Whole-soil warming leads to substantial soil carbon emission in an alpine grassland.

The sensitivity of soil organic carbon (SOC) decomposition in seasonally frozen soils, such as alpine ecosystems, to climate warming is a major uncertainty in global carbon cycling. Here we measure soil CO2 emission during four years (2018-2021) from the whole-soil warming experiment (4 °C for the top 1 m) in an alpine grassland ecosystem. We find that whole-soil warming stimulates total and SOC-derived CO2 efflux by 26% and 37%, respectively, but has a minor effect on root-derived CO2 efflux. Moreover, experimental warming only promotes total soil CO2 efflux by 7-8% on average in the meta-analysis across all grasslands or alpine grasslands globally (none of these experiments were whole-soil warming). We show that whole-soil warming has a much stronger effect on soil carbon emission in the alpine grassland ecosystem than what was reported in previous warming experiments, most of which only heat surface soils.

Contrasting impacts of fertilization on topsoil and subsoil greenhouse gas fluxes in a thinned Chinese fir plantation.

Globally, forest soils are considered as important sources and sinks of greenhouse gases (GHGs). However, most studies on forest soil GHG fluxes are confined to the topsoils (above 20 cm soil depths), with only very limited information being available regarding these fluxes in the subsoils (below 20 cm soil depths), especially in managed forests. This limits deeper understanding of the relative contributions of different soil depths to GHG fluxes and global warming potential (GWP). Here, we used a concentration gradient-based method to comprehensively investigate the effects of thinning intensity (15% vs. 35%) and nutrient addition (no fertilizer vs. NPK fertilizers) on soil GHG fluxes from the 0-40 cm soil layers at 10 cm depth intervals in a Chinese fir (Cunninghamia lanceolata) plantation. Results showed that forest soils were the sources of CO2 and N2O, but the sinks of CH4. Soil GHG fluxes decreased with increasing soil depth, with the 0-20 cm soil layers identified as the dominant producers of CO2 and N2O and consumers of CH4. Thinning intensity did not significantly affect soil GHG fluxes. However, fertilization significantly increased CO2 and N2O emissions and CH4 uptake at 0-20 cm soil layers, but decreased them at 20-40 cm soil layers. This is because fertilization alleviated microbial N limitation and decreased water filled pore space (WFPS) in topsoils, while it increased WFPS in subsoils, ultimately suggesting that soil WFPS and N availability (especially NH4+-N) were the predominant regulators of GHG fluxes along soil profiles. Generally, there were positive interactive effects of thinning and fertilization on soil GHG fluxes. Moreover, the 35% thinning intensity without fertilization had the lowest GWP among all treatments. Overall, our results suggest that fertilization may not only cause depth-dependent effects on GHG fluxes within soil profiles, but also impede efforts to mitigate climate change by promoting GHG emissions in managed forest plantations.

Warming inhibits the priming effect of soil organic carbon mineralization: A meta-analysis.

Fresh organic carbon (C) input will accelerate or inhibit the mineralization of native soil organic carbon (SOC), which is called positive or negative priming effect (PE), respectively. However, little is known about how warming affects the PE. Here, we adopted a widely-used ratio of SOC mineralization between substrate-added and unadded-control treatments to represent PE intensity and used the PE difference between ambient-control temperature and elevated temperature to indicate the effect of warming on PE (ΔPE). By conducting a meta-analysis of 146 observations from 57 independent soils worldwide, we found that experimental warming significantly decreased the PE by 0.26 (unitless). Among ecosystems, warming significantly suppressed the PE of cropland and grassland soils by 0.43 and 0.21 respectively, but did not change the PE of forest soils. Moreover, we found significant positive correlations of ΔPE with the initial soil C/N ratio and the effect size of warming on microbial biomass. Between substrate types (i.e., containing N or not), warming significantly decreased the PE induced by N-containing substrates. These results suggested that the response of PE to warming is likely regulated by soil N availability and warming-induced changes in microbial biomass. As such, we proposed a conceptual framework-the microbial N mining hypothesis dominates in soils with low C/N ratio where warming inhibits PE by promoting N mineralization, while the stoichiometric decomposition hypothesis dominates in soils with high C/N ratio where warming stimulates PE by promoting N mineralization. Collectively, these findings provide important insights into how warming affects SOC dynamics via inhibiting PE, which may weaken the positive feedback between soil C emission and climate warming.

Whole-soil-profile warming does not change microbial carbon use efficiency in surface and deep soils.

