Most field studies of residential water-saving devices rely on aggregate meter data that cannot isolate device-level effects. We present a tap-level monitoring framework applied at HSB Living Lab, a 29-apartment facility in Gothenburg, Sweden. Over 32 months, 205,097 water use events were recorded across 80 taps (31 treated, 49 control) for nine water-saving products installed in three rounds. Event-level non-parametric tests, a difference-in-differences (DiD) framework, and three user experience surveys (n = 26) were combined.Event-level tests found that 65–81% of treated taps showed statistically significant shifts in the distribution of individual use events (median volume change –7.1%); because an event here reflects a ∼10-minute reporting window rather than a fully independent discrete use, we treat this as exploratory description rather than confirmatory evidence. The aggregate, tap-clustered difference-in-differences estimate—our confirmatory analysis—was not significant. Product-specific effects were heterogeneous; in this small, exploratory survey (n = 26 respondents, 3–10 per product), the least-liked product showed the largest measured reduction and the most-liked showed none, a suggestive pattern that would need confirmation in larger samples. Event-level monitoring can surface intervention effects that aggregate data may not resolve; we interpret this as a consequence of finer measurement resolution rather than evidence of savings hidden within aggregate figures. We emphasise, however, that statistically significant event-level shifts indicate changes in the distribution of individual water-use events, not necessarily reductions in total water use at the building scale.
Achieving a sustainable water future requires placing Africa’s water-related risks at the forefront of global sustainability agendas. Despite the urgency, systematic, continental-scale assessments remain limited. Drawing on the IPCC framework, water-related risks were evaluated through the combined influence of eight water stressors, human exposure, and societal adaptive capacity. The highest water-related risk areas were predominantly concentrated within the equatorial and sub-equatorial belt of the continent. At the basin level, the Nile, Congo, and Niger basins exhibited the greatest risk levels, while at the country level, Nigeria, the Democratic Republic of Congo, Ethiopia, and Egypt were identified as the most severely affected. These areas are further characterized by high population density, low HDI scores, limited governance capacity, and the concurrent presence of multiple water stressors. In contrast, northern African countries exhibit low WASH-service risks alongside high groundwater risks, indicating that adequate water resources do not inherently reflect underlying resource abundance, but is instead driven by effective infrastructure, governance, and investment. This study provides an analytical and cross-disciplinary framework and unveils spatial patterns and geographic archetypes for guiding future water policies and transboundary cooperation across Africa.
The COVID-19 pandemic and the Russia–Ukraine conflict challenged global food security and prompted substantial reorganization of global cereal trade. Using annual bilateral trade data and country-level production and reserve data, all converted to kilocalories, we constructed weighted directed networks for nine cereals across 194–195 countries. We combined network indicators, community detection, core–periphery analysis, and an improved cascading-failure model incorporating reserve absorption and preference-weighted trade reallocation to examine structural evolution, shock propagation, country-level exposure, and distributional inequality. The network expanded through 2021 and partially contracted in 2022–2023, while community composition and core rankings shifted markedly. Stronger export shocks produced more propagation rounds, affected more countries, increased trade losses, and reduced post-shock connectivity. Export influence remained concentrated among major exporters, whereas import vulnerability was more geographically variable. Domestic supply impacts were highly unequal and disproportionately concentrated among lower-income populations, with most inequality arising within income groups.
Battery energy storage systems (BESS) are essential for renewable energy integration, but their rapid deployment may intensify critical metal supply risks. This study combines a Gompertz growth model with dynamic material flow analysis to project lithium, cobalt, nickel, and vanadium demand in China’s BESS sector from 2025 to 2050 under ten scenarios integrating energy-transition pathways, technology substitution, and application-specific preferences. By 2050, cumulative demand is projected to reach 223-535 kt for lithium, 46-203 kt for cobalt, and 386-1689 kt for nickel, while annual vanadium demand reaches 41-269 kt across scenarios. More ambitious decarbonization pathways substantially increase resource pressure. Technology substitution reduces reliance on lithium, cobalt, and nickel but shifts part of the pressure toward vanadium. Cobalt exhibits relative scarcity across all scenarios, whereas vanadium scarcity is highly sensitive to technology mix and transition pathway; lithium and nickel are not directly constrained by resource scarcity but face rapidly increasing demand pressure. Under the medium recycling scenario, recovered metals can meet 55-63% of annual demand by 2050, although recycling provides limited near-term mitigation because of the lag between deployment and retirement. Second-life utilization provides earlier mitigation, reducing cumulative primary demand by 21-107 kt for lithium, 19-89 kt for cobalt, and 154-727 kt for nickel during 2025-2050. These findings indicate that future resource risks in China’s BESS sector will be shaped not only by deployment scale, but also by technology choice and application structure, underscoring the need to align storage planning with resource security strategies.
