Effects of Mixing Hippophae rhamnoides and Pinus tabuliformis on Ecosystem Carbon, Nitrogen, and Phosphorus Sequestration and Storage Capacity in the Loess Hilly Region, China

Abstract

Mixed forests created by incorporating nitrogen-fixing tree species offer enhanced ecological advantages compared with forests consisting of only one type of tree species. These benefits include habitat rehabilitation and the promotion of biodiversity. Nevertheless, the impact of introducing nitrogen-fixing tree species on ecosystem carbon (C), nitrogen (N), and phosphorus (P) sequestration and storage capacity in the Loess Plateau of China remains inadequately explored. To examine changes in the sequestration and storage capacity of ecosystem C, N, and P, the mixed plantations of P. tabulaeformis and H. rhamnoides (H_rP_t) were selected as the research object, and the pure plantations of H. rhamnoides (H_r) and P. tabulaeformis (P_t) were selected as the control. The results indicated that in comparison to the pure forest, the ecosystem in H_rP_t significantly increased C and N stocks but decreased P stocks. In addition, C, N, and P stocks in the soil layer accounted for more than 60% of the C, N, and P stocks in the pure and mixed forest ecosystems compared with the vegetation layer. Moreover, H_rP_t significantly improved ecosystem C and N sequestration rates relative to the pure forest but decreased P sequestration rates. Furthermore, the soil physicochemical properties can be inferred from the redundancy analysis showing 66.79% and 0.06% in H_r, 44.84% and 0.06% in P_t, as well as 44.28% and 0.04% in H_rP_t, respectively. In conclusion, compared with the pure forest, the introduction of N-fixing tree species was more conducive to the accumulation of C and N. The results offer substantial significance for the scientific guidance of vegetation restoration in degraded landscapes and the stewardship of mixed forests in the Loess Hilly Region, providing essential data for nutrient storage in ecosystems.

1. Introduction

Forest ecosystem C stock accounts for about 70% of soil C stock in terrestrial ecosystems and plays an important role in global warming [1]. N, P, and other nutrients are important factors for maintaining plant survival and growth. They interact with the structural element C and change cooperatively in the process of plant growth, jointly restricting the key processes of nutrient cycling between plants and soil [2]. Forest ecosystems accumulate organic compounds with long C residence time and contribute to global C cycling [3]. Therefore, exploring and analyzing ecosystem C, N, and P stocks can help us determine the function of ecosystem nutrient cycling as well as the relevance between elements and the ecosystem [4].

Mixed-species tree planting has attracted widespread attention due to its superior stability and sustainability compared with pure forest planting, and it is becoming increasingly popular worldwide [5]. Conversely, changes in the sequestration and storage capacity of ecosystem nutrients will impact ecosystem functions and processes. It is essential to conduct a thorough evaluation of the nutrient sequestration and storage capacity in ecosystems that are in the process of ecological restoration as this will aid in comprehending the ecological effects of mixed afforestation [6]. This assessment could have a considerable impact on the enduring availability of nutrients and the ongoing sustainable stewardship of plantations.

The impact of different tree species on nutrient levels in mixed forests varies [7]. Introducing N₂-fixing tree species into Eucalyptus plantations in subtropical China increased soil C and N contents [8]. Soil organic C in Pinus massoniana can be changed by mixing cultivation with Erythrophleum fordii [9]. It can be seen that a reasonable mixed planting pattern is conducive to improving the nutrient status of the ecosystem. However, the existing research primarily concentrates on the impact of mixed planting on alterations in nutrient content [10]. Enhancing the comprehension of the influence on C, N, and P sequestration, as well as their storage capacity, is crucial for ecosystem nutrient availability and cycling.

Mixed planting of P. tabulaeformis and H. rhamnoides has become a widely distributed planting method in the Loess Plateau [11]. Research has indicated that combining P. tabulaeformis and H. rhamnoides in a mixed forest significantly and positively affected community structure, species diversity, and soil physical and nutrient conditions [12]. Furthermore, the planting method is beneficial for promoting an increase in the leaves’ N:P stoichiometric ratio and enhancing the nutrient content of C, N, and P in P. tabulaeformis [13]. Nevertheless, the existing study on the impact of mixed afforestation of P. tabulaeformis and H. rhamnoides primarily focuses on changes in nutrient content and stoichiometric characteristics. The response of C, N, and P storage to the mixed forest of H. rhamnoides and P. tabulaeformis at the ecosystem level remains unclear.

In this study, we investigated the response of P. tabuliformis to mixed afforestation of N-fixing species by measuring C, N, and P stocks in tree layers, litter layers, and 0–200 cm soil layers. Our objective is to (1) analyze the changes in ecosystem C, N, and P sequestration, as well as their storage capacity between pure and mixed forests, and (2) identify the primary influencing elements of sequestration and their storage capacity. The research is helpful for the ecological restoration of vegetation degradation and the sustainable management of mixed forests in the Loess Hilly Region.

2. Materials and Methods

2.1. Site Description

The study was conducted at the Ansai National Ecological Experimental Station (36°51′30″ N, 109°19′23″ E), which is located in Shaanxi Province, China. The region belongs to the typical temperate semi-arid climate with a mean annual temperature of 8.8 °C and an average annual rainfall of (458 ± 172.5) mm. The primary soil type is loessial soil, which was developed from wind-deposited loessial parent material, and has a homogeneous silt loam texture. Moreover, the soil was a salt loam (USDA NRCS) with 9.28% clay (<0.002 mm), 61.08% silt (0.002–0.05 mm), and 29.65% sand content (>0.05 mm). In addition, the median soil grain size, soil bulk density, and soil cohesion were 36.82 μm, 1.19 g cm⁻³, and 12.79 kPa, respectively [14], which correlates with “Entisols” in the American soil taxonomy (1975) and “Regosols” in the United Nations world soil tutu unit/FAO/UNESCO (1988) [15]. The main woody plants are H. rhamnoides, P. tabuliformis, Robinia pseudoacacia, and Platycladus orientalis. Herbaceous plants mainly include Bothriochloa ischaemum, Artemisia gmelinii, etc. The original plantation density was widely used in the central part of the Loess Plateau: the stem densities of P_t and H_r in pure plantations were 1667 and 3300 stems ha⁻¹, respectively. The stem densities in the corresponding mixed plantations were 1111 and 1900 stems ha⁻¹. Since the artificial afforestation began in 2000, environmental changes or human activities may have led to poor tree growth or mortality. Consequently, during the growth process, the initial densities have been changed. Based on the previous survey in April 2018, the existing stem densities of P_t and H_r were 1200 and 1733 stems ha⁻¹, respectively. The existing stem densities of H_rP_t were 850 and 1775 stems ha⁻¹.

