Varied response of carbon dioxide emissions to warming in oxic, anoxic and transitional soil layers in a drained peatland

Liu, Liangfeng; Chen, Huai; Tian, Jianqing

doi:10.1038/s43247-022-00651-y

Download PDF

Article
Open access
Published: 10 December 2022

Varied response of carbon dioxide emissions to warming in oxic, anoxic and transitional soil layers in a drained peatland

Communications Earth & Environment volume 3, Article number: 313 (2022) Cite this article

2887 Accesses
4 Citations
10 Altmetric
Metrics details

Subjects

Abstract

Peatlands are an important natural store of carbon. Here, we investigate how carbon dynamics in soils from three distinct layers from a drained peatland in the Qinghai-Tibet Plateau might respond to global warming. We incubated Zoige Plateau soil samples that represent oxic surface soil, permanently waterlogged anoxic deep soil, and a transitional interlayer, which varies between oxic and anoxic conditions with water table fluctuations, in the laboratory at 8 °C and 18 °C for 154 days. We find that carbon quality and microbial activity were highest in surface soils and lowest in interlayer soils. Carbon dioxide emissions were lowest in interlayer soils, where they were less sensitive to warming and more sensitive to nitrogen content than in the other layers. Conversely, carbon quality and microbial activity were the primary predictors of carbon dioxide emissions from the surface and deep soil, respectively. Our findings suggest a low vulnerability of soil carbon in interlayer soils of drained peatlands to warming.

Enhanced response of soil respiration to experimental warming upon thermokarst formation

Article Open access 30 April 2024

Climate warming and elevated CO₂ alter peatland soil carbon sources and stability

Article Open access 20 November 2023

Massive peatland carbon banks vulnerable to rising temperatures

Article Open access 12 May 2020

Introduction

Peatlands make up a critical part of the world’s carbon pool, storing about 600 Pg carbon and covering 15–30% of the terrestrial biosphere¹. However, against the background of continuous climate change and human activities, 15% of worldwide peatlands have been drained^2,3. The CO₂ release from drained peatlands should be a high priority to investigate, because it contributes about 5% of the global anthropogenic CO₂ release⁴, and has been considered the most important source of atmospheric CO₂³.

Compared with pristine peatlands, the water table, which is one of the primary regulators of peat carbon accumulation⁵, severely reduces in drained peatlands⁴. Water table drawdown exposes peatlands profile into three layers with distinct hydrological and oxygenation characteristics: surface soil is predominantly oxic and contains living plants and roots; deep soil is perpetually waterlogged and anoxic, and lacks plant roots; and the interlayer between surface and deep soil alternates between being oxic and anoxic during water table fluctuation^6,7,8. Long-term of these distinct characteristics lead to separations in soil biochemistry⁶, which altogether may translate to different soil CO₂ release under climate change. However, the CO₂ release from these three layers of drained peatlands still remains unclear, because the stratification of the vertical profile in previous research of drained peatlands is two and is based on depth, like soil 20⁹, 25¹⁰, or 30 cm below the surface^11,12, neglecting the interlayer (Fig. 1). These considerations highlight the need to enclose the interlayer to clarify the magnitude and determinants of CO₂ release from drained peatlands.

**Fig. 1: Carbon release from the profile of drained peatlands.**

The distribution of carbon quality^13,14 and microbes^9,15,16,17 are reported as two primary factors regulating the magnitude of CO₂ release. The study reports that carbon quality¹⁸ and microbial activity always decrease with depth, because surface soil contains abundant labile carbon and nutrients from plants¹⁹, whereas long-term decay leads to the accumulation of recalcitrant/inaccessible carbon in deep soil^18,20. However, this simplistic assumption may not always hold for drained peatlands, because several studies find that carbon at the interlayer is most degraded during water table rise and fall^6,21.

The rise and fall of the water table may also complicate the distribution of microbes since they are regulated by carbon accessibility²² and sensitive to local environmental changes^23,24,25. For example, labile carbon like dissolved organic carbon (DOC) and water-soluble phenolics (H₂O-PHE) are easily accessible to microbes^26,27 and enzymes²⁸, however, mineral-bound organic carbon is inaccessible to microbes^29,30. Water table fluctuation can also create oxic-anoxic cycles at the interlayer that affect microbial metabolism pathways and activity. The potential for different carbon quality and microbial activity across the layers of drained peatlands implies that the layers should be analyzed separately in order to predict the carbon stability of drained peatlands.

Moreover, nitrogen (N) nutrition is becoming vital in regulating soil organic carbon (SOC) decomposition against the background of increasing N deposited³¹. Soil N could originate from plants or be released from soil aggregates or carbon macromolecules by microbes³². Research shows that the available N could significantly increase SOC decomposition and CO₂ release in arctic soils³³, due to N is necessary for microbes to produce extracellular enzymes that decompose soil carbon to get energy^16,34. Therefore, any models of CO₂ release in water table-dependent soil layers must take nitrogen into account.

The distribution of carbon, microbial activity, and nitrogen nutrient also regulate the response of CO₂ release to climate warming. In a five years whole-soil warming experiment in a mixed coniferous forest, a study suggests that carbon loss is driven by the increases in microbial activity and DOC content³⁵, and in a decade warming experiment in a tallgrass prairie, the shift in microbial communities is reported largely associated with soil carbon loss¹⁵. The response of CO₂ release to warming in subtropical peatlands depends strongly on carbon quality and nutrient status¹³, while the response of permafrost peatlands depends strongly on carbon quality in the surface layer but microbial activity in the deep layer³⁶. These results greatly advance our knowledge of the determinant of CO₂ loss under a warming climate. However, the relative importance of the determinants across the soil profile, especially with respect to the water table, is unknown. To integrate these various determinants of CO₂ release into coherent climate models, we should examine them for the separate layers of drained peatlands.

Here, we examined the magnitude and determinants of CO₂ release from the three water table-dependent soil layers in a drained peatland on the Zoige plateau under a simulated warming climate (Supplementary Fig. 1). We explored the distribution of carbon and microbial composition in the three layers and quantified the magnitude of CO₂ release under warming conditions. We hypothesized that (1) carbon quality, microbial diversity, and CO₂ release would be lowest at the interlayer because of a periodically oxic decomposition during water table fluctuation⁶; (2) CO₂ release at the surface would depend primarily on microbial and enzyme activity, and CO₂ release from deep soil would depend primarily on carbon quality; (3) CO₂ release from the interlayer would be primarily regulated by nitrogen because the severe disturbance during water table fluctuation. We found that the CO₂ release in the interlayer is stable and less vulnerable to warming than at the surface or deep soil. The CO₂ released from the surface and deep soil are primarily controlled by carbon quality and decomposers, respectively. And the CO₂ release from the interlayer was sensitive to nitrogen variation.

Results

Carbon degradation indexes and chemical composition

Soil carbon degradation indexes, including C/N ratio, ash content, bulk density, δ¹³C, and δ¹⁵N were shown in Supplementary Fig. 4. The C/N ratio was lower at the surface than the deeper two layers. Ash content and bulk density were significantly higher at the surface than in the other two layers. Deep soil showed the highest δ¹³C and lowest δ¹⁵N of the three layers.

SOC was higher at the interlayer and deep soil than the surface (Supplementary Fig. 4). The SOC composition as acquired by Solid-State ¹³C CPMAS differed in certain respects among the three layers. The proportion of alkyl-C was significantly higher at the interlayer (45.8%) than at the surface (38.4%) or deep soil (38.1%; both p < 0.05). Conversely, the proportion of O-alkyl-C (39.8%) was significantly lower at the interlayer than at the surface (47.5%) or deep soil (43.2%; both p < 0.05). The calculated alkyl/O-alkyl ratio was thus significantly higher at the interlayer than surface or deep soil (p < 0.01), indicating greater recalcitrant carbon in the interlayer soil (Fig. 2a–c). The three soil layers did not differ significantly in proportions of aromatic or carboxylic C (Supplementary Fig. 5). Fe-bound organic carbon (Fe-OC), which could reflect soil organic carbon protection by minerals was 1.5-1.9 times higher at the interlayer than the other two layers (p < 0.01; Fig. 2d).

