Soil Comprehensive Fertility Changes in Response to Stand Age and Initial Planting Density of Long-Term Spacing Trials of Chinese Fir Plantations

Abstract

The growing demand for wood products and ecosystem services in Chinese fir plantations has led to longer rotation ages and density control practices, raising concerns about their impact on soil fertility. This study assessed soil fertility of Chinese fir plantations in Fujian, Jiangxi, and Sichuan Provinces using the Nemerow index. The effects of stand age and initial planting density on soil fertility were analyzed using statistical models. In Fujian and Jiangxi, soil fertility was significantly higher at 11 and 30 years than at 5 and 25 years, while in Sichuan, it was higher at 25 and 30 years than at 5 and 11 years. In Fujian, soil fertility was higher at 6667 trees ha⁻² than at 1667 trees ha⁻². No significant differences were observed in Jiangxi, while in Sichuan, soil fertility at 6667 trees ha⁻² was significantly higher than at 5000 and 1667 trees ha⁻², and soil fertility at 10,000 trees ha⁻² exceeded that at 1667 trees ha⁻². Soil fertility typically increased with stand age, especially in Fujian and Sichuan. Soil fertility also increased with initial planting density, especially in Jiangxi and Sichuan. A structural equation model (SEM) explained 88% of the variance in soil fertility, with stand age directly affecting soil fertility and soil organic matter mediating the effects of stand age and planting density. These findings suggest that adjusting rotation age and planting density could help improve soil fertility, offering practical implications for sustainable forest management in Chinese fir plantations.

1. Introduction

As demand for wood production and ecosystem services increases, intensive forest management has become a global trend [1]. While forest management supports tree growth, forest carbon stocks, and forest biodiversity [2,3,4], improper practices can disrupt soil structure, reduce organic matter content, and deplete soil fertility [5,6]. Soil fertility degradation has become a major issue in global forests [7,8,9], with recent studies emphasizing the need for sustainable management practices to mitigate soil health decline. For instance, rotation breaks are essential for maintaining the organism community of soil [10], and logging residues aid in restoration of soil health after clear-cutting [11]. Additionally, balanced fertilizer management can also improve soil health in plantations [12].

Rotation period and density control are key plantation management strategies [13]. The length of the rotation period determines the cycling and restoration of soil nutrients [14]. Optimal rotation periods can harness the decomposition of tree litter to return organic matter to the soil, facilitate nutrient cycling within the plantation ecosystem, and prevent excessive depletion of soil nutrients [15]. Initial planting density influences plant growth, stand structure, and intra-specific competition [16,17]. It alters the intensity of competition for soil nutrients and the amount of litter produced by stands [18,19], thereby affecting the utilization and accumulation of soil nutrients. Therefore, stand age and initial planting density play crucial roles in sustainable soil resource management in forestry.

Comprehensive evaluations of soil fertility provide reliable indicators for managing rotation age and planting density. These indicators include soil organic matter, soil mineral elements such as nitrogen, phosphorus, and potassium, and soil pH, which considerably affect plant development and are commonly used to assess soil status [20]. The recycling of nutrients from soil organic matter is essential for maintaining soil fertility [21,22]. Mineral elements, such as nitrogen, phosphorus, and potassium, are crucial macro- and micronutrients provided to plants through fertilizers [23]. Soil pH is a critical factor in soil science, directly influencing plant growth [24]. The Nemerow index method comprehensively evaluates the soil comprehensive fertility index (SCFI) by combining multiple nutrient indicators, effectively reducing subjective influences and identifying the least limiting nutrient for a thorough soil fertility assessment [25,26].

Chinese fir (Cunninghamia lanceolata (Lamb.) Hook. f.), a species in the Cupressaceae family, a key timber species in southern China with over a millennium of planting history and covering 10 million hectares [27,28], has experienced soil fertility depletion in forest management [29]. The impact of various forest management practices on soil microbial communities, plant growth, understory diversity, biomass, and stand spatial structure has been extensively studied in Chinese fir plantations [30,31,32]. Research has also focused on the effects of stand age and density on soil nutrients in Chinese fir plantations, showing that soil fertility is higher at the 7-year stage than at the 15-year and 25-year stages [33] and that middle-aged plantations (12 years) have higher soil organic carbon than young (6 years) and mature (25 years) plantations [34]. Increased density reduces soil fertility in Chinese fir stands [35]. Actual density often differs considerably from initial planting density due to density-dependent mortality during forest self-thinning [36]. While previous studies have mainly focused on the effects of actual stand density, our research stands out by examining the influence of initial planting density, which has been relatively underexplored. Additionally, prior studies often rely on spatial data as a proxy for temporal data and tend to use experimental sites with inconsistent ecological conditions, introducing potential bias. In contrast, this study is based on long-term trials in fixed experimental plots, providing a consistent and reliable dataset, and offering more accurate insights into the relationship between stand age, initial planting density, and soil fertility.

