Objective The processes and mechanisms by which ant nesting influences soil CO₂ emission dynamics in the montane slope forests adjacent to the Napahai Wetland remain poorly understood. This study aims to elucidate the key pathways and driving mechanisms through which ant nesting regulates soil CO₂ fluxes in these forests.

Method We selected Formica manchu in the spruce–fir forests of the Napahai montane slope and established two treatments: ant nest soil and non-nest soil. Soil CO₂ fluxes and bacterial community characteristics were measured using a Li-6400-09 soil respiration chamber and high-throughput sequencing during both dry and wet seasons. Correlation analysis and structural equation modeling were used to identify key drivers and regulatory pathways of soil CO₂ emissions in ant nest soils.

Result (1) Ant nesting significantly promoted soil CO₂ emissions (p < 0.05), with clear dry–wet seasonal variations. Compared to non-nest soils, ant nest soils showed a 37.0% increase in CO₂ flux during the dry season and a 22.0% increase during the wet season. (2) Ant nesting significantly increased the relative abundance of Proteobacteria (annual increase of 58.8%–96.3%) and significantly raised soil organic matter, total nitrogen, available phosphorus, available potassium, ammonium nitrogen, nitrate nitrogen, and pH (annual increase of 9.0%–635.0%). Conversely, ant nesting significantly reduced the relative abundance of Acidobacteria (annual decrease of 12.9%–29.6%) and significantly decreased the bacterial ACE and Chao1 indices, soil bulk density, and water content (annual decrease of 34.7%–42.8%). (3) During the dry season, CO₂ flux in ant nest soils was significantly positively correlated with soil nutrients such as carbon and nitrogen and with the relative abundance of Proteobacteria, and significantly negatively correlated with soil bulk density and water content. During the wet season CO₂ flux in ant nest soils was significantly positively correlated with soil temperature, carbon and nitrogen nutrients, and the relative abundance of Proteobacteria, and significantly negatively correlated with soil bulk density, water content, bacterial ACE and Chao1 indices, and the relative abundances of Acidobacteria and Actinobacteria. (4) Structural equation modeling revealed that the effect of ant nesting on soil CO₂ flux differed significantly between dry and wet seasons. During the dry season, ant nesting mainly influenced CO₂ emissions by regulating soil bulk density, nitrogen nutrients, and Proteobacteria. The relative abundance of Proteobacteria had a significant positive direct effect on CO₂ emissions, whereas soil water content, temperature, and nitrogen nutrients indirectly affected emissions by influencing Proteobacteria. During the wet season, the regulatory pathway became more complex: bacterial α diversity exerted a significant negative direct effect on CO₂ emissions, while dominant bacterial phyla (Proteobacteria and Acidobacteria) showed positive direct effects. Meanwhile, soil bulk density, temperature, and nitrogen and phosphorus nutrients indirectly regulated CO₂ emissions by influencing the bacterial community.

Conclusion  Ant nesting significantly promotes soil CO₂ emissions in the montane slope forests adjacent to Napahai Wetland, with a greater increase in the dry season (37.0%) than in the wet season (22.0%). The underlying pathways differ markedly between seasons: the dry season is dominated by a positive direct effect of Proteobacteria, whereas the wet season features positive direct effects from both Proteobacteria and Acidobacteria, accompanied by negative direct regulation by bacterial α diversity. Soil physicochemical factors primarily exert indirect effects by modulating the bacterial community.