Tree Species Diversity and Tree Growth Affected Element Compositions in Glomalin-Related Soil Protein–Soil pH Interaction

Abstract

Glomalin-related soil protein (GRSP), a glycoprotein derived from mycorrhizal fungal hyphae, is a mixture of substances rich in various elements essential for plant growth. However, the impacts of tree diversity and forest structure on the element content and storage of GRSP are not well understood. To investigate this, we collected soil samples from 720 plots (10 m × 10 m) and determined the relative content and storage of elements (C, N, O, Si, P, Fe, Al, Na, Mg, Ca, and K) in GRSP. Additionally, the tree diversity, tree size and density, tree assemblage, and soil physicochemical properties were determined. The results show the following: (1) Plots with lower diversity had 1.27 times higher storage of 11 elements in GRSP compared to those with higher diversity. Plots with higher soil electrical conductance (EC) plots had 28–35% higher storage of 11 elements in GRSP. (2) The relative content of Na, C, and N in GRSP showed a positive relationship with pH, while they exhibited a negative relationship with soil EC, available phosphorus (AP), and tree density. Other elements generally showed contrasting patterns. (3) Path analysis reveals that tree diversity and tree growth had stronger effects on the elemental composition of GRSP than tree spatial assemblage. The magnitude of the driving path coefficients depended on the factors closely related to soil pH. This study demonstrates that the elemental composition of GRSP can be dynamically affected by tree diversity and stand structure, with soil pH playing a crucial interactive role.

1. Introduction

Soil fungi networks are important for plant community stability [1], and arbuscular mycorrhizal (AM) fungi can impact the diversity of forest communities [2]. Glomalin-related soil protein (GRSP) is primarily produced by the death of AM mycelium [3] and is a mixture of abundant extracellular protein and other substances [4,5,6]. High levels of GRSP in soils indicate higher soil aggregate stability, as well as enhanced long-term storage of soil carbon and nitrogen [4]. GRSP is also an excellent source of various macro- and micro-elements, such as C, H, O, S, K, P, Ca, Si, Fe, Cu, and Mg, which are necessary for plant growth and help in the storage of heavy metal pollution in soils and sediments [7,8,9,10]. Some studies have found that GRSP contributes 4–23% of the carbon content in different soils [11,12]. Lovelock et al. determined that GRSP contained about 4% nitrogen in tropical soils, accounting for 5% of the total nitrogen in the soil [13]. Similarly, Zhang et al. found a notably high nitrogen concentration of 13.13% in GRSP samples obtained from soils with high salinity [7]. Currently, limited data on GRSP composition have hindered a precise understanding of its role in soil functions. Determining how GRSP composition changes with environmental factors is essential for enhancing our knowledge of this important soil organic matter (SOM) fraction [14].

Tree diversity plays a vital role in increasing forest carbon storage [15] and ecosystem productivity stability [16,17]. Plant–fungi networks are important for plant community stability [1]. Understanding the complex associations between taxonomic tree species diversity, tree size and density, tree spatial distribution, and mycorrhizal-fixed elements (GRSP-fixed ones) under different soil physicochemical conditions is important for exploring the interaction between aboveground plants and underground soil functions. Diversity-related niche complementarity allows ecosystems to utilize various resources more effectively [18]. Tree spatial distribution patterns can also affect soil properties [19]. Some commonly used parameters have been developed, such as neighborhood comparison (U), the mingling index (M), and the uniform angle index (W) [20,21,22,23]. Tree size and density-related forest structures strongly explain the variation in the water-holding capacity of forests [24]. Our previous studies found that tree diversity and forest structure significantly affect the GRSP content [25]. Studies have found that GRSP constitutes a significant portion (approximately 4–5%) of carbon in forest soils. The contribution of GRSP to soil organic carbon (SOC) is more than 20 times greater than that of microbial biomass [11]. However, how tree diversity and forest structure affect the GRSP element composition is still less understood. Quantifying the difference in the element composition of GRSP and exploring its relationship with plant diversity, forest growth, and spatial assemblages of trees could favor the understanding of associations between changes in tree identity and nutrient retention from the community scale to the ecosystem scale [26].

GRSP, a significant component produced by arbuscular mycorrhizal fungi (AMF) and abundant in diverse elements, plays a crucial role in the growth and development of plants [9]. Some studies have found that the element composition in total GRSP (TG) is determined by soil properties [7,11]. However, the impact of aboveground tree characteristics remains unknown. Based on this, we hypothesized that the diversity of tree species, their spatial distribution, and forest growth may collectively influence the storage and relative content of various elements in GRSP. Moreover, these influences may closely interact with soil pH. We aimed to (1) explore how the diversity index, tree size, spatial distribution, community structure, and soil pH affect the storage and relative content of 11 elements in TG and (2) identify the possible driving paths affecting element storage and relative content in TG and their possible implications. The results of this study can contribute to forest management and enhance our comprehension of the underground soil processes related to glomalin composition in high-latitude forests in the northern hemisphere.

2. Materials and Methods

2.1. Description of the Research Site, Soil Sample Collection, and Experimental Design

Our research area was situated within the Harbin Experimental Forest Farm of Northeast Forestry University, spanning the coordinates of 126°15′ to 127°30′ east longitude and 45°20′ to 46°20′ north latitude. To mitigate the impact of edge effects on tree growth, we selected a 7.2 ha sample plot in the central area of a 35 ha experimental forest to prevent edge effects from affecting tree growth. This sample plot was further divided into 720 subplots (10 m × 10 m). Within these subplots, we investigated the structural indicators of the forest, comprising four community structure indicators: diameter at breast height (DBH), tree height (TH), height under branches (UBH), and stand density (SD). Additionally, we examined three spatial assemblage indicators labeled U, M, and W. To characterize plant diversity within the plot, we calculated the Simpson index (SP), Shannon Wiener index (SW), Pielou evenness index (PL), and species richness (SR) using field plot data. We assessed five soil indicators: bulk density (BD), water content of soil (WC), soil acidity (pH), electrical conductivity (EC), and available phosphorous (AP). The diversity and structural traits were derived from field survey data, which included taxonomic species information (Figure 1).

