Abstract: Least cost theory (LCT) has been proposed to understand the functional traits of leaves, which assumes an optimal balance of investments in water flow and photosynthetic capacity for plants, and has received strong support at global or regional scales. However, relatively little is known about nutrient and water resource acquisition and utilization of plant species in various grassland types in arid and semi-arid regions. In this study, we focused on 71 herbaceous and shrub species in foreststeppe, typical steppe, and desert steppe that occur along the natural rainfall gradient from east to west in Gansu Province. We examined how climate and soil variables modulated leaf functional traits associated with acquisition and utilization of carbon and water resources and whether there existed trait interrelationships across different grassland types in temperate regions under the framework of LCT. The results showed that leaf mass per area (LMA) and nitrogen concentration per unit area (N_area) increased with increasing of photosynthetically active radiation (PAR), and decreased with increasing mean annual precipitation, mean annual minimum temperature, soil clay content, soil water content, and soil organic carbon content. An opposite pattern was observed for the ratio of intercellular CO₂ to air CO₂ concentration (c_i∶c_a). Across the three grassland types, N_area increased as LMA increased and decreased inversely to c_i∶c_a. There was no significant difference in the linear fitted slopes of these traits from the different grasslands. Multivariate linear fitting of leaf N_area with environmental gradient and structural factors showed that LMA and c_i∶c_a had the greatest influence on the variation of N_area, followed by precipitation and soil moisture. The PAR and mean annual minimum temperature also exerted a significant influence on N_area. Our results will contribute to understanding the optimal balance of investments in water and carbon resources for photosynthesis at the regional scale by combining important soil factors with key leaf functional traits in the study region, providing scientific guidance for successful ecosystem management under future climates to ensure ecological security in the region.

Key words: leaf functional trait, least-cost theory, regional scale, climate gradient, soil attribute.