Impact of canopy vertical height on leaf functional traits in a Cunninghamia lanceolata common garden experiment in Dagangshan

Abstract

Plant functional traits represent resource acquisition strategies and, consequently, reflect ecosystem functions. As a critical focus on functional trait research, the study of intraspecific variation and vertical gradients in canopies can better reveal plant adaptation mechanisms to environmental conditions. In this study, we examined five provenances of Cunninghamia lanceolata in the common garden of Dagangshan, Jiangxi. We measured and analyzed nine functional trait indicators: the leaf dry matter content, relative chlorophyll content, specific leaf area, leaf tissue density, water use efficiency, leaf carbon content, leaf nitrogen content, carbon-to-nitrogen ratio, and equivalent water thickness. The results revealed that canopy height has a minor effect on the variation in leaf functional traits, with most of the variation attributed to geographic provenances. Notably, 50% of the leaf functional traits presented significant correlations, which were mainly associated with photosynthetic capacity-related traits. The correlations of leaf functional traits between the average canopy values and the middle part of the canopy are similar. The influence of canopy height on leaf functional traits in subtropical Cunninghamia lanceolata artificial forests is limited, but the provenances had a certain impact on the leaf functional traits. The consistency of trait relationships between middle part and the entire canopy making it feasible to analyze forest ecosystem functions by collecting leaf samples from the middle canopy. This study provides scientific evidence for research on the functional dynamics of Cunninghamia lanceolata artificial forest ecosystems and the selection of superior provenances.

Introduction

Plant functional traits elucidate the mechanisms of species coexistence and community dynamics within forests¹. Leaves, which are sensitive to environmental changes², exhibit functional traits that are relatively straightforward to measure, making them crucial for research on plant functional traits³. Since Wright et al. (2004) combined a series of related leaf functional traits to represent plant resource balancing strategies and proposed the Leaf Economic Spectrum (LES)⁴, this theory has been extended to include traits of wood and other plant organs^5–7. The Plant Economic Spectrum(PES) can predict the dynamics of community composition and ecosystem functions⁸. Reich (2014) found a broad trade-off between slow growth (conservative strategy) and rapid growth (acquisition strategy) among species by integrating the characteristics of roots, stems, and leaves, including traits related to water, carbon, and nutrient resources⁹. The theory of the plant economic spectrum provides a basic theoretical framework for understanding plant adaptation strategies by integrating the trade-off between plant functional traits in resource acquisition and resource conservation^10,11.Leaves have received the most widespread attention due to their ease of sampling and measurement¹². Leaf traits not only reflect resource acquisition capabilities but also characterize plant survival strategies^13,14. The primary categories of research on plant leaf traits include morphological, physiological, biochemical, and stoichiometric traits¹⁵. The individual or combined effects of these different traits can reveal adaptation strategies to environmental conditions^16,17.

Research has focused on the impact of environmental factors on leaf functional traits^18,19, as well as differences in species traits²⁰ and adaptation strategies at a regional scale²¹. However, investigations into trait variation and adaptation characteristics within small habitats at the individual scale are relatively limited. Light is a key factor influencing plant growth and development, and is closely linked to photosynthesis²². Previous studies have indicated that light intensity affects plant functional traits²³. Notably, the nitrogen and phosphorus contents per unit leaf area of canopy plants are 60% and 90% higher than the average contents in understory plants, respectively²⁴. As the plant part that is most directly connected to the external environment, the canopy itself has altered light intensity because of its shape and structure²⁵. Consequently, variations in light, temperature, moisture, and other conditions may occur at different heights within the canopy²⁶. Leaf functional traits may change in response to these habitat variations²⁷. Some studies have shown that leaves situated in the upper part of the canopy tend to have smaller leaf areas, greater leaf thickness, and higher leaf mass per area, whereas leaves in the lower canopy exhibit higher specific leaf area and nitrogen content to increase resource utilization efficiency^28,29.

