Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Spatial distribution pattern of colonized native semi-shrubs in two artificial vegetation restoration patterns in Mu Us sandy land, North China

  • Ercha Hu,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations College of Forestry, Inner Mongolia Agricultural University, Hohhot, Inner Mongolia, China, Inner Mongolia Academy of Forestry Sciences, Hohhot, Inner Mongolia, China

    ⨯
  • Runhong Gao

    Roles Data curation, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – review & editing

    grhzwdm@163.com

    Affiliation College of Forestry, Inner Mongolia Agricultural University, Hohhot, Inner Mongolia, China

    ⨯

Abstract

Vegetation construction is a key process for restoring and rehabilitating degraded ecosystems. However, the spatial pattern and process of native plants colonized by different vegetation restoration methods in semi-arid sandy land are poorly understood. In this study, two artificial vegetation restoration patterns (P1: row belt restoration pattern of Salix matsudana with low coverage; P2: a living sand barrier pattern of Caryopteris mongolica with low coverage) were selected to analyze the spatial distribution pattern and interspecific association of the colonizing native shrubs. The effects of the two restoration models on the spatial patterns of the main native semi-shrubs of the colonies (i.e., Artemisia ordosica and Corethrodendron lignosum var. leave) were studied using single variable and bivariate transformation point pattern analysis based on Ripley’s L function. Our results showed that two restoration patterns significantly facilitated the establishment of A. ordosica and C. lignosum var. leave, with their coverage reaching 17.04% and 22.62%, respectively. In P1, the spatial distribution pattern of colonial shrubs tended to be a random distribution, and there was no spatial correlation between the species. In P2, the colonial shrub aggregation distribution was more dominant, and with the increase in scale, the aggregation distribution changed to a random distribution, whereas the interspecific association was negatively correlated. The differences in the spatial distribution patterns of colonized native semi-shrubs in these two restoration patterns could be related to the life form of planted plants, configuration methods, biological characteristics of colonized plants, and intra- and interspecific relationships of plants. Our results demonstrated that the nurse effect of artificially planted vegetation in the early stage of sand ecological restoration effectively facilitated the near-natural succession of communities. These findings have important implications for ecological restoration of degraded sandy land in the semi-arid region of northern China.

Citation: Hu E, Gao R (2024) Spatial distribution pattern of colonized native semi-shrubs in two artificial vegetation restoration patterns in Mu Us sandy land, North China. PLoS ONE 19(6): e0304204. https://doi.org/10.1371/journal.pone.0304204

Editor: Xiao-Dong Yang, Ningbo University, CHINA

Received: December 12, 2023; Accepted: May 8, 2024; Published: June 6, 2024

Copyright: © 2024 Hu, Gao. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript.

Funding: This study was funded by the Inner Mongolia Science and Technology Major Project to EH (grant no. 2019ZD007), Natural Science Foundation of Inner Mongolia Autonomous Region to RG (grant no. 2020ZD04).

Competing interests: The authors have declared that no competing interests exist.

Introduction

Desertification is a major environmental problem in arid, semi-arid, and dry sub-humid areas, and has a widespread impact on regional society and economy [1, 2]. China is one of the most seriously desertified countries in the world, and desertification has affected more than 25% of the national land area [3]. The Mu Us sandy land is located in the agro-pastoral ecotone of semi-arid regions of northern China, where ecosystems are typically fragile and sensitive to periodic drought [4] and unsustainable human activities [5]. For decades, this area has suffered from serious desertification, primarily due to improper exploitation and wind erosion [5, 6] implying that the restoration of the degraded ecosystem is urgently needed, as well as this being a big challenge [7].

Vegetation construction is commonly used as a windbreak to stabilize sand [8], and is proven to be one of the most low-cost and effective ecological restoration measures [9, 10]. It is generally believed that vegetation construction plays an important role in reducing wind erosion, improving soil nutrient content, and regulating climate [11]. However, inappropriately implemented vegetation restoration may cause negative ecological effects, while large scale and high-density artificial afforestation may lead to the reduction of deep percolation and groundwater recharge in arid and semi-arid areas [12–14]. Plant interactions and their underlying mechanisms must be understood when formulating a vegetation restoration plan. Further research is needed to clarify the processes and mechanisms of plant community establishment under different restoration patterns, to guide the selection of appropriate technologies.

Intra- and inter-species interactions are key processes that affect the structure and dynamics of plant communities [15, 16]. Plant interactions include positive and negative effects, which together with environmental factors affect the spatial patterns and succession of communities [17, 18]. The “facilitation” model suggests that established vegetation can facilitate the integration of new individuals into the community, while the “inhibition” model suggests that once the earlier colonists are established, they will continue to exclude or restrain later colonists until the former is dead or damaged [19]. Many studies have demonstrated that facilitation and inhibition are common in natural communities; they determine the direction and speed of community succession [20, 21]; their manipulation can be used to affect the development of degraded ecosystems into desired communities [22]. The life form of interacting species largely influenced the interaction outcomes [23]. Shrubs had large facilitative effects, while herbs had strong inhibitive effects, especially on other herb species [23]. The nurse plant effect of woody plants on their underlying herbs have been well documented in desert plant communities with different life forms [24, 25]. In most cases, competition plays a dominant role in a productive environment with more available resources, while facilitation effects would be more important under greater environmental stresses [26–28]. Many empirical studies and meta-analyses have been conducted to test this prediction in arid and semiarid ecosystems [23, 29]. However, the role of positive and negative interactions in different artificial vegetation restoration measures remains uncertain. An examination of spatial distribution patterns and interactions among species in different artificial vegetation restoration measures would provide valuable information on how communities are shaped, which is important for the restoration and management of degraded ecosystems.

A spatial distribution pattern is the specific manifestation of a plant community structure, mainly determined by competition (negative effect) and facilitation (positive effect) [30]. If facilitation is the main factor determining the spatial pattern of plants, distribution tends to be aggregated, and if competition is more dominant, the distribution tends to be uniform. If both facilitation and competition are absent, the distribution tends to be random [31]. As a typical example, the seedlings of trees or shrubs are commonly found clustered near the adult individuals [32]. Studying the spatial distribution pattern of species is helpful in understanding the specific performance of plant interactions in life strategies such as recruitment, growth, death, and resource utilization [33, 34].

