Legacy Effect of Long-Term Elevated CO 2 and Warming on Soil Properties Controls Soil Organic Matter Decomposition

: Plant litter quality is one of the key factors that control soil organic matter (SOM) decomposition. Under climate change, although signiﬁcant change in litter quality has been intensively reported, the effect of litter quality change on SOM decomposition is poorly understood. This limits our ability to model the dynamics of soil carbon under climate change. To determine the effect of litter quality and soil property change on SOM decomposition, we performed a controlled, reciprocal transplant and litter decomposition experiments. The soils and plant litters were collected from a long-term ﬁeld experiment, where four treatments were designed, including: (1) the control without warming at ambient CO 2 ; (2) elevated atmospheric CO 2 up to 500 ppm (C); (3) warming plant canopy by 2 ◦ C (T); (4) elevated CO 2 plus warming (CT). We found that elevated CO 2 and warming altered the litter quality signiﬁcantly in terms of macronutrients’ content and their stoichiometry. Elevated CO 2 decreased the concentration of N in rice and wheat straw, while warming decreased the concentration of N and K in wheat straw. However, the change in plant litter quality did not lead to a shift in SOM decomposition. On the contrary, the legacy effect of long-term elevated CO 2 and warming on soil properties dominated the decomposition rate of SOM. Elevated atmospheric CO 2 suppressed SOM decomposition mainly by increasing phosphorous availability and lowering the soil C/N, fungi/bacteria ratio, and N-acetyl-glucosaminidase activity, while warming or elevated CO 2 plus warming had no effect on SOM decomposition. Our results demonstrated that the changes in soil property other than litter quality control the decomposition of SOM under climate change, and soil property change in respond to climate change should be considered in model developing to predict terrestrial soil carbon dynamics under elevated atmospheric CO 2 and warming.


