Plant functional traits are measurable characteristics formed from long-term plant-environment interactions and are closely linked to the growth, reproduction, and survival processes of plants (Violle et al., 2007; Garnier et al., 2016; Kattge et al., 2020). A single trait and the combinations of multiple traits not only respond to plant life history strategy, biodiversity, vegetation, environment, and disturbance changes but also affect ecosystem processes and functions (Cornelissen et al., 2003; Vandewalle et al., 2010; Mensens et al., 2017; Wang et al., 2020). Therefore, the study of plant functional traits is at the forefront of modern ecology and global change ecology (Meng et al., 2007; Liu and Ma, 2015; Blondeel et al., 2020).

Forests are the largest vegetated areas in global terrestrial ecosystems and play important roles in global biodiversity maintenance and climate change mitigation (Melillo et al., 1993; Anderson-Teixeira et al., 2015). Plant functional traits, such as morphological traits and C: N: P ecological stoichiometry of leaves, roots, stems, and twigs (especially leaves) in forests worldwide have been intensively investigated at local, regional and global scales (McGroddy et al., 2004; Chave et al., 2009; Díaz et al., 2016; He et al., 2019; Guerin et al., 2021; Liu et al., 2022). Compared to interspecific trait variations (trait differences among species) of woody plants in forests, intraspecific trait variations (trait differences among independent individuals within a species) are yet to be fully investigated. However, intraspecific trait variations are non-negligible, and they may account for an average of 32% of the total trait variations based on a global database (Siefert et al., 2015). Environmental factors, such as climate, topography, and soil nutrition, greatly affect intraspecific trait variations (Hultine and Marshall, 2000; Thuiller et al., 2004; Quested et al., 2007; Wang et al., 2016a).

In addition, most plant functional trait studies conducted in forests have focused on dominant trees, and understory shrubs have rarely been described. As an important component of forest ecosystems, understory shrubs influence the energy and water cycles of forests and participate in forest renewal, development, and succession (Cao et al., 2020). In limited trait studies on understory shrubs, the latitudinal distribution pattern and driving factors of leaf traits (Luo et al., 2019), the correlation between morphological traits and nitrogen and phosphorus nutrients (Wang et al., 2017), relationships between twigs and leaf traits at different succession stages (Ma, 2014), and different phenotypes of understory shrubs (Wang et al., 2016b) have been reported. Light conditions and canopy coverage are the main environmental factors driving intraspecific variations in leaf traits (Burton et al., 2017). However, studies on intraspecific trait variation in understory shrubs in forests at different succession stages in different habitats are scarce (Kumordzi et al., 2019; Vanneste et al., 2019; Cao et al., 2020).

Environments (or habitats), species composition, and the structure of a forest community change with progressive succession. In early-stage forests, illumination is sufficient, but soil nutrients are poor, and soil temperature and humidity fluctuate. Such forests are committed to gap formation, understory initiation, and regeneration (Moridi et al., 2015; Zhang et al., 2016). In medium-stage forests, soil humidity and nutrients are greatly improved, but light availability is the lowest because of the highest canopy coverage and stand density. Such forests are in volume accumulation, lightning, and stem exclusion phases (Zhang et al., 1999; Wang et al., 2004; Yan et al., 2008; Moridi et al., 2015). Light conditions improve in late-stage forests because of slightly lower canopy coverage and stand density (Moridi et al., 2015; Zhang et al., 2016). Some species (turnover species) only appear in forests at one succession stage and disappear in the next stage, whereas, some species (overlapping species) grow in forests at many succession stages (Yang et al., 2014). Overlapping species in the early-stage forest may form a series of trait combinations for a resource conservative strategy to adapt to fragile habitats with poor soil nutrients and changeable soil temperature and humidity, and those in the medium- and late-stage forests may form converse trait combinations for a resource acquisitive strategy to adapt to low light availability and strongly competitive habitats (Li and Ma, 2002).

Originating from the uplift of the Qinghai–Tibet Plateau, a broad subtropical evergreen broad-leaved forest was found in China. This forest is a unique vegetation type on Earth and is essential to the global gene pool and regional environments (Song, 2013). In the present study, dominant overlapping species, including tree and understory shrub species, in three evergreen broad-leaved forests at different succession stages in eastern China were investigated as examples. The leaf, twig, stem, and root traits were comprehensively measured. Here, the life strategies of plants growing in forests at different succession stages in various habitats were determined from the perspective of plant functional traits. Specifically, the following predictions were made. (1) Overlapping species in the forests at three succession stages present significant differences in plant functional traits. (2) Plants (trees and understory shrubs) in the early-stage forest form a series of trait combinations with higher leaf thickness (LT), dry matter contents and tissue densities of leaves, roots, twigs, and stems and lower leaf area (LA) and specific leaf area (SLA) for a resource conservative strategy with a low growth rate. Plants in the medium- and late-stage forests form converse trait combinations for a resource acquisitive strategy with a high growth rate. This study provides basic data to global (Kattge et al., 2020) and China (He et al., 2019) plant trait databases and guides the sustainable management and ecological restoration of evergreen broad-leaved forests in eastern China.

