Fertilization and Residue Management Improved Soil Quality of Eucalyptus Plantations

Abstract

The problem of soil degradation caused by continuous planting of Eucalyptus has a long history in southwest China. It is of great significance to explore reasonable management methods to improve soil quality and forest productivity in Eucalyptus plantations. In this study, the third-generation Eucalyptus plantation in the Qipo state-owned forest farm of Shangsi County, Guangxi Autonomous Region, was used as the research object to explore the effects of fertilization and residue management on soil quality. Therefore, a cross-over test between fertilization (In-O, inorganic fertilizer; O, organic fertilizer; M, mixed fertilizer) and residue management (T, residues were tiled; R, residues were removed; S, residues were stacked) was designed. One-way ANOVA was used to detect the difference in each soil indicator between the three fertilization groups or between the three residue management groups, and two-way ANOVA was used to test whether the indicators were affected by the interaction of these two variables. The soil quality of Eucalyptus plantations was evaluated by principal component analysis (PCA) and the minimum data set (MDS). The results showed that inorganic fertilizer significantly increased the soil nutrient content, such as N, P and K, while organic fertilizer significantly increased soil enzyme activity. Compared with Group R, the retention of residues (T, S) improved the soil pore structure and promoted soil biochemical reactions. The order of soil quality indexing (SQI) was O × T (0.697) > M × T (0.618) > In-O × T (0.557) > O × S (0.490) > M × S (0.439) > O × R (0.362) > In-O × S (0.324) > M × R (0.290) > In-O × R (0.138). Fertilization, residue management and their interaction had significant effects on the soil quality index; among them, residue management was the main factor in the variation of SQI, with a variance contribution rate of 41.6%. In the management of Eucalyptus plantations, soil quality can be improved by applying organic fertilizer and tiling plant residues.

1. Introduction

Eucalyptus is an evergreen tree belonging to the genus Eucalyptus in Myrtaceae, with a tall tree shape and a straight trunk. It is one of the three famous fast-growing artificial species in China. Since 1890, Eucalyptus has been introduced and widely cultivated in red-soil hilly areas and karst landscape areas in southern China. By 2018, its planting area exceeded 5.46 million ha, accounting for about 2.5% of the total forest area; however, it provided one-third of the total wood in China [1,2]. The problem of soil erosion and fertility decline caused by Eucalyptus continuous planting has a long history [3,4]. Morales et al. (2015) showed that the pure Eucalyptus forest would reduce the diversity of understory vegetation and that surface soil and nutrients would be lost in heavy rainfall [5]. The short-cycle management mode consumes a lot of water and fertilizer in the soil of Eucalyptus plantations, and the nutrient elements, such as N, P and K, taken away by wood harvesting cannot be supplemented by nature [6]. Gao et al. (2021) showed that N, P and K were taken away by each harvest of Eucalyptus and were 0.58~1.02 t·ha⁻¹, 0.07~0.12 t·ha⁻¹ and 0.13~0.24 t·ha⁻¹, respectively [7]. Autotoxic substances produced by Eucalyptus roots can also lead to an imbalance of soil microbial flora [8], which in turn affects the biogeochemical cycle of carbon and nitrogen [9]. Therefore, exploring new management models for Eucalyptus plantations is of great significance for improving soil quality, increasing forest productivity and promoting the sustainable development of the wood industry.

The traditional method of burning plant residues can make the nutrients contained in the residues return to the soil surface quickly; however, it also destroys other vegetation, resulting in ash and surface soil loss under heavy rainfall [10]. Retaining plant residues has become the main way to deal with them at present. The effect of residue decomposition on soil physicochemical properties is not absolute. Some studies have shown that the nutrients released by plant residues are not enough to change the soil nutrient content, such as P, K, N and other elements, because these nutrient elements are insignificant compared with the soil nutrient bank [11]. While more reports proved that covering the forest land surface with plant residues was conducive to intercepting precipitation and avoiding soil erosion, which also provides material and space for soil microbial reproduction, thus improving the soil pore structure as well as increasing the soil’s available nutrient content and enhancing microbial metabolic activity [12,13,14], Parhizkar et al. (2021) showed that the lowest runoff was observed for a mulch layer of 3 Mg·ha⁻¹ of rice straw with a length of 200 mm—the lowest soil loss was found with the same application rates but with a 10 mm length in a simulation study [15]. Fertilization can supplement the available nutrients required for the growth of Eucalyptus quickly, which delays the continuous decline in soil fertility caused by continuous planting as much as possible [16,17], and is one of the keys to maintaining the productivity of Eucalyptus plantations.

