Depth‐Dependent Soil C:N:P Stoichiometry in <i>Robinia pseudoacacia</i> Mixed Versus Pure Stands and Its Drivers in the Chinese Loess Hilly Region

ABSTRACT Soil carbon (C): nitrogen (N): phosphorus (P) stoichiometry, which reflects nutrient availability, responds species‐specifically to mixed‐species afforestation. However, how mixed species affect soil C:N:P stoichiometry with depth remains poorly understood. This knowledge gap is critical in the Loess Hilly Region, where understanding depth‐dependent nutrient dynamics is essential for ecological restoration and sustainable management. Here, we investigated C:N:P dynamics across soil depths (0–20, 20–40, and 40–60 cm) and associated influencing factors (i.e., afforestation measures and soil physicochemical properties) in mixed stands ( Robinia pseudoacacia/Prunus davidiana ; R. pse/Prunus sibirica ), pure Robinia ( R. pse ), and Prunus ( P. dav ; P. sib ). The results showed that stand type significantly influenced the vertical migration of N and P elements in soil profiles ( p &lt; 0.05). Compared to pure Robinia, mixed stands ( R. pse/P. dav ) enhanced N cycling, while mixtures with P. sib demonstrated N complementarity. In contrast, P . dav and P . sib mixtures promoted P cycling due to contrasting soil organic matter (SOM) decomposition effects between R. pse (inhibitory) and the other two species (enhancing). Soil organic C and total N (TN) across soil depths primarily derived from SOM, whereas soil total P originated predominantly from mineral sources. The nodule bacteria of R. pse enhanced TN concentrations, whereas mixed planting accelerated P release from SOM and minerals. Soil C:N:P stoichiometry at 20–40 cm depth was more influenced by the proportion of clay, associated with improved microenvironment, for example, soil water content. These findings demonstrate that mixing R. pse and P . dav enhances soil nutrient cycles through soil physical‐dominated properties, which provides new insights for optimizing mixed afforestation strategies in arid land ecosystems.