2.2. Experimental Material

Based on the long-term fixed monitoring forest land in the field of the mountain test site, the mixed plantations of P. tabulaeformis and H. rhamnoides (H_rP_t) were selected as the research object, and the pure plantations of H. rhamnoides (H_r) and P. tabulaeformis (P_t) were selected as the control (Figure 1). Each plot was planted around 2000. The fundamental properties of the plots are displayed in

Table 1. The basic characteristics of the sample plots.

Sample Plots	Tree Species	Tree Height (m)	DBH/DGH (cm)	Crown Diameter of (EW × SN/m × m)	Area of Sample Plot (m × m)
HrPt	Hippophae rhamnoides	3.08 ± 0.14	5.43 ± 0.33	(2.01 ± 0.09) × (2.04 ± 0.11)	20 × 20
	Pinus tabuliformis	4.01 ± 0.17	7.11 ± 0.48	(2.62 ± 0.12) × (2.60 ± 0.14)	20 × 20
Hr	Hippophae rhamnoides	3.46 ± 0.12	5.96 ± 0.33	(2.40 ± 0.12) × (2.29 ± 0.11)	20 × 20
Pt	Pinus tabuliformis	3.69 ± 0.09	7.43 ± 0.27	(2.96 ± 0.12) × (3.36 ± 0.11)	20 × 20

Note: P_t and H_r represent the pure plantations of P. tabuliformis and H. rhamnoides, respectively. H_rP_t represents the mixed plantations of P. tabuliformis with H. rhamnoides. DBH and DGH represent diameter at breast height and ground height, respectively. The abbreviations mean the same below. The data in the table are mean ± standard deviation. The survey of basic information of sample plots was conducted before the growing season (April).

.

In this study, for each forest type, three repeated plots were selected, and each plot was 20 m × 20 m for observation. The research was carried out during the 2018 growing season (May–October).

2.3. Sampling and Laboratory Analyses

To investigate the alterations in the sequestration and storage capacity of C, N, and P in P. tabuliformis after mixed afforestation with H. rhamnoides, plant, litter, and soil sampling was conducted in May–October 2018 at the end of each month. Therefore, a total of 6 times samples were collected.

According to the test results of each tree in each plot, 4 standard trees were selected according to average DBH/DGH, average tree height, and average crown width, and leaf, stem, and root samples of each standard tree were collected at the end of each month (healthy mature leaves were selected at the crown with high branch shears and picked and mixed in 4 directions, east, west, north, and south). After the leaves are removed, they are stems. The sampling depth of the roots was set to 2 m, which was consistent with the depth of the nutrient content. In each sample plot, three representative litter samples of 1 m × 1 m were selected according to the diagonal method; the litter was collected using a 20 cm high grid and positioned above the ground. A soil core of 2 m depth (0–10, 10–20, 20–50, 50–100, 100–150, and 150–200 cm) was drilled into the corresponding sample plot by using the soil drilling method. For each P_t plot, three 2 m soil drills were carried out. Therefore, there were 9 soil samples for pure P_t. The design and sampling method for H_r and H_rP_t are the same as P_t, so there are 27 soil samples in H_r, P_t, and H_rP_t. About 300 g of each sample was collected. After being brought back to the laboratory, the samples of leaf, stem, root, and litter were dried at 85 °C to constant weight, crushed with a grinder, and used for the measurement of plant C, N, and P contents and litter biomass. Samples of soil were split into two parts. Before measuring the C, N, and P contents, one part was allowed to be air-dried naturally, and the rocks and plant remnants were removed. The remaining part was stored in a refrigerator at 4 °C to be analyzed for other physicochemical properties of the soil.

The undisturbed soil was obtained by the soil drilling method and then drained to constant weight on the sand after soaking in water for a day and night. After oven-dried at 105 °C for 48 h until constant weight, the soil bulk density (SBD), soil total porosity (STP), soil capillary porosity (SCP), soil non-capillary porosity (SNCP), soil water content (SWC), and field capacity (FC) were calculated. The pH value was determined by the potentiometry method (HJ 962-2018). C was determined by the wet oxidation method; N was determined using the Kjeldahl method. The HClO₄-H₂SO₄ colorimetric and NaHCO₃-H₂SO₄ colorimetric methods were employed to measure P and the available P (AP), respectively. Furthermore, continuous flow injection analysis was used to assess the amounts of ammonium N (AN) and nitrate N (NN) [16]. The fundamental soil physicochemical properties of the measured soil are detailed in

Table 2. Physicochemical properties in the soil profile under pure and mixed forests.

