Sustainable Restoration of Typical Degraded Grasslands: An Evaluation of Ecological Benefits from Bio-Organic Fertilizer Applications

Abstract

To investigate the effects of restoration measures on degraded grasslands in the typical steppe of Inner Mongolia, two restoration treatments—enclosure (EN) and enclosure with additional bio-organic fertilizer application (EF)—were established with moderately grazed degraded grassland as the control (MG). The impacts of these measures on vegetation community characteristics and soil physicochemical properties were systematically analyzed, and the post-restoration ecological benefits were quantitatively evaluated using an assessment system. The results showed that (1) over the two experimental years, both EN and EF treatments improved vegetation community characteristics and soil physicochemical properties, with the overall restoration effect ranking as EF > EN > MG; (2) compared with the first experimental year, the restoration of plant and soil characteristics was more significant in the second year; (3) ecological benefit analysis showed that the Soil Quality Index (SQI) decreases with soil depth. Compared to MG, the EF treatment increased topsoil SQI by 38.89–65.71% in the first and second years of restoration, respectively. Comparative evaluation indicated that EF led to more significant increases in vegetation community characteristics than MG. Thus, the EF treatment can significantly enhance the ecological benefits of typical degraded grasslands.

1. Introduction

Driven by climate change and overgrazing, global grassland ecosystem degradation is intensifying, threatening regional ecological security and constraining sustainable development goals centered on grassland the “ecology–production–livelihoods” synergy; while China has achieved partial phased progress in grassland degradation control (e.g., increased vegetation coverage in some areas), overall efforts still face long-term challenges [1], and moderately degraded grasslands in Inner Mongolia—critical for ecology (windbreak, sand fixation, and sustaining animal husbandry) yet complex to restore—are key regional restoration targets but lack targeted research. Both fertilization (scientific use of inorganic, organic–inorganic compound, and farmyard manure-based organic fertilizers) and enclosure with grazing prohibition are core degraded grassland restoration measures: the former boosts soil nutrients and improves vegetation community structure and productivity [1], while the latter promotes natural grassland recovery by blocking overgrazing. However, existing studies mostly focus on their individual effects, leaving a notable gap in research on the “enclosure + bio-organic fertilizer”-combined restoration mechanism for Inner Mongolia’s moderately degraded grasslands—especially the insufficient exploration of its integrated role in “ecological protection–resource utilization–sustainable restoration” (e.g., soil–vegetation–benefit-coupled responses)—which limits theoretical support for precise grassland restoration and sustainable development in this region. In this study, the Soil Quality Index (SQI) model was used for soil quality assessment, with a comparative evaluation method integrated to analyze ecological benefits; the SQI model, designed for quantitative soil quality evaluation in degraded ecosystems, builds a comprehensive index to reflect soil health by selecting core soil physical and chemical indicators, making it well-suited for soil quality monitoring and restoration effect evaluation in ecologically sensitive areas like grasslands. It outperforms other methods and, unlike single-index approaches (e.g., focusing solely on organic matter), it integrates multi-dimensional indicators to better capture overall soil quality. Moreover, in contrast to pure statistical methods such as principal component analysis (PCA), it incorporates the study area’s ecological traits (e.g., grassland soil nutrient cycling) and restoration goals (e.g., enhancing soil fertility for vegetation recovery) during indicator selection, yielding more targeted and practically useful results. Filling this combined restoration mechanism gap and providing a sustainability-focused restoration scheme for Inner Mongolia’s moderately degraded grasslands via the SQI model is the core value of this study.

In vegetation restoration, studies have confirmed that cattle manure biogas slurry regulates forage moisture, protein, fat, and ash content ratios to optimize nutritional composition [2], and that enclosed fertilization significantly improves plant communities [3]. For soil improvement, research has found that Nitrogen (N)–Phosphorus (P)–Potassium (K) compound fertilizer effectively ameliorates soil properties [4], and long-term farmyard manure application promotes soil microbial carbon accumulation, altering organic matter component ratios by 3–16% relative to control groups and enhancing carbon sequestration potential through microbial community regulation [5]; additionally, relevant studies have further verified that in the Loess Plateau region, the intervention measure of applying bio-organic fertilizer can effectively reduce soil bulk density and significantly increase soil water content, laying a foundation for subsequent vegetation restoration and soil quality improvement [6]. Regarding restoration duration, studies have noted positive correlations between increases in vegetation height/aboveground biomass and recovery time [7]; in meadow steppes treated with biofertilizers, aboveground vegetation biomass increased annually, and community height also increased significantly [8]; and extended organic fertilization was found to sustainably elevate soil carbon and nitrogen pools [9]. For comprehensive benefits, research has indicated that fertilization significantly increases the Soil Quality Index (SQI), driven by elevated organic matter, zinc, and manganese, among which organic matter shows a particularly strong positive correlation with SQI [10]; meanwhile, economic benefit analysis before and after grassland restoration demonstrated that all fertilization treatments boosted net economic income, verifying synergistic optimization of ecological and economic benefits [11].

Based on the above, this study, after systematically analyzing the current status, issues, and experiences in ecological protection and restoration of Inner Mongolia grasslands, focuses on a moderately degraded typical steppe. Specifically, from 2023 to 2024, restoration measures, including enclosure and enclosure with bio-organic fertilizer application, were implemented consecutively for two years in experimental plots in Huade County. By monitoring changes in vegetation community characteristics and soil physicochemical properties under these measures and evaluating the ecological benefits in each study area using an ecological assessment index system, this research systematically reveals the linkage mechanisms of ecosystem components under different measures. It aims to provide a practical paradigm integrating ecological benefits and sustainable management for degraded grassland governance, contributing to consolidating China’s important ecological security barrier in the north.

