Abstract: Slope aspect (topographic factor) and formation stage (temporal factor) are critical determinants of vegetation development and succession on engineered slopes. Understanding how these variables influence population formation rates and characteristics of Albizia julibrissin provides a scientific foundation for vegetation restoration on exposed slopes in purple soil hilly regions. This study investigated woody A. julibrissin populations established on three aspects (EN40°, WS33°, and WN29°) of the same mountain using a "tree + grass" ecological slope protection model. Based on establishment age, four successional stages were delineated: foundation and introduction (Stage I, 1–3 years), adaptation and colonization (Stage II, 4–6 years), expansion and establishment (Stage III, 7–9 years), and stability and evolution (Stage IV, 10–12 years). Population characteristics—including density, age structure, population dynamics, and spatial distribution—along with individual morphological indicators (plant height, crown width, diameter at breast height, DBH) were measured across aspects and stages to assess the effects of slope aspect on population formation rate and stable-stage morphological traits. The results revealed that: 1) Significant differences existed across the four successional stages in population height, coverage, birth and mortality rates, immigration and emigration rates, age structure, and distribution coefficients (P < 0.05). During Stages I and II, the WS population exhibited lower birth rates but higher mortality rates than the WN population. In Stage III, population density and height on the WS slope were lower than on the WN slope, whereas the spatial aggregation of young trees was higher on WS than on WN and EN slopes (P < 0.05). In Stage IV, projected coverage and basal coverage of the WS population were lower than those of the WN population, while the number and proportion of annual and biennial plants were greater on WS than on WN and EN slopes. Population formation rate followed the order: WN slope > EN slope > WS slope. 2) During the stability and evolution stage (Stage IV), significant differences were detected in the density and proportion of A. julibrissin individuals across four height classes (P < 0.05). The WN population was dominated by individuals >200 cm in height, whereas WS and EN populations were dominated by individuals of 100–200 cm in height. Notably, the WS population had the highest proportion of individuals <50 cm in height. For individuals of equivalent height, significant among-aspect differences were found in crown height, crown width, and the numbers of bipinnately compound leaves, compound leaves, and leaflets (P < 0.05), resulting in aspect-dependent morphological differentiation at both population and individual levels. 3) In the stability and evolution stage, the WN population exhibited significantly greater plant height, leaf number, crown width, and leaf area index than EN and WS populations (P < 0.05), yet its basal coverage was lower than that of the WS population. Additionally, individuals >200 cm in height on the WN slope had larger DBH but smaller ground diameter compared to their WS counterparts, indicating lower ecological protection potential and higher susceptibility to crown breakage and lodging. In conclusion, the "tree + grass" model represents an effective configuration for vegetation reconstruction on exposed slopes. Populations of A. julibrissin and associated protective communities matching the slope habitat conditions successfully established on all three aspects. Although A. julibrissin populations on the WN slope exhibited faster formation rates, their community basal coverage and individual stability were inferior to those on the WS slope. These findings underscore the necessity of continuous and systematic evaluation of vegetation reconstruction efficacy for ecological governance, accounting for topographic factors and spatiotemporal dynamics.