Soil Layer (cm)	Sample Plots	STP (%)	SCP (%)	SNCP (%)	SWC (%)	SBD (g cm−3)	FC (%)	pH	NN (g kg−1)	AN (g kg−1)	AP (g kg−1)
0–10	Hr	0.545	0.424	12.140	0.086	1.204	19.175	8.390	0.931	1.841	0.614
	Pt	0.515	0.383	13.240	0.104	1.138	17.942	8.800	0.213	1.863	0.310
	HrPt	0.519	0.382	13.710	0.072	1.225	17.141	8.319	0.389	1.493	0.406
10–20	Hr	0.553	0.424	12.980	0.094	1.127	19.470	8.297	0.357	1.980	0.410
	Pt	0.471	0.360	11.100	0.082	1.345	15.287	8.877	0.258	1.860	0.355
	HrPt	0.518	0.394	12.340	0.089	1.172	18.211	8.400	0.335	1.407	0.367
20–50	Hr	0.457	0.348	10.830	0.082	1.423	14.326	8.497	0.224	1.487	0.323
	Pt	0.477	0.369	10.790	0.099	1.355	15.447	8.898	0.147	1.733	0.304
	HrPt	0.503	0.386	11.660	0.088	1.281	16.562	8.500	0.152	1.268	0.366
50–100	Hr	0.468	0.351	11.690	0.101	1.386	14.626	8.593	0.295	1.394	0.343
	Pt	0.484	0.363	12.100	0.104	1.368	15.509	8.995	0.123	1.725	0.346
	HrPt	0.529	0.421	10.860	0.094	1.294	18.081	8.490	0.166	1.364	0.364
100–150	Hr	0.467	0.358	10.840	0.104	1.486	14.361	8.680	0.256	1.265	0.350
	Pt	0.445	0.335	11.020	0.098	1.440	13.669	9.020	0.112	1.971	0.316
	HrPt	0.512	0.401	11.130	0.065	1.266	17.517	8.530	0.146	1.547	0.274
150–200	Hr	0.492	0.383	10.930	0.105	1.418	15.857	8.790	0.128	1.171	0.337
	Pt	0.448	0.340	10.840	0.092	1.400	14.061	9.100	0.103	1.697	0.361
	HrPt	0.515	0.403	11.220	0.070	1.250	17.881	8.650	0.112	1.359	0.286

Note: STP, soil total porosity; SCP, soil capillary porosity; SNCP, soil non-capillary porosity; SWC, soil water content; SBD, soil bulk density; FC, field capacity; pH, soil acidity and alkalinity; NN, nitrate nitrogen; AN, ammonium nitrogen; AP, available phosphorus. The abbreviations mean the same below.

.

2.4. Data Processing

2.4.1. Biomass of the Tree Layer

The biomass of each forest type was calculated as follows [17] (

Table 3. Allometric equations are used to calculate the biomass of the various tree components.

Sample Plots	Component	Allometric Equation	R2
Hr	Leaf Stem Root	WL = 0.006 (D2H)0.8403	0.986
		WS = 0.0302 (D2H)0.9474	0.990
		WR = 0.0119 (D2H)0.9501	0.957
Pt	Leaf Stem Root	WL = 0.05 (D2H)0.6308	0.989
		WS = 0.0485 (D2H)0.8510	0.993
		WR = 0.0298 (D2H)0.7866	0.994
HrPt(Hr)	Leaf Stem Root	WL = 0.0193 (D2H)0.7608	0.840
		WS = 0.0477 (D2H)0.8526	0.986
		WR = 0.0293 (D2H)0.8025	0.953
HrPt(Pt)	Leaf	WL = 0.0262 (D2H)0.7727	0.967
	Stem	WS = 0.048 (D2H)0.8494	0.997
	Root	WR = 0.0301 (D2H)0.7767	0.990

Note: D and H represent diameter at breast height and ground height, respectively.

):

2.4.2. C, N, P Stocks and Sequestration Rate

The C, N, and P stocks of leaves, stems, and roots were obtained by multiplying the biomass of leaves, stems, and roots by the C, N, and P contents of leaves, stems, roots, and the corresponding stand density.

The C, N, and P stocks of the litter layer were obtained by multiplying the litter layer’s dry mass by the litter layer’s C, N, and P contents.

The C, N, and P stocks of each soil layer were calculated as follows [18]:

$$C_{istock}=0.1\times C_i\times B{D}_i\times E_i$$

(1)

$$N_{istock}=0.1\times N_i\times B{D}_i\times E_i$$

(2)

$$P_{istock}=0.1\times P_i\times B{D}_i\times E_i$$

(3)

where C_i, N_i, and P_i are the C, N, and P contents of the ith layer of soil (g kg⁻¹), respectively. BD_i is the SBD of the ith layer (g cm⁻³) and E_i is the soil depth of the ith layer (cm). The soil stock of the 0–200 cm layer was equal to the sum of the six layers.

The C, N, and P sequestration rates were calculated as follows [19]:

$$C_{sr}=(S{A}_{1}CS-S{A}_{2}CS)/SAI$$

(4)

$$N_{sr}=(S{A}_{1}NS-S{A}_{2}NS)/SAI$$

(5)

$$P_{sr}=(S{A}_{1}PS-S{A}_{2}PS)/SAI$$

(6)

where C_sr, N_sr, and P_sr are the C sequestration rate (Mg ha⁻¹ year⁻¹), N sequestration rate (Mg ha⁻¹ year⁻¹), and P sequestration rate (Mg ha⁻¹ year⁻¹), respectively. SA₁CS, SA₂CS, SA₁NS, SA₂NS, SA₁PS, and SA₂PS represent stand age₁ C stock (Mg ha⁻¹), stand age₂ C stock (Mg ha⁻¹), stand age₁ N stock (Mg ha⁻¹), stand age₂ N stock (Mg ha⁻¹), stand age₁ P stock (Mg ha⁻¹), and stand age₂ P stock (Mg ha⁻¹), respectively. SAI is the stand age interval.

2.5. Statistical Analysis

One-way analysis of variance (ANOVA) polynomial linear and Tukey’s multiple comparison method were selected to compare significant differences in the biomass of the tree layer and litter layer and the C, N, and P contents, and stocks of the tree layer, litter layer, and soil layer between the pure and mixed forests. Linear fitting was used to analyze the coupling relationship between ecosystem C, N, and P stocks in pure forest and mixed afforestation models. Correlation analysis was used to study the influencing factors of ecosystem C, N, and P stocks. RDA redundancy analysis was used to research the explanatory factors of soil physicochemical properties for C, N, and P stocks in the research plot ecosystem. Finally, all statistical analyses were performed using SPSS 26.0 software, and a redundancy analysis (RDA) diagram was drawn by Canoco 5, In addition, all the remaining diagrams were drawn with Origin 2022 software.

3. Results

3.1. The Tree Biomass

There were significant differences in the biomass between a pure forest and a mixed forest (Figure 2). The biomass (kg) between pure and mixed forests for leaves, stems, and roots varied from 0.84 to 2.91, 3.38 to 12.02, and 1.59 to 5.58, respectively. In contrast to P_t, H_rP_t significantly increased the biomass of leaves, stems, and roots, with increases of 97.96%, 150.42%, and 169.57%, respectively (p < 0.05). Furthermore, the comparison against H_r, reveals that H_rP_t has significantly increased the biomass of leaves, stems, and roots, with increases of 246.43%, 255.62%, and 250.94%, respectively (p < 0.05).

