Correlation Between CO2 Efflux and Net Nitrogen Mineralization and Its

Home , Priming (microbiology)

Pedosphere 21(5): 666–675, 2011 ISSN 1002-0160/CN 32-1315/P c 2011 Soil Science Society of China Published by Elsevier B.V. and Science Press

Correlation Between CO2 Eﬄux and Net Nitrogen Mineralization and Its Response to External C or N Supply in an Alpine Meadow Soil∗1

SONG Ming-Hua1,2,∗2, JIANG Jing1,3, XU Xing-Liang1 and SHI Pei-Li1 1Key Laboratory of Ecosystem Network Observation and Modelling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101 (China) 2State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008 (China) 3Graduate University of Chinese Academy of Sciences, Beijing 100049 (China) (Received November 30, 2010; revised June 15, 2011)

ABSTRACT In nutrient-limited alpine meadows, nitrogen (N) mineralization is prior to soil microbial immobilization; therefore, increased mineral N supply would be most likely immobilized by soil microbes due to nutrient shortage in alpine soils. In addition, low temperature in alpine meadows might be one of the primary factors limiting soil organic matter decomposition and thus N mineralization. A laboratory incubation experiment was performed using an alpine meadow soil from the Tibetan Plateau. Two levels of NH4NO3 (N) or glucose (C) were added, with a blank without addition of C or N as the control, ◦ before incubation at 5, 15, or 25 C for 28 d. CO2 efflux was measured during the 28-d incubation, and the mineral N was measured at the beginning and end of the incubation, in order to test two hypotheses: 1) net N mineralization is negatively correlated with CO2 efflux for the control and 2) the external labile N or C supply will shift the negative correlation to positive. The results showed a negative correlation between CO2 efflux and net N immobilization in the control. External inorganic N supply did not change the negative correlation. The external labile C supply shifted the linear correlation from negative to positive under the low C addition level. However, under the high C level, no correlation was found. These suggested that the correlation of CO2 efflux to net N mineralization strongly depend on soil labile C and C:N ratio regardless of temperatures. Further research should focus on the effects of the types and the amount of litter components on interactions of C and N during soil organic matter decomposition. Key Words: C:N ratio, inorganic N, labile C, organic matter, temperature

Citation: Song, M. H., Jiang, J., Xu, X. L. and Shi, P. L. 2011. Correlation between CO2 eﬄux and net nitrogen mineralization and its response to external C or N supply in an alpine meadow soil. Pedosphere. 21(5): 666–675.

INTRODUCTION 1994). The connection between C and net N mineralization is reﬂected in most N mineralization models, Mineralization of soil organic matter (SOM) re- which assume constant C:N ratios of the microorga- leases carbon (C) and nitrogen (N) to the soil and en- nisms or variation of C:N ratios within certain limits vironment. SOM mineralization is directly relevant to (Parnas, 1975; Neill and Gignoux, 2006). Other mo- the increasing CO2 concentration in the atmosphere dels are based on empirical relationships between net leading to global warming. N mineralization poten- N and C mineralization (Bosatta and Staaf, 1982). tial of soil is often used as an index of N availability to Net N mineralization is the result of two opposing plants in terrestrial ecosystems (Palm et al., 1993). Net processes: gross N mineralization and gross N immobi- C and N mineralization have often been studied to- lization. Therefore, net N mineralization is diﬃcult to gether due to the closely related processes (Hart et al., predict. In a great number of decomposing litter ma-