Nanomaterials are commonly used to modify anodes to enhance interfacial electron transfer and pollution removal in microbial electrochemical systems (MES). However, the effectiveness of nanomaterial-modified anodes at large scales under real field conditions has yet to be investigated. Here, we fabricated one of the largest reported nanomaterial-modified anodes (projected area: 2.4 m2) and applied it in a field-scale ecological floating bed MES (ECOFB-MES) for near-natural ecological restoration of real polluted water and sediment. Compared with the unmodified system, the carbon nanotube-modified anode improved conductivity, promoted electroactive biofilm activity, and enhanced direct extracellular electron transfer and potential flavin-associated indirect electron transfer, enabling the system to achieve a higher output voltage and an increased maximum power density of 3.433 ± 0.569 mW m−2. The enhanced interfacial electron transfer was associated with improved removal of sediment organic matter, nitrate reduction in the overlying water, and promoted the accumulation of more stable sediment-bound phosphorus fractions. The CNT-modified anode proposed in this study is facile to prepare at a large scale and showed six-month field operability, low biotoxicity, and good mechanical stability under accelerated shaking and hydraulic scouring tests. This study demonstrates the field-scale potential of large-area nanomaterial-modified MES anodes for enhancing interfacial electron transfer and pollutant removal, and provides useful insights into the application of carbon nanomaterials in near-natural ecological restoration.
Leguminous green manure (LGM) can enhance soil fertility in winter wheat (Triticum aestivum L.) systems but may also increase carbon dioxide (CO2) emissions. Soil amendments, such as Bacillus velezensis and sepiolite, offer a potential strategy for mitigating emissions and improving LGM-derived carbon (C) retention. However, the mechanisms by which they regulate microbial carbon-use efficiency (CUE) and nutrient limitation pathways remain unclear. A two-year field trial conducted on the Loess Plateau examined four treatments: bare fallow (CK), black bean (Phaseolus vulgaris L.) as LGM alone (BB), BB + Bacillus velezensis (BBW), and BB + sepiolite (BBH), and assessed soil nutrient dynamics, microbial enzyme stoichiometry, CUE, and CO2 emissions. BBW and BBH increased soil nutrient availability, microbial biomass, and enzyme activity relative to BB, thereby modifying microbial nutrient limitation patterns. BBH significantly reduced cumulative CO2 emissions by 17.1% over two years and increased soil organic carbon (SOC) by 12.9% at the LGM incorporation stage in 2024-2025, while enhancing CUE by 12.2%. Furthermore, BBW and BBH significantly increased grain yield by 9.8%-12.9% and 16.5%-17.4%, respectively, across the two years. The partial least squares path model demonstrated that amendments regulated CO2 emissions primarily through soil properties, enzyme activities, and CUE pathways, rather than direct microbial nutrient limitation. These findings highlight that sepiolite enhances microbial CUE and reduces CO2 losses, supporting direct C sequestration, whereas Bacillus velezensis promotes nutrient cycling and biological activity. Together, they enhance both soil C sustainability and grain yield, providing low-emission management strategies for sustainable dryland winter wheat production.
Sustainability rating schemes offer structured methods for assessing the environmental, social and economic performance of infrastructure projects. In practice, infrastructure sustainability assessment often relies heavily on manual interpretation of requirements, evidence collection, and aggregation of calculations, which leads to long assessment cycles and high costs. Therefore, this study focuses on the Infrastructure Sustainability Rating Scheme (IS Rating Scheme) and systematically examines how digital technologies can support and accelerate sustainability assessment workflows for infrastructure projects. A PRISMA-guided evidence-mapping is conducted to identify quantifiable credits, define their data needs, and map them to applicable digital technologies. The results show that core platforms, such as Building Information Modelling, Geographic Information Systems, Internet of Things, and Sensor Networks, have a strong capacity to facilitate sustainability assessments on credits related to the environment, resources, and costs. Specialised tools such as Unmanned Aerial Vehicles, LiDAR and Cloud Platforms provide targeted support for specific measurements. However, several Data Needed items, particularly those related to social equity, policy or contractual commitments, and complex economic or supply-chain information, still lack clearly reported digital support. This study proposes a conceptual framework that maps quantifiable credits in the IS Rating Scheme to digital technologies, providing evidence for integrating digital technologies into the sustainability assessment workflow, which could potentially reduce manual involvement, shorten assessment cycles, and lower costs.
