Altura dominante como impulsionador estrutural dos perfis de diversidade efetiva na Amazônia brasileira

Abstract

Understanding how forest structure shapes biodiversity remains a central challenge in tropical ecology, yet predictive frameworks linking continuous diversity profiles to stand-level attributes are limited. We present the Structural Hill Diversity Model (SHDM), a novel nonlinear approach predicting effective diversity (Hill numbers) from dominant tree height in Brazilian Amazon forests. Using National Forest Inventory data from three Amazonian states for parameterization (559 clusters; 133,121 stems) and independent validation across six additional states (455 clusters; 102,005 stems), we evaluated hierarchical formulations with increasing structural complexity. The SHDM integrates two ecological processes: asymptotic diversity saturation with canopy complexity (179 effective species per 0.8 ha) and height-driven modulation of diversity decay along the Hill continuum (= 0.42). Taller forests maintain higher diversity and exhibit slower decline under increasing ecological dominance, resulting in greater evenness and structural stability. Among diversity orders, the Simpson index (q = 2) showed the most consistent predictive performance. External validation confirmed strong generalizability across heterogeneous Amazonian forest types. Dominant tree height emerged as a robust structural driver of effective biodiversity patterns, supporting its application in remote sensing–based assessments. The SHDM provides a parsimonious framework unifying structural and compositional diversity for scalable biodiversity monitoring and conservation.