The paucity of investigations of carbon (C) dynamics through the soil profile with warming makes it challenging to evaluate the terrestrial C feedback to climate change. Soil microbes are important engines driving terrestrial biogeochemical cycles; their carbon use efficiency (CUE), defined as the proportion of metabolized organic C allocated to microbial biomass, is a key regulator controlling the fate of soil C. It has been theorized that microbial CUE should decline with warming; however, empirical evidence for this response is scarce, and data from deeper soils are particularly scarce. Here, based on soil samples from a whole-soil-profile warming experiment (0 to 1 m, +4 °C) and 18O tracing approach, we examined the vertical variation of microbial CUE and its response to ~3.3-y experimental warming in an alpine grassland on the Qinghai-Tibetan Plateau. Microbial CUE decreased with soil depth, a trend that was primarily controlled by soil C availability. However, warming had limited effects on microbial CUE regardless of soil depth. Similarly, warming had no significant effect on soil C availability, as characterized by extractable organic C, enzyme-based lignocellulose index, and lignin phenol-based ratios of vanillyls, syringyls, and cinnamyls. Collectively, our work suggests that short-term warming does not alter microbial CUE in either surface or deep soils, and emphasizes the regulatory role of soil C availability on microbial CUE.

Ecosystem-dependent responses of soil carbon storage to phosphorus enrichment.

Phosphorus deposition can stimulate both plant carbon inputs and microbial carbon outputs. However, how P enrichment affects soil organic carbon (SOC) storage and the underlying mechanisms remain unclear. We conducted a meta-analysis of 642 SOC observations from 213 field P addition experiments world-wide and explored the regulations of plant inputs, microbial outputs, plant characteristics, and environmental and experimental factors on SOC responses. We found that, globally, P addition stimulated SOC by 4.0% (95% CI: 2.0-6.0%), but the stimulation only occurred in forest and cropland rather than in grassland. Across sites, the response of SOC correlated with that of plant aboveground rather than belowground biomass, suggesting that the change in plant inputs from aboveground was more important than that from belowground in regulating SOC changes due to P addition. Among multiple factors, plant N fixation status and mean annual temperature were the best predictors for SOC responses to P addition, with SOC stimulation being higher in ecosystems dominated by symbiotic nitrogen fixers and ecosystems in high-temperature regions like tropical forests. Our findings highlight the differential and ecosystem-dependent responses of SOC to P enrichment and can contribute to accurate predictions of soil carbon dynamics in a P-enriched world.

A distinct sensitivity to the priming effect between labile and stable soil organic carbon.

Soil organic carbon (SOC) is a mixture of various carbon (C) compounds with different stability, which can be distinctly affected by the priming effect (PE). However, little is known about how the PE changes with SOC stability. We address this issue by combining results from two experiments and a metaanalysis. We found that the PE increased with the prolongation of soil preincubation, suggesting that higher PE occurred for more stable SOC than for labile SOC. This was further supported by the metaanalysis of 42 observations. There were significant negative relationships between the difference in PE (ΔPE) between labile and more stable SOC and their differences in SOC, microbial biomass C and soil C : N ratio, indicating that soil C availability exerts a vital control on ΔPE. We conclude that, compared with labile SOC, stable SOC can be more vulnerable to priming once microbes are provided with exogenous C substrates. This high vulnerability of stable SOC to priming warrants more attention in future studies on SOC cycling and global change.

Changes in plant inputs alter soil carbon and microbial communities in forest ecosystems.

Global changes can alter plant inputs from both above- and belowground, which, thus, may differently affect soil carbon and microbial communities. However, the general patterns of how plant input changes affect them in forests remain unclear. By conducting a meta-analysis of 3193 observations from 166 experiments worldwide, we found that alterations in aboveground litter and/or root inputs had profound effects on soil carbon and microbial communities in forest ecosystems. Litter addition stimulated soil organic carbon (SOC) pools and microbial biomass, whereas removal of litter, roots or both (no inputs) decreased them. The increased SOC under litter addition suggested that aboveground litter inputs benefit SOC sequestration despite accelerated decomposition. Unlike root removal, litter alterations and no inputs altered particulate organic carbon, whereas all detrital treatments did not significantly change mineral-associated organic carbon. In addition, detrital treatments contrastingly altered soil microbial community, with litter addition or removal shifting it toward fungi, whereas root removal shifting it toward bacteria. Furthermore, the responses of soil carbon and microbial biomass to litter alterations positively correlated with litter input rate and total litter input, suggesting that litter input quantity is a critical controller of belowground processes. Taken together, these findings provide critical insights into understanding how altered plant productivity and allocation affects soil carbon cycling, microbial communities and functioning of forest ecosystems under global changes. Future studies can take full advantage of the existing plant detritus experiments and should focus on the relative roles of litter and roots in forming SOC and its fractions.

Changes in soil total, microbial and enzymatic C-N-P contents and stoichiometry with depth and latitude in forest ecosystems.

Soil microorganisms and their extracellular enzymes are key factors determining the biogeochemical cycles of carbon (C), nitrogen (N) and phosphorus (P). Relevant studies mainly focus on surface soils (0-20 cm), while deep soils (>20 cm) are often neglected, let alone comparing multiple ecosystems simultaneously. In this study, we studied the latitudinal (19-48°N) and vertical (0-100 cm) patterns of soil total, microbial and enzymatic C-N-P contents and ratios (stoichiometry) in eight temperate, subtropical and tropical forest ecosystems in eastern China. We found that the C-N-P contents and their stoichiometry in soil, microbial biomass and extracellular enzymes all varied significantly with depth and latitude. Soil total C, N and P declined with depth, as did microbial biomass and enzyme activity, while microbial and enzymatic C:N ratios showed increasing or no trend with increasing soil depth. Moreover, soil total and microbial C-N-P contents in surface soils (0-20 cm) showed positive correlations with increasing latitude, and such correlations tended to be weaker or disappeared in deep soils (>20 cm). Overall, changes in total, microbial and enzymatic C-N-P contents and ratios among latitudes suggested a shift from relative N limitation in the north to relative P limitation in the south.