The influence of single residue management or fertilization on the soil quality of Eucalyptus plantations has attracted wide attention; however, the effects of their interaction on the soil’s physical and chemical properties and microbial metabolic activity have not been reported. It is reasonable to believe that the combined application of these two measures has a positive effect on improving the soil quality of Eucalyptus plantations. Therefore, two sets of variables (residue management and fertilization) were selected, and nine test groups were designed to explore the effects of different management combinations on soil’s physical, chemical and enzyme activities in Eucalyptus plantations. The minimum data set and indicator weight were determined by correlation analysis and principal component analysis, and the soil quality index, under different management combinations, was further calculated. The relative contribution of the residue treatments or fertilization methods and their interactions in improving the soil quality was analyzed. Therefore, three hypotheses were proposed: (1) the retention of residues is beneficial to improve the soil’s physical properties; (2) the application of inorganic fertilizers can increase the soil’s nutrient content, and inorganic fertilizers can increase the soil’s microbial activity; (3) there is an interaction between residue management and fertilization, which is beneficial to the improvement in soil quality. The purpose of this study is to provide effective management measures for the restoration of Eucalyptus plantation soil fertility.

2. Materials and Methods

2.1. Study Area

The study area is located in the Qipo state-owned Forest Farm in Shangsi County, Guangxi Autonomous Region (108°24′4″~108°57′25″ E, 22°10′33″~22°45′25″ N), which is close to Beibu Gulf and borders northern Vietnam. The region is a typical subtropical monsoon climate, with an annual average precipitation of 1200–1300 mm, annual average temperature of 21.7 °C, annual average sunshine hours of 1800~2000 h, and sufficient water and heat, which has obvious seasonal changes and dry–wet alternation characteristics. The region is a typical southeast mountainous hilly landform with an altitude of 195–395 m and a slope of 14–31°. The zonal soil is mainly latosolic red soil with a thickness of 60–100 cm and medium soil fertility, which is developed from Cambrian sandstone, in which where planting contiguous Eucalyptus plantations, the understory vegetation is sparse. Table 1 shows some soil properties before the experiment, which were measured in November 2016.

Table 1. Some soil properties before the experiment.

Indicator	Numerical Value	Indicator	Numerical Value
BD/g·cm−3	1.18 (0.17)	TP/g·kg−1	0.75 (0.18)
NMC/%	16.2 (3.4)	TK/g·kg−1	1.94 (0.52)
TTP/%	38.4 (5.3)	AN/mg·kg−1	20.3 (5.5)
pH	4.61 (0.22)	NN/mg·kg−1	3.94 (0.89)
LLC	73.9 (21.5)	AP/mg·kg−1	1.85 (0.43)
TN/g·kg−1	1.51 (0.32)	AK/mg·kg−1	9.4 (1.7)

The meaning of the abbreviation is shown in Table 4. Data are means, followed by standard deviations in parentheses.

The Eucalyptus plantation was first planted in 2005, with clone DH32-28 cutting seedlings. The first and second harvests were in 2011 and 2016, respectively. In 2017, miscellaneous shrubs and Eucalyptus sprouts were cut down. In June 2018, the third generation of Eucalyptus plantation was planted, with clone DH32-29 cutting seedlings and with a planting density of 1750 trees per ha. From 2018 to 2021, the cumulative fertilization was 3500 kg·ha⁻¹, with an annual fertilization of about 0.5 kg per tree. Table 2 shows the type and quantity of fertilization and the mode of residue management.

Table 2. Basic situation of sample plot and experimental design.

Fertilization	Residue	Orthogonal Group	Total Fertilization Mass/kg·ha−1	Altitude/m	Slope/°
In-O	T	In-O × T	inorganic fertilizer, 3500	377	15
	R	In-O × R		382	12
	S	In-O × S		367	15
O	T	O × T	organic fertilizer, 3500	375	10
	R	O × R		370	15
	S	O × S		364	14
M	T	M × T	mixed fertilizer, 3500	374	15
	R	M × R		382	10
	S	M × S		382	12

2.2. Test Design

Residue management and fertilization were regarded as two variables. The residue management modes comprised (T) residues that were evenly tiled on the surface of the forest land, (R) residues that were removed from the forest land, and (S) residues that were stacked like points on the surface of the forest land. The fertilization methods were as follows: In-O, inorganic fertilizer (65% CH₄N₂O and 35% KH₂PO₄), 0.5 kg per tree per year, N:P:K = 20:12:16; O, organic fertilizer (fermentation of cow manure, sheep manure and corn straw), 0.5 kg per tree per year, N:P:K = 15:6:9; and M, mixed fertilizer (inorganic fertilizer and organic fertilizer mixed at a 1:1 ratio), 0.5 kg per tree per year. A total of nine groups of orthogonal experiments were designed with two variables, and three 20 m × 20 m standard plots were set in each group. Table 2 indicates the basic situation of the sample plots. Table 3 shows the contents of the main nutrient elements in Eucalyptus residues.

Table 3. The main nutrient element content of Eucalyptus residues.

Element	C/g·kg−1	N/g·kg−1	P/g·kg−1	K/g·kg−1
Content	514 (67)	7.15 (1.44)	0.83 (0.15)	6.33 (0.22)

The nutrient elements of Eucalyptus residue are determined after it is crushed. Data are means, followed by standard deviations in parentheses.

