 mL ultrapure water and filtered through 0.45 membrane. The extractable SOC and TN concentrations in the ultrapure water extracts were measured 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. We also analyzed dissolved inorganic N (DIN) in the ultrapure water extracts on a LACHAT Quikchem Automated Ion Analyzer. Then DON was calculated as the difference between DTN and DIN. The potassium dichromate method was used to determine SOC [24]. Soil TN was measured on a CN analyzer (Elementar Variomax CN). Soil microbial biomass (MBC) and N (MBN) data were obtained from Fu et al. [20].

Statistical Analysis.
In order to examine the elevation effect, repeated-measures ANOVA with experimental warming and elevation as the between subject factors and with sampling date as the within subject factor was performed for a specific soil property (i.e., SOC, TN, DOC, DON, ratio of DOC to DON (DOC/DON), NH 4 , and N min ). At each site, repeated-measures ANOVA with experimental warming (i.e., OTCs versus control) as the between subject factor and with sampling date as the within subject factor was conducted for each soil property. Single factor linear regressions were performed between soil properties and , SM, MBC, and MBN. In addition, multiple stepwise regression analyses were conducted for soil properties to examine the relative importance of , SM, MBC, and MBN in affecting the variations of soil properties. All data were examined for normality and homogeneity before analysis and natural logarithm transformations were made if necessary. The level of significance was < 0.05. All the statistical tests  The Scientific World Journal   (Table 1). In contrast, the sensitivity of N min to experimental warming increased with increasing elevation (Table 1). In detail, experimental warming significantly decreased N min by 29.2% and 23.5% at the low and mid-elevation, NO 3 − -N by 36.4%, 29.5% at the low and mid-elevation, and NH 4 + -N by 16.7% at the mid-elevation across all the three sampling dates, respectively. In contrast, experimental warming had little effects on NO 3 − -N and N min at the high elevation.  The multiple stepwise regression analyses were listed in Table 3. Both SOC and TN were simultaneously affected by MBC and , whereas MBC explained more variation of the two soil properties than . Only MBC was included in the multiple regression equations for DOC, DON, and NH 4 + -N/NO 3 − -N, while only MBN was included in the regression equation for NO 3 − -N. Soil microbial biomass C explained the variation of NH 4 + -N more than SM. Both MBC and MBN were simultaneously and positively correlated with N min . In addition, all the five concerned variables were excluded for DOC/ DON.

Effects of Experimental Warming on SOC, TN, DOC, and DON.
Recently, some studies showed that short-term (<3 years) experimental warming had little effects on SOC, TN, DOC, and/or DON in a tallgrass prairie with a silt loam soil (∼2 ∘ C increase of 5 cm soil temperature) in USA [25], in a dragon spruce plantation with a mountain brown soil (∼0.6 ∘ C increase of 5 cm soil temperature) on the Tibetan Plateau [8], in an alpine treeline with a sandy Ranker and Podzols soil (∼4 ∘ C increase of 5 cm soil temperature) in Switzerland [13], and in a lichen-rich dwarf shrub tundra with Gleyic Cryosols soils (∼0.9 ∘ C increase of 5 cm soil temperature) in Siberia [15]. However, other studies with long-term (>3 years) experimental warming indicated that warming significantly increased or decreased SOC, TN, DOC, and/or DON in a temperate steppe with a Calcic Kastanozems soil in Inner Mongolia (∼1.4 ∘ C increase of 10 cm soil temperature) [2], in an alpine meadow (∼3 ∘ C increase of 5 cm soil temperature) on the Tibetan Plateau [3], and in a temperate steppe with chestnut soil in Inner Mongolia (∼1.2 ∘ C increase of 10 cm soil temperature) [1]. Therefore, the insignificant responses of SOC, TN, DOC, and DON to warming (Table 1) may be due to the short period of warming treatment (14-16 months). A meta-analysis showed that the effects of experimental warming on N min , net N mineralization, and nitrification were significantly and positively correlated with raised soil temperature (∼0.6-6.7 ∘ C for N min , ∼0.6-5.5 ∘ C for net mineralization, and ∼1.3-5.5 ∘ C for net nitrification) across all biomes [14]. Similarly, we found that experimental warming-induced change of soil temperature tended to be negatively correlated with that of TN ( 2 = 0.43, = 0.057) and positively correlated with that of MBN ( 2 = 0.43, = 0.056) [20]. In addition, MBN was significantly correlated with SOC, TN, DOC, and DON (Table 2). Therefore, the negligible responses of soil C and N pools to experimental warming (Table 1) may be also due to lower warming magnitude in this alpine meadow.
Microbial activity regulates the production of dissolved organic matter [5,8,26] and experimental warming-induced decline in soil moisture may suppress soil microbial activity [20,27]. Similarly, we also found that soil C and N pools increased with increasing soil microbial biomass and soil moisture ( Figure 2, Table 2). Moreover, short-term experimental warming had little effect on soil microbial biomass in this system [20]. Therefore, the negligible responses of SOC, TN, DOC, and DON to short-term experimental warming may be also related to that of soil microbial biomass [8,20]. Moreover, experimental warming-induced soil drying may also suppress the production of DOC and DON [8,20].