The reconstruction of artificial vegetation is an effective measure for the restoration of degraded ecosystems in sandy land. However, artificial promotion of near-natural vegetation restoration has not received enough attention. To better understand and clarify the effects of different artificial vegetation restoration patterns on native plant restoration in Mu Us sandy land, two artificial vegetation restoration patterns were selected to analyze the spatial distribution pattern and interspecific association of the colonized native shrubs. The objectives of this study were: (1) to analyze the spatial distribution patterns and inter-species associations of naturally restored native shrubs A. ordosica and C. lignosum var. leave in two vegetation restoration patterns, and (2) to reveal the recruitment mechanisms of native shrubs in the early stage of artificial vegetation restoration in sandy land. The research can provide a scientific base and theoretical support model for vegetation restoration in Mu Us sandy land.

Materials and methods

Study sites and species

The study area was located in the Wushen Banner (109°15’37.71″E, 38°51’5.07″N), an agro-pastoral region of Inner Mongolia, northern China (Fig 1). The area lies in the belly of the Mu Us sandy land and is characterized by a mid-temperate, semi-arid continental monsoon climate. The mean annual temperature is approximately 6.8°C, and the annual precipitation is 350–400 mm, 62% of which falls from July to September. The average wind speed is 2.7–3.5 m/s, with predominantly northwesterly winds in the spring. The terrain is relatively flat, with fixed, semi fixed, semi mobile dunes and inter dune lowlands alternately distributed. The dominant vegetation comprises A. ordosica, C. lignosum var. leave, Salix psammophila, and other desert shrubs.

thumbnail
Download:
Fig 1. The current landscape of the sampling sites.

https://doi.org/10.1371/journal.pone.0304204.g001

Since the 1970s, an aerial seeding sand control project has been implemented in Mu Us sandy land. After more than 40 years of restoration and succession, fixed sand dunes and semi fixed sand dunes with A. ordosica, C. lignosum var. leave and S. cheilophila as the dominant species were created but due to human factors, such as overgrazing and the effects of climatic drought, resulting in degradation of some areas [4]. Two ecological restoration patterns, S. matsudana plantations and C. mongolica living sand barriers, were selected to study the spatial patterns of colonial native shrubs. Both sites were located in the middle and lower parts of the connected dunes, with similar environmental conditions, and each pattern area is approximately 0.5 hm2. Project implementation began in 2017. Before restoration, the two plots were mobile dunes without vegetation but from the third year it was apparent that the bare sand was gradually being colonized by the native dominant shrubs A. ordosica and C. lignosum var. leave. After 4 years of restoration, the number and crowns of the native shrubs A. ordosica and C. lignosum var. leave had increased significantly, but herbs were still rare, with only a small number of annual herbs, such as Agriophyllum squarrosum and Corispermum hyssopifolium, being found. Because of this, these herbs were not used in this study.

A. ordosica is a semi-shrub, 50–100 cm in height, with a strong woody stock, much branched. The stems of A.ordosica seedlings are brown or dark purple, while the adults are dark gray or dark brown. A. ordosica has tap root system, and its root system is mainly distributed in the upper 30 cm of the sand, while its main roots may reach 1–3 m deep. The recruitment of A. ordosica is generally realized by seed reproduction and is characterized by R-strategy species, and the plant begins to reproduce at the age of 2–3 years. A. ordosica has a high flowering and seed setting rate, and the volume and quality of achenes are very small. These are also the special structures of A. ordosica seeds to adapt to wind-borne transmission. Small seeds can also reduce the need for water for germination, thereby increasing the germination rate of seeds under drought conditions [35, 36].

C. lignosum var. leave, semi-shrub, 40–80 cm tall, much branched, seedling stems green, while adults gray white or gray brown. C. lignosum var. leave is a perennial deciduous semi-shrub with strong adaptability and can grow on extremely arid and barren semi-fixed and fixed sandy land. C. lignosum var. leave has strong natural regeneration ability, recruitment includes seed reproduction and clonal reproduction, and clonal reproduction plays an important role in its life history. C. lignosum var. leave has underground rhizomes, and buds are on the nodes and tops of rhizomes. These buds can grow new plants upward. In natural habitats, genet generally grows for 2 to 3 years and then sprouts 1 to several rhizomes at the root neck. These rhizomes grow horizontally underground for a period of time and then turn upward to develop into daughter ramets, which continuously produce new plants. The study of clonal growth of C. lignosum var. leave in Mu Us Sandy Land found that the average number of underground rhizomes per individual plant of C. lignosum var. leave can reach 14, the total length is 14.36 m, and the average length of single underground rhizome is about 0.82 m [37, 38].

Experimental design and field surveys

A set of 30 m × 30 m sampling plots was established in each of the two restoration patterns, using a coordinate system with due east as the x-axis (perpendicular to rows), and due north as the y-axis (parallel to rows), with the southwest vertex of the plot as the origin (0,0) (Fig 1). Each sample plot was divided into 9 contiguous 10 m × 10 m subplots. In each subplot, the relative position coordinates, canopy length (L), canopy width (W), and canopy height (H), ground diameter (or diameter at breast height) and growth status of each A. ordosica and C. lignosum var. leave was measured. The canopy was assumed to be an ellipse, and its area calculated using the formula (π × L × W)/4 (Table 1). Previous research has found a significant relationship between the canopy width, height of shrubs, and the development stage of shrubs in arid environments [39, 40]. Using growth indexes such as height, canopy, branch number and lignification degree of A. ordosica and C. lignosum var. leave, combined with the results from previous studies[10, 35], individual plants were divided into two groups, namely adults and seedlings (seedling group: average heights and crown were less than 30 cm, single plant, low lignification, and the stems of A. ordosica were brown or dark purple, and the stems of C. lignosum var. leave were green; adult group: average heights and crown were more than 30 cm, and the stems of A. ordosica were dark gray or dark brown, and the stems of C. lignosum var. leave were gray white or gray brown).

thumbnail
Download:
Table 1. General situation of sampling sites.

https://doi.org/10.1371/journal.pone.0304204.t001

During afforestation, 6 wind erosion piles were evenly set between the rows of the two sample plots, and 6 wind erosion piles were set up in the downwind bare sand dunes as a control. The initial height of the wind erosion pile was 40 cm. The height of the wind erosion pile was measured during the vegetation survey to obtain the amount of wind erosion.