Soil organic carbon (SOC) sequestration in agroecosystems has been regarded as an effective strategy to restore soil fertility, sustain productivity and meet the targets of climate change. However, it remains unclear on the underlying mechanisms of long-term fertilization regulating the formation of SOC fractions (particulate organic carbon, POC; mineral-associated organic carbon, MAOC). Here, a one-year laboratory incubation experiment was conducted to quantity the POC and MAOC formation and uncover underlying mechanisms using a 13C-labelled straw tracing approach in two texture-contrasting soils collected from 29-year history fertilization field trials (unfertilized, control; mineral fertilizer, NPK; mineral fertilizer plus manure application, NPKM) at the Urumqi (UQ) and Gongzhuling (GZL) sites. Results showed that NPKM treatment strongly increased straw-derived SOC by 38–42%, POC by 39–76% and MAOC by 28–53% compared to the corresponding NPK and control treatment, with more pronounced increase in MAOC in the GZL relative to UQ sites. The 52–81% of the straw-derived SOC was stored as MAOC across soils and treatments after year-around incubation. The straw-derived MAOC showed positive relationships with PLFAs abundances, fungal k-strategies taxa and microbial carbon use efficiency, implying close linkage between MAOC formation and microbial C anabolism. In contrast, the increased straw-derived POC in the manure-amended soils primarily resulted from enhanced aggregate stability. The relative higher contribution of straw-derived POC to SOC in the GZL soil primarily resulted from greater physical protection via aggregation as compared to UQ soil. Overall, these findings suggested that long-term manure amendment help facilitate POC and MAOC formation via physical protection and microbial metabolism mechanisms, respectively, providing perspectives for optimizing fertilization practices to enhance soil C sequestration in agricultural ecosystems.
Mining is essential to economic growth and the energy transition but presents complex Environmental, Social, and Governance (ESG) risks that vary across the mining life cycle. This study presents a PRISMA-based systematic review of 88 publications to examine how ESG factors are integrated into technical, operational, and economic decision-making across mining stages, from exploration and development to closure and post-closure. ESG integration is defined as the incorporation of measurable ESG variables into decision processes, rather than as general sustainability or disclosure practices. Unlike prior reviews, this study develops a decision-oriented synthesis linking ESG metrics to life cycle stages and project-level decisions through a three-dimensional (3 × 3 × 3) ESG integration matrix that maps three ESG dimensions across three life cycle stages and three integration levels - measurement, technical model embedding, and economic valuation - yielding 27 cells.The review evaluates ESG metrics, indicators, indices, and analytical tools according to their decision relevance, auditability, and forward-looking capacity. The results reveal a structural imbalance: environmental indicators are mature, standardized, and embedded in engineering models, particularly for emissions, energy, water, and waste. Social and governance dimensions remain less operationalized despite their importance for permitting, conflict risk, and long-term viability. Three structurally distinct barriers emerge: an economic valuation gap for environmental indicators, a measurement gap for social indicators, and a scale mismatch for governance indicators. The effectiveness of these tools is further conditioned by regional governance capacity.
Wind erosion constitutes a critical driver of land degradation in arid regions, prompting the Chinese government to implement policies such as the Grain for Green Project (GFGP) and the Grassland Ecological Compensation Policy (GECP) in Inner Mongolia. However, the efficacy of these interventions remained inadequately quantified. This study revealed a pronounced seesaw response in policy efficacy across Inner Mongolia by systematically evaluating policy impacts and clarifying their operational pathways. Key findings revealed an annual average wind erosion modulus of 62.61 t/ha across Inner Mongolia from 1982 to 2022. Regional disparities were pronounced, with the pastoral zone demonstrating severe wind erosion (106.08 t/ha), in stark contrast to the agro-pastoral ecotone (5.95 t/ha). The GFGP achieved 25.86% wind erosion reduction in the agro-pastoral ecotone through afforestation-driven mechanism, although long-term afforestation exhibited diminishing wind erosion suppression effect. Conversely, limited cropland availability nullified policy efficacy in the pastoral zone. The GECP reduced wind erosion by 11.61% in pastoral zone but paradoxically increased wind erosion by 19.85% in the agro-pastoral ecotone. These findings underscore the necessity for regionally differentiated governance strategies: prioritizing grassland compensation in pastoral zone while optimizing afforestation implementation in the agro-pastoral ecotone.