**Fig. 2: Comparison of carbon composition, mineral-protected organic carbon, and carbon pool sizes in the three water table-dependent soil layers.**

The carbon composition of three layers of soil was further confirmed by the size of the carbon fraction pool (Fig. 2e–g). The fast carbon pool was significantly larger at the surface than in the other two layers (p < 0.01; Fig. 2e), and the slow carbon pool was significantly smaller in the interlayer than in the other two layers (Fig. 2f). Conversely, the passive carbon pool was significantly larger in the interlayer than in the other two layers (Fig. 2g).

Labile carbon levels also differed among the three layers. DOC was lowest at the surface and highest in the deep soil, while H₂O-PHE was lowest at the surface and highest in the interlayer (Supplementary Table 1). These results suggest that DOC in the interlayer is less biodegradable than that at the surface or in deep soil.

Microbial diversity and community composition

The alpha diversity of bacteria was significantly lower in the interlayer than in the other two layers (Fig. 3a and Supplementary Fig. 6a–c). The composition of bacterial communities also differed significantly among the three layers (Fig. 3b). In contrast, the fungal alpha diversity did not differ (Fig. 3c and Supplementary Fig. 6d–f), and its community composition differed significantly among the three layers (Fig. 3d). The numbers of specific OTUs were lowest in the interlayer soil for both bacteria and fungi. The highest number of specific OTUs were at the surface for bacteria and in deep soil for fungi (Fig. 3e, g).

**Fig. 3: Comparison of bacterial and fungal diversity and composition in the three water table-dependent soil layers.**

Among the bacterial phyla showing a relative abundance >0.5%, Proteobacteria, Bacteroidetes, and Atribacteria were significantly more abundant in the interlayer than in the other two layers. And the surface contained significantly more Actinobacteria, Acidobacteria, Verrucomicrobia, Nitrosirae, and Planctomycetes than the other two layers (Fig. 3f and Supplementary Table 2). Among the bacterial genera showing relative abundance >0.5%, Pseudolabrys, Bradyrhizobium, Nitrosospira, and Natronoflexus were significantly higher at the surface than the other two layers. Conversely, Geobacter was significantly higher in the deep soil than the other two layers. Burkholderia, Variovorax, Duganella, Pseudomonas, and Mucilaginibacter were highest, and Dehalococcoides was lowest at the interlayer (Fig. 4a and Supplementary Table 2).

**Fig. 4: Comparison of the bacterial and fungal genera abundance in the three water table-dependent soil layers.**

The dominant functional groups of bacteria (>1%) did not differ among the three layers, except for groups of Replication and repair and Cell motility, both of which were significantly higher at the interlayer than the other two layers, and the group of Cell growth and death which was significantly lower at the interlayer than the other two layers (Supplementary Table 3).

Among fungal classes showing relative abundance >0.5%, Dothideomycetes was significantly more abundant in the interlayer and deep soil than at the surface. Conversely, Leotiomycetes and Sordariomycetes were more abundant at the surface than the other two layers, and Tremellomycetes, Pezizomycetes, and Microbotryomycetes were more abundant in the interlayer than in the surface and deep soil (Fig. 3h and Supplementary Table 4). Among the fungal genera showing relative abundance >0.5%, Paraphaeosphaeria was significantly more abundant in the deep soil than in the other two layers, while Dioszegia, Filobasidium, and Phoma were significantly more abundant in the interlayer than in the other two layers (Fig. 4b and Supplementary Table 4).

CO₂ emission and its response to warming

Under anoxic conditions, the CO₂ emission was significantly lower at the interlayer than at the surface and deep soil (Fig. 5a, b). 10 °C warming accelerated CO₂ emission of three layers, with a significant increment of 32.53 μg g⁻¹ d⁻¹ at the surface and 19.2 μg g⁻¹ d⁻¹ in the deep soil. The increment at the interlayer was 12.3 μg g⁻¹ d⁻¹ (Fig. 5a). The Q₁₀ parameter of CO₂ emission was higher at the surface than the emission from the interlayer or deep soil (Fig. 5b).

**Fig. 5: Comparison of CO₂ emission rate from the three water table-dependent soil layers under different oxygen and temperature conditions.**

Under oxic conditions, the CO₂ emission rate increased in all three layers than under anoxic conditions, with an increased proportion of 12.3% at the surface, 61.7% at the interlayer, and 35.7% in the deep soil (Fig. 5c, d). Warming further accelerated CO₂ emission by 108.7, 80.5, and 84.8% of the soil at the surface, interlayer, and deep, respectively. The Q₁₀ value for soil CO₂ emission was higher under oxic conditions than under anoxic conditions (Fig. 5d).

Soil variables response to warming

In all three layers, warming and oxygen significantly altered soil variables (Table 1). At the surface, nearly all these parameters increased under warming conditions, DOC, and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ also significantly increased under oxic conditions. H₂O-PHE, MBC, peroxidase activity, ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$ and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ significantly increased under the combination of oxic and warming. Similar results were observed in the interlayer, except that the activity of phenolic oxidase significantly decreased under warming conditions. In deep soil, DOC and cellobiohydrolase significantly increased under warming conditions, and MBN, phenolic oxidase, peroxidase, and cellobiohydrolase significantly increased under oxic combined warming conditions. In addition, MBC significantly increased under warming, oxic, and their combined conditions.

Table 1 Significance of variation in variables across the three water table-dependent soil layers during incubation.

Full size table

Determinant of CO₂ emission at each layer

Pearson correlation was used to detect relationships between original properties and CO₂ emission of three layers of soil (Fig. 6). Regardless of the presence or absence of warming and the presence or absence of oxygen, the CO₂ emission rate correlated with Alkyl-C, the size of the fast carbon pool, MBC, MBN, bacterial diversity, H₂O-PHE and ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$ (Fig. 6 and Supplementary Table 5). CO₂ emission rate also correlated with O-alkyl-C and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ under warming in the presence or absence of oxygen. In contrast, the CO₂ emission rate did not correlate with degradation indexes under any of the conditions tested.

**Fig. 6: Correlations between CO₂ emission rate and soil properties under different oxygen and temperature conditions.**

The taxon of bacteria and fungi and functional groups of bacteria also correlated with CO₂ emission of three layers of soil. Under anoxic conditions, with or without warming, the CO₂ emission rate significantly correlated with the abundance of the bacterial phyla Proteobacteria, Acidobacteria, Bacteroidetes, Verrucomicrobia, Nitrospirae, and Planctomycetes, as well as the bacterial genera Natronoflexus, Mucilaginibacter, Pseudomonas, Variovorax, and Clostridium. Under oxic conditions, the CO₂ emission rate correlated with the bacterial phyla Atribacteria and bacterial genera Clostridium and Dehalococcoides. Under oxic-warming conditions, the CO₂ emission rate also correlated with the bacterial phyla Proteobacteria, Bacteroidetes, and Verrucomicrobia, as well as the bacterial genera Pseudolabrys, Nitrosospira, Burkholderia, Mucilaginibacter, Pseudomonas, Variovorax, and Duganella (Supplementary Table 6). The CO₂ emission only correlated with the bacterial functional group of Replication and repair under warming conditions, and groups of Cell motility and Cell growth and death under anoxic, warming, and oxic-warming conditions (Supplementary Table 7).