∗1Supported by the National Basic Research Program (973 Program) of China (Nos. 2010CB951704 and 2010CB833502), the National Natural Science Foundation for Young Scientists of China (No. 30600070), and the West Light Joint Scholarship of the Chinese Academy of Sciences. ∗2Corresponding author. E-mail: [email protected]. SOIL CO2 EFFLUX AND N MINERALIZATION 667 materials, two distinct phases can be distinguished: A nutrient-limited environment during the entire growing period of N accumulation is followed by a period of net season, with particular deficiencies in mineral N at the N release (Barrett and Burke, 2000; Bengtsson et al., beginning of the growing season (Zhou, 2001). Soil CO2 2003). The duration and magnitude of N accumula- efflux increased from July to August in the growing tion depends on the C/N ratio of the substrates (Bar- period, and decreased in September in the plant senes- rett and Burke, 2000; Bengtsson et al., 2003; Craine et cence period (Hu et al., 2008). al., 2007). SOM decomposition is modeled as a first- The relationships between soil respiration and net order process, with a constant decomposition rate per N mineralization and to what extent the correlation gram soil organic matter (SOM) (Van Veen and Paul, is regulated by external C or N supply are still not 1981; Parton et al., 1987; Moorhead and Reynolds, clear. In this study an incubation experiment under 1991). CO2 efflux rate decreases exponentially as the controlled condition was conducted to investigate 1) available fractions of soil C are consumed in the pro- the effects of temperature and external C or N supply cesses of SOM decomposition (Paul et al., 1998). Cor- on soil CO2 efflux and net N mineralization; 2) the relations between net N mineralization and CO2 pro- relationship between soil CO2 efflux and net N mine- duction were found to be positive (Scott et al., 1998; ralization. Two hypotheses were tested: 1) N minera- Parfitt et al., 2003), negative (Scott et al., 1998), or lization is characterized by an extensive soil microbe none (Kelliher et al., 2004). Murphy et al. (2003) have immobilization in nutrient-limited alpine meadow soils suggested that the relationship between mineralized N and therefore the correlation between soil respiration and respiration largely depends on the C:N ratio of the and net N mineralization should be negative during a decomposing pool and the microbial C use efficiency. short-term incubation and 2) due to the priming effect Basic stoichiometric decomposition theory predicts of labile N or C additions on N mineralization (Bosatta that ecosystems will store more C if increasing atmo- and Staaf, 1982), the external labile N or C supply will spheric CO2 leads to greater substrate C:N ratios (Hes- shift the correlation from negative to positive. sen et al., 2004), while increasing N availability would MATERIALS AND METHODS decrease soil C storage (Mack et al., 2004). The common idea is that N addition causes the priming ef- Site description fect (Kuzyakov et al., 2000) on old soil organic matter decomposition due to enhanced microbial activity Soil samples for this study were collected from (Westerman and Tucker, 1974). However, some studies an experimental site at the Haibei Research Station showed that N addition resulted in decline in decom- of Alpine Meadow Ecosystem, Chinese Academy of position and CO2 evolution (Westerman and Tucker, Sciences, located in the northeastern region of the 1974; Wang et al., 2004). Qinghai-Tibet Plateau (37◦ 32 N, 101◦ 15 E, 3 240 The Tibetan Plateau is one of the most sensitive m above sea level). Annual precipitation is 618 mm, places to global changes (IPCC, 2007). Fluctuations of which 85% is concentrated in the growing season in temperature have had clear effect on the Tibetan from May to September in the study site. Mean an- Plateau during the past 40 years. For example, the nual temperature is −1.7 ◦C. The soil is classified rate of increase in temperature amounts to 0.016 ◦C as a Mat Cry-gelic Cambisol (Chinese Soil Taxonomy year−1, which is higher than that of other regions in Research Group, 1995) corresponding to Gelic Cam- China; the average rate of temperature increase in bisol (FAO/ISRIC/ISSS, 1998). A detailed description Chinaasawholeis0.004◦Cyear−1 (Tang et al., of the soil characteristics is given in Table I. The 1998). Average N deposition from precipitation is 4.6 study site is dominated by a perennial herb Kobre- kgNha−1 year−1 in alpine meadows (Jiang, 2010). The sia humilis Serg. (Cyperaceae). Other common species plateau covers approximately 2.5 million km2 with an include grasses such as Stipa aliena, Elymus nutans average altitude of more than 4 000 m. Approximately Griseb., and Festuca ovina Linn. and herbs such as 35% of the plateau is occupied by alpine meadows Saussurea superba Anth., Gentiana lawrencei Burk. (Zheng, 2000) and the decomposition rate of the soil var. farreri T. N. Ho, Gentiana straminea Maxim., is very low due to low temperature. N mineralized in Potentilla nivea Linn., Potentilla saundersiana Royle, this meadows is subject to microbial immobilization Scirpus distigmaticus Tang et Wang, Cyperaeae herbs during the growing season (−27 mg m−2 d−1)(Cao such as Kobresia pygmaea C. B. Clarke in Hook f., and and Zhang, 2001). Generally, alpine plants grow in a Carex sp. (Zhou, 2001). Total vegetation coverage is 668 M. H. SONG et al.