2. Materials and Methods

2.1. Study Area Overview

The study site of degraded grassland, located in Huade County, is situated at the Sheep Breeding Farm of Changshun Town (114°00′13″ E, 41°40′40″ N), with an average elevation of 1500 m a.s.l. Climatically, it falls within a typical semi-arid continental monsoon climate zone. The dominant soil type is chestnut soil, with a mean pH value of 7.38. The grassland in this region is a moderately degraded typical steppe, and its dominant plant species include Artemisia frigida, Cleistogenes squarrosa, Leymus chinensis, and Stipa krylovii, among others. The geographical location is shown in Figure 1.

2.2. Overview of Temperature and Precipitation Characteristics

When applying the SQI (Soil Quality Index) model through soil sample analysis, considerations of climate and environmental factors are of critical importance. The model incorporates the impacts of climatic factors, with particular focus on two core variables: precipitation and temperature variation. By altering soil hydrothermal conditions, these two factors directly affect soil physicochemical properties and composition, thereby exerting a correlated influence on both soil sample analysis results and model evaluation conclusions.

As shown in Figure 2, climatic data during the study period revealed minimal differences in seasonal temperature variation between the two years, leading to limited interference with the analysis and model outcomes. In contrast, precipitation differed significantly: rainfall in the second year was substantially higher than in the first, and its impact on the soil system was more pronounced due to abundant annual precipitation. Specifically, sufficient precipitation modifies soil moisture content and aeration, accelerates nutrient leaching, and thereby affects the determination of key indicators such as organic soil matter content and pH value. This resulted in soil sample analysis data from the second year being far more affected by precipitation than that from the first year. Ultimately, this caused discrepancies in the SQI model’s soil quality evaluations between the two years, with the fluctuation of the second-year results showing a stronger correlation with precipitation. The temperature and precipitation data were obtained from the Wulanchabu Municipal Meteorological Bureau and the Huade County Bureau of Statistics.

2.3. Experimental Design and Sampling

This experiment was conducted from 2023 to 2024 in a moderately degraded typical steppe at the Sheep Breeding Farm of Changshun Town, Huade County, Ulanqab City, Inner Mongolia, covering an area of 10,000 mu (≈667 hm²) and adopting a “control-treatment” experimental design principle: a moderately grazed degraded grassland was designated as the control plot (MG) with a stocking rate strictly controlled at 2.5 sheep units/hm² to simulate the local traditional grazing pattern, while two distinct restoration treatments were established, namely (1) Enclosure treatment (EN), where grazing disturbance was blocked by fencing to allow the grassland to restore its vegetation and soil via its inherent natural recovery capacity, and (2) Enclosure with bio-organic fertilizer application treatment (EF), in which bio-organic fertilizer was applied at a rate of 3 t/hm² on the basis of enclosure—this bio-organic fertilizer, derived from the resource utilization of sheep manure (i.e., sheep manure-based organic fertilizer) and holding a valid product registration certificate (Registration No.: Mengnongfei (2013) Zhunzi No. 0430), met relevant standards for nutrient indicators and quality, specifically with a total content of nitrogen (N), phosphorus pentoxide (P₂O₅), and potassium oxide (K₂O) of ≥4% and an organic matter content of ≥40%. Fertilization management followed standardized operations, with a single annual application carried out in April (the early regreening stage of the steppe) using a fertilizer spreader for uniform broadcasting, and after broadcasting, natural precipitation or artificial supplementary irrigation was used to promote the gradual infiltration of the fertilizer into the 0–30 cm soil layer (the main distribution zone of root systems). Additionally, three 1.33 hm² monitoring plots (i.e., three replicates) were set up in each treatment group (EN, EF) and the control plot (MG), and from the implementation of the treatment measures until the completion of data collection at key experimental stages, human disturbances such as grazing, reclamation, and trampling were strictly excluded from the treatment plots. In each monitoring plot of the control and restoration areas, 5 random 1 m × 1 m quadrats were selected using a systematic random sampling method for plant community surveys. Within each quadrat, 5 sampling points were randomly set to collect 0–10 cm, 10–20 cm, and 20–30 cm soil layers, (each soil layer collected from different sampling points is an independent sample, with 75 samples per soil layer, 150 samples per treatment, and 450 samples per year), which were sealed, frozen, and transported to the laboratory for analysis. Due to limitations of the plot monitoring conditions, only the ecological benefits were evaluated in this study.

2.4. Determination Items and Methods

In mid-August 2023 and 2024, plant height, coverage, and density were measured within the designated quadrats. Subsequent to vegetation surveys, aboveground biomass was determined by oven-drying at 65 °C until constant [12]. Standard soil analytical methods were employed for the determination of physical soil properties and nutrients: soil bulk density (SBD) was measured using the core method [13]; soil water content (SWC) was determined via the oven-drying method [14]; electrical conductivity (EC) was assayed with a conductivity meter [15]; soil pH was measured using the potentiometric method [16]; soil organic matter (SOM) was analyzed through the potassium dichromate oxidation–external heating method [17]; total nitrogen (TN) was determined by the Kjeldahl digestion method [18]; alkaline hydrolysable nitrogen (AN) was quantified via the alkaline diffusion method [19]; available phosphorus (AP) was assayed using molybdenum-antimony anti-spectrophotometry [20]; and available potassium (AK) was measured with flame photometry [21,22]. Ecological benefits were evaluated using the Comprehensive Soil Quality Index (SQI) and comparative assessment methodology [23]. The specific experimental procedures and methods described above were performed in accordance with Soil Agrochemical Analysis [24].