3.2. C, N, and P Contents

3.2.1. Tree and Litter Layers

There were significant differences in the C, N, and P contents of the tree layer (leaves stems, and roots) and the litter layer between the pure forest and the mixed forest (

Table 4. C, N, and P contents in tree and litter layers of the pure and mixed forest.

Component	Sample Plots	C (g kg−1)	N (g kg−1)	P (g kg−1)	C:N	C:P	N:P
Leaf	Hr	487.15 ± 10.55 a	35.06 ± 0.57 a	1.79 ± 0.04 a	13.91 ± 0.21 a	275.44 ± 6.29 a	19.79 ± 0.34 a
	HrPt(Hr)	475.29 ± 6.13 a	35.12 ± 0.54 a	1.83 ± 0.04 a	13.57 ± 0.11 a	263.29 ± 4.54 a	19.44 ± 0.36 a
	Pt	549.08 ± 10.57 a	10.97 ± 0.34 b	0.82 ± 0.03 b	54.69 ± 2.36 a	729.77 ± 26.37 a	13.57 ± 0.27 a
	HrPt(Pt)	531.64 ± 7.40 a	12.40 ± 0.27 a	0.99 ± 0.05 a	43.88 ± 1.32 b	569.04 ± 20.57 b	13.15 ± 0.41 a
Stem	Hr	491.20 ± 5.09 a	16.43 ± 0.72 b	1.03 ± 0.06 a	31.62 ± 1.28 a	517.86 ± 22.08 a	16.64 ± 0.64 b
	HrPt(Hr)	472.52 ± 11.99 a	17.88 ± 0.54 a	0.96 ± 0.05 a	27.15 ± 0.98 a	516.96 ± 19.65 a	19.31 ± 0.58 a
	Pt	545.97 ± 5.56 a	4.14 ± 0.32 a	0.37 ± 0.02 a	157.19 ± 10.15 a	1536.07 ± 57.65 a	11.47 ± 0.97 a
	HrPt(Pt)	537.29 ± 4.46 a	4.17 ± 0.30 a	0.37 ± 0.02 a	152.26 ± 9.90 a	1553.78 ± 65.78 a	11.65 ± 0.90 a
Root	Hr	483.58 ± 10.55 a	29.94 ± 1.02 a	0.69 ± 0.04 a	16.57 ± 0.48 b	798.89 ± 57.05 a	47.84 ± 2.88 a
	HrPt(Hr)	488.34 ± 10.89 a	25.25 ± 1.39 b	0.69 ± 0.03 a	35.79 ± 9.89 a	751.05 ± 34.62 a	39.07 ± 2.74 b
	Pt	474.17 ± 9.89 a	5.76 ± 0.23 a	0.58 ± 0.02 b	90.05 ± 5.80 a	865.58 ± 42.83 a	10.29 ± 0.56 a
	HrPt(Pt)	473.96 ± 12.46 a	6.54 ± 0.33 a	0.62 ± 0.03 a	81.38 ± 5.56 a	850.67 ± 58.62 a	10.91 ± 0.52 a
Litter layer	Hr	458.64 ± 5.98 A	25.18 ± 0.53 A	1.12 ± 0.03 A	18.36 ± 0.27 A	416.34 ± 10.16 A	22.72 ± 0.50 A
	Pt	527.93 ± 7.50 a	7.18 ± 0.31 b	0.49 ± 0.02 b	77.86 ± 3.44 a	1129.73 ± 41.20 a	14.74 ± 0.32 b
	HrPt	431.38 ± 17.09 Ab	21.31 ± 1.06 aB	0.91 ± 0.03 aB	21.32 ± 0.74 Ab	482.50 ± 20.02 Ab	23.49 ± 1.06 Aa

Note: Among leaf, stem, and root components, different lowercase letters indicate that there are significant differences between pure and mixed forests at 0.05 levels, and different capital letters indicate that there are significant differences between H_r and H_rP_t at 0.05 levels.

). There was no significant difference in the C content of the tree layer between the pure forest and the mixed forest. The N contents of leaves and the litter layer between the pure and mixed plantations of P. tabuliformis varied from 10.97 to 12.40 and from 7.18 to 21.31 (g kg⁻¹). In comparison to the P_t, H_rP_t(P_t) resulted in a significant increase in the N content of leaves and the litter layer, amounting to 13.04% and 196.80%, respectively (p < 0.05). The range of N content in the stems of pure and mixed H. rhamnoides plantations was 16.43 to 17.88 (g kg⁻¹). When compared with H_r, H_rP_t(H_r) led to a significant increase in stem N content, amounting to 8.83% (p < 0.05). The P contents of leaves and roots in the pure and mixed plantations of P. tabuliformis varied from 0.82 to 0.99 and from 0.58 to 0.62 (g kg⁻¹). In contrast to the P_t, H_rP_t(P_t) significantly increased the P content of leaves and roots, with increases of 20.73% and 6.90%, respectively (p < 0.05). In comparison to P_t, H_rP_t(P_t) significantly increased the P content of the litter layer, with an increase of 85.71% (p < 0.05). The stoichiometric ratio C:N and N:P in the roots and stems of pure and mixed H. rhamnoides plantations were within the ranges of 16.57 to 35.79 and 16.64 to 19.31 (g kg⁻¹). As opposed to the H_r, H_rP_t(H_r) significantly improved the C:N for roots and N:P for stems, with increases of 115.99% and 16.05%, respectively (p < 0.05). The N:P for the litter layer between the pure and mixed plantations of P. tabuliformis varied from 14.74 to 23.49 (g kg⁻¹). Moreover, the N:P for the litter layer significantly improved by 59.36%, with H_rP_t(P_t) in contrast to P_t. (p < 0.05).

3.2.2. Soil Layer

For the 0–10 cm soil layer in both pure and mixed plantations, the content ranges of soil elements (g kg⁻¹) were 3.57 to 12.79 for C, 0.18 to 0.65 for N, and 0.51 to 0.54 for P, as per (

Table 5. C, N, and P contents in different soil layers of the pure and mixed forests.

