Leaf and hydraulic traits are key determinants of growth rates, and hence potentially exhibit significant associations with wood density (WD) and its intraspecific variation (ITV). However, the extent to which functional traits could improve WD prediction accuracy, and how ITV in WD correlates with functional traits remain incompletely understood. We investigated WD and its ITV across 10,218 plant species, mapped the global distribution of WD, and analyzed the association of ITV in WD with niche breadth and functional traits. Plant species with an acquisitive resource-use strategy, characterized by higher specific leaf area (SLA), leaf nitrogen concentration (LN), and leaf maximum stomatal conductance (gmax), exhibited lower WD. Associations of WD with hydraulic traits indicated species with greater hydraulic safety exhibited higher WD. Moreover, the integration of leaf traits (i.e., SLA and LN) and hydraulic traits with environmental factors substantially enhanced WD prediction accuracy in a random forest model, raising the explained variance from 55% to 95%. Furthermore, resource-acquisitive species demonstrated higher ITV for WD. ITV was positively related to relative niche breadth concerning both climatic factors and soil properties. Overall, functional traits significantly improve WD prediction accuracy, and plant species with an acquisitive resource-use strategy exhibit lower WD but greater intraspecific variation.

Afforestation is gaining global attention for its role in carbon sequestration and timber production. Afforestation affects hydrological processes in forest ecosystems, including water yield. However, there is a paucity of longterm observation data to examine the effects of afforestation on water yield. Moreover, drought caused by global warming could change the runoff generation process and make the forest-water relationship hotly debated. To fully assess these effects, we used 22-year data from a long-time observation from 1997 to 2019 (excluding missing data in 2017) in a Chinese fir (Cunninghamia lanceolata (Lamb) Hook.) plantation in subtropical China. We analyzed the response of runoff to drought and stand growth post afforestation. Average runoff during drought declined by 41.2%, 37.5%, and 3.7% lower than nondrought events at monthly, seasonal, and yearly timescales, respectively. The drought affected runoff significantly when a drought event lasted 1.2 months, and drought severity was larger than 1.5. Annual runoff and runoff anomaly decreased with stand age during the first 14 years after afforestation and thereafter started to recover. A similar change pattern was found for the relationship between annual runoff and stand biomass. Annual runoff also decreased with an increase in the basal area. Monthly runoff was primarily controlled by rainfall and soil water rather than stand characteristics, while annual runoff largely depended on stand biomass. These findings clarify the relationship between planted forests and water resources and provide insightful information for sustainable water management in forests under global climate change.

Transpiration is a key process that couples the land-atmosphere exchange of water and carbon, and its maximum water transport ability affects plant productivity. Functional traits significantly influence the maximum transpiration rate; however, which factor plays the dominant role remains unknown. SAPFLUXNET dataset, which includes sap flux density of diverse species worldwide, provides fundamental data to test the importance of photosynthetic and hydraulic traits on maximum tree-level sap flux density (Js_max). Here, we investigated variations in Js_max of 2194 trees across 129 species using data from the SAPFLUXNET dataset, and analysed the relationship of Js_max with photosynthetic and hydraulic traits. Our results indicated that Js_max was positively correlated with photosynthetic traits at both leaf and tree level. Regarding hydraulic traits, Js_max was positively related to xylem hydraulic conductivity (Ks), leaf-specific hydraulic conductivity (Kl), xylem pressure inducing 50 % loss of hydraulic conductivity (P50), xylem vessel diameter (Vdia), and leaf-to-sapwood area ratio (AlAs). Random forest model showed that 87 % of the variability in Js_max can be explained by functional traits, and hydraulic traits (e.g., P50 and sapwood area, As) exerted larger effects on Js_max than photosynthetic traits. Moreover, trees with a lower sapwood area or depth could increase their sap flux density to compensate for the reduced whole-tree transpiration. Js_max of the angiosperms was significantly higher than that of the gymnosperms. Mean annual total precipitation (MAP) were positively related to Js_max with a weak correlation coefficient. Furthermore, Js_max showed a significant phylogenetic signal with Blomberg's K below 0.2. Overall, tree species with acquisitive resource economics or more efficient hydraulic systems show higher water transport capacity, and the efficiency of xylem hydraulic system rather than the demand for carbon uptake predominantly determines water transport capacity.