Investigating the variations in soil fertility concerning stand age and initial planting density provides a theoretical basis for implementing scientific management practices in Chinese fir plantations. This research aims to (1) assess how stand age and initial planting density affect the SCFI in Chinese fir plantations in three provinces, (2) investigate their effects on the SCFI and individual soil fertility factors in subtropical Chinese fir plantations, and (3) examine how stand age and initial planting density influence the SCFI through various fertility factors. This study seeks to enhance understanding of the effects of stand age and initial planting density on soil fertility in Chinese fir plantations and to provide a theoretical basis for implementing scientific management practices.

2. Materials and Methods

2.1. Study Sites

This study was conducted at three sites (Figure 1). The first site was located in the Weimin Town National Forest Farm, Shaowu City, Fujian Province (27°05′ N, 117°43′ E), with an elevation of 500 m above sea level. The soil type at this site is red soil. The second site was located at Nianzhu Forest Farm, Fenyi County, Xinyu City, Jiangxi Province (27°34′ N, 114°33′ E), at an elevation of 500 m above sea level, with red soil. The third site was situated on a forest farm in Minsheng Village, Luzhou City, Sichuan Province (28°15′ N, 105°20′ E), at an elevation of 440 m above sea level, with red soil. All sites were located in southern China and had a subtropical monsoon climate characterized by hot summers, mild winters, an average temperature above 22 °C in the hottest month, an average temperature between 0 °C and 15 °C in the coldest month, and annual precipitation ranging from 800 to 1600 mm.

The Chinese fir plantations were established in the spring of 1981 using 1-year-old bare-root seedlings from the planting site. The area was originally forested with native vegetation. All sample plots underwent site preparation prior to afforestation, which involved burning the underbrush and clearing the land. Five initial planting densities were applied at each study site: 1667, 3333, 5000, 6667, and 10,000 trees ha⁻², with corresponding plant spacings of 2.0 m × 3.0 m, 2.0 m × 1.5 m, 2.0 m × 1.0 m, 1.0 m × 1.5 m, and 1.0 m × 1.0 m. Three sample plots were established for each density. The experimental design used a randomized block layout to minimize variability and ensure statistical reliability. Each plot was randomly assigned to the five planting densities, and blocks were systematically arranged across the study area to account for environmental heterogeneity.

2.2. Sample Collection and Analysis

Soil samples were collected in October 1985, 1991, 2005, and 2010, within a 30-year rotation cycle that balances sustaining forest productivity and long-term soil health. A diagonal sampling method was used to collect surface soil samples (0–40 cm), with three sampling points along the diagonal combined from each plot. Each region contained 15 composite samples. After the soil samples were brought to the laboratory, they were dried, and the gravel, plant roots, and other debris were removed before being filtered through a 100-mesh nylon sieve. Each sample was measured three times, and the average value was calculated. Soil organic matter (SOM), total nitrogen (TN), available nitrogen (AN), total phosphorus (TP), available phosphorus (AP), total potassium (TK), available potassium (AK), and pH were determined using various methods. The methods considered were the potassium dichromate method, Kjeldahl digestion, potassium chloride extraction, near-infrared spectroscopy, sodium bicarbonate extraction with molybdenum-antimony resistance colorimetry, sodium hydroxide alkali melting, flame photometry, and the glass electrode method [37]. The measured soil fertility indicators, including SOM, TN, AN, TP, AP, TK, AK, and pH (

Table 1. Descriptive statistics of different soil fertility indicators for Chinese fir plantations.

Item	Min	Max	Mean	SD	CV/%
SOM g·kg−1	8.51	50.81	25.88	7.94	30.68
TN g·kg−1	0.32	2.35	0.98	0.34	34.52
AN mg·kg−1	9.50	286.91	89.68	49.03	54.67
TP g·kg−1	0.02	0.97	0.29	0.17	60.30
AP mg·kg−1	0.10	9.38	1.53	1.49	97.24
TK g·kg−1	1.30	35.10	12.35	8.28	67.03
AK mg·kg−1	5.99	214.07	58.48	36.81	62.94
pH	3.44	5.53	4.30	0.42	9.83

Note: Min is the minimum value; Max is the maximum value; SD is the standard deviation; CV is the coefficient of variation; SOM is the soil organic matter; TN is the total nitrogen; AN is the available nitrogen; TP is the total phosphorus; AP is the available phosphorus; TK is the total potassium; AK is the available potassium; pH is the potential of hydrogen.