2.3. Soil Sampling and Analysis

In July 2021, soil samples were collected during the continuous rainless period. In view of the fact that the topsoil is a nutrient-rich area and relatively susceptible to residues, only 0–20 cm of topsoil was collected in this study. Eight sampling points were selected along the ‘S’ route in each plot. After the humus layer and stones on the soil surface were removed, one core soil (100 cm³) and one mixed soil (500 g) were taken from each sampling point, providing a total of 216 core soil samples and 216 mixed soil samples. The core soil samples were used to determine the soil’s physical properties, some of the mixed soil samples were dried at room temperature for chemical analysis, and the remaining mixed soil samples were stored at 4 °C for other chemical and microbial metabolism analyses. The determination methods of the soil quality indicators are shown in Table 4.

Table 4. Determination methods of soil’s physical, chemical and enzyme activity indicators.

Indicator	Unit	Method [18,19]
Bulk density (BD)	g·cm−3	Soil core method
Natural moisture content (NMC)	%	Soil core method
Mass water-holding capacity (MWC)	%	Soil core immersion method
Effective water-holding capacity (EWC)	%	Soil core immersion method
Capillary porosity (CP)	%	Soil core immersion method
Non-capillary porosity (NCP)	%	Soil core immersion method
The total porosity (TTP)	%	Soil core immersion method
pH		Potentiometry (soil: H2O = 1:2.5)
Leaching liquid conductivity (LLC)		Leaching, potentiometry
Total nitrogen (TN)	g·kg−1	Semi-micro Kjeldahl method
Total phosphorus (TP)	g·kg−1	Alkali fusion-interrupt analyzer
Total kalium (TK)	g·kg−1	Alkali fusion-flame photometer method
Ammonia nitrogen (AN)	mg·kg−1	KCl-interrupt analyzer
Nitrate nitrogen (NN)	mg·kg−1	CaSO4-interrupt analyzer
Available phosphorus (AP)	mg·kg−1	Double acid extraction-molybdenum antimony anti-colorimetric method
Available kalium (AK)	mg·kg−1	Ammonium acetate extraction-flame photometer method
Catalase	nmol·g−1	Potassium permanganate titration
Urease	nmol·g−1	Phenol-sodium hypochlorite colorimetry
Sucrase	nmol·g−1	3,5-Dinitrosalicylic acid colorimetry
Acid phosphatase	10−3 nmol·g−1	Phosphoric acid-disodium benzene colorimetry

2.4. Soil Quality Evaluation Method

2.4.1. Construct MDS

The 20 measured soil indicators (Table 4) were regarded as the total data set (TDS) for the soil quality evaluation. The minimum data set (MDS) with independent variables was constructed by a principal component analysis; only the principal component (PC) with an eigenvalue > 1 was considered for identifying the MDS, and a VARIMAX rotation was performed to enhance the interpretability of the uncorrelated components [20]. In each selected PC, the indicator whose absolute value of factor loading was within 10% of the maximum factor loading was selected as the high-load indicator [18]. When one high-factor loading indicator was left in a PC, it was selected for the MDS. When more than one high-factor loading indicator was left in a PC, a correlation analysis was performed. If the correlation coefficient was less than 0.7, all high-factor loading indicators were selected for the MDS; if the correlation coefficient was greater than 0.7, the high-factor load indicator with the largest sum of correlation coefficients was selected for the MDS [21].

2.4.2. Indicator Scores and Weights

The fuzzy membership function (FMF) was used to standardize the indicators in the MDS [22], and the function type was determined according to the positive and negative effects of each indicator on soil quality. For the indicators with positive effects on the soil quality, use Equation (1); otherwise, use Equation (2).

$$F\left(X_i\right)=\left(X_{ij}-X_{i\mathrm{m}\mathrm{i}\mathrm{n}}\right)/\left(X_{i\mathrm{m}\mathrm{a}\mathrm{x}}-X_{i\mathrm{m}\mathrm{i}\mathrm{n}}\right)$$

(1)

$$F\left(X_i\right)=\left(X_{i\mathrm{m}\mathrm{a}\mathrm{x}}-X_{ij}\right)/\left(X_{i\mathrm{m}\mathrm{a}\mathrm{x}}-X_{i\mathrm{m}\mathrm{i}\mathrm{n}}\right)$$

(2)

where X_ij represents the average value of the jth indicator of the ith test group, and X_imax and X_imin represent the maximum and minimum values of all measured values of the jth indicator, respectively.

The weight of the indicators in the MDS was calculated by principal component analysis [20,23]. The weight of each indicator was equal to the proportion of its principal component variance contribution with the cumulative variance contribution (with an eigenvalue greater than 1).

2.4.3. Calculating SQI

SQI was calculated by Equation (3).

$$SQI=\sum _{i=1}^n{W}_{i}F\left(X_i\right)$$

(3)

where W_i represents the weight of the indicators, F(X_i) represents the indicator scores, and n is the number of indicators in the MDS.

2.5. Statistical Analysis

All data were statistically analyzed by Origin 2021b and SPSS 22.0. Significant differences among the different treatments (within 3 fertilization groups or within 3 residue management groups) for each indicator (Table 4) were analyzed by one-way ANOVA. The effects of different residue management modes, different fertilization methods and their interactions on the physical and chemical soil properties, enzyme activity and soil quality index were analyzed by a two-way ANOVA. Both significance levels were p = 0.05. The correlation between the 20 soil quality indicators was tested using Pearson’s correlation analysis.