The soil samples were collected in September 2021. In a 10 m × 10 m sample plot, we gathered soil samples from the topsoil layer (0–20 cm) using a 200 cm³ cutting ring. After removing visible stones and roots, the soil samples were air-dried in a cool, well-ventilated place until they reached a constant weight. Subsequently, the samples were passed through a 0.25 mm sieve and stored for GRSP element composition and soil physicochemical properties.

To identify the effects of tree diversity, size, spatial distribution, and soil properties on the TG amount and elemental composition, we categorized these parameters into three levels of treatment: low, medium, and high. For all parameters, the sample sizes were distributed as follows: 50 to 231 for the low level, 195 to 456 for the medium level, and 96 to 310 for the high-level treatments (

Table 1. Criteria for distinguishing the sample sizes into three treatment levels for each parameter. SR, species richness; SW, Shannon Wiener index; SP, Simpson index; PL, Pielou evenness index; TH, tree height; DBH, diameter at breast height; SD, stand density; UBH, height under branches; U, neighborhood comparison; M, mingling index; W, uniform angle index; WC, water content of soil; BD, bulk density; pH, soil acidity; and AP, available phosphorous.

Parameters	Low		Middle		High
	Criteria	Sample Size	Criteria	Sample Size	Criteria	Sample Size
Tree diversity
SR	<3	53	3–6	376	>6	222
SW	<0.9	105	0.9–1.6	353	>1.6	193
SP	<0.4	50	0.4–0.7	291	>0.7	310
PL	<0.78	206	0.78–0.87	214	>0.87	230
Community structure
TH	<5	162	5–8	359	>8	130
DBH	<5	99	5–12	456	>12	96
SD	<1500	152	1500–3000	309	>3000	190
UBH	<2	136	2–4	399	>4	116
Spatial structure
U	<0.47	174	0.47–0.5	195	0.5–0.75	282
M	<0.5	137	0.5–0.75	374	0.75–1	140
W	<0.5	118	0.5–0.6	418	0.6–1	115
Soil traits
WC	<9.5	165	9.5–12.6	301	>12.6	185
BD	<1.5	172	1.5–1.62	263	>1.62	216
pH	<6.3	154	6.3–7	298	>7	199
EC	<100	189	100–200	341	>200	121
AP	<6	231	6–11	264	>11	156

).

2.2. Plant Diversity and Forest Structure Parameter Calculation

The formulas for calculating diversity indicators are as follows:

SW = −∑P_ilnP_i,

(1)

SP = 1 − ∑(P_i²),

(2)

PL = SW/lns,

(3)

SR = s,

(4)

where P_i represents the percentage of species i to the total number of all tree species in their abundance. ‘s’ is the number of taxonomic species in each plot.

The community structure and spatial structure indices were calculated. The community structure is primarily associated with tree size and density, which are represented by the UBH (unit basal area), TH (tree height), SD (stand density), and DBH (diameter at breast height). We utilized the average values of these parameters to represent the plot-level forest structure.

The spatial structure indices included U (neighborhood comparison), which reflects the proportion of dominant trees at the community level, and W (the uniform angle index) is a spatial structural parameter that describes the distribution pattern of individual trees, with values ranging from 0 to 1, where a smaller value indicates a more uniform distribution pattern of trees. M (the mixing degree) represents the probability of the nearest four adjacent tree species types of the reference tree. Detailed calculation formulas have been described in our previous research [22,27].

2.3. TG Storage and Its Element Relative Content Measurement

2.3.1. T-GRSP Storage Determination

The determination of TG in the soil samples was conducted following the method described by Vasconcellos, et al. [28]. The extraction and determination process is detailed as follows: Initially, 0.10 g of soil was used to extract GRSP, to which 4 mL of 50 mM citrate buffer (pH, 8.00) was added. The mixture was then subjected to a high-pressure treatment at 121 °C for 1 h. Afterward, the supernatant was centrifuged at 4000 rpm for 6 min. This process was repeated several times for the same soil sample (autoclaving at 121 °C for 30 min). Then, all the extracts from the same sample were merged for the determination of TG storage in one unit weight of bulk soil.

2.3.2. Purification of TG and Determination of Element Concentrations Using X-Ray Photoelectron Spectroscopy (XPS)

The purification of TG was carried out following the method described by Gillespie, et al. [29]. The concentrations of 11 elements within the TG were measured using X-ray photoelectron spectroscopy (XPS). The spectral processing and the atomic concentration of elements (Al, Si, C, N, P, O, Fe, Na, Mg, Ca, and K) were calculated by Casa XPS software (V2.3.12, Casa Software Ltd., Teignmouth, UK) (Figure A1) [30].

2.3.3. Element Relative Content and Storage Calculation

Each element’s relative content was calculated by dividing its concentration by the total concentration of all the elements. Element storage was calculated by multiplying each element’s relative content by TG storage.

2.4. Determination of Soil Properties

Soil bulk density (BD), water content (WC), pH, electrical conductance (EC), and available phosphorus (AP) were measured using the methods described by Peng, et al. [31] and Zhu, et al. [32].

