Abstract: The trend of artificial forest degradation in the southern regions of our country is intensifying, and targeted transformation is key to enhancing the productivity of forest land. Exploring the vertical root distribution characteristics and belowground competitive relationships of important afforestation tree species in the southern hilly mountainous region provides a reference basis for the restoration and reconstruction of degraded artificial forest vegetation. This study focused on six typical mixed afforestation species in southwest Zhejiang: Magnolia grandiflora, Michelia figo, Liquidambar formosana, Chamaecyparis obtusa, Cunninghamia lanceolata, and Pinus tabulaeformis. The vertical distribution patterns of root length density, root surface area density, and root biomass density of each tree species were systematically analyzed. The Levins indices, Shannon indices, and Pianka indices were used to quantify belowground competitive relationships among these species. Results indicated that M. grandiflora and C. obtusa showed a high proportion of root length and root surface area density distribution in the topsoil layer (0-40 cm), accounting for over 50%. L. formosana and P. tabulaeformis exhibited a higher proportion in the deeper soil layers below 60 cm, while the distribution of M. figo and C. lanceolata was relatively uniform. The root biomass density of each species was primarily concentrated in the topsoil layer (0-40 cm), accounting for 53.63-84.74% of the total soil depth. P. tabulaeformis and C. obtusa had high total root biomass densities of 73.90 and 55.89 mg/cm³, respectively, while M. figo and C. lanceolata had much lower densities of only 13.54 and 18.54 mg/cm³. In root niche analysis, M. grandiflora was the most competitive species with Levins index and Shannon index reaching 3.89 and 1.48, respectively, while L. formosana demonstrated the weakest competitiveness (Levins index = 2.13, Shannon index = 0.97). Low competition intensities were observed between M. figo and L. formosana, M. figo and C. lanceolata, and M. figo and P. tabulaeformis, with Pianka niche overlaps indices of 0.37, 0.34, and 0.35, respectively. Moreover, the high biomass density of P. tabulaeformis, along with the complementary root system effects with M. figo, leads to greater stability for ecosystems formed by M. figo × P. tabulaeformis. Therefore, mixed planting models of M. figo × P. tabulaeformis are recommended for afforestation in the research area.