Introduction
Climate change, mainly characterized by the rapid increase in the atmospheric CO 2 concentration and the elevation of global surface temperature, is challenging the sustainable development of global agriculture.The concentration of CO 2 in the atmosphere has been increasing since the 1840s, and it has exceeded 400 ppm in 2013 [1].In the meantime, the global temperature is continuously rising.It is predicted that the atmospheric CO 2 concentration will exceed 700 ppm [2], and the global temperature will increase by 1.1-6.4• C by the end of this century [3].
Soil organic matter (SOM) in terrestrial ecosystems plays an important role in the global carbon cycle.Approximately 2000 petagrams of carbon are stored in the top two meters of global soils as SOM, and more than twice as much carbon is stored in the soil as in the world's vegetation and atmosphere combined.Therefore, a slight change in SOM will have a profound impact on atmospheric CO 2 concentration, which in turn influences the global climate.Although a great number of research studies have been conducted to investigate the effect of climate change on soil carbon cycling, it remains an open question whether elevated CO 2 and global warming will promote soil carbon sequestration [4].Several studies reported that elevated atmospheric CO 2 could increase soil organic carbon (SOC) storage by increasing net CO 2 uptake [5,6].Liu et al. (2018) and Luo et al. (2006) predicted that the SOC stock would increase by approximately 5%, although it is quite small compared to the increase in the rate of plant biomass carbon under elevated CO 2 [7,8].However, Koyama et al. (2018) found that elevated atmospheric CO 2 did not affect the SOC pool in a Mojave Desert ecosystem [9].Similar findings were reported in cropland and temperate grassland ecosystems [10,11].Furthermore, increased soil CO 2 flux under elevated CO 2 has been frequently reported [7].Kuzyakov et al. (2019) argued that elevated atmospheric CO 2 has no (or little) effect on the soil carbon pool but strongly increases the CO 2 fluxes and accelerates carbon cycles [12].Similarly to elevated CO 2 , recent metaanalyses have shown that global warming generally has no [13][14][15][16][17] or negative [13,15] effects on the SOC pool.Long-term warming decreased the SOC pool by stimulating microbial utilization of the recalcitrant C pool [13].However, most of the studies involved in these meta-analyses were conducted in forest or grassland ecosystems.It remains unclear whether global warming will affect the pools and fluxes of SOC in cropland ecosystems.This uncertainty limits our accurate prediction of soil carbon stock change under elevated CO 2 and warming.
The concentration of CO 2 in soil is much higher than that in the atmosphere (10-50 times), and elevated atmospheric CO 2 (+200 ppm) will probably not affect soil carbon cycling directly [18].Its effect on soil carbon cycling is indirect, through the plant growth changes.Elevated CO 2 and warming affect plant growth by altering leaf stomatal conductance and the photosynthesis rate [19].Elevated CO 2 can increase crop yield by photosynthesis rate and soil nutrients use efficiency increase [5].As the atmospheric CO 2 concentration increases, the nutrient conditions of grains and the shoot biomass will change accordingly.Therefore, some studies have predicted that plants would be exposed to a global nutrient imbalance with lower N contents or higher ratios of C:N and C:P in plant litters under elevated CO 2 [20,21].In addition to macronutrients, the micronutrients in plant litter will also decrease under elevated CO 2 [22].He et al. (2015) even found that elevated CO 2 and warming reduced the content of crude protein and the in vitro digestibility of wheat straw [23].Plant litter with different chemical properties would likely affect the decomposition rate of SOM.For example, Elias et al. (2020) found that plant litter with higher P content and lower lignin to N ratios decompose faster in a forest soil [24].Fanin et al. (2011) also found that the content of C, N and P and their stoichiometry in plant litters were important factors that regulate soil microbial respiration [25].However, under elevated CO 2 and warming, the effect of litter quality change on SOM decomposition has never been tested.
In addition to plant litter quality, the soil microbial community also regulates the decomposition of SOM.Under elevated CO 2 or warming, significant changes in soil microbial communities have been reported intensively [26][27][28].Several studies found that elevated CO 2 altered the soil microbial composition [29][30][31][32][33][34][35][36].Soils exposed to elevated CO 2 had higher relative abundances of fungi and higher enzyme activity [29,37], which led to more soil carbon loss [30,31,36,38].Lipson et al. (2005) observed that elevated CO 2 had no effect on bacterial diversity but increased fungal biomass in a Chaparral Ecosystem [33].Sun et al. (2021) found that the soil microbial community evolves from dominating K-strategists to r-strategists under elevated CO 2 , with decreasing ratios of fungi to bacteria, Gram-positive to Gram-negative bacteria and Acidobacteria to Proteobacteria [28].Warming generally increases the abundance of microorganisms related to soil carbon and nitrogen cycling, leading to soil carbon loss and greater N 2 O emissions [39,40].Some studies showed that warming reduced bacterial and fungal abundance in forest ecosystems [41,42].The soil microbial community structure was also altered by warming [43].Deslippe et al. (2012) found that warming decreased bacterial communities evenness while it increased fungal communities evenness [44].Cheng et al. (2017) showed that warming increased the relative abundance of key functional genes involved in soil carbon degradation [39].Sheik et al. (2011) found that warming increased the soil microbial population size but decreased diversity under wet conditions, whereas it reduced the microbial population size under drought conditions [45].Under elevated atmospheric CO 2 and warming, the abundance of some dominant phyla was significantly increased, and the effect of combined elevated CO 2 and warming on soil functional processes was similar to that of only elevated CO 2 [46].
Under elevated CO 2 or warming, the changes in the soil microbial community and plant litter quality have been observed as mentioned above.Understanding the effect of plant litter quality and the soil microbial community on soil organic carbon decomposition can help us to model soil carbon dynamics under elevated CO 2 and warming.To our knowledge, there was no report investigating the effect of plant litter quality and soil microbial community change on soil organic carbon mineralization under these elevated conditions.Three manipulated incubation experiments were conducted to answer the following questions: (1) Does plant litter quality (C:N and nutrient content) change affect SOM decomposition under elevated CO 2 and warming?(2) Does soil microbial community change affect SOM decomposition under elevated CO 2 and warming?(3) Does plant litter have a greater effect on SOM decomposition than soil microbial community?We hypothesized that plant litter with decreased quality under elevated CO 2 and warming would suppress SOM decomposition, whereas the change in the soil microbial community would promote SOM decomposition.The results of this study can be used in soil carbon cycling model developing to predict terrestrial carbon dynamics under future climate change of elevated CO 2 and warming more precisely.