), were used to assess the SCFI, providing a comprehensive understanding of the overall fertility status of the soil.

2.3. Assessment of Soil Fertility

2.3.1. Assessment of a Single Index of Soil Fertility

Grading standards for various soil fertility indices in Chinese fir plantations were established (

Table 2. Classification criteria for soil fertility indicators in the improved Nemerow index.

Soil Properties	Grade	SOM g/kg	TN g/kg	AN mg/kg	TP g/kg	AP mg/kg	TK g/kg	AK mg/kg	pH
Classification index for Nemerow	xe	40	2	150	1.0	40	25	200	8.5
	xd	30	1.5	120	0.8	20	20	150	7.5
	xc	20	1	90	0.6	10	15	100	6.5
	xb	10	0.75	60	0.4	5	10	50	5.5
	Xa	6	0.5	30	0.2	3	5	30	4.5

Note: as defined in Table 1.

). This grading was based on the second soil census standard of China (National Earth System Science Data Center, 2005, http://gre.geodata.cn [accessed on 24 January 2025]) (

Table 3. Classification criteria for soil fertility indicators in the second soil census of China.

Grade	SOM g/kg	TN g/kg	AN mg/kg	TP g/kg	AP mg/kg	TK g/kg	AK mg/kg	pH	Description
1	>40	>2	>150	>1	>40	>25	>200	8.5–9	Extremely
2	30–40	1.5–2	120–150	0.8–1	20–40	20–25	150–200	7.5–8.5	Rich
3	20–30	1–1.5	90–120	0.6–0.8	10–20	15–20	100–150	6.5–7.5	Medium
4	10–20	0.75–1	60–90	0.4–0.6	5–10	10–15	50–100	5.5–6.5	Moderately inferior
5	6–10	0.5–0.75	30–60	0.2–0.4	3–5	5–10	30–50	4.5–5.5	Poor
6	<6	<0.5	<30	<0.2	<3	<5	<30	<4.5	Very poor

Note: as defined in Table 1.

) [25]. Figure 2 provides an overview of the data analysis steps and methodology used in this study.

2.3.2. Assessment of Soil Comprehensive Fertility

The improved Nemerow index was used to calculate the SCFI. The formula for the improved Nemerow index is as follows:

$$F(i)=\sqrt{{\displaystyle \frac{{Fiave}^2+{Fimin}^2}{2}}}·\left({\displaystyle \frac{n-1}{n}}\right)$$

where F(i) represents the SCFI value, Fiave is the mean value of an individual soil fertility indicator, Fimin is the minimum value of fertility indicators, i denotes the serial number of sampling sites, and n represents the total number of involved indicators. In this improved formula, Fimin replaces Fimax with the original Nemerow index to emphasize the influence of the minimum soil fertility indicator on plant growth. Additionally, the term $\left({\displaystyle \frac{n-1}{n}}\right)$ was added to highlight the change in calculation results when the number of indicators varied.

To eliminate dimensional effects, soil fertility indicators were standardized. Standardization was performed according to the criteria established based on the second soil census standard of China (Table 3). The standardization method varied at different index (ci) values as in Equations (1)–(5).

ci ≤ xa, Fi = ci/xa, (Fi ≤ 1)

(1)

xa < ci ≤ xb, Fi = 1 + (ci − xa)/(xb − xa), (1 < Fi ≤ 2)

(2)

xb < ci ≤ xc, Fi = 2 + (ci − xb)/(xc − xb), (2 < Fi ≤ 3)

(3)

xc < ci ≤ xd, Fi = 3 + (ci − xc)/(xd − xc), (3 < Fi ≤ 4)

(4)

xd < ci, Fi = 4 + (ci − xd)/(xe − xd),(4 < Fi)

(5)

where Fi represents the standardized value for each indicator, ci is the measured value, and xa, xb, xc, xd, and xe represent the classification thresholds.

2.4. Data Statistics and Analysis

A two-way ANOVA was performed to evaluate the effects of stand age and initial planting density on the SCFI across three provinces. If an interaction effect was observed, a simple main effect analysis was conducted to examine the influence of each factor at different levels of the other factor. If no interaction was detected, the significance of the main effects was evaluated, and the impact of each factor on the SCFI was analyzed. Fisher’s LSD post hoc test, appropriate for 3–5 groups, was applied to the main and simple main effects [38]. This study included four stand age groups and five initial planting density groups, satisfying the criteria for Fisher’s LSD post hoc test.