Although plant–microbial interactions are essential to sustain terrestrial ecosystem functions, root–microbial interactions remain poorly understood. Therefore, understanding how functional trait diversity of fine roots mediates soil microbial diversity and composition provides insightful information to maintain forest ecosystem functions. A total of 87 plots with 1–12 tree species richness gradient in three subtropical secondary forests were selected. Functional trait diversity (functional dispersion, FDis, and community-weighted mean, CWM) of fine roots was quantified. Soil microbial diversity and community composition were analyzed by high-throughput Illumina sequencing. Bacterial and fungal richness was primarily predicted by CWM of root diameter (CWM_diam) in the humus horizon, but by FDis of specific root length (FDis_SRL) in the mineral horizon. Microbial community composition in two soil horizons increased with FDis_SRL, including the abundance of copiotrophic bacteria (r-strategists, e.g., Proteobacteria, Actinobacteria), Chloroflexi, saprotrophs and pathogens. FDis_SRL was also indirectly mediated microbial richness and composition via decreasing carbon:nitrogen ratio and increasing pH, total nitrogen, total and available phosphorus contents. Overall, functional trait diversity of fine roots plays a pivotal role in mediating soil microbial richness and community composition in subtropical secondary forests. These results deepen our understanding of the important role of plant–microbial interactions in terrestrial biogeochemical processes.

Tree species diversity is assumed to be an important component in managing forest ecosystems because of effects on multiple functions or ecosystem multifunctionality. However, the importance of tree diversity in determining multifunctionality in structurally complex subtropical forests relative to other regulators (e.g., soil microbial diversity, stand structure, and environmental conditions) remains uncertain. In this study, effects of aboveground (species richness and functional and structural diversity) and belowground (bacterial and fungal diversity) biodiversity, functional composition (community-weighted means of species traits), stand structure (diameter at breast height and stand density), and soil factors (pH and bulk density) on multifunctionality (including biomass production, carbon stock, and nutrient cycling) were examined along a tree diversity gradient in subtropical forests. The community-weighted mean of tree maximum height was the best predictor of ecosystem multifunctionality. Functional diversity explained a higher proportion of the variation in multifunctionality than that of species richness and fungal diversity. Stand structure -played an important role in modulating the effects of tree diversity on multifunctionality. The work highlights that species composition and maximizing forest structural complexity are effective strategies to increase forest multifunctionality while also conserving biodiversity in the management of multifunctional forests under global environmental changes.

Annual gross primary productivity (AGPP) is the basis for grain production and terrestrial carbon sequestration. Mapping regional AGPP from site measurements provides methodological support for analysing AGPP spatiotemporal variations thereby ensures regional food security and mitigates climate change. Based on 641 site-year eddy covariance measuring AGPP from China, we built an AGPP mapping scheme based on its formation and selected the optimal mapping way, which was conducted through analysing the predicting performances of divergent mapping tools, variable combinations, and mapping approaches in predicting observed AGPP variations. The reasonability of the selected optimal scheme was confirmed by assessing the consistency between its generating AGPP and previous products in spatiotemporal variations and total amount. Random forest regression tree explained 85 % of observed AGPP variations, outperforming other machine learning algorithms and classical statistical methods. Variable combinations containing climate, soil, and biological factors showed superior performance to other variable combinations. Mapping AGPP through predicting AGPP per leaf area (PAGPP) explained 86 % of AGPP variations, which was superior to other approaches. The optimal scheme was thus using a random forest regression tree, combining climate, soil, and biological variables, and predicting PAGPP. The optimal scheme generating AGPP of Chinese terrestrial ecosystems decreased from southeast to northwest, which was highly consistent with previous products. The interannual trend and interannual variation of our generating AGPP showed a decreasing trend from east to west and from southeast to northwest, respectively, which was consistent with data-oriented products. The mean total amount of generated AGPP was 7.03 ± 0.45 PgC yr−1 falling into the range of previous works. Considering the consistency between the generated AGPP and previous products, our optimal mapping way was suitable for mapping AGPP from site measurements. Our results provided a methodological support for mapping regional AGPP and other fluxes.

