*2.1. Research Characteristics*

As shown in Figure 1, the number of studies on the structure and stability of grassland ecosystems showed an overall fluctuating upward trend. From 1995 to 2004, the average annual number of articles did not exceed two, and it was in the budding stage. The number of papers published from 2005–2014 was 26, and the number of papers published during this period increased rapidly with the UN Millennium Ecosystem Assessment (MEA) synthesis report [68], with large fluctuations; this is the development phase. The number of articles published from 2015 to present was 97, mainly due to the convening of the United Nations Sustainable Development Goals (SDGs) [69]; the number of studies on ecosystem

structure and stability shows a rapid upward trend, and the degree of research on concepts and mechanisms is gradually expanding and entering a diversified development stage. on concepts and mechanisms is gradually expanding and entering a diversified development stage.

number of papers published from 2005–2014 was 26, and the number of papers published during this period increased rapidly with the UN Millennium Ecosystem Assessment (MEA) synthesis report [68], with large fluctuations; this is the development phase. The number of articles published from 2015 to present was 97, mainly due to the convening of the United Nations Sustainable Development Goals (SDGs) [69]; the number of studies on ecosystem structure and stability shows a rapid upward trend, and the degree of research

*Plants* **2023**, *12*, x 4 of 33

**Figure 1.** Trend in the annual distribution of literature related to the structure and stability of grassland ecosystems (MEA: Millennium Ecosystem Assessment; SDGs: Sustainable Development **Figure 1.** Trend in the annual distribution of literature related to the structure and stability of grassland ecosystems (MEA: Millennium Ecosystem Assessment; SDGs: Sustainable Development Goals).

#### Goals). 2.1.1. Stage Characteristics

#### Budding Stage

2.1.1. Stage Characteristics Budding Stage Research during the budding stage was mainly focused on theoretical studies of ecosystem structure and stability, and mechanisms of impact. The aim is mainly to improve the yield and quality of grassland forage through these studies and to meet the needs of human food security. The relationship between biodiversity and ecosystem function and productivity and the mechanisms that influence them have long been debated and have been shown to be positively correlated in national and international studies [70]. For example, through field experiments controlling for plant species diversity, functional diversity and functional composition, Tilman et al. concluded that these three factors are the main determinants of plant productivity, total plant nitrogen and light infiltration capacity, which, in turn, have a significant impact on ecosystems [71]. Spehn et al. monitored range biomass production, resource use (space, light and nitrogen) and decomposition processes on eight European pastures with different plant species richness, showing that higher species and functional group diversity also resulted in higher yields and more efficient use of resources [72]. These scholars have mainly observed changes in grassland biodiversity and productivity from the field scale, which is only a single-sided influence process, and have not addressed the interaction mechanisms between grassland species diversity, and biodiversity, etc., and stability. Moreover, the scale of the study is limited Research during the budding stage was mainly focused on theoretical studies of ecosystem structure and stability, and mechanisms of impact. The aim is mainly to improve the yield and quality of grassland forage through these studies and to meet the needs of human food security. The relationship between biodiversity and ecosystem function and productivity and the mechanisms that influence them have long been debated and have been shown to be positively correlated in national and international studies [70]. For example, through field experiments controlling for plant species diversity, functional diversity and functional composition, Tilman et al. concluded that these three factors are the main determinants of plant productivity, total plant nitrogen and light infiltration capacity, which, in turn, have a significant impact on ecosystems [71]. Spehn et al. monitored range biomass production, resource use (space, light and nitrogen) and decomposition processes on eight European pastures with different plant species richness, showing that higher species and functional group diversity also resulted in higher yields and more efficient use of resources [72]. These scholars have mainly observed changes in grassland biodiversity and productivity from the field scale, which is only a single-sided influence process, and have not addressed the interaction mechanisms between grassland species diversity, and biodiversity, etc., and stability. Moreover, the scale of the study is limited to the field scale, which does not involve the regional or global scale. Therefore, it remained uncertain at this stage how the findings would be extended to the landscape and regional level, and how they would be extended across different ecosystem types and processes [73].

#### mained uncertain at this stage how the findings would be extended to the landscape and Development Stage

regional level, and how they would be extended across different ecosystem types and processes [73]. Facing the threat of increased environmental pollution and the endangerment of cherished flora and fauna, the United Nations hosted the MEA, which motivated the conservation and sustainable development of ecosystems and was a phase of development in the study of the structure and stability of grassland ecosystems.

to the field scale, which does not involve the regional or global scale. Therefore, it re-

The study of factors that influence the structure and stability of grassland ecosystems was becoming progressively more advanced. How to manage pastures scientifically so that they can maintain healthy grassland ecology while balancing pasture production were the main issue of research during this period. For example, Isbell et al. conducted a longterm N enrichment experiment on grasslands to analyze the effects of N enrichment on productivity, plant diversity and the species composition of natural grasslands. The results showed that in the early stages, nitrogen enrichment increased grassland productivity, but over time, nitrogen enrichment was negatively correlated with species diversity and productivity [74]. Thus, scientific anthropogenic fertilization, generalizable results derived from experimental fertilization, is an important factor driving grassland biodiversity and species composition. Facing the problem that the productivity of natural grasslands cannot meet the current demand for livestock feed, the improvement of natural grasslands and the optimization of the planting structure of artificial grasslands became the focus of research at this stage [75]. For example, Albayrak et al. conducted experiments with seed mixes of oats and vetch on pastures in the highlands of Madagascar. The experiments showed that the mixed planting not only enhanced forage yield and quality, but also improved resource utilization [76]. Meanwhile, methods for evaluating the stability of grassland ecosystems are gradually arising. Zheng et al. used the fuzzy comprehensive evaluation method to evaluate the stability of mixed-seeded grassland in terms of community components, function and resistance to invasion. The results showed that the mixed seeding species and the proportion of mixed seeding could influence the community stability; however, this did not play a decisive role [77]. Therefore, how to scientifically quantify the stability of legume–grass mixed grassland communities, considering the temporal scale, spatial scale and their corresponding sensitive indicators, etc., is one of the issues that needed to be urgently explored at this stage.