Second, a linear mixed-effects model was used to assess the effects of stand age and initial planting density on the SCFI and soil fertility indicators in subtropical Chinese fir plantations. The linear mixed-effects model is appropriate for handling hierarchical data from 45 sample plots across three provinces, with model fitting conducted using the “lmer” function from the “lmerTest” package [39]. The model included stand age and initial planting density as fixed effects. Given the differences in soil and climate among provinces, as well as variations within provinces due to factors such as elevation and slope, two models were constructed: Model 1, with random effects for study sites and sample plots within each site, and Model 2, with random effects for sample plots only. The likelihood ratios of these two models were compared using the “Anova” function to assess the impact of provincial heterogeneity on the SCFI and soil fertility factors. If the p-value of the likelihood ratio was less than 0.05, provincial factors were considered to considerably affect the SCFI. The model was considered optimal, including random effects for the study sites. The dataset was divided into subsets by province, and linear fixed-effect models were constructed for each subset to explore regional differences in response. The coefficients, directions, and statistical significance of stand age and initial planting density were compared across models to assess regional differences. The formulas are presented below:

$$Y_{ij}={\beta }_0+{\beta }_1{X}_{1ij}+{\beta }_2{X}_{2ij}+u_{0i}+u_{1ij}+{ϵ}_{ij}$$

where Y_ij represents the response variable, denoting the observed values of the SCFI and various soil fertility indicators for the jth sample plot in the ith province. X_1ij and X_2ij represent the stand age and initial planting density, respectively, for the jth sample plot at the site. u_0i represents the random effect of the research site. u_1ij represents the random effect among different sample plots within the same site. β₀ is the overall intercept, and β₁ and β₂ are the fixed slope coefficients corresponding to stand age and initial planting density, respectively.

Finally, structural equation models (SEMs) were used to analyze the standardized direct, indirect, and total effects of SCFI concerning stand age and initial planting density. The SEMs were analyzed using a combined dataset from three provinces and separate datasets for each province. In constructing the model, the close associations of SOM, soil mineral nutrients (including TN, AN, TK, AK, TP, and AP), and pH with soil health and plant growth were considered, and these variables were designated as mediators. Direct effects were quantified using standardized regression coefficients that link independent and dependent variables, while indirect effects were obtained by multiplying the coefficients along the indirect pathways within the model. The total effect was computed as the sum of direct and indirect effects. The SEM analysis was conducted using the “plspm” package [40]. Previous research recommends eliminating variables with factor loadings below 0.4, as this approach reduces complexity and improves model clarity, particularly in cases with small sample sizes and numerous variables [41].

3. Results

3.1. Effects of Stand Age, Initial Planting Density, and Region on SCFI

3.1.1. Effects of Stand Age on SCFI

The two-way ANOVA results indicated that stand age considerably affected the SCFI in the three provinces (F = 3.543, p < 0.001 for Fujian Province; F = 13.853, p < 0.001 for Jiangxi Province; F = 3.343, p < 0.001 for Sichuan Province). In contrast, the interaction between stand age and initial planting density had no significant effect on the SCFI (F = 0.385, p = 0.980 for Fujian Province; F = 0.928, p = 0.544 for Jiangxi Province; F = 0.680, p = 0.829 for Sichuan Province) (

Table 4. Summary of two-way ANOVA for the effects of five initial planting densities (I) and four stand ages (S) on the SCFI.

Parameters	Stand Age (S)		Initial Planting Density (I)		S × I
	F	p	F	p	F	p
Fujian Province	3.543	***	14.103	*	0.385	0.980
Jiangxi Province	13.853	***	0.825	0.516	0.928	0.544
Sichuan Province	6.343	***	29.684	***	0.680	0.829

Note: The p-values represent the significance of the statistical tests. Significant p-value at: * p < 0.05; *** p < 0.001. F = calculated F value.

). As no interaction effect was observed, a main effect analysis was conducted on stand age. The main effect analysis revealed that in Fujian and Jiangxi Provinces, the mean SCFI values at 11 years (1.71 and 1.74) and 30 years (1.71 and 1.67) were significantly higher than those at 5 years (1.16 and 1.46) and 25 years (1.40 and 1.37) (Figure 3a,c). In contrast, in Sichuan Province, SCFI values at 25 years (1.10) and 30 years (1.06) were significantly higher than those at 5 years (0.86) and 11 years (0.81) (Figure 3e). Further, the likelihood ratio analysis of the two linear mixed-effects models indicated that Model 1, which includes the study site factor, provided a better fit than Model 2 (

Table 5. Likelihood ratio test between two mixed-effects models with different random effects structures.

