Neighborhood competition modulates the link between crown structure and tree ring variability in monospecific and mixed forest stands

3D crown shape and tree ring development are autobiographies of the growth conditions. With advancements in terrestrial laser scanning, modeling growth from 3D crown structure offers insights into trees’ structural and functional responses in a non-destructive and repetitive way. However, how the trees in different growing conditions respond in 3D structure and manifest their responses into tree rings remains unexplored, which would help to better explain tree-level growth dynamics and forest management. To enhance this understanding, we tested a set of hypotheses: (HI) that 3D crown shape (CS) and tree ring (width) patterns (TRP) are correlated across different growing conditions or forest stands like monospecific, provenances trials, and mixed forest stands; (HII) that stand types influence the CS-TRP link; and (HⅢ) local neighborhood competition (LNC) modulates the CS-TRP link. (HⅣ) 3D crowns manifest local growth conditions; therefore, 3D crown structures can be used to predict tree ring growth. We assessed these hypotheses by employing terrestrial laser scanning-based 3D crown shape and dendrochronology-based tree ring width patterns from Norway spruce (Picea abies [L.] Karst.) trees growing in pure spruce (unthinned and thinned), provenances trial, and mixed-species trial (with European Beech, Fagus sylvatica [L.]) stands covering a large-scale competition gradient. We first show that 3D crown shape and TRP metrics differed significantly across forests (p<0.05) but were correlated (p<0.05). Neighborhood competition among the forest types influences the link between 3D crowns and tree rings. Pathway-based analyses revealed that neighborhood competition indirectly influences ring variability by modifying crown structure (p<0.05), suggesting local growth conditions are mostly manifested into crown shapes, leading to 3D crown shape-based low-error growth predictions (0.44 mm) across forest types. However, incorporating competition legacy information (competition over the last 30 years) in the model slightly improved the prediction performance (error reduced to 0.41 mm), further explaining that the crown likely loses growth information due to growing conditions (competition it faces and species with it is growing). This study reveals how trees in different growing conditions differed structurally and mechanized their responses in tree rings, providing crucial insights into tree-level growth dynamics and management.