The calculation formula was defined as follows: Initially, normalization methods were employed to standardize the datasets of the nine soil indicators:

$$\begin{array}{c}{Y}_i={\displaystyle \frac{X_i-\mathit{min}\left\{X_i\right\}}{max\left\{X_i\right\}-min\left\{X_i\right\}}}\left(1\le i\le n\right)\end{array}$$

(1)

where Yi represents the standardized value of each soil indicator, Xi denotes the value of the soil indicator, and n indicates the number of elements for horizontal comparison.

Due to the differences in the importance and contribution of each soil indicator to the SQI before and after the implementation of degraded grassland restoration measures, the standard deviation coefficient method was adopted to separately calculate the weight values of each soil indicator before and after the restoration treatment. First, the standard deviation coefficient Vj was calculated using Equation (2); then the weight Wj was obtained through normalization according to Equation (3); finally, the SQI was calculated using Equation (4). A larger SQI indicates a higher soil quality.

$$\begin{array}{c}{V}_j={\displaystyle \frac{\sqrt{\frac{1}{n}{\sum }_{i=1}^n{\left(X_{ij}-{\overline{x}}_j\right)}^2}}{{\overline{x}}_j}}\end{array}$$

(2)

$$\begin{array}{c}{W}_j={\displaystyle \frac{V_j}{{\sum }_{j=1}^m{V}_j}}\end{array}$$

(3)

$$\begin{array}{c}SQI={\sum }_{i=1}^m\left(Y_i\times W_j\right)\end{array}$$

(4)

In the above equations, $X_{ij}$ represents the original value of the soil nutrient indicator and ${\overline{x}}_j$ denotes the mean value of the j-th indicator.

Changes in vegetation community characteristics were primarily evaluated using a comparative assessment approach, which quantified the relative increments in vegetation height, coverage, density, and aboveground biomass (dry weight) relative to the control plots before and after restoration.

2.5. Ecological Benefit Assessment Methods

This study employed the SQI model to quantitatively assess the effectiveness of soil nutrient improvement. Based on the obtained soil analysis data (including SBD, SWC, EC, pH, SOM, TN, AN, AP, and AK), the standardized values, weight coefficients, and SQI values for each parameter within the 0–30 cm soil layer were computed before and after the implementation of restoration measures, as calculated following Equations (1)–(4).

2.6. Data Analysis

The experimental data analyses were performed using SPSS 19.0 (IBM Corp., Armonk, New York, NY, USA) (i.e., maximum, minimum, mean, and standard error). One-way analysis of variance, followed by the Duncan test, was applied to test differences in soil quality indicators relative to the control among depths, restoration methods, and years, as well as differences in vegetation relative to the control among restoration methods and years (p < 0.05). The correlation coefficients between soil and vegetation indicators were analyzed via Pearson correlation analysis. The figures were drawn by using Origin 2021b (Origin Lab Corp., Northampton, MA, USA).

3. Result

3.1. Vegetation Community Response to Restoration Measures

As shown in Figure 3, the order of improvement effects of different treatments on vegetation community characteristics in the Huade County sample plots over the two years was EF > EN > MG. Among them, all indicators in the EF treatment showed significant differences compared with MG (p < 0.05); the EN treatment only showed significant differences from MG in terms of vegetation height and aboveground biomass (dry weight) in the second year of restoration (p < 0.05). The specific increment data are as follows: for the EN treatment, the vegetation height in the first and second years of restoration was 7.23% and 45.71% increase, respectively, the vegetation coverage was 2.97% and 10.56%, respectively, the vegetation density was 5.44% and 16.91%, respectively, and the aboveground biomass (dry weight) was 10.02% and 27.97%, respectively; the increments of the above indicators in the EF treatment were 50.39% and 137.43%, 17.91% and 30.56%, 40.68% and 71.59%, 39.68% and 55.09% in sequence. The improvement effects of all treatments were more significant in the second year of restoration. Moreover, over the two years, the improvement effect of the EF treatment on vegetation community characteristics was significantly better than that of the EN treatment (p < 0.05). (The visualization is available in the Supplementary Materials.)

3.2. Impact of Restoration on Soil Nutrients

In the 0–30 cm soil layer, the EN and EF treatments exerted effects on SBD, SWC, and EC, with the overall improvement efficacy generally in the order EF > EN > MG (moderate grazing, control). For SBD, all treatments showed reductions compared with MG in the 0–30 cm layer across the two restoration years. Compared with MG, EN decreased SBD by 0.73–2.92% (first year) and 6.00–6.80% (second year) in the 0–30 cm layer, while EF decreased it by 7.30–9.49% and 14.00–14.97%, respectively. The improvement in SBD was more significant in the second restoration year across different soil layers for all treatments. Over the two years, SBD in the 0–30 cm layer under EF and in the 0–10 cm layer under EN (second year) was significantly lower than that under MG (p < 0.05); meanwhile, significant differences in SBD were observed between EN and EF treatments in the 0–10 cm layer (both years) and the 10–20 cm layer (first year) (p < 0.05). For SWC, all treatments exhibited increases relative to MG in the 0–30 cm layer during the two years. Compared with MG, EN increased SWC by 34.62–45.19% (first year) and 26.82–81.27% (second year) in the 0–30 cm layer, with EF increasing it by 116.59–174.66% and 168.76–197.58%, respectively. Similarly to SBD, the improvement in SWC was more prominent in the second restoration year across various soil layers for all treatments. Over the two years, SWC in the 0–30 cm layer under EF and in the 20–30 cm layer under EN (second year) was significantly higher than that under MG (p < 0.05), and significant differences in SWC between EN and EF treatments were detected in the 0–30 cm layer (both years) (p < 0.05). For EC, compared with MG, the EN treatment decreased EC in the 0–30 cm layer by 1.58–18.10% (first year) and 7.23–9.80% (second year), while the EF treatment resulted in negative reductions (i.e., EC increases) of −16.34% to −8.77% (first year) and −27.77% to −11.26% (second year) for the same layer; except for the 0–20 cm layer under EN and the 10–20 cm layer under EF, the EC improvement effects of all other treatment–soil layer combinations were more significant in the second restoration year.