Data analysis

Spatial distribution patterns were analyzed using Ripley’s L-function [41, 42], which is developed from Ripley’s K-function. Ripley’s K-function is based on two-dimensional coordinate points of individuals and was used to analyze population distribution pattern and spatial association at different scales [41]. It is the most common method of population spatial distribution pattern analysis, and is estimated from the function K(r): (1)

Where A is the sample plot area, n is the total number of plant individuals, uij is the distance between point i and point j, when uij≤r, Ir(uij) = 1, when uij≥r, Ir(uij) = 0; Wij is the ratio of the arc length of the circle with point i as the center and uij as the radius in area A to the whole circumference, which can correct the error caused by boundary effect.

The expression of L(r) function is: (2)

when L(r) = 0, it indicates that the population is randomly distributed, when L(r) > 0, the distribution is aggregated, and when L(r) < 0, it is uniformly distributed.

Bivariate K12(r) function and modified L12(r) were used to analyze the spatial correlation of different age groups and different populations. The calculation formulas are as follows: (3) (4) where n1 and n2 are the total number of individuals in species 1 (or seedlings) and species 2 (or adults), respectively, other symbolic meanings are the same as in Formulas (3) and (4). When L12(r) = 0, the two populations had no correlation at r scale, when L12(r) > 0, the two populations were positively correlated at the r scale, when L12(r) < 0, the two populations were negatively correlated at r scale.

The Monte−Carlo fitting test was used to randomly simulate 1000 iterations to calculate the upper and lower envelopes, with a confidence level of 99%. For each r value, the values of L(r) and L12(r) were calculated and the process repeated until the predetermined number of iterations was reached. The L(r) value falls above the envelope line as an aggregated distribution, is randomly distributed within the envelope line, and is uniformly distributed below the envelope line. For the spatial association of the two populations, the value of L12(r) indicates a positive association above the envelope, no association between the envelopes, and a negative association below the envelope. All the special distribution calculations were performed using ADE-4 package (http://pbil.univ-lyon1.fr/ADE-4/) [43].

Results

Community characteristics and wind erosion

The individual number of A. ordosica and C. lignosum var. leave was significantly different in the two restoration patterns, with 157 individuals mapped for the P1 and 231 for the P2 (Fig 2), and coverage of 17.04% and 22.62%, respectively, accounting for 70.20% and 86.45% of the total coverage of the community. In P1, there were 56 and 101 individuals of A. ordosica and C. lignosum var. leave, respectively, while in P2, their numbers 134 and 97. This suggests that P1 is more conducive to colonization by C. lignosum var. leave, and P2 is more suitable for colonization by A. ordosica. There was no significant difference in their growth between the two patterns. In P1 and P2, the seedlings of colonized shrubs accounted for 40.76% and 26.84% of the total individuals. The proportion of seedlings indicates that the population is in the growth stage. In addition, a small number of annual herbs such as A. squarrosum and C. hyssopifolium appeared in the two plots.

thumbnail
Download:
Fig 2. Mapped point pattern of C. lignosum var. leave and A. ordosica in sampling sites.

AO: A. ordosica, CL: C. lignosum var. leave. Symbol sizes represent the crown of individuals.

https://doi.org/10.1371/journal.pone.0304204.g002

The results of wind erosion measurement showed that the average wind erosion amount of P1, P2 and mobile sand were -3.67 cm, 0.33 cm and -81.35 cm. There was no significant difference in wind erosion amount between the two vegetation restoration modes, but it was significantly lower than that of the mobile sand (P < 0.05) (Fig 3). The survey also found that although there were some residual Salix sand barriers and litters on the ground in P1, there was still obvious wind erosion, while P2 generally formed obvious sand ridges at the downwind of sand barriers due to the better wind-proof effect of C. mongolica living sand barriers, but the wind erosion in the plot was uneven and had obvious heterogeneity. The mobile sand was set in the downwind direction above the two restoration modes. Because the artificial vegetation changed the direction and size of the wind, the top of the dune at the downwind mouth was flattened, and the wind erosion was aggravated. The amount of wind erosion was significantly higher than that of the two vegetation restoration modes.

thumbnail
Download:
Fig 3. Wind erosion of sample plot.

Dash line indicates the initial height of the wind erosion piles.

https://doi.org/10.1371/journal.pone.0304204.g003

Spatial pattern analysis of colonized shrubs

In P1, all A. ordosica and C. lignosum var. leave individuals were distributed randomly at a scale of 4–12 m and aggregated at smaller and larger scales. (Fig 4A). Seedlings were randomly distributed at 5–11 m, and aggregated at other scales, while adults were randomly distributed at the study scale (Fig 4M and 4G). In P2, all individuals were aggregated below 10 m and were randomly distributed at a larger scale (Fig 4B). The seedlings were below 3m, and the adults were below 9m, showing aggregated distribution, and the larger scale was randomly distributed (Fig 4N and 4H). In P1, all individuals, adult and seedling, of A. ordosica were distributed randomly at all scales (Fig 4C, 4I and 4O), whereas in P2, seedlings were distributed randomly at all scales (Fig 4P), but adult individuals were only distributed randomly below 3 m, and aggregated on a larger scale (Fig 4J). All individuals of A. ordosica were aggregated at 2–13 m, and randomly distributed at other scales (Fig 4D). C. lignosum var. leave was randomly distributed at all scales in P1 (Fig 4E), but all individuals, adults and seedlings, were aggregated at scales below 7–10 m and randomly distributed at larger scales in P2 (Fig 4F, 4L and 4R). In P1, adult individuals were randomly distributed at all scales (Fig 4K), seedlings were randomly distributed at 5–10 m, and aggregated at smaller and larger scales (Fig 4Q). In P2, adult individuals were aggregated below 9 m while seedlings were aggregated below 3 m and randomly distributed at a larger scale (Fig 4F and 4R).

thumbnail
Download:
Fig 4. Spatial distribution patterns of A. ordosica and C. lignosum var. leave in two restoration patterns.

