Soil erosion has caused the loss of black soil and exposed the soil parent material in the cultivated layer of sloping farmland in Northeast China. Straw return (STR) and manure fertilization (MF) are critical measures to improve soil quality and crop yield. However, the effect of STR and MF on the soil properties of the parent material remains unclear. We conducted a 1-year pot experiment in the field using the soil parent material of degraded black soil to evaluate the effects of STR and MF on soil nutrients, microbial community, and soybean yield. We analyzed these effects using two treatments (STR and MF) in three soybean growth stages (seedling, flowering, and maturity) and a control group (CK). The MF treatment had higher α and β diversity of soil microbial than the CK during all soybean growth stages. Similarly, STR had higher soil microbial α diversity at the maturity stage and lower diversity at the seedling stage. Co-occurrence network analysis suggested that STR and MF increased the proportion of positively correlated edges in soil bacterial and fungal networks compared to the CK. Notably, the treatments enriched beneficial taxa, such as Schizothecium (fungi) and Massilia (bacteria), which are associated with organic matter decomposition and nitrogen cycling. STR and MF significantly improved soil organic matter, total nitrogen, and carbon-nitrogen ratio (p < 0.05). Structural equation modeling (SEM) revealed that STR and MF directly increased soybean yield. This effect was primarily mediated by the significantly higher soil organic matter, total carbon, total nitrogen, and carbon-to-nitrogen ratio in the treatments than in the CK (p < 0.05). In summary, STR and MF improved soil fertility and soil microbial community diversity of degraded black soil. This study provides scientific methods to improve the fertility of degraded black soil and increase soybean production in the short term.

Soil microorganisms are essential for maintaining the function and health of agricultural ecosystems. However, the responses of microbial communities to long-term changes in land use have been insufficiently explored. Hence, based on a 15 years of field experiments in the northeast Mollisol region of China, we applied the Illumina high-throughput sequencing technology to study the effects of different land use types, including conventional tillage (CT), bare land (BL), no tillage (NT), natural vegetation restoration (NVR), and afforestation (AF), on bacterial communities along the soil profile (0–5 cm, 5–10 cm, 10–20 cm, and 20–30 cm) and co-occurrence networks and identified their relationships with soil physicochemical properties. The findings indicated that the land use type as well as soil depth affected the diversity and structure of bacterial communities significantly. There was no marked difference in the diversity of bacterial communities between CT and NT at different soil depths, except for a depth of 20–30 cm. In NT, NVR, and AF, the relative abundance of Actinomycetota and Firmicutes was higher than that in CT. Conversely, CT showed a remarkably higher abundance of Proteobacteria and Acidobacteriota than BL, NT, NVR, and AF. Compared with CT and BL, increased stability and complexity of the community co-occurrence networks was identified for NT, NVR, and AF. Additionally, the diversity and composition of bacterial communities were correlated with the soil’s total nitrogen (TN), pH as well as total organic carbon (TOC). Our study revealed the potential mechanism by which long-term land use changes affected the distribution of soil bacterial communities, which was of high importance for sustainable development of agriculture and optimal management of land resources.

Distinct paradigms, such as the “enzymic latch” and “iron gate” theories, have been proposed to elucidate SOC loss or accumulation, but their relative significance and whether they are mutually exclusive in permafrost peatlands remain unclear. To address this, we evaluated their relative importance and identified the dominant factors controlling SOC stability. Therefore, we employed a space-for-time substitution approach across a permafrost gradient (continuous, discontinuous, and isolated) by systematically quantifying extracellular enzyme activities, iron (Fe) phases, and iron-bound soil organic carbon (Fe-SOC) at various depths (0–10, 10–30, and 30–50 cm) in peatlands. Our results did not support the “enzymic latch” theory, with hydrolytic enzyme activities (β-glucosidase (BG), cellobiohydrolase (CBH), and β-N-acetylglucosaminidase (NAG)) showing positive correlations with phenolics but negative correlations with phenol oxidase (PHO) activity. However, ferrous iron (Fe(II)) was significantly positively correlated with PHO activity, and ferric iron (Fe(III)) stabilized SOC through co-precipitation with it to form Fe-SOC, supporting the “iron gate” theory. Moreover, Fe-SOC decreased from the continuous to the isolated permafrost zone, and with soil depth from 0–10 cm to 30–50 cm. Partial least squares path modeling (PLS-PM) analysis indicated that Fe(III) directly and indirectly (via Fe-SOC and phenolics) affected SOC. Our study demonstrated the primacy of the “iron gate” mechanism in controlling carbon stability in the Great Hing’an Mountains permafrost peatlands, providing new insights for projecting carbon-climate feedback.

Understanding of the efficacy of short-term strip tillage (ST) is essential for its adoption in Northeast China. A two-year field experiment (2023–2024) with soybean–maize rotation was conducted using a randomized complete block design to explore the effects of short-term ST on soil physicochemical properties and crop yields compared with no-till (NT) and conventional tillage (CT). Soil samples in ST were collected from the seedbed (tilled without straw mulching, ST-IS) and between the seedbed (no-till with straw mulch, ST-BS), respectively. Results showed that in the 0–10 cm layer, soil temperature in ST-IS was 1.61–1.65 °C higher than NT, and soil moisture in ST-BS was 4.20–8.52% higher than CT. ST-IS had lower bulk density and penetration resistance than NT. Meanwhile, aggregate stability, saturated water content, and soil nutrients were greater under ST and NT than those under CT in the 0–5 cm layer. Moreover, maize yield was significantly higher under CT compared to NT, while ST maintained intermediate yields. In contrast, NT achieved the highest soybean yield. Furthermore, structural equation modeling (SEM) showed short-term tillage primarily affected crop yield by altering soil temperature and structure (not direct or nutrient-mediated effects), with a more pronounced impact on maize than soybean. Notably, the total standardized effects of soil temperature, moisture, and structure are completely opposite between soybean and maize. In conclusion, ST is an appropriate tillage practice for maize cultivation, while NT is more suitable for soybean cultivation in Northeast China.