Soils and Plants Litter
The soils and plants litter used in this study were collected from the long-term field experiment of Nanjing Agricultural University, which was located in Kangbo Village (31 • 30 48 N, 120 • 33 36 E), Changshu City, Jiangsu Province of China.The field experiment facility was constructed in 2010, and the objective of this facility was to simulate Free Air CO 2 Enrichment and plant canopy warming in an open field (Figure 1).The soil is a Gleyic Stagnic Anthrosol (WRB-FAO) derived from clayey lacustrine deposit and cultivated with summer rice-winter wheat rotation dating back hundreds of years.The basic properties of the topsoil before the experiment onset were: pH (H 2 O) 7.0, bulk density of 1.2 g cm −3 , and concentration of organic C and total N of 16.0 g kg −1 and 1.9 g kg −1 , respectively.There were four treatments, including elevated CO 2 up to 500 ppm (C), warming plant canopy by 2 • C (T), elevated CO 2 plus plant canopy warming (CT), and ambient CO 2 without warming as the control (Control).The soils were collected from the top 15 cm in June 2018, after 7 years of treatment.The plant litters (rice and wheat straw) were collected at harvest.Rice straw (Cultivar: Changyou 5) was collected in October 2017, and wheat straw (Cultivar: Yangmai 16) was collected in June 2018.

Experimental Design
Three incubation experiments were designed (Table 1).In the first experiment (Experiment Ⅰ), the soils from the control, C, T and CT treatments were incubated with the addition of crop straw from the control, C, T and CT treatments, respectively.In the second experiment (Experiment Ⅱ), the soils from the control were incubated with the addition of crop straw from the control, C, T and CT treatments.In the third experiment (Experiment Ⅲ), the soils from the control, C, T and CT treatments were incubated with the addition of crop straw from the control.All the treatments were replicated three times.
Table 1.Experimental design.Control represents the soils or litter that were collected from the ambient atmospheric CO2 without warming; C represents the soils or litter that were collected from elevated CO2; T represents the soils or litter that were collected from plant canopy warming; CT represents the soils or litter that were collected from CO2 plus warming.

Soils Litters Abbreviation
Fifty grams of air-dried soils were mixed with 0.06 g of rice straw, and the mixture was placed in a 500 mL flask.All flasks were incubated at 25 °C in the dark.The bottle was sealed with a cap, and two rubber tubes (16 cm and 7 cm in length) were inserted into the bottle cap.A three-way valve was sleeved above the rubber tube for fresh air and gas sample collection.To simulate the soil respiration process during the whole crop growing season in the study area, two soil water conditions were designed.The soils mixed with rice straw were incubated first at aerobic conditions with soil water content maintained at

Experimental Design
Three incubation experiments were designed (Table 1).In the first experiment (Experiment I), the soils from the control, C, T and CT treatments were incubated with the addition of crop straw from the control, C, T and CT treatments, respectively.In the second experiment (Experiment II), the soils from the control were incubated with the addition of crop straw from the control, C, T and CT treatments.In the third experiment (Experiment III), the soils from the control, C, T and CT treatments were incubated with the addition of crop straw from the control.All the treatments were replicated three times.
Table 1.Experimental design.Control represents the soils or litter that were collected from the ambient atmospheric CO 2 without warming; C represents the soils or litter that were collected from elevated CO 2 ; T represents the soils or litter that were collected from plant canopy warming; CT represents the soils or litter that were collected from CO 2 plus warming.Fifty grams of air-dried soils were mixed with 0.06 g of rice straw, and the mixture was placed in a 500 mL flask.All flasks were incubated at 25 • C in the dark.The bottle was sealed with a cap, and two rubber tubes (16 cm and 7 cm in length) were inserted into the bottle cap.A three-way valve was sleeved above the rubber tube for fresh air and gas sample collection.To simulate the soil respiration process during the whole crop growing season in the study area, two soil water conditions were designed.The soils mixed with rice straw were incubated first at aerobic conditions with soil water content maintained at 80% of the soil water holding capacity.Then, they were mixed with wheat straw (0.06 g) and incubated under flooded conditions.During aerobic incubation, gas sampling was performed on Days 1, 1.5, 2, 3, 4, 5, 6, 8, 9, 11, 13, 15, 17, 19, 23, 28, 33, 43 and 64.During anaerobic incubation, gas sampling was performed on Days 65, 65.5, 66, 66.5, 67.5, 69, 71, 73, 82, 89, 98, 115, 123, 131, 139 and 147.Gas samples were collected with a syringe 2 h after ventilation.