Most studies have focused on the variation in plant functional traits at the community level along environmental gradients^30,31, often neglecting the variation in individual functional traits within species. The adaptability of individuals to the micro-environment leads to variations in plant functional traits^32,33, whereas genetic differences, phenotypic plasticity, and other factors result in intraspecific variation^34,35. Many studies have demonstrated that nearly one-third of trait variation within a community stems from intraspecific trait variation^36,37, which is especially pronounced when the number of species in the community is low or when environmental stress is minimal³⁸. In the process of adapting to diverse environments, plants develop geographical provenances³⁹. By transplanting these provenances from different regions into common gardens, researchers can observe variations in growth⁴⁰, enabling them to elucidate the adaptation strategies of these provenances to identical environmental conditions, which is a crucial consideration for selecting superior provenances in the context of climate change^41,42. Cunninghamia lanceolata (commonly known as Chinese fir), an important timber resource, is used for artificial afforestation in southern China, which is widely distributed in 19 provinces⁴³. Chinese fir is the main dominant tree species in the national plantation forests of China⁴⁴. China has conducted extensive long-term research on the provenances of Chinese fir, with studies examining the variation in growth, resistance, and functional traits⁴⁵. In particular, investigating leaf functional traits at different canopy heights from various provenances provides valuable insights into the developmental characteristics and growth strategies of tree canopies across these provenances.

Leaves from the middle canopy are frequently sampled to represent plant functional traits^46,47, yet vertical trait variation across canopy gradients remains understudied. From lower to upper canopy, SLA decreases, leaf thickness increases and leaf photosynthesis ability relative traits are adjusted to optimize carbon gain under the strong vertical light gradient^17,25,26. This study examines five Cunninghamia lanceolata provenances in a common garden (Dagangshan, Jiangxi), analyzing leaf functional traits at multiple canopy heights. We specifically test three hypotheses driven by adaptation theory^19,21: (1) functional traits will differ significantly among provenances due to genetic adaptation to native environments; (2) functional traits will be affected by vertical height in the canopy; and (3) the relationships among functional traits will change along the canopy gradient.

Materials and methods

Study site

This study was conducted at the Dagangshan National Key Field Observation and Research Station for Forest Ecosystem, JiangXi Province, China(114°31′23″E ~ 114°44′31″E, 27°32′2″N ~ 27°49′31″N). The mean annual air temperature and precipitation are 15.8 ◦C and 1600 mm, respectively. Most precipitation occurs from April to June, accounting for 45% of the annual precipitation. In 1981, the Chinese Academy of Forestry conducted a geographical provenance experiment of Chinese fir here⁴⁵. The provenances were collected across 13 provinces in China, using a grid-based method for site selection based on latitude and longitude. Seedlings were raised in 1980 and planted in the spring of 1981. The trial employed a balanced incomplete block design (BIBD) with 183 blocks and 14 replications. Each block contained 14 plots, with 4 trees planted per plot at a spacing of 2 m × 2 m. In this study, we randomly selected 5 provenances growing in the same block with nearly the same climatic and soil conditions (Table 1). The height (H) and diameter at breast height (DBH) of each selected individual were measured before sampling.

Table 1.

Oringin of provenances, tree height (m; mean ± standard deviation) and diameter at breast height (cm; mean ± standard deviation) of sample trees. H indicates the height of sample trees, DBH indicates the diameter at breast height of sample trees. Different letters indicate significant differences among provenances (P < 0.05), same letters indicate non-significant differences among provenances (P > 0.05).

Provenances	H(m)	DBH(cm)
ZJYH(Zhejiang Yunhe)	16.17 ± 1.84a	21.54 ± 3.52a
JXAF(Jiangxi Anfu)	18.33 ± 1.65a	28.09 ± 3.27a
JXLA(Jiangxi Lean)	18.10 ± 1.37a	23.76 ± 1.88a
HBWF(Hubei Wufeng)	17.67 ± 2.36a	25.83 ± 6.38a
GXYF(Guangxi Yongfu)	17.50 ± 1.08a	22.27 ± 2.54a

Samples collection

In this study, considering the actual conditions and heights of the selected provenances, the canopy was categorized vertically into three layers: the lower layer (＜ 6 m above ground), the middle layer (6 ~ 11 m above ground), and the upper layer (＞11 m above ground). In September 2023, three individuals from each provenance were selected as replicates (15 Chinese fir individuals in total). All selected trees were in good and similar growth condition. Leaves were collected from the upper, middle, and lower layers of the canopy for each individual, by pruning and climbing techniques. During the sampling process, healthy branches were chosen from various layers and directions of each individual, with at least 50 mature, healthy, and intact leaves at each layer were collected from this year’s growth segments, which were immediately placed in envelopes. Bring back to the laboratory for measuring functional traits within two hours.

