Deterministic assembly of grassland soil microbial communities driven by climate warming amplifies soil carbon loss.

Perturbations in soil microbial communities caused by climate warming are expected to have a strong impact on biodiversity and future climate-carbon (C) feedback, especially in vulnerable habitats that are highly sensitive to environmental change. Here, we investigate the impact of four-year experimental warming on soil microbes and C cycling in the Loess Hilly Region of China. The results showed that warming led to soil C loss, mainly from labile C, and this C loss is associated with microbial response. Warming significantly decreased soil bacterial diversity and altered its community structure, especially increasing the abundance of heat-tolerant microorganisms, but had no effect on fungi. Warming also significantly increased the relative importance of homogeneous selection and decreased "drift" of bacterial and fungal communities. Moreover, warming decreased bacterial network stability but increased fungal network stability. Notably, the magnitude of soil C loss was significantly and positively correlated with differences in bacterial community characteristics under ambient and warming conditions, including diversity, composition, network stability, and community assembly. This result suggests that microbial responses to warming may amplify soil C loss. Combined, these results provide insights into soil microbial responses and C feedback in vulnerable ecosystems under climate warming scenarios.

Natural restoration exhibits better soil bacterial network complexity and stability than artificial restoration on the Loess Plateau, China.

Natural restoration (NR, e.g., secondary succession) and artificial restoration (AR, e.g., afforestation) are key approaches for rehabilitating degraded land; however, a comparative assessment of microbial network between these approaches is lacking. We compared bacterial networks under NR and AR in two different watersheds on the Loess Plateau. Our findings revealed significantly heightened network complexity under NR compared to AR, including metrics such as node, edge, modularity, degree, centrality, and keystone nodes. NR's network robustness exceeded AR by 19.45-35.9% and 7.79-17.74% in the two watersheds, aligning with the ecological principle that complexity begets stability. The significantly higher negative/positive cohesion and natural connectivity under NR also support its better network stability than AR. Integrated analysis of paired sequencing data from five Loess Plateau studies conducted on the Loess Plateau further confirmed the higher complexity and stability of bacterial networks under NR. Further analysis unveiled "biological interactions" as primary drivers of bacterial co-occurrence (on average 84.21% of links), surpassing the influence of environmental filtering (5.17%) or dispersal limitation (4.2%). Importantly, networked communities under NR exhibited generally stronger linkages with various ecosystem function than AR. Overall, our study provides insights into vegetation restoration strategies from the perspective of microbial network, underscoring natural regeneration's potential as a superior remedy for degraded-land restoration.

Decreased soil multifunctionality is associated with altered microbial network properties under precipitation reduction in a semiarid grassland.

Our results reveal different responses of soil multifunctionality to increased and decreased precipitation. By linking microbial network properties to soil functions, we also show that network complexity and potentially competitive interactions are key drivers of soil multifunctionality.

[Effects of Warming and Increased Precipitation on Soil Respiration of Abandoned Grassland in the Loess-Hilly Regions].