Soils
The concentration of CO 2 in the gas samples was detected in a gas chromatograph (Agilent 7890A).The emission rate of CO 2 was calculated using the following equation:

Plant and Soil Sample Analysis
Plant and soil samples were analyzed following the protocol described by Lu (2000) [47].The plant samples were digested with sulfuric acid and hydrogen peroxide.The contents of nitrogen, phosphorus and potassium in the digestion were determined by the micro-Kjeldahl determination method, colorimetric method and flame photometer method, respectively.Total organic carbon and total nitrogen were measured by a CNS Macro Elemental Analyzer (Elementar, Germany).Dissolved organic carbon (DOC) was extracted with 0.05 mol•L −1 K 2 SO 4 solution.The mixture was shaken at 180 r•min −1 for 30 min and then passed through a 0.45 µm filter.The concentration of DOC in the liquid was measured in a TOC analyzer.Soil pH was measured in distilled water (soil/water ratio of 1/2.5 w/w) with a pH meter (Seven Easy Mettler Toledo, China, 2008).Soil available K was extracted with 1.0 mol L −1 ammonium acetate (pH 7.0) and determined with a flame photometer (FP6410, Company of Shanghai Jingke, China).Soil available P was extracted with 1.0 mol L −1 sodium bicarbonate and determined using colorimetric method.
Soil microbial biomass carbon (MBC) was determined using the chloroform fumigationextraction method.Fresh soils were fumigated at 25 • C for 24 h.The fumigated soils were extracted with 0.5 mol•L −1 K 2 SO 4 solution for 30 min in a shaker (180 r•min −1 ).Then, the mixture was filtered through a 0.45 µm water-based filter membrane.The concentration of carbon in the extract was measured with a TOC analyzer (Multi N/C 3100).MBC = (fumigated C-unfumigated C)/0.45 Microbial metabolic quotient is the ratio of carbon emitted by soil respiration to soil microbial biomass during incubation time.Soil PLFA was determined according to the method of Frostegård and Bååth (1996) [48].PLFAs were extracted from freeze-dried soil samples (2 g) with a single-phase chloroform/methanol/citric acid buffer (15 mL at a 1:2:0.8vol ratio).Total concentration of PLFAs (nmol•g −1 ) was set to account for total microbial biomass.Bacterial/fungal ratio (B/F) was calculated by dividing the bacterial biomass by the fungal biomass.Soil enzyme activity was determined by fluorescence microplate method with MUB (4-methylumbelliferone) and L-DOPa (L-3,4-dihydroxyphenylalanine) substrates.Fresh soil samples equivalent to 2.0 g of dry soil were weighed into a glass beaker, 300 mL buffer was added to make soil suspension, and this was homogenized thoroughly on a magnetic stirrer.Then, 200 µL soil suspension and 50 µL of 200 µmol•L −1 MUB substrate were siphoned off with a pipette gun into a 96-well black polystyrene microplate.At the same time, MUB standard solution was used to make the standard curve of each soil sample to be tested.The fluorescence values were measured by Perkinelmer EnSight (Perkinelmer, MA, USA) with excitation and absorption wavelengths of 365 nm and 450 nm after 3 h of culture at 25 • C under dark conditions.

Statistical Analysis
Data are expressed as the mean plus/minus one standard deviation of three replicates.One-way ANOVA followed by the least significant difference (LSD) was used to test the difference among the various treatments.Statistical significance was set at p < 0.05.All statistical analyses were carried out in SPSS 20.0, and figures were made by Origin 2021.