Response Variable	AIC		BIC		logLik		Pr (>Chisq)
	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
SCFI	3.14	72.02	22.30	87.98	4.43	−31.01	***
SOM	1259.20	1258.60	1278.30	1274.60	−623.59	−624.30	0.23
TN	10.77	37.01	29.93	52.97	0.62	−13.50	***
AN	1935.40	1943.90	1954.60	1959.90	−961.72	−966.96	**
TP	−204.76	−143.51	−185.60	−127.54	108.38	76.75	***
AP	1026.30	1059.60	1045.50	1075.50	−507.17	−524.79	***
TK	1174.40	1197.80	1193.60	1213.70	−581.22	−593.88	***
AK	1761.50	1804.30	1780.60	1820.20	−874.73	−897.14	***
pH	165.92	198.95	185.08	214.91	−76.96	−94.47	***

Note: As defined in Table 1 and Figure 2. Significant p-values are indicated as follows: ** p < 0.01; *** p < 0.001.

). This result indicates that Model 1 provides a more accurate representation of the effects of the SCFI. The linear mixed-effects model (Model 1) for the complete sample set revealed that stand age had a significant positive impact on the SCFI (β = 0.01 **, 95% CI = 0.00, 0.02), as well as on the soil fertility factors AN (β = 1.57 ***, 95% CI = 0.84, 2.30), TK (β = 0.47 ***, 95% CI = 0.38, 0.56), and AK (β = 0.91 ***, 95% CI = 0.47, 1.35) (Figure 4a,d,g,h). Conversely, stand age had a significant negative impact on the soil fertility factors TP (β = −0.01 **, 95% CI = 0.00, −0.02) and AP (β = −0.23 **, 95%CI = −0.29, −0.18) (Figure 4e,f).

3.1.2. Effects of Initial Planting Density on SCFI

The two-way ANOVA results indicated that initial planting density significantly affected the SCFI in Fujian and Sichuan Provinces (F = 14.103, p < 0.05 for Fujian Province; F = 29.684, p < 0.001 for Sichuan Province), but had no significant effect in Jiangxi Province (F = 0.825, p = 0.516 for Sichuan Province) (Table 4). As no significant interaction effect was observed between initial planting density and stand age, the analysis focused on the main effect of initial planting density on the SCFI. In Fujian Province, SCFI was significantly higher at a planting density of 6667 trees per hectare (1.63) compared to 1667 trees per hectare (1.30) (Figure 3b). In Sichuan Province, the SCFI at 6667 trees per hectare (1.07) was significantly higher than at 5000 trees per hectare (0.95) and 1667 trees per hectare (0.92). Additionally, the SCFI at 10,000 trees per hectare (1.05) was significantly higher than at 1667 trees per hectare (0.92) (Figure 3f). In Jiangxi Province, the mean SCFI values at densities of 1667, 3333, 5000, 6667, and 10,000 trees per hectare were 1.49, 1.53, 1.47, 1.55, and 1.57, respectively, with no significant difference observed between initial planting densities (Figure 3d). The linear mixed-effects model (Model 1) revealed that initial planting density significantly and positively influenced the SCFI (β = 0.02 **, 95% CI = 0.01, 0.03) and the soil fertility factors SOM (β = 0.60 **, 95% CI = 0.20, 1.00), TN (β = 0.02 **, 95% CI = 0.01, 0.03), and TP (β = 0.01 *, 95% CI = 0.00, 0.02) (Figure 5a–c).

3.1.3. Effects of the Region on SCFI

Model 1 showed that Jiangxi Province had the highest random intercepts for the SCFI, SOM, TN, AN, AP, and TK across the three study sites, followed by Fujian, with Sichuan having the lowest intercepts. Fujian ranked first for soil TP, AK, and pH intercepts, followed by Jiangxi, with Sichuan ranking last (

Table 6. Random intercepts for study sites across three provinces.

Response Variable	Sampling Sites
	Fujian	Jiangxi	Sichuan
SCFI	0.16	0.22	−0.38
SOM	−0.46	1.49	−1.03
TN	0.00	0.15	−0.16
AN	5.62	14.75	−20.37
TP	0.11	0.06	−0.18
AP	−0.23	2.63	−2.39
TK	0.65	2.88	−3.53
AK	24.79	7.62	−32.41
pH	0.17	0.12	−0.29

Note: As defined in Table 1 and Figure 2.