Under anoxic conditions, the CO₂ emission rate significantly correlated with the fungal class Tremellomycetes and Microbotryomycetes, as well as the fungal genera Vishniacozyma, Vishniacozyma, Dioszegia, and Phoma. Under warming conditions, the CO₂ emission rate significantly correlated with the fungal genera Dioszegia, Phoma, Zymoseptoria, and Filobasidium. Under oxic combined warming conditions, the CO₂ emission rate significantly correlated with the fungal class Sordariomycetes and Pezizomycetes, as well as the fungal genera Lulworthia, Otidea, Dioszegia, and Phoma (Supplementary Table 8).

CO₂ emission from each soil layer seemed to be regulated by different soil variables (Table 2). At the surface soil, CO₂ emission mainly correlated with H₂O-PHE, DOC, MBC, MBN, peroxidase, β−1-4-glucosidase, ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$, and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$. At the interlayer, CO₂ emission mainly correlated with DOC, MBC, MBN, phenolic oxidase, peroxidase, β−1-4-glucosidase, ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$, and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$. CO₂ emission from deep soil correlated with MBC and enzyme activities. After grouping the soil variables into three groups through principal component analysis (Supplementary Table 9), the first principal component explaining most of the variation in emission from each soil layer was entered into the structural equation model (Fig. 7).

Table 2 Correlations between soil CO₂ emission and soil variables at each layer (n = 12).

Full size table

**Fig. 7: Structural equation modeling to describe the effects of factors on the CO₂ emission rate of three layers soil.**

The resulting model explained 88% of the variation in CO₂ emission at the surface. Soil carbon quality and decomposers had a direct positive effect on CO₂ emission, while carbon quality also indirectly affected CO₂ emission through its negative correlation with decomposers and its positive correlation with N nutrients. Soil N nutrients indirectly affected CO₂ emission through its positive correlation with decomposers. Warming and oxygen indirectly affected CO₂ emission through their positive correlation with soil carbon quality. Taken together, carbon quality emerged as the most influential predictor of CO₂ emission at the surface, based on standardized effect coefficients (Fig. 7a).

The structural equation model explained 94% of the variation in CO₂ emission in the interlayer, where decomposers and N nutrients affected the emission rate directly as well as indirectly through their negative correlation with each other. Temperature indirectly affected CO₂ emission through its positive correlation with decomposers, and oxygen indirectly affected CO₂ emission through its positive correlations with decomposers and N nutrients. In contrast to the results at the surface, N nutrients emerged as the most influential predictor of CO₂ emission from the interlayer (Fig. 7b).

The structural equation model explained 79% of the variation in CO₂ emission in deep soil, in which decomposers were the major determinant. Neither carbon quality nor N nutrients correlated significantly with CO₂ emission, whereas temperature and oxygen affected CO₂ emission indirectly through their positive correlation with decomposers. Thus, decomposers emerged as the most influential predictors of CO₂ emission from deep soil (Fig. 7c).

Discussion

We measured mean CO₂ emission rates from 24.97 ± 0.21 to 43.87 ± 3.86 μg g⁻¹ d⁻¹ across the three water table-dependent soil layers during 154-day incubation, which falls at the low end of the range reported for tropical peatlands (193–1864.8 μg g⁻¹ d⁻¹)^37,38. Higher emission from tropical peatlands has been attributed to higher carbon quality, greater nutrient availability, and lower proportions of recalcitrant carbon³⁸. Consistent with this, the C/N ratio, which can indicate carbon quality²¹, ranges from 20.3 to 47 on tropical peatlands^39,40,41, much higher than the range of 15.09–19.45 in our study.

Our mean CO₂ emission rates are at the high end of the range reported for boreal peatlands^42,43. This suggests that carbon on boreal peatlands is less vulnerable than carbon on Zoige peatlands, which may reflect differences in vegetation, the main source and predictor of carbon liability. The dominant vegetation on Zoige peatlands is herbaceous, featuring graminoids and sedges, whereas dwarf shrubs and moss predominate on peatlands at higher latitudes⁴⁴. The higher carbon quality of herbaceous litter in comparison to moss and shrubs litter could then result in a higher CO₂ emission rate on the Zoige Plateau. Another contributor to the greater CO₂ emission from the Zoige Plateau is minerotrophic peat, abundant in nutrients from rivers⁴⁵; boreal peatlands, in contrast, is ombrotrophic with poor nutrients⁴⁶. In addition, the higher incubation temperature also is a reason of higher CO₂ emission rate on the Zoige Plateau.

Our study illuminates a relatively stable interlayer and relatively vulnerable surface and deep soil in terms of CO₂ emission on the Tibetan Plateau. In our laboratory incubation, CO₂ emission was greater from the surface than from the other two layers. This may reflect the greater carbon quality and microbial activity at the surface^12,14. Plant roots and litter supply abundant labile carbon to surface soil⁴⁷, increasing the content of O-alkyl-C and N nutrient and enhancing microbial activity⁴⁸. Greater CO₂ emission rate at the surface soil may also reflect the predominance of aerobic microbes, such as the bacteria of Bradyrhizobium, Nitrosospira, and Natronoflexus⁴⁹ and fungi of Lulworthia and Acephala⁵⁰. Lower CO₂ emission rate at deeper two layers mainly because of the exhaustion of labile nutrients for microbes during long-term slow decay⁵¹. This result further reflects that the carbon at the deeper two layers was more degraded, with a higher C/N ratio and δ¹³C, low bulk density. Higher δ¹⁵N at surface soil may result from the increasing human activity and nitrogen deposition⁵².

Surprisingly, our study found that the CO₂ emission rate was lowest in the interlayer, rather than in deep soil. Such phenomena could attribute to the highest proportion of recalcitrant/passive carbon at the interlayer. Long-term water table fluctuation exposes interlayer soil to secondary oxic decomposition, during which labile carbon and nutrients had been consumed and recalcitrant carbon was accumulated^21,53. Accompanied by a high content of recalcitrant carbons, the bacterial diversity and number of specific OTU were also lowest at the interlayer, partially explaining the lowest CO₂ emission at interlayer^25,47,54. This result suggests that the interlayer in drained peatlands is stable in CO₂ emission.

Compared with the interlayer, higher CO₂ emission in the deep soil may be because of the preferential flow of nutrients from the surface^17,55. The accumulation of nutrients in anoxic deep soil increases labile carbon content and CO₂ emission rate³⁵. In addition, the deep soil in our study contained highly diverse fungal communities and high numbers of specific OTUs. These factors also contribute to the higher CO₂ emission, given that fungi prefer to metabolize complex carbons to obtain energy⁵⁶.

Our study further illuminates that the three layers responded differently to warming. In specific, warming greatly increased CO₂ emission at the surface soil, presumably because of its greater abundance of microbes, which are reported to be temperature responsive²⁸. Warming also greatly increased CO₂ emission in deep soil, because of its higher content of inherently biochemical recalcitrant carbon, such as aromatic compounds, which are more responsive to warming than physically protected carbons^57,58. Consistent with these explanations, the concentration of DOC, which could be accessed by microbes directly¹², also greatly increased by warming at the surface and deep soil, supporting the enhancement of CO₂ emission. At deep soil, the increased DOC even could serve as exogenous carbon inputs to cause positive priming¹⁰, accelerating CO₂ emission^17,18. Together with other studies on alpine peatlands, which also reported that deep soil was temperature sensitive^15,35, our results suggest that deep soil carbon is recalcitrant, but still vulnerable to warming.

At the interlayer, warming increased microbial activity and, thereby, CO₂ emission, whereas to a smaller extent than in the other two soil layers. This may be attributed to the fact that the interlayer contained the highest proportion of mineral-protected carbon, here measured as Fe-OC, and this form of carbon is reported as less sensitive to warming than other forms, such as biochemically recalcitrant carbons^57,58. We postulate that following long-term secondary oxic decay, the severely degraded carbon in the interlayer is not only recalcitrant but also insensitive to warming. In this case, previous depth-dependent studies may have overestimated the vulnerability^10,59 and warming sensitivity^10,19,35 of carbon in the deeper soil.