In the 0–30 cm soil layer, all treatments improved soil conditions except soil pH in the 10–20 cm layer under the EN in the second year of restoration, with improvement effects ordered EF > EN > MG. Versus MG, EN increased soil pH in 0–30 cm by −1.08% to −0.54% (first year) and −0.79% to 0.27% (second year); EF showed −5.82% to −3.23% and −10.90% to −5.51%, respectively. Except for the EN treatment in the 10–20 cm layer, improvements in all other treatments and layers were more significant in the second year. The pH of EF in the 0–30 cm layer was significantly different from that of MG over two years (p < 0.05), with significant differences between EN and EF observed in the 0–20 cm layer (first year) and 10–30 cm layer (second year) (p < 0.05). For SOM, both EN and EF increased the content of SOM in the 0–30 cm over the two years (EF > EN > MG). Versus MG, EN raised SOM by 2.06% to 4.39% (first year) and 4.87% to 22.48% (second year); EF showed −10.36% to 25.81% and 24.39% to 36.93%, respectively. Second-year improvements across all layers were more significant in the second year, with EF in the 10–20 cm layer showing significant differences from both EN and MG in the first year (p < 0.05).

In the 0–30 cm soil layer, EN and EF treatments improved TN and AN contents, with improvement effects generally in the order EF > EN > MG. For TN, all treatments except the EN treatment in the 10–20 cm layer in the second year of restoration showed improvements over MG. Compared with MG, EN changed TN by 2.13–14.29% (first year) and −10.99–28.57% (second year) in the 0–30 cm, while EF changed it by 9.52–26.70% and 20.88–76.40%, respectively. Except for the EN treatment in the 20–30 cm layer, improvements in TN content were more significant in the second year. The TN content in the 0–10 cm layer under EF was significantly higher than that under MG and EN over the two years (p < 0.05). For AN, all treatments except the EN treatment in the 10–20 cm in the second year showed improvements over MG. EN changed AN by 14.00–25.85% (first year) and −1.54–46.85% (second year) in the 0–30 cm layer, with EF changing it by 40.16–52.24% and 36.97–94.63%, respectively. Except for the EN treatment in the 10–20 cm layer, improvements in AN content were more significant in the second year. The AN content in the 0–30 cm layer under both EN and EF was significantly higher than that under MG over the two years (p < 0.05), with significant differences in AN content between EN and EF observed in the 0–10 cm layer (first year) and 0–30 cm (second year) (p < 0.05).

In the 0–30 cm soil layer, EN and EF treatments altered AP and AK contents to varying degrees, with overall improvement effects ordered EF > EN > MG. For AP versus MG, EN increased AP in 0–30 cm by 1.69–7.46% (first year) and 5.78–37.42% (second year); EF showed increases of 7.30–82.02% and 20.23–144.65%, respectively. Improvements in AP content across all layers were more significant in the second year of restoration. The AP content in the 0–20 cm layer under EF was significantly higher than that under MG over the two years (p < 0.05), with its 0–10 cm AP differing significantly from EN over two years (p < 0.05). For AK versus MG, EN increased AK in the 0–30 cm layer by 1.28–4.82% (first year) and 6.69–11.01% (second year); EF showed increases of 10.80–30.71% and 32.78–74.04%, respectively. Improvements in AK content across all layers were more significant in the second year of restoration. The AK content in the 0–10 cm layer (first year) and 0–30 cm layer (second year) under EF was significantly higher than that under MG (p < 0.05), with significant differences in AK content from EN observed in the 0–10 cm layer (first year) and 0–20 cm layer (second year) (p < 0.05) (

Table A1. Changes in soil nutrient content under each remediation measure in different years.