In summary, research on the structure and stability of grassland ecosystems has gone through a budding stage and a development stage, which have developed towards diversification. In the budding stage, qualitative theoretical studies and influence mechanisms are the focus; in the development stage, with the aim of maintaining the stability of grassland ecosystems and protecting and promoting their sustainable development, a series of studies on the structure, function, stability influence mechanisms and stability evaluation of grassland ecosystems were carried out. Since then, the research directions and themes of ecosystem structure and stability have gradually developed in a diversified manner.

#### Diversification Stage

In order to thoroughly solve the development problems in the social, economic and environmental dimensions and shift to a sustainable development path, the United Nations Sustainable Development Summit put forward the SDGs, and many countries and regions have responded to the goals of poverty eradication, hunger eradication and climate change, etc. How to optimize the structure and stability of grassland ecosystems to promote sustainable development of ecosystems has become an issue that needs to be solved at the stage of diversified development [78,79].

With the gradual progress of research, the research at the stage of diversified development is mainly focused on the control experiments of grassland, to reveal the influence mechanism of species diversity and stability [80], and to quantitatively study the nutrients and productivity of pasture, and the regulatory mechanism of stability [24]. In addition, with the development of technology, data disclosure and the rise of multiple research methods (GIS technology, etc.), different spatial and temporal comparative studies have been widely performed.

The exploration of the mechanisms influencing species configuration and productivity and stability is a prerequisite for exploring the scientific management of grasslands and improving their productivity and stability. For example, Prieto et al. studied the effects of grassland productivity and sustainable supply capacity through species and genetic diversity. The results show that the complementary effects of taxonomy and genetic diversity can increase productivity under conditions of multi-grass species configurations, which, in turn, increases the productivity and resilience of grasslands in the face of environmental hazards [81]. Quantitative studies of nutrient limitation and balance and water–fertilizer coupling in forages are important processes to improve forage quality, productivity, and

resistance to invasion [23]. Niu et al. assessed the effect of nitrogen enrichment on the stability of semi-arid grassland ecosystems in northern China and its potential mechanisms by simulating atmospheric nitrogen enrichment. The results showed that community stability was non-linearly related to nitrogen enrichment and that this relationship was positively correlated with species asynchrony, species richness and species diversity, as well as the stability of dominant species and the stability of grassland functional groups [82]. Therefore, it is important to re-evaluate the mechanisms by which multiple levels of nitrogen deposition affect the stability of natural ecosystems to gain a deeper understanding of the multiple nutrient inputs to grasslands and their stability response mechanisms.

In summary, the optimization and stabilization of grassland ecosystem structure can provide important basic research to elucidate the mechanisms of the complementarity and functional diversity of functional traits of forage grasses, as well as the synergy and tradeoff between productive and ecological functions of grassland, thus promoting the healthy and sustainable development of grassland. We believe that, based on the idea of cascading benefits of grassland ecosystem pattern, process, function, and services, we should explore the mechanisms of multi-species configurations or natural grassland improvement and stability maintenance, and clarify how the species configuration of grassland (leguminous– grass, annual–perennial, and deep-rooted–shallow-rooted, etc.) affects the nutrient flow of grassland, which, in turn, affects community stability, and ultimately promotes the synergistic development of grassland-ecosystem productivity and stability.

#### 2.1.2. Stage Characteristics

Table 1 illustrates the global distribution of studies related to the structure and stability of grassland ecosystems. China and the USA have the highest number of published literature, with over 20 articles, which shows the concern for global issues such as the sustainable development of grasslands and food security in those countries [5,83]. Developed countries such as Europe and Oceania also published a relatively significant amount of literature. In addition, countries such as Brazil and South Africa account for a relatively small number of publications.


**Table 1.** Distribution of the number of publications issued by countries or region.

Due to regional differences in natural economic and social conditions, research on grassland ecosystems has developed unevenly and has prominent regional characteristics. In terms of the number of publications, Asia accounts for 47.8%, which is related to national policy support and the attention of research institutions [84]. In the Asian region, China is the main publication country, with most of the research focus on large-scale natural grasslands in the provinces of Tibet, Inner Mongolia and Xinjiang [85], where the restoration of these grasslands is of great importance to China in realizing the "two mountains theory" of "Lucid waters and lush mountains are invaluable assets." [86]. The emergence of global issues (global warming, land degradation and resource crises) has led to sustainable and green development being a crucial issue in the 21st century, as exemplified by the United

Nations SDGs. As a result, publications focused on the restoration of grassland ecosystems are gradually increasing, with particular attention being paid by countries in North America. In addition, with a large proportion of grasslands in North and South America, particularly in the Pampas, which has always been an export area for beef cattle, the sustainability of grassland ecosystems determines the economic lifeblood of the entire region. Similarly, although Europe does not have the same area of grassland as Asia or the Americas, food mainly comes from pastureland due to dietary habits, etc. As a result, the stable output of pastureland is vital to Europe's survival [87]. Oceania is the region with the highest exports of wool and dairy products, and the quality of forage is related to the value of trade in export commodities and is equally important to the economy of the region as a whole.