Leaf carbon isotope composition (δ13C) provides an integrative record on the carbon and water balance of plants over long periods. Photosynthetic ability and hydraulic traits which are highly associated with stomatal behavior could affect leaf δ13C. Association between photosynthetic ability and leaf δ13C has been examined, however, how hydraulic traits influence leaf δ13C has not been fully understood. To fill this gap, we investigated the variations in leaf δ13C among 2591 woody species (547 shrub and 2044 tree species), and analyzed the link of leaf δ13C with leaf photosynthetic and xylem hydraulic traits. Our result showed that leaf δ13C was positively correlated to leaf photosynthetic ability and capacity. For hydraulic traits, leaf δ13C was negatively related to hydraulic conductivity (Ks), xylem pressure inducing 50 % loss of hydraulic conductivity (P50) and vessel diameter (Vdia). Associations of leaf δ13C with xylem hydraulic traits indicate woody species with stronger hydraulic safety discriminated less against 13C, while woody species with higher hydraulic efficiency had more negative leaf δ13C. Shrub species, which showed a lower Vdia and P50, had a significant less negative leaf δ13C than tree species. Furthermore, woody species inhabiting in dry regions discriminated less against 13C than those growing in humid regions. Moreover, leaf δ13C displayed a low phylogenetic signal based on Blomberg's K statistic. Overall, woody species with a higher leaf photosynthetic ability or stronger hydraulic safety system discriminated less against 13C and adopt the provident water use strategy.

The annual gross primary productivity (AGPP) is the basis of food production and carbon sequestration in terrestrial ecosystems. An accurate assessment of regional AGPP can provide a theoretical basis for analyzing the spatiotemporal variation of AGPP and ensuring regional food security and mitigating climate change trends. Based on Chinese Flux Observation and Research Network (ChinaFLUX) measurements and public datasets, we produced a dataset of annual gross primary productivity over China’s terrestrial ecosystems was constructed. In combination with biological, climatic, and soil factors, we used the random forest regression tree to construct the assessment model of China AGPP by simulating the AGPP of unit leaf area. The dataset of annual gross primary productivity over China’s terrestrial ecosystems during 2000-2020 was generated with a spatial resolution of 30arcsecond and a data format of tiff. The dataset can provide validation data for model simulation, as well as data support for regional productivity, ecological quality, and assessment and management of terrestrial carbon sinks.

In order to accurately evaluate the characteristics of evapotranspiration （ET） and water consumption of Chinese fir （Cunninghamia lanceolata） plantation and explore the possibility of estimating its evapotranspiration by Penman-Monteith method （PM）. Based on the observation data of meteorological gradient and flux from the National Field Scientific Observation and Research Station of Chinese Fir Plantation Ecosystem in Huitong， Hunan Province from 2016 to 2019， this study used the eddy covariance method （EC） to analyze the temporal variation characteristics of Chinese fir plantation evapotranspiration， and compared the ET obtained with PM. The results showed that the ET of Chinese fir plantation showed single-wave diurnal and annual variations. During the day， the maximum value appeared at 12:30–13:00 in the afternoon， and the minimum value appeared at 0:00-2:30 after midnight. In a year， the maximum value of ET appeared in July–August， and the minimum value appeared in December–February. The ET increased with the increase of net ratio (R n )， air temperature (T a ) and vapor pressure difference （VPD）， and decreased with the increase of soil water content （SWC）， and the main meteorological factors affecting the ET of Chinese fir forest were R n ， T a and VPD. The ET obtained by PM and EC were consistent in spring and autumn， but significantly different in summer and winter. The difference of ET obtained by PM and EC basically showed a very significant linear positive correlation with R n ， T a and VPD， and a very significant linear negative correlation with SWC. The research results provided theoretical support for exploring the hydrological process of Chinese fir plantation and the feasibility of estimating its ET by PM.

Carbon and water cycles in forest ecosystems are tightly coupled, but global warming-induced soil and atmospheric droughts alter the coupling, thereby greatly increasing uncertainty in predicting carbon and water cycles. Therefore, a carbon–water coupled model (TRIPLEX-CW-Flux) was developed, and an R package (rTRIPLEXCWFlux) was created to facilitate model application. TRIPLEX-CW-Flux integrates vapor pressure deficit and soil moisture into a stomatal conductance submodule to estimate forest carbon and water fluxes. Prediction accuracy of TRIPLEX-CW-Flux and rTRIPLEXCWFlux application were evaluated in a Chinese fir (Cunninghamia lanceolata) plantation. Simulated net ecosystem production (NEP) and evapotranspiration (ET) were in good agreement with flux observations (R2: 0.76 for NEP; 0.71 for ET). Thus, the TRIPLEX-CW-Flux model can be used to predict and quantify effects of global warming-induced droughts on forest carbon and water cycles. The open-access rTRIPLEXCWFlux package facilitates estimations of carbon sequestration and water consumption in forest ecosystems using the observed flux data.