Exposing the three layers of soil to oxic conditions hugely increased CO₂ emission, which was further greatly increased under oxic combined warming conditions. Sufficient oxygen activates aerobic microbes such as the bacteria Rhizobiales, Nitrospirales, and Caulobacterales, and fungi of Helotiales and Pleosporales⁶⁰. These aerobic microbes, especially the fungi, secrete a wide range of enzymes to break down recalcitrant compounds to meet their demands for energy and nutrients⁵⁶, and enhancing the CO₂ emission. Indeed, we found that the activity of peroxidase and β−1-4-glucosidase in the interlayer and the activity of peroxidase and phenol oxidase in the deep soil substantially increased when under oxic conditions. These results suggest that microbial and enzyme activity are severely constrained under anoxic conditions, which then maintain the stability of peatlands carbon stock.

At last, our study presents direct evidence of different determinants mediating CO₂ emission in each soil layer. As shown by SEM, carbon quality was the primary determinant of CO₂ emission from the surface soil. Soil carbon can be divided into three pools^61,62 according to their turnover time, and the fast pool is small and can be depleted in a few months^26,63. Thus, the remained slow and passive carbon pool, which size is large and with a turnover time of thousands to centuries years⁶⁴, may be the direct regulator of CO₂ emission from the surface soil. During incubation in the present study, the contribution of the fast carbon pool to total respired CO₂ was reduced, and the contribution from the slow and passive carbon pool progressively increased (Supplementary Fig. 7). Thus, we postulate that carbon quality, especially the slow and passive carbon pools, will determine CO₂ loss from drained peatlands in the long term. Previous studies came to divergent conclusions about whether the major determinants of CO₂ emission from surface soil of drained peatlands were carbon quality, nutrient content¹³, or microbial activity⁶⁵. Their divergent findings may reflect a failure to account for the environmental gradient caused by water table fluctuation.

In contrast to the situation at the surface, we found that decomposers became the primary determinant of CO₂ emission in deep soil. The study reports that genes coding for enzymes are greatly higher in recalcitrant carbon-rich soils than in other soils⁶⁶, because microbes should strongly upregulate enzyme secretion to decompose recalcitrant carbon to obtain energy and nutrients^36,66. And warming could further upregulate such genes in bacteria and fungi¹⁵, which help explain the enhanced CO₂ emission from deep soil in our study. In this way, our study indicates that decomposers are the primary drivers of CO₂ emission from deep soil, especially fungi which has enzymatic potential to decompose recalcitrant carbon^36,61.

Similarly, decomposers were a major determinant of CO₂ emission from the interlayer, because of its rich recalcitrant carbon and microbes need to produce a wide range enzymes to obtain energy and nutrients^36,66. Interestingly, our results showed that the N nutrient, with the highest standardized path coefficients, was the primary factor mediating the CO₂ emission at the interlayer. The content of ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$ and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ increased under oxic or warming conditions and correlated with MBC and CO₂ emission rates. This result suggests the sensitivity of microbial metabolism to N nutrients⁶⁷, in part because nitrogen is required for microbes in order to produce extracellular enzymes^17,68.

While our study provides experimental evidence that the carbon stability is distinct among water table-dependent three layers of drained peatlands. At the same time, our results should be interpreted with caution in light of several limitations: first, our incubation experiment did not consider the influence of vegetation, leading to uncertainty in explaining the determinant of CO₂ release at each layer on a large scale. Second, serving the average water table as the base to divide the surface and interlayer, may lead to inaccuracy in evaluating the magnitude and determinant of CO₂ release of each layer when in field, because such surface soil also could be waterlogged after heavy rain. In situ experiments, including the influence of vegetation and reasonable stratification of the vertical profile, are needed to clarify the carbon stability of drained peatlands.

Despite these limitations, our study suggests that the carbon stability of drained peatlands differs across their water table-dependent soil layers, because of the differences in soil carbon quality as well as microbial diversity and activity. We find that oxic surface and anoxic deep soil are vulnerable and sensitive to warming in CO₂ emission, for the highest MBC at the surface soil and chemically recalcitrant carbon in the deep soil. The major determinant of CO₂ emission at these two layers are carbon quality and decomposers, respectively. However, experiencing periodic oxic decay, the CO₂ emission at the interlayer is stable and less sensitive to warming than the other two layers, and N nutrients become its major driver of CO₂ emission.

Our findings suggest that the CO₂ loss from drained peatlands in previous depth-dependent studies may be unfaithful because they omit the interlayer along the soil profile which is caused by the heterogeneity of environments like oxygen and hydrology. Future research should include the environmental gradient along the soil profile. Our findings also highlight the need to incorporate N nutrient into Earth System Models for precisely predicting carbon loss from drained peatlands under climate change.

Methods

Site description and sampling

Our study site was located at Ruokeba on the Zoige Plateau, which lies in the northeastern part of the Qinghai-Tibet Plateau (33°03′58′′N, 102°34′39′′E, avg. 3470 m a.s.l; Supplementary Fig. 2a). Annual mean temperature on the Zoige plateau is 3.3 °C and annual mean precipitation is 705 mm. Peatlands on the Zoige Plateau are minerotrophic fens where peat has been accumulating since the early Holocene due to the low temperature and abundant precipitation. Currently peat on the Zoige Plateau covers 4605 km² to a mean depth of 1.39 m (range, 0.2 to 6.0 m)⁶⁹. Dominant vegetation at Ruokeba peatlands is Equisetum ramosissimun, Scirpus triqueter, and Blysmus sinocompressus.

The peatlands on Zoige plateau, including Ruokeba, experienced severe drainage in the 1970s⁷⁰. Currently, the ditch has a depth of 30–50 cm and acts as the main channel for transporting water and DOC from peatlands to lower regions (Supplementary Fig. 2b). Based on measurements since 2015 using a 1-m grid, the water table on Ruokeba peatland averages −25 cm, ranging from −4 to −67 cm below the soil surface (Supplementary Fig. 2c). As a result, we divided the soil layers for the present study as follows: surface, above a depth of 25 cm; deep, deeper than a depth of 75 cm; and interlayer, from a depth of 25 cm to a depth of 75 cm.

Three plots with similar water table fluctuation range at Ruokeba were randomly selected, and five soil columns (105 cm) were collected at each plot using a vibracore with a diameter of 8 cm. For each soil layer, the soil from the five columns at each plot was combined into a single sample. Each sample was transported to the laboratory in an ice cooler and divided into two fractions, one of which was subjected to microbial analysis, while the other was subjected to lab incubation and soil properties analyses.

Lab incubation

After removing large roots, 80 g of homogenized soil of each layer was put into a 500-ml glass jar, which was then sealed with a rubber stopper with two openings for sampling and gas exchange. For 154 days, jars were incubated in the dark under both anoxic conditions in which the headspace was flushed with N₂ (99.9%) and oxic conditions in which the headspace was flushed with CO₂-free air (containing O₂: 21%; N₂: 79%; 99.9%), and at two temperatures: 8 °C as control (the average temperature during the growing season on the Zoige Plateau) and 18 °C as warming (10 °C warming to get a reliable Q₁₀ of CO₂ emission). The flow rate and timing of gas flush was set through a continuous monitor⁷¹. The four treatments were anoxic (control), warming, oxic, oxic combined warming, and these experiments involved 36 samples: 3 soil layers × 2 conditions × 2 temperatures × 3 replicates.

During the incubation, headspace gas was sampled weekly. At 12 h before each sampling, the headspace was evacuated and flushed with N₂ or CO₂-free air again to remove the accumulated CO₂. CO₂ concentration in the headspace gas was measured using gas chromatography (Agilent 7890 A, Wilmington, DC, USA) as described in ref. ⁷¹. The Q₁₀ parameter was roughly estimated using the ratio of CO₂ emission at 18 to 8 °C.