The low number of publications in Africa is mainly related to the economic development of Africa, where most of the countries are still developing countries and the primary concern is to solve the problem of food and maintain national security and stability. Thus, although the area of grassland in Africa is large, and the sustainable development of savannah, in particular, is of great importance for maintaining the global climate, it is mostly studied by developed countries, such as the USA and Denmark [88–90]. South Africa is one of the few African countries with a significant volume of publications due to its better economic conditions.

#### *2.2. Research Progress and Major Landmark Results*

Based on the titles, keywords, abstracts and previous studies of the literature [91], drawing on the studies of Wang et al. and Fu et al. and the changes in grassland ecosystem patterns and their relationship with ecological processes [92,93], the 133 papers were divided into five aspects: structural studies (component structure, trophic structure, spatiotemporal structure), structural optimization, stability studies, structure–stability relationships and influencing factors (Figure 2), according to structural composition, ecological processes (replaced by stability) [94,95], the relationship between structure and stability, and factors influenced by the environment. The largest portion of literature focused on the factors influencing ecosystem structure and stability, at 31.5%, followed by structure studies, which accounted for 30% of this type of literature. In addition, studies on ecosystem stability accounted for 17.2% of the literature, while studies on ecosystem structure optimization and structure–stability relationships accounted for 12.7% and 8%, respectively. *Plants* **2023**, *12*, x 8 of 33

Grasslands include the degraded, natural, and artificial, which determine the type of

Grassland ecosystem structure refers to the relatively ordered and stable state of the various components of an ecosystem in space and time [98], including component struc-

The structure of grassland ecosystems consists of biotic (plants, animals, micro-organisms) and abiotic (water, air, soil) factors, and a certain hierarchy and structural pattern is maintained between the components of each system. The composition of grasslands is the basis of all grassland research, and the analysis of changes in species composition and their response to grassland has always been an important part of grassland research [100–102]. Silva Mota et al. evaluated changes in floristic composition, structure, diversity, and life-forms spectra along an altitudinal gradient in the rupestrian grasslands in the south-eastern Espinhaço Mountains Range, and the results showed that differences in vegetation structure, diversity, species composition, frequency and the richness of each life form in relation to soil attributes and elevation. Plant height, species richness, diversity and evenness, frequency and richness of phanerophytes and chamaephytes decreased with elevation. The related results indicated that soil is an important driver of community change [103]. At the same time, human activities affect the species composition of grassland, and, thus, changing the cascading benefits of research into ecosystem services has

spatial and temporal structure and nutrient structure [96], influencing the degree of ecosystem stability and, thus, the supply of ecosystem service capacity [97]. It also combines with artificial capital to form multiple forms of animal husbandry, which enter the market and, further, form an industrial chain which provides ecosystem services and benefits to humans. Therefore, attention to the structural composition and stability of grassland ecosystems is a prerequisite for interpreting the processes that shape grassland-ecosystem

ture, nutrient structure and spatial and temporal structure [99].

services.

2.2.1. Structure Research

Component Structure

Grasslands include the degraded, natural, and artificial, which determine the type of ecosystem structure and species composition of their habitats. This alters the process of ecosystem material cycling and information transfer through the component structure, spatial and temporal structure and nutrient structure [96], influencing the degree of ecosystem stability and, thus, the supply of ecosystem service capacity [97]. It also combines with artificial capital to form multiple forms of animal husbandry, which enter the market and, further, form an industrial chain which provides ecosystem services and benefits to humans. Therefore, attention to the structural composition and stability of grassland ecosystems is a prerequisite for interpreting the processes that shape grassland-ecosystem services.

#### 2.2.1. Structure Research

Grassland ecosystem structure refers to the relatively ordered and stable state of the various components of an ecosystem in space and time [98], including component structure, nutrient structure and spatial and temporal structure [99].

#### Component Structure

The structure of grassland ecosystems consists of biotic (plants, animals, microorganisms) and abiotic (water, air, soil) factors, and a certain hierarchy and structural pattern is maintained between the components of each system. The composition of grasslands is the basis of all grassland research, and the analysis of changes in species composition and their response to grassland has always been an important part of grassland research [100–102]. Silva Mota et al. evaluated changes in floristic composition, structure, diversity, and life-forms spectra along an altitudinal gradient in the rupestrian grasslands in the south-eastern Espinhaço Mountains Range, and the results showed that differences in vegetation structure, diversity, species composition, frequency and the richness of each life form in relation to soil attributes and elevation. Plant height, species richness, diversity and evenness, frequency and richness of phanerophytes and chamaephytes decreased with elevation. The related results indicated that soil is an important driver of community change [103]. At the same time, human activities affect the species composition of grassland, and, thus, changing the cascading benefits of research into ecosystem services has become an emerging research direction. However, it is difficult to solve the various problems of grassland composition alone at the present stage. Therefore, the above studies were conducted in combination with spatial and temporal changes [104].

#### Spatial and Temporal Structure

The characteristics of ecosystems depend on biodiversity, that is, the functional characteristics of organisms present in the ecosystem, and the distribution and abundance of these organisms in space and time [105]. Due to species evolution and turnover, and a lack of scientific grassland management measures, the species diversity of grassland is gradually homogenizing, the soils are gradually being degraded, and the productivity is gradually decreasing [106]. Therefore, study on the spatial and temporal structure of grassland ecosystem provides a good reference for improving the service capacity of grassland ecosystem in the KDC area.

