Optimizing public investment in urban green infrastructure under water scarcity is a core challenge in resource economics. This study addresses the critical knowledge gap in quantifying the economic returns on physiological adaptations of urban trees, which are central to their value as natural capital. We integrate dual water isotopes (δ²H, δ¹⁸O) and leaf carbon isotope (δ¹³C) analysis to mechanistically decode the water-use strategy of Machilus yunnanensis in drought-prone Kunming, China. Results show strategic seasonal plasticity: a shift from shallow soil water (10–50 cm) in the wet season to deeper soil sources (50–90 cm) and stem reserves in the dry season, coupled with a dynamic, diurnally variable water-use efficiency (WUE). We then construct a transparent economic valuation model translating these traits into three quantifiable benefit streams: (1) Operational cost savings (EV₁) from reduced irrigation demand; (2) Enhanced marginal productivity of water (EV₂) in ecosystem service generation; and (3) Climate resilience value (EV₃) via mitigated mortality risk. Our “Water-Carbon-Economy” nexus framework provides a generalizable methodology for assessing the cost-effectiveness and risk-adjusted returns of urban forest species. It demonstrates that selecting trees based on such eco-efficient traits is not merely an ecological choice but a sound economic investment, offering direct implications for budget-constrained municipalities seeking to maximize the benefits of green infrastructure under climate uncertainty.