Clarifying the changing trends and driving factors of soil respiration in fragile habitats under the background of climate change is of great significance for understanding the regional carbon cycle and the conversion of ecosystem carbon source and sink functions. This research focused on grasslands that had been naturally abandoned and restored for 12 years in the loess hilly region of northern Shaanxi, using an open top chamber (OTC) and artificially increased natural rainfall to simulate climate warming and precipitation increase and their interaction. Furthermore, we used a combination of field monitoring and indoor analysis to explore soil water content, temperature, and nutrient characteristics and the response characteristics of soil respiration rate to warming and increased precipitation and further analyzed the key factors driving changes in soil respiration. The results showed that:① warming (W) significantly increased the 5 cm soil temperature, with an average increase of 1.34℃ throughout the sampling year, whereas the increased precipitation (P50%) treatment significantly reduced the 5 cm soil temperature, reducing the average 5 cm soil temperature during the entire sampling year by 0.88℃ and increasing the soil water content (SWC) at the same time. The SWC was 13.12% and 16.45% higher than that in the control (CK), respectively. In addition, compared with that in the CK, the treatment of warming and increased precipitation (WP50%) not only increased soil temperature but also increased SWC; in general, the increase in temperature and precipitation played an antagonistic effect on the influence of soil temperature and humidity. ② P50% significantly increased the content of soil organic carbon, dissolved organic carbon, and labile organic carbon, causing changes in the soil stoichiometric ratio and the distribution characteristics of labile-recalcitrant carbon components, whereas W did not have a significant impact on organic carbon. In addition, soil total nitrogen and phosphorus and available nitrogen and phosphorus nutrients were not significantly different between treatments. ③ P50% significantly increased the Rs rate, and the effect of W on the soil respiration rate mainly depended on the seasonal precipitation and temperature. It was demonstrated that warming in winter and seasons with abundant rainfall had a significant promotion effect on the soil respiration rate. The exponential fitting of soil respiration rate and 5 cm soil temperature found that the soil respiration temperature sensitivity (Q10) was the highest under the precipitation treatment, reaching 1.68, whereas the Q10 was the lowest under the warming treatment (1.50). ④ Linear regression analysis showed that soil organic carbon, dissolved organic carbon, and labile organic carbon were all significantly positively correlated with soil respiration rate. Variation partitioning analysis showed that soil temperature, SWC, and nutrient characteristics explained 64.43% of the variation in soil respiration rate. The soil temperature and SWC were the main controlling factors of the change in soil respiration rate, with an explanation degree of 31.16%. Correlation analysis also showed that there was a significant correlation between SWC, soil temperature and respiration rate, soil organic carbon, dissolved organic carbon, labile organic carbon, C:N, and C:P. In summary, the climate prediction of abandoned grassland tending toward warm temperatures and high humidity in the loess hilly region will significantly affect the regional hydrothermal environment and nutrient characteristics, change the distribution ratio of soil labile and recalcitrant carbon, and promote regional soil carbon emissions. The analysis results showed that the key factor driving the change in soil respiration rate of abandoned grassland in the loess hilly region was soil temperature and SWC characteristics.

Microbial traits determine soil C emission in response to fresh carbon inputs in forests across biomes.

Soil priming is a microbial-driven process, which determines key soil-climate feedbacks in response to fresh carbon inputs. Despite its importance, the microbial traits behind this process are largely undetermined. Knowledge of the role of these traits is integral to advance our understanding of how soil microbes regulate carbon (C) emissions in forests, which support the largest soil carbon stocks globally. Using metagenomic sequencing and 13 C-glucose, we provide unprecedented evidence that microbial traits explain a unique portion of the variation in soil priming across forest biomes from tropical to cold temperature regions. We show that microbial functional profiles associated with the degradation of labile C, especially rapid simple sugar metabolism, drive soil priming in different forests. Genes involved in the degradation of lignin and aromatic compounds were negatively associated with priming effects in temperate forests, whereas the highest level of soil priming was associated with ß-glucosidase genes in tropical/subtropical forests. Moreover, we reconstructed, for the first time, 42 whole bacterial genomes associated with the soil priming effect and found that these organisms support important gene machinery involved in priming effect. Collectively, our work demonstrates the importance of microbial traits to explain soil priming across forest biomes and suggests that rapid carbon metabolism is responsible for priming effects in forests. This knowledge is important because it advances our understanding on the microbial mechanisms mediating soil-climate feedbacks at a continental scale.

[Effects of Farmland Abandonment on Soil Enzymatic Activity and Enzymatic Stoichiometry in the Loess Hilly Region, China].

Clarifying the characteristic of soil enzymatic activity and stoichiometry variations as well as their influencing factors following farmland abandonment have important implications for understanding soil nutrient availability after revegetation and for illuminating the underlying mechanisms of soil nutrient cycling in ecosystems. To determine microbial nutrient limitations after farmland abandonment and to explore the driving factors of the variations in soil enzymatic activity and stoichiometry along a chronosequence of abandoned farmlands (0-, 10-, 20-, and 30-year-old) in the Loess Hilly Region, China, the potential activities of carbon (C)-, nitrogen (N)-, and phosphorus (P)-acquiring enzymes, soil physicochemical properties, and plant diversity and family composition were measured. The results showed that the activities of ß-1,4-N-acetylglucosaminidase (NAG), leucine aminopeptidase (LAP), and alkaline phosphatase (ALP) increased significantly with the increasing years of land abandonment, whereas the activity of ß-1,4-glucosidase (BG) showed the opposite change trend. Additionally, the ratios of BG:(NAG+LAP) and BG:ALP had the same variation trend with BG activity, which decreased significantly with increasing time, but the ratio of (NAG+LAP):ALP showed an increasing trend and then decreased, with the highest values observed in the 20-year sites. Moreover, the vector length of soil enzymatic stoichiometry decreased significantly as the years of land abandonment inceased, suggesting a reduced microbial C limitation after farmland abandonment. The vector angles <45°were observed at farmlands (0-year sites) and 10-year sites, whereas angles >45°were detected at 20-and 30-year sites, indicating that soil microbial communities were N-limited in the first 10 years of land abandonment and thereafter were P-limited. The redundancy analysis (RDA) reveled that soil organic C content, total N content, the C:N and C:P ratios, soil pH values, and plant diversity had significant effects on soil enzymatic activity and stoichiometry. A variation partitioning analysis (VPA) further demonstrated that edaphic and vegetation factors explained 62.0% of the total variance of soil enzymatic activity and stoichiometry. It should be noted that the interaction between vegetation characteristics and soil physicochemical properties was the major factor affecting soil enzymatic activity and stoichiometry, which explained 37.1% of the variance of the soil enzyme characteristics. Collectively, the application of P fertilizer should be considered to mitigate the deficiency of available P in the ecosystem during farmland abandonment, and these findings may provide a theoretical basis for understanding the mechanisms underlying microbe-mediated biogeochemical cycles as well as guiding soil nutrient management and the sustainable development of the ecological environment.