2.5. Data Statistical Analysis

The normal distribution of the initial data was assessed using the Shapiro–Wilk test. If the data were not normally distributed, logarithmic transformations were applied to standardize and normalize the data. The differences in the TG amount and element composition (both storage and relative content) across different gradients of tree diversity, tree size, and spatial structure at low, medium, and high levels were analyzed using one-way ANOVA and Duncan’s multiple range test with SPSS 25 (SPSS Analytics Solutions, Chicago, IL, USA) (Table 1). The influences of soil properties, tree density, and tree size on element storage were further explored through linear regression and Pearson’s correlation analysis using Origin 2019 (Electronic Arts Inc., Northampton, MA, USA). Cluster analysis was used to classify the element relative content based on their relations with independent variables using TBtools-II (Toolbox for Biolo-gists v1.120, China). Redundancy ordination analysis (RDA) and variance partitioning analysis were used to determine the relative importance of plant diversity, forest structure, and soil properties to the TG amount and element composition. Both analyses were performed using Canoco 5 (Biometrics, Wageningen, The Netherlands). The partial least squares path model (PLS-PM) was employed to identify potential driving pathways of tree diversity, tree assemblage patterns, and soil properties on the variations of TG and its elemental composition. The PLS-PM was constructed with ‘plspm’ packages in R-4.2.3 (University of Auckland, New Zealand).

3. Results

3.1. Effects of Plant Diversity Variations

Plots with high diversity exhibited lower TG storage of the 11 elements, particularly in plots with a high SW (

Table 2. The storage of TG and 11 elements in TG at diverse levels (low, mid, and high) of plant diversity (SW, SP, SR, and PL). SW, Shannon Wiener index; SP, Simpson index; SR, species richness; and PL, Pielou evenness index. Different lowercase letters indicate significant differences at different levels of the same indicator (p < 0.05). A shallow green background indicates significant decreases from low to high.

	SW			PL			SP			SR
	Low	Mid	High	Low	Mid	High	Low	Mid	High	Low	Mid	High
TG storage (mg g−1)
TG	8.96 a	8.59 ab	7.89 b	8.73 a	8.49 a	8.15 a	8.31 a	8.68 a	8.24 a	8.45 a	8.54 a	8.28 a
Element storage in TG (mg g−1)
Al	0.34 a	0.32 ab	0.27 b	0.33 a	0.31 a	0.29 a	0.30 a	0.32 a	0.29 a	0.31 a	0.31 a	0.29 a
C	3.53 a	3.43 ab	3.26 b	3.46 a	3.42 a	3.32 a	3.35 a	3.45 a	3.35 a	3.37 a	3.42 a	3.37 a
N	0.26 a	0.25 a	0.24 a	0.25 a	0.25 a	0.25 a	0.25 a	0.25 a	0.25 a	0.25 a	0.25 a	0.25 a
O	3.45 a	3.28 ab	2.95 b	3.35 a	3.23 a	3.08 a	3.16 a	3.32 a	3.12 a	3.24 a	3.26 a	3.13 a
Si	0.65 a	0.60 ab	0.51 b	0.62 a	0.59 a	0.55 a	0.57 a	0.61 a	0.56 a	0.60 a	0.60 a	0.56 a
Fe	0.13 a	0.12 ab	0.10 b	0.12 a	0.12 a	0.11 a	0.11 a	0.12 a	0.11 a	0.12 a	0.12 a	0.11 a
Na	0.23 a	0.23 ab	0.21 b	0.23 a	0.22 a	0.22 a	0.22 a	0.23 a	0.22 a	0.22 a	0.23 a	0.22 a
P	0.04 a	0.04 ab	0.03 b	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a
Mg	0.05 a	0.04 ab	0.04 b	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a
Ca	0.05 a	0.05 ab	0.04 b	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a
K	0.35 a	0.33 ab	0.30 b	0.34 a	0.33 a	0.31 a	0.32 a	0.34 a	0.32 a	0.33 a	0.33 a	0.32 a

). However, there were no significant differences in TG, and the storage of the 11 elements low, middle, and high in the PL, SP, and SR plots (p < 0.05). Specifically, in plots with a low SW, TG storage was 13.94% higher compared to the high SW plots (p < 0.05). Elements such as Al, C, and O were 1.26, 1.08, and 1.17 times higher than those in the high SW plots. Similarly, Si, Fe, Na, P, Mg, Ca, and K in the low SW plots were 1.09 to 1.33 times higher than those in the high SW plots.

We also performed a linear correlation analysis to examine the relationships between tree diversity and TG storage (Figure A2 and

Table 3. Pearson correlations between plant diversity, forest structure, soil properties, and TG and its elemental storage. Bold fonts indicate significant correlations. * is at p < 0.05, and ** is at p < 0.01.