Globally abundant tea by-products (TBP) are an underexploited reservoir of natural flavonoids with methane (CH4) mitigation potential. This study innovatively developed a closed-loop framework integrating chemical characterization, in vitro screening, in vivo validation, and mechanistic elucidation. Systematic chemical profiling of 30 TBP identified distinct proximate and flavonoid compositions, revealing that green TBP are rich in catechins, whereas black TBP are dominated by theaflavins. In vitro screening determined 2% DM as the optimal dose (CH4 reduction up to 22.29%), and this effect was significantly correlated with tea polyphenol content. 16S rRNA analysis showed TBP supplementation increased Succinivibrionaceae_UCG-002 and decreased Methanobrevibacter relative abundance. Subsequently, a 150-day feeding trial was conducted with finishing Angus steers supplemented with Anshun green tea residue (ASGT, selected for in vitro efficacy and palatability). Forty-five individually housed Angus steers were randomized into three groups (0, 75, and 150 g/d ASGT). The 75 g/d supplementation enhanced average daily gain and improved feed efficiency, while the 150 g/d dose achieved greater CH4 reduction (14.41 kg CH4/steer/year, 10.1% lower than control). Metagenomic analysis elucidated the microbial mechanism: ASGT enriched propionate-producing bacteria (e.g., Prevotella), suppressed hydrogenotrophic methanogens (e.g., Methanobrevibacter), altered carbohydrate-active enzyme profiles and hydrogenase group composition, and functionally downregulated key methanogenic enzymes (e.g., methyl-coenzyme M reductase) while upregulating pathways for the TCA cycle and propionate metabolism. This study provides the first complete evidence chain supporting TBP as a natural dual-purpose additive that reduce beef cattle carbon footprint and improve productivity, addressing agricultural waste valorization and climate-smart livestock production for sustainable beef industry intensification.
Ecological restoration is increasingly implemented through integrated, multi-actor governance, yet existing evaluations that rely on indicators such as vegetation greening remain insufficient to capture multidimensional socio-ecological responses. Focusing on the Huaying area, a key mountainous region for ecological conservation in the upper Yangtze River, this study investigates the spatiotemporal dynamics of vegetation greening, ecological function, and well-being enhancement across different governance stages from 2013 to 2023. XGBoost-SHAP and partial correlation network analysis are employed to identify driver patterns, nonlinear responses, and temporal changes in association structure across governance stages. Results show that vegetation greening and ecological function increased substantially over the study period and exhibited relatively consistent spatial patterns. In contrast, well-being enhancement exhibited stronger spatial heterogeneity and more pronounced phase fluctuations. During the later governance stage, coinciding with the implementation of the Shan-Shui initiative, ecological restoration measures became increasingly important for vegetation greening and ecological function, whereas well-being enhancement remained primarily linked to human activity. Network analysis further indicates a shift from a pattern more closely associated with climate in the earlier stage toward a more balanced multi-factor co-regulation pattern. These findings demonstrate the multifaceted role of the Shan-Shui initiative in jointly advancing vegetation greening, ecological function, and well-being enhancement, while ecological gains remained more pronounced and consistent than well-being enhancement during the study period. This study enhances the understanding of multidimensional socio-ecological responses under integrated ecological restoration and provides insights for evaluating and managing restoration efforts.
Lithium, cobalt, molybdenum, graphite and rare earths are strategic resources that support high-tech manufacturing and the global clean energy transition. It is essential to fully understand the trade patterns of these five critical raw materials. Horizontal comparisons help explore their differences in spatiotemporal evolution, organizational structure and impact mechanisms within the global trade landscape. This paper therefore classifies these critical raw materials into three categories, which are energy-oriented materials (lithium and graphite), high-end manufacturing-oriented materials (cobalt and molybdenum) and heterogeneous materials (rare earth elements), based on their core application fields and constructs global trade networks for each category. This study uses multiple network analysis methods and the Temporal Exponential Random Graph Model (TERGM). It deeply investigates the spatiotemporal evolution, organizational structure and associated factors of their trade networks. Results show that the three material categories have temporal mismatches in their development stages and present spatial decoupling risks between the supply side and the demand side. All three trade networks show significant core-periphery structures. Most nodes are at a disadvantage in trade. The core associated forces differ across the three material categories. Technological level and geographic distance exert significant impacts on cobalt and molybdenum trade network. Investment and energy productivity boost the trade development of lithium and graphite. Geopolitical factors serve as important drivers for rare earth trade. Finally, this paper puts forward targeted policy recommendations based on the differences of the three material categories. It provides a scientific basis and decision support for the sustainable development of the five critical raw materials.