). The likelihood ratio test indicated that the region significantly affected the SCFI and soil nutrient factors (TN, AN, TP, AP, TK, AK, and pH) (Pr (>Chisq) < 0.05) (Table 5). Sensitivity analysis showed that stand age significantly positively affected the SCFI in Fujian (β = −0.01 ***, 95% CI = 0.00, 0.02) and Sichuan (β = 0.01 ***, 95% CI = 0.00, 0.02) but not in Jiangxi (β = 0.00, 95% CI = −0.01, 0.01). Initial planting density had a significant positive effect on the SCFI in Jiangxi (β = 0.02 *, 95% CI = 0.00, 0.04) and Sichuan (β = 0.02 ***, 95% CI = 0.01, 0.03) but not in Fujian (β = 0.02, 95% CI = −0.01, 0.05) (

Table 7. Sensitivity analysis evaluating the consistency of results across provinces using linear fixed-effects models.

Dependent Variable	Independent Variable	Region
		FJ	JX	SC
SCFI	standage	0.01 *** (0.00, 0.02)	0.00 (−0.01, 0.01)	0.01 *** (0.00, 0.02)
	Initial planting density	0.02 (−0.01, 0.05)	0.02 * (0.00, 0.04)	0.02 *** (0.01, 0.03)
SOM	standage	0.03 (−0.09, 0.15)	−0.04 (−0.20, 0.12)	−0.01 (−0.26, 0.24)
	Initial planting density	0.601 ** (0.199, 1)	0.592 * (0.0226, 1.16)	0.653 * (−0.195, 1.5)
TN	standage	0.00 (0.00, 0.00)	0.00(0.00, 0.01)	0.01 * (0.00, 0.02)
	Initial planting density	0.02 (0.00, 0.04)	0.02 (0.01, 0.04)	0.02 (0.00, 0.04)
AN	standage	1.57 *** (0.81, 2.33)	2.39 ** (1.03,3.75)	2.05 *** (1.43, 2.66)
	Initial planting density	1.45 (−1.25, 4.14)	1.47 (−3.33,6.28)	1.79 (−0.37, 4.00)
TP	standage	−0.01 *** (−0.01, 0.00)	−0.01 *** (0.01, 0.00)	0.01 *** (0.00, 0.02)
	Initial planting density	0.01 (−0.01, 0.02)	0.00 (−0.01, 0.01)	0.00 (−0.01, 0.01)
AP	standage	−0.23 *** (−0.30, −0.17)	−0.55 *** (−0.68, −0.41)	−0.03 *** (−0.04, −0.02)
	Initial planting density	0.07 (−0.17, 0.31)	0.14 (−0.34, 0.62)	0.01 (−0.02, 0.04)
TK	standage	0.47 *** (0.38, 0.57)	0.04 (−0.04, 0.12)	0.37 *** (0.22, 0.51)
	Initial planting density	0.15 (−0.21, 0.51)	−0.06 (−0.42, 0.29)	0.22 (−0.28, 0.73)
AK	standage	0.91 *** (0.47, 1.35)	1.29 *** (0.61, 1.97)	0.08 (−0.37, 0.53)
	Initial planting density	1.11 (−1.96, 4.17)	4.35 ** (1.57, 7.14)	0.364 (−1.22, 1.95)
pH	standage	0.00 (−0.01, 0.01)	−0.01 ** (−0.02, −0.00)	−0.01 (−0.01, 0.00)
	Initial planting density	−0.01 (−0.04, 0.01)	−0.01 (−0.05, 0.02)	0.00 (−0.02, 0.02)

Note: As defined in Table 1 and Figure 2. Significant p-value at: * p < 0.05; ** p < 0.01; *** p < 0.001; FJ is Fujian Province, JX is Jiangxi Province, SC is Sichuan Province.

). Additionally, the response coefficients of soil nutrient factors (TN, TP, TK, AK, and pH) to stand age varied in direction and significance across the three provinces. The significance of the response of AK to initial planting density also varied across provinces.

3.2. Pathway Analysis of the Impact of Stand Age and Initial Planting Density on SCFI

The SEM for the complete sample set explained 88% of the variance in the SCFI for Chinese fir plantations (Figure 6a). The analysis revealed that stand age significantly and directly affected the SCFI (direct effect = 0.18 ***). Stand age and initial planting density significantly influenced SOM, which affected the SCFI. Changes in SOM significantly influenced soil mineral nutrients and pH, impacting the SCFI (indirect effect of stand age = −0.06 **, indirect effect of initial planting density = 0.09 **). In summary, stand age and initial planting density significantly positively affected the SCFI (total effect of stand age = 0.12 **, total impact of initial planting density = 0.09 **) (

Table 8. Standardized indirect, direct, and total effects of stand age and initial planting density on the SCFI.