AG: adult group, SG: seedling group, AO: A. ordosica, CL: C. lignosum var. leave. The solid line indicates the L(r), and the dashed lines represent the confidence intervals.

https://doi.org/10.1371/journal.pone.0304204.g004

Intra- and interspecific spatial associations

The association analysis included the spatial association of adult and seedling A. ordosica and C. lignosum var. leave. Under the two restoration patterns, there were differences in the spatial associations of A. ordosica and C. lignosum var. leave. P1 had no spatial association (Fig 5A), but P2 had negative spatial association (Fig 5B). In the two restoration patterns, the spatial association between all seedlings and adults of A. ordosica and C. lignosum var. leave was different (Fig 6A and 6B). In the two restoration patterns, there was no spatial correlation between the adults and seedlings of A. ordosica (Fig 6C and 6D). The analysis of spatial association between adults and seedlings of C. lignosum var. leave showed negative association within 5 m and no association above 5 m in P1 (Fig 6E). In P2, there was no spatial association between adults and seedlings (Fig 6F).

thumbnail
Download:
Fig 5. Spatial association of A. ordosica and C. lignosum var. leave in two restoration patterns.

https://doi.org/10.1371/journal.pone.0304204.g005

thumbnail
Download:
Fig 6. Association between seedlings and adults of A. ordosica and C. lignosum var. leave in two restoration patterns.

https://doi.org/10.1371/journal.pone.0304204.g006

Discussion

Effects of artificial measures on natural restoration of native shrubs

Our study demonstrated that artificial measures such as low-density tree afforestation and the construction of living shrub sand barriers could promote the colonization by native shrubs and quickly fix mobile dunes during the early stages of vegetation restoration. Sandy land in semi-arid areas degraded below a certain threshold could not be naturally restored, and artificial measures were required to re-establish vegetation [44]. Studies had shown that the construction of straw checkerboards and shrub seedlings plantations could significantly improve topsoil development on the dune surface, as well as accelerate the plant diversity, vegetation cover, and plant density [11]. A sand dune was a dynamic environment, determined by wind erosion, sand accumulation, and dune encroachment, which directly impacted the spatial distribution and growth of plants [45, 46]. We found that planting native trees (S. matsudana) at low densities (about 50% lower than conventional afforestation) and low coverage of living sand barriers (C. mongolica) could provide a nurse effect to facilitate the colonization by A. ordosica and C. lignosum var. leave. After four years, the cover of the two native shrubs in the two plots had reached 17.04% and 22.62%, respectively. A. ordosica and C. lignosum var. leave remained the dominant species in Mu Us sandy land [35], and their colonization was conducive to the formation of more stable near-natural vegetation.

The interactions between plants and their environment determined the composition and structure of plant communities and affected their dynamics [18, 47]. Artificial plantings provided a living environment for the invasion of native plants in the early stages of vegetation restoration by changing the microenvironment through wind speed reduction, sand flow reduction, seed capture, and seedling protection [30, 48]. The number and distribution patterns of A. ordosica and C. lignosum var. leave in different restoration projects were different in this study (Table 1 and Fig 2), a finding consistent with previous research [49]. In the fourth year of vegetation restoration, it was observed that the living shrub sand barrier of C. mongolica was more conducive to the colonization and settlement of A. ordosica and C. lignosum var. leave than the S. matsudana plantation pattern. This was due to the dwarf shrub C. mongolica being closer to the ground and forming a dense barrier, effectively reducing wind speed and surface sand flow; the microenvironment created between rows was more favorable for seed capture and germination. The fallen branches and leaves of artificially planted trees formed a natural barrier under the trees, which helped seed capture and seedling recruitment, allowing native shrubs to grow near the trees (Fig 1). Regional vegetation without artificial measures did not show signs of recovery (Fig 1). This suggested that viable artificial vegetation reconstruction was crucial to the restoration of a degraded sandy ecosystem.

Spatial patterns of native shrubs in two restoration patterns

Population structure and spatial pattern were the results of the interaction of biological characteristics, environmental factors, and interspecific competition, and they reflected population dynamics and community succession [47, 50]. Analysis of the spatial pattern and population structure at different stages of population growth could provide basic population information, such as recruitment, individual mortality, and intraspecies competition [33, 35].

A. ordosica, a native semi-shrub species, was the main dominant species that was broadly distributed in the Mu Us sandy land [10]. It adapted to the moving sand dune environment by changing the biomass allocation under burial and denudation stress [51]. This study showed that the adults and seedlings of A. ordosica were randomly distributed at all scales in PI, which may be primarily related to its reproductive mode. The seeds of A. ordosica were small and light, being suitable for wind dispersal, which could be over a significant distance. Seeds were captured by artificially planted vegetation, where they germinated and survived in the plot, a process that occured randomly. However, in P2, adult individuals were randomly distributed at small scales but clustered at larger scales (Fig 4J). These results were different from the previous research [10, 35]. Liu and Zhang studied the population pattern and structure of A. ordosica in enclosures and found that seedlings and adult populations showed a small-scale aggregation distribution with a random distribution at a large scale [35]. It was generally understood that small scale aggregation distribution was beneficial for seedlings as it assisted them to resist adverse natural conditions [52]. In this study, the sand barrier changed wind speed and sand burial, thereby influencing the distribution pattern of A. ordosica. The main reason for this difference might be that the artificial restoration model of C. mongolica sand barrier had an impact on the spatial distribution pattern of colonial native semi-shrubs. In addition, there were relatively more A. ordosica settled in the early stage of restoration, which was aggregated in a limited space, and the settlement growth time was shorter. A. ordosica had not yet entered the self-thinning stage of aging and death. Therefore, the spatial distribution pattern of the population was determined by many factors, such as human disturbance, intraspecific and interspecific competition, environmental factors and plant biological characteristics.