Measurement of functional traits

Nine functional traits of leaves were chosen for measurement in this study: leaf dry matter content (LDMC, g/kg), relative chlorophyll content (SPAD), specific leaf area (SLA, cm²/g), leaf tissue density (LTD, g/cm³), water use efficiency (WUE, µmolCO₂/molH₂O), leaf carbon content (LCC, g/kg), leaf nitrogen content (LNC, g/kg), carbon-to-nitrogen ratio (C: N), and equivalent water thickness (EWT, mm). The methods for obtaining basic data on leaf functional traits followed the guidelines of Cornelissen⁴⁶ and Perez-Harguindeguy⁴⁷. After the petiole removed, the fresh weight of the leaves (LFW) was measured via an electronic balance (the accuracy was 0.001 g). The chlorophyll content was assessed at three different positions (upper, middle, and lower) on the leaf using a SPAD-502, with the average value taken as the chlorophyll content. Leaf images were obtained using a CanoScan scanner (the image resolution is 300DPI), and the area of the leaves (LA) was calculated using the image processing software ImageJ. The thickness of each leaf was measured with a caliper at three random positions, with the average value taken as the leaf thickness (LT).

The processed leaves were first subjected to removing green at 105 ℃ and then dried to a constant weight at 65 ℃. Afterward, the leaves were weighed using an electronic balance to obtain the dry weight (LDW). A grinder was used to pulverize the samples, which were then sieved through a 100-mesh screen. The leaf carbon content was determined using the dichromate heating method, whereas the leaf nitrogen content was evaluated using a Kjeldahl nitrogen analyzer. The stable carbon isotope values of the leaves were measured using a stable isotope ratio mass spectrometer.

The calculation formulas for functional traits are as follows. LDMC was calculated by LDW and LFW, referring to Eq. 1, SLA was calculated by LA and LDW, referring to Eq. 2, LTD was calculated by LDW, LT, and LA, referring to Eq. 3, EWT was calculated by LFW, LDW, and LA, referring to Eq. 4.:

1

2

3

4

The WUE was calculated by the relative abundance of ¹³C/¹²C in the sample, referring to Eqs. 5,6, and7, The calculation formula for functional traits are as follows⁴⁸:

5

where R_sam is the relative abundance of ¹³C/¹²C in the sample; and R_std is the relative abundance of ¹³C/¹²C in the internationally recognized standard substance VPDB (Vienna Pee Dee Belemnite)

6

7

where C_a is the atmospheric CO₂ concentration; a is the fractionation coefficient during the diffusion process of CO₂; b is the fractionation coefficient during the carboxylation reaction; 1.6 is the conductivity ratio of stomata to water vapor and CO₂; ∆¹³C is the discrimination ratio of plant carbon isotopes; δ¹³C_p is the carbon isotope ratio of plant leaves; and δ¹³C_a is the carbon isotope ratio of CO₂ in the atmosphere.

Data analysis

All data were expressed as the mean ± standard error (SE). After a normality test was conducting on the data, two-way univariate analysis of variance with provenance and canopy height as factors was used to test for differences in leaf traits between provenances and across canopy levels. The Pearson correlation coefficient was used to assess the correlations among functional traits. The significance level for the tests was set at P = 0.05. We also examined the differences between functional traits at different levels and average canopy traits. All the statistical analyses were conducted using SPSS software (version 22.0, IBM SPSS, Chicago, Illinois), figures are made on https://cnsknowall.com.

Results

Vertical variation patterns of leaf functional traits

Our analysis revealed no significant differences (P > 0.05) in the nine selected leaf functional traits among the different canopy heights (Fig. 1). Overall, LDMC, LTD, WUE, LCC, LNC, C:N, and EWT in the upper canopy were slightly higher than those in the lower and middle canopies. Conversely, SLA was lower in the upper canopy than in the other heights, whereas SPAD values were slightly lower in the lower canopy than in the upper canopy but higher than in the middle canopy. Leaves situated in the middle of the canopy presented the lowest values for LDMC, SPAD, LCC, LNC, C:N, and EWT among the three heights.

Fig. 1.