Understanding stomatal regulatory mechanisms across various woody species is helpful for explaining their adaptations to diverse environmental conditions. Stomatal opening and closing are driven by the requirements for maintaining water transport integrity and carbon uptake; however, distinguishing which factor plays a dominant role in the regulation of tree-level stomatal behavior has seldom been explored. To address this knowledge gap, we investigated differences in tree-level iso/anisohydric stomatal regulation among various tree species (61 and 59 species, at daily and hourly timescales, respectively) across diverse biomes, and analyzed the association of tree-level stomatal regulation with xylem hydraulic and leaf photosynthetic traits. Xylem hydraulic traits were closely related to tree-level stomatal regulation, whereas leaf photosynthetic traits showed non-significant correlations. Reduction in tree-level stomatal conductance with the same vapor pressure deficit increment (ranging from 0.6 to 2 kPa; Lcond), representing the degree of iso/anisohydry, was positively correlated to xylem pressure inducing 50% loss of hydraulic conductivity and vessel diameter, but negatively correlated to the hydraulic safety margin. The associations between xylem hydraulic traits and Lcond revealed that tree species with greater xylem hydraulic efficiency were more likely to adopt an avoidance strategy for tree-level stomatal regulation, whereas tolerance strategy occurred in species with a stronger hydraulic safety system. Furthermore, Lcond was positively correlated to mean annual precipitation and temperature, suggesting that species inhabiting humid and warm regions rely upon isohydric stomatal regulation. Moreover, Lcond displayed a phylogenetic signal, suggesting that variation in Lcond has been influenced by evolutionary history. Overall, tree-level stomatal regulation is more closely related to xylem hydraulic traits than to leaf photosynthetic traits, and maintaining water transport integrity rather than fulfilling requirements for carbon uptake is the major factor impacting tree-level stomatal regulation.

The process of rapid urbanization in China has led to various impacts on the urban ecological environment and one of these significant effects is the urban heat island (UHI). Pursuing the construction of a ecological city is recognized as an important direction to relieve the contradictions between environmental protection and city development. Yet how the landcape configuation and composition in such city development influences the UHI has not been well understood. This paper attempted to evaluate the spatial patterning of land surface temperature (LST) and to discuss the effects of source and sink landscape patterns upon LST in a smart city demonstrated at the local level in China. The results revealed that the UHI was spatially associated with the land use changes driven by rapid urbanization and gradually declines along the urban–rural gradient. When the urban–rural gradient reached the range of 7.5 to 8.75 km from the city center, there exhibited a tipping point where the combinative effects of source and sink landscapes on UHI significantly decreased. Through exploring the impact of source-sink landscape area ratio on UHI, we also found a tipping point that when source-sink area ratio in a patch was smaller than the golden separation value (1.30), UHI effect was almost mitigated because the warming effect of source area would be weaker than the cooling effect of sink area. Overall, this study highlights that controlling the optimal area ratio of source-sink landscape are very important strategy for mitigating the UHI effect during urban development and planning.

Forest farms are the main body of forest management at the landscape-scale in China, and have long sought ways to jointly maximize timber production and other ecosystem services. Carbon sequestration is of particular interest because China has established the goal of becoming carbon neutral by 2060. However, optimizing the balance between timber production and carbon sequestration is a challenge due to a lack of methods that account for the spatial arrangement of forest management activities within a landscape. This study developed a multi-objective planning model for forest management approaches to promote natural forest regeneration, adjusting the stand age structure of plantation forests to an even distribution across a landscape, and quantified the effects of social preferences and market prices on optimal management. The results show that depending on the dominant ecosystem services provided by the forests, forest management could promote an increased timber volume and stand carbon stock that is sustainable in the long-term. Social preferences have a relatively small influence on optimal management plans after forests are restructured to a normal forest age class distribution. In addition, forest management dynamics are more sensitive to carbon prices than timber prices. Hence, the spatial allocation of different management practices and forest restructuring, rather than protecting forests or managing them less intensively, is more effective for improving forest carbon stocks. More importantly, an urgent need also exists to increase the carbon price to guide and strengthen the attention of forest managers to forest carbon sequestration, with the goal of achieving a win–win strategy.