C. lignosum var. leave was a rhizomatous clonal semi-shrub widely dispersed by aerial seeding in Mu Us sandy land in northern China since the 1950s. The high clonal propagation and population growth rate of C. lignosum var. leave enabled it to quickly colonize in mobile dunes and to realize rapid spatial expansion [53]. The spatial distributions of the two patterns were different in this study. Adults were randomly distributed in P1, whereas in P2 they were aggregated within 7 m and randomly distributed as the scale increased (Fig 4K and 4L). The nurse effect of planted trees created an appropriate microenvironment for the seed germination and settlement of C. lignosum var. leave. As a result, the distribution of adult C. lignosum var. leave was more similar to that in the natural environment and was random. In P2, due to the height limitation of the shrub sand barrier, the nurse effect range was limited, meaning C. lignosum var. leave aggregation occurred only within a certain range (Fig 4R). This was also related to the growth status of C. mongolica in the plot, including height differences and sand barrier death fragments. Studies showed that populations of C. lignosum var. leave, once established, were able to realize rapid population expansion through high clonal propagation [53]. The first settled individuals produced more seedlings through asexual reproduction in its surroundings, resulting in seedlings aggregated on a small scale.

In comparison to the mechanical sand barrier, the plant sand barrier used living plants as raw materials, which significantly shortened the vegetation restoration cycle, and had the advantages of low cost, rapid effect, high stability, and strong ecological function in the practice of desertification control. In this study, a living sand barrier with a large row spacing (5 m) was used, which not only reduced costs but also provided enough space for the colonization by native shrubs, thereby increasing species diversity. These were also the most significant differences compared to conventional dense grid sand barriers. The study also found that living sand barriers affected the spatial distribution pattern of plants by controlling the direction and speed of wind, resulting in changes in the erosion and deposition patterns. Therefore, afforestation species and configuration methods had an important impact on the spatial patterns of colonized plants. Once the native shrubs were planted, they gradually became nurse plants and promoted rapid vegetation restoration.

The differences in the spatial distribution patterns of colonized native semi-shrubs in these two restoration patterns could be related to the life form of planted plants, configuration methods and biological characteristics of colonized plants. S. matsudana and C. mongolica were native tree species in Mu Us sandy land, which had good windbreak and sand fixation function and adaptability. The results of this study showed that the colonial native semi-shrubs tended to be more randomly distributed at P1 and more clustered at P2. This might be related to the fact that S. matsudana, as a tree, had thin and high branches, and its wind erosion resistance was lower than that of low semi-shrub C. mongolica. Moreover, C. mongolica was planted in sand barrier mode, and its wind erosion resistance function was strong (Fig 3). In addition, there were also differences in the spatial distribution of wind erosion and sand burial in the two plots. The wind erosion under the S. matsudana forest was more uniform and dominated by wind erosion, while the C. mongolica was lower and formed a sand ridge in the downwind near the sand barrier. The uneven distribution of erosion and sand burial increased heterogeneity, which was also conducive to the settlement of native colonial plants and also affected its distribution pattern. In summary, the differences in morphological characteristics and planting patterns of S. matsudana and C. mongolica leaded to wind erosion and sand burial patterns in the plots, thus affecting the spatial distribution pattern and quantity of colonial native semi-shrubs.

Intra- and interspecific spatial association of colonial native shrubs

Plant-plant interactions were a major driving force in plant community succession or vegetation restoration [54]. Positive (promotion) and negative (competition) effects existed and occured simultaneously between plants in the community, and the environmental conditions determined which effect was more important [26]. The positive effect of nurse plants increased under stressful abiotic conditions, which were evidently greater in higher mountains, particularly during the earlier stages of degraded ecosystem restoration. As the main type of facilitation, nurse effect was used widely in ecological restoration of degraded habitats [55], and it can effectively shorten the recovery time. Nurse plants primarily improved the surrounding microenvironment, enabling adjacent species to settle in their original environment. This effect was particularly important in areas with harsh environmental conditions, such as arid and semi-arid areas, alpine areas, and salt marshes [25, 56]. Plants that were the first established in an area will impact the recruitment, growth, and reproductive success of species that arrive later [57]. This study also supported the stress gradient hypothesis. The vegetation established by different artificial measures in the harsh semi-arid sandy environment played the role of nurse plants and promoted the colonization by native shrubs A. ordosica and C. lignosum var. leave.

The stress gradients and biological characteristics of nurse and target plants determined the results of plant interactions, which in turn affected the success or failure of vegetation restoration [58]. The target plants in this study were native shrubs and herbs, and they were selected because the establishment of near-natural communities was the key to vegetation restoration success. In the early stage of sand dune vegetation restoration, both restoration methods promoted the colonization by native shrubs A. ordosica and C. lignosum var. leave. The spatial distribution pattern and spatial association of colonial shrubs could explain the facilitation mechanism of nurse plants [59]. Nurse plants could make the probability of target plants appearing under their canopy higher than those on open land [60], and their survival rates in different directions were also significantly different [61], thus affecting the spatial distribution pattern of target plants. In this study, there was no association between the populations of C. lignosum var. leave and A. ordosica in P1. Tree stems were thin and tall, forming different synusia structures in C. lignosum var. leave and A. ordosica, and had no direct competition with shrubs. The space and resources in the sample plot were sufficient and A. ordosica and C. lignosum var. leave could grow in their own spaces, with no evident interspecific competition. A. ordosica and C. lignosum var. leave had a negative association in P2 (Fig 5B), because the spacing of C. mongolica sand barrier planting was small, and the age and individual size was similar to the target plant [62], indicating that the colonial shrubs were excluded from the inter row distribution. The number of individuals of A. ordosica and C. lignosum var. leave was greater than in P1, and their spatial distribution pattern demonstrated aggregation. With community succession, the competition for space and resources between the two species increases. However, low-density (large row spacing) planting methods not only minimized the direct competition between nurse plants and target plants in the early stages of sand vegetation restoration, but also provided suitable seed germination, seedling recruitment, and growth conditions for the colonization by native plants in our study. Low-density planting methods not only effectively resolved competition between artificially planted vegetation and native plants in the early stages of vegetation restoration, but also improved the effectiveness of vegetation restoration. Other studies in the study area concluded that dune stabilization can change the spatial pattern of shrub populations by weakening the spatial association between native shrub individuals [63].