Group	Parameter	Correlation Coefficients												Sig. No
		TG	Al	C	N	O	Si	Fe	Na	P	Mg	Ca	K
Tree diversity	SW	−0.057	−0.061	−0.048	−0.035	−0.059	−0.061	−0.06	−0.051	−0.06	−0.059	−0.057	−0.059	0
4 sig./48 total	PL	−0.068	−0.068	−0.063	−0.052	−0.069	−0.068	−0.068	−0.065	−0.068	−0.069	−0.068	−0.069	0
	SP	−0.072	−0.078 *	−0.057	−0.038	−0.075	−0.078 *	−0.077 *	−0.061	−0.077 *	−0.075	−0.072	−0.074	4
	SR	−0.048	−0.056	−0.033	−0.017	−0.051	−0.056	−0.054	−0.037	−0.055	−0.051	−0.048	−0.051	0
Tree size and density	DBH	−0.061 *	−0.06	−0.084 *	−0.086 *	−0.069	−0.06	−0.064	−0.082*	−0.063	−0.069	−0.073	−0.07	3
19 sig./48 total	TH	0.021	0.043	−0.011	−0.038	0.03	0.043	0.038	−0.003	0.04	0.029	0.021	0.028	0
	UBH	0.065	0.080 *	0.039	0.011	0.072	0.080 *	0.077 *	0.045	0.078 *	0.071	0.065	0.07	4
	SD	0.179 **	0.188 **	0.151 **	0.111 **	0.184 **	0.188 **	0.187 **	0.159 **	0.187 **	0.183 **	0.179 **	0.183 **	12
Tree clustering pattern	U	0.046	0.051	0.035	0.022	0.048	0.051	0.05	0.038	0.05	0.048	0.046	0.048	0
0 sig./36 total	M	0.024	0.035	0.006	−0.01	0.029	0.035	0.033	0.011	0.033	0.028	0.024	0.028	0
	W	−0.019	−0.014	−0.023	−0.024	−0.017	−0.014	−0.015	−0.022	−0.015	−0.017	−0.019	−0.017	0
Soil properties	BD	−0.003	−0.025	0.027	0.051	−0.013	−0.025	−0.02	0.02	−0.022	−0.011	−0.003	−0.01	0
36 sig./60 total	WC	−0.016	0.002	−0.038	−0.055	−0.008	0.002	−0.002	−0.033	0	−0.009	−0.016	−0.01	0
	pH	−0.263 **	−0.312 **	−0.175 **	−0.079 *	−0.286 **	−0.312 **	−0.302 **	−0.198 **	−0.305 **	−0.282 **	−0.264 **	−0.281 **	12
	EC	0.173 **	0.134 **	0.207 **	0.219 **	0.158 **	0.135 **	0.144 **	0.201 **	0.141 **	0.160 **	0.173 **	0.161 **	12
	AP	0.158 **	0.143 **	0.162 **	0.151 **	0.153 **	0.143 **	0.147 **	0.162 **	0.146 **	0.154 **	0.158 **	0.154 **	12

). We found that the SP was negatively correlated with Al, Si, Fe, and P (p < 0.05), while other diversity parameters did not significantly relate to the elemental storage in TG. Both linear regression and ANOVA results indicate that tree diversity decreased TG and all element storage.

3.2. Influence of Tree Size and Density

TG and storage of the 11 elements displayed a consistent pattern across low, middle, and high tree sizes and SD plots (

Table 4. Differences in TG and the storage of 11 elements at diverse levels (low, mid, and high) of community structure (DBH, UBH, TH, and SD). DBH, diameter at breast height; UBH, height under branches; TH, tree height; and SD, stand density. Different lowercase letters indicate significant differences at different levels of the same indicator (p < 0.05). A shallow green background indicates significant decreases from low to high, while a yellow-brown background indicates significant increases from low to high treatments.

	DBH			TH			UBH			SD
	Low	Mid	High	Low	Mid	High	Low	Mid	High	Low	Mid	High
TG storage (mg g−1)
TG	8.89 a	8.46 ab	7.89 b	8.78 a	8.17 a	8.79 a	8.04 b	8.44 ab	8.92 a	7.07 b	8.70 a	9.12 a
Element storage in TG (mg g−1)
Al	0.33 a	0.31 a	0.27 a	0.33 a	0.29 a	0.34 a	0.28 b	0.31 ab	0.35 a	0.22 b	0.32 a	0.36 a
C	3.52 a	3.40 ab	3.24 b	3.50 a	3.33 a	3.44 a	3.32 a	3.40 a	3.47 a	3.07 b	3.47 a	3.53 a
N	0.26 a	0.25 ab	0.24 b	0.26 a	0.25 a	0.25 a	0.25 a	0.25 a	0.25 a	0.23 b	0.25 a	0.25 a
O	3.41 a	3.22 ab	2.97 b	3.36 a	3.08 a	3.40 a	3.01 b	3.21 ab	3.46 a	2.57 b	3.33 a	3.54 a
Si	0.63 a	0.58 a	0.52 a	0.62 a	0.55 a	0.64 a	0.52 b	0.58 ab	0.66 a	0.41 b	0.61 a	0.68 a
Fe	0.13 a	0.12 ab	0.11 b	0.12 a	0.11 a	0.13 a	0.11 b	0.12 ab	0.13 a	0.09 b	0.12 a	0.13 a
Na	0.23 a	0.22 ab	0.21 b	0.23 a	0.22 a	0.23 a	0.22 a	0.22 a	0.23 a	0.20 b	0.23 a	0.23 a
P	0.04 a	0.04 ab	0.03 b	0.04 a	0.04 a	0.04 a	0.03 b	0.04 ab	0.04 a	0.03 b	0.04 a	0.04 a
Mg	0.05 a	0.04 ab	0.04 b	0.05 a	0.04 a	0.05 a	0.04 b	0.04 ab	0.05 a	0.03 b	0.04 a	0.05 a
Ca	0.05 a	0.05 ab	0.04 b	0.05 a	0.05 a	0.05 a	0.04 b	0.05 ab	0.05 a	0.04 b	0.05 a	0.05 a
K	0.35 a	0.33 ab	0.30 b	0.34 a	0.31 a	0.34 a	0.31 b	0.33 ab	0.35 a	0.26 b	0.34 a	0.36 a

). Small-diameter but high trees and dense forests typically had higher TG and elemental storage. Some of these differences were statistically significant, particularly those related to SD treatments, and most plots were associated with DBH and UBH. However, no significant differences were found in the low, middle, and high TH plots (Table 4). In small-diameter-tree plots, TG and the storage of all 11 elements were 1.1 to 1.25 times higher than those in large-diameter-tree plots. TG and elemental storage in high UBH plots were 1.05–1.33 times higher than those in low UBH plots, and the most pronounced difference was observed in P storage (1.33 fold). High-density forest plots had 1.08–1.67-fold higher TG and elemental storage compared to low-density forests, with the peak differences noted in Mg and Si storage (1.66–1.67 fold) (p < 0.05).