Total solar radiation is an important factor affecting carbon exchange in forest ecosystem. In order to understand the effects of radiation change on carbon exchange in Chinese fir plantation, long-term monitoring data of carbon dioxide flux and meteorological factors measured by open eddy covariance system and meteorological gradient observation system were used in this study. The clearness index (kt) was used to represent the condition of solar radiation. We analyzed the effects of kt on net ecosystem exchange of carbon dioxide (NEE) in the central subtropical Chinese fir plantation during the growing season (from April to October). The results showed that total solar radiation in clear sky was usually higher in the morning than that in the afternoon, and that NEE was lower in the morning than in the afternoon. Such difference in NEE reached the maximum when the solar elevation angle was about 50°. At the medium kt(0.42-0.52), carbon absorption of Chinese fir plantation was the strongest. The ave-rage maximum relative change of NEE in 10 years in different solar elevation angles ranged from 11.0% to 29.4%, while the minimum and maximum critical values appeared at 35°-40° and 45°-50°, respectively. When kt was at the moderate degree due to the existence of clouds, carbon absorption and diffuse photosynthetically active radiation of Chinese fir plantation reached the maximum, and the latter might be the main reason for the former. Moderate radiation condition with the presence of cloud clould promote NEE of Chinese fir plantation and lead to largest carbon absorption.

Canopy interception loss is an important hydrological process that affects rainfall redistribution, nutrient cycling, and soil and water conservation in forest ecosystems. However, the manner in which interception loss is affected in forests that are in restoration stages dominated by different tree species has not been fully understood. In this study, interception loss, throughfall, and stemflow were measured consecutively for 2 years (May 2017 through April 2019) in three secondary forests in subtropical China: a coniferous and broadleaved mixed forest ( Pinus massoniana , Lithocarpus glaber , PM) in early restoration stage, a deciduous broadleaved forest ( Choerospondias axillaris , CA) in middle restoration stage, and an evergreen broadleaved forest ( L. glaber – Cyclobalanopsis glauca , LG) in late restoration stage. For a given amount of rainfall, the relative interception loss in the CA was significantly lower than that in the evergreen forests in summer, spring, or winter. The amount of interception loss during light rain, moderate rain, or heavy storms was also significantly lower in CA as compared to that in evergreen forests, owing to the lesser leaf area and phenology of deciduous tree species as well as a higher leaf area index (LAI) and stand density in PM and LG. Meanwhile, the highest relative interception loss, interception loss amount, canopy storage capacity, canopy cover fraction, and mean evaporation rate was observed in the climax forest (LG), which has a higher LAI and stand density. Further, the revised Gash's analytical model (RGAM) and the sparse Rutter models were successfully applied to simulate interception loss in these forests, with the RGAM performing better as compared to the sparse Rutter model. The results of the field measurements and model simulations depict that the climax community at the late restoration stage can effectively reduce the amount of rainwater reaching the soil surface, thus weakening the effect of flood peaks.

Afforestation is an effective method to increase carbon (C) sinks and address climate change. It is crucial to understand how the stand growth affects C sequestration capacity, especially when the trade-offs with timber production from plantations have not been fully examined. We used a chronosequence approach to estimate C storage in Chinese fir (Cunninghamia lanceolata (Lamb.) Hook.) plantations (including the trees, understory, litter, and soils) at seven stand ages (3, 8-11, 16, 21, 25, 29, and 32 years). Ecosystem C storage increased nonlinearly from 76.4 to 282.2 t ha(-1) with stand age and was fitted with a logistic model that had a maximum C storage and age of 271.9 t ha(-1) and 33 years, respectively, to reach 95% of the maximum stored C. The C increment was mainly contributed by an increase in tree biomass, which ranged from 2.8 to 177.7 t ha(-1) and comprised 4-64% of the total ecosystem C. Live root C (sum of the stump, coarse, and fine root C) showed a logistic increase from 2.0 to 26.3 t ha(-1) with stand age and constituted 2.5-9.3% of ecosystem C. Understory plants and litter represented a small pool (< 2% of ecosystem C). The C storage in shrubs and litter slightly increased, while that in herbs decreased as the stands aged. Soil C storage was an important and relatively stable pool, ranging from 69.6 to 130.1 t ha(-1). Stand volume was also best fitted with a logistic model with a maximum value of 552.6 m(3 )ha(-1). Additionally, the time needed to reach 95% of the maximum volume was 25 years. Hence, extending the rotation age to over 30 years for Chinese fir plantations could potentially maximize the synergistic benefits of C storage to mitigate climate change and obtain timber products for economic profit.