Soil carbon degradation indexes

Indexes indicating soil carbon degradation degree were C/N ratio, ash content, bulk density (BD), δ¹³C, and δ¹⁵N⁷². The C/N ratio was calculated as the ratio of %C to %N, both of which were determined using an elemental analyzer (Elementar, Langenselbold, Germany)⁷³. Ash content was measured using thermogravimetry⁷², and BD was measured using a bulk soil sampler⁷⁴. Stable isotopes of δ¹³C and δ¹⁵N in soil were analyzed using an isotope ratio mass spectrometer (Thermo Scientific, Waltham, MA, USA) and the Vienna Pee Dee Belemnite standard.

Soil chemical components

Chemical components in soil carbon were quantified using solid-state ¹³C CPMAS NMR. Magic-angle spinning experiments were conducted on an Avance III 400WB solid-state spectrometer (Bruker, Berlin, Germany) using 3.2-mm double-resonance probes spinning at 15 kHz. Spectra were collected at the ¹³C Larmor frequency of 100.6 MHz. ¹³C cross-polarization contact time was 2 ms, and ¹H decoupling power was 104.2 kHz. ¹³C signals were referenced to the methylene signal of adamantane at 38.5 ppm, and the spectra were analyzed in the following chemical shift regions using MestReNova 9.0 (MESTRELAB Research, Santiago de Compostela, Galicia, Spain): alkyl-C, 0 to 50 ppm; O-alkyl-C, 50 to 110 ppm; aromatic-C, 110 to 165 ppm; and carbonyl-C, 165–220 ppm⁷⁵. The concentration of Fe-OC was measured using the citrate-bicarbonate-dithionite method⁷⁶.

Levels of labile organic carbon present as H₂O-PHE and DOC. The concentration of H₂O-PHE in soil was estimated using the Folin-Ciocalteu method, with p-hydroxybenzoic acid as the standard. Incubated fresh soil (5 g) and deionized water (25 ml) were shaken at a frequency of 200 rpm for 20 h, filtered through a 0.45-μm filter, and diluted 1:5. An aliquot (1 ml) was mixed with 2 ml Na₂CO₃-NaHCO₃ buffer (pH 10), 1 ml Folin-Ciocalteu reagent, and 1 ml Na₂CO₃ (0.4 mol l⁻¹), then incubated for 15 min at 37 °C in the dark. Absorbance at 750 nm was measured using a spectrophotometer (Thermo Spectronic 200, Waltham, MA, USA). Standard phenolic absorbance was also measured with the same procedure. The concentration of H₂O-PHE, expressed as mg g⁻¹ soil, was calculated as described in ref. ⁷⁷. Levels of DOC was measured using a continuous flow analyzer (San++, SKALAR, Delft, Netherlands).

Soil carbon pool model

SOC could be conceptually grouped into three pools: fast pool with a turnover time of a few days to a few months, slow pool with a turnover time of about a few years, and passive pool with a turnover time of a few years to thousands of years⁷⁸. Carbon in passive pools also belongs to recalcitrant carbons whose turnover time is about centuries to millennia years⁷⁹. The size of different carbon pools of each layer was estimated using the three-pool model⁶³, which integrates data from three replicates incubated under anoxic conditions into Eq. (1):

$${{{{{\rm{R}}}}}}({{{{{\rm{t}}}}}})={k}_{1}{f}_{1}{C}_{tot}{e}^{-k1t}+{k}_{2}{f}_{2}{C}_{tot}{e}^{-k2t}+{k}_{3}(1-{f}_{1}-{f}_{2}){C}_{tot}{e}^{-k3t}$$

(1)

where R(t) is the CO₂ emission rate at time t (mg C g⁻¹ d⁻¹); C_tot is the initial soil carbon content; f₁ and f₂ refer to the fast pool fraction and slow pool fraction, respectively; and k₁, k₂, and k₃ are the decay rates for fast, slow and passive carbon pools, respectively. C_tot and R(t) were measured experimentally, while the other parameters were estimated using a Markov Chain Monte Carlo approach⁸⁰. The prior parameter range of Markov Chain Monte Carlo analysis was set in Supplementary Table 10, which was as wide as possible to cover the possibility of three-layer soils.

Soil microbial properties

Microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), microbial diversity, and community composition were determined. MBC and MBN were measured using chloroform fumigation.

Microbial diversity and community composition were determined using high-throughput sequencing of soil DNA, which was extracted using the PowerSoil Kit (MoBio Laboratories, Carlsbad, CA, USA) following the manufacturer’s instructions. The V4 region of the bacterial 16 S rRNA gene and the ITS1 region of fungi were amplified using PCR to characterize the bacterial and fungal communities, respectively. The primers of V4 region was 515 F (5′- GTGCCAGCMGCCGCGGTAA-3′) and 806 R (5′-GGACTACHVGGGTWTCTAAT-3′)⁸¹, and ITS1 region was amplified with primers ITS1f (5’-CTTGGTCATTTAGAGGAAGTAA-3′)⁸² and ITS2 (5’GCTGCGTTCTTCATCGATGC-3′)⁸³. Sequencing was performed using the 2 × 300 paired-end Miseq PE300 platform (Illumina, USA) at Allwegene Tech Beijing (Beijing, China).

QIIME was used to filter raw sequences to obtain higher-quality clean tags (length, 250–500 bp, average base quality score >25). After removing chimeric sequences using the UCHIME algorithm, sequences with ≥97% similarity were assigned to the same operational taxonomic unit (OTU) based on the SILVA database (bacteria) and UNITE database (fungi). The OTU abundances were normalized using a standard sequence number corresponding to the sample with the fewest sequences (14,130 for bacteria and 22,169 for fungi). The normalized OTU was used in the downstream analysis of bacteria and fungi. The bacterial and fungal sequences of three layers of soil were retrieved from published 16 S and ITS sequences (deposited in the NCBI Sequence Read Archive (SRA) database with accession number SRP211776)⁶⁰.

Soil enzyme activity

Four enzymes activity were measured as described in refs. ^84,85 in a 96-well plate assay. Phenol oxidase and peroxidase were assayed spectrophotometrically, while β−1,4-glucosidase and cellobiohydrolase were assayed fluorimetrically.

Soil nitrogen levels

Levels of ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$ and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ were measured using a continuous flow analyzer (San++, SKALAR, Delft, Netherlands).

Data analysis

During data analysis, carbon degradation indexes, carbon chemical components (alkyl, O-alkyl, aromatic, carboxylic), Fe-bound organic carbon (Fe-OC), and size of carbon pools^78,86,87 were used to describe soil carbon quality. Microbial properties and enzyme activities were used to describe soil decomposers. Nitrogen levels of NO⁻₃-N and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ were listed as N nutrients. In addition, soil labile carbon levels, MBC, MBN, enzyme activity, ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$, and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$ varied during incubation. Thus, these indexes were listed as soil variables and were measured to identify determinants of CO₂ release from each layer (Supplementary Fig. 1).

The diversity of bacteria and fungi were assessed using the vegan package (version 2.4-5) in R⁸⁸. The composition of bacterial and fungal communities were assessed using ordination by nonmetric multidimensional scaling with the “metaMDS” function in vegan⁸⁹, and their difference among layers were assessed using permutational multivariate analysis of variance (PERMANOVA) with 999 permutations and an “ADONIS” function in vegan. Bacterial OTU were also assigned to a different functional group based on Tax4Fun package⁹⁰.