Changes in Litter Quality under Elevated CO 2 and Warming
Table 2 shows the nutrient concentration of rice and wheat straw under elevated CO 2 and warming.Elevated CO 2 decreased the N concentration of rice and wheat straw by 16.5% and 39.7%, respectively.Under elevated CO 2 , the K concentration of wheat also decreased significantly.Warming decreased the N and K concentration of wheat straw by 25.2% and 52.9%, respectively.Under elevated CO 2 plus warming, the N and P concentration of rice straw and the N and K concentration of wheat straw decreased significantly compared to the control.

The Effect of Elevated CO 2 and Warming on Soil Respiration (Experiment I)
The average CO 2 emission rate during the aerobic stage was 66.39 mg C•kg −1 •d −1 , which was about 13 times higher than the one during the anaerobic stage (Figure 2A).During the aerobic stage, the emission peak occurred on the first day of incubation, and since then, it decreased dramatically until Day 2. From Day 4 to Day 64, soil CO 2 emission rate gradually decreased.During the anaerobic stage, soil CO 2 emission rate dramatically increased in the first 15 days and then gradually declined.The emission peak was observed at Day 82.
The cumulative release of CO 2 (soil respiration hereafter) from the soil is shown in Figure 3A.Much more CO 2 was released during the aerobic stage, which accounted for about 90% of the overall release rate.During the aerobic process, elevated CO 2 decreased soil respiration by 27.60% compared to the control, while warming or elevated CO 2 plus warming had no effect on it.During the anaerobic process, all the treatments had no effect on soil respiration.

The Effect of Litter Quality Change on Soil Respiration (Experiment II)
As shown in Figure 2B, the CO 2 released dynamics across treatments were very similar to Experiment I.During the anaerobic stage, the CO 2 release rate increased dramatically in the first 15 days and was then gradually declined.The emission peak was observed at Day 82.Adding litters from different climate change treatments to the control soil had no effect on the soil respiration rate (Figure 3B).

The Effect of Soil Property Change on Soil Respiration (Experiment III)
As shown in Figure 2C, the CO 2 release dynamics across treatments were very similar to Experiment I and Experiment II.However, soil respiration varied greatly across treatments during the aerobic incubation stage.Compared to the ambient control, soils treated with elevated CO 2 plus warming emitted higher amounts of CO 2 .The accumulated CO 2 emission of soils treated with elevated CO 2 was 2874 mg C•kg −1 , which was significantly lower than the values from soils under warming and elevated CO 2 plus warming.During the anaerobic stage, there were no significant treatment effects (Figure 3C).The cumulative release of CO2 (soil respiration hereafter) from the soil is shown in Figure 3a.Much more CO2 was released during the aerobic stage, which accounted for about 90% of the overall release rate.During the aerobic process, elevated CO2 decreased soil respiration by 27.60% compared to the control, while warming or elevated CO2 plus warming had no effect on it.During the anaerobic process, all the treatments had no effect on soil respiration. .CO 2 released rate during the aerobic and anaerobic stage.Control represents the soils or litter that were collected from the ambient atmospheric CO 2 without warming; C represents the soils or litter that were collected from elevated CO 2 ; T represents the soils or litter that were collected from plant canopy warming; CT represents the soils or litter that were collected from CO 2 plus warming.The letters (A-C) represent Experiment I, Experiment II and Experiment III, respectively.The inset represents the period of flooded incubation from Day 65 to Day 147.

Correlation between Soil Respiration and Soil Characteristics
In Experiment I, soil respiration rate was positively correlated with microbial metabolic quotient, soil C:N, fungi to bacteria ratio and N-acetyl-glucosaminidase activity, but was negatively correlated with soil available P (Table 3).In Experiment III, soil respiration rate was positively correlated with SOC, dissolved organic carbon, microbial metabolic quotient, soil available K and β-Glucosidase activity, but it was negatively correlated with soil microbial biomass carbon and available P content.  1 for the treatment abbreviations.The letters A, B and C represent Experiment Ⅰ, Experiment Ⅱ and Experiment Ⅲ, respectively.Different lower-case letters indicate significant differences among treatments (p < 0.05).

The Effect of Litter Quality Change on Soil Respiration (Experiment Ⅱ)
As shown in Figure 2b, the CO2 released dynamics across treatments were very similar to Experiment Ⅰ.During the anaerobic stage, the CO2 release rate increased dramatically in the first 15 days and was then gradually declined.The emission peak was observed at Day 82.Adding litters from different climate change treatments to the control soil had no effect on the soil respiration rate (Figure 3b).