3. Results

3.1. Effect of Fertilization on Soil Quality Indicators

The results of one-way ANOVA with fertilization used as a variable showed that the fertilization methods had different effects on the soil physical properties, chemical properties and enzyme activities of Eucalyptus plantations (Table 5). In the soil physical properties, there were no significant differences in BD, NMC, EWC, CP, NCP, and TTP among the different fertilization methods, while MWC was significantly higher in Group O than in Group In-O. In the soil chemical properties, there were no significant differences in the pH, LLC, TP, TK, AN, NN, AP and AK among the different fertilization treatments; however, the values of these indicators in Group In-O were relatively higher than those in Group O and Group M. In the soil enzyme activity, catalase, urease, sucrase and ACP in Group O were significantly higher than those in Group In-O, which increased by 68.4%, 52.8%, 59.2% and 48.3%, respectively. Urease, sucrase and ACP in Group M were significantly higher than those in Group In-O, which increased by 20.7%, 42.4% and 43.6%, respectively.

Table 5. Comparison of soil quality indicators under different fertilization methods.

Indicator	Fertilization Method			Minimum	Maximum	Variation Coefficient
	Group In-O	Group O	Group M
BD	1.19 (0.14)	1.17 (0.12)	1.22 (0.08)	0.96	1.40	0.100
NMC	17.9 (2.8)	19.1 (3.3)	16.2 (3.0)	11.8	26.0	0.184
MWC	32.5 (3.4) b	36.8 (4.0) a	34.8 (4.6) ab	28.0	45.1	0.119
EWC	29.3 (3.6)	30.1 (3.0)	28.7 (2.9)	22.9	37.8	0.123
CP	35.3 (4.9)	36.4 (2.6)	34.7 (4.8)	26.1	41.3	0.122
NCP	7.9 (2.5)	7.5 (2.9)	8.5 (2.3)	3.4	12.0	0.328
TTP	43.2 (6.4)	44.0 (3.2)	43.2 (6.6)	31.3	53.3	0.131
pH	4.54 (0.13)	4.34 (0.16)	4.42 (0.09)	4.07	4.80	0.035
LLC	89.1 (28.5)	86.4 (21.2)	79.6 (12.2)	58.4	134.0	0.259
TN	2.00 (0.30) a	1.49 (0.14) b	1.50 (0.25) b	1.08	2.60	0.204
TP	0.92 (0.23)	0.96 (0.25)	0.87 (0.09)	0.55	1.49	0.227
TK	2.12 (0.57)	1.84 (0.48)	1.95 (0.19)	1.02	3.93	0.271
AN	28.0 (6.19)	21.4 (2.79)	26.9 (5.47)	17.7	38.2	0.228
NN	3.70 (0.80)	5.05 (1.86)	4.24 (0.82)	2.55	8.73	0.318
AP	3.54 (1.21)	2.52 (1.30)	2.80 (0.99)	1.03	3.93	0.271
AK	12.1 (2.74)	10.5 (1.93)	10.1 (2.31)	6.50	17.62	0.232
Catalase	4.05 (1.04) b	6.82 (2.73) a	5.80 (1.38) ab	2.27	11.71	0.394
Urease	0.053 (0.012) c	0.081 (0.030) a	0.064 (0.021) b	0.034	0.147	0.391
Sucrase	2.45 (0.60) b	3.90 (1.19) a	3.49 (0.92) a	1.53	6.19	0.342
ACP	2.34 (0.67) b	3.47 (0.94) a	3.36 (0.93) a	1.49	5.58	0.327

Data are means, followed by standard deviations in parentheses. Different lowercase letters indicate that a certain indicator is significantly different among different fertilization methods (one-way ANOVA, p < 0.05, n = 9).

3.2. Effect of Residue Management on Soil Quality Indicators

The results of one-way ANOVA with residue management used as a variable showed that the residue management modes had different effects on the soil physical properties, chemical properties and enzyme activities of Eucalyptus plantations (Table 6). In the soil physical properties, BD in Group T was significantly lower than that in Group R and Group S, which decreased by 19.1% and 12.4%, respectively. While NMC, MWC, EWC, CP, NCP and TTP in Group T were significantly higher than those in Group R, which increased by 36.2%, 17.4%, 19.2%, 17.8%, 78.9% and 27.3%, respectively. In the soil chemical properties, there was no significant difference in the pH, TN, TK or AN among the residue management modes, and LLC, TP, NN, AP and AK in Group T were significantly higher than those in Group R by 33.8%, 35.4%, 36.1%, 146.8% and 34.7%, respectively. In the soil enzyme activity, the four enzyme activities in Group T were significantly higher than those in Group R, which increased by 110.7%, 91.3%, 75.6% and 43.9%, respectively. At the same time, catalase, urease and sucrase in Group S were also significantly higher than those in Group R, which increased by 37.5%, 34.1% and 39.2%, respectively.