Linear regression and Pearson correlation analyses were used to determine the impact of tree size and density on element storage (Figure A2 and Table 3). We found that DBH was negatively related to TG, C, Na, and N storage, and UBH was positively related to Al, Si, Fe, and P storage (p < 0.05). In comparison with tree sizes, SD had a stronger positive association with the storage of the 11 elements (p < 0.001), and the coefficients ranged from 0.11 to 0.19.

3.3. Effect of Tree Spatial Structure

The ANOVA results of TG and the storage of the 11 elements in different spatial distribution structures are shown in

Table 5. Differences in TG and storage of 11 elements under various spatial structure levels (low, mid, and high) of U, M, and W. U, neighborhood comparison; M, mingling index; and W, uniform angle index. Different lowercase letters signify significant differences at various levels of the same indicator (p < 0.05). A light green background indicates significant differences between the middle and high M plots.

	U			M			W
	Low	Mid	High	Low	Mid	High	Low	Mid	High
TG storage (mg g−1)
TG	7.95 a	8.65 a	8.60 a	8.36 a	8.24 a	8.96 a	8.05 a	8.65 a	8.05 a
Element storage in TG (mg g−1)
Al	0.28 a	0.32 a	0.32 a	0.29 a	0.30 a	0.34 a	0.28 a	0.32 a	0.29 a
C	3.26 a	3.47 a	3.43 a	3.43 ab	3.31 b	3.55 a	3.31 a	3.46 a	3.25 a
N	0.24 a	0.25 a	0.25 a	0.25 ab	0.24 b	0.26 a	0.25 a	0.25 a	0.24 a
O	2.99 a	3.30 a	3.29 a	3.14 a	3.13 a	3.44 a	3.02 a	3.31 a	3.05 a
Si	0.52 a	0.60 a	0.61 a	0.55 a	0.57 a	0.64 a	0.53 a	0.61 a	0.55 a
Fe	0.11 a	0.12 a	0.12 a	0.11 a	0.11 a	0.13 a	0.11 a	0.12 a	0.11 a
Na	0.21 a	0.23 a	0.23 a	0.22 a	0.23 a	0.23 a	0.22 a	0.23 a	0.21 a
P	0.03 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.03 a	0.04 a	0.04 a
Mg	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.05 a	0.04 a	0.04 a	0.04 a
Ca	0.04 a	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.04 a	0.05 a	0.04 a
K	0.30 a	0.34 a	0.33 a	0.32 a	0.32 a	0.35 a	0.31 a	0.34 a	0.31 a

. Across three levels of U and W, the TG amount in the plots with a middle level of U and W was the highest. The storage of C and N in TG in plots with a middle-level of M was 6.76% and 7.69% lower, respectively, compared to those in the high-level plots.

Overall, plots with high U and high M had 7% to 8% higher TG storage compared to low plots. The storage of the 11 elements in TG in high U plots showed an average increase of 13% compared to low U plots, ranging from 0% for Mg storage to 33% for P storage. Similarly, high M plots had an average increase of 10% in the storage of the 11 elements compared to low M plots, ranging from 0% for P and Ca storage to 17% for Al storage (Table 5). However, the differences did not reach a statistically significant level (p > 0.05), except for the storage of C and N in TG between the middle and high M plots (p < 0.05).

The relationships between U, M, W, and elemental composition were also analyzed. We found that TG and the storage of the 11 elements showed a positive correlation with U and M, while they exhibited a negative correlation with W. However, all these correlations were too weak to be statistically significant (p > 0.05) (Table 3).

3.4. Effect of Soil Properties

TG storage in high EC plots was 1.28 times greater than in low EC plots, and the storage of the 11 elements was 1.22 to 1.67 times higher in high EC plots (p < 0.05). Similarly, TG storage in high AP plots was 1.24 times greater than in low AP plots, with all 11 elements showing a 1.18- to 1.4-fold increase in high EC plots (p < 0.05). The ANOVA results of TG and the storage of the 11 elements across various soil physicochemical plots are shown in Table 5. In plots with high pH, TG and the storage of the 11 elements were 8% to 84% higher than those in low pH plots, with Al element storage showing the most significant decrease (p < 0.05). TG storage in high EC plots was 1.28 times greater than in low EC plots, and the storage of the 11 elements in high EC plots was 1.22–1.67 fold higher than that in low EC plots (p < 0.05). TG storage in high AP plots was 1.24 times higher than that in low AP plots, and the storage of the 11 elements was 1.18–1.4 fold higher in high EC plots compared to low EC plots (p < 0.05) (

Table 6. Differences in TG storage and the storage of 11 elements in TG in low, middle, and high soil property plots. Differences in TG storage and the storage of 11 elements in TG in low, middle, and high soil property plots (BD, WC, pH, EC, and AP). BD, bulk density; WC, water content of soil; pH, soil acidity; EC, electrical conductivity; and AP, available phosphorous. Different lowercase letters indicate significant differences at different levels of the same indicator (p < 0.05). A shallow green background indicates significant decreases from low to high, while a yellow-brown background indicates significant increases from low to high treatments.