#### Time Structure

Quantitative assessment of the interannual variability in grassland landscape patterns in response to their ecosystem service values is a current research hot spot [107]. Monitoring the spatial and temporal dynamics of grassland areas can help decision makers to plan the use of grassland in a rational manner and achieve the sustainable development of grassland [108]. Landscape changes directly affect changes in the demand and supply of ecosystem services. Compared to the spatial scale of ecosystem services, little attention has been paid to how interannual changes in land use affects ecosystem service trade-offs, but this is necessary to facilitate the rationalization of decisions, especially when those decisions aim to re-establish diverse services and restore biodiversity [109]. Therefore, it is important to analyze the spatial and temporal evolution characteristics of grasslands and elucidate the main drivers of grassland-ecosystem service capacity to develop reasonable ecological restoration measures for grasslands. Recent studies have shown that current research has focused too much on global, national, and provincial spatial and temporal dynamics at large scales, neglecting research on the mechanisms of trade-off synergy and value transfer of grassland-ecosystem service functions at small scales in counties and sample sites [107]. Therefore, it is important to conduct overall monitoring of grassland landscape patterns in different periods in a specific region, analyze the temporal variability and heterogeneity of grassland-ecosystem landscapes at spatial scales, and grasp the structuralchange characteristics of grassland ecosystems, to promote the optimization of the overall ecosystem structure and stability enhancement.

Seasonal changes in grasslands are also a current research hotspot in ecological restoration. The physical characteristics of plant growth in grassland determine its seasonal variation, which can lead to changes in ecosystem structure in the short term [110]. Thus, the changes in the species composition of grasslands in the short term can affect not only the above- and below-ground productivity supply of grasslands, but also the quality of forage and, in turn, the supply of multiple ecosystem functions. Rodriguez Barrera et al. compared the effects of groundhog and grassland use types and their seasonal changes on grassland vegetation structure and diversity in the semi-arid grassland in northern Mexico. The results of this study showed that grassland use type and its seasonal variation were the main determinants of grassland vegetation structure and cover [111]. However, current research on ecosystem services is mainly focused on inter-annual variability, with less research on monthly and seasonal variability [112]. Therefore, the interannual and seasonal changes in grassland-ecosystem structure should be strengthened in future studies.

#### Spatial Structure

Some studies have specifically classified grassland structure into vertical canopy structure and horizontal planting structure [113,114]. Therefore, the spatial structure of grassland ecosystems is discussed in terms of vertical and horizontal structure.

In terms of vertical structure, the three-dimensional structure of forest grassland is globally recognized as a stable state of ecosystems [115,116]. Studies have shown that a reasonable three-dimensional structure can effectively use light, heat, water, air and other environmental factors to improve productivity and maintain the stability of ecosystems [117], and that a composite ecosystem structure has a better soil retention effect [118]. The ecosystems of economic fruit forests and agroforestry are typical composite ecosystems, of which grassland is an essential and important component. The shallow roots of grassland combined with the deep roots of trees provide effective use of natural elements such as light and hot water and air, and intercept water through the multi-layered canopy, thus preventing flooding and providing resistance to erosion and, thus, having a good soil and water conservation effect [119].

In terms of horizontal structure, the optimal combination of multiple species can improve the productivity of grassland ecosystems and promote the efficient recovery of degraded grasslands, thus enhancing the stability and service functions of grassland ecosystems [120]. It has been proven that the persistence and stability of forage production in grassland ecosystems is generally higher than mixed planting with perennial grass species and single-forage cultivation [121,122]. Mixing planting with leguminous forages and replanting natural grasslands with leguminous forages can increase the productivity of grasslands, enhance their resilience and biodiversity, maintain the stability of grassland ecosystems and improve their service capacity [123]. By monitoring light cut-offs in the canopy of grassland communities with high species diversity, researchers found a positive correlation between plant species richness and multiple ecosystem functions, at which time grassland resource utilization was high. The relationship between grassland biodiversityproductivity and management intensity was also positively correlated when supplemented by sound grassland management [124]. Thus, a rational vertical canopy structure and a

scientific horizontal structure not only maintain the biodiversity of the grassland, but also ensure its productivity, thus providing the basis for multi-functional grassland management [125]. In a protected area in central Italy ("Laghi di Suviana e Brasimone" regional park), Cervasio et al. showed that mixed seeding had a significant contribution to improving grassland quality and had a positive impact on grassland ecosystem biodiversity by changing the vertical and horizontal characteristics of grassland ecosystem species [126].

In addition, ecological corridors with a predominantly grassland landscape structure are also an important part of current spatial-structure research [127,128]. Ecological corridors are important links between the structure and function of a system, and grasslands can be used as buffer zones to improve the connectivity of ecological functions, allowing species to use corridors for long-distance dispersal or to provide shelter for some animal movements [129], thus maintaining and protecting biodiversity and improving the quality of their habitats [130,131]. At the same time, the spatial pattern of the landscape influences the magnitude of ecosystem service provision at the landscape scale through grassland use patterns and connectivity [132–135]. By quantifying inter-annual and seasonal changes in grassland landscape patterns in northern China, Hao et al. showed that degraded vegetation can influence changes in ecosystem service functions by changing land use patterns, increasing or dividing patch sizes of grassland and agricultural land, and increasing the size of forest areas where appropriate [136]. This is a very good reference for the grassland ecological protection of KDC. At this stage, however, most studies are still at the stage of qualitative description, and these mainly explore the factors influencing ecological functions (such as biomass production and other ecosystem services) and plant community characteristics [137]. There is also an urgent need to use these insights to develop combinations of grass species for high-yielding, high-quality communities [138], as the research subjects and geographical environments change and whether regional research results can be adapted to global environments or specific geographical areas, such as karst desertification areas and alpine grassland regions.