Background Forest ecosystems play an important role in carbon sequestration, climate change mitigation, and achieving China's target to become carbon (C) neutral by 2060. However, changes in C storage and net primary production (NPP) in natural secondary forests stemming from tree growth and future climate change have not yet been investigated in subtropical areas in China. Here, we used data from 290 inventory plots in four secondary forests [evergreen broad-leaved forest (EBF), deciduous and evergreen broad-leaved mixed forest (DEF), deciduous broad-leaved forest (DBF), and coniferous and broad-leaved mixed forest (CDF)] at different restoration stages and run a hybrid model (TRIPLEX 1.6) to predict changes in stand carbon storage and NPP under two future climate change scenarios (RCP4.5 and RCP8.5). Results The runs of the hybrid model calibrated and validated by using the data from the inventory plots suggest significant increase in the carbon storage by 2060 under the current climate conditions, and even higher increase under the RCP4.5 and RCP8.5 climate change scenarios. In contrast to the carbon storage, the simulated EBF and DEF NPP declines slightly over the period from 2014 to 2060. Conclusions The obtained results lead to conclusion that proper management of China’s subtropical secondary forests could be considered as one of the steps towards achieving China’s target to become carbon neutral by 2060.

Accurate estimation of regional and global patterns of ecosystem respiration (ER) is crucial to improve the understanding of terrestrial carbon cycles and the predictive ability of the global carbon budget. However, large uncertainties still exist in regional and global ER estimation due to the drawbacks of modeling methods. Based on eddy covariance ER data from 132 sites in China from 2002 to 2020, we established Intelligent Random Forest (IRF) models that integrated ecological understanding with machine learning techniques to estimate ER. The results showed that the IRF models performed better than semiempirical models and machine learning algorithms. The observed data revealed that gross primary productivity (GPP), living plant biomass, and soil organic carbon (SOC) were of great importance in controlling the spatiotemporal variability of ER across China. An optimal model governed by annual GPP, living plant biomass, SOC, and air temperature (IRF‐04 model) matched 93% of the spatiotemporal variation in site‐level ER, and was adopted to evaluate the spatiotemporal pattern of ER in China. Using the optimal model, we obtained that the annual value of ER in China ranged from 5.05 to 5.84 Pg C yr−1 between 2000 and 2020, with an average value of 5.53 ± 0.22 Pg C yr−1. In this study, we suggest that future models should integrate process‐based and data‐driven approaches for understanding and evaluating regional and global carbon budgets.

Despite the advances in technologies to derive the maximum rate of Rubisco carboxylation (V-cmax) seasonality at large-scale, factors controlling the temporal dynamics of V-cmax is largely unknown without extensive field measurements at stand scale. In addition, state-of-the-art process-based terrestrial ecosystem models had not accounted the complex canopy structure for estimating radiation interception by vegetation. Both of which could lead to uncertainties in gross ecosystem production (GEP) estimations. Here, we examined the respective and combined effects of leaf age and canopy structure on GEP by integrating the Farquhar photosynthesis model with a two-leaf (sunlit and shaded) canopy radiation interception model based on the Geometric Optical and Radiative Transfer (GORT) theory. We observed that the V-cmax of new leaves was approximately 34.4% higher than that of mature leaves. Considering the seasonal dynamics of V-cmax caused by new leaf expansion, the average bias between the modeled GEP and that derived from the eddy covariance (EC) in the growing season at 8-day temporal scale was reduced from -8% to -1%, with slightly higher correlation (from 0.9 to 0.92) and reduction in root mean squared error (from 6.0 to 3.9 g C m(-2) 8d(-1)). Most importantly, the total effect of new leaf expansion and canopy gaps during the growing season was +322 g C m 2 yr(-1) and -114 g C m(-2) yr(-1), which was approximately 22.5% and 8.1% of the total GEP, respectively. Thus, it is important to consider both leaf age and canopy structure when developing a robust terrestrial ecosystem model. We highlighted that V-cmax seasonality and realistic canopy radiation interception simulation are critical for accurate estimation of canopy GEP for evergreen forests.