	BD			WC			pH			EC			AP
	Low	Mid	High	Low	Mid	High	Low	Mid	High	Low	Mid	High	Low	Mid	High
TG storage (mg g−1)
TG	8.27 a	8.74 a	8.22 a	8.33 a	8.56 a	8.36 a	10.2 a	8.02 b	7.59 b	7.71 b	8.35 b	9.85 a	7.59 c	8.61 b	9.44 a
Element storage in TG (mg g−1)
Al	0.30 a	0.33 a	0.29 a	0.30 a	0.31 a	0.31 a	0.44 a	0.28 b	0.24 b	0.28 b	0.30 b	0.38 a	0.26 c	0.31 b	0.36 a
C	3.32 a	3.46 a	3.38 a	3.39 a	3.44 a	3.33 a	3.76 a	3.28 b	3.26 b	3.12 c	3.40 b	3.83 a	3.13 c	3.45 b	3.69 a
N	0.24 a	0.25 a	0.25 a	0.25 a	0.25 a	0.24 a	0.26 a	0.25 b	0.24 b	0.23 c	0.25 b	0.28 a	0.23 c	0.25 b	0.27 a
O	3.15 a	3.35 a	3.09 a	3.15 a	3.26 a	3.20 a	4.08 a	3.02 b	2.77 b	2.92 b	3.16 b	3.83 a	2.84 c	3.28 b	3.65 a
Si	0.57 a	0.62 a	0.54 a	0.56 a	0.59 a	0.59 a	0.83 a	0.54 b	0.45 b	0.52 b	0.56 b	0.73 a	0.50 c	0.60 b	0.69 a
Fe	0.11 a	0.12 a	0.11 a	0.11 a	0.12 a	0.12 a	0.16 a	0.11 b	0.09 b	0.10 b	0.11 b	0.14 a	0.10 c	0.12 b	0.14 a
Na	0.22 a	0.23 a	0.22 a	0.22 a	0.23 a	0.22 a	0.25 a	0.22 b	0.21 b	0.20 c	0.22 b	0.25 a	0.20 c	0.23 b	0.24 a
P	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.04 a	0.05 a	0.03 b	0.03 b	0.03 b	0.04 b	0.05 a	0.03 c	0.04 b	0.04 a
Mg	0.04 a	0.05 a	0.04 a	0.04 a	0.04 a	0.04 a	0.06 a	0.04 b	0.04 b	0.04 b	0.04 b	0.05 a	0.04 c	0.04 b	0.05 a
Ca	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.05 a	0.06 a	0.04 b	0.04 b	0.04 b	0.05 b	0.06 a	0.04 c	0.05 b	0.05 a
K	0.32 a	0.34 a	0.31 a	0.32 a	0.33 a	0.32 a	0.41 a	0.31 b	0.28 b	0.30 b	0.32 b	0.39 a	0.29 c	0.33 b	0.37 a

).

We conducted Pearson correlation and linear regression analyses to explore the relationships between soil properties and TG and elemental storage (Table 3, Figure A2 and Figure 2). As shown in Figure 2, pH was negatively correlated with TG storage and the storage of the 11 elements in TG. Conversely, EC and AP were positively correlated with these variables, while no significant relationships were found with BD and WC (Figure A3 and Table 3). Among the correlations, pH exhibited the most substantial coefficients, ranging from −0.31 to −0.08, followed by EC with coefficients from 0.13 to 0.22 and AP with coefficients from 0.14 to 0.16 (Table 3).

3.5. Cluster Analysis for the Relationship Between Elemental Relative Content in TG and Various Factors

Cluster analysis classified the Pearson correlations into two groups (Figure 3). The first group included elements that were positively related to pH, i.e., Na, C, and N, which were negatively correlated with EC, SD, and AP. The second group included elements that were negatively correlated with pH, i.e., Al, O, Si, Fe, P, Mg, Ca, and K, which were positively correlated with EC, SD, and AP (p < 0.01) (Figure 3). We also calculated the mean values of the relative content of the 11 elements. The relative content of C in TG was the highest at 44.1%, followed by O at 35.9%, and Si, K, N, Na, Al, and Fe ranged from 1.2% to 5.6%. A lower relative content of P, Mg, and Ca was found at 0.4–0.6%.

3.6. Redundancy Ordination for Elemental Storage and Relative Content Variations

Among all the significant parameters, the RDA and Variation partitioning results indicate that diversity, stand structure, and soil properties accounted for the variations in element storage and relative content in TG (Figure 4 and Figure A4). Soil physicochemical properties were identified as the primary controlling factors for variations in element storage, with a unique explanatory power of 78%. Diversity and tree size–spatial assemblage contributed to the explanation, with 2.8% and 2.0%, respectively (Figure A4). Among all the significant parameters, pH and EC were the main explainers, with explanatory rates of 5.4% and 6.1% under conditional term effects (F > 20, p = 0.002) (Figure 4).

3.7. PLS-PM Analysis: Driving Pathways for the Element Composition Variations in TG

We selected the most probable parameters (diversity traits of the SP and SW, structural traits of DBH, UBH, SD, M, and W, and soil properties of pH, EC, and AP) associated with GRSP element composition and constructed a PLS-PM to identify the driving pathways (Figure 5 and

Table 7. The relative contributions of each influencing factor to the changes in the standard coefficients. pH↑ represents the relative content of the element positively correlated with soil pH; pH↓ represents the relative content of the element negatively correlated with soil pH. *: p < 0.05, and **: p < 0.01.

Relationships	Direct	Indirect	Total
Plant Diversity → Soil Properties	0.019202	0	0.0192
Plant Diversity → TG Storage	−0.082492 *	0.00747	−0.07502
Plant Diversity → Element Storage	−0.000834	−0.07484	−0.07568
Plant Diversity → Relative Content of the Element (pH↑)	0.082492 *	−0.00747	0.07502
Plant Diversity → Relative Content of the Element (pH↓)	−0.082492 *	0.00747	−0.07502
Community Structure → Soil Properties	0.239296 **	0	0.2393
Community Structure → TG Storage	0.105079 **	0.09308	0.19816
Community Structure → Element Storage	0.000807	0.19853	0.19934
Community Structure → Relative Content of the Element (pH↑)	−0.105079 **	−0.09308	−0.19816
Community Structure → Relative Content of the Element (pH↓)	0.105079 **	0.09308	0.19816
Spatial Structure → Soil Properties	0.063033	0	0.06303
Spatial Structure → TG Storage	−0.012083	0.02452	0.01244
Spatial Structure → Element Storage	0.000346	0.0126	0.01295
Spatial Structure → Relative Content of Element (pH↑)	0.012083	−0.02452	−0.01244
Spatial Structure → Relative Content of Element (pH↓)	−0.012083	0.02452	0.01244
Soil Properties → TG Storage	0.388974 **	0	0.38897
Soil Properties → Element Storage	0.002943 **	0.38832	0.39127
Soil Properties → Relative Content of Element (pH↑)	−0.388974 **	0	−0.38897
Soil Properties → Relative Content of Element (pH↓)	0.388974 **	0	0.38897
TG storage → Element Storage	0.998329 **	0	0.99833