The spatial association analysis between individuals at different growth stages could help us to explain the mechanism of spatial pattern formation in the early stages of vegetation restoration [64]. C. lignosum var. leave and A. ordosica were dominant plants in different succession stages in Mu Us sandy land. C. lignosum var. leave was a typical guerrilla clonal plant, and clonal reproduction plays an important role in its life history, while A. ordosica mainly relied on seed reproduction to maintain population and population diffusion. The results of this study showed that different ecological restoration methods strongly affected the spatial association of colonial native shrubs, and that there was no spatial association between adults and seedlings of A. ordosica (Fig 6C and 6D). In P1, C. lignosum var. leave adults and seedlings showed a negative spatial correlation on a smaller scale, which was contrary to the results of natural population studies [35]. These results were due to the nurse effect observed in different restoration methods, and the ecological strategy of colonizing shrubs may jointly affect the spatial pattern of the population. First, there were many natural A. ordosica distributed around the plot. The seeds of A. ordosica were small and light, which were easy to reach the plot in the wind. Secondly, the seeds had mucilage, and the germination water demand was small [65, 66]. These breeding strategies adapted to the arid sandy environment maked it quickly settle in the plot, and the establishment process tended to be random. The reproductive strategy of C. lignosum var. leave was different from that of A. ordosica. Even if the number of colonies was small, it relied on its strong rhizome cloning ability to quickly occupy limited resources and expanded the population size. In the two restoration modes, the spatial correlation between adult and seedling of C. lignosum var. leave was negatively correlated at a small scale (5m). This may be a strategy that when the diffusion space was large enough in the early stage of vegetation restoration, the clonal foraging behavior of C. lignosum var. leave made the daughter plant colonize as far as possible from the genet to avoid premature competition. In this study, a small number of C. lignosum var. leave seeds may first settle in the plot and start clonal reproduction 3–4 years later. Or at the same time, the rhizomes of C. lignosum var. leave around the plot entered the plot for clonal reproduction [37, 63]. However, the landscape heterogeneity and the species pool in the restoration area also affected the restored species. Sufficient seed sources around the restoration area were a prerequisite for the colonization by native plants.

In this study, both vegetation restoration models promoted the colonization of native shrubs A. ordosica and C. lignosum var. leave, and had different effects on the spatial distribution pattern of the two colonial native semi-shrubs. However, the process of vegetation restoration was a dynamic process. With the succession of the community, the intra- and interspecific relationships would continue to change. It was not enough to study the population pattern at a certain time point. Therefore, we recommended that future studies should be carried out from more time series to grasp the whole process of vegetation restoration in order to better guide production activities.

Conclusion

This study provided novel insights into the ecological restoration of degraded sandy ecosystems by analyzing the spatial patterns of early colonized semi-shrubs, A. ordosica and C. lignosum var. leave, in two restoration patterns. In the early stages of restoration, the nurse effect of artificial plantings in the two restoration patterns of low-density artificial afforestation of S. matsudana and the living sand barrier of C. mongolica promoted the colonization by local semi-shrubs A. ordosica and C. lignosum var. leave. Two restoration patterns have important effects on the number and spatial distribution pattern of colonizing native shrubs. In S. matsudana plantation pattern, the spatial distribution pattern of colonial shrubs tended to be random, and there was no spatial association between species. In the C. mongolica living sand barrier pattern, the colonial shrubs aggregation distribution was more dominant, and with the increase of scale, the aggregation distribution changed to random distribution, while the interspecific association was negatively correlated. In the ecological restoration of degraded sandy land, the life form and configuration of plant species and the biological characteristics of colonial plants jointly affected the community structure and succession process. This study clearly demonstrated that low-cost restoration measures could achieve better restoration results in ecological restoration of sandy land in semi-arid areas. As it was necessary to evaluate the promotion effect of different artificial measures on dominant native plants to improve the effectiveness of vegetation restoration, our results provided a theoretical basis for ecological restoration of degraded sandy land.

Acknowledgments

We thank Zihao Li, Guangyu Hong, and Haifeng Yang for their help during field survey and data collections. We thank Dr. Shaokun Wang (Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences) and Dr. Zhaoping Yang (Nanjing University of Information Science & Technology) for their kind help in revising and improving the manuscript.