Effects of Experimental Warming on Soil Inorganic N.
Bai et al. [14] demonstrated that experimental warming did not significantly increase net N nitrification in grasslands. Similarly, experimental warming did not increase net N mineralization in an alpine meadow on the Tibetan Plateau [28]. In the same alpine meadow as this study, the finding that experimental warming did not increase ecosystem photosynthesis and aboveground plant biomass [22] also indirectly supported that experimental warming may not increase soil N availability because it has been observed that plant productivity is positively correlated with net N mineralization [29]. Therefore, the negligible or negative effect of experimental warming on soil inorganic N ( Figure 1, Table 1) may result from the suppression of net N mineralization and nitrification under warming. The suppression of net N mineralization and nitrification may be owing to decreases in soil moisture and microbial activity because N min , NH 4 + -N, and NO 3 − -N increased significantly with increasing soil moisture and microbial biomass ( Figure 2, Table 2). Similarly, the experimental warming-induced significant reductions or insignificant changes of inorganic N ( Figure 1, Table 1) were also partly attributed to experimental warming-induced decline in soil microbial biomass [20] and soil drying [10,29,30]. This was in line with the finding that the effect of experimental warming on soil moisture was significantly correlated with that on soil nitrification [14]. On the other hand, microbial biomass was more closely related to soil inorganic N than soil moisture (Table 3). This implied that microbial biomass may dominate the variation of soil inorganic N in this study. However, our previous study showed that short-term experimental warming tended to reduce microbial biomass due to soil drying in the same alpine meadow as this study [20]. Therefore, the experimental warming-induced changes of soil inorganic N, net N mineralization, and nitrification may be directly related to that of microbial activity and indirectly related to that of soil moisture.
The different responses of N min to experimental warming among the three elevations across the sampling dates could be attributed to several probable underlying mechanisms. First, DON is high-quality N source for N mineralization [8,31]. This was supported by the positive relationships between DON and N min and NH 4 + -N and NO 3 − -N ( Figure 3). DON under warmed plots tended to be decreased by 10.3% at the low elevation and by 28.7% at the mid-elevation but to be increased by 4.4% at the high elevation across all the three sampling dates, compared to control plots. Second, experimental warming-induced different changes in soil microbial biomass N (MBN) among three elevations [20] could partly explain this phenomenon considering that the production of DON and the immobilization of soil inorganic N were regulated by MBN [3,32,33]. This viewpoint was confirmed by the positive correlations between MBN and DON, N min , NH 4 + -N, and NO 3 − -N (Table 2). Third, the response of soil N availability to warming could be strongly related to the initial conditions [8,34]. In our system, N min , DON, and microbial biomass at the high elevation were significantly larger compared to the low and mid-elevation, whilst there were insignificant differences between the latter two [20].

Conclusions
In summary, short-term experimental warming had no obvious effects on topsoil organic C, total N, dissolved organic C, and N pools for the alpine meadow in this study. The insignificant responses of these C and N pools to warming may be due to short-term warming treatment, experiment warminginduced soil drying, and lower warming magnitude. In contrast, the response of soil inorganic N to experimental warming differed among the three elevations, which may be attributed to different response trends of dissolved organic N and microbial biomass and different initial soil inorganic N.