Leaf functional traits at different canopy heights. Where LDMC, SPAD, SLA, LTD, WUE, LCC, LNC, C:N, and EWT represent leaf dry matter content, relative chlorophyll content, specific leaf area, leaf tissue density, water use efficiency, leaf carbon content, leaf nitrogen content, carbon-to-nitrogen ratio, and equivalent water thickness, respective. Data represent the mean ± S.E., and different lowercase letters above error bars indicate significant different among canopy heights by One-way ANOVA test at the P = 0.05 level.

Differences in leaf functional traits among provenances

We calculated the average leaf functional traits from the upper, middle, and lower canopy layers of the same provenance to represent the leaf functional traits of that provenance, investigating the differences in functional traits among various provenances (Fig. 2). There were no significant differences in LDMC among the five provenances (P > 0.05). The SPAD value of GXYF was significantly lower than those of the other provenances (P < 0.05). The SLA of HBWF and GXYF were higher than other provenances (P < 0.01). JXAF, JXLA, and GXYF presented significantly greater LTD than the other two provenances (P < 0.01). ZJYH has the highest WUE, LCC and LNC among these provenances. With respect of the C: N ratio, significant differences were observed only between provenances JXAF and HBWF (P < 0.05). The EWT of JXLA was the highest.

Fig. 2.

Violin plots showing nine leaf functional traits of different provenances. Where ZJYH, JXAF, JXLA, HBWF, and GXYF represent the provenances from Zhejiang Yunhe, Jiangxi Anfu, Jiangxi Lean, Hubei Wufeng, and Guangxi Yongfu, respectively. Refer to Fig. 1 for trait abbreviations.

The combined effects of provenance and canopy height on leaf functional traits

Statistical analysis (Tables 2 and 3) of the leaf functional traits across the five provenances at three different vertical positions (upper, middle, and lower) in the canopy revealed that LDMC (except for GXYF), SPAD (except for ZJYH), WUE, LCC, EWT (except for ZJYH), and C: N tended to decrease trend with increasing canopy height. In contrast, SLA, LTD, and LNC did not show pattern changes. Provenance had a significant effect on leaf functional traits (SPAD, SLA, LTD, LNC, and EWT) (P < 0.05), whereas the effect of canopy height on these traits was not significant (Table 3).

Table 2.

Leaf functional traits of different provenances at different vertical locations(Means ± SE). u, m, and d represent the locations of the upper, middle and lower of canopy heights, respectively. Refer to Fig. 1 for trait abbreviations. Refer to Fig. 2 for provenances abbreviations.

