Abstract: Branch length and branch diameter are important characteristics that determine wood quality and yield. Development of static branch length and diameter models by incorporating individual tree variables, site quality and competition have been widely studied, while the climate effect has rarely been reported. In this study, mixed-effects climate-sensitive branch length and diameter models were developed based on 228 sample trees of Larix kaempferi from three latitude regions in China (approximate 42°N in Liaoning Province, 33°N in Gansu Province, and 30°N in Hubei Province). Results revealed that diameter at breast height, and crown ratio, sum of the basal areas of trees larger than the subject trees, dominant tree height, mean warmest month temperature, and summer precipitation substantially improved branch length model. Diameter at breast height, and crown ratio, ratio of the sum of DBH in sample plot to the subject tree, dominant tree height, mean warmest month temperature, and spring precipitation significantly improved branch diameter model. Compared with base model, mean square error reduction of mixed-effects branch length and diameter models were 32.9% and 44.1%, respectively. The relative contributions of covariates to branch length model were tree size (59.1%), site quality (25.7%), competition (13.5%), and climate (1.7%), and branch diameter model were tree size (57.0%), competition (21.9%), site quality (18.3%), and climate (2.8%). Relative contributions of covariates on branch length and diameter models from different latitude regions were different. Effects of competition on branch length model in Liaoning and Hubei were larger than climate, whereas climate in Gansu was larger than competition. As for branch diameter model, competition in Liaoning was larger than site quality, whereas site quality in Hubei and Gansu was larger than competition. The present study strengthened the importance of considering climate variables in developing branch length and diameter model. It is desirable to disentangle the different sources of variations in affecting branch length and diameter from different latitude regions to reduce the uncertainty in predicting branch characteristics under the condition of climate changing.

Key words: Branch characteristic model, Latitude regions, Climate, Competition index, Larix kaempferi