Year	Soil Nutrient Indicators		MG (cm)			EN (cm)			EF (cm)
		0–10	10–20	20–30	0–10	10–20	20–30	0–10	10–20	20–30
	SBD (g·cm−3)	1.37 ± 0.01 Aa	1.37 ± 0.03 Aa	1.41 ± 0.05 Aa	1.34 ± 0.02 Aa	1.36 ± 0.02 Aa	1.37 ± 0.03 Aab	1.24 ± 0.03 Ab	1.27 ± 0.02 Ab	1.30 ± 0.02 Ab
	SWC (%)	3.85 ± 0.54 Ab	4.42 ± 0.45 Ab	6.09 ± 0.82 Ab	5.59 ± 0.96 Ab	5.95 ± 0.41 Ab	8.28 ± 0.66 Ab	9.27 ± 1.32 Aa	12.14 ± 1.28 Ba	13.19 ± 0.84 Ba
	EC (μs/cm)	77.56 ± 7.08 Aa	60.56 ± 12.15 Aa	53.72 ± 6.99 Aa	63.52 ± 8.61 Aa	57.22 ± 4,81 Aa	52.14 ± 8.89 Aa	84.36 ± 8.39 Aa	66.46 ± 2.85 Aa	62.50 ± 3.62 Aa
	pH	7.22 ± 0.06 Aa	7.40 ± 0.11 Aa	7.43 ± 0.06 Aa	7.18 ± 0.12 Aa	7.32 ± 0.11 Aa	7.39 ± 0.09 Aab	6.80 ± 0.07 Ab	7.03 ± 0.05 Ab	7.19 ± 0.06 Ab
2023	SOM (g·kg−1)	17.97 ± 1.82 Aa	15.38 ± 0.52 Ab	15.03 ± 2.19 Aa	18.34 ± 2.26 Aa	16.01 ± 0.72 Ab	15.69 ± 1.23 Aa	20.60 ± 1.23 Aa	19.35 ± 0.66 Aa	16.44 ± 1.87 Aa
	TN (g·kg−1)	1.91 ± 0.06 Ab	1.89 ± 0.04 Aa	1.40 ± 0.11 Aa	1.99 ± 0.05 Ab	1.93 ± 0.06 Aa	1.60 ± 0.19 Aa	2.42 ± 0.15 Ba	2.07 ± 0.18 Aa	1.65 ± 0.21 Aa
	AN (mg·kg−1)	61.05 ± 3.59 Ac	56.14 ± 4.88 Ab	47.79 ± 7.66 Ab	76.83 ± 4.80 Ab	65.59 ± 7.42 Aab	54.48 ± 5.29 Aab	92.94 ± 3.09 Ba	79.89 ± 6.32 Ba	66.98 ± 1.40 Ba
	AP (mg·kg−1)	2.28 ± 0.25 Ab	1.83 ± 0.28 Ab	1.78 ± 0.16 Aa	2.45 ± 0.21 Ab	1.96 ± 0.25 Aab	1.81 ± 0.13 Aa	4.15 ± 0.43 Aa	2.82 ± 0.37 Aa	1.91 ± 0.13 Aa
	AK (mg·kg−1)	132.74 ± 4.11 Ab	97.93 ± 4.79 Aa	81.78 ± 7.82 Aa	137.56 ± 1.21 Ab	99.18 ± 8.91 Aa	83.58 ± 8.19 Aa	173.50 ± 6.31 Ba	109.72 ± 3.79 Ba	90.61 ± 3.79 Ba
	SBD (g·cm−3)	1.42 ± 0.02 Aa	1.47 ± 0.05 Aa	1.50 ± 0.04 Aa	1.33 ± 0.03 Ab	1.37 ± 0.04 Aab	1.41 ± 0.03 Aab	1.21 ± 0.03 Ac	1.25 ± 0.03 Ab	1.29 ± 0.06 Ab
	SWC (%)	4.93 ± 0.63 Ab	6.19 ± 0.61 Ab	6.78 ± 0.66 Ac	6.78 ± 0.56 Ab	7.85 ± 1.93 Ab	12.29 ± 1.96 Ab	13.25 ± 1.36 Aa	18.42 ± 1.76 Aa	20.03 ± 1.39 Aa
	EC (μs/cm)	59.05 ± 9.54 Aa	54.83 ± 4.51 Aa	55.93 ± 5.70 Aa	54.78 ± 4.14 Aa	49.78 ± 6.14 Aa	50.45 ± 5.08 Aa	75.45 ± 5.49 Aa	63.20 ± 3.82 Aa	62.23 ± 6,47 Aa
	pH	7.28 ± 0.04 Aa	7.43 ± 0.04 Aa	7.44 ± 0.10 Aa	7.15 ± 0.13 Aab	7.45 ± 0.04 Aa	7.37 ± 0.08 Aa	6.56 ± 0.33 Ab	6.62 ± 0.06 Bb	7.03 ± 0.13 Ab
2024	SOM (g·kg−1)	17.66 ± 2.07 Aa	15.19 ± 2.84 Aa	13.61 ± 1.27 Aa	18.52 ± 1.44 Aa	16.87 ± 2.17 Aa	16.67 ± 2.97 Aa	21.97 ± 4.15 Aa	20.80 ± 1.82 Aa	16.93 ± 0.88 Aa
	TN (g·kg−1)	1.61 ± 0.14 Ac	1.82 ± 0.31 Aa	1.36 ± 0.15 Aa	2.07 ± 0.19 Ab	1.62 ± 0.26 Aa	1.48 ± 0.12 Aa	2.84 ± 0.07 Aa	2.20 ± 0.08 Aa	1.68 ± 0.06 Aa
	AN (mg·kg−1)	57.14 ± 4.15 Ac	73.85 ± 6.69 Ab	44.71 ± 8.03 Ab	83.91 ± 4.85 Ab	72.71 ± 6.65 Ab	60.64 ± 1.99 Ab	111.21 ± 5.49 Aa	101.15 ± 2.44 Aa	85.14 ± 7.24 Aa
	AP (mg·kg−1)	2.15 ± 0.44 Ab	1.55 ± 0.61 Ab	1.73 ± 0.46 Aa	2.61 ± 0.27 Ab	2.13 ± 0.32 Aab	1.83 ± 0.87 Aa	5.26 ± 0.35 Aa	3.28 ± 0.31 Aa	2.08 ± 0.18 Aa
	AK (mg·kg−1)	129.58 ± 8.99 Ab	95.20 ± 1.00 Ab	79.53 ± 0.39 Ab	143.84 ± 9.18 Ab	101.57 ± 4.57 Ab	87.30 ± 9.52 Aab	225.52 ± 15.53 Aa	135.83 ± 12.13 Aa	105.60±2.32 Aa

Note: EN denotes enclosure; EF refers to enclosure with bio-organic fertilizer application; MG represents moderately grazed degraded grassland (serving as control). Different lowercase letters indicate significant differences (p < 0.05) among different treatments within the same year, whereas different uppercase letters denote significant differences (p < 0.05) for the same treatment across different years. In 2023, the number of valid samples was 64 per soil layer, 128 per treatment, and 384 for the year; in 2024, these figures were 67 per soil layer, 134 per treatment, and 402 for the year.

