Abstract
Under the dual pressures of urbanization and climate change, vertical greening plays a crucial role in compensating for limited urban green space and in enhancing both landscape quality and ecological functions. To improve plant selection strategies for vertical greening, this study quantified the photosynthetic adaptability of ten green wall species across light gradients on all four building facades. Over three years of in situ monitoring, combined with analyses of photosynthetic parameter variation (P(max), LCP, and LSP) and biomass, we evaluated the physiological and ecological response mechanisms of plants with different growth forms at multiple scales ("trait-species-community"). The main findings are as follows: (1) Photosynthetically active radiation intensity had the order of south wall > east wall > west wall > north wall, with differences of 3-5 fold. (2) Cluster analysis of photosynthetic traits revealed a sensitivity hierarchy: monocotyledonous herbs > dicotyledonous herbs and vines > woody species. Woody plants such as Ligustrum sinense, Ligustrum japonicum, and Rhododendron spp. showed significantly lower variability in photosynthetic parameters (45.8-64.5%) than herbaceous species, thereby maintaining strong stability under light gradients. Muehlenbeckia complexa (Mc) adapted to intense south-facing light, while Acorus gramineus 'Ogon' (Ag) preferred the weaker light of north-facing walls. In contrast, Farfugium japonicum (Fj), Carex oshimensis (Co), Trachelospermum jasminoides (Tj), and Vinca major (Vm) displayed substantial physiological fluctuations. (3) Based on three years of monitoring data, we developed a quantitative model of light adaptation driven by the coefficient of variation (CV) of photosynthetic parameters. Together with PCA-based clustering, we proposed a "growth form-orientation" synergistic configuration framework. Woody plants with high photosynthetic stability are recommended as the structural backbone for cost-efficient green walls, supplemented by vine/herbaceous species selected according to wall orientation. This study not only provides a scientific basis for accurate plant selection and low-maintenance design of green walls but also provides technical strategies for integrating multifunctional green infrastructure with low-carbon urban development. Furthermore, the proposed approach can be standardized as a Nature-Based Solutions (NbS) pathway for widespread application to building facades in high-density cities worldwide.