Green manuring is a cost-effective strategy to enhance soil organic carbon (SOC) sequestration in paddy systems, yet its effectiveness depends on water management–induced redox dynamics that regulate microbial decomposition and utilization of plant inputs. Here, we used a 13C-labeled microcosm experiment to test how brief drainage prior to flooding influences green manure-derived C transformation and native SOC mineralization. Four treatments were applied: continuous flooding without (Control) or with green manure incorporation (GM), and delayed flooding by five days without (DW) or with green manure incorporation (GMDW). Compared with GM, GMDW reduced priming-induced CO2 and CH4 emissions from native SOC by 12.4% and 97.1%, respectively, while increasing green manure-derived 13C-CO2 emissions (+21.0%) and strongly suppressing 13C-CH4 emissions (-96.6%). GMDW enhanced the incorporation of green manure C into microbial necromass, increasing 13C-bacterial and fungal necromass C by 64.0% and 204.5%, respectively, compared with GM. Therefore, the efficient accumulation of microbial necromass and the suppression of priming effects jointly underpin green manure–derived C sequestration. This shift was driven by transient aerobic conditions that stimulated plant-C degradation, increased green manure–derived dissolved organic C by 80.9%, and favored r-strategist and fungal-dominated communities, thereby accelerating microbial growth and necromass formation. Subsequently, under flooded conditions, the declined abundance of microbial C-degrading genes suggests a lower potential risk of necromass C decomposition, promoting necromass preservation and increasing the necromass accumulation coefficient. In contrast, delayed flooding without green manure enhanced microbial C-degrading gene abundance and potentially elevate the risks of native organic C decomposition. Collectively, owing to the readily decomposable character of milk vetch, delayed flooding induces a metabolic shift that accelerates the conversion of plant-derived C into microbial necromass during aerobic phases, and enhances its stabilization under subsequent anoxic conditions. This mechanism provides a practical strategy to increase SOC formation in paddy soils.
Long-term no-tillage (NT) alters soil habitat and biota organization, yet its consequences for cross-kingdom interactions and soil organic carbon (SOC) storage in humid Andisols remain unclear. We investigated a 22-year field experiment comparing NT and conventional tillage (CT), with and without cover cropping, across topsoil (0–5 cm) and subsurface soil (5–15 cm) over three seasons. Soil physicochemical properties, microbial biomass C (MBC), β-glucosidase activity (BG), bacterial and fungal gene copy numbers were analyzed. We characterized bacterial, fungal, protistan, and nematode communities, inferred SPIEC-EASI-based cross-kingdom networks, and quantified microbial necromass as an indicator of accumulated microbial-derived SOC.NT exerted stronger effects than cover cropping and created a topsoil-oriented system. In topsoil, NT increased SOC, moisture, available N & P, MBC, BG, and bacterial and fungal gene copies, whereas CT maintained relatively higher values in subsurface soil. Bacterial, but not fungal, copies strongly correlated with MBC (r = 0.86***), indicating a bacteria-associated agroecosystem. Tillage significantly altered all four biotic groups, and NT topsoil bacteria were enriched in taxa associated with polymer decomposition. NT topsoil also developed a larger, denser, and more robust network than CT topsoil, with more bacterial nodes, within-bacteria links, and bacteria–protist/nematode links. Functional composition of protistan and nematode nodes indicated a more developed bacteria-based predator–prey web under NT. Procrustes analysis suggested shared edaphic filtering for bacteria and fungi, whereas bacteria–protist/nematode concordance reflected bacteria-based trophic coupling that was maintained under NT but weakened under CT. Despite these shifts, BG/MBC remained similar, indicating limited changes in microbial metabolic efficiency. In contrast, bacterial and fungal necromass increased under NT in topsoil but decreased in subsurface soil. These results indicate that, in humid Andisols, NT promotes topsoil carbon storage mainly by increasing microbial biomass, strengthening bacteria-centered trophic coupling, and enhancing microbial necromass accumulation, rather than by shifting microbial metabolic efficiency.
