Key message
A climate-sensitive height to crown base (HCB) model developed by combining a nonlinear mixed-effects model and dummy variable approach led to higher prediction accuracy of HCB than those without climatic variables for moso bamboo.
Height to crown base (HCB) is one of the important variables used in forest growth and yield models, as it is crucial for assessing vitality, competition, growth and development stage, stability, and production efficiency of the individuals. As climate impact is substantial on HCB, its inclusion of any forest model is crucial to make the model climate sensitive. However, existing HCB models do not consider climate impact on Phyllostachys pubescens (moso bamboo) HCB. With data collected from 26 moso bamboo sample plots in Jiangsu and Fujian provinces in China, we used five common HCB functions to develop climate sensitive HCB models. Modeling showed the significant effects of two individual variables (height—H, diameter at breast height—DBH), two stand-level variables (quadratic mean DBH—QMD, canopy density—CD), and two climate variables (extreme maximum temperature—EXT and Hargreaves’ climatic moisture deficit—CMD) on HCB. Compared with the basic model, the introduction of covariates (QMD, CD, EXT and CMD), dummy variable (regions), and random effects (block- and sample plot-level random effects) resulted in increased R2 by 5.01%, 7.13%, 7.14%, and 13.34%, respectively. The logistic model provided better fit statistics than other models we evaluated. Two-level nonlinear mixed-effects (NLME) models significantly improved fit statistics. Response calibration (model localization) with two medium-sized bamboos per sample plot provided the optimal prediction accuracy. This strategy can be considered as a reasonable compromise between the measurement costs and errors for HCB prediction.