Table 6. Comparison of soil quality indicators under different residue management modes.

Indicator	Residue Management Mode			Minimum	Maximum	Variation Coefficient
	Group T	Group R	Group S
BD	1.06 (0.07) c	1.31 (0.07) a	1.21 (0.03) b	0.96	1.40	0.100
NMC	20.7 (2.9) a	15.2 (2.1) c	17.7 (2.1) b	11.8	26.0	0.184
MWC	39.2 (3.3) a	33.4 (3.2) b	32.9 (2.6) b	28.0	45.1	0.119
EWC	32.3 (2.7) a	27.1 (3.2) b	28.5 (2.8) b	22.9	37.8	0.123
CP	37.8 (2.8) a	32.1 (4.4) b	35.9 (3.6) ab	26.1	41.3	0.122
NCP	10.2 (1.4) a	5.7 (2.1) c	7.7 (2.1) b	3.4	12.0	0.328
TTP	48.0 (3.3) a	37.7 (4.1) c	43.7 (3.9) b	31.3	53.3	0.131
pH	4.40 (0.21)	4.46 (0.06)	4.43 (0.15)	4.07	4.80	0.035
LLC	105.2 (21.9) a	78.6 (12.4) b	71.2 (13.5) b	58.4	134.0	0.259
TN	1.80 (0.44)	1.55 (0.26)	1.65 (0.22)	1.08	2.60	0.204
TP	1.07 (0.22) a	0.79 (0.13) b	0.89 (0.15) ab	0.55	1.49	0.227
TK	2.11 (0.57)	2.25 (0.68)	1.90 (0.46)	1.02	3.93	0.271
AN	28.9 (6.7)	25.0 (4.6)	22.4 (3.8)	17.7	38.2	0.228
NN	5.16 (1.83) a	3.79 (0.77) b	4.04 (0.82) b	2.55	8.73	0.318
AP	4.32 (0.49) a	1.75 (0.38) c	2.80 (1.00) b	1.03	3.93	0.271
AK	13.2 (2.3) a	9.8 (1.8) b	9.6 (1.6) b	6.50	17.62	0.232
Catalase	7.46 (2.26) a	3.54 (0.67) c	5.67 (1.08) b	2.27	11.71	0.394
Urease	0.088 (0.029) a	0.046 (0.008) c	0.064 (0.014) b	0.034	0.147	0.391
Sucrase	4.25 (1.06) a	2.42 (0.60) c	3.17 (0.76) b	1.53	6.19	0.342
ACP	3.80 (1.08) a	2.64 (0.85) b	2.73 (0.53) ab	1.49	5.58	0.327

Data are means, followed by standard deviations in parentheses. Different lowercase letters indicate that a certain indicator is significantly different among different residue management modes (one-way ANOVA, p < 0.05, n = 9).

3.3. Effect of Interaction between Fertilization and Residue Management on Soil Quality Indicators

The results of the two-way ANOVA with fertilization and residue management used as variables showed that the effects of fertilization or residue on the soil physical properties, chemical properties and enzyme activities in Eucalyptus plantations were basically consistent with the results of the one-way ANOVA (Table 7). The interaction of these two variables had different effects on 20 soil quality indicators of Eucalyptus plantations (Table 7), where the interaction between the two variables showed significant effects on the variations of soil CP, LLC, AN, NN, AP, catalase, urease and ACP.

Table 7. Two-way ANOVA of the effects of fertilization, residues and their interactions on soil properties.

