Models considering the theoretical stand age will underestimate the future forest carbon sequestration potential

The future biomass carbon sequestration (BCS) in forests were often predicted by the theoretical stand age (TSA, based on the aging of the stand). Due to tree regeneration and various disturbances, however, the real stand age (RSA, calculated by averaging the age of single individuals in the stand) is often inconsistent with TSA in a given forest, and its effect on BCS prediction was poorly understood. Here, this study analysed the variations in RSA in three forest types (i.e., coniferous, broadleaf, and mixed broadleaf forests) of two forest origins (i.e., planted and natural forests) using the National Forest Inventory dataset of China between 1999 and 2018, and evaluated their effects on BCS between 2020 and 2060 using a random forest model. The ratio of RSA to TSA differed in forest origins and age groups. For all forest types, the ratio was higher and increased with an increase of age group in planted forests, while it showed opposite trends in natural forests, indicating a high variability in stand age in natural forests. The differences in predicted biomass carbon (C) storage between RSA and TSA varied over time. The enhanced C storage from RSA in natural forests in 2060 was characterized by a trend of mixed broadleaf forests (45.61 TgC) > broadleaf forests (17.62 TgC) > coniferous forests (8.16 TgC). Moreover, the predicted BCS in all forests were higher in the RSA scenario than in the TSA scenario and showed various trends between 2020 and 2060. Especially in natural forests, broadleaf forests showed a high and stable BCS (from 15.14 TgC yr−1 to 15.44 TgC yr−1) and mixed broadleaf forests exhibited an increased BCS (from 60.18 TgC yr−1 to 63.90 TgC yr−1) during this period. Our results confirmed the widespread phenomenon of inconsistency between RSA and TSA in China's forests and underlined their various effects on future forest BCS. More importantly, these results suggested that considering the RSA rather than the TSA is more scientific for forest C accounting.