Soil priming effect and its responses to nutrient addition along a tropical forest elevation gradient.

Priming plays important roles in terrestrial carbon cycling, but the patterns and drivers of priming and its responses to nutrient addition in tropical forests remain unclear. By collecting soils along a tropical forest elevation gradient, we conducted an incubation experiment with 13 C-labeled glucose and nutrient (N and/or P) additions. Results showed that priming effects increased soil organic matter decomposition by 44 ± 12% across elevations, and priming intensity decreased significantly with elevation. Among soil and microbial properties, soil organic carbon (SOC) content and pH were two key factors negatively and positively regulating priming, respectively. Across elevations, the additions of N, P, or both of them (NP) did not significantly change priming. However, the variations in the effects of nutrient (N and/or P) addition on priming significantly correlated with initial soil nutrient (N or P) availability. The intensity for the effects of N addition on priming decreased significantly with initial soil N availability, and that for the effects of P and NP addition on priming decreased with initial soil P availability. Based on these relationships, we proposed a conceptual framework linking stoichiometric decomposition and nutrient mining hypotheses, in which the former dominates in low-nutrient availability soils and the latter dominates in high-nutrient availability soils. This conceptual framework can help to explain the contrasting effects of nutrient addition on priming. Collectively, our findings highlight the roles of SOC content and soil pH in regulating priming intensity, and the role of initial soil nutrient availability in regulating the effects of nutrient addition on priming.

Responses of arbuscular mycorrhizal fungi to nitrogen addition: A meta-analysis.

Arbuscular mycorrhizal (AM) fungi play important roles in carbon (C), nitrogen (N) and phosphorus (P) cycling of terrestrial ecosystems. The impact of increasing N deposition on AM fungi will inevitably affect ecosystem processes. However, generalizable patterns of how N deposition affects AM fungi remains poorly understood. Here we conducted a global-scale meta-analysis from 94 publications and 101 sites to investigate the responses of AM fungi to N addition, including abundance in both intra-radical (host roots) and extra-radical portion (soil), richness and diversity. We also explored the mechanisms of N addition affecting AM fungi by the trait-based guilds method. Results showed that N addition significantly decreased AM fungal overall abundance (-8.0%). However, the response of abundance in intra-radical portion was not consistent with that in extra-radical portion: root colonization decreased (-11.6%) significantly, whereas extra-radical hyphae length density did not change significantly. Different AM fungal guilds showed different responses to N addition: both the abundance (spore density) and relative abundance of the rhizophilic guild decreased significantly under N addition (-29.8% and -12.0%, respectively), while the abundance and relative abundance of the edaphophilic guild had insignificant response to N addition. Such inconsistent responses of rhizophilic and edaphophilic guilds were mainly moderated by the change of soil pH and the response of root biomass, respectively. Moreover, N addition had an insignificant negative effect on AM fungal richness and diversity, which was strongly related with the relative availability of soil P (i.e. soil available N/P ratio). Collectively, this meta-analysis highlights that considering trait-based AM fungal guilds, soil P availability and host plant C allocation can greatly improve our understanding of the nuanced dynamics of AM fungal communities under increasing N deposition.

Tropical forest soils serve as substantial and persistent methane sinks.

Although tropical forest soils contributed substantially global soil methane uptake, observations on soil methane fluxes in tropical forests are still sparse, especially in Southeast Asia, leading to large uncertainty in the estimation of global soil methane uptake. Here, we conducted two-year (from Sep, 2016 to Sep, 2018) measurements of soil methane fluxes in a lowland tropical forest site in Hainan island, China. At this tropical forest site, soils were substantial methane sink, and average annual soil methane uptake was estimated at 2.00 kg CH4-C ha-1 yr-1. The seasonality of soil methane uptake showed strong methane uptake in the dry season (-1.00 nmol m-2 s-1) and almost neutral or weak soil methane uptake in the wet season (-0.24 nmol m-2 s-1). The peak soil methane uptake rate was observed as -1.43 nmol m-2 s-1 in February, 2018, the driest and coolest month during the past 24 months. Soil moisture was the dominant controller of methane fluxes, and could explain 94% seasonal variation of soil methane fluxes. Soil temperature could not enhance the explanation of seasonal variation of soil methane fluxes on the top of soil moisture. A positive relationship between soil methane uptake and soil respiration was also detected, which might indicate co-variation in activities of methanotroph and roots and/or microbes for soil heterotrophic respiration. Our study highlights that tropical forests in this region acted as a methane sink.