[Relationship Between the Composition of Soil Aggregates and the Distribution of Organic Carbon Under Long-Term Abandoned Restoration].

Soil aggregates are important carriers of soil organic carbon (SOC) accumulation, and play an important role in the evaluation of soil structure and quality. Natural recovery can promote change in soil aggregate structure and quantity via the redistribution of SOC in the aggregates. Natural restoration from farmland is an important vegetation restoration model on the Loess Plateau. The changes in soil aggregate structure and soil carbon stock after natural restoration have received extensive attention. However, little is known about the continuous study of soil changes on the abandoned grassland during the recovery process. Therefore, to understand how SOC accumulates in the process of natural recovery and quantitatively analyze the contribution of aggregates to the total soil carbon pool, we selected four abandoned grasslands of different restoration ages on the Loess Plateau, China, and studied the changes in soil structure, soil total organic carbon (TOC), soil C:N, soil aggregate distribution, soil aggregate stable index (mean weight diameter, MWD; geometric mean diameter, GMD), and aggregate-associated SOC changes as well as their correlations from 0-20 cm and 20-40 cm soil layers in abandoned grasslands. In addition, we calculated the contribution of aggregates with different sizes to soil TOC stock. The results showed that:① natural restoration increased the macroaggregate amount, MWD, and GMD, but decreased the amount of microaggregate and silt-and clay-sized fractions. There are significant differences in the distribution and stability of aggregates between different soil layers; the promotion effect of the surface was higher than that of the subsurface soils. ② In the 42 years after abandoning recovery, soil TOC stock, macaggregate-and mesaggregate-associated SOC stock increased significantly, and varied with soil depth and years of abandonment (1.92 times, 10.2 times, and 3.61 times). In contrast, micaggregate-associated SOC stock decreased significantly, and silt-and clay-sized fractions-associated SOC stock showed no distinct change. In addition, natural restoration promoted the ratio of C:N; nevertheless, the ratio of C:N under the surface showed a reduced phenomenon after 42 years of abandonment. ③ The improvement in soil TOC stock depends primarily on changes in the macaggregate-associated organic carbon stocks, which account for 80% of macaggregate, and the significant increase in the amount of macaggregate is the main reason for the high contribution.The results of our study suggest that natural restoration is conducive to the accumulation of soil organic carbon, and improvement in soil structure and stability. Macroaggregate is the key factor in soil organic carbon accumulation and soil structure improvement in the process of natural restoration.

Effects of Robinia pseudoacacia afforestation on aggregate size distribution and organic C dynamics in the central Loess Plateau of China: A chronosequence approach.

Afforestation has been proven to have enormous potential for carbon (C) sequestration; however, the dynamics of aggregate-associated organic carbon (OC) following afforestation and their contribution to changes in bulk soil OC are not well understood in regions with serious soil erosion. Therefore, we investigated the dynamics of OC associated with aggregates along a Robinia pseudoacacia (RP) afforestation chronosequence in the Loess Plateau. Soil aggregate size distribution and OC dynamics in bulk soil were analyzed 10, 18, 28, and 42 years after RP afforestation at depths of 0-20 cm and 20-40 cm. Results showed that total macroaggregates (>0.25 mm), mean weight diameter, and geometric mean diameter increased significantly with stand age, after 42 years of afforestation, increased by 433.5%, 437.2%, 302.1% in the 0-20 cm depth, respectively, while microaggregate amounts decreased by 52.9%, and the proportions of silt + clay fraction showed no obvious changes. Long-term afforestation increased OC content and stock, both in bulk soil (245.6% and 222.9% in the 0-20 cm depth, respectively) and soil aggregates. The improvement of soil structure and enrichment of OC stocks were greater at the 0-20 cm depth than the 20-40 cm depth. In addition, small macroaggregates (2-0.25 mm) contained the highest OC content and microaggregates (<0.025 mm) had the highest OC stocks regardless of soil depth and stand age. Across the afforestation chronosequence, OC content and stock in bulk soil positively correlated with large macroaggregate (>2 mm) amounts and small macroaggregate (2-0.25 mm) associated OC dynamics (P < 0.01). These results indicated that changes in bulk soil OC dynamics mainly depend on changes in the proportion of large macroaggregates and in the OC dynamics associated with small macroaggregates after RP afforestation.

