Effects of tree size heterogeneity on carbon sink in old forests

Old-growth forests are carbon sinks as supported by increasing evidence in recent years. However, finding a resolution of the contradiction between the carbon sink function and neutral hypothesis at the community and individual scales is still a challenge. Tree size heterogeneity may offer an alternative answer. To determine this point, we studied subalpine (above an altitude of 3000 m) primeval Abies fabri forests located in the western part of China by comparing theoretical predictions and actual results. In theory, we first derived a model of the general carbon use efficiency (CUE) based on an individual scale, which allowed us to obtain the stand CUE by incorporating the carbon budget of all individuals. Afterwards, we quantified the effects of the gradual disturbance on the mortality of individuals with different sizes and predicted that the size of most individuals will trend to be medium (relative to the largest with the state of carbon neutral), meaning that the CUE of old stands has a high probability of tending to be a certain value (e.g. 0.4). In practice, the CUEs of old and middle-aged communities (0.405 and 0.602) calculated by the model were exceedingly close to the actual (0.401 and 0.597), indicating the effectiveness of the model. Further model-based analyses were performed, showing that the CUE of two old communities at different altitudes during the period from 2005 to 2015 were around 0.40, which are different from the reduced CUE from 0.64 to 0.60 in the middle-aged community. Meanwhile, in old forests, heterogeneous individuals dominated by medium-sized individual trend to be stable. Our findings indicated that with the increases in gradual disturbance events, a stable distribution of medium-sized individuals is an important cause for the carbon sink of old stands, and most of the carbon uptake by trees may further be stored in the soil. In addition, our model implied that environmental factors may change the forest carbon sink capacity by affecting the individual's potential maximum biomass.