Before statistical analyses, data were confirmed to be normally distributed (Kolmogorov–Smirnov test); if not, they were log-transformed. The log-transformed indexes included DOC and bacterial diversity. All statistical analyses were performed using SPSS 20.0 for Windows (IBM, Chicago, IL, USA) with a significance definition of p < 0.05. Results were depicted graphically using Origin 8.0 (OriginLab, Northampton, MA, USA).

Differences in soil carbon quality, bacterial/fungal diversity, the relative abundance of each taxon of bacteria and fungi, bacterial functional groups, and CO₂ emission among the three layers were assessed for significance using one-way analysis of variance (ANOVA) with the Tukey HSD test. Differences in CO₂ emission between control and treatments were assessed using the paired t-test. The potential relationships between soil carbon quality/microbial properties of the original soil and CO₂ emission were accessed together of three layers of soil using Pearson correlation analysis. Pearson correlation analysis was also used to access the relationships between soil variables and CO₂ emission separately for each layer.

Piecewise structural equation model was conducted to identify determinants of CO₂ emission in each layer under different temperatures and incubation conditions. In this modeling, treatment was the exogenous variable, while soil variables were the endogenous variables. The modeling was carried out using the semPlot package in R.

The primary structural equation model was developed based on prior knowledge of regulators of CO₂ emission (Supplementary Fig. 3). Paths were drawn from temperature (T) and incubation conditions (C) to all endogenous variables, then to CO₂. This model was used in linear mixed-effect models. The least significant paths were removed stepwise from the model until none of the remaining parameters was associated with p > 0.1.

To simplify the structural equation model, soil endogenous variables were grouped into carbon quality, comprising H₂O-PHE and DOC; decomposers, comprising MBC, MBN, phenolic oxidase, peroxidase, β−1-4-glucosidase and cellobiohydrolase; and N nutrients, comprising ${{{{{{\rm{NO}}}}}}}_{3}^{-}{\mbox{-}}{{{{{\rm{N}}}}}}$ and ${{{{{{\rm{NH}}}}}}}_{4}^{+}{\mbox{-}}{{{{{\rm{N}}}}}}$. Within each of these three groups, principal component analysis was conducted using only those variables that correlated significantly with CO₂ emission. The first component, which could explain more than 60% of the total variation (Supplementary Table 9), was introduced as a new variable in the structural equation model³⁶.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The microbial data were available in the NCBI Sequence Read Archive (SRA) database (SRP211776). The remaining data supporting the findings of this study were deposited on figshare (https://doi.org/10.6084/m9.figshare.21580938.v1).