Outcome Variables	Independent Variables	Standardized Effects
		Direct	Indirect	Total
SCFI (combined area)	stand age	0.18 ***	−0.06 **	0.12 **
	initial planting density	0.03	0.09 **	0.09 **
SCFI (Fujian Province)	stand age	0.44 ***	−0.19 **	0.25 **
	initial planting density	0.06	0.11	0.17
SCFI (Jiangxi Province)	stand age	−0.24 ***	0.25 **	0.01 **
	initial planting density	0.00	0.02	0.02
SCFI (Sichuan Province)	stand age	−0.02	0.32 ***	0.32 ***
	initial planting density	0.10 *	0.08	0.10 *

Note: As defined in Figure 2. Significant p-value at: * p < 0.05; ** p < 0.01; *** p < 0.001.

).

The SEMs for Fujian, Jiangxi, and Sichuan Provinces explained 75%, 91%, and 83% of the variance in the SCFI of the Chinese fir plantations, respectively (Figure 6b–d). The SEM results for each province indicated that stand age significantly and directly affected the SCFI in Fujian and Jiangxi (direct effects of Fujian = 0.44 ***, direct effects of Jiangxi = −0.24 ***). In all provinces, stand age significantly influenced the SCFI through soil mineral nutrients. Additionally, in Fujian, pH, influenced by soil mineral nutrients, also significantly altered the SCFI (indirect effects: Fujian = −0.19 **, Jiangxi = 0.25 **, Sichuan = 0.32 ***). The overall effect of stand age on the SCFI was significantly positive across all three provinces (total effects in Fujian = 0.25 **, Jiangxi = 0.01 **, Sichuan = 0.32 **). In Sichuan, the initial planting density significantly and directly affected the SCFI (direct effects = 0.32 ***). The total effect of the initial planting density on the SCFI in Sichuan was significantly positive (total effect = 0.10 *) (Table 8). Notably, although stand age and initial planting density did not significantly impact SOM in any province, SOM still significantly influenced the SCFI by affecting soil mineral nutrients.

4. Discussion

4.1. The Impact of Stand Age on SCFI

The analysis showed that stand age led to an overall increase in the SCFI in subtropical Chinese fir plantations. This finding is consistent with previous studies reporting significantly higher soil fertility indices in over-mature stands than in young ones [42]. Previous studies indicate that as forest age increases, soil microbial communities become more stable and diverse, potentially enhancing the SCFI of the Chinese fir plantations in Fujian [43]. In Jiangxi Province, stand age directly reduced the SCFI. This decrease may result from age-related changes in tree size and biomass, leading to soil nutrient depletion and significantly lowering soil fertility [44]. Pathway analysis revealed that SOM, a crucial intermediary variable, influenced the SCFI by affecting soil minerals and pH. This effect may be closely related to the mineralization of SOM and its impact on soil fertility [45]. The total effect of stand age on the SCFI was positive, indicating that soil fertility in Chinese fir plantations has been maintained throughout a 30-year rotation period. We observed a significant negative impact of stand age on soil AP. This finding aligns with studies on Pinus massoniana Lamb. (Pinaceae) and Pinus sylvestris L. (Pinaceae) plantations [46,47]. This result is because, with the increasing age of Chinese fir plantations, phosphorus uptake rises. In contrast, phosphorus returned to soil decreases due to enhanced internal phosphorus cycling within the trees [48]. This result highlights the need for phosphorus supplementation in long-rotation Chinese fir plantations. Due to significant phosphorus depletion in older stands, regular phosphorus supplementation, particularly in plantations older than 20 years, may help mitigate nutrient deficits. The lack of a significant interaction effect between stand age and initial planting density across the three provinces may be due to the influence of other variables, such as regional soil characteristics, climatic conditions, or micro-environmental factors, which were not accounted for in our current model. These factors warrant further investigation to better understand the drivers of soil fertility.

4.2. The Impact of Initial Planting Density on SCFI

Our study found that the SCFI increased with higher initial planting density in Chinese fir plantations in subtropical China, which contrasts with previous findings indicating that actual stand density reduces soil nutrient content in such plantations [35]. This discrepancy may be explained by the difference between initial planting density and actual stand density. Previous studies typically focused on actual stand density, which is influenced by factors like self-thinning and natural mortality [36], whereas our study specifically examined initial planting density, which may have a different impact on soil fertility. Additionally, our study demonstrated that initial planting density indirectly influences the SCFI by affecting SOM in subtropical Chinese fir plantations. This effect may be due to increased litterfall from a high initial planting density, which elevates SOM content [49]. Another reason might be that a high initial planting density could hinder SOM decomposition, possibly due to a shift in fungal community composition from saprophytic to a mixed trophic mode as density increases [50]. Furthermore, gene expression related to circadian rhythms, carbon metabolism, and amino acid biosynthesis negatively correlates with stand density, indirectly influencing tree growth, development, and nutrient cycling [51]. Managing planting density to optimize tree spacing and incorporating strategies such as organic amendments may improve SOM retention and soil health.