+ − TABLE I NH4 -N and NO3 -N with an auto analyzer (Bremner, 1965). Dissolved organic nitrogen (DON) was calcu- Some basic properties of the soil (top 10 cm) at the study n lated as the difference between the total N and ex- site ( = 6–8) + − changeable inorganic N (NH4 -N + NO3 -N) in the ex- a) Property Mean Standard error tracts. The basic properties of the soil studied are pre- pH 8.0 0.1 sented in Table I. Bulk density (g cm−3) 0.70 0.05 C:N ratio 19.6 0.3 Incubation experiment SOC (kg m−2) 11.8 0.3 Total N (kg m−2) 0.60 0.04 A 28-d incubation experiment was carried out in Microbial biomass N (g m−2)6.50.3 the laboratory. A two-way factorial design was used, −2 DON (g m )1.80.1including temperature, N addition, and C addition. −2 Extractable inorganic N (g m )1.4 0.4 The sieved soil samples equivalent to 30 g oven-dry Microbial biomass C:N ratio 6.2 0.8 weight were placed into 250-mL Schott jars after 5 mL a)SOC = soil organic carbon; DON = dissolved organic of deionized water was carefully mixed into them. The nitrogen. N addition levels were 0.024 and 0.118 mg NH4NO3 per subsample, equivalent to 0.005 (N1) and 0.025 (N2)mg greater than 95%. Rooting depth is shallow, with more Ng−1 soil. C addition levels were 5 and 25 mg glucose than 90% of the root mass concentrated within the top (C6H12O6) per subsample, equivalent to 2 (C1)and 15 cm (Zhou, 2001). The plants are grown under N- −1 10 (C2)mgCg soil. We choose N levels based on limited conditions, with a marked decrease in growth the mean mineralized N level in the alpine meadow occurring in the month of August (Zhou, 2001). soil during the growing season (0.0066 mg N g−1 soil) Soil sampling and analysis (Zhou, 2001). The C levels were based on the amount of litter accumulated in the alpine meadow soil (1 400 g Intact soil cores, 7.5 cm in diameter and 10 cm in m−2) and C:N ratio in litter materials (Zhou, 2001). A depth, were collected from five random points along 5 blank without N and C addition was taken as the con- arbitrarily laid transects on July 10, 2007. They were trol (CK). Small cups filled with 15 mL of 1 mol L−1 transported in cooled containers (4 ◦C) to the labo- NaOH were placed into the jars to trap the evolved ratory in Beijing. The five soil cores from each tran- CO2. The jars were fastened airtight and incubated at sect were mixed together in the laboratory. After the three different temperatures: 5, 15, and 25 ◦C. The roots and stones were removed, the soil samples were selection of soil temperatures were based on the range passed through a 2-mm sieve. The sieved soil samples of temperatures experienced during the growing sea- were kept at a temperature of 4 ◦C until incubation son in the study site. For each temperature, a series of started shortly thereafter. Soil gravimetric moisture blanks without soil were included, using cups only filled content was measured (105 ◦C, 24 h). Soil pH values with 15 mL of 1 mol L−1 NaOH in the Schott jars to were measured using a glass electrode at a soil-to-water trap the CO2 from the atmosphere enclosed in the jars. ratio of 1:2. Total N was measured using Kjeldahl di- Moisture of the samples was periodically adjusted to a gestion with a salicylic acid modification (Pruden et value of 30% water holding capacity (WHC) since this al., 1985). Soil organic C (SOC) was measured follo- is generally considered to imitate the moisture content wing the method described by Kalembasa and Jen- of the field conditions. The constant soil moisture was kinson (1973). Microbial biomass N was estimated u- maintained by weighing each sample once a week and sing chloroform fumigation direct extraction technique adjusting the water content to the target value. Four (Brookes et al., 1985; Davidson et al., 1998). Nitrogen replicates were setup for each treatment. The CO2 that −1 in 0.5 mol L K2SO4 extracts (soil:extractant, 1:4) evolved from the soil was measured at 1, 4, 7, 14, 21, was determined using Kjeldahl digestion of a salicylic and 28 d by titrating the NaOH solution against 0.1 −1 acid modification (Pruden et al., 1985). Exchangeable mol L HCl after the addition of BaC12. NaOH so- + − NH4 -N and NO3 -N were extracted with 20 mL 0.5 lution, incubated as described above, was also titrated −1 mol L K2SO4 on a rotary shaker for 1 h. After but without incorporating any soil (blank). Soil sam- centrifugation, the supernatant was filtered through a ples before and after incubation of 28 d were collected −1 filter paper, and the filtrates were then analyzed for and extracted with 0.5 mol L K2SO4 (10 g wet soil SOIL CO2 EFFLUX AND N MINERALIZATION 669

with 50 mL K2SO4). The extracts were analyzed for lization (Q10). Regression analysis was used to analyze + − NH4 -N and NO3 -N to determine the amount of N the relationship of CO2 efflux to net N mineralization. mineralized during the 28-d incubation. Because there is no significant effect of N addition on CO2 efflux and net N mineralization, we put the data Calculation and statistical analysis of the N1 and N2 treatments together for analysis of the correlation between CO2 efflux and net N mine- The following exponential function was used to de- ralization. All the statistical analyses were carried out scribe the temperature dependence of C mineraliza- using SPSS10.0 for Windows (SPSS Inc., USA). tion: RESULTS bT CMRme =CMRbae (1) Effects of temperature and external C or N addition on where CMRme is the measured C mineralization rate C mineralization −1 −1 (mg C g soil d ); CMRba is the basal C mineralization rate at 0 ◦C; T is the incubation temperature The SOM mineralization rate decreased over time. (◦C), and b is a parameter relevant to the increase in C decomposition fitted first-order decay equation very ◦ respiration or C mineralization with a 10 C increase in well (Table II). The coefficient C0, which indicates the temperature (temperature coefficient, Q10) as follows: initial labile C pool available to microbial activity in soil, was lower under the control (CK) and N addition Q 10b 10 =e (2) (N1 and N2) treatments than the treatments with C addition (C1 and C2), and it was much lower under The following first-order decay function was used the C1 treatment than C2 treatment at each of three to describe the size of C pools, such as added C and C temperatures (Table II). The parameter k,whichisthe in soil organic pools: decomposition factor of labile C, was the lowest under ◦ −kt the C2 treatment at 5 C, the highest under the C1 C = C0e (3) treatment at 15 ◦C, and much higher under C addi- ◦ where C is the CO2 efflux (primarily from microbes, tion than CK and N addition at 25 C(TableII). −1 mg C g soil), C0 is the amount of available labile C Cumulative CO2 efflux increased over time (Fig. in soil (mg C g−1 soil), k is the decay constant, and t 1). It was significantly affected by temperature and is the incubation time (d). external labile C addition (Table III). The cumulative CO2 efflux increased with increases in incubation CO2 efflux was calculated by dividing the amount temperature. In addition, it increased with C levels of CO2-C released by the days between every two sampling times. The amount of N mineralized (net N mi- under all three temperatures (Fig. 1). The significant neralization) was calculated as the difference between interaction between temperature and C addition re- + − sulted in further increase in total CO2 efflux (Table III; the amount of NH4 -N and NO3 -N in soil at the end of incubation and that before incubation was started Fig. 1). External N addition seemed to cause a little in- plus external NH+-N and NO−-N addition. Net N min- crease in CO2 efflux (Fig. 1); however, the increase was 4 3 P> . eralization rate was calculated from the amount of N not statistically significant ( 0 05) (Table III). The Q mineralized during the incubation divided by the incu- calculated 10 values were significantly higher under C F . bation days (28 d). addition than the control and N addition ( =3101, P<0.001), indicating that the labile external C sup- The general linear model (GLM) (Anderson, 1984) ply increased the sensitivity of SOM decomposition to was used to examine the effects of temperature and temperature (Table IV). external N or C supply on CO2 efflux and net N mineralization. Additionally, Tukey’s post hoc test (with a Effects of temperature and external C or N addition on confidence limit of 95%) was applied to determine dif- net N mineralization ferences in CO2 efflux and net N mineralization under N or C supply. A two-way analysis of variance Temperature and external C addition as well as (ANOVA) was used to analyze the effects of tempera- their interaction had significant effects on net N mine ture and external C or N supply on C decay coefficient ralization (Table III). Increases of temperature stimu- (C0 and k) and temperature sensitivity of C minera- lated microbial N immobilization rate. N immobiliza- 670 M. H. SONG et al.

