Xiangqi Kong

The increase in global carbon emissions has intensified the effects of CO₂ fertilization on the carbon cycle. CO₂ fertilization is shaped by several factors, including the physiological differences among trees of varied forest ages and types, as well as the influence of different climatic conditions. It is essential to investigate the differences in CO₂ fertilization effects across diverse climate zones and delve into the association between these effects and forest age and type. Such exploration will deepen our knowledge of forest responses to environmental changes. This study used annual ring width data from the International Tree-Ring Data Bank, employing the generalized additive mixed models and the Random Forest model to discern the pattern of the CO₂ fertilization effect concerning forest age in the Northern Hemisphere. This study also explored the variations in the effect of CO₂ fertilization across unique climate zones and the disparities among various forest types within the same climatic zone. The results indicated a link between forest age and the CO₂ fertilization effect: it tends to increase in sapling forests and middle-aged forests and diminish in mature forests. Warmer, drier environments had a more marked effect of increased CO₂ on tree fertilization. Additionally, coniferous forests demonstrated a more substantial CO₂ fertilization effect than broadleaf forests, and deciduous needle-leaf forests surpassed evergreen needle-leaf forests in this regard. This research is pivotal in understanding the shifting patterns of CO₂ fertilization effects and how forests respond to atmospheric changes.

The CO₂ fertilization effect (CFE) plays a crucial role in the amelioration of climate change. Many physiological and environmental factors, such as stand age, stand type, and site conditions, may affect the extent of the CFE. However, the relationship between the CFE and these factors remains elusive. In this study, we used the emerging gross primary production (GPP) remote sensing products, with GPP predicted using eddy covariance–light use efficiency models (EC-LUE GPP) based on satellite near-infrared reflectance of vegetation (NIRv GPP) and assessed with a random forest model to explore the CFE trends with stand age in a coniferous forest and a broad-leaved forest in Heilongjiang Province, China. We additionally compared the differences among the CFEs under different site conditions. The CFEs in coniferous forests and broad-leaved forests both showed a rapid increase in stands of 10 to 20 years of age, followed by a decline after reaching a maximum, with the rate of decline reducing with age. Eventually, CFE remained stable in stands near 100 years of age. However, the CFE in coniferous forests exhibited more extended periods of rapid increase and a higher maximum than in broad-leaved forests. Moreover, in this study, we used the site class index (SCI) to grade site conditions. The results demonstrate that the CFE differed significantly under different levels of site conditions, and these differences gradually decreased with age. The site with the highest SCI had fewer environmental restrictions on the CFE, and consequently, the CFE rate of decline was faster. Our results are of significance in understanding the CFE and adapting to future changes in atmospheric CO₂ concentration.