Drought and nitrogen (N) limitation frequently co-occur in terrestrial ecosystems and can interactively alter plant–soil–microbe interactions that influence soil carbon (C) cycling. However, how combined stress influences root exudation, rhizosphere bacterial communities, and microbial respiration remains poorly resolved. We conducted a factorial drought × N availability experiment using wheat (Triticum aestivum cv. RB07) encompassing both drought and recovery phases. We examined treatment effects on plant growth, root exudate quantity and chemistry, rhizosphere bacterial community structure, and exudate-induced soil respiration. Nitrogen limitation imposed the strongest constraint on plant biomass and explained more variation in rhizosphere bacterial community structure than drought. In contrast, important rhizosphere functions responded non-additively to interacting stresses: combined drought × N limitation substantially increased carbon exudation per unit biomass and elevated exudate C:N during drought, while reshaping exudate metabolomic composition and bacterial community structure across stress and recovery phases. Reciprocal exudate–soil assays revealed that variation in exudate-induced respiration was associated more strongly with exudate origin than with the prior drought × nitrogen conditioning history of the soil, highlighting the importance of plant-derived inputs for microbial carbon processing. Structural equation modelling indicated a temporal shift in factors associated with respiration, from stronger associations with exudate carbon supply during drought to stronger associations with exudate composition and bacterial community structure after recovery. Together, these results demonstrate that interacting drought and nitrogen limitation reorganize rhizosphere carbon fluxes across stress and recovery, with plant-derived inputs emerging as important correlates of microbial respiration under multiple environmental stressors.
Community-level growth potential is a central bacterial life-history trait, yet its environmental controls and responses to land use remain uncertain at continental scales. Here, we analysed 264 topsoil metagenomes across Australia to test how climatic factors, edaphic habitat and land use shape bacterial growth potential. Genome-inferred growth potential showed a clear spatial pattern, increasing towards higher latitude. At the environmental level, aridity index emerged as the primary predictor, with a distinct transition near the dryland threshold (aridity index ≈ 0.34). Consistently, the stress–resource habitat analysis showed that water stress, rather than resource availability, was the primary constraint on growth potential. Beyond these environmental effects, agricultural land use was associated with an average 6.1% reduction in growth potential across habitats. These findings show that climate sets the broad-scale distribution of bacterial growth potential, habitat conditions modify its environmental controls, and agricultural land use reduces growth potential. This framework improves our understanding of how bacterial life-history strategies are organised across environmental gradients and land-use change.
Arbuscular mycorrhizal fungi (AMF) are commonly described as nutrient pipelines that transfer mineral nutrients to plants in exchange for photosynthetically derived carbon. Although this framework has been highly productive, it underrepresents the role of extraradical AMF hyphae as microbial interfaces within soil. AMF hyphae release carbon-rich compounds, shape hyphosphere microbiomes, and interact with microorganisms capable of mobilizing organic nutrient pools that AMF alone may access inefficiently. Here, we propose the necromass-broker hypothesis, whereby AMF hyphae generate carbon-subsidized microbial hotspots in which microbial growth, nutrient immobilization and turnover produce necromass-derived mineral nutrient pools that can be mobilized by hyphosphere decomposers and partly recaptured through the mycorrhizal pathway. This Perspective does not suggest that AMF are classical saprotrophs, nor that direct nutrient transfer is secondary. Rather, it proposes that AMF-mediated nutrient cycling may include an overlooked microbial-turnover component linking plant carbon allocation, hyphosphere community assembly, microbial necromass decomposition and nutrient recovery. Under this framework, AMF may alleviate plant–microbe nutrient competition not by preventing microbial immobilization, but by spatially coupling microbial turnover with delayed nutrient recycling. Recognizing AMF as potential microbial recycling interfaces provides a new perspective on nutrient cycling in the hyphosphere and may help improve soil fertility management, organic amendment strategies and microbiome-based agriculture.
Arbuscular mycorrhizal (AM) fungal communities are highly patchy at microscales, meaning sampling decisions can strongly shape the communities detected. We used a long-term fertilisation experiment to compare three sampling strategies and two sample preservation methods to assess their influence on AM fungal diversity and taxon-specific detection. We found that sequencing individual subsamples detected substantially more taxa than compositing, which missed rare fungi and introduced phylogenetically structured biases that underrepresented certain fungal lineages. Short-term drying and freezing produced comparable community profiles. Together, these results show that AM fungal detection depends strongly on sampling strategy, highlighting the need to align sampling scales with the microscale processes structuring these symbioses.