TABLE II

−kt Coefficients obtained from fitting first-order decay equation, C = C0e , where C is the CO2 efflux, C0 is the amount of available labile C in soil, k is the decay constant, and t is the incubation time, for C decomposition in an alpine meadow ◦ soil with or without addition of N (NH4NO3) or C (glucose) incubated at 5, 15, and 25 C

a) 2d) Temperature C or N supply C0 kR ◦CmgCg−1 soil d−1 5CK 0.23± 0.09b) ac) 0.097 ± 0.003 b 0.852*** N1 0.24 ± 0.08 a 0.094 ± 0.002 b 0.849*** N2 0.23 ± 0.02 a 0.096 ± 0.001 b 0.769** C1 0.32 ± 0.09 b 0.097 ± 0.005 b 0.878*** C2 0.41 ± 0.06 c 0.047 ± 0.001 a 0.850*** 15 CK 0.20 ± 0.09 a 0.079 ± 0.004 a 0.609** N1 0.22 ± 0.05 a 0.089 ± 0.002 ab 0.646** N2 0.25 ± 0.04 a 0.088 ± 0.001 ab 0.694** C1 0.37 ± 0.03 b 0.099 ± 0.001 b 0.801*** C2 0.74 ± 0.06 c 0.082 ± 0.003 a 0.992*** 25 CK 0.24 ± 0.09 a 0.067 ± 0.001 a 0.659** N1 0.23 ± 0.02 a 0.069 ± 0.001 a 0.607** N2 0.23 ± 0.03 a 0.068 ± 0.001 a 0.580** C1 0.36 ± 0.06 b 0.079 ± 0.004 b 0.640** C2 0.91 ± 0.01 c 0.086 ± 0.002 b 0.916***

**, *** Significant at the 0.01 and 0.001 probability levels, respectively. a) −1 −1 −1 CK: the control without C or N addition; C1:2mgCg soil; C2:10mgCg soil; N1: 0.005 mg N g soil; N2: 0.025 mg N g−1 soil. b)Mean ± standard error. c)Means followed by the same letter(s) are not significantly different at P =0.05. d)Coefficient of determination. tion rate was significantly promoted by increasing C under each of the three temperatures, indicating that level (Fig. 2a). Compared with CK, the C1 treatment all mineral N added was immobilized by soil microbes significantly increased microbial immobilized N under (Table III, Fig. 2b). both 5 and 15 ◦C, but it did not change microbial im- ◦ mobilized N under 25 C. In the C2 treatment, there Relationships of CO2 efflux with N mineralization is a significant increase in N immobilization rate under each of the three temperatures (Fig. 2a). The external Soil CO2 efflux was negatively correlated to net N N did not bring any changes to N mineralization rate mineralization for the control without C or N addition

TABLE III

General linear model analysis for the eﬀects of temperature and external glucose (C) or NH4NO3 (N) addition on cumulative CO2-C eﬄux and net N mineralization during the 28-d incubation (n =4)

Variable Degree of Cumulative CO2-C eﬄux Net N mineralization freedom Type III sum of squares F value P value Type III sum of squares F value P value Temperature (T) 2 3.40 72.91 < 0.0001 0.56 20.61 < 0.0001 N supply (N) 2 0.12 2.65 0.089 0.04 1.37 0.27 T × N 4 0.12 1.32 0.29 0.12 2.11 0.11 Error 26 0.61 0.36 Temperature (T) 2 9.86 266.03 < 0.0001 0.69 5.06 0.016 C supply (C) 2 111.07 2 997.85 < 0.0001 62.34 458.54 < 0.0001 T × C 4 1.42 19.19 < 0.0001 4.56 16.76 < 0.0001 Error 21 0.39 1.43 SOIL CO2 EFFLUX AND N MINERALIZATION 671

mineralization, but the correlation changed to positive by the addition of C1 (Fig. 3). There was no signiﬁcant correlation between CO2 ﬂux and net N mineralization under the addition of C2 (Fig. 3).