The study of the spatial structure of grassland ecosystems should focus on the vertical and horizontal cropping structure of grasslands and the distribution of grassland landscape patterns. This not only helps researchers to monitor the dynamics of grasslands comprehensively from a micro to macro level, but also helps grasslands (as an ecological linkage corridor) to provide better connectivity to other ecosystems, guaranteeing the healthy development of the whole terrestrial ecosystem and providing more and better service functions for humans [139]. Studies have found that increasing the abundance, evenness and diversity of dominant species not only effectively improves the productivity of the system [140,141], but also ensures the harmonious development of the "productionliving-ecological function" of grassland ecosystems [142,143], maintaining the stability of their productive functions and the diversity of biological species. In summary, the vertical and horizontal structural characteristics of grasslands change the species composition and biodiversity of grasslands, affecting the quality of grassland habitats and their ability to provide services.

#### Trophic Structure

The study of trophic structure, food webs and ecological networks in ecosystems is the theoretical basis for understanding the composition, structure and dynamics of ecosystems, and provides indispensable scientific support for biodiversity conservation, ecosystem management and restoration, as well as response to global change [99]. The trophic structure of an ecosystem refers to the food chains and food webs formed by producers, consumers and decomposers in a biome with food as the link, which constitutes the main pathway for material cycling and energy flow [144]. Food webs are based on the interactions between species at different trophic levels, forming upstream and downstream regulatory mechanisms to maintain their structural stability [145].

Species interactions or alterations are one of the most important areas in the study of food webs. Food webs, which depict networks of trophic relationships in ecosystems, provide complex yet tractable depictions of biodiversity, species interactions, and ecosystem structure and function [146]. Nearly all ecosystems have been altered by human activities; these changes may modify species interactions and food-web stability [147]. De Castro et al. started from the classic soil food-web model of Hunt to formulate a plausible topology of soil food-web models and then compare the effect of this topology with those of random topologies [148,149]. The results showed that the stronger the species interactions in the soil, the more stable the food web, and that disrupting the functional-group assemblages and strength of interactions in the soil food web inevitably has a significant impact on species and their relative abundance. Duchardt et al. applied structural equation models (SEM) to disentangle direct and indirect effects of prairie dogs on multiple trophic levels (vegetation, arthropods, and birds) in the Thunder Basin National Grassland, and the results indicated that prairie dogs directly or indirectly influence associated vegetation, arthropods, and avifauna [150].

In summary, changes in key traits that sustain interactions are one of the most important factors in determining the stability of food webs [151]. Therefore, the study of species interactions is important for clarifying inter-species relationships and food-web driving mechanisms, which is one of the priorities that should be focused on in the future.

#### 2.2.2. Structure Optimization

Structural-optimization measures can change the species composition of grasslands, thus affecting their ecosystem biodiversity and altering their capacity to supply ecosystem functions. There are many ways to optimize structure, but this study will only discuss biological measures (optimal allocation of species structure) and engineering measures (pasture management systems).

Scientific and rational species allocation is the key to the high productivity and stability of grassland ecosystems [152]. Optimal species structure allocation includes the optimal vertical and horizontal allocation of grassland species. The three-dimensional structure of composite ecosystems has proven to be an important strategy for reconciling environmental protection and economic development in ecologically fragile areas [153], with grasslands being the most crucial aspect, and their strong renewal rate making them a large surface area within the composite ecosystem. The resulting three-dimensional ecosystem structure not only plays an important role in maintaining and providing ecosystem services, but its multiple ecosystem services also provide an effective way to promote the restoration of degraded ecosystems, which is an important biological measure to make full use of resources [117,154]. A diverse three-dimensional and horizontal structure is an effective measure to restore grassland ecosystems, that is, through a combination of different species of grasses from different families (in intercropping, crop rotation, and mixed sowing, etc.; see Figure 3) [155]. The most studied of these is the mixed cropping pattern of legumes and grasses. This mixed cropping pattern makes full use of sunlight, heat, water and air, creates complementarity between species, and promotes the uptake of soil nutrients by forage grasses, which, in turn, improves their annual growth and nitrogen use efficiency [156]. Multi-species planting with different habits is an effective means of restoring grassland ecosystems, and it can effectively improve the degradation of grasslands, protect and improve soil quality, enrich inter-species and intra-species species diversity, and effectively increase the resilience and recovery of degraded grassland ecosystems [157]. Scientific seeding is an effective measure to restore the dynamic balance of grass species [158]. Under the same circumstances, mixed-seeded grasslands have higher forage density and species richness, and it has more competitive and better grassland vegetation restoration compared to single-seeded grasslands. However, it has been demonstrated that species diversity and stability in grassland ecosystems with different rates of mixed seeding are positively correlated, and that seeding rates are negatively correlated with resistance to invasion [159]. Therefore, understanding forage seeding rates and their effective mix relationships is an important area of vegetation-restoration research [160]. Meanwhile, the mixing and intercropping of annual and perennial forages are also efficient management practices that

take advantage of the temporal structure of grassland ecosystems, reducing inter-specific competition and increasing the persistence of forage production [161]. Mixing annual and perennial legume-grass is recognized as a highly productive and robust management practice [162], which can maintain a diverse supply of ecosystem services, and the leguminous forage grasses provide a more adequate supply of nitrogen to other grass species, thus maintaining high grass biomass and meeting the production requirements of farmers [163]. *Plants* **2023**, *12*, x 13 of 33