China's rapid urbanization has turned the construction sector—including upstream industries, building operations, on-site construction, and demolition—into one of the most carbon-intensive systems globally. Yet, the stark urban-rural disparities in CO2 footprint remain poorly understood, hindering low-carbon development strategies in the sector. This study introduces a comprehensive framework to uncover the spatiotemporal disparities in construction-related CO2 emissions across China, demonstrating how urbanization, energy-mix shifts, industrial changes, and rural revitalization have shaped the country's emission trajectory. The results show that in 2020, construction-related emissions accounted for 57% of China's total anthropogenic CO2 emissions. Urban areas not only account for 89.4% of the total emissions, but also exhibit 4.8 times greater per capita emissions (Ecap) than rural settings. This disproportionate imbalance is largely driven by urban emission reliance on carbon-intensive building materials and energy systems (68%), while rural areas are more dependent on building operation (57%). During 1990–2020, this imbalance drove an 5.6-fold increase in construction-related emissions shaped by population redistribution, construction demand, and supply-chain structures. Post-2015, urban Ecap began to slow down (7.7%) due to industrial upgrades and cleaner energy transition, while rural emissions continued to rise at 7.9%, fueled by expanding construction activities and increased operational energy demand. Our findings underscore the importance of promoting sustainable practices in urban regions and improving efficiency in rural areas to drive the construction sector's low-carbon transition.
To meet food demand, China has reclaimed marginal land for cropland in recent decades while converting cropland to ecological land via ecological retirement and agricultural restructuring, yet the rationality of this dual practice remains unevaluated. This study first analyzed the spatiotemporal patterns of cropland reclamation and ecological loss in China from 1990 to 2024, then employed the XGBoost-SHAP model to identify their underlying drivers. It also used the MaxEnt model to assess cropland suitability and the substitution potential of cropland restoration for reclamation, and quantified effects on grain production and ecosystem services via GAEZ-PCA-RF and InVEST models. Results showed cropland reclamation and ecological loss accounted for 26.25% and 22.46% of the base-period cropland area, respectively, with net reclamation concentrated in Northwest and Northeast China, and the two processes had opposing drivers. Ecologically lost cropland had 1.14 times higher suitability than reclaimed cropland, with the former's suitability rising and the latter's falling. Notably, 50.12% (25.95 Mha) of planned new reclamation could be substituted by cropland restoration, mainly in the Middle and Lower reaches of the Yangtze River drainage basin and South China (90.11%). This substitution would boost national grain production by 3.07% (23.26 Mt, 74.87% from Southern China) and raise the integrated ecosystem service index by 0.65%. These findings suggest that China should place greater emphasis on the importance of cropland restoration in the future to maintain stable cropland area, an approach that would better balance food security and ecological conservation.
Industrial process-related emissions represent a critical and hard-to-abate segment of global greenhouse gas emissions, traditionally managed through supply-side technological interventions. However, achieving deep emission reduction requires a paradigm shift beyond the factory gate. This study conducts an integrated assessment of demand-driven industrial process-related emissions by source for 80 global economies from 1995 to 2022, encompassing both peak pathway identification and the decomposition of demand-side drivers. Results show that global demand-driven industrial process-related emissions in capital formation, final consumption and abroad expenditure increased by 2091.23 Mt CO2-eq, 549.91 Mt CO2-eq and 0.76 Mt CO2-eq. The peaking analysis indicates that only ten economies have peaked in demand-driven industrial process-related emissions. Yet, trade has led to a growing reliance on imported emissions among some of these economies. For cement, lime and metal industries, the emission intensity emerges as the primary driver of emission reductions, contributing averages across 80 economies of 69.88 Mt CO2-eq, 33.34 Mt CO2-eq and 20.31 Mt CO2-eq, respectively. However, for product uses as substitutes for ozone depleting substances, the emission intensity is identified as a major driving factor, contributing an average across 80 economies of 4.82 Mt CO2-eq. These findings call for a unified industrial process-related emission accounting beyond the factory gate, with measures that reduce demand-driven emissions in capital formation and final consumption by increasing green investment and promoting green consumption worldwide.