Fig. 2 N mineralization rate in an alpine meadow soil with or without addition of glucose (C1 and C2) (a) or NH4NO3 ◦ (N1 and N2) (b) incubated at 5, 15, and 25 C. Vertical Fig. 1 Cumulative CO2 efflux in an alpine meadow soil bars represent standard error of the mean (n = 4). Bars with or without addition of glucose (C1 and C2)or ◦ with the same letter indicate no significant difference for NH4NO3 (N1 and N2) incubated at 5, 15, and 25 C n C or N addition (P<0.05). CK: the control without C or ( = 4). CK: the control without C or N addition; C1: −1 −1 −1 −1 N addition; C1:2mgCg soil; C2: 10mgCg soil; N1: 2mgCg soil; C2: 10mgCg soil; N1: 0.005 mg N −1 −1 −1 −1 0.005 mg N g soil; N2: 0.025 mg N g soil. g soil; N2: 0.025 mg N g soil. DISCUSSION (Fig. 3), which meant that the more the increase in CO2 efflux, the more the inorganic N immobilized by Our results showed that both soil respiration and the soil microbes. External N addition did not change net N mineralization were sensitive to temperatures. the negative correlation between CO2 efflux and net N The higher temperatures increased CO2 efflux and N

Fig. 3 Relationships between CO2 efflux and net N mineralization in an alpine meadow soil with or without addition of glu- ◦ cose (C1 and C2)orNH4NO3 (N1 and N2) incubated at 5, 15, and 25 C(n = 4). CK: the control without C or N addition; −1 −1 −1 −1 C1:2mgCg soil; C2:10mgCg soil; N1: 0.005 mg N g soil; N2: 0.025 mg N g soil. Because there is no significant effect of N addition on CO2 efflux and net N mineralization, all the data of N1 and N2 were put together. 672 M. H. SONG et al.

TABLE IV tion in this study, external N supply did not change either CO2 efflux or net N mineralization, which indi- Parameters obtained from fitting temperature-dependent functions to the exponential equation of measured C min- cated that the N added was totally immobilized by soil bT eralization rate (CMRme), CMRme =CMRbae , where microbes. ◦ CMRba is the basal C mineralization rate at 0 C; T is the Nutrients in mineral forms are taken up by the de- incubation temperature, and b is a parameter relevant to composers (immobilized). Typically, net N release (net the increase in C mineralization with a 10 ◦C increase in mineralization) in mineral forms (ammonium and ni- temperature (temperature coefficient, Q10), and Q10 val- trate) only occurs after N demand of soil microbes was ues for an alpine meadow soil with or without addition of satisfied (Berg and McClaugherty, 2003). The biologi- glucose (C1 and C2)orNH4NO3 (N1 and N2) incubated at cal degradation is primarily carried out by microbial different temperatures decomposers, including bacteria and fungi and their 2a) NorC C0 bRQ10 value grazers, which have lower C:N ratios compared with supplya) most litter types (Berg and McClaugherty, 2003). This mg C g−1 soil d−1 creates a high N demand, and even though a conside- CK 2.213 0.0163 0.999 1.177 ac) rable fraction of assimilated C is respired, the decom- N1 2.425 0.0121 1 1.128 a posers often still require inorganic N at least at the N2 2.536 0.0128 0.952 1.136 a early phases of decomposition. The decomposer C:N C1 3.004 0.0203 1 1.225 b ratio and the respiration rate (complementary to the C2 5.188 0.0256 0.994 1.292 b C-use efficiency) define the actual nutrient requirement a) CK: the control without C or N addition; C1:2mgC of the decomposers (Bosatta and Staaf, 1982; Craine et −1 −1 −1 g soil; C2:10mgCg soil; N1: 0.005 mg N g soil; al., 2007; Manzoni and Porporato, 2007). The C:N ra- −1 N2: 0.025 mg N g soil. tio was 13–34 in the senesced plant materials, 10–20 in b) Coefficient of derermination. the soil, and 4–7 in the soil microbial biomass for the c) Values followed by the same letter are not significantly alpine meadow studied. The lower microbial biomass different at the 0.05 probability level. C:N ratio in the soil made the soil microbes immobi- immobilization. This suggested that low temperature lize all available N during the short-term incubation to was an important factor limiting soil organic matter suffice to meet their high N demand. (SOM) decomposition and nutrient release in alpine Parton et al. (2007) analyzed global-scale N release meadow soils. Another factor limiting SOM decompo- patterns. They found similarities in N release patterns sition was lack of labile C components. Increased CO2 during long-term decomposition. The initial N concen- efflux and N immobilization under external labile C tration of leaf litter is a dominant driver of net N immo- supply supported the above-mentioned results. Previ- bilization and release during long-term litter decompo- ous studies showed that the supply of labile C through sition at global scale, regardless of climate, other litter leaf and root turnover is an important source of la- chemical properties, edaphic conditions, and soil mi- bile C substrate that drives the growth and activity crobial communities. Their results also suggest that N of heterotrophic soil microorganisms (Hooker et al., can be released early in decomposition of high-quality 2008). Consumption of labile C results in increase of litters in environments that support low decomposi- microbial N demand and immobilization of available N tion rates. The time required to initiate net N release (Hart et al., 1994; Schaeffer and Evans, 2005). Howe- was predicted from initial N of the litter and the cli- ver, N supply may result in greater N availability, matic condition across biomes (Parton et al., 2007). In which is likely to lead to a net decrease in decom- our study, both labile C supply and increased tempera- position rates (Hart et al., 1994; Saetre and Stark, ture promoted microbe N immobilization (Fig. 1). Net 2005). Microbial N mining is the process whereby some N release did not happen during the 28-d incubation microbes use labile C to decompose recalcitrant organic period, which might need more mineral N as well as a matter in order to acquire N (Fontaine and Barot, longer time period. 2005; Moorhead and Sinsabaugh, 2006). With greater In our study, the relationship between CO2 efflux N availability, the process of recalcitrant C mining for and net mineralization was analyzed across tempera- N was suppressed, which might explain observed de- tures but not time. The results showed that the cor- clines in decomposition with N addition (Hagedorn et relations of CO2 efflux to net N mineralization were al., 2003; Wang et al., 2004). During the 28-d incuba- relevant to substrate amendment regardless of changes SOIL CO2 EFFLUX AND N MINERALIZATION 673