**Figure 3.** Artificial grassland planting structure. (**a**) Forage mono sowing: Grass forage, *Lolium perenne* L., (**b**) Pasture intercropping: Grasses-Leguminosae Forage, *Lolium perenne* L. + *Medicago sativa* L., *Trifolium repens* L.+ *Lolium perenne* L., (**c**) Pasture mix sowing: Grasses-Leguminosae Forage, *Lolium perenne* L. *+ Medicago sativa* L., (**d**) Forage high and low mix sowing: *Lolium perenne* L. + *Medicago sativa* L. + *Trifolium repens* L. + *Lolium perenne* L. (The grass species of (**b**–**d**) are the same, but the arrangement of planting order is different. (**b**,**c**) are planted in a horizontal way, but the density and proportion of planting are different. (**b**) is good for personnel to manage, and (**c**) is good for the nutrient complementation of grass species. (**d**) is planted in a vertical way, which can fully utilize light, heat, water and air). **Figure 3.** Artificial grassland planting structure. (**a**) Forage mono sowing: Grass forage, *Lolium perenne* L., (**b**) Pasture intercropping: Grasses-Leguminosae Forage, *Lolium perenne* L. + *Medicago sativa* L., *Trifolium repens* L.+ *Lolium perenne* L., (**c**) Pasture mix sowing: Grasses-Leguminosae Forage, *Lolium perenne* L. *+ Medicago sativa* L., (**d**) Forage high and low mix sowing: *Lolium perenne* L. + *Medicago sativa* L. + *Trifolium repens* L. + *Lolium perenne* L. (The grass species of (**b**–**d**) are the same, but the arrangement of planting order is different. (**b**,**c**) are planted in a horizontal way, but the density and proportion of planting are different. (**b**) is good for personnel to manage, and (**c**) is good for the nutrient complementation of grass species. (**d**) is planted in a vertical way, which can fully utilize light, heat, water and air).

A scientific system of pasture management can effectively enhance the sustainable

supply of grassland resources. Pasture management not only includes grass seed cultivation and fertilization, but also includes the full use of time and space, such as rotational grazing, rest grazing and ban grazing. In the early days, pasture management was mainly aimed at improving pasture production. In the new context of global climate change and increased environmental pollution, balancing ecological protection and green economic development has become a necessary path to improve the overall ecological quality of regions [164]. The general system of rotational and rest grazing is only an artificial strategic rest and tactical grazing of grasslands [165]. Therefore, some studies have proposed models such as the sustainable grazing system (SGS), DairyMod, GRAZPLAN, GrassGro, DairyNZ whole-farm model and EverGraze [166–172]. The above systems encourage the sustainable development of grassland ecosystems, and are key to preventing soil erosion and promoting soil improvement, as well as being important measures to conserve grassland plant biodiversity. Michalk et al. detailed sustainable and permanent grazing systems and answer the key questions currently facing Australia: (1) whether increasing the number of paddocks and implementing rotational grazing results in higher grazing rates, higher yields per hectare and better economic benefits; (2) which combination of grazing methods and grazing rates is most appropriate to create higher and more stable perennial A scientific system of pasture management can effectively enhance the sustainable supply of grassland resources. Pasture management not only includes grass seed cultivation and fertilization, but also includes the full use of time and space, such as rotational grazing, rest grazing and ban grazing. In the early days, pasture management was mainly aimed at improving pasture production. In the new context of global climate change and increased environmental pollution, balancing ecological protection and green economic development has become a necessary path to improve the overall ecological quality of regions [164]. The general system of rotational and rest grazing is only an artificial strategic rest and tactical grazing of grasslands [165]. Therefore, some studies have proposed models such as the sustainable grazing system (SGS), DairyMod, GRAZPLAN, GrassGro, DairyNZ whole-farm model and EverGraze [166–172]. The above systems encourage the sustainable development of grassland ecosystems, and are key to preventing soil erosion and promoting soil improvement, as well as being important measures to conserve grassland plant biodiversity. Michalk et al. detailed sustainable and permanent grazing systems and answer the key questions currently facing Australia: (1) whether increasing the number of paddocks and implementing rotational grazing results in higher grazing rates, higher yields per hectare and better economic benefits; (2) which combination of grazing methods

grasslands to improve yields and environmental benefits in different parts of the landscape; and (3) can landscape variability be identified, mapped and effectively managed

grassland grazing and farmers' livelihoods in KDC.

and grazing rates is most appropriate to create higher and more stable perennial grasslands to improve yields and environmental benefits in different parts of the landscape; and (3) can landscape variability be identified, mapped and effectively managed on native grasslands in high-rainfall areas [173]? These can provide some inspiration for grassland grazing and farmers' livelihoods in KDC.

Therefore, optimizing the structural configuration of grasslands in terms of both biological and engineering measures is currently an unavoidable and important issue for improving the sustainable development of grasslands and economic development. It will not only solve the shortage of supply capacity for ecosystem services and promote regional economic development, but also issues related to the global supply of food.

#### 2.2.3. Stability Studies

Stability refers primarily to the ability of a community or ecosystem to maintain its original structural and functional state and resist disturbance after a disturbance, known as resistance stability, and the ability of a community or ecosystem to return to its original state after a disturbance, known as resilience stability.