Indicator	Source of Variation	df	Mean Square	F Value	p Value
BD	Fertilization	2	0.006	1.867	0.183
	Residue	2	0.145	42.102	0.000 **
	Fertilization × Residue	4	0.005	1.456	0.257
NMC	Fertilization	2	19.864	3.439	0.054 *
	Residue	2	67.242	11.643	0.001 **
	Fertilization × Residue	4	3.477	0.602	0.666
MWC	Fertilization	2	18.998	1.803	0.193
	Residue	2	110.459	10.481	0.001 **
	Fertilization × Residue	4	6.817	0.647	0.636
EWC	Fertilization	2	4.484	0.612	0.553
	Residue	2	64.449	8.801	0.002 **
	Fertilization × Residue	4	20.902	2.854	0.054
CP	Fertilization	2	8.982	0.888	0.429
	Residue	2	76.491	7.566	0.004 **
	Fertilization × Residue	4	38.572	3.815	0.02 *
NCP	Fertilization	2	2.176	0.516	0.605
	Residue	2	44.87	10.644	0.001 **
	Fertilization × Residue	4	3.001	0.712	0.594
TTP	Fertilization	2	4.972	0.312	0.736
	Residue	2	239.435	15.005	0.000 **
	Fertilization × Residue	4	22.123	1.386	0.278
pH	Fertilization	2	0.089	4.852	0.021 *
	Residue	2	0.009	0.466	0.635
	Fertilization × Residue	4	0.03	1.612	0.215
LLC	Fertilization	2	214.721	1.952	0.171
	Residue	2	2878.221	26.168	0.000 **
	Fertilization × Residue	4	1239.382	11.268	0.000 **
TN	Fertilization	2	0.767	14.381	0.000 **
	Residue	2	0.145	2.722	0.093
	Fertilization × Residue	4	0.083	1.551	0.23
TP	Fertilization	2	0.021	0.722	0.499
	Residue	2	0.185	6.485	0.008 **
	Fertilization × Residue	4	0.064	2.228	0.107
TK	Fertilization	2	1.187	4.174	0.032 *
	Residue	2	0.188	0.662	0.528
	Fertilization × Residue	4	0.219	0.772	0.558
AN	Fertilization	2	111.729	10.775	0.001 **
	Residue	2	98.177	9.468	0.002 **
	Fertilization × Residue	4	75.417	7.273	0.001 **
NN	Fertilization	2	4.16	8.741	0.002 **
	Residue	2	4.758	9.998	0.001 **
	Fertilization × Residue	4	6.212	13.053	0.000 **
AP	Fertilization	2	2.484	16.958	0.000 **
	Residue	2	15.092	103.016	0.000 **
	Fertilization × Residue	4	1.114	7.605	0.001 **
AK	Fertilization	2	11.241	3.582	0.049 *
	Residue	2	36.677	11.686	0.001 **
	Fertilization × Residue	4	4.961	1.581	0.222
Catalase	Fertilization	2	17.82	36.59	0.000 **
	Residue	2	34.665	71.176	0.000 **
	Fertilization × Residue	4	4.011	8.237	0.001 **
Urease	Fertilization	2	0.002	6.262	0.009 **
	Residue	2	0.004	14.167	0.000 **
	Fertilization × Residue	4	0.001	5.178	0.015 *
Sucrase	Fertilization	2	5.048	13.328	0.000 **
	Residue	2	7.684	20.289	0.000 **
	Fertilization × Residue	4	0.424	1.119	0.379
ACP	Fertilization	2	3.535	6.411	0.008 **
	Residue	2	3.753	6.806	0.006 **
	Fertilization × Residue	4	0.636	5.153	0.014 *

* indicates that the source of variation has a significant effect on soil indicators (p < 0.05), and ** indicates that the source of variation has an extremely significant effect on soil indicators (p < 0.01).

3.4. Soil Quality Assessment

3.4.1. Determining the MDS

Principal component analyses of 20 selected soil quality indicators showed that the eigenvalues of four principal components were greater than 1, the variance contribution rates were 44.73%, 20.15%, 9.90% and 6.78%, respectively, and the cumulative variance contribution rate was 81.56%. The extracted principal components explained most of the variations occurring from the indicators (Table 8). PC1 had three high-load indicators, including sucrase, catalase and urease. The correlation coefficients of sucrase with catalase and urease were all greater than 0.7, and the sum of the correlation coefficients of sucrase with catalase and urease was the largest (Figure 1); then, it was selected and placed into the MDS (Table 9). PC2 had three high-load indicators, too, including CP, EWC and TTP. The correlation coefficients of CP with EWC and TTP were all greater than 0.7, and the sum of the correlation coefficients of CP with EWC and TTP was the largest (Figure 1); then, it was selected and placed into the MDS (Table 9). Both PC3 and PC4 have only one high-load indicator, TP and AN, respectively, which were selected and placed into the MDS (Table 9). These four indicators are highly representative. Through a PCA, the dimension of the data set was reduced, and the interference of the correlation between the indicators was excluded.

Table 8. Principal component factor rotation load matrix, eigenvalue and variance contribution rate.

	PC1	PC2	PC3	PC4
Eigenvalue	8.060	3.630	1.784	1.222
Percent	44.73	20.15	9.90	6.78
Cumulative percent	44.73	64.88	74.78	81.56
Eigenvector
Sucrase	0.880	0.282	0.051	−0.197
Catalase	0.839	0.180	0.262	−0.313
Urease	0.798	0.242	0.214	−0.236
ACP	0.768	0.073	0.080	0.022
MWC	0.659	0.449	−0.044	0.200
NCP	0.649	0.264	0.137	0.292
NN	0.619	−0.241	0.546	−0.122
PH	−0.564	0.224	−0.102	0.328
CP	0.034	0.932	0.217	−0.070
EWC	0.166	0.894	0.166	0.161
TTP	0.306	0.842	0.264	0.073
TP	0.295	−0.006	0.826	−0.240
AP	0.318	0.341	0.698	0.269
TN	−0.234	0.181	0.686	0.221
LLC	0.157	0.391	0.599	0.297
BD	−0.522	−0.459	−0.582	−0.178
NMC	0.290	0.453	0.569	0.048
AK	0.050	0.392	0.541	0.491
AN	−0.009	0.144	0.168	0.874
TK	−0.300	−0.069	0.059	0.634

Figure 1. Correlation of soil quality indicators.

Table 9. Indicator scores and weights in MDS.