). Plant diversity and forest structure had both direct and indirect effects on the element composition in TG, while soil properties largely mediated these processes indirectly. For the storage of the 11 elements, no direct effects were found in plant diversity, community structure, and spatial structure. Their associations were primarily achieved through TG storage and soil properties. Plant diversity indirectly influenced the storage of the 11 elements via TG (−0.08 × 0.99) and soil properties (0.23 × 0.002). For the relative content of the 11 elements in TG, elements positively correlated with pH (Na, C, and N) and those negatively correlated with pH (Al, O, Si, Fe, Mg, P, Ca, and K) showed different patterns. Elements positively correlated with soil pH could be directly enhanced by plant diversity (0.08) but directly decreased by soil properties (−0.38) and forest growth (−0.10). Conversely, elements negatively correlated with soil pH could be directly declined by plant diversity (−0.08) but directly increased by soil properties (0.38) and forest growth (0.10).

Soil properties and TG storage were identified as significant interaction factors influencing elemental storage and relative content (Table 7). Compared to the impacts of forest growth and plant diversity, spatial structure had a more limited effect on element changes, with no significant coefficients found in the PLSM model. In terms of total coefficients, plant diversity exhibited a coefficient of 0.075, forest growth demonstrated coefficients between 0.18 and 0.20, and spatial structure had a coefficient of 0.01.

4. Discussion

More studies on GRSP have emphasized its multiple ecological benefits, such as enhancing carbon sequestration, immobilizing heavy metals in the soil, and improving soil physicochemical properties [33]. These benefits are largely contingent upon the relative content of elements within GRSP [30]. In this study, the element composition in TG was measured, and the relationship between the storage and relative content of 11 elements in TG and plant diversity, forest structure, and soil physicochemical indexes was investigated. We found that the 11 elements in TG could be dynamically regulated by tree diversity and stand structure, and soil pH plays a crucial mediating role in the element composition of TG.

4.1. Plant Diversity Affected Elemental Storage in TG

Understanding the relationship between plant diversity and GRSP and its components can inform strategies for soil improvement [34,35]. Our study reveals that plots with lower plant diversity had 1.01 to 1.27 times higher elemental storage in TG compared to those with higher diversity, with all 11 elements exhibiting a similar trend (Table 2). Our results underscore that plant diversity directly reduced the storage of all 11 elements, primarily driven by changes in TG storage rather than differences in the relative content of the individual elements. This was evidenced by the strong correlation between TG storage and the storage of these 11 elements in the PLSPM analysis. Previous studies have indicated that variations in soil properties are linked to changes in tree species diversity, which, in turn, can affect the GRSP content. For instance, higher soil water content can accelerate the mycelia and spore breakdown in ecosystems, leading to the release of more GRSP and promoting the accumulation of TG [36]. Soil moisture tends to decrease with an increase in tree species diversity [37], which could explain why high tree species diversity is not favorable for GRSP accumulation.

The impact of plant diversity on TG and its element storage might be attributed to niche complementation among diverse trees for nutrient exploitation. The niche complementarity hypothesis posits that increased species diversity enhances productivity, as each additional species utilizes different resources, such as nutrients, leading to a more thorough utilization of the resources in the ecosystem [38]. Some studies have confirmed that increased species diversity is beneficial for improving productivity [39,40]. This implies that tree species diversity facilitates the soil nutrient uptake by plants [41], potentially resulting in a reduction in nutrients associated with GRSP. Our study also supports this, demonstrating that as plant diversity increased, the elemental content in GRSP decreased.

4.2. Forest Growth Influenced Elemental Storage in TG

We found that plots with large-sized trees and lower-tree-density plots decreased TG and its elemental storage by 11% to 25% and 8% to 67%, respectively, compared to plots with smaller-diameter trees and higher-density plots. Forest growth is commonly characterized by an increase in diameter and a decrease in density, while tree pruning and self-thinning are associated with an increase in UBH. Our results indicate that TG and its elemental storage were negatively correlated with forest growth but positively correlated with pruning-related UBH. Trees with larger diameters can transfer chemical elements from the soil to aboveground vegetation by absorbing large amounts of soil nutrients [42,43]. Soil nutrient elements directly affect the production and chemical composition of GRSP [44,45,46], which, in turn, might affect how tree growth and density influence GRSP and its elemental storage. However, in this study, forest growth had varied effects on the relative content of different elements in TG, which were closely related to soil pH. We found that pH was positively related to the relative content of Na, C, and N in TG but negatively related to Al, K, and Mg (Table 3). This could be due to the impact of tree growth and density on soil pH changes [47,48].

Compared to forest growth characteristics, the spatial assemblage of trees produced weaker effects on element composition in TG. The importance of U, M, and W in regulating forest ecosystem functions has been highlighted in many studies [29,49], but their effects on GRSP are still less known. Our study found that high mixed plots (high M) with smaller neighbor trees (high U) had 7–8% higher TG storage and a 10–13% increase in the storage of the 11 elements in TG than the low mixed plots. However, most of these increases were not statistically significant (p > 0.05). Previous research has indicated that mixed stands could enhance soil development more than pure stands [49]. The presence of smaller neighboring trees indicates less competition for resources and a higher nutrient content, which are conducive to the growth and colonization of AMF and the accumulation of GRSP and its element components [50,51].