Dynamics of bacterial community in litter and soil along a chronosequence of Robinia pseudoacacia plantations.

As the driver of plantation ecosystem function, microorganisms can decompose plant residues and soil organic matter. To identify dynamics of microbial communities in litter and soil and its influence by vegetation and soil at regional scales, the plantations of Robinia pseudoacacia at different successional stages (13, 19, 29, and 44 y) was selected on the Loess Plateau. High-throughput sequencing of the 16S rRNA gene was used to examine bacterial communities in litter and soil, and changes in vegetation, litter, and soil characteristics were analyzed. With increase of stand age, coverage and biomass of understory vegetation increased significantly and peaked at 44-y. Concentrations of carbon (C), nitrogen (N), and phosphorus (P) in litter and soil increased significantly, whereas pH values decreased significantly. Composition and diversity of bacterial communities in litter and soil were significantly different. Diversity and richness of litter bacterial communities were higher than that of soils. Relative abundances of Actinobacteria and Proteobacteria in litter were higher than that in soil; relative abundance of Acidobacteria exhibited the reverse trend. The diversity and richness index of vegetation significantly affected that of litter bacterial communities. Soil C/P significantly affected the Simpson and Shannon index of soil bacterial communities. The C/P and pH of litter and soil were significantly correlated with bacterial composition, primarily including Actinobacteria, Acidobacteria, and Gemmatimonadetes. Diversity of litter bacterial communities was more sensitive to the diversity and richness of vegetation flora than that of soil in the succession of R. pseudoacacia. Canopy density, vegetation, and litter and soil nutrients might directly or indirectly affect bacterial communities. Carbon, phosphorus, and pH may be critical factors influencing the composition of bacterial communities in litter and soil.

Differential responses of soil microbial biomass, diversity, and compositions to altitudinal gradients depend on plant and soil characteristics.

Alt'itudinal gradients strongly affect plant biodiversity, but the effects on microbial patterns remain unclear, especially in the large scale. We therefore designed an altitudinal gradient experiment that covered three climate zones to monitor soil microbial community dynamics and to compare those with plant and soil characteristics. Illumina sequencing of the 16S rRNA gene and ITS gene was used to analyze soil microbial (bacterial and fungal) diversity and composition, and fumigation-extraction was used to determine microbial biomass; the plant community metrics (i.e., percent cover, Shannon-Wiener, grass biomass, and carbon/nitrogen in leaf and biomass) and soil properties (i.e., soil moisture, soil temperature, bulk density, organic carbon, total nitrogen, and available nitrogen) were determined. The results showed that carbon/nitrogen in microbial biomass was higher at medium altitude and was positively related to carbon and nitrogen in both soil and grass biomass along the altitudinal gradients. Soil bacterial alpha diversity was significantly higher at medium altitude but fungal alpha diversity did not affected by altitudinal gradients; the effect of altitudinal gradients on bacterial beta diversity was larger than that on fungal beta diversity, although both groups were significantly affected by altitudinal gradients. Moreover, Alpha-proteobacteria, Beta-proteobacteria, and Gemmatimonadetes were significantly more abundant in higher altitude than in lower altitude, both Acidobacteria and Actinobacteria significantly declined with increasing altitude; other bacterial taxa such as Chloroflexi, Nitrospirae, Gamma-proteobacteria, and Delta-proteobacteria were significantly higher at medium altitudes. For fungal taxa, Basidiomycota and Ascomycota were the dominant phyla and responded insignificantly to the altitudinal gradients. The responses of microbial alpha diversity were mostly associated with plant Shannon index, organic carbon, and total nitrogen, whereas microbial beta diversity and composition mainly depended on soil moisture and temperature. Overall, these results suggest that soil bacteria rather than fungi can reflect changes in plant and soil characteristics along altitudinal gradients.