	LDMC(g/kg)	SPAD	SLA(cm2/g)	LTD(g/cm3)	WUE(µmol/mol)	LCC(g/kg)	LNC(g/kg)	C: N	EWT(mm)
ZJYHu	408.43 ± 22.38a	71.67 ± 7.29a	62.19 ± 10.69a	0.33 ± 0.04a	3.99 ± 0.05a	598.98 ± 41.91a	13.15 ± 0.14c	45.57 ± 3.64a	0.23 ± 0.02b
ZJYHm	402.87 ± 3.64a	69.63 ± 7.37a	60.05 ± 1.54a	0.34 ± 0.01a	3.88 ± 0.05b	554.16 ± 14.45a	13.66 ± 0.07b	40.57 ± 0.92b	0.24 ± 0.01b
ZJYHd	381.44 ± 4.75a	72.2 ± 2.44a	54.94 ± 2.49a	0.32 ± 0.02a	3.84 ± 0.03b	555.69 ± 18.93a	15.93 ± 0.26a	34.90 ± 1.75c	0.29 ± 0.01a
JXAFu	411.04 ± 22.70a	77.27 ± 1.82a	45.99 ± 3.44b	0.43 ± 0.03a	3.90 ± 0.06a	558.02 ± 54.83a	12.15 ± 0.63b	46.02 ± 5.50b	0.31 ± 0.05ab
JXAFm	369.86 ± 16.70a	70.00 ± 3.50ab	52.93 ± 4.31b	0.44 ± 0.02a	3.70 ± 0.03b	564.46 ± 11.45a	10.28 ± 0.24c	54.92 ± 1.38a	0.32 ± 0.04a
JXAFd	380.05 ± 35.73a	54.50 ± 12.75b	69.71 ± 8.33a	0.45 ± 0.09a	3.49 ± 0.14c	504.88 ± 4.32a	14.12 ± 0.94a	35.84 ± 2.16c	0.24 ± 0.03b
JXLAu	396.73 ± 2.98a	75.70 ± 0.82a	29.72 ± 4.78b	0.63 ± 0.10a	4.08 ± 0.03a	551.14 ± 8.64a	13.42 ± 0.32a	41.09 ± 1.54b	0.52 ± 0.08a
JXLAm	387.46 ± 12.45a	61.97 ± 12.04ab	47.58 ± 15.89b	0.5 ± 0.16ab	3.76 ± 0.04b	560.66 ± 11.24a	11.44 ± 0.53b	49.12 ± 3.18a	0.36 ± 0.13ab
JXLAd	369.29 ± 37.89a	46.23 ± 8.42b	86.92 ± 10.57a	0.37 ± 0.02b	3.04 ± 0.16c	496.14 ± 17.07b	12.32 ± 1.15ab	40.43 ± 2.49b	0.20 ± 0.03b
HBWFu	428.26 ± 19.30a	74.93 ± 0.76a	71.39 ± 4.62a	0.31 ± 0.02a	3.68 ± 0.05a	566.73 ± 2.73a	14.09 ± 0.24a	40.22 ± 0.67a	0.19 ± 0.01a
HBWFm	403.25 ± 19.28a	67.70 ± 4.42a	96.89 ± 8.91a	0.25 ± 0.02a	3.59 ± 0.17a	542.26 ± 37.84a	13.14 ± 0.35b	41.25 ± 2.39a	0.15 ± 0.01a
HBWFd	403.75 ± 52.11a	31.07 ± 9.86b	91.68 ± 2.89b	0.35 ± 0.10a	3.08 ± 0.18b	493.91 ± 11.17b	14.47 ± 0.26a	34.14 ± 0.70b	0.16 ± 0.03a
GXYFu	373.91 ± 33.11b	54.50 ± 5.62a	86.72 ± 29.54a	0.41 ± 0.09a	3.88 ± 0.02a	532.08 ± 27.95ab	10.82 ± 0.42c	49.27 ± 4.05a	0.21 ± 0.05a
GXYFm	350.33 ± 6.43b	47.77 ± 2.38a	92.79 ± 13.81a	0.39 ± 0.05a	3.75 ± 0.03a	536.98 ± 9.34a	12.50 ± 0.14b	42.97 ± 0.31b	0.20 ± 0.03a
GXYFd	474.75 ± 35.70a	30.13 ± 7.95b	80.86 ± 13.06a	0.52 ± 0.03a	3.08 ± 0.14b	492.89 ± 19.69b	15.25 ± 0.57a	32.38 ± 2.49c	0.14 ± 0.01a

Table 3.

Leaf functional traits of different provenances at different vertical locations.

Indicators	LDMC(g/kg)		SPAD		SLA(cm2/g)		LTD(g/cm3)		WUE(µmol/mol)		LCC(g/kg)		LNC(g/kg)		C: N		EWT(mm)
	F	P	F	P	F	P	F	P	F	P	F	P	F	P	F	P	F	P
Height	0.968	0.391	0.265	0.769	0.514	0.603	0.068	0.935	0.142	0.868	0.704	0.503	0.014	0.986	0.106	0.9	0.011	0.989
Provenances	0.692	0.603	4.034	< 0.01	8.447	< 0.001	7.555	< 0.001	1.955	0.127	2.391	0.073	2.813	0.043	1.35	0.274	6.956	< 0.001
Provenances×Height	0.253	0.976	0.231	0.982	0.981	0.47	0.613	0.76	0.053	0.999	0.851	0.566	0.064	0.999	0.17	0.993	0.187	0.991

Relationships among leaf functional traits

Overall, 50% of the leaf functional traits presented significant correlations (Fig. 3A). Notably, C:N was significantly correlated with LDMC and the EWT at the P < 0.05 level, whereas the remaining traits were displayed highly significant correlations (P < 0.01). Differences in relationships exist among traits along the vertical gradient (Fig. 3B ~ 3D). Consistent relationships were found between SPAD and WUE, SPAD and EWT, as well as LNC and C: N. Although the relationships among SPAD and LCC, LTD and EWT, and WUE and LCC also remained unchanged, their significance levels varied.

Fig. 3.

Relationships among leaf functional traits at (A) overall canopy, (B) upper canopy, (C) middle canopy and (D) lower canopy. Refer to Fig. 1 for trait abbreviations. * indicates a significance level of P < 0.05, ** indicates a significance level of P < 0.01.