References

  1. 1. Reynolds JF, Smith DMS, Lambin EF, Turner B, Mortimore M, Batterbury SP, et al. Global desertification: building a science for dryland development. Science. 2007;316(5826):847–51. pmid:17495163
  2. 2. Vogt J, Safriel U, Von Maltitz G, Sokona Y, Zougmore R, Bastin G, et al. Monitoring and assessment of land degradation and desertification: towards new conceptual and integrated approaches. Land Degradation Development. 2011;22(2):150–65.
  3. 3. Liu J, Diamond J. China’s environment in a globalizing world. Nature. 2005;435(7046):1179–86. pmid:15988514
  4. 4. Wang X, Chen FH, Dong Z, Xia D. Evolution of the southern Mu Us Desert in North China over the past 50 years: an analysis using proxies of human activity and climate parameters. Land Degradation Developmental. 2005;16(4):351–66.
  5. 5. Wu B, Ci LJ. Landscape change and desertification development in the Mu Us Sandland, Northern China. Journal of Arid Environments. 2002;50(3):429–44.
  6. 6. Jia F, Lu R, Gao S, Li J, Liu X. Holocene aeolian activities in the southeastern Mu Us Desert, China. Aeolian Research. 2015;19:267–74.
  7. 7. Choi YD, Temperton VM, Allen EB, Grootjans AP, Halassy M, Hobbs RJ, et al. Ecological restoration for future sustainability in a changing environment. Ecoscience. 2008;15(1):53–64.
  8. 8. Sun Z, Mao Z, Yang L, Liu Z, Han J, Wanag H, et al. Impacts of climate change and afforestation on vegetation dynamic in the Mu Us Desert, China. Ecological indicators. 2021;129:108020.
  9. 9. Visser N, Botha J, Hardy M. Re-establishing vegetation on bare patches in the Nama Karoo, South Africa. Journal of Arid Environments. 2004;57(2):155–77.
  10. 10. Wu B, Yang H. Spatial patterns and natural recruitment of native shrubs in a semi-arid sandy land. PloS one. 2013;8(3):e58331. pmid:23505489
  11. 11. Li Y, Cui J, Zhang T, Okuro T, Drake S. Effectiveness of sand-fixing measures on desert land restoration in Kerqin Sandy Land, northern China. Ecological Engineering. 2009;35(1):118–27.
  12. 12. Wang X, Zhang C, Hasi E, Dong Z. Has the Three Norths Forest Shelterbelt Program solved the desertification and dust storm problems in arid and semiarid China? Journal of Arid Environments. 2010;74(1):13–22.
  13. 13. Liu X, He Y, Sun S, Zhang T, Luo Y, Zhang L, et al. Restoration of sand-stabilizing vegetation reduces deep percolation of precipitation in semi-arid sandy lands, northern China. Catena. 2022;208:105728.
  14. 14. Lu C, Zhao T, Shi X, Cao S. Ecological restoration by afforestation may increase groundwater depth and create potentially large ecological and water opportunity costs in arid and semiarid China. Journal of Cleaner Production. 2018;176:1213–22.
  15. 15. Callaway RM. Positive interactions among plants. The botanical review. 1995;61:306–49.
  16. 16. Lv X, Zuo Z, Sun J, Ni Y, Dong G. Spatial patterns of dominant species and their implication for natural secondary forest ecosystem dynamics in a reserved forest of north China. Ecological engineering. 2019;127:460–7.
  17. 17. Sterner RW, Ribic CA, Schatz GE. Testing for life historical changes in spatial patterns of four tropical tree species. The Journal of Ecology. 1986:621–33.
  18. 18. Zhang J, Song B, Li B-H, Ye J, Wang X-G, Hao Z-Q. Spatial patterns and associations of six congeneric species in an old-growth temperate forest. Acta Oecologica. 2010;36(1):29–38.
  19. 19. Connell JH, Slatyer RO. Mechanisms of succession in natural communities and their role in community stability and organization. The american naturalist. 1977;111(982):1119–44.
  20. 20. Holzapfel C, Mahall BE. Bidirectional facilitation and interference between shrubs and annuals in the Mojave Desert. Ecology. 1999;80(5):1747–61.
  21. 21. Montesinos-Navarro A, Storer I, Perez-Barrales R. Benefits for nurse and facilitated plants emerge when interactions are considered along the entire life-span. Perspectives in plant ecology, evolution and systematics. 2019;41:125483.
  22. 22. Siles G, Rey P, Alcántara J, Ramírez J. Assessing the long‐term contribution of nurse plants to restoration of Mediterranean forests through Markovian models. Journal of Applied Ecology. 2008;45(6):1790–8.
  23. 23. Gómez‐Aparicio L. The role of plant interactions in the restoration of degraded ecosystems: a meta‐analysis across life‐forms and ecosystems. Journal of Ecology. 2009;97(6):1202–14.
  24. 24. Tielbörger K, Kadmon R. Temporal environmental variation tips the balance between facilitation and interference in desert plants. Ecology. 2000;81(6):1544–53.
  25. 25. Holzapfel C, Tielbörger K, Parag HA, Kigel J, Sternberg MJB, ecology a. Annual plant–shrub interactions along an aridity gradient. Basic and applied ecology. 2006;7(3):268–79.
  26. 26. Bertness MD, Callaway R. Positive interactions in communities. Trends in ecology & evolution. 1994;9(5):191–3. pmid:21236818
  27. 27. Maestre FT, Valladares F, Reynolds JF. Is the change of plant–plant interactions with abiotic stress predictable? A meta‐analysis of field results in arid environments. Journal of Ecology. 2005;93(4):748–57.
  28. 28. Barbier N, Couteron P, Lefever R, Deblauwe V, Lejeune O. Spatial decoupling of facilitation and competition at the origin of gapped vegetation patterns. Ecology. 2008;89(6):1521–31. pmid:18589517
  29. 29. Liancourt P, Tielbörger K. Ecotypic differentiation determines the outcome of positive interactions in a dryland annual plant species. Perspectives in Plant Ecology, Evolution and Systematics. 2011;13(4):259–64.
  30. 30. Barot S, Gignoux J, Menaut J-C. Demography of a savanna palm tree: predictions from comprehensive spatial pattern analyses. Ecology. 1999;80(6):1987–2005.
  31. 31. Wiegand T, A. Moloney K. Rings, circles, and null‐models for point pattern analysis in ecology. Oikos. 2004;104(2):209–29.
  32. 32. Schenk HJ, Holzapfel C, Hamilton JG, Mahall BE. Spatial ecology of a small desert shrub on adjacent geological substrates. Journal of Ecology. 2003;91(3):383–95.
  33. 33. Druckenbrod DL, Shugart HH, Davies I. Spatial pattern and process in forest stands within the Virginia piedmont. Journal of Vegetation science. 2005;16(1):37–48.
  34. 34. Jara‐Guerrero A, De la Cruz M, Espinosa CI, Méndez M, Escudero A. Does spatial heterogeneity blur the signature of dispersal syndromes on spatial patterns of woody species? A test in a tropical dry forest. Oikos. 2015;124(10):1360–6.