Freshwater scarcity and decarbonization targets jointly constrain staple-crop provisioning, yet water footprint (WF) and greenhouse-gas (GHG) burdens are often evaluated separately, obscuring coupled pressures and trade-mediated responsibility shifts. We developed an integrated water-carbon sustainability accounting framework for China's five staple crops (wheat, rice, maize, soybean, and potato) using a 2020 baseline. Spatially explicit production, blue/green/gray WF, province-crop GHG intensities, edible-portion calorie denominators, and interprovincial trade flows were linked to estimate irrigated and rainfed burdens under mass- and calorie-based functional units. WF and GHG intensities were generally positively coupled but varied across crops and production systems, reflecting shared irrigation-energy and nitrogen-related pathways. Interprovincial trade reallocated 362.7 Gm(3) of WF and shifted coupled burdens from exporting to importing provinces; consumption-based attribution attenuated co-burden inequality (G-WSI: -0.124 to -0.033; edible-corrected G-GHGIcal: +0.291 to +0.175), but did not remove upstream burden concentration. Edible-corrected calorie-scale coupling was significant for rice (R-2 = 0.407, p < 0.001), soybean (R-2 = 0.304, p < 0.001), and maize (R-2 = 0.112, p = 0.043), but weaker for wheat and potato. Inequality signals remained sensitive to accounting perspective, ordering criterion, functional-unit choice, and spatial aggregation, with regional summaries masking substantial provincial heterogeneity. The findings support equity-aware, scale-sensitive interventions that combine irrigation efficiency, low-carbon pumping, improved nutrient management, methane mitigation, scarcity-informed sourcing, and responsibility-sharing mechanisms across staple supply chains.
The large-scale resource utilization of coal gangue is urgently needed. This study proposes using coal gangue as an amendment for saline-alkali soil improvement, coupled with in-situ layered double hydroxide (LDH) formation. Soil incubation, laboratory-simulated plant growth experiments and leaching tests were conducted to assess the feasibility and efficacy of this approach. The nutrient and pyrite-rich characteristics of coal gangue provide nutrients by chemical weathering and reduce soil alkalinity by the oxidation of pyrite, respectively. The physical incorporation of coal gangue increases soil organic carbon from 7.10 to 51.48 g/kg. Chemical weathering of coal gangue via super-stable mineralization, enhanced by CaO and Fe(NO3)3 & sdot;9H2O as mineralizers, releases available nutrients, enhances the bioavailability of water-soluble organic carbon, and elevates available potassium (from 290.44 to 350.78 mg/kg). The OH-, CO32-, and SO42-ions are immobilized via in-situ formation of layered double hydroxides (LDHs), lowering soil salinization. Meanwhile, super-stable mineralization by LDHs reduces the heavy metal content and mobility, mitigating pollution. Maize (Zea mays L.) cultivation experiments and field trials verify the above results and demonstrate that the strategic application of coal gangue and mineralizers can effectively ameliorate saline-alkali soils, providing a safe and sustainable pathway for the largescale resource utilization of coal gangue.
Dissolved organic carbon (DOC) fuels soil micro-food webs, yet how its chemical heterogeneity mediates the multi-trophic regulation of microbial necromass carbon (MNC) remains a major scientific challenge. Specifically, it is unclear how DOC quality links bottom-up resource supply with top-down predatory control to regulate MNC accumulation. Here, we leveraged a natural temperature gradient across Mollisol regions of Northeast China to generate a diverse range of DOC profiles and microbial communities. We found that increasing mean annual temperature (MAT) significantly reduced DOC, soil organic carbon (SOC), and both bacterial and fungal necromass carbon (BNC and FNC). The BNC and FNC contents increased with DOC, particularly with humic- and protein-like fractions. The qPCR analysis revealed that the abundances of lower-trophic groups (bacteria and fungi) were positively correlated with these DOC components, while higher-trophic groups (protists and nematodes) showed a negative correlation. These shifts simplified and destabilized micro-food webs by increasing intra-trophic associations and decreasing cross-trophic associations, ultimately contributing to BNC and FNC accumulation. Collectively, this study proposes a conceptual framework where higher DOC content, particularly humic- and protein-like fractions, stimulates lower-trophic growth, reduces cross-trophic interactions, and enhances MNC formation. By integrating resource heterogeneity with multi-trophic associations, this study provides a mechanistic explanation for how soil food webs regulate carbon stabilization, suggesting that strategic management of DOC quality and trophic network structure can enhance soil carbon sequestration under sustainable management.
