To enhance phenotypic plasticity, it is vital to maximize the genetic growth potential of trees and understand their adaptive responses to environmental conditions. Tree species adapt to dynamic environmental conditions by leveraging the interactions among the environment, genotype, and genotype-by-environment. A total of 25 improved varieties of Chinese fir were transplanted and developed through multi-generational breeding into four types of artificial forest soils. Through a quantitative analysis of genotypic, soil environmental conditions, and genotype-by-environment interaction effects on variations in growth, biomass, and root functional traits, the key drivers of phenotypic plasticity were identified. The results indicate that soil environmental conditions and genotype-by-environment interactions are the primary factors influencing trait variation, explaining 55.89% to 93.94% of the observed variation, while the family effect is relatively minor. Notably, pronounced phenotypic plasticity drives divergent selection in both aboveground and belowground growth strategies. Critical traits influencing root dry weight include root average diameter, total root volume, and root-to-shoot ratio. Although root dry weight does not directly affect plant height, it has a substantial impact on aboveground dry weight. These findings highlight that the changes in the aboveground and belowground growth strategies of Chinese fir during the seedling stage are closely linked to the plasticity of root functional traits. For multi-generational genetically improved varieties, this study examined how the integration of genetic effects, soil environmental conditions, and genotype-by-environment interactions in the selection of aboveground growth and root functional traits influences the mechanisms driving biomass accumulation. The results provide actionable insights for selecting soil-specific families in subtropical plantations.

This study aimed to reveal the rhizosphere microbial community structure, carbon–nitrogen–phosphorus (C-N-P) nutrient cycling processes, and functional gene characteristics of Pinus massoniana and Schima superba in mixed forests. Furthermore, we sought to elucidate the microbial mechanisms by which mixed-species afforestation enhances soil quality improvement, providing a theoretical basis in soil microbiology for the cultivation of these mixed forests. The research subjects included pure P. massoniana plantations (CLPs), pure S. superba plantations (CLSs), and individual P. massoniana (HJP) and S. superba (HJS) trees within mixed plantations (HJLs). We collected rhizosphere and bulk soil samples to analyze their physicochemical properties and enzyme activities. Metagenomic sequencing was employed to profile the rhizosphere microbial communities and functional genes involved in C-N-P cycling. Furthermore, by integrating a functional gene co-occurrence network analysis with structural equation modeling (SEM), we systematically elucidated the coupling relationships among the stand types, soil properties, microbial communities, and nutrient cycling. Mixed planting significantly improved soil quality; compared to the CLP and CLS forests, the nitrate nitrogen (NO₃⁻-N) content in the mixed forest soils increased by 121.01% and 120.10% (p < 0.05), and the activity of urease (URE) also significantly increased by 123.99% and 49.56%, respectively. Mixing significantly altered the microbial community structure. In the bacterial community of the mixed forests, the abundance of nitrogen-fixing and potentially phosphorus-solubilizing bacteria from the genera Paraburkholderia and Burkholderia increased. In the fungal community, the arbuscular mycorrhizal fungus Rhizophagus, which possesses a nutrient absorption advantage, exhibited absolute dominance, with its relative abundance ranging from 14.84% to 88.81%. The abundances of genes associated with denitrification and phosphorus starvation regulation were significantly upregulated in the mixed forests; notably, the abundance of phosphorus starvation regulation genes in the HJSs was 18.84% higher than that in the CLSs. A co-occurrence network analysis demonstrated that the proportion of positive correlation edges in the HJP nitrogen cycling network reached as high as 75.0%, and the average degree of the HJS phosphorus cycling network (2.691) surpassed that of the CLSs. The structural equation modeling further revealed that the association strength between the fungi and phosphorus cycling genes in the mixed forests increased to R² = 0.915 (p < 0.01) from R² = 0.213 in the pure forests. This mixed planting practice transforms nutrient cycling from a resource-competitive mode to a microbially synergized mode, thereby forming an efficient endogenous nutrient cycling system. This synergistic rhizosphere microbial effect is a key internal mechanism for overcoming nutrient bottlenecks and should serve as a diagnostic indicator of soil recovery in the ecological restoration of degraded pine forests.