).

3.3. Correlation Analysis of Vegetation and Soil Indicators

As shown in Figure 4, in the first-year restoration plots of Huade County, vegetation height exhibited a highly significant positive correlation with coverage, density, and above-ground biomass (dry weight) (all p < 0.001); coverage showed highly significant positive correlations with density, above-ground biomass (dry weight), (AP), and AK (all p < 0.001), while displaying a highly significant negative correlation with SBD (p < 0.001); density had highly significant positive correlations with above-ground biomass (dry weight), AP, and AK (all p < 0.001), and also a highly significant negative correlation with SBD (p < 0.001); above-ground biomass (dry weight) was in highly significant positive correlation with available nitrogen (AN) (p < 0.001); and SBD presented a highly significant negative correlation with AK (p < 0.001).

As indicated in Figure 5, in the second-year restoration plots of Huade County, vegetation height correlated highly significantly positively with aboveground biomass (dry weight), SWC, AN, and AK (all p < 0.001), while exhibiting a highly significant negative correlation with SBD (p < 0.001); coverage showed highly significant positive relationships with density and aboveground biomass (dry weight) (both p < 0.001); density was highly significantly positively correlated with aboveground biomass (dry weight), SWC, TN, AN, and AK (all p < 0.001), and had a highly significant negative correlation with SBD (p < 0.001); aboveground biomass (dry weight) displayed a highly significant positive correlation with AN (p < 0.001); SBD was highly significantly negatively correlated with SWC and AN (both p < 0.001); SWC showed highly significant positive correlations with TN, AN, available phosphorus (AP), and AK (all p < 0.001); TN had highly significant positive associations with AN and AK (both p < 0.001); and AN correlated highly significantly positively with AK (p < 0.001).

3.4. Evaluating Ecological Outcomes of Grassland Rehabilitation

3.4.1. Weighting Coefficients of Soil Indicators

Following EN and EF treatments in degraded grasslands, certain changes were observed in the weights of soil indicators. In the 0–10 cm soil layer during the first year of restoration, the top three weights in the MG experimental area were SWC, AP, and SOM in sequence; those in the EN treatment area were SWC, EC, and SOM; and those in the EF treatment area were SWC, AP, and EC. In the 10–20 cm soil layer, the top three weights in the MG experimental area were AP, SWC, and AN; those in the EN treatment area were AP, AN, and AK; while those in the EF treatment area were AP, TN, and AN. In the 20–30 cm soil layer, the top three weights in the MG experimental area were AN, SOM, and SWC (tied with EC); those in the EN treatment area were EC, TN, and AN (tied with AK); and those in the EF treatment area were TN, SOM, and AP (Figure 6a).

In the 0–10 cm soil layer during the second year of restoration, the top three weights in the MG experimental area were AP, EC, and SWC in sequence; those in the EN treatment area were AP, TN, and SWC; and those in the EF treatment area were SOM, SWC, and AK. In the 10–20 cm soil layer, the top three weights in the MG experimental area were AP, SOM, and TN in sequence; those in the EN treatment area were SWC, TN, and AP; and those in the EF treatment area were SWC, AP, and SOM. In the 20–30 cm soil layer, the top three weights in the MG experimental area were AP, AN, and SOM (tied with TN and SWC); those in the EN treatment area were AP, SOM, and SWC; and those in the EF treatment area were AP, EC, and AN (Figure 6b).

3.4.2. Integrated Soil Quality Index

In the first restoration year, relative to the MG, both the EN treatment (except in the 20–30 cm soil layer) and EF treatment enhanced the SQI across the 0–10, 10–20, and 20–30 cm soil layers, with the improvement efficacy following the order EF > EN > MG and SQI exhibiting a decreasing trend as soil depth increased. Specifically, the EN treatment increased SQI by 5.55%, 6.25%, and 0.00% in the respective layers, while the EF treatment yielded SQI increments of 38.89%, 28.13%, and 17.24%. In the second restoration year, the SQI improvement efficacy in the 0–20 cm layer remained EF > EN > MG, whereas the order shifted to EF > MG > EN in the 20–30 cm layer—with the EN treatment increasing SQI by 14.29%, 10.34%, and −12.50% (a decrease) and the EF treatment inducing increments of 65.71%, 58.62%, and 21.88% across the three layers, respectively. Across different soil layers, all treatments except the EN treatment in the 20–30 cm layer showed more significant improvement in ecological benefits in the second restoration year (

Table 1. Changes in SQI under each restoration measure in different years.

Year	Treatment	0–10 cm	10–20 cm	20–30 cm
2023	MG	0.36	0.32	0.29
	EN	0.38	0.34	0.29
	EF	0.50	0.41	0.34
2024	MG	0.35	0.29	0.32
	EN	0.40	0.32	0.28
	EF	0.58	0.46	0.39

Note: EN denotes enclosure; EF refers to enclosure with bio-organic fertilizer application; MG represents moderately grazed degraded grassland (serving as control).

