Spatial and age-dependent tree-ring growth responses of Larix gmelinii to climate in northeastern China

Tree-ring width chronologies from 276 Larix gmelinii cores taken in northeastern China were used to analyze spatial and age-dependent growth–climate response relationships. Tree radial growth from five localities showed similar patterns, while exhibiting different tree-ring growth responses to local climate. The rotated principal component analysis (RPCA) indicated that tree age, growing season moisture conditions, and ambient air temperature variations resulted from location differences (e.g., longitude, latitude, and altitude), which could explain the non-stationary spatial climate–growth relations observed. The study tested the fundamental assumption that the climate–growth of L. gmelinii was age independent after the removal of size trends and disturbance signals. The age-related climate–growth relationship might potentially improve the veracity of past climate reconstructions. Bootstrapped correlation function analyses suggested that the response of L. gmelinii radial growth to climate differed between trees ≥150 years old and <150 years old. Mean sensitivity and standard deviation for trees increased with age in the <150 years old tree class; whereas trees ≥150 years old had no significant relationship with age. These results showed that the assumption of age-independent climate–growth relationship is invalid at these sites. Physiological processes and/or hydraulic constraints dependent on tree age, together with detrending techniques could be the possible causal factors of clear age-dependent responses. These results suggested the importance of incorporating trees of all ages into the chronology to recover a detailed climatic signal in a reconstruction of L. gmelinii for this region.