The Effect of Soil Property Change on Soil Respiration (Experiment Ⅲ)
As shown in Figure 2c, the CO2 release dynamics across treatments were very similar to Experiment Ⅰ and Experiment Ⅱ.However, soil respiration varied greatly across treatments during the aerobic incubation stage.Compared to the ambient control, soils treated with elevated CO2 plus warming emitted higher amounts of CO2.The accumulated CO2 emission of soils treated with elevated CO2 was 2874 mg C•kg −1 , which was significantly lower than the values from soils under warming and elevated CO2 plus warming.During the anaerobic stage, there were no significant treatment effects (Figure 3c).

Correlation between Soil Respiration and Soil Characteristics
In Experiment Ⅰ, soil respiration rate was positively correlated with microbial metabolic quotient, soil C:N, fungi to bacteria ratio and N-acetyl-glucosaminidase activity, but was negatively correlated with soil available P (Table 3).In Experiment Ⅲ, soil respiration rate was positively correlated with SOC, dissolved organic carbon, microbial metabolic quotient, soil available K and β-Glucosidase activity, but it was negatively correlated with soil microbial biomass carbon and available P content.

Discussion
Under future climate change of elevated CO 2 and warming, the changes in soil condition and litter quality have been observed, and they were supposed to alter the decomposition of SOM.Then, a new balance between organic carbon inputs and soil carbon losses, which can be used to predict the dynamics of SOC under climate change conditions, might be reached.However, this hypothesis was not fully supported by the current study.We found that the legacy effect of long-term elevated CO 2 and warming on soil conditions dominated the decomposition of SOM.Plant litter quality change had no effect on SOM mineralization, although significant changes in plant litter quality were observed in this study and among others [21,49].Hillstrom et al. (2010) also found that elevated CO 2 had a minimal effect on microbial respiration in a forest system, although it affected litter quality significantly [50], whereas Cornwell et al. (2008) found that the decomposition rate of litter caused by litter quality is three times that of climate factors [51].This may be true for ecosystems on a large scale, but for small areas of field, like that in the current study, this might not be true.
The response of soil respiration to elevated CO 2 varied across studies [7,[52][53][54][55].A recent study had shown that elevated CO 2 increased soil respiration by 25% on average [7]; however, this study showed that elevated CO 2 suppressed soil respiration compared with the ambient control, although neutral or negative effects have also been reported.Two reasons accounted for the higher soil respiration rate found under elevated CO 2 levels.Firstly, elevated CO 2 stimulated soil respiration by increasing the labile carbon pools.This carbon derived mainly from fine roots development and their exudates; most of it was decomposed by soil microbes and released to the atmosphere directly without forming aggregates with soil minerals [56,57].Therefore, no net carbon gains were observed in soils under elevated CO 2 .Secondly, elevated CO 2 stimulated soil respiration via the water saving effect.Under elevated CO 2 , leaf stoma closure reduced plant transpiration, and more water could be stored in soil, which facilitated soil microbial respiration [53].However, the water saving effect can only be observed in dry soil conditions; under wet soil conditions, it will decrease soil respiration because of low soil aeration.Therefore, Bader and Körner (2010) argued that there was no overall simulation of soil respiration under elevated CO 2 in a mature deciduous forest ecosystem [53].Furthermore, the magnitude of the soil respiration stimulating effect does not persist forever, and it will decline over the years with atmospheric CO 2 enrichment [58].This suggests that soil microbial community can adapt to long-term elevated CO 2 , and a new balance between carbon inputs and outputs can be reached.In the current study, there was no water saving effect as described in previous studies, because the soils were incubated at the same water condition, and there were no carbon additions via root exudates.Therefore, no stimulation effect was observed.The soils under long-term elevated CO 2 had higher phosphorous availability and lower soil C:N, fungi-to-bacteria ratio, and N-acetyl-glucosaminidase activity, which collectively led to the lower soil respiration rates (Table 3).
This study also demonstrated that soil respiration under elevated CO 2 plus warming responded differently to litter addition (Figure 3; Experiment I, Experiment III).The soil incorporated with litter from the control had a significant higher CO 2 emission rate than the soil with litter from the treatment of elevated CO 2 plus warming.In experiment III, the soil respiration rate of soil under elevated CO 2 plus warming was even higher than the rate of soil under the control and elevated CO 2 alone, which was different from the results in experiment I.We attributed this to the adaptation of the soil microbial community to long-term elevated CO 2 and warming [59].The soil microbes in this study under 7 years of elevated CO 2 , warming or both elevated CO 2 and warming have to obtaining nutrients and energy from SOM and litter in a more efficient way, and CO 2 was emitted.In contrast, a sudden change in food resource (adding litter from other environments, such as the litter from the control in this study) led to a lower carbon use efficiency, which caused a high soil respiration rate, especially for the warming treatment soils.Therefore, the soil microbes need to decompose more organic matter to obtain similar amounts of nutrients after food change.