Exploring the causes that affect the stability of grassland ecosystems and suggesting targeted restoration strategies for degraded outcomes is a major focus of current research [83]. Community stability increases species diversity, species heterogeneity and population size, so maintaining the long-term stability of communities is crucial to maintaining ecosystem function. Researchers have explored the potential drivers and mechanisms of ecosystem stability by studying changes in the relationship between biodiversity and ecosystem service functions [174–176]. Biodiversity is a major driving factor of ecosystem stability [177]. Biodiversity consists mainly of above- and below-ground components. In terms of above-ground biodiversity, an increasing number of studies have shown that the more above-ground biodiversity (plant diversity) there is, the more stable the ecosystem is [178]. The results of Garcia-Palacios et al. showed that the diversity of leaf traits may promote the stability of an ecosystem under low drought conditions, while the species richness may play a greater role in the stability of an ecosystem under high drought conditions [179]. Similarly, below-ground biodiversity (soil biodiversity) altered grassland-ecosystem resistance and resilience through direct and indirect effects on plant diversity, net ecosystem productivity and plant species interactions, and, thus, changed grassland ecosystem stability [180]. In addition, recent studies have shown that phenological variation can also reconcile plant diversity with the seasonal stability of ecosystems, and, thus, affect the stability of the whole ecosystem [181].

Evaluating grassland-ecosystem stability has also become an effective means of detecting changes in grassland landscapes or the effectiveness of grassland restoration [182]. When evaluating the stability of grassland ecosystems, timescale variation is a factor that must be considered. By comparing the changes (years, seasons, or months) in grassland landscape patterns in different periods combined with evaluation models and methods, researchers have selected indicators of ecosystem vitality, resistance, resilience and variability to evaluate ecosystem stability [32], revealing the processes of change in grassland ecosystems under extreme weather conditions and their driver factors, quantifying the correlation between grassland ecosystem stability and biodiversity and their spatial patterns, and providing important prerequisites for improving sustainable ecosystem services [183,184]. The traditional evaluation methods of ecosystem stability are mainly qualitative analysis methods, including empirical methods and expert consultation methods, which are, generally, using expert consultation methods to quantify measures. The other is mainly quantitative evaluation including the comprehensive-evaluation method, gray-information analysis, ecological models and other methods, among which the most common methods of comprehensive analysis are hierarchical analysis, the entropy weighting method, the weighted comprehensive average method, and the comprehensive index evaluation method, etc. New kinds of ecosystem stability evaluation are diverse, and an increasing number of methods are becoming more credible and scientific in their assessment of ecological stability. These methods for evaluating the stability of ecosystems can provide good theoretical references for the ecosystem evaluation of KDC, and they can provide insights into the establishment of stability evaluation index systems for karst grassland ecosystems.

#### 2.2.4. The Relationship between Structure and Stability

A good system structure maintains its relative stability, and ecosystems with high stability generally have a more rational structure. Tilman et al. argued that diversity leads to higher community productivity, ecosystem stability and resistance to invasion [74]. Before the 1970s, ecologists believed that communities with higher diversity had more stable ecosystems [185]. Since then, ecologists have focused more on the relationship between species diversity and ecosystem stability [186,187].

Biodiversity affects ecosystem services by altering ecosystem function and stability [188]. Biodiversity–stability relationships showed that above-ground productivity and temporal stability increased significantly with increasing species richness, while biodiversity was largely influenced by ecosystem structure [85,189]. Therefore, understanding the relationship between grassland-ecosystem structure and stability and its influencing factors is essential for the sustainable development of grassland ecosystems [190]. However, how plant species diversity regulates the stability of ecosystems (such as biomass reproduction and nutrient cycling) has become one of the challenging questions in ecology [191,192]. Experiments have demonstrated that the higher the complexity of the ecological network of grasslands, the higher the ecological stability, by influencing plant physiological conditions, as well as species generation, diversity and variation [193,194]. Of course, grasslandecosystem stability is also vulnerable to the influence of ecosystem components (ecosystem species diversity, composition), climate change and anthropogenic activities, mainly in the form of lowering the productivity of grassland, weakening biodiversity and declining service functions, which, in turn, can affect grassland-ecosystem services [195]. Therefore, studying the structure of grassland ecosystems and their material and energy flows and cycles among different components, and exploring the relationship between grasslandecosystem structure, function, and stability is one of the research areas that should be focused on in the future [196]. Recent studies have shown that the positive relationship between biodiversity and stability is also influenced by spatial scale [197]. Therefore, the study of the relationship between structure and stability should also consider different spatial scales and timescales.

#### 2.2.5. Factors Affecting Structure and Stability

The structure and stability of grassland ecosystems are influenced by multiple factors, such as the natural environment and human activities. Therefore, the heterogeneity of spatial environments inevitably leads to variability in the degree of stability. In the case of grassland ecosystems, their structure and stability are mainly influenced by climate change, nitrogen deposition or nutrient addition, grazing and grassland management, plant invasion and natural disasters (Figure 4).

Global climate change is affecting ecosystem function and stability, indirectly altering species diversity, species composition, and functional plant traits [198,199], thus reducing the capacity of ecosystem services and reducing the various benefits that humans derive from them [200]. The impact of climate change on ecosystem function and biodiversity is highly dependent on grazing history and natural conditions [190]. By comparing changes in the spatial pattern of grasslands and simulating different climatic conditions on grassland ecosystem resistance, recovery times and rates, White et al. concluded that future climate change will have a significant impact on the resistance and recovery of disturbed ecosystems; however, the spatial pattern of impacts varies widely [201].

**Figure 4.** The structural components of grassland ecosystems and the factors that influence them combine to influence grassland ecosystem stability.