Item	Sucrase	CP	TP	AN
In-O × T	0.334 (0.048)	0.921 (0.088)	0.617 (0.119)	0.866 (0.164)
In-O × R	0.052 (0.050)	0.265 (0.121)	0.213 (0.109)	0.219 (0.116)
In-O × S	0.207 (0.071)	0.533 (0.271)	0.367 (0.277)	0.423 (0.017)
O × T	0.807 (0.149)	0.612 (0.132)	0.754 (0.181)	0.142 (0.076)
O × R	0.239 (0.083)	0.736 (0.172)	0.228 (0.179)	0.258 (0.076)
O × S	0.480 (0.078)	0.698 (0.158)	0.346 (0.025)	0.146 (0.016)
M × T	0.612 (0.133)	0.777 (0.158)	0.309 (0.142)	0.635 (0.094)
M × R	0.279 (0.102)	0.190 (0.159)	0.333 (0.047)	0.592 (0.195)
M × S	0.370 (0.177)	0.729 (0.179)	0.383 (0.067)	0.115 (0.010)
Weights	0.549	0.247	0.121	0.083

Data are means, followed by standard deviations in parentheses.

3.4.2. Calculating Soil Quality Index

The weight of the indicator in the MDS was equal to the proportion of its principal component variance contribution with the cumulative variance contribution. The weights of sucrase, CP, TP and AN were 0.549, 0.247, 0.121 and 0.083, respectively (Table 9). The indicator scores in the MDS were calculated by Equation (1) (Table 9), and the soil quality indexes from different test groups were calculated by Equation (3). The soil quality index of 27 plots ranged from 0.117 to 0.816, with an average of 0.435 and a coefficient of variation of 41.7% (Figure 2), and the soil quality indexes of nine orthogonal test groups were as follows: O × T (0.697) > M × T (0.618) > In-O × T (0.557) > O × S (0.490) > M × S (0.439) > O × R (0.362) > In-O × S (0.324) > M × R (0.290) > In-O × R (0.138). In the residue management modes, the soil quality index in Group T was significantly higher than that in Group S and Group R. In the fertilization methods, the soil quality index in Group O was significantly higher than that in Group In-O.

Figure 2. Soil quality index of different management combinations (different lowercase letters indicate there are significant differences among different management combinations (one-way ANOVA, p < 0.05, n = 3)).

The results of the two-way ANOVA showed that fertilization and residue management had significant effects on the soil quality index, and the interaction between them also had significant effects on the soil quality index (Table 10). Among them, the residue management model explained 41.6% of the soil quality index variation, the fertilization method explained 25.7% of the soil quality index variation, and the interaction between them explained 16.2% of the soil quality index variation (Figure 3). This shows that the residue management mode is the main factor leading to the variations in the soil quality index within the scope of the experimental design.

Table 10. Results of two-way ANOVA on soil quality index by fertilization, residue and their interaction.

Indicator	Source of Variation	df	Mean Square	F Value	p Value
SQI	Fertilization	2	0.072	10.785	0.007 **
	Residue	2	0.295	26.189	0.000 **
	Fertilization × Residue	4	0.002	8.256	0.013 *

* indicates that the source of variation has a significant effect on SQI (p < 0.05), and ** indicates that the source of variation has an extremely significant effect on SQI (p < 0.01).

Figure 3. Variance decomposition of variation in soil quality index.

4. Discussion

4.1. Effects of Fertilization on Soil Physical and Chemical Properties and Enzyme Activities

A long-term application of nitrogen-based inorganic fertilizers would lead to soil compaction and reduce soil permeability in Eucalyptus plantations [24]. Zhao et al. showed that the soil porosity and water-holding capacity of Eucalyptus plantations with a continuous application of nitrogen fertilizer decreased by 10.3%~17.1% and 9.4%~15.0% [25], respectively. In this study, although there was almost no significant difference in the soil physical properties between the inorganic fertilizer group and the organic fertilizer group, the application of inorganic fertilizer still increased soil BD and reduced soil porosity and MWC, while the organic fertilizer relatively improved soil pore structure, which was similar to the previous research results [26,27].

The soil pH was different among the inorganic fertilizer group, organic fertilizer group and mixed fertilizer group. By the action of microorganisms, inorganic fertilizers continuously release humic acid and fulvic acid into the soil [28], resulting in a decrease in soil pH, which is consistent with this study, although there is no significant difference in the pH between the three fertilization methods. Inorganic fertilizer can quickly supplement the nutrient elements needed by plants in the soil and increase the total amount of soil cations. In this study, the contents of LLC, TN, TP, AN, AP and AK in the inorganic fertilizer group were higher than those in the organic fertilizer group. Meanwhile, NN was opposite to them, which may be the result of organic fertilizer under microbial nitrification [29].

Soil enzymes are a type of efficient catalytic protein secreted by plant roots and microorganisms, and their activity reflects the efficiency of soil biochemical reactions [30]. Organic fertilizer enriches the species and quantity of soil microorganisms and provides material and space for microbial reproduction [31]. Chang et al. (2007) showed that the application of organic fertilizer increased soil urease activity by 10.2%~68.5% and catalase activity by 49.2%~145.3% [32]. In this study, the long-term application of organic fertilizer increased soil catalase, urease, sucrase and ACP activity (Table 5). This is consistent with the previous research results. This biochemical process provides a steady stream of nutrients for plant growth, although it is less efficient than inorganic fertilizers.