in temperatures. Gross N immobilization has been (Q10) was higher under the C addition than under the found to be correlated to respiration (Hart et al., 1994; N addition and control, suggesting that C mineraliza- Recous et al., 1999; Barrett and Burke, 2000; Bengts- tion rates were more sensitive to changes in soil labile son et al., 2003) because the microbial biomass needs C content. The correlation between CO2 efflux and net N and energy for growth. Gross N mineralization has N immobilization was linearly negative in the control also been found to be linked to the respiration in soils without C or N addition. Inorganic N amendment did (Hart et al., 1994; Bengtsson et al., 2003). Whether not change the negative correlation; however, labile C net N mineralization or net N immobilization occurs amendment shifted the linear correlation from negative is closely linked with C availability and has been re- to positive under the low C level. Under the high C lated to the C:N ratio of the decomposing organic level, no correlation was found. Our results suggested matter (van Veen et al., 1984). The correlations be- that the correlation of CO2 efflux and net N minerali- tween respiration and net N mineralization in the pre- zation strongly depended on soil C content and C:N vious works are summarized as follows. In an N-limi- ratios. ted soil, negative correlations between respiration and net N mineralization were observed (Staaf and Berg, ACKNOWLEDGEMENTS 1977; Kowalenko et al., 1978; Foster et al., 1980; John- We thank Professor Yakov Kuzyakov from the son et al., 1980; Nömmik and Möller, 1981; Bosatta AgroEcoSystem Research Department, University of and Berendse, 1984; Schimel, 1986). Positive correla- Bayreuth, Germany, for his great help in improving the tions were detected on soils limited by C availability English of this manuscript. We thank three anonymous or supplied with labile C (Bosatta and Berendse, 1984; reviewers for their insight comments and suggestions. Schimel et al., 1986; Gao et al., 2009). No correlations were observed on soils with great variability in soil C:N ratios during decomposition (Hart et al., 1994; Trin- REFERENCES soutrot et al., 2000; Jensen et al., 2005; Luxhøi et al., Anderson, T. W. 1984. An Introduction to Multivariate 2006). Soil organic substrates with high C:N ratios of- Statistical Analysis. 2nd ed. Wiley, New York. ten support microbial communities that are N limi- Barrett, J. E. and Burke, I. C. 2000. Potential nitrogen ted and generally exhibit higher rates of N immobi- immobilization in grassland soils across a soil organic lization because microbes require additional mineral N matter gradient. Soil Biol. Biochem. 32: 1707–1716. to metabolize materials with high C content (Janssen, Bengtsson, G., Bengtson, P. and M˚ansson, K. F. 2003. 1996). Higher N immobilization should be accompa- Gross nitrogen mineralization, immobilization, and ni- nied by high rates of C mineralization since C sub- trification rates as a function of soil C/N ratio and microbial activity. Soil Biol. Biochem. 35: 143–154. strates presumably facilitate N retention by microbes Berg, B. and McClaugherty, C. A. 2003. Plant Litter: De- (Barrett and Burke, 2000). Negative correlations be- composition, Humus Formation, Carbon Sequestration. tween CO2 efflux and net N mineralization were ob- Springer, Berlin. served in the N-limited alpine meadow soil in our study. Bosatta, E. and Berebdse, F. 1984. Energy or nutri- C but not N supply changed the shape of the CO2 ent regulation of decomposition: implications for the efflux vs. net N mineralization curve, which is in ac- mineralization-immobilization response to perturba- cordance with the theory that inorganic N does not tions. Soil Biol. Biochem. 16: 63–67. appear in the definition of critical C:N ratio (Bosatta Bosatta, E. and Staaf, H. 1982. The control of nitrogen 39 and Staaf, 1982; Manzoni et al., 2008). The critical turnover in forest litter. Oikos. : 143–151. Bremner, J. M. 1965. Inorganic forms of nitrogen. In ratio is affected by organic C supply, which can thus Black, C. A. (ed.) Methods of Soil Analysis. Volume change the relationship between soil respiration and 2. American Society of Agronomy, Madison. pp. 1179– net N mineralization. 1237. Brookes, P. C., Landman, A., Pruden, G. and Jenkinson, D. CONCLUSIONS S. 1985. Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure External inorganic N addition had no significant microbial biomass nitrogen in soil. Soil Biol. Biochem. effect on soil CO2 efflux and net N mineralization. 17: 837–842. However, external labile C addition significantly in- Craine, J. M., Morrow, C. and Fierer, N. 2007. Microbial creased CO2 efflux and microbial immobilization, es- nitrogen limitation increases decomposition. Ecology. pecially for the high C dose. Temperature coefficient 88(8): 2105–2113. 674 M. H. SONG et al.