Soil Layer (cm)	Sample Plots	C Content (g kg−1)	N Content (g kg−1)	P Content (g kg−1)	C:N	C:P	N:P
0–10	Hr	12.79 ± 0.75 A	0.65 ± 0.03 A	0.54 ± 0.01 A	19.61 ± 0.62 A	23.71 ± 1.40 A	1.20 ± 0.05 A
	Pt	8.78 ± 0.33 b	0.45 ± 0.02 b	0.53 ± 0.01 a	19.59 ± 0.60 a	16.86 ± 0.74 b	0.87 ± 0.04 b
	HrPt	10.78 ± 0.52 Aa	0.53 ± 0.01 Aa	0.54 ± 0.01 Aa	20.43 ± 1.13 Aa	20.17 ± 0.88 Aa	1.01 ± 0.03 Aa
10–20	Hr	6.83 ± 0.37 A	0.35 ± 0.02 A	0.52 ± 0.01 A	19.55 ± 0.55 A	13.16 ± 0.68 A	0.68 ± 0.03 A
	Pt	7.52 ± 0.40 a	0.38 ± 0.01 a	0.53 ± 0.01 a	19.92 ± 0.95 a	14.43 ± 0.89 a	0.73 ± 0.03 a
	HrPt	6.06 ± 0.29 Ab	0.31 ± 0.01 Ab	0.51 ± 0.01 Aa	19.34 ± 0.71 Aa	11.94 ± 0.65 Ab	0.61 ± 0.02 Ab
20–50	Hr	4.78 ± 0.36 A	0.23 ± 0.01 A	0.51 ± 0.01 A	20.92 ± 0.91 A	9.40 ± 0.67 A	0.44 ± 0.02 A
	Pt	5.11 ± 0.27 a	0.27 ± 0.02 a	0.51 ± 0.01 a	22.19 ± 3.10 a	10.18 ± 0.58 a	0.53 ± 0.03 a
	HrPt	4.28 ± 0.21 Ab	0.22 ± 0.01 Ab	0.51 ± 0.01 Aa	20.04 ± 1.23 Aa	8.51 ± 0.48 Aa	0.43 ± 0.01 Ab
50–100	Hr	3.99 ± 0.19 A	0.18 ± 0.01 A	0.52 ± 0.01 A	22.20 ± 1.11 A	7.74 ± 0.41 A	0.36 ± 0.01 A
	Pt	3.66 ± 0.15 a	0.19 ± 0.00 a	0.51 ± 0.00 a	19.63 ± 0.78 a	7.23 ± 0.34 a	0.37 ± 0.01 a
	HrPt	3.57 ± 0.15 Aa	0.19 ± 0.00 Aa	0.51 ± 0.01 Aa	18.92 ± 0.68 Aa	6.97 ± 0.31 Aa	0.37 ± 0.01 Aa
100–150	Hr	3.81 ± 0.19 A	0.19 ± 0.01 A	0.52 ± 0.01 A	20.64 ± 0.99 A	7.38 ± 0.42 A	0.36 ± 0.01 A
	Pt	3.74 ± 0.18 a	0.19 ± 0.00 a	0.52 ± 0.01 a	20.33 ± 1.09 a	7.28 ± 0.41 a	0.36 ± 0.01 a
	HrPt	3.68 ± 0.19 Aa	0.18 ± 0.00 Aa	0.53 ± 0.01 Aa	20.32 ± 0.98 Aa	6.95 ± 0.38 Aa	0.34 ± 0.01 Aa
150–200	Hr	3.73 ± 0.14 A	0.19 ± 0.01 A	0.53 ± 0.01 A	19.34 ± 0.71 A	7.13 ± 0.33 A	0.37 ± 0.01 A
	Pt	3.60 ± 0.20 a	0.18 ± 0.00 a	0.53 ± 0.01 a	19.92 ± 1.02 a	6.91 ± 0.42 a	0.34 ± 0.01 a
	HrPt	3.66 ± 0.17 Aa	0.18 ± 0.00 Aa	0.53 ± 0.00 Aa	20.04 ± 0.84 Aa	6.90 ± 0.32 Aa	0.34 ± 0.01 Aa

Different lowercase letters indicate that there are significant differences between P_t and H_rP_t at 0.05 levels, and different capital letters indicate that there are significant differences between H_r and H_rP_t at 0.05 levels.

). H_rP_t significantly increased the contents of C and N by 22.78% and 17.78%, respectively, compared with P_t, with no significant difference in the N content. Furthermore, H_rP_t significantly improved the stoichiometric ratio C:P and N:P in the soil layer (0–10 cm), with increases of 19.63% and 16.09%, respectively. Furthermore, there was no significant difference in C, N, P, C:N, C:P, and N:P between H_r and H_rP_t in the same soil layer.

3.3. Effects of Mixed Afforestation on C, N, and P Stocks

3.3.1. Tree Layer

There were significant differences in the C, N, and P stocks of the tree layer between the pure forest and the mixed forest (Figure 3).

The mean stocks (kg ha⁻¹) between pure and mixed plantations are in the range of 709.99 to 4103.96 for leaves, 2880.61 to 17,111.32 for stems, and 1176.12 to 7215.91 for roots in C stocks; 19.40 to 271.08 for leaves, 23.83 to 569.54 for stems, and 14.28 to 373.03 for roots in N stocks; 1.45 to 14.09 for leaves, 2.16 to 30.49 for stems, and 1.45 to 10.15 for P stocks. In relation to P_t, H_rP_t(P_t) significantly increased the C stock of leaves, stems, and roots, with increases of 322.85%, 444.11%, and 495.47%, respectively (p < 0.05). Contrasting with H_r, H_rP_t(H_r) significantly increased the C stock of leaves, stems, and roots, with increases of 416.76%, 422.42%, and 441.19%, respectively (p < 0.05). When compared with P_t, H_rP_t(P_t) led to a significant increase in the N stock in leaves, stems, and roots, with increases of 393.30%, 457.07%, and 577.17%, respectively (p < 0.05). The comparison with H_r revealed that H_rP_t(H_r) significantly improved the N stock in leaves, stems, and roots, with increases of 430.59%, 491.30%, and 351.83% respectively (p < 0.05). Moreover, in comparison to P_t, H_rP_t(P_t) led to a significant increase in the P stock across leaves, stems, and roots, with increases amounting to 429.66%, 449.07%, and 529.66% respectively. (p < 0.05). In comparison to H_r, H_rP_t(H_r) significantly increased the P stock of leaves, stems, and roots, with increases of 439.85%, 402.31%, and 434.21%, respectively (p < 0.05).