).

3.4.3. Plant Community Improvement Relative to Control

As shown in

Table 2. Characteristics of vegetation communities under each restoration measure in different years.

Year	Treatment	Height (cm)	Coverage (%)	Density (Plant/m2)	Aboveground Biomass (Dry Weight) (g/m2)
2023	MG	12.68	74.00	216.80	211.83
	EN	13.60	76.20	228.60	233.05
	EF	19.07	87.25	305.00	295.88
2024	MG	10.02	72.00	208.20	210.94
	EN	14.60	79.60	243.40	269.94
	EF	23.79	94.00	357.25	327.14

Note: EN denotes enclosure; EF refers to enclosure with bio-organic fertilizer application; MG represents moderately grazed degraded grassland (serving as control).

, over two years, vegetation community characteristics in Huade County plots improved under EN and EF treatments relative to MG. In the first restoration year, EN increased height, coverage, density, and aboveground biomass (dry weight) by 0.92 cm, 2.20%, 11.80 plant/m², and 21.22 g/m², respectively, relative to MG; EF yielded corresponding increments of 6.39 cm, 13.25%, 88.20 plant/m², and 84.05 g/m². In the second year, EN increased these indicators by 4.58 cm, 7.60%, 35.20 plant/m², and 59.00 g/m² relative to MG, while EF showed increments of 13.77 cm, 22.00%, 149.05 plant/m², and 116.20 g/m². All treatments exhibited more significant increments relative to MG in the second restoration year.

4. Discussion

4.1. Restoration Effects on Degraded Steppe Vegetation

In degraded grassland ecological restoration, plant community structure dynamics are core biological indicators for assessing improvement effects. This study indicates that measures like enclosure (EN) and enclosure + bio-organic fertilizer (EF) in Huade County plots promote ecological restoration, significantly improving community traits such as vegetation height, coverage, density, and aboveground biomass (dry weight), with more prominent effects in the second restoration year. These results are highly consistent with previous studies: scholars have confirmed and quantified the significant promotion of vegetation growth by fertilization [25,26]; short-term enclosure (1–4 years) effectively increases aboveground biomass and coverage by eliminating grazing disturbances (e.g., foraging, trampling), providing a self-recovery window for vegetation [27]; studies on alpine meadow show both enclosure and enclosure + fertilization increase community aboveground biomass, but only the latter has a significant effect [28]; other scholars note that community coverage increases with extended fertilization, especially grass biomass [29], which is consistent with relevant research [30]. These findings collectively suggest that degraded grassland restoration depends on the synergy of human intervention and natural recovery: fertilization directly stimulates plant growth by supplementing key nutrients, while enclosure indirectly optimizes community structure by eliminating disturbances. Both driving positive vegetation succession, with their restoration effects on vegetation structure and function, continuously enhanced with the short-term extension of enclosure and fertilization duration.

4.2. Rehabilitation Effects on Soil Nutrients

In grassland ecosystems, soil, as the core carrier for plant community development, has fertility and nutrient supply capacity that directly affect vegetation succession, while rational fertilization is key to ensuring stable crop yield increase and improving soil fertility. This study shows that in Huade County plots, except for the improvement in TN, AN, and pH in the 10–20 cm soil layer under EN treatment in the second restoration year, and the improvement in EC under EF treatment across the two restoration years, which were lower than those in MG, all other indicators were improved relative to MG. Under the same treatment, except for soil EC in the 0–10 cm layer, soil pH in the 10–20 cm layer, TN in the 20–30 cm layer, AN in the 10–20 cm layer, and EC in the 20–30 cm layer under EF treatment, the improvements in the second restoration year were more significant.

Previous studies have shown that SOM content increased after fertilization restoration of degraded grasslands [31]. Other scholars have experimentally confirmed that fertilization can effectively improve soil nutrient status, significantly increasing the contents of SOM, N, P, K, and other nutrients (p < 0.05) [32,33]. However, other studies have found that after enclosure, the contents of SOM, total phosphorus (TP), and AP in moderately and severely degraded desert steppes significantly decreased (p < 0.05), while no significant changes were observed in lightly degraded areas [34]. This is inconsistent with the findings of the present study, and the discrepancy is presumably attributed to differences in enclosure duration, grassland type, and inherent soil nutrient contents. Additionally, studies on fertilization and crop yields in arid-semiarid soils showed fertilization significantly increased soil organic carbon (SOC), TN, and TP in 0–20 cm topsoil [35]. Furthermore, relevant research results have demonstrated that fertilization-based restoration measures can improve SBD, significantly increase SWC, and also exert a positive effect on soil salt content [36]. However, the present study found that soil salt content increased after fertilization application yet effectively decreased in the second restoration year, particularly in the topsoil layer. This phenomenon can be explained as follows: in the first year, the soluble ions introduced by the fertilizer directly increased soil salt content, while the processes of leaching and vegetation uptake were relatively weak; in the second year, the significantly increased precipitation facilitated intense leaching to remove salts, and this effect was synergistically enhanced by the strengthened uptake of salt ions by vegetation. In terms of restoration duration, soil C, N, and P contents are significantly affected by fertilization years [37], which is consistent with field research [38], indicating that long-term organic fertilization effectively increases AP. In the restoration of degraded grasslands, long-term fertilization improves the contents of SOC, TP, and TN at a faster rate and to a greater extent than fencing enclosure. Furthermore, the combination of fertilization and fencing enclosure can synergistically enhance restoration efficiency [39]. Short-term fencing (3 years) slightly reduced TN and AP [40], with all the above conclusions aligning with this study.