Cao, G. M. and Zhang, J. X. 2001. Soil nutrition and sub- and plant quality in a broad range of plant materials stance cycle of Kobresia meadow. In Zhou,X.M.(ed.) on C and N mineralization in soil. Plant Soil. 273: Chinese Kobresia Meadows (in Chinese). Science Press, 307–326. Beijing. pp. 188–216. Johnson, D. W., Edwards, N. T. and Todd, D. E. 1980. Ni- Chinese Soil Taxonomy Research Group. 1995. Chinese trogen mineralization immobilization and nitrification Soil Taxonomy (in Chinese). Science Press, Beijing. following urea fertilization of a forest soil under field Davidson, E. A., Belk, E. and Boone, R. D. 1998. Soil water and laboratory conditions. Soil Sci. Soci. Am. J. 44: content and temperature as independent or confounded 610–616. factors controlling soil respiration in a temperate mixed Jiang, C. M. 2010. The effects of N deposition on C and N hardwood forest. Glob. Change Biol. 4: 217–227. cycles in alpine meadow ecosystem of Haibei on the Ti- FAO/ISRIC/ISSS. 1998. World Reference Base for Soil betan Plateau (in Chinese). Ph. D. Thesis, Institute of Resources. FAO/ISRIC/ISSS, Rome. Geographic Sciences and Natural Resources Research, Fontaine, S. and Barot, S. 2005. Size and functional diver- Chinese Academy of Sciences. sity of microbe populations control plant persistence Kalembasa, S. J. and Jenkinson, D. S. A. 1973. Compara- and long-term soil carbon accumulation. Ecol. Lett. 8: tive study of titrimetric and gravimetric methods for 1075–1087. determination of organic carbon in soil. J. Sci. Food Foster, N. W., Beauchamp, E. G. and Corke, C. T. 1980. Agr. 24: 1085–1090. Microbial activity in a Pinus banksiana Lamb. forest Kelliher, F. M., Ross, D. J., Law, B. E., Baldocchi, D. D. floor amended with nitrogen and carbon. Can. J. For. and Rodda, N. J. 2004. Limitations to carbon min- Res. 60: 199–209. eralization in litter and mineral soil of young and old Gao, J. Q., Ouyang, H., Xu, X. L., Zhou, C. P. and ponderosa pine forests. Forest Ecol. Manag. 191: 201– Zhang, F. 2009. Effects of temperature and water satu- 213. ration on CO2 production and nitrogen mineralization Kowalenko, C. G., Ivarson, K. C. and Cameron, D. R. in alpine wetland soils. Pedosphere. 19: 1–8. 1978. Effect of moisture content, temperature and ni- Hagedorn, F., Spinnler, D., Bundt, M., Blaser, P. and Sieg- trogen fertilization on carbon dioxide evolution from wolf, R. 2003. The input and fate of new C in two field soils. Soil Biol. Biochem. 10: 417–423. forest soils under elevated CO2. Global Change Biol. Kuzyakov, Y., Friedel, J. K. and Stahr, K. 2000. Review of 9: 862–872. mechanisms and quantification of priming effects. Soil Hart, S., Stark, J., Davidson, E. and Firestone, M. 1994. Biol. Biochem. 32: 1485–1498. Nitrogen mineralization, immobilization, and nitrifica- Luxhøi, J., Brunn, S., Stenberg, B., Breland, T. A. and tion. Soil Sci. Soc. Am. J. 5: 983–1018. Jensen, L. S. 2006. Prediction of gross and net mi- Hessen, D. O., Agren, G. I., Anderson, T. R., Elser, J. J. neralization-immobilization-turnover from respiration. and De Ruiter, P. C. 2004. Carbon sequestration in Soil Sci. Soci. Am. J. 70: 1121–1128. ecosystems: The role of stoichiometry. Ecology. 85: Mack, M. C., Schuur, E. A. G., Bret-Harte, M. S., Shaver, 1179–1192. G. R. and Chapin III, F. S. 2004. Ecosystem carbon Hooker, T. D., Stark, J. M., Norton, U., Leffler, A. J., storage in arctic tundra reduced by long-term nutrient Peek, M. and Ryel, R. 2008. Distribution of ecosystem fertilization. Nature. 433: 440–443. C and N within contrasting vegetation types in a semi- Manzoni, S. and Porporato, A. 2007. A theoretical analy- arid rangeland in the Great Basin, USA. Biogeochem. sis of nonlinearities and feedbacks in soil carbon and 90: 291–308. nitrogen cycles. Soil Biol. Biochem. 39: 1542–1556. Hu, Q. W., Wu, Q., Cao, G. M., Li, D., Long, R. J. and Manzoni, S., Jackson, R. B., Trofymow, J. A., Porporato, Wang, Y. S. 2008. Growing season ecosystem respi- A. and Manzoni, S. 2008. The global stoichiometry of rations and associated component fluxes in two alpine litter nitrogen mineralization. Science. 321: 684–686. meadowsontheTibetanPlateau.J. Integr. Plant Biol. Murphy, D. V., Recous, S., Stockdale, E. A., Fillery, I. R. 50: 271–279. P., Jensen, L. S., Hatch, D. J. and Goulding, K. W. T. IPCC. 2007. Climate Change 2007. Synthesis Report. 2003. Gross nitrogen fluxes in soil: Theory, measure- A Contribution of Working Groups I, II, and III to ment and application of 15N pool dilution techniques. the Fourth Assessment Report of the Intergovernmen- Adv. Agron. 79: 69–118. tal Panel on Climate Change. Cambridge University Moorhead, D. L. and Reynolds, J. F. 1991. A general model Press, Cambridge. of litter decomposition in the northern Chihuahuan Janssen, B. H. 1996. Nitrogen mineralization in relation desert. Ecol. Model. 56: 197–219. to C-N ratio and decomposability of organic materials. Moorhead, D. L. and Sinsabaugh, R. L. 2006. A theoretical Plant Soil. 181: 39–45. model of litter decay and microbial interaction. Ecol. Jensen, L. S., Salo, T., Palmason, F., Breland, T. A., Hen- Monog. 76: 151–174. riksen, T. M., Stenberg, B., Pedersen, A., Lundström, Neill, C. and Gignoux, J. 2006. Soil organic matter decom- C. and Esala, M. 2005. Influence of biological origin position driven by microbial growth: a simple model for SOIL CO2 EFFLUX AND N MINERALIZATION 675