4.2. Effects of Residue Management on Soil’s Physical and Chemical Properties and Enzyme Activities

In undisturbed forest ecosystems, plant residues play an important role in improving soil pore structure, maintaining ecosystem stability and promoting nutrient cycling [33]. Covering plant residues on the surface of forestland avoids the direct erosion of precipitation, especially in the young forest period. The microenvironment, composed of plant residues and soil, can reduce soil moisture evaporation, prevent soil compaction and improve soil permeability [34]. In this study, compared with removing residues, spreading them on the soil surface significantly reduced soil BD and increased soil porosity, natural water content, effective water-holding capacity and maximum water-holding capacity (Table 6), which is consistent with previous experience [34,35].

Long-term studies have shown that removing plant residues limits nutrient supply and plant growth. In the process of decomposition of plant residues, organic carbon, N, P, K and other nutrients are delivered to the surface soil [36,37]. At the same time, the coverage of residues on the surface of forestland can reduce surface runoff, thus reducing soil nutrient loss [38]. In this study, the effect of retention residues on the soil pH was similar to that of organic fertilizer, and the pH was lower in Group T than in Group R. Soil LLC, TP, AN, NN, AP and AK in Group T were significantly higher than those in Group R, indicating that the decomposition of Eucalyptus residues provided N-, P- and K-available nutrients to the soil, or slowed down the removal of these nutrients and other cations by surface runoff.

Eucalyptus residue also provides material and space for soil microbial reproduction [39], which is similar to the effect of organic fertilizer on soil enzyme activity. Retaining residues can significantly improve soil enzyme activity, and it is more significant than organic fertilizer [14]. The study of Eucalyptus plantations in the Guangxi Autonomous Region by Zhao et al. (2022) [40] showed that soil microbial diversity and enzyme activity were higher when the residues were retained than burned. In this study, the activities of catalase, urease, sucrase and ACP in soil increased by 37.5%~110.7%, 34.1%~91.3%, 39.2%~75.6% and 4.2%~39.2%, respectively, when Eucalyptus residues were retained.

4.3. Soil Quality Evaluation of Eucalyptus Plantation

Soil quality evaluation results are different if the indicator system and evaluation methods are different [20,41]. The physical, chemical and biological properties of soil are usually selected as indicators for soil quality evaluation, and sometimes site conditions and climate are also considered [42,43]. Legaz et al. (2017) summarized the indicators and methods for an effective soil quality evaluation based on different objectives [44]. In this study, 20 physical, chemical and enzyme activity soil indicators were used to evaluate the soil quality of Eucalyptus plantations. Sucrase, CP, TP and AN were selected for the MDS by a PCA. Sucrase represents the intensity of the soil’s biochemical reaction, CP represents soil permeability and water-holding capacity, and TP and AN reflect the soil’s nutrient supply capacity. These four indicators can basically represent soil quality. The effect of fertilization on the soil quality of Eucalyptus plantations was significant. The effect of inorganic fertilizer was mainly reflected in the supply capacity of soil nutrients, while organic fertilizer was reflected in the soil’s biochemical process. Therefore, the soil quality index in Group O was significantly higher than that in Group In-O. Residue management also had a significant effect on the soil quality of Eucalyptus plantations. Compared with removing the residues, retaining them could improve the soil pore structure, increase the soil nutrient content and promote soil biochemical reactions. The soil quality index in Group T was significantly higher than that in Group S and Group R, which indicated that the arrangement of the residues also affected the soil quality. Compared with point stacking, tiling residues could affect the soil indicators and soil quality in a wider range. Analyzing the contributions of the two variables in the context of the SQI variation in Eucalyptus plantations has not been reported. The management of residues is the main factor leading to the SQI change, which is consistent with its influence on soil indicators.

5. Conclusions

Fertilization increased the soil nutrient supply capacity, and organic fertilizer further accelerated the soil biochemical reaction rate (Table 5). The effect of the retained residues on soil is mainly reflected in improving the soil pore structure, increasing soil nutrient supply capacity, and promoting soil biochemical reactions (Table 6). The interaction of fertilization and residue also affected the soil quality indicators and showed significant variations in soil CP, LLC, AN, NN, AP, catalase, urease and ACP. (3) The soil quality index of the Eucalyptus plantation was: O × T (0.697) > M × T (0.618) > In-O × T (0.557) > O × S (0.490) > M × S (0.439) > O × R (0.362) > In-O × S (0.324) > M × R (0.290) > In-O × R (0.138), which indicated that the soil quality index was higher when the residues were retained than when they were removed, which was 18.6%–33.5% higher, and the soil quality index was higher when applying organic fertilizer than when applying mixed fertilizer and inorganic fertilizer, which was 7.9%–16.6% higher. The result of the two-way ANOVA showed that the residue management method significantly changed the soil quality, and its contribution reached 41.6%. (4) In the management of Eucalyptus plantations, soil quality can be improved by applying organic fertilizer and tiling plant residues.