Introduction

Under the background of global climate change, ecological management and protection are confronted with more significant challenges1. Particularly in some regions with fragile environments, vegetation restoration and management come across numerous unpredictable situations2. For instance, the complex and variable habitats in plateau regions have a substantial influence on the diversity and ecological functions of vegetation, thereby posing difficulties in the vegetation restoration work on plateaus around the world3,4. It has become a common understanding among scholars around the world to explore the ways for sustainable vegetation restoration and protection, from the aspect of the response mechanisms of plant growth to changes of environmental factors5. In alpine rocky desertification areas, there are large-scale exposed rocks, thin soil, and large daily temperature differences6. There exist some unique types of natural vegetation with the characteristics of drought and cold tolerance7. However, due to the limiting environment, the vegetation community is fragile7. In recent years, the Chinese government has proposed improving the quality of forestry ecological engineering, especially forest protection and restoration in areas with fragile environments8. But in this fragile environmental area, the high-quality development of natural vegetation still faces many challenges, such as vegetation degradation and water and soil loss9.

Non-tree plants including shrubs and herbaceous plants play a crucial role in alpine areas, their diversity and coverage are essential indicators of ecosystem health10. Research indicated that there is a complex correlation between non-tree plant diversity and coverage in alpine areas11. Diverse non-tree plant communities can utilize resources such as light, water, and nutrients more efficiently, leading to denser vegetation cover12. It is also possible that due to niche compression, the diversity of plant communities with high coverage may actually decrease11. The relationship is often observed in alpine areas, such as the Qinghai-Tibet Plateau, where high-diversity non-tree plant communities can enhance ecosystem productivity and stability through complementary effects and optimized resource allocation13. Maintaining stable vegetation coverage and species diversity helps reduce soil erosion and water loss, improving soil structure and microclimate conditions, thereby creating a favorable environment for non-tree plant growth14,15. However, in alpine areas, scarce resources in limited environments may lead to reduced diversity and coverage, exacerbating soil erosion and soil poverty, forming a vicious cycle16. Therefore, protecting and enhancing non-tree plant diversity is key to restoring and managing ecosystems in alpine areas.

Specific leaf area is an important parameter of leaf morphology, representing the leaf area per unit of leaf mass, reflecting the efficiency of plants in acquiring and utilizing resources17. In alpine areas, specific leaf area becomes a crucial physiological trait that reflects how non-tree plants adapt to harsh environments. Generally, high specific leaf area values indicate large photosynthetic areas and high growth rates but also high water and nutrient consumption18. In alpine areas with resource-scarce environments such as the Apennines, low specific leaf area helps plants gain a competitive advantage, supporting high diversity and coverage19. Further, vegetation degradation is typically accompanied by a decrease in diversity and coverage, making plant leaves susceptible to environmental stress, which in turn reduces the specific leaf area in alpine areas20. These leaf adaptive strategies align with the environmental stress theory21, which posits those plants enhance their adaptability and maintain ecosystem stability through morphological and physiological adjustments in adverse conditions. However, the specific leaf area varies among different species, and merely observing the size of specific leaf area cannot determine whether the leaf adopts a growth strategy to enhance resource acquisition or a defense strategy to adapt to harsh environments. For example, shrubs with similar specific leaf area in alpine regions exhibit differences in the allometric relationship between leaf dry weight and leaf area22.

However, the phenomena supported by the environmental stress theory in alpine areas may not limit to these. Especially in limited environments with alpine rocky desertification, as environmental stress increases, the non-tree plant communities that the environment can support undergo significant changes23. This may be accompanied by the rapid decline of shrubs and the rapid rise of herbaceous plants. It may lead to an apparent increase in community specific leaf area and species diversity, yet the risk of vegetation degradation is overlooked. To test this hypothesis and fill the gap in how environmental stress drives changes in adaptation strategies of specific leaf area in alpine karst areas, this study focuses on non-tree plants at different altitudes and vegetation types in alpine karst desertification areas of the Jinsha River Basin. Exploring the complex relationships between habitat factors, community structure, and specific leaf area, combined with the allometric growth relationship between leaf dry weight and leaf area in non-tree plants, aims to reveal how leaf adaptive strategies of non-tree plant communities change due to community structure in alpine environments. This research aims to provide scientific basis for predicting vegetation degradation and high-quality management of natural vegetation in alpine karst desertification areas.

Materials and methods

Study area

The study area is located in the Yunnan-Kweichow Plateau, within the Jinsha River Basin. The coordinates were within the range of 26.503634°N − 26.789644°N and 100.107873°E-100.22724139°E (Fig. 1). This area is an alpine desertification zone, with an altitude varying from 2220 m to 3240 m (Fig. 1). The rocks are composed of dolomite and limestone, and the rock exposure rate can exceed 50% (Fig. 1). Study area pertains to a low - latitude plateau monsoon climate, which is characterized by a frequent alternation between dry and wet periods24. At different altitudes, the annual average temperature range is 9.2–14.6℃, and the annual average precipitation range is 763.2–790.8 mm (Fig. 1)25. Precipitation and high temperatures are concentrated from June to October, exhibiting a concurrent pattern of rainfall and heat. The vegetation is composed of coniferous or small-leaved plants (Table S1). Soil type is brown soil, and the physical properties can be found in Table S2.

Fig. 1
figure 1

Location of investigated plots, annual average precipitation and annual average temperature at different altitudes.

Full size image

Field investigation

Selection of vegetation plots

The study area consists of three vegetation types: grassland, Pinus yunnanensis Franch. forest, and Quercus rehderiana Hand.-Mazz. forest (Fig. 1). The grassland area has the highest proportion, whereas the distributions of Pinus yunnanensis Franch. forest and Quercus rehderiana Hand.-Mazz. forest were sparse. Ultimately, the number of plots was determined based on the observable community quantities of different vegetation types. In July 2024, 12 grassland plots, 8 Pinus yunnanensis Franch. forest plots, and 4 Quercus rehderiana Hand.-