Conclusions
The study showed that under elevated CO 2 and warming, the change in plant litter has no effect on the decomposition of soil organic matter, even though significant changes in litter quality have been observed.The decomposition of soil organic matter is controlled by the legacy effect of soil properties changes under climate change conditions.Changes in soil phosphorous availability and C/N, fungi/bacteria ratio and N-acetyl-glucosaminidase activity may be attributed to the alternation of SOM decomposition under elevated CO 2 .Elevated atmospheric CO 2 may promote soil carbon sequestration by suppressing soil microbial respiration under no temperature elevation conditions.

Figure 1 .
Figure 1.A photograph of the experimental set-up from the air.

Figure 1 .
Figure 1.A photograph of the experimental set-up from the air.
where F represents the CO 2 emission rate (mg C•kg −1 •d −1 ); ρ represents the density of CO 2 , which is 1.997 g•L −1 ; V represents the volume of air above the flask (L); m represents the mass of soil (g); ∆C represents the change in CO 2 concentration in the gas sample (µmol•mol −1 ); ∆t represents the sampling time (d) of the closed flask; and T is the temperature of the incubation (25 • C). α represents the conversion coefficient, 12/44 (C/CO 2 ).

Figure 2 .
Figure 2. CO2 released rate during the aerobic and anaerobic stage.Control represents the soils or litter that were collected from the ambient atmospheric CO2 without warming; C represents the soils or litter that were collected from elevated CO2; T represents the soils or litter that were collected from plant canopy warming; CT represents the soils or litter that were collected from CO2 plus warming.The letters A, B and C represent Experiment Ⅰ, Experiment Ⅱ and Experiment Ⅲ, respectively.The inset represents the period of flooded incubation from Day 65 to Day 147.

1 •day − 1 )Figure 2
Figure 2. CO 2 released rate during the aerobic and anaerobic stage.Control represents the soils or litter that were collected from the ambient atmospheric CO 2 without warming; C represents the soils or litter that were collected from elevated CO 2 ; T represents the soils or litter that were collected from plant canopy warming; CT represents the soils or litter that were collected from CO 2 plus warming.The letters (A-C) represent Experiment I, Experiment II and Experiment III, respectively.The inset represents the period of flooded incubation from Day 65 to Day 147.

Figure 3 .
Figure 3.The cumulative CO2 emission during aerobic and anaerobic stage.Please refer to Table1for the treatment abbreviations.The letters A, B and C represent Experiment Ⅰ, Experiment Ⅱ and Experiment Ⅲ, respectively.Different lower-case letters indicate significant differences among treatments (p < 0.05).

Table 2 .
Nutrients concentration of plant litter under elevated CO 2 and warming.
Different lower-case letters indicate significant differences among treatments (p < 0.05).

Table 3 .
Person correlation between soil respiration during the aerobic period and soil characteristics.The cumulative CO 2 emission during aerobic and anaerobic stage.Please refer to Table1for the treatment abbreviations.The letters (A-C) represent Experiment I, Experiment II and Experiment III, respectively.Different lower-case letters indicate significant differences among treatments (p < 0.05).

Table 3 .
Person correlation between soil respiration during the aerobic period and soil characteristics.