  35. 35. Liu J, Zhang K. Spatial pattern and population structure of artemisia ordosica shrub in a desert grassland under enclosure, Northwest China. International Journal of Environmental Research and Public Health. 2018;15(5):946. pmid:29747420
  36. 36. Li SL, Yu FH, Werger MJ, Dong M, Zuidema PA. Habitat‐specific demography across dune fixation stages in a semi‐arid sandland: understanding the expansion, stabilization and decline of a dominant shrub. Journal of Ecology. 2011;99(2):610–20.
  37. 37. Chen Y, Dong M. Genet Characters of Hedysarum Laeve and the Characters of Its Ramet Population in Different Habitation Mu Us Sandland. Acta Phytoecologica Sinica. 2000;42:40–5.
  38. 38. Zhang C, Yang C, Dong M. The clonal integration of photosynthates in the rhizomatous half-shrub Hedysarum laeva. Acta Ecologica Sinica. 2001;21(12):1986–93.
  39. 39. Reynolds JF, Virginia RA, Kemp PR, De Soyza AG, Tremmel DC. Impact of drought on desert shrubs: effects of seasonality and degree of resource island development. Ecological Monographs. 1999;69(1):69–106.
  40. 40. Wiegand T, Milton SJ, Esler KJ, Midgley GF. Live fast, die young: estimating size-age relations and mortality pattern of shrubs species in the semi-arid Karoo, South Africa. Plant Ecology. 2000;150:115–31.
  41. 41. Ripley BD. The second-order analysis of stationary point processes. Journal of applied probability. 1976;13(2):255–66.
  42. 42. Haase P. Spatial pattern analysis in ecology based on Ripley’s K‐function: Introduction and methods of edge correction. Journal of vegetation science. 1995;6(4):575–82.
  43. 43. Thioulouse J, Chessel D, Dole´ dec S, Olivier J-M. ADE-4: a multivariate analysis and graphical display software. Statistics and computing. 1997;7:75–83.
  44. 44. Snyman H. Revegetation of bare patches in a semi-arid rangeland of South Africa: an evaluation of various techniques. Journal of arid environments. 2003;55(3):417–32.
  45. 45. Liu Z, Li X, Yan Q, Wu J. Species richness and vegetation pattern in interdune lowlands of an active dune field in Inner Mongolia, China. Biological Conservation. 2007;140(1–2):29–39.
  46. 46. Zuo X, Zhao H, Zhao X, Guo Y, Li Y, Luo Y. Plant distribution at the mobile dune scale and its relevance to soil properties and topographic features. Environmental Geology. 2008;54:1111–20.
  47. 47. Abeli T, Cauzzi P, Rossi G, Adorni M, Vagge I, Parolo G, et al. Restoring population structure and dynamics in translocated species: learning from wild populations. Plant ecology. 2016;217:183–92.
  48. 48. Lu R, Zheng J, Jia C, Liu Y, Huang Z, He H, et al. Nurse effects of patch-canopy microhabitats promote herbs community establishment in sandy land. Ecological Engineering. 2018;118:126–33.
  49. 49. Miyasaka T, Okuro T, Miyamori E, Zhao X, Takeuchi K. Effects of different restoration measures and sand dune topography on short-and long-term vegetation restoration in northeast China. Journal of Arid Environments. 2014;111:1–6.
  50. 50. Mwavu EN, Witkowski ET. Population structure and regeneration of multiple-use tree species in a semi-deciduous African tropical rainforest: implications for primate conservation. Forest Ecology and Management. 2009;258(5):840–9.
  51. 51. Li S-L, Werger MJ, Zuidema PA, Yu F-H, Dong M. Seedlings of the semi-shrub Artemisia ordosica are resistant to moderate wind denudation and sand burial in Mu Us sandland, China. Trees. 2010;24:515–21.
  52. 52. DeFalco L, Esque T, Kane J, Nicklas M. Seed banks in a degraded desert shrubland: influence of soil surface condition and harvester ant activity on seed abundance. Journal of Arid Environments. 2009;73(10):885–93.
  53. 53. Li S-L, Yu F-H, Werger MJ, Dong M, During HJ, Zuidema PA. Mobile dune fixation by a fast-growing clonal plant: a full life-cycle analysis. Scientific reports. 2015;5(1):8935.
  54. 54. Ren H, Yang L, Liu N. Nurse plant theory and its application in ecological restoration in lower subtropics of China. Progress in Natural Science. 2008;18(2):137–42.
  55. 55. Zhao H-L, Zhou R-L, Su Y-Z, Zhang H, Zhao L-Y, Drake S. Shrub facilitation of desert land restoration in the Horqin Sand Land of Inner Mongolia. Ecological engineering. 2007;31(1):1–8.
  56. 56. Hacker SD, Gaines SD. Some implications of direct positive interactions for community species diversity. Ecology. 1997;78(7):1990–2003.
  57. 57. Fukami T. Historical contingency in community assembly: integrating niches, species pools, and priority effects. Annual review of ecology, evolution, and systematics. 2015;46:1–23.
  58. 58. He Q, Bertness MD, Altieri AH. Global shifts towards positive species interactions with increasing environmental stress. Ecology letters. 2013;16(5):695–706. pmid:23363430
  59. 59. Niering W, Whittaker R, Lowe C. The Saguaro: A Population in Relation to Environment: Reproduction and survival are more affected by man’s intrusion than by environmental extremes. Science. 1963;142(3588):15–23.
  60. 60. Castro J, Zamora R, Hodar JA, Gomez JM, Gómez‐Aparicio L. Benefits of using shrubs as nurse plants for reforestation in Mediterranean mountains: a 4‐year study. Restoration Ecology. 2004;12(3):352–8.
  61. 61. Franco‐Pizaña J, Fulbright TE, Gardiner DT. Spatial relations between shrubs and Prosopis glandulosa canopies. Journal of Vegetation Science. 1995;6(1):73–8.
  62. 62. Blignaut A, Milton S. Effects of multispecies clumping on survival of three succulent plant species translocated onto mine spoil in the succulent Karoo Desert, South Africa. Restoration Ecology. 2005;13(1):15–9.
  63. 63. Zhang J, Zhang Y, Fan D, Qin S, Jia X, Wu B, et al. Effects of sand dune stabilization on the spatial pattern of Artemisia ordosica population in Mu Us Desert, Northwest China. PLoS One. 2015;10(6):e0129728. pmid:26102584
  64. 64. Alados CL, ElAich A, Papanastasis VP, Ozbek H, Navarro T, Freitas H, et al. Change in plant spatial patterns and diversity along the successional gradient of Mediterranean grazing ecosystems. Ecological Modelling. 2004;180(4):523–35.
  65. 65. Huang Z, Gutterman Y. Water absorption by mucilaginous achenes of Artemisia monosperma: floating and germination as affected by salt concentrations. Israel Journal of Plant Sciences. 1999;47(1):27–34.
  66. 66. Schenk HJ. Clonal splitting in desert shrubs. Plant Ecology. 1999;141:41–52.