In contrast to the overall results, the correlations of the leaf functional traits in the middle canopy were generally consistent with the overall findings, whereas the correlations in the lower canopy differed considerably from the overall results.

Discussions

Impact of canopy height on variability in leaf functional traits

Research suggests that variation in leaf functional traits may be influenced by the leaf micro-environment, which may be affected by canopy structure^49–51. Furthermore, leaf variation at the canopy level can be influenced by multiple variables, such as light availability, hydraulic limitations, or leaf age²⁶. The high plasticity of leaf traits is crucial for making full use of available environmental resources. In this study, we demonstrated that the impact of canopy height on leaf functional traits in Chinese fir plantations is limited (Fig. 1). This result, however, is not consistent with some researches that have revealed significant differences in leaf traits at varying heights within the canopy^52–54. The canopy structure and leaf morphology of Chinese fir appear to be the main reasons for this result.

Owing to the canopy structure is approximately conical with relatively larger inter-leaf gaps, leaves located in the middle and lower sections of the canopy still receive effective light radiation^55,56. Moreover, lower-canopy leaves exhibit higher quantum-use efficiency under low light, which can compensate for reduced irradiance and diminish vertical differences⁵⁷, resulting in limited variation in traits across canopy heights. The needles of Chinese fir are arranged in dense whorls, are small and thick, and have well-developed cuticles. This morphology constrains rapid leaf responses to micro-environmental changes, leading to limited within-canopy trait variation⁵⁸. Previous research indicates that as canopy height increases, the specific leaf area (SLA) tends to decrease, while leaf thickness (LTD) tends to increase^59–61; we found similar results in this study. As shown in Fig. 1, leaves in the upper layer, which are exposed to stronger light, typically adopt strategies such as increased thickness, larger leaf area, and stomatal closure to mitigate water loss^62,63, resulting in lower SLA and higher equivalent water thickness (EWT). Some studies showed that shade leaves have 43.3% higher SLA than sun leaves²⁶, in this study, this gap was reduced to 8.57%, which mainly because of the experiment conducted within one species in a common garden. Some studies also found that LDMC did not show significant changes with the increase of canopy height^64,65, which is consistent with our results, but is inconsistent with some previous research conclusions^66,67, which may because the variation of functional traits may influenced by sampling time, sampling location, and plant selection⁶⁴. Furthermore, upper-conapy leaves of the canopy must survive high radiation and evaporation, achieving a balance between carbon fixation and water consumption through stomatal regulation⁶², leading to higher water use efficiency than that in lower layers. The functional traits closely related to photosynthetic rates, such as leaf nitrogen content and SPAD values, also vary with height⁶². Yang et al.⁶⁸ found that with increasing tree height, Cunninghamia lanceolata exhibits hydraulic limitations that lead to decline in photosynthetic rate, transpiration rate, and stomatal conductance in the upper-crown leaves. Leaf nitrogen content did not show a consistent change pattern with vertical height, which may be due to the comprehensive influence of factors such as light intensity, water content, wind speed, nutrient utilization mode, plant characteristics, and soil nitrogen and phosphorus content^69,70. Additionally, severe nitrogen deposition in subtropical regions of China may be an important factor affecting the results⁷¹.

Impact of provenances on variability in leaf functional traits

In this study, we find that intraspecific differences occur in leaf functional traits among provenances, which is also demonstrated in the previous study that leaves exhibit considerable plasticity³⁴. When different genetic provenances are transplanted into a common garden, the environmental heterogeneity of the provenance location results in genetic effects reflected in functional traits^72,73.

The phenotypic plasticity of plants is closely related to the environment and represents an adaptation to environmental conditions throughout evolution. Tree after long-term growth in different ecological environments undergo adaptive geographic variation to the local habitat, forming different geographic provenances³⁹. In this study, LDMC, WUE, and LCC did not have significant regulatory effects on the provenances, primarily because provenances ZJYH, JXAF, JXLA, and HBWF are all located in provenance zone VII⁴⁵. The zones are classified on the basis of the growth parameters of various provenances, and the resource utilization strategies within the same zone are relatively similar⁴⁵, leading to minimal differences. As shown in Fig. 2 and Table 2, ZJYH presented relatively high LCC, LNC, WUE, and SPAD values, indicating strong photosynthetic capacity. HBWF has a higher SLA and the highest LDMC but the lowest LTD, WUE, and EWT, indicating that this provenance is water-intensive and high-yielding, with strong resource acquisition capability but poor drought stress resistance.