4.3. Ecological Benefits of Steppe Restoration

Using an ecological benefit evaluation system, this study systematically revealed the application effects of degraded grassland restoration technologies. Results showed that compared with MG (control), the SQI and vegetation community characteristics of degraded grasslands under all restoration treatments were improved (characterized by increased values), with only the SQI in the 20–30 cm soil layer under the EN treatment failing to exhibit this trend over the two restoration years. Meanwhile, SQI displayed a decreasing pattern as soil depth increased, and the improvement in ecological benefits was more significant in the second restoration year. The lower SQI in the 20–30 cm soil layer under the EN treatment (relative to MG) over the two restoration years is presumably due to grazing prohibition, leading to reduced C/N mineralization rates in deep soil (particularly in deep soil layers) and impaired nutrient cycling. Animal activities in grazed areas facilitate nutrient mixing in sub-surface soil layers and enhance microbial activity.

Previous studies support these patterns: fertilized grasslands had higher SQI than unfertilized ones (p < 0.05), with SOM, Zn, Mn sensitive to SQI improvement and SOM highly correlated with SQI [10]; research on alpine grasslands research in the Qilian Mountains verified a 11.78% SQI increase in 0–30 cm layers after freezing–thawing via a soil function-based framework [41]; enclosed grasslands showed higher SQI than grazed or croplands, linked to increased SOC, AN, and microbial biomass C, with SQI decreasing with depth [42]; fertilization enhanced ecological benefits more stably in the second restoration year [8]; and the herb–shrub–tree mixed model (MSII) had a higher SQI (0.54 vs. 0.04 in pure herb restoration) by boosting litter, root activity, soil stability, and nutrient cycling, making it suitable for degraded grassland restoration [43].

4.4. Impacts of Long-Term Field Experiments and Analysis of Long-Term Effects of Restoration Measures

Conducting long-term field experiments is a core method for analyzing the evolutionary patterns of degraded grassland ecosystems and ensuring the sustainability of restoration measures; in studies on typical degraded grasslands in Inner Mongolia, such experiments, through multi-year in situ observations, not only clarify the accumulation patterns of key nutrients like SOC, SWC, and TN under fertilization interventions but also systematically demonstrate the dynamic changes in vegetation community characteristics (e.g., height, coverage, aboveground biomass), providing a solid scientific basis for optimizing restoration schemes and effectively avoiding the one-sidedness of conclusions from short-term observations.

From the perspective of specific measure effects, long-term enclosure has dual impacts: on one hand, it significantly improves community growth indicators—for example, a study in the Hongsongwa National Nature Reserve showed that the dry weight of aboveground vegetation biomass in control plots was 368 g/m², and after 1–5 years of enclosure, this value increased to 412, 594, 620, 628, and 638 g/m² respectively, with a more pronounced increase in the early stage [44]—and other studies have also confirmed that long-term enclosure leads to significant increases in community height, coverage, and aboveground biomass [45]; on the other hand, studies have found that in the “4-year enclosure plus fertilization” mode, enclosure weakens productivity stability [46]. In contrast, long-term fertilization exerts a positive driving effect: it not only significantly increases the contents of soil organic carbon and total nitrogen to restore the soil nutrient pool [47], but also offsets the negative impact of enclosure on productivity; additionally, long-term nitrogen (N) addition enhances the sink capacity of the vegetation nitrogen pool [48], and long-term fertilization significantly increases the contents of SOM, TN, TP, and total potassium (TK) [49], while reducing SBD and significantly increasing SWC [50], further confirming the key role of fertilization in the long-term restoration of degraded grasslands.

5. Conclusions

Based on the study results, this paper concludes the following: (1) Within the short-term restoration period, all rehabilitation measures effectively improved the vegetation community characteristics and soil physical-chemical properties in the experimental plots of Huade County, except for the soil electrical conductivity (EC) under the EF treatment. Overall, the EF treatment exhibited particularly significant comprehensive improvement effects on vegetation communities and soil physical–chemical properties (excluding EC), and the comprehensive improvement effectiveness of all rehabilitation measures was more prominent in the second year of restoration. (2) Ecological benefit analysis indicated that during the short-term restoration period, the EF treatment achieved the most significant enhancement in ecological benefits, showing the optimal performance. From the perspective of vertical distribution characteristics, the ecological benefits of all treatments generally presented a decreasing trend with the increase in soil depth. In terms of temporal dynamic changes, the enhancement effects of ecological benefits for all treatments were particularly significant in the second year of restoration. This phenomenon was closely related to the climatic conditions in the second restoration year, especially the extremely sufficient precipitation in that year, which provided crucial water support for vegetation growth and soil improvement, further amplifying the enhancement effects of rehabilitation measures on ecological benefits.

The results indicate that artificial restoration measures can effectively promote the recovery of vegetation and soil in degraded grasslands, with the EF treatment achieving the optimal restoration effect. Notably, the significant enhancement of restoration benefits in the second year of the study was closely associated with the extremely sufficient precipitation during that period—abundant rainfall provided crucial water support for vegetation growth and soil improvement, which synergistically amplified the positive effects of artificial restoration measures. This finding provides an important basis for the targeted restoration of moderately degraded grasslands in typical steppes, demonstrating that selecting appropriate restoration measures, ensuring sufficient recovery time, and taking into account the regulatory role of key climatic factors, such as precipitation, can significantly enhance ecological restoration benefits, and holds guiding significance for regional ecological restoration practices.