Abstract
Community biomass allocation is jointly determined by habitat conditions and plant functional traits. Studies of biomass allocation patterns in topographic-soil climax communities of karst ecosystems remain scarce. According to the trait-driven paradigm, topographic gradients and soil properties indirectly influence karst forest biomass, via their control over community-level functional structure. In the 25-ha Maolan Dynamic Plot of the Karst Forest Ecosystem in South China, we compiled 1255 high-quality trait records for six key plant functional traits related to biomass from 48 dominant species, individual biomass data for 12,354 stems, and fine-scale environmental variables. Partial least-squares structural equation modeling (PLS-SEM) was used to quantify the direct and indirect factors affecting biomass allocation in this climax karst forest community. We observed that the trade-offs in biomass among different forest layers were more effective in predicting the biomass status of natural communities (R (2) = 0.69). Topographic heterogeneity acted as an environmental filter, driving the assembly of distinct karst climax communities. Community-level trait distributions and abiotic variables significantly influenced both community biomass and its trade-offs, although trait patterns explained biomass trade-offs more effectively than environmental factors. PLS-SEM identified slope position as the primary driver of biomass trade-offs in the karst climax communities, with community-level variation in specific leaf area (SLA) mediating biomass allocation. Slope position decline reduced the community-weighted mean of functional traits (SLA, Wood density, Leaf nitrogen content) and concurrently increased biomass of the stable layer. In parallel, lower community-weighted variance of traits (SLA) attenuated biomass loss in the regeneration layer. These results underscore the pivotal role of trait composition in mediating biomass partitioning at the community scale.