Research showed that variation in plant leaf functional traits is not only related to genetic diversity but also influenced by habitat heterogeneity^69,74. Environmental factors at provenances location can significantly affect the variation in the leaf functional traits of Chinese fir provenances. After long-term geographical isolation and natural selection, different provenances of Chinese fir have formed, with nearly one-third of the variation still explained by the hydrothermal conditions of the location of provenances even after they were transplanted into a homogeneous garden environment^72,73. Therefore, exploring intraspecific variation in plant functional traits helps capture the full spectrum of functional diversity^2,12, improving predictions of vegetation responses under future climate change.

Correlations among leaf functional traits

As shown in Fig. 3, correlations among traits differed across the upper, middle, and lower canopy layers, reflecting the ecological strategies by which plant leaves adapt to varying environmental factors^14,17 such as light and moisture at different canopy heights^25,26.

The synergistic associations among leaf functional traits indicate the survival strategies of plants⁴ Research has revealed a significant negative correlation between SLA and EWT, reflecting trade-offs between light capture and water management⁷⁵. In subtropical regions characterized by intense sunlight and rapid evaporation, plants tend to allocate more dry matter to guard cells⁵⁹, resulting in a reduced SLA and increased leaf thickness or density. This adaptation fosters efficient water storage. The SPAD values were positively correlated with the WUE, LCC, and C: N ratios. Since SPAD values reflect a plant’s photosynthetic capacity, Chinese fir, which has high photosynthetic ability, assimilates more carbon dioxide through photosynthesis while regulating stomatal conductance to minimize water loss, which enhances its WUE. Furthermore, more carbon is allocated to leaf tissue construction, resulting in a higher nitrogen utilization rate⁷⁶, which is consistent with previous research. In this study, we found that the correlations of leaf functional traits in the middle canopy layer were similar to the entire canopy (Fig. 3), further confirming the feasibility of collecting samples from middle canopy leaves to examine the variation patterns of functional traits in subtropical Cunninghamia lanceolata forests.

Implications

In this study, we collected leaf samples from various geographical provenances of Chinese fir at different canopy heights and measured relevant functional traits. These findings confirm that the impact of canopy height on leaf functional traits of Chinese fir plantations is limited. This limitation is due to the leaf morphology and canopy structures⁵⁶, which allow some leaves in the middle and lower canopies to receive adequate light. However, significant intraspecific variation in functional traits is evident, influenced by genetic background⁷⁰. Notably, some functional traits exhibited minor differences among provenances, which diverges from the findings of previous studies. The age of the leaves and their position on the branches (which determines the distance of water transmission) all affect the photosynthetic capacity, which also have influence on functional traits. Moreover, this study focused on the variation of functional traits among provenances but sampled only 3 individuals per provenances, which may affect the results robustness. Future studies should increase the sample size to investigate the patterns of intraspecific variation.

Additionally, this study revealed that leaf samples from the middle canopy can effectively represent the functional traits of Chinese fir at the whole-canopy level. Regardless of whether single trait values or the relationships among multiple traits are considered, the differences are minimal compared with the overall average values across different canopy heights. These findings have significant implications for understanding the ecological functions of Chinese fir plantations in subtropical regions.

Conclusions

The variation in canopy height has a limited effect on Chinese fir leaf functional traits, with most of the functional traits variation attributed to the differences of provenances. Overall, 50% of the leaf functional traits presented significant correlations, which were associated mainly with photosynthetic capacity-related traits. The correlations among traits at the middle part of the canopy are consistent with the average canopy values. These results provide useful insights about adaptation of functional traits of Chinese fir to canopy mircoenvironment.

Author contributions

Tingyu Xu: Conceptualization, Data curation, Formal analysis, Writing-original draft, Writing-review & editing. Xiang Niu: Conceptualization, Investigation, Writing-review & editing. Bing Wang: Investigation, Project administration. Yihui Wang: Data curation.

Funding

This study was supported by the the Central Non-profit Research Institution of CAF, grant number CAFYBB2020ZE003, and Jiangxi Dagangshan National Key Field Observation and Research Station for Forest Ecosystem, grant number 2060204-25-201.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.