3.3.2. Litter Layer

There were significant differences in the C, N, and P stocks of the litter layer between the pure forest and the mixed forest (Figure 4). The mean stocks (kg ha⁻¹) in the litter layer of pure and mixed plantations are in the range of 9.86 to 38.79 for C stock, 0.13 to 1.80 for N stock, and 0.01 to 0.08 for P stock. In comparison to P_t, H_rP_t significantly improved the C, N, and P stocks of the litter layer, with increases of 262.98, 1284.62%, and 700%, respectively (p < 0.05). In addition, in relation to H_r, H_rP_t significantly improved the C and N stocks of the litter layer, with increases of 36.34 and 25%, respectively (p < 0.05). Moreover, there was no significant difference in the litter layer between H_r and H_rP_t (Figure 4c).

3.3.3. Soil Layer

For the 10 cm soil layer, the mean stock (Mg ha⁻¹) in pure and mixed plantations for C, N, and P stocks are within 9.99 to 15.40, 0.52 to 0.78, and 0.60 to 0.66 ranges respectively (Figure 5). In comparison to P_t, H_rP_t significantly improved the C, N, and P stocks of the 0–10 cm soil layer, with increases of 32.13, 25, and 10%, respectively (p < 0.05). Nevertheless, the stocks of C, N, and P in soil layers at depths other than those mentioned showed a downward trend, with the C stock displaying significant differences except for the 10–20 and 20–50 cm layers; the N stock, except for the 0–10 and 50–100 cm layers; and the P stock, except for the 0–10 cm layer (p < 0.05). In contrast to H_r, H_rP_t significantly decreased the C stock in 20–50, 50–100, and 100–150 cm soil layers, the N stock in 0–10, 100–150, and 150–200 cm soil layers, and the P stock, except for 0–10 and 10–20 cm soil layer (p < 0.05).

3.3.4. C, N, and P Stocks and Distribution Patterns in the Plantation Ecosystem

There were significant differences in the C, N, and P stocks of the vegetation layer, soil layer (0–200 cm), and the ecosystem between the pure forest and the mixed forest (

Table 6. C, N, and P stocks and distribution patterns of pure and mixed forests in the plantation ecosystem.

Plantation Ecosystem	Sample Plots	C Stock (Mg ha−1)	N Stock (Mg ha−1)	P Stock (Mg ha−1)
Vegetation layer	Hr	4.89 ± 0.05 B	0.22 ± 0.004 B	0.01 ± 0.0003 B
	Pt	5.26 ± 0.06 b	0.06 ± 0.002 b	0.005 ± 0.0001 b
	HrPt	53.68 ± 0.62 Aa	1.50 ± 0.03 Aa	0.08 ± 0.001 Aa
Soil layer (0–200 cm)	Hr	127.24 ± 6.19 A	6.17 ± 0.16 A	14.70 ± 0.18 A
	Pt	120.54 ± 5.58 a	6.08 ± 0.16 a	14.32 ± 0.16 a
	HrPt	107.83 ± 4.88 Ba	5.40 ± 0.08 Bb	13.26 ± 0.14 Bb
Ecosystem	Hr	132.14 ± 6.18 B	6.40 ± 0.16 B	14.71 ± 0.18 A
	Pt	125.80 ± 5.59 b	6.14 ± 0.15 b	14.33 ± 0.16 a
	HrPt	161.50 ± 4.94 Aa	6.90 ± 0.09 Aa	13.34 ± 0.14 Bb

Note: Different lowercase letters indicate that there are significant differences between P_t and H_rP_t at 0.05 levels, and different capital letters indicate that there are significant differences between H_r and H_rP_t at 0.05 levels.

). The mean stocks (Mg ha⁻¹) in the vegetation layer of pure and mixed plantations are in the range of 4.89 to 53.68 for the C stock, 0.06 to 1.50 for the N stock, and 0.005 to 0.08 for the P stock. In contrast to P_t, H_rP_t significantly improved the C, N, and P stocks of the vegetation layer, with increases of 920.53, 2400, and 1500%, respectively (p < 0.05). In comparison to H_r, H_rP_t substantially increased the C, N, and P stocks of the vegetation layer, with increases of 997.75, 581.82, and 700%, respectively (p < 0.05). The mean stocks (Mg ha⁻¹) in the soil layer (0–200 cm) of pure and mixed plantations are in the range of 107.83 to 127.24 for the C stock, 5.40 to 6.17 for the N stock, and 13.26 to 14.70 for the P stock. However, in comparison to P_t, H_rP_t significantly decreased N and P stocks, and in relation to H_r, H_rP_t significantly decreased C, N, and P stocks (p < 0.05). Furthermore, the mean stocks (Mg ha⁻¹) in the ecosystem of pure and mixed plantations are in the range of 125.80 to 161.50 for the C stock, 6.14 to 6.90 for the N stock, and 13.34 to 14.71 for the P stock. In H_r, soil layers (0–200 cm) accounted for 96.29%, 96.41%, and 99.93% of the ecosystem C, N, and P stocks, respectively. In P_t, soil layers (0–200 cm) accounted for 95.82%, 99.02%, and 99.93% of the ecological system C, N, and P stocks. In H_rP_t, soil layers (0–200 cm) accounted for 66.77%, 78.26%, and 99.40% of the ecological system C, N, and P stocks, respectively. In contrast to P_t, H_rP_t significantly improved the C and N stocks of the ecosystem, with increases of 28.38 and 12.38%, respectively (p < 0.05). In comparison to H_r, H_rP_t significantly increased the C and N stocks of the ecological system, with increases of 22.22 and 7.81%, respectively (p < 0.05). On the contrary, a contrasting trend was observed for the P stock where significant differences between mixed and pure forests were noted.