a complex network of interactions. Soil Biol. Biochem. Schimel, D. S., Parton, W. J., Adamsen, F. J., Wood- 38: 803–811. mansee, R. G., Senft, R. I. and Stillwell, M. A. 1986. Nömmik, H. and Mliller, G. 1981. Nitrogen recovery in The role of cattle in the volatile loss of nitrogen from a soils and needle biomass after fertilization of a Scats shortgrass steppe. Biogeochemistry. 2: 39–52. pine stand and growth responses obtained. Stud. For. Scott, N. A., Parfitt, R. L., Ross, D. J. and Salt, G. J. 1998. Suecia. 159: 5–37. Carbon and nitrogen transformations in New Zealand Palm, C., Robertson, G. P. and Vitousek, P. M. 1993. Ni- plantation forest soils from sites with different N status. trogen availability. In Anderson, J. M. and Ingram, J. Can. J. For. Res. 28: 967–976. S. I. (eds.) Tropical Soil Biology and Fertility: A Hand- Staaf, H. and Berg, B. 1977. Mobilization of plant nutri- book of Methods. 2nd ed. CAB International, Oxford. ents in a Scats pine forest mor in central Sweden. Silva pp. 158–163. Fenn. 11: 210–217. Paul, E. A., Collins, H. P. and Haile-Mariam, S. 1998. An- Tang, M. C., Cheng, G. D. and Lin, Z. Y. 1998. Con- alytical determination of soil C dynamics. Trans. Int. temporary Climatic Variations over Tibetan Plateau 16 Congr. Soil Sci. : 1–8. and Their Influences on Environments (in Chinese). Parfitt, R. L., Scott, N. A., Ross, D. J., Salt, G. J. and Tate, Guangdong Science and Technology Press, Guangzhou. K. R. 2003. Land-use change effects on soil C and N Trinsoutrot, I., Recous, S., Bentz, B., Linères, M., Chèneby, transformations in soils of high N status: comparisons D. and Nicolardot, B. 2000. Biochemical quality of crop under indigenous forest, pasture and pine plantation. residues and carbon and nitrogen mineralization kine- Biogeochemistry. 66: 203–221. tics under nonlimiting nitrogen conditions. Soil Sci. Parton, W., Silver, W. L., Burke, I. C., Grassens, L., Har- Soc. Am. J. 64: 918–926. mon, M. E., Currie, W. S., King, J. Y., Carol Adair, E., Van Veen, J. A., Ladd, J. N. and Frissel, M. J. 1984. Mode- Brandt, L. A., Hart, S. C. and Fasth, B. 2007. Global- lling C and N turnover through the microbial biomass scale similarities in nitrogen release patterns during in soil. Plant Soil. 76: 257–274. long-term decomposition. Science. 315(19): 361–364. van Veen, J. A. and Paul, E. A. 1981. Organic carbon Parnas, H. 1975. Model for decomposition of organic ma- dynamics in grassland soils. 1. Background informa- terial by microorganisms. Soil Biol. Biochem. 7: 161– tion and computer simulation. Can. J. Soil Sci. 61: 169. 185–201. Pruden, G., Powlson, D. S. and Jenkinson, D. S. 1985. The measurement of 15N in soil and plant material. Fertil. Wang, W. J., Baldock, J. A., Dalal, R. C. and Moody, Res. 6: 205–218. P. W. 2004. Decomposition dynamics of plant materi- Recous, S., Aita, C. and Mary, B. 1999. In situ changes in als in relation to nitrogen availability and biochemistry gross N transformations in bare soil after addition of determined by NMR and wet-chemical analysis. Soil 36 straw. Soil Biol. Biochem. 31: 119–133. Biol. Biochem. : 2045–2058. Saetre, P. and Stark, J. M. 2005. Microbial dynamics and Westerman, R. L. and Tucker, T. C. 1974. Effects of salts carbon and nitrogen cycling following re-wetting of soils plus nitrogen-15-labelled ammonium chloride on mi- beneath two semi-arid plant species. Oecologia. 142: neralization of soil nitrogen, nitrification and immobi- 247–260. lization. Soil Sci. Soci. Amer. Proc. 38: 602–605. Schaeffer, S. M. and Evans, R. D. 2005. Pulse additions of Zheng, D. 2000. Mountain Geoecology and Sustainable De- soil carbon and nitrogen affect soil nitrogen dynamics in velopment of the Tibetan Plateau. Kluwer, Dordrecht. an arid Colorado Plateau shrubland. Oecologia. 145: Zhou, X. M. 2001. Chinese Kobresia Meadows (in Chinese). 425–433. Science Press, Beijing.