References

Yu, Z., Loisel, J., Brosseau, D. P., Beilman, D. W. & Hunt, S. J. Global peatland dynamics since the Last Glacial Maximum. Geophys. Res. Lett. 37, L13402 (2010).
Article Google Scholar
Joosten, H., Tapio-BiströmM, L. & Susanna, T. Peatlands: guidance for climate change mitigation through conservation, rehabilitation and sustainable use. Food and Agriculture Organization of the United Nations and Wetlands International. FAO (2012).
IUCN. Issues brief: peatlands and climate change. www.icun.org (2017).
Joosten, H. Peatlands, Climate Change Mitigation and Biodiversity Conservation. An Issue Brief on the Importance of Peatlands for Carbon and Biodiversity Conservation and the Role of Drained Peatlands as Greenhouse Gas Emission Hotspots (Nordic Council of Ministers, 2015).
Moore, T. R. & Knowles, R. The influence of water table levels on methane and carbon dioxide emissions from peatland soils. Can. J. Soil Sci. 69, 33–38 (1989).
Article CAS Google Scholar
Tfaily, M. M. et al. Organic matter transformation in the peat column at Marcell Experimental Forest: humification and vertical stratification. J. Geophys. Res. Biogeosci. 119, 661–675 (2014).
Article CAS Google Scholar
Clymo, R. S. & Bryant, C. L. Diffusion and mass flow of dissolved carbon dioxide, methane, and dissolved organic carbon in a 7-m deep raised peat bog. Geochim. Cosmochim. Acta 72, 2048–2066 (2008).
Article CAS Google Scholar
Clymo, R. S. The limits to peat bog growth. Philos. Trans. R. Soc. B 303, 605–654 (1984).
Google Scholar
Qin, S. et al. Temperature sensitivity of SOM decomposition governed by aggregate protection and microbial communities. Sci. Adv. 5, eaau1218. 1211–1219 (2019).
Article Google Scholar
Dorrepaal, E. et al. Carbon respiration from subsurface peat accelerated by climate warming in the subarctic. Nature 460, 616–619 (2009).
Article CAS Google Scholar
Luo, Z. K., Wang, G. C. & Wang, E. L. Global subsoil organic carbon turnover times dominantly controlled by soil properties rather than climate. Nat. Commun. 10, 3688 (2019).
Article Google Scholar
Wilson, R. M. et al. Stability of peatland carbon to rising temperatures. Nat. Commun. 7, 13723 (2016).
Article CAS Google Scholar
Sihi, D., Inglett, P. W. & Inglett, K. S. Carbon quality and nutrient status drive the temperature sensitivity of organic matter decomposition in subtropical peat soils. Biogeochemistry 131, 103–119 (2016).
Article CAS Google Scholar
Wang, Q., Liu, S. & Tian, P. Carbon quality and soil microbial property control the latitudinal pattern in temperature sensitivity of soil microbial respiration across Chinese forest ecosystems. Glob. Chang. Biol. 24, 2841–2849 (2018).
Article Google Scholar
Cheng, L. et al. Warming enhances old organic carbon decomposition through altering functional microbial communities. ISME J. 11, 1825–1835 (2017).
Article Google Scholar
Luan, J., Wu, J., Liu, S., Roulet, N. & Wang, M. Soil nitrogen determines greenhouse gas emissions from northern peatlands under concurrent warming and vegetation shifting. Commun. Biol. 2, 132 (2019).
Article Google Scholar
Meyer, N. et al. Nitrogen and phosphorus supply controls soil organic carbon mineralization in tropical topsoil and subsoil. Soil Biol. Biochem. 119, 152–161 (2018).
Article CAS Google Scholar
Fontaine, S. et al. Stability of organic carbon in deep soil layers controlled by fresh carbon supply. Nature 450, 277–280 (2007).
Article CAS Google Scholar
Moni, C. et al. Temperature response of soil organic matter mineralisation in arctic soil profiles. Soil Biol. Biochem. 88, 236–246 (2015).
Article CAS Google Scholar
Xu, X., Sherry, R. A., Niu, S., Zhou, J. & Luo, Y. Long-term experimental warming decreased labile soil organic carbon in a tallgrass prairie. Plant Soil 361, 307–315 (2012).
Article CAS Google Scholar
Broder, T., Blodau, C., Biester, H. & Knorr, K. H. Peat decomposition records in three pristine ombrotrophic bogs in southern Patagonia. Biogeosciences 9, 1479–1491 (2012).
Article CAS Google Scholar
Adamczyk, M., Perez-Mon, C., Gunz, S. & Frey, B. Strong shifts in microbial community structure are associated with increased litter input rather than temperature in High Arctic soils. Soil Biol. Biochem. 151, 108054 (2020).
Article CAS Google Scholar
Hug, L. A. et al. Community genomic analyses constrain the distribution of metabolic traits across the Chloroflexi phylum and indicate roles in sediment carbon cycling. Microbiome 1, 22 (2013).
Article Google Scholar
Yun, J. L., Ju, Y. W., Deng, Y. C. & Zhang, H. X. Bacterial community structure in two permafrost wetlands on the Tibetan Plateau and Sanjiang Plain, China. Microb. Ecol. 68, 360–369 (2014).
Article Google Scholar
Zhong, Q. et al. Water table drawdown shapes the depth-dependent variations in prokaryotic diversity and structure in Zoige peatlands. FEMS Microbiol. Ecol. https://doi.org/10.1093/femsec/fix049 (2017).
Article Google Scholar
Karhu, K. et al. Temperature sensitivity of soil respiration rates enhanced by microbial community response. Nature 513, 81–84 (2014).
Article CAS Google Scholar
Thiessen, S., Gleixner, G., Wutzler, T. & Reichstein, M. Both priming and temperature sensitivity of soil organic matter decomposition depend on microbial biomass – An incubation study. Soil Biol. Biochem. 57, 739–748 (2013).
Article CAS Google Scholar
Walker, T. W. N. et al. Microbial temperature sensitivity and biomass change explain soil carbon loss with warming. Nat. Clim. Chang. 8, 885–899 (2018).
Article CAS Google Scholar
Dungait, J. A. J., Hopkins, D. W., Gregory, A. S. & Whitmore, A. P. Soil organic matter turnover is governed by accessibility not recalcitrance. Glob. Chang. Biol. 18, 1781–1796 (2012).
Article Google Scholar
Conant, R. T. et al. Temperature and soil organic matter decomposition rates - synthesis of current knowledge and a way forward. Global Chang. Biol. 17, 3392–3404 (2011).
Article Google Scholar
Hietz, P. et al. Long-term change in the nitrogen cycle of tropical forests. Science 4, 334 (2011).
Google Scholar
Manzoni, S., Taylor, P., Richter, A., Porporato, A. & Agren, G. I. Environmental and stoichiometric controls on microbial carbon-use efficiency in soils. New Phytol. 196, 79–91 (2012).
Article CAS Google Scholar
Sistla, S. A., Asao, S. & Schimel, J. P. Detecting microbial N-limitation in tussock tundra soil: Implications for Arctic soil organic carbon cycling. Soil Biol. Biochem. 55, 78–84 (2012).
Article CAS Google Scholar
Chen, L. et al. Nitrogen availability regulates topsoil carbon dynamics after permafrost thaw by altering microbial metabolic efficiency. Nat. Commun. 9, 3951 (2018).
Article Google Scholar
Soong, J. L. et al. Five years of whole-soil warming led to loss of subsoil carbon stocks and increased CO₂ efflux. Sci. Adv. 7, eabd1343 (2021).
Chen, L. et al. Determinants of carbon release from the active layer and permafrost deposits on the Tibetan Plateau. Nat. Commun. 7, 13046 (2016).
Article CAS Google Scholar
Girkin, N. T. et al. Interactions between labile carbon, temperature and land use regulate carbon dioxide and methane production in tropical peat. Biogeochemistry 147, 87–97 (2019).
Article Google Scholar
Swails, E. et al. Will CO₂ emissions from drained tropical peatlands decline over time? Links between soil organic matter quality, nutrients, and C mineralization rates. Ecosystems 21, 868–885 (2017).
Article Google Scholar
Ismawi, S., Gandaseca, S. & Ahmed, O. Effects of deforestation on soil major macro-nutrient and other selected chemical properties of secondary tropical peat swamp forest. Int. J. Phys. Sci. 7, 2225–2228 (2012).
Article CAS Google Scholar
Kimura, S., Melling, L. & Goh, K. Influence of soil aggregate size on greenhouse gas emission and uptake rate from tropical peat soil in forest and different oil palm development years. Geoderma 185, 1–5 (2012).
Article Google Scholar
Takakai, F. et al. Effects of agricultural land-use change and forest fire on N₂O emission from tropical peatlands, Central Kalimantan, Indonesia. Soil Sci. Plant Nutr. 52, 662–674 (2006).
Article CAS Google Scholar
Knoblauch, C., Beer, C., Sosnin, A., Wagner, D. & Pfeiffer, E. M. Predicting long-term carbon mineralization and trace gas production from thawing permafrost of Northeast Siberia. Glob. Chang. Biol. 19, 1160–1172 (2013).
Article Google Scholar
Treat, C. C. et al. Temperature and peat type control CO₂ and CH₄ production in Alaskan permafrost peats. Glob. Chang. Biol. 20, 2674–2686 (2014).
Article CAS Google Scholar
Hobbie, S. E., Schimel, J. P., Trumbore, S. E. & Randerson, J. Controls over carbon storage and tureover in high-latitude soils. Glob. Chang. Biol. 6, 196–210 (2000).
Article Google Scholar
Keller, J. K., Bauers, A. K., Bridgham, S. D., Kellogg, L. E. & Iversen, C. M. Nutrient control of microbial carbon cycling along an ombrotrophic-minerotrophic peatland gradient. J. Geophys. Res. https://doi.org/10.1029/2005jg000152 (2006).
Chen, H. et al. A historical overview about basic issues and studies of mires (in Chinese). Sci. Sin. 51, 15–26 (2020).
Google Scholar
Ridl, J. et al. Plants rather than mineral fertilization shape microbial community structure and functional potential in legacy contaminated soil. Front. Microbiol. 7, 1–10 (2016).
Article Google Scholar
Kane, E. S. et al. Response of anaerobic carbon cycling to water table manipulation in an Alaskan rich fen. Soil Biol. Biochem. 58, 50–60 (2013).
Article CAS Google Scholar
Carrell, A. A. et al. Experimental warming alters the community composition, diversity, and N₂ fixation activity of peat moss (Sphagnum fallax) microbiomes. Glob. Chang. Biol. 25, 2993–3004 (2019).
Article Google Scholar
Lamit, L. J. et al. Patterns and drivers of fungal community depth stratification in Sphagnum peat. FEMS Microbiol. Ecol. 93, fix082 (2017).
Article Google Scholar
Harrison, R. B., Footen, P. W. & Strahm, B. D. Deep soil horizons: contribution and importance to soil carbon pools and in assessing whole-ecosystem response to management and global change. Forest Sci. 57, 67–76 (2011).
Google Scholar
Krüger, J. P., Leifeld, J., Glatzel, S., Szidat, S. & Alewell, C. Biogeochemical indicators of peatland degradation – a case study of a temperate bog in northern Germany. Biogeosciences 12, 2861–2871 (2015).
Article Google Scholar
Franzén, L. G. Increased decomposition of subsurface peat in Swedish raised bogs: are temperate peatlands still net sinks of carbon? Mires Peat 1, 3 (2006).
Google Scholar
Eilers, K. G., Lauber, C. L., Knight, R. & Fierer, N. Shifts in bacterial community structure associated with inputs of low molecular weight carbon compounds to soil. Soil Biol. Biochem. 42, 896–903 (2010).
Article CAS Google Scholar
de Graaff, M. A., Jastrow, J. D., Gillette, S., Johns, A. & Wullschleger, S. D. Differential priming of soil carbon driven by soil depth and root impacts on carbon availability. Soil Biol. Biochem. 69, 147–156 (2014).
Article Google Scholar
Peay, K. G., Kennedy, P. G. & Brun, T. D. Fungal community ecology: a hybrid beast with a molecular master. BioScience 58, 799–810 (2008).
Article Google Scholar
Gillabel, J., Cebrian, B., Six, J. & Merckx, R. Experimental evidence for the attenuating effect of SOM protection on temperature sensitivity of SOM decomposition. Glob. Chang. Biol. 16, 2789–2798 (2010).
Article Google Scholar
Pries, C. E. H., Castanha, C., Porras, R. C. & Torn, M. S. The whole-soil carbon flux in response to warming. Science 355, 1420–1423 (2017).
Article Google Scholar
Hicks Pries, C. E., Schuur, E. A. G. & Crummer, K. G. Thawing permafrost increases old soil and autotrophic respiration in tundra: partitioning ecosystem respiration using δ¹³C and ∆¹⁴C. Global Chang. Biol. 19, 649–661 (2013).
Article Google Scholar
Tian, J. et al. Aerobic environments in combination with substrate additions to soil significantly reshape depth-dependent microbial distribution patterns in Zoige peatlands, China. Appl.Soil Ecol. 170, 104252 (2022).
Article Google Scholar
Feng, W. et al. Enhanced decomposition of stable soil organic carbon and microbial catabolic potentials by long-term field warming. Glob. Chang. Biol. 00, 1–12 (2017).
Google Scholar
Feng, W. et al. Methodological uncertainty in estimating carbon turnover times of soil fractions. Soil Biol. Biochem. 100, 118–124 (2016).
Article CAS Google Scholar
Liang, J. et al. Methods for estimating temperature sensitivity of soil organic matter based on incubation data: A comparative evaluation. Soil Biol. Biochem. 80, 127–135 (2015).
Article CAS Google Scholar
Cai, A., Feng, W., Zhang, W. & Xu, M. Climate, soil texture, and soil types affect the contributions of fine-fraction-stabilized carbon to total soil organic carbon in different land uses across China. J. Environ. Manag. 172, 2–9 (2016).
Article CAS Google Scholar
Liu, L. et al. Response of anaerobic mineralization of different depths peat carbon to warming on Zoige plateau. Geoderma 337, 1218–1226 (2019).
Article CAS Google Scholar
Waldrop, M. et al. Molecular investigations into a globally important carbon pool: permafrost protected carbon in Alaskan soils. Glob. Chang. Biol. 16, 2543–2554 (2014).
Google Scholar
Mooshammer, M., Wanek, W., Zechmeister-Boltenstern, S. & Richter, A. Stoichiometric imbalances between terrestrial decomposer communities and their resources: mechanisms and implications of microbial adaptations to their resources. Front. Microbiol. 5, 22 (2014).
Article Google Scholar
Blagodatskaya, E. & Kuzyakov, Y. Mechanisms of real and apparent priming effects and their dependence on soil microbial biomass and community structure: critical review. Biol. Fertil. Soils 45, 115–131 (2008).
Article Google Scholar
Chen, H. et al. The carbon stock of alpine peatlands on the Qinghai–Tibetan Plateau during the Holocene and their future fate. Quat. Sci. Rev. 95, 151–158 (2014).
Article Google Scholar
Sun, G. A study on the mineral formation law, classifictation and reserves of the peat in the Rouergai Plateau. J. Nat. Res. 7, 334–345 (1992).
Google Scholar
Liu, L. et al. Responses of peat carbon at different depths to simulated warming and oxidizing. Sci. Total Environ. 548-549, 429–440 (2016).
Article CAS Google Scholar
Liu, L. et al. Water table drawdown reshapes soil physicochemical characteristics in Zoige peatlands. Catena 170, 119–128 (2018).
Article CAS Google Scholar
Liu, L. et al. Carbon stock stability in drained peatland after simulated plant carbon addition: Strong dependence on deeper soil. Sci. Total Environ. 848, 157539 (2022).
Article CAS Google Scholar
Yang, Z. et al. Soil properties and species composition under different grazing intensity in an alpine meadow on the eastern Tibetan Plateau, China. Environ. Monit. Assess 188, 678 (2016).
Article Google Scholar
Simpson, M. J. & Simpson, A. J. The chemical ecology of soil organic matter molecular constituents. J. Chem. Ecol. 38, 768–784 (2012).
Article CAS Google Scholar
Lalonde, K., Mucci, A., Ouellet, A. & Gelinas, Y. Preservation of organic matter in sediments promoted by iron. Nature 483, 198–200 (2012).
Article CAS Google Scholar
Deforest, J. L., zak, D. R., Pregitzer, K. S. & Burtonf, A. J. Atomspheric nitrate deposition and enhanced dissolved organic carbon leaching: test of a potential mechanism. Soil Sci. Soc. Am. J. 69, 1233–1237 (2005).
Article CAS Google Scholar
Schadel, C. et al. Circumpolar assessment of permafrost C quality and its vulnerability over time using long-term incubation data. Glob. Chang. Biol. 20, 641–652 (2014).
Article Google Scholar
Bell, M. & Lawrence, D. Soil carbon sequestration - myths and mysteries. Department of Primary Industries and Fisheries, Queensland Government (2009).
Schadel, C., Luo, Y., David Evans, R., Fei, S. & Schaeffer, S. M. Separating soil CO₂ efflux into C-pool-specific decay rates via inverse analysis of soil incubation data. Oecologia 171, 721–732 (2013).
Article Google Scholar
Caporaso, J. G. et al. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl Acad. Sci. USA 108, 4516–4522 (2011).
Article CAS Google Scholar
Gardes, M. & Bruns, T. D. ITS primers with enhanced specificity for basidiomycetes - application to the identification of mycorrhizae and rusts. Mol. Ecol. 2, 113–118 (1993).
Article CAS Google Scholar
White, T. J. in PCR-Protocols: A Guide to Methods and Applications (Academic Press, 1990).
Bell, C. et al. High-throughput fluorometric measurement of potential soil extracellular enzyme activities. J. Vis. Exp. 81, e50961 (2013).
Google Scholar
DeForest, J. L. The influence of time, storage temperature, and substrate age on potential soil enzyme activity in acidic forest soils using MUB-linked substrates and L-DOPA. Soil Biol. Biochem. 41, 1180–1186 (2009).
Article CAS Google Scholar
Amundson, R. The carbon budget in soils. Annu. Rev. Earth Planet. Sci. 29, 535–562 (2001).
Article CAS Google Scholar
Trumbore, S. E. Potential responses of soil organic carbon to global environmental change. Proc. Natl Acad. Sci. USA 94, 8284–8291 (1997).
Article CAS Google Scholar
R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. https://www.r-project.org (2017).
Oksanen, J. et al. vegan: community ecology package. R Packag version 24-1 (2016).
Asshauer, K. P., Wemheuer, B., Daniel, R. & Meinicke, P. Tax4Fun: predicting functional profiles from metagenomic 16S rRNA data. Bioinformatics 31, 2882–2884 (2015).
Article CAS Google Scholar