Tree species diversity and community structure can directly affect plant nutrient use efficiency [52], thereby altering soil physicochemical properties. The spatial structure often indirectly modifies these soil properties by mediating competition among plants [53]. GRSP varies with changes in soil properties and serves as a reliable indicator of soil health [54]. Previous studies have found that GRSP is closely associated with soil properties, and its accumulation is influenced by geographical climate and vegetation type [55]. In this study, the path analysis further demonstrates that tree diversity and community structure can affect TG and element composition directly or indirectly via soil properties, while the effects of spatial assemblages were relatively minor (16% of the effect was from diversity, 6% was from community structure, and 3% was from soil properties) (Figure 5). These findings elucidate that tree species diversity and community structure have more prominent effects on GRSP than spatial structure.

4.3. Soil pH Importance: Direct Effects and Mediation Roles in Element Composition Variation in TG

Soil pH is frequently regarded as the primary factor due to its significant impact on plant nutrients and growth [56]. Soils with low pH values can reduce the availability of nutrients or cause aluminum toxicity in highly weathered soils [57]. The activity of soil microorganisms, which are responsible for residue decomposition, and the cation exchange capacity of the soil are influenced by soil pH [58]. Diverse forests tend to have better soil nutrient conditions, as indicated by higher pH levels. The variety of tree species contributes positively to the soil’s pH and base saturation [59]. In this study, we also found that tree species diversity was positively correlated with pH (Figure 4). It has been reported that pH could affect the GRSP amount and composition [7]. Forest structure is closely related to soil nutrient changes. Changes in stand density can enhance the positive effects of tree species mixing on tree growth and stand productivity [60]. Afforestation can lead to a change in forest structure and has a significant neutralizing effect on soil pH [48] Our results also testify to this, as soil pH strongly affected TG storage and its element composition (Table 3 and Figure 5). Additionally, the relative content of elements in TG varied with pH, especially for abundant elements such as C and P. Soil pH had both direct effects and mediating roles in element composition variation in TG. As demonstrated by the clustering and pathway analyses, the effects of plant diversity and community structure on the relative content of the 11 elements in TG were pH-dependent (Figure 5). Our study emphasizes that pH is crucial not only for soil nutrients but also in regulating the accrual of mycorrhizal-derived glomalin in soils and various elements in TG.

4.4. Driving Pathways Analysis

By employing RDA ordination and PLSPM analysis, we identified the relative contribution of tree diversity, community structure and spatial assemblage patterns, and soil properties to variations in GRSP elemental composition. We found that soil physicochemical properties exhibit a strong impact, with the largest coefficients from 0.36 to 0.39, which were 2-fold, 5.5-fold, and 31-fold higher than those from forest community structure, tree diversity, and tree spatial assemblage patterns, respectively (Table 7). Moreover, the soil physicochemical properties had the strongest explanatory power (78.8%) for TG storage and its elemental composition, which is considerably higher than that of plant diversity and forest structure (Figure 5). Soil serves as the storage medium for GRSP and can directly affect its accumulation and elemental components [10]. The explanatory power of plant diversity and forest structure might be because of their indirect effects on GRSP variations, primarily through soil factors.

4.5. Implications

Our findings provide an important basis for soil improvement based on GRSP in Northeast China, focusing on the perspectives of element storage and relative content within GRSP. The accumulation of these elements in GRSP improves their sequestration stability in the soil, as microbial residues are known for their high stability in fluctuating environments [7,61]. Some studies found that GRSP is important for soil C sequestration and nutrient retention in Northeast China [62]. Our study reveals that the increase in plant diversity promoted the ability of soil nutrient absorption, resulting in a certain decrease in TG element storage (Table 2 and Figure 5). We also found that forest structure produced greater effects on soil properties and TG storage (Figure 5). Hence, for future forest management in Northeast China, it is recommended to strategically increase plant diversity and construct suitable spatial structures, such as decreasing stand density, to improve the soil C sink and nutrient utilization provided by GRSP.

However, our study also has some limitations. For instance, it was conducted at a single location and did not investigate seasonal changes in the elemental storage and relative content of glomalin. In future work, we should conduct studies at multiple locations and monitor the seasonal changes in the elemental storage and relative content of glomalin influenced by plant diversity and stand structure. This will enhance the representativeness and accuracy of our research findings. Additionally, the specific indices we used in this study to quantify diversity, forest structure, and spatial assemblage may not fully capture the complexity of these ecological factors. In future work, for example, we should consider the effect of the proportion of large and small trees in plots on TG and its elemental composition.

5. Conclusions

The significance of mycorrhizal associations in soil carbon sequestration, nutrient cycling, and heavy metal immobilization has been underscored in previous research. In this study, we define the importance of tree diversity, forest structure, and soil properties in regulating the storage and relative content of 11 elements in GRSP. We found that plant diversity and forest growth affected the storage of these 11 elements. Except for Na, C, and N, the relative content of the remaining elements in GRSP was negatively correlated with pH but positively correlated with EC, tree density, and AP. In contrast, the relative content of Na, C, and N showed opposite patterns. Among soil properties, soil pH was the most significant factor influencing the storage and relative content of the elements in GRSP. Soil properties had a stronger impact on the glomalin traits, as indicated by coefficients from 0.36 to 0.39. Plots with high pH had 26–31% lower elemental storage in GRSP compared to those with low pH, with a similar trend observed for the relative content of the 11 elements. Our findings provide new insights into the factors that affect TG storage and its elemental composition and highlight that increasing tree species diversity and reducing stand density may enhance nutrient utilization associated with GRSP in high-latitude temperate forests.