Understanding how soil biogeochemical processes develop from microscale heterogeneity remains a central challenge in soil science. Many key transformations, including microbial respiration, nutrient cycling, and greenhouse gas production, occur in spatially structured microenvironments that are difficult to capture with conventional bulk measurements. Planar optodes have emerged as a powerful tool to address this challenge by enabling two-dimensional imaging of chemical parameters such as oxygen (O2), pH, ammonia (NH3), and carbon dioxide (CO2) across soil profiles with high spatial and temporal resolution.In this Perspective, we critically examine what planar optodes can reveal about soil biogeochemistry and what they cannot. We discuss the sensing principles underlying optode measurements and review recent applications that use optode imaging to visualize redox heterogeneity, rhizosphere processes, and biogeochemical hotspots that remain invisible to conventional approaches. These observations provide a dynamic view of soil chemistry and open new opportunities to generate mechanistic hypotheses about the controls of microbial activity and nutrient transformations. Particular emphasis is placed on the strengths of optodes for identifying microsites and guiding targeted sampling strategies.At the same time, we highlight key limitations, including challenges related to calibration, optical artefacts, and the interpretation of two-dimensional concentration fields in inherently three-dimensional soil systems. We also discuss how optode imaging can be integrated with complementary techniques such as microsensors, molecular analyses, gas flux measurements, and structural imaging to better resolve links between soil structure and biogeochemical processes.Finally, we outline emerging directions that could expand the role of optodes in soil research, including their integration with data-driven and process-based models. Used thoughtfully and in combination with other approaches, planar optodes can become a central tool for investigating soil biogeochemistry at the microscale.
Management practices such as cover cropping can reshape soil microbial niches by modifying the distribution of carbon (C) and nitrogen (N) substrates, oxygen availability, and redox conditions that regulate microbial metabolism. Yet how these shifts in microbial functional groups alter the trade-off between soil C storage and nitrous oxide (N2O) production remains poorly resolved. Cover cropping provides a particularly relevant model to address this question because it is widely promoted as a climate-smart strategy, yet its net climate impact depends on how cover crop traits modify residue inputs, N availability, and microsite redox heterogeneity that govern microbial C and N transformations. In this perspective, we synthesize recent advances in understanding how cover crop functional traits regulate microbial N cycling, soil C–N interactions, and N2O emissions. Legume and grass cover crops differ in N acquisition strategy, biomass accumulation, and residue chemistry, generating contrasting effects on soil C inputs, N availability, residue decomposition, oxygen diffusion and microsite redox conditions. These differences can shift the substrate and redox conditions that govern nitrification, denitrification, and coupled N2O-producing pathways during cover crop growth and residue decomposition. As a case study of how cover crop traits may filter microbial functional groups, we examined field studies evaluating the response of ammonia-oxidizing microorganisms. Legume cover crops tended to increase the abundance of ammonia-oxidizing bacteria and potentially increased their contributions to nitrification-derived N2O emissions, whereas responses of ammonia-oxidizing archaea were more variable. We argue that predicting the net climate impact of cover crops requires microbial–biogeochemical approaches that integrate microbial community analyses with gross N transformation rates, N2O source partitioning, and measurements of soil microsite conditions, thereby clarifying the biological processes that regulate the trade-offs between soil C storage and N2O emissions.
Soil fungi are highly diverse and functionally pivotal in terrestrial ecosystems, yet the biogeographical patterns in anoxic wetlands remain poorly understood. Here, we synthesized a global dataset comprising 1,276 fungal amplicon sequencing samples from 239 sites and 311 biomass samples from 118 sites across natural wetlands. Using machine learning models, we identified key environmental drivers and resolved the global distribution patterns of fungal diversity and biomass. We found that fungal diversity peaked in inland wetlands, followed by coastal wetlands and peatlands. Notably, fungal amplicon sequence variants (ASVs) richness reached its maximum in mid-latitude regions (30–60°N) and was driven primarily by temperature variability and vegetation characteristics. In contrast, fungal biomass peaked in high-latitude (>60°N) regions and was positively associated with soil organic carbon (SOC) in anoxic wetlands. These findings refine our global understanding of mycological patterns and reveal predictable shifts in fungal communities along climatic, vegetation, and soil environmental gradients in anoxic wetlands.