Download references

Acknowledgements

This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA200500404), China Postdoctoral Science Foundation (2020M683370), the National Natural Science Foundation of China (42101063), the Second Tibetan Plateau Scientific Expedition (2019QZKK0304), and the Key Research Program of Frontier Sciences of the Chinese Academy of Sciences (QYZDB-SSW-DQC007).

Author information

Authors and Affiliations

CAS Key Laboratory of Mountain Ecological Restoration and Bioresource Utilization & Ecological Restoration and Biodiversity Conservation Key Laboratory of Sichuan Province, Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu, 610041, China
Liangfeng Liu & Huai Chen
Zoige Peatland and Global Change Research Station, Chinese Academy of Sciences, Hongyuan, 624400, China
Liangfeng Liu
CAS Center for Excellence in Tibetan Plateau Earth Sciences, Chinese Academy of Sciences (CAS), Beijing, 100101, China
Huai Chen
State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, Beijing, 100093, China
Jianqing Tian

Authors

Liangfeng Liu
View author publications
You can also search for this author in PubMed Google Scholar
Huai Chen
View author publications
You can also search for this author in PubMed Google Scholar
Jianqing Tian
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

H.C. and L.L. conceptualized the idea and designed this research; L.L. collected field samples, conducted incubation, and analyzed soil chemistry; J.T. did microbial experiments; L.L., H.C., and J.T. wrote and discussed the results of the manuscript.

Corresponding author

Correspondence to Huai Chen.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Communications Earth & Environment thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editors: Clare Davis.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Liu, L., Chen, H. & Tian, J. Varied response of carbon dioxide emissions to warming in oxic, anoxic and transitional soil layers in a drained peatland. Commun Earth Environ 3, 313 (2022). https://doi.org/10.1038/s43247-022-00651-y

Download citation

Received: 21 August 2022
Accepted: 30 November 2022
Published: 10 December 2022
DOI: https://doi.org/10.1038/s43247-022-00651-y