Key limiting elements can alter species interactions [202], change the spatial and temporal patterns of terrestrial plant communities [203], and influence ecosystem function [204]. Atmospheric nitrogen deposition and the application of nutrients (phosphorus, potassium, etc.) can affect species diversity in grassland communities [205]. The results of related studies show that the addition of nitrogen or phosphorus individually can increase the productivity of grasslands by 0–20%, while the addition of both nitrogen and phosphorus can increase the productivity of grasslands by 60% [206,207]. Nitrogen is an essential nutrient for plant growth, and an appropriate addition of nitrogen deposition will directly promote rapid plant growth and increase the net primary productivity of terrestrial ecosystems. However, excessive nitrogen deposition can also lead to soil acidification, altering the effectiveness of soil mineral nutrients and the structure and function of microbial communities, which, in turn, affects plant growth and changes plant diversity [208–210]. Plant diversity is closely related to ecosystem stability and productivity [211], and ultimately drives changes in ecosystem structure and function [212]. Therefore, paying attention to the differences in supply between different nutrient elements and the resulting differences in the resistance and resilience of plant diversity that result is extremely important for both productivity stability and the temporal stability of grasslands [213].

Plant invasion and vegetation succession are also among the factors that alter the species composition of grasslands and directly change the structure of grassland ecosystems [214,215]. Since the early 20th century, the invasion of woody plants into grasslands and their impact on the carrying capacity of grasslands has become a serious problem for savannas [216]. As reported in savanna ecosystems, the large-scale invasion of shrubs, trees, and other plants has led to significant changes in the functioning of global ecosystems, which will have profound impacts on the biodiversity, carbon storage capacity, and supply of these ecosystems [217]. Invasions of poisonous weeds has been considered one of the most important causes of economic losses by inhibiting forage production and killing

livestock. However, a recent study concluded that toxic weeds can also have some potential positive ecological impacts on grasslands, such as promoting soil and water conservation, improving nutrient cycling and biodiversity, and protecting rangelands from excessive livestock damage [218]. Therefore, appropriate actions are needed by policy makers, managers and stakeholders to assess the ecological functions of invasive toxic weeds and to reconcile the long-term trade offs between livestock development and maintaining the ecological services provided by grasslands.

Animal husbandry plays an important role in eradicating hunger and improving malnutrition [219]. Protein and energy from livestock products are the main sources of human nutrition. Therefore, improving and sustaining livestock development is critical to advancing the United Nations SDGs, especially in addressing zero hunger and mitigating climate change [220]. Rational livestock management is a major initiative to promote healthy ecosystem development. Therefore, rangeland management, as a determinant of maintaining biodiversity, ecosystem services and landscape [221], is more important to focus on exploring its driving mechanisms on grassland species structure and ecosystem stability [222]. In addition, natural hazards are of great importance to our understanding of global biogenic burning and its emissions, carbon cycling, biodiversity, conservation and land management, especially as ecosystem succession and biodiversity changes associated with grassland fires are critical to the patterns and dynamics of ecosystem functions and services [223].

In summary, sustainable management of grasslands requires a deep understanding of the functional relationships among these factors due to the spatial heterogeneity of environmental conditions, production potential and flora composition [224]. It is more important to jointly explore the similarities and differences in the structural composition of grassland ecosystems in multiple factors and scales, extract the key factors affecting the service capacity of grassland ecosystems, and apply them to the restoration of the grassland vegetation of KDC.

#### **3. Materials and Methods**

Based on the research of Khan et al. and Chapman et al. [225,226], we performed a systematic literature search and review following the protocol from Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) including quantitative statistics and qualitative content analysis. Systematic reviews have an advantage over traditional reviews and commentaries in that they cover studies by following an explicitly formulated procedure, which can help readers to understand the whole protocol followed for the literature review [227].

#### *Literature Search and Selection*

The first step was identifying records. We conducted a systematic search of peerreviewed literature (articles, reviews) using the key words in WOS and CNKI databases (Table 2). A total of 99 repeated references were excluded; 3821 references were selected for review. The search timeframe for both databases was the maximum timeframe of the databases, and the search deadline was 1 November 2022.

The second step was screening. We screened the titles and abstracts of each article to select articles that assessed, depicted, and quantified or mapped grassland ecosystem structure and stability which were eligible for full-text reading (*n* = 278). The third step was reading the full text of each of the selected publications, and, finally, a total of 133 references were chosen as case studies. To summarize the information about main structural characteristics, stability, structure–stability relationship and influencing factors and suggest future directions of ecosystem service-capacity enhancement from case studies, we recorded the annual distribution of the literature, distribution of countries of publication, types of grassland structure, disturbance factors of grassland stability, and research methods (Figure 5).


*Plants* **2023**, *12*, x 18 of 33

"Landscape pattern" OR "Forage mixes" OR "ecological corridors" OR "Time change" OR "Interannual variation" OR

#### **Table 2.** Literature search terms and results. "Month change" AND "Ecosystem services"

methods (Figure 5).

tion, types of grassland structure, disturbance factors of grassland stability, and research

**Figure 5.** The flow diagram showing the steps of the methodology and selection process used for the systematic literature review on the left and the literature review process on the right. **Figure 5.** The flow diagram showing the steps of the methodology and selection process used for the systematic literature review on the left and the literature review process on the right.

#### **4. Prospects 4. Prospects**

#### *4.1. Key Scientific Issues That Need to Be Addressed 4.1. Key Scientific Issues That Need to Be Addressed*

Research on the structure and stability of grassland ecosystems has achieved great success, but there are still many scientific questions that need to be addressed. Based on the previous systematic analysis, this paper categorizes the key scientific problems to be solved in the structure and stability of grassland ecosystem into three aspects: structure optimization and stability improvement and their relationship. Research on the structure and stability of grassland ecosystems has achieved great success, but there are still many scientific questions that need to be addressed. Based on the previous systematic analysis, this paper categorizes the key scientific problems to be solved in the structure and stability of grassland ecosystem into three aspects: structure optimization and stability improvement and their relationship.
