Modeling stand biomass for Moso bamboo forests in Eastern China

Abstract

Stand biomass models can be used as basic decision-making tools in forest management planning. The Moso bamboo (Phyllostachys pubescens) forest, a major forest system in tropical and subtropical regions, represents a substantial carbon sink, slowing down the rise of greenhouse gas concentrations in the earth's atmosphere. Bamboo stand biomass models are important for the assessment of the contribution of carbon to the terrestrial ecosystem. We constructed a stand biomass model for Moso bamboo using destructively sampled data from 45 sample plots that were located across the Yixing state-owned farm in Jiangsu Province, China. Among several bamboo stand variables used as predictors in the stand biomass models, mean diameter at breast height (MDBH), mean height (MH), and canopy density (CD) of bamboo contributed significantly to the model. To increase the model's accuracy, we introduced the effects of bamboo forest block as a random effect into the model through mixed-effects modeling. The mixed-effects model described a large part of stand biomass variation (R2 = 0.6987), significantly higher than that of the ordinary least squares regression model (R2 = 0.5748). Our results show an increased bamboo stand biomass with increasing MH and CD, confirming our model's biological logic. The proposed stand biomass model may have important management implications; for example, it can be combined with other bamboo models to estimate bamboo canopy biomass, carbon sequestration, and bamboo biomass at different growth stages.

Keywords: Moso bamboo; aboveground biomass; belowground biomass; heteroscedasticity; random effect.