3.4. Factors Correlated with the C, N, and P Stocks, and the Relationships of C, N, and P in Stocks

3.4.1. Relationships of C, N, and P in Stocks and Soil Physicochemical Properties

The RDA analysis revealed that Axis-1 and Axis-2 explained 66.79% and 0.06% in H_r, 44.84% and 0.06% in P_t, and 44.28% and 0.04% in H_rP_t, respectively, of the variation in the C, N, and P stocks in the ecosystem (Figure 6). In H_r, the C stock was negatively correlated with SWC, the N stock was negatively correlated with SWC, SBD, and pH and positively correlated with FC, NN, AN, AP, and temperature. However, the P stock was not correlated with soil physicochemical properties. In P_t, the N stock was negatively correlated with SBD and positively correlated with most soil physicochemical properties. In H_rP_t, the C stock was negatively correlated with SCP and FC, and the N stock was negatively correlated with SCP, SBD, and SNCP. NN and AP were positively correlated, while P stock was positively correlated with SWC (Figure 7).

3.4.2. Regression Relationships of C, N, and P in Stocks

The relationships among C, N, and P stocks exhibited diverse trends between the pure forest and the mixed forest (Figure 8). In H_r, significant relationships were found between the shifts in the C and N stocks (R² = 0.41, p < 0.01, y = 25.05x − 28.11; Figure 8a), and P and N stocks (R² = 0.12, p < 0.05, y = 0.39x + 12.19; Figure 8c), but there was no significant correlation between C and P stocks. In P_t, a significant relationship was found in C and P stocks (R² = 0.17, p < 0.05, y = −14.58x + 334.66; Figure 8e), but there was no significant correlation between C, N stocks, and P, N stocks. In H_rP_t, significant relationships were found between the shifts in the C and N stocks (R² = 0.16, p < 0.05, y = 21.58x + 12.56; Figure 8g), and P and N stocks (R² = 0.27, p < 0.01, y = 0.78x + 7.96; Figure 8i), but there was no significant correlation between C and P stocks.

3.5. C, N, and P Sequestration Rate

There were significant differences in the C, N, and P sequestration rates of ecosystems between the pure forest and the mixed forest (

Table 7. C, N, and P sequestration rates in the ecosystem of pure and mixed forests.

	Sample Plots	C (Mg ha−1 year−1)	N (Mg ha−1 year−1)	P (Mg ha−1 year−1)
Ecosystem	Hr	8.81 ± 0.41 B	0.43 ± 0.01 B	0.98 ± 0.01 A
	Pt	8.39 ± 0.37 b	0.41 ± 0.01 b	0.96 ± 0.01 a
	HrPt	10.77 ± 0.33 Aa	0.46 ± 0.01 Aa	0.89 ± 0.01 Bb

Note: Different lowercase letters indicate that there are significant differences between P_t and H_rP_t at 0.05 levels, and different capital letters indicate that there are significant differences between H_r and H_rP_t at 0.05 levels.

). The mean sequestration rates (Mg ha⁻¹ year⁻¹) in the ecosystem of pure and mixed plantations are in the range of 8.39 to 10.77 for C, 0.41 to 0.46 for N, 0.89 to 0.98 for P stock. In comparison to P_t, H_rP_t significantly enhanced the C and N sequestration rates of the ecosystem, with increases of 28.37 and 12.20%, respectively (p < 0.05). In contrast to H_r, H_rP_t significantly increased the C and N sequestration rates of the ecosystem, with increases of 22.25 and 6.98%, respectively (p < 0.05). Nonetheless, the P sequestration rate of the ecosystem displayed a downward trend, with a significant difference in P levels being observed.

4. Discussion

4.1. Effects of Mixed Afforestation on C, N, and P Stocks

In the forest ecosystem, replanting N-fixing tree species in artificial pure forests can improve soil C stocks after the formation of mixed forests [20]. In comparison to the pure forest, H_rP_t significantly improved the C stock of the tree layer (Figure 3). It may be that interspecific interactions in mixed forests enhance nutrient absorption and utilization by tree species, leading to increased C stock in the tree layer [21]. After P_t and H_r mixed planting, there may be N-sharing between non-N-fixing and N-fixing tree species. The N-fixation of non-N-fixing plants can promote the secretion of organic acids, causing soil acidification in the rhizosphere. This acidification boosts the activity of insoluble P in the soil, improving the plants’ P absorption capacity [22].

Similarly, in comparison to P_t, H_rP_t significantly enhanced the C, N, and P stocks in the litter layer and 0–10 cm soil layer (Figure 4 and Figure 5). This increase may be due to H_r being a deciduous broad-leaved forest tree species with rich litter decomposition [23]. C and N stocks decreased in the 10–20 cm soil layer in contrast to the surface layer due to increased litter accumulation. Deeper soils store slightly more C and N than the top 20 cm. H_r had higher C and N stocks than H_rP_t in deeper layers, likely influenced by plant roots. Fine particles in deep soils restrict root exudate processes. Furthermore, there was no significant difference in the P stock between H_r and H_rP_t because the effective P produced by litter decomposition was rapidly absorbed by plants [24].

4.2. Effects of Mixed Afforestation on C, N, and P Stocks Distribution Pattern

It is of great significance to study the distribution of nutrient elements in the mixed forest ecosystem with drought-resistant tree species to indicate its stability [25]. In relation to H_r and P_t, H_rP_t significantly improved C, N, and P stocks in the vegetation layer. However, in the 0–200 cm soil layer, there was a decrease in C, N, and P stocks for H_rP_t in contrast to H_r and P_t (Table 6). At the ecosystem level, H_rP_t increased C and N stocks but decreased P stocks significantly compared with H_r and P_t (p < 0.05) (Table 6). C, N, and P stocks in the 0–200 cm soil accounted for over 95% of the ecosystem for H_r and P_t, and for H_rP_t, they accounted for 66.77%, 78.26%, and 99.40%, respectively. Such a pattern of reserve allocation is similar to the high proportion of C allocated to the underground found by predecessors in dry years [26]. More C is allocated by trees from underground in an environment with limited water resources [27]. After over 15 years, the dense forest canopy resulting from long-term afforestation may intensify forest competition, decrease soil organic matter fixation, and alter C storage in the mixed soil layer. Thus, optimizing the mixed forest ratio can mitigate interspecific effects [28]. At the same time, trees in these conditions have narrow root systems rather than shorter branches [29]. N-fixing such as H. rhamnoides store N in 0–200 cm soil layers with their deep roots, while non-N-fixing species benefit from N competition through root exudations or mycorrhizal networks. Drought significantly reduces N transfer from H. rhamnoides to Populus tomentosa [30]. Accordingly, the distribution of H_rP_t is speculated to be influenced by drought, affecting soil N transfer from H_r to P_t. P stock distribution is mainly influenced by the soil matrix. After mixed planting, increased litter leads to rapid plant absorption of available P from litter decomposition [31]. Soil P fluidity is weaker than that of soil C and N, resulting in distinct P stock distribution patterns.

