Biomass is a direct reflection of community productivity, and the allocation of aboveground and belowground biomass is a survival strategy formed by the long-term adaptation of plants to environmental changes. However, under global changes, the patterns of aboveground–belowground biomass allocations and their controlling factors in different types of grasslands are still unclear. Based on the biomass data of 182 grasslands, including 17 alpine meadows (AMs) and 21 desert steppes (DSs), this study investigates the spatial distribution of the belowground biomass allocation proportion (BGBP) in different types of grasslands and their main controlling factors. The research results show that the BGBP of AMs is significantly higher than that of DSs (p < 0.05). The BGBP of AMs significantly decreases with increasing mean annual temperature (MAT) and mean annual precipitation (MAP) (p < 0.05), while it significantly increases with increasing soil nitrogen content (N), soil phosphorus content (P), and soil pH (p < 0.05). The BGBP of DSs significantly decreases with increasing MAP (p < 0.05), while it significantly increases with increasing soil phosphorus content (P) and soil pH (p < 0.05). The random forest model indicates that soil pH is the most important factor affecting the BGBP of both AMs and DSs. Climate-related factors were identified as key drivers shaping the spatial distribution patterns of BGBP by exerting an influence on soil nutrient availability. Climate and soil factors exert influences not only on grassland biomass allocation directly, but also indirectly by impacting the availability of soil nutrients.

Specific leaf area (SLA) and leaf dry matter content (LDMC) are key leaf functional traits commonly used to reflect tree resource utilization strategies and predict forest ecosystem responses to environmental changes. Previous research on tree resource utilization strategies (SLA and LDMC) primarily focused on the species level within limited spatial scales, making it crucial to quantify the spatial variability and driving factors of these strategies. Whether there are discrepancies in resource utilization strategies between trees in planted and natural forests, and the dominant factors and mechanisms influencing them, remain unclear. This study, based on field surveys and the literature from 2008 to 2020 covering 263 planted and 434 natural forests in China, using generalized additive models (GAMs) and structural equation models (SEMs), analyzes the spatial differences and dominant factors in tree resource utilization strategies between planted and natural forests. The results show that the SLA of planted forests is significantly higher than that of natural forests (p < 0.01), and LDMC is significantly lower (p < 0.0001), indicating a “faster investment–return” resource utilization strategy. As the mean annual high temperature (MAHT) and mean annual precipitation (MAP) steadily rise, trees have adapted their resource utilization strategies, transitioning from a “conservative” survival tactic to a “rapid investment–return” model. Compared to natural forests, planted forest trees exhibit stronger environmental plasticity and greater variability with forest age in their resource utilization strategies. Overall, forest age is the dominant factor influencing resource utilization strategies in both planted and natural forests, having a far greater direct impact than climatic factors (temperature, precipitation, and sunlight) and soil nutrient factors. Additionally, as forest age increases, both planted and natural forests show an increase in SLA and a decrease in LDMC, indicating a gradual shift towards more efficient resource utilization strategies.