Temperature and soil phosphorus strongly influence microarthropod communities and may affect their trophic structure and energy channeling. In this study, we investigated the two most abundant detritivore microarthropod taxa, Collembola and Oribatida, and examined changes in their trophic structure and basal resource use across temperature and phosphorus gradients. We applied two state-of-the-art approaches at the community level: bulk stable isotope analysis and compound-specific stable isotope analysis of amino acids. Samples were taken at 12 forest sites in central Europe along three temperature and four soil phosphorus levels. Estimates of trophic positions from both amino acid-specific stable isotope analysis (TPCSIA) and bulk tissue Δ15N values (calculated by calibrating δ15N values to litter as baseline) showed significant responses to temperature and phosphorus availability, but this depended on microarthropod groups. Generally, TPCSIA of Collembola (3.00 ± 0.25; mean ± SD) was higher than that of Oribatida (2.66 ± 0.24) but this was not pronounced at sites with low temperature and rich phosphorus. Bayesian mixing models identified fungi as the main source of essential amino acids (eAAs) in Collembola (37.3–59.6%), while in Oribatida it was bacteria (39.9–50.2%). With increasing temperature, the contribution of fungal and bacterial eAAs increased, while the contribution of plant eAAs declined. At low temperature, plants comprised the second most important basal resource in both Collembola and Oribatida, accounting for 17.2 – 46.1% and 26.0 – 46.2% of total basal resources, respectively. Our findings reveal how Collembola and Oribatida partition resources and adjust trophic positions in response to environmental gradients. Collembola occupy higher but less flexible trophic positions at low and intermediate temperature with strong fungal reliance, whereas Oribatida are more plastic and rely more on bacterial pathways. Both groups are able to shift towards plant-derived resources when microbial processing or nutrient availability is limited. The results advance our understanding of soil microarthropod ecology and highlight the interplay between environmental conditions, resource use and trophic structure in shaping soil food webs.
While litter-dwelling earthworms are widespread in northern Scandinavian forests, burrowing earthworms, i.e., species commonly considered as ecosystem engineers, are largely missing in these ecosystems. We hypothesized that their absence results not only from environmental filtering by unfavorable soil and climatic conditions but also from dispersal limitation. We examined earthworm presence-absence across Swedish forests by environmental DNA (eDNA) sequencing of 312 soil inventory samples. We then used statistical models to identify factors best explaining their occurrence among a set of variables representing soil properties, climatic factors, vegetation and distances to anthropogenic activities (e.g. roads and agricultural fields). Soil pH and texture were strong explanatory variables for the presence of burrowing earthworms, while distance to agricultural fields explained a significant part of the remaining variation. The importance of proximity to agricultural fields was especially pronounced in northern Sweden (> 60° latitude). In contrast, occurrence of litter-dwelling earthworms was only weakly linked to environmental variables. Additionally, based on random forest modelling we predicted earthworm presence for almost 7000 soil inventory sites, and created high-resolution probability maps of earthworm occurrence in Swedish forests, including a scenario without potential dispersal limitation. These maps show that occurrence of burrowing earthworms would be more likely in large parts of the northern inland forests of Sweden without dispersal limitations, suggesting that these forests may be more susceptible to earthworm establishment upon introduction than previously anticipated. Our results highlight the importance of anthropogenic activities in dispersion and distribution of burrowing earthworms in northern forests. While environmental filtering largely governs burrowing earthworm occurrence, their current distribution is also substantially influenced by dispersal limitation - an underestimated factor to date.
The combination of extracellular depolymerization and intracellular amino acid metabolism jointly shapes soil organic nitrogen (SON) mineralization, but whether agricultural management alters the relative importance of these stages remains unclear. We combined measurements of N mineralization rates, SON pools, and extracellular enzyme activities with metagenomic and metabolomic analyses in a nine-year field experiment comprising three rice-based rotations: fallow-rice (FR), oilseed rape-rice (OR), and wheat-rice (WR), each managed with or without NPK fertilization. Relative to the WR rotation, the FR and OR rotations significantly increased soil N mineralization rates by 36%–60% and 16%–23%, respectively. The FR and OR rotations were characterized by larger pools of total N, protein, and dissolved organic N, higher potential activities of β-N-acetylglucosaminidase and protease, and greater relative abundances of genes encoding these enzymes. Amino acids and peptides were relatively depleted during incubation. Compared with WR rotation, the FR and OR rotations showed greater relative enrichment of amino acid and peptide related metabolites and higher relative abundances of genes involved in amino acid metabolic pathways, particularly alanine, aspartate, and glutamate metabolism and arginine biosynthesis. The relative abundances of genes encoding key enzymes, including glutamate dehydrogenase and acetylornithine aminotransferase, were positively associated with N mineralization rates. NPK fertilization increased N mineralization rates and was associated with higher relative abundances of genes encoding leucine aminopeptidase and genes involved in arginine biosynthesis, but it decreased the relative abundance of genes encoding protease. Collectively, our results show that crop rotation and fertilization regulate different stages of the SON mineralization pathway. Crop rotation effects are linked to coordinated variation in substrate supply, activities of organic N-depolymerizing enzymes, and potential for amino acid metabolism, while fertilization shifts microbial N acquisition toward low-molecular-weight organic N.