4.3. Factors Controlling C, N, and P Stocks after Mixed Afforestation

In contrast to the pure forests, the quantity and quality of litter and root exudates in mixed forests will vary, impacting soil physicochemical properties and soil microbial composition and potentially influencing C storage [32]. There was a significant positive correlation between the C stock and the N stock in H_r and H_rP_t(H_r). Higher TN levels positively impacted organic C stock by reducing the C-N ratio in the litter. This reduction helps avoid competition for N between microorganisms and plants, facilitating litter mineralization and decomposition and, ultimately, promoting the C stock [33]. In H_r, there was a significant negative correlation between the C stock and soil water content, which may be attributed to the lack of soil moisture in the Loess Hilly Region and increased soil porosity, promoting the mineralization and decomposition of organic C, which hinders organic C [34]. The study results indicated a significant negative correlation between the N stock of H_r, P_t, and H_rP_t and SBD, which may be due to the increase in SBD, resulting in low soil porosity, while the exchange of N in the soil is blocked, resulting in a decrease in the N stock. The N stock was significantly and positively correlated with NN. AN and NN are the main forms of effective N absorbed by roots. H. rhamnoides are a few woody plants with N-fixing ability in arid areas [35]. Although H. rhamnoides is not a leguminous plant, it can form a relationship with N-fixing bacteria. These bacteria grow on H. rhamnoides roots to create nodules that fix N from the atmosphere, explaining the positive correlation between NN and N stock. The N stock in H_r and H_rP_t is also positively correlated with AP. H. rhamnoides root exudates have high acid substance content, lowering soil pH, which, in turn, affects soil P chemical form. Changes in inorganic P components significantly increase AP content [36].

The interaction of SOC, TN, and TP plays a key role in the sustainability of terrestrial ecosystems [37]. In this study, there is a partial correlation between the C, N, and P stocks of H_r, P_t, and H_rP_t, which is further processed by linear regression. (Figure 6). The results showed that there was a linear relationship between C and N stocks in H_r and H_rP_t, and the C stock increased with the increase in the N stock (Figure 8a,g). The coupling of SOC stock and TN is due to the decomposition of soil organic matter [38]. A higher N content in the soil decreases soil microorganism activity and mineralization rate, leading to higher soil C content [39]. In addition, there is a linear relationship between P stock and N stock in H_r and H_rP_t, and the P stock increases with the increase in the N stock, but the C stock of P_t decreases with the increase in the P stock. This may be due to the influence of soil N stock by N fixation, deposition input, and inorganic N leaching [40]. Previous studies have also suggested that soil N and P dynamics may strictly limit the accumulation of soil C during vegetation replanting in natural terrestrial ecosystems [41].

4.4. Effects of Mixed Afforestation on C, N, and P Sequestration Rate

Integrating and estimating organic C sequestration rates in mixed plantations helps managers accurately monitor changes in forest C pools and allocate effectively [42]. In comparison to H_r and P_t, H_rP_t significantly increases ecosystem C and N sequestration rates (p < 0.05). The C sequestration rate of the H_rP_t ecosystem in this study area surpasses the annual average rate of Shaanxi Province, possibly due to the faster growth rate of H_rP_t leading to increased C input until reaching equilibrium. N-fixing tree species like H. rhamnoides can enhance N availability in ecosystems by fixing atmospheric N, enriching the soil, and improving N storage rates [43]. However, the P sequestration rate in the H_rP_t ecosystem decreased in contrast to pure forest due to the high demand for P by biological N-fixation, despite N-fixing tree species promoting the accumulation of soil organic P fractions. The interaction between species-specific soil microbial composition of different mixture species was not considered [44]. A significant proportion of soil P stock in the ecosystem contributes to the decline in the P sequestration rate.

5. Conclusions

Afforestation is an important measure to achieve vegetation restoration and sustainable development in the Loess Plateau of China. This paper discusses the effects of mixed afforestation on C, N, and P sequestration and storage capacity, as well as their controlling factors in drought-resistant tree species ecosystems in this region. The results showed that in comparison to P_t H_rP_t significantly increased C, N, and P stocks in the vegetation layer and 0–10 cm soil layer. In contrast with H_r, H_rP_t significantly enhanced the C, N, and P stocks in the vegetation layer. But in the 20–200 cm soil layer, the C, N, and P stocks of H_rP_t were lower than those of H_r and P_t. Moreover, in contrast with the pure forest, H_rP_t significantly increased ecosystem C and N stocks but decreased P stocks. In addition, C, N, and P stocks in the soil layer accounted for more than 60% of the C, N, and P stocks in the pure and mixed forest ecosystems compared with the vegetation layer. At the same time, in comparison to pure forest, H_rP_t significantly increased ecosystem C and N sequestration rates but decreased P sequestration rates. Furthermore, the soil physicochemical properties can be inferred from the redundancy analysis, with 66.79% and 0.06% in H_r, 44.84% and 0.06% in P_t, and 44.28% and 0.04% in H_rP_t, respectively. The research establishes a foundational theory for the rational and sustainable management of regional artificial mixed-species restoration efforts. However, due to the intricate and variable nature of internal ecological environments in mixed forest ecosystems, our understanding of nutrient mechanisms may be limited by sampling data collected only during periods of vigorous growth. Therefore, future research should focus on enhancing the frequency of inter-annual or intra-annual observations, coupled with isotope technology, to investigate the influence mechanism of introducing N-fixing tree species on ecosystem reserves, sequestration capacity, and storage capacity.