Few studies on intraspecific trait variations have been conducted compared with studies on interspecific trait variation in the world’s forests, and most trait studies have focused on trees and their leaf traits, such as LA, LDMC, and SLA (Rosbakh et al., 2015; Siefert et al., 2015; Zhou, 2016; Heilmeier, 2019; Zhang et al., 2019). The intraspecific trait variation characteristics of understory shrubs and those of the twigs, stems, and roots of trees have rarely been reported (Burton et al., 2017; Zhong et al., 2018; Cao et al., 2020). The present study comprehensively investigated the intraspecific variation characteristics of 12 traits of leaves, twigs, stems, and roots of six common species (two tree species and four understory shrub species) in three typical subtropical evergreen broad-leaved forests in eastern China. This study could fill the gap in plant functional trait studies in subtropical China.

Since the application of succession in plant ecology research, studies on vegetation succession have been widely performed. Biodiversity and carbon storage were often used as indicators to evaluate vegetation succession processes in early studies (Ni et al., 2015; Capellesso et al., 2021). Plant functional traits and trait-based community ecology (a theory using trait-based approaches to determine community composition, structures and functions) offer an alternative approach to reveal the colonization, survival, growth, and death processes of plants and plant communities in different succession vegetation (Song et al., 2018). However, species compositions vary in different succession vegetations, and traits of different species were measured in previous studies. Rather than considering several dominant species in different succession vegetation (Chazdon, 2008; Muscarella et al., 2017), the present study focused on overlapping species and provided another pathway to understand the interaction of plant functional traits and vegetation succession.

Plant functional traits are codetermined by species identity (genetic factors) and environment (Scheiner and Lyman, 1991; Weiher and Keddy, 1995). Environments promote the coexistence of species with similar functional traits in local habitats, but intraspecific trait variations cause differences in the life history strategy of a certain species or different species (Ackerly and Cornwell, 2007; Auger and Shipley, 2013; Siefert et al., 2015; Kumordzi et al., 2019). However, the species composition of a plant community differs with the environment. The effects of environments on traits are often discussed without considering the effects of species identity, or some statistical analysis methods, such as two-way ANOVA, are used in species–environment interactive analysis in other studies (Zhong et al., 2018; Cao et al., 2020). In the present study, overlapping species in three typical subtropical evergreen broad-leaved forests with different habitats in eastern China are investigated as an example to explore the effects of the environment on traits. This investigation method is preferred because of the direct exclusion of the effects of species identity.

Root traits often exhibit the largest interspecific and intraspecific variations resulting from complex and diverse below-ground habitats (Comas and Eissenstat, 2004). Leaf traits display larger interspecific and intraspecific variations than twig and stem traits (Ackerly and Reich, 1999; Garnier et al., 2001; Zhong et al., 2018). In the present study, variations in the intraspecific traits of overlapping species in the forests at three succession stages with different habitats ranged from 2.8% to 56.1%, with an average of 13.5%. Some traits of roots and leaves considerably varied. SRL, RTD, LA, and SLA also differed, but RDMC, LT, LTD, and LDMC (especially the last leaf trait) slightly varied. Consistent with early studies (Garnier et al., 2001), our study showed that intraspecific variations in twig and stem traits slightly differed.

A single trait and the combinations of multiple traits can indicate the adaptation and response of plants to certain environments, expose the plant trade-off strategies of resource acquisition and utilization, and reveal the different life history strategies of plants in different environments (Wright et al., 2004; Díaz et al., 2016). Some species (resource acquisitive strategy adopters) present a high SLA, leaf nitrogen content, photosynthetic and respiratory rates, and short lifespans. Some other species (resource conservative strategy adopters) display low SLA, leaf nitrogen content, photosynthetic and respiratory rates, and long lifespan (Chen and Xu, 2014). Such plant and trait combinations favor niche overlap and competition reduction, rich biodiversity maintenance, and ecosystem stability enhancement (Liu and Ma, 2015). In the present study, the early-stage forest is a relatively fragile ecosystem with poor soil nutrients and fluctuating soil temperature and humidity. Trees and understory shrubs in such forests allocate more nutrients to resist adverse environments. Thus, they adopt a slow investment-return (resource conservation) strategy and form a series of trait combinations with higher LT, LDMC, LTD, TDMC, TTD, SDMC, STD, RDMC, and RTD and lower SLA and SRL for storing more nutrients and growing slower. However, in the medium- and late-stage forests, trees and understory shrubs are exposed to lower light availability and stronger competition. They adopt a quick investment-return (resource acquisitive) strategy and are associated with a series of trait combinations with higher LA, SLA, and SRL and lower LT, LDMC, LTD, TDMC, TTD, SDMC, STD, RDMC, and RTD for faster growth. Plants in medium-stage forests are more resource acquisitive for lower light availability and stronger competition than those in late-stage forests.

The remaining primary evergreen broad-leaved forests only account for 4% of its total distribution area in eastern China because of intensive human disturbances. Degraded secondary forests, shrub and grass communities, and bare land are distributed everywhere. However, excellent hydrothermal conditions promote secondary forest succession once human disturbances stop. Evergreen broad-leaved forests may recover gradually in this area. Therefore, research on the secondary succession mechanism and restoration ecology of evergreen broad-leaved forests should be strengthened (Song, 2013). Understanding the trend of changes in trait characteristics with vegetation succession and the life strategies of plants adapted to different succession vegetation is important not only for providing data to the global and China plant trait databases but also for establishing regional forest management and restoration in eastern China.

The original contributions presented in this study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

LL and YW conceived and designed the research. LL and HX analyzed the data. XQ contributed to the field work. LL wrote the first draft with substantial input from YW. All authors contributed to the article and approved the submitted version.