Microbial communities are central to soil ecosystem function. However, the extent to which functional diversity is conserved across communities, providing resilience to environmental change, remains uncertain. Here, we investigated how microbial legacy and soil properties shape community assembly and function, by cross-inoculating distinct microbial communities into sterilised soils from agricultural and semi-natural habitats. Over a 10-month incubation, the soil environment drove microbial community convergence at high taxonomic ranks, but fine-scale community composition and functional outcomes remained distinct. Microbial communities showed a ‘home-field advantage’ in soil carbon use that increased cumulative respiration by 16-26% in agricultural soils and by 26-84% in semi-natural soils, demonstrating limited redundancy of broad ecological function between soil communities. Distinct communities also caused significant shifts in soil pH associated with contrasting inorganic nitrogen transformations, exposing limited conservation of specialised metabolic functions. In summary, microbial community legacy had a lasting influence on carbon and nitrogen cycling, and thus, the effects of anthropogenic land use change on soil microbial functional diversity will likely have substantial impacts on these key ecosystem processes.
Pasture soils are major sources of nitrous oxide (N2O), a potent greenhouse gas mainly produced through microbial denitrification. However, interactions between soil aeration status and microbial responses involved in regulating N2O emissions remain poorly understood. We investigated denitrifier gene expression together with N2O and dinitrogen (N2) emissions using repacked cores of three pasture soils with contrasting bulk densities incubated across ten matric potentials (−1 to −10 kPa). Peak N2O fluxes occurred at matric potentials of −1 to −2 kPa and were strongly related to relative gas diffusivity (Dp/Do) and water-filled pore space (WFPS). In contrast, the expression of nirS, norB, and nosZ was more closely associated with matric potential and volumetric water content than with Dp/Do and WFPS. However, denitrifier gene transcription alone did not explain the observed gaseous N emissions, indicating that microbial transcriptional responses and measured N fluxes were decoupled under changing soil aeration conditions. These findings suggest bulk soil aeration indices (e.g. Dp/Do or WFPS) do not fully represent the local conditions regulating microbial response, whereas measured gaseous N emissions are additionally constrained by soil physical processes governing gas transport. Overall, these results highlight the importance of integrating microbial responses with soil physical processes to improve our mechanistic understanding of N2O emissions from grazed pasture soils.
Microbial extracellular polymeric substances (EPS) are key chemical bonding compounds in soil, yet their capacity to form and stabilize soil aggregates independently of microbial processes remains unclear. Here, we used a controlled microcosm experiment with selective biocide treatments to disentangle the roles of EPS and soil microbiota in aggregation. We found that bacterial-dominated soil produced substantial polysaccharide-rich EPS, which facilitated microaggregate formation. However, despite elevated EPS-polysaccharide concentrations, aggregates lacking fungal cellular integrity and activity rapidly disintegrated upon wetting. This suggests that fungal hyphal networks are critical for macroaggregate stability, likely through a combination of physical enmeshment and continuous production of protein-rich EPS. While EPS represent promising soil management tools, their effectiveness critically depends on the presence of active, interacting microbial communities.
Ecoenzyme stoichiometry (EEST) models are widely used to estimate microbial resource limitation, nutrient use efficiencies, threshold element ratios, and soil carbon sequestration potential, but uncertainties complicate their application and interpretation. For example, (1) the diagnostic enzyme activities only approximate target resource acquisition, with some enzymes involved in acquiring multiple nutrients; (2) evaluating resource availability is imprecise with neither dissolved nor bulk nutrient measurements fully representing availability to microbes; (3) microbial biomass stoichiometry may not consistently define maximum microbial growth because the stoichiometry of growing and non-growing microbes may differ; and (4) multiple environmental factors (e.g., temperature, pH, texture) simultaneously and differentially influence microbial metabolism, enzyme activities and resource availability. Based on a careful examination of the available evidence, we recommend the simultaneous evaluation of individual enzyme activities linked to the carbon (C)-, nitrogen (N)-, and phosphorus (P) cycles, together with enzyme C:N, C:P, and N:P activity ratios to better identify microbial resource limitation. Moreover, we anticipate that future research should unravel ecoenzyme induction mechanisms at the molecular level because environmental cues and resource availability strongly influence enzyme expression and thus the accuracy of EEST model predictions. To that end, parallel, high-resolution analyses of resource composition, corresponding enzyme activities, and key environmental drivers, using advanced multi-omics technologies such as metaproteomics, meta-transcriptomics, exometabolomics, and high-resolution mass spectrometry, are needed to more clearly link the mechanisms underlying process controls defining EEST models. Given the expanding use of EEST models, these considerations are timely to ensure their appropriate use and interpretation as well as maximize their potential contributions to understanding ecosystem functioning under changing environmental conditions.