4.3. The Impact of Region on SCFI

The likelihood ratio test showed that region significantly affected the SCFI, suggesting that regional differences play an important role in the SCFI and nutrient availability. Fujian and Jiangxi had higher initial SCFI and soil fertility levels than Sichuan. This result is likely due to the greater geographical distance between Sichuan and the other two provinces, resulting in different parent rocks and mineral weathering processes and soil fertility differences [52]. In Jiangxi, the SCFI remains stable with stand age. This may be due to the high SCFI values, which support plantation growth and maintain stable soil nutrient levels. In Fujian, the SCFI does not change significantly with initial planting density. This may be due to the high levels of TP and AK in the soil. With sufficient phosphorus and potassium, competition for nutrients among trees is low. As a result, even at higher planting densities, the nutrient supply remains adequate, minimizing the impact of density changes on soil nutrients. These findings highlight the need for adaptive forest management strategies in the context of climate change. For example, increased temperatures and changing precipitation patterns may affect soil nutrient availability. Regions experiencing more intense rainfall may face higher risks of soil erosion and nutrient leaching, requiring adjustments in soil management practices, such as erosion control measures and enhanced organic matter inputs. Therefore, management practices that avoid excessive tillage and promote organic matter retention will be increasingly important to ensure soil health and long-term fertility under changing climate conditions. Soil fertility, which underpins forest growth, is influenced by various forest management practices. Regional natural conditions, such as climate, site-specific attributes, and biodiversity, may further influence plantation growth, soil physicochemical properties, and microbial communities [53,54,55,56,57], thereby diversifying soil fertility responses to forest management across regions.

It is important to acknowledge the limitations of using the Nemerow index method to evaluate SCFI. Although the method effectively integrates multiple soil fertility indicators, it oversimplifies the complex nature of soil fertility across regions. The Nemerow index treats all soil fertility indicators equally, potentially overlooking the varying importance of different nutrients and soil properties in different ecological contexts. Soil fertility is influenced by factors such as microbial communities, climate, and management practices. Relying solely on a composite index may overlook the diversity of soil conditions and nutrient cycling processes across regions.

5. Conclusions and Suggestions

5.1. Conclusions

This study examined the effects of stand age and initial planting density on the comprehensive soil fertility of Chinese fir plantations. Key findings revealed that soil fertility increases with stand age and initial planting density in subtropical China. In Fujian, where soil nutrient levels were relatively high, planting density had a minor effect on soil fertility. Conversely, the initial planting density significantly improved the SCFI in Jiangxi and Sichuan. In Jiangxi, where soil was already fertile, stand age had a minimal impact on the SCFI, indicating that nutrient levels remained stable in mature plantations in Jiangxi. However, the SCFI improved significantly with stand age in Fujian and Sichuan. SOM is crucial in determining how stand age and initial planting density affect soil fertility; thus, practices that preserve SOM should be prioritized. Forest managers should adjust planting densities and consider plantation age based on regional soil fertility conditions, focusing on SOM maintenance and proper nutrient cycling to optimize the SCFI and support forest health.

5.2. Suggestions

The following recommendations are made to enhance forest management and the sustainability of Chinese fir plantations based on the findings of this study. (1) Optimize planting density: In regions such as Fujian and Sichuan, increasing planting density can improve soil fertility and the SCFI, thereby enhancing carbon sequestration. In Jiangxi, where no significant effects were observed, alternative strategies, such as species selection or soil amendments, should be explored. (2) Focus on stand age and soil management: Since stand age positively influences the SCFI, promoting healthy stand development through thinning and pest management is essential. Enhancing soil fertility through fertilization or organic amendments can increase the SCFI in younger stands. (3) Consider regional differences: Forest management strategies should be tailored to the specific conditions of each region, taking into account local soil, climate, and forest characteristics. In regions such as Jiangxi, where the SCFI is high, optimizing stand density and age may be less effective, and broader ecosystem management strategies should be explored. (4) Further research on soil nutrient dynamics: Future studies should investigate the effects of planting density and stand age on soil nutrients and the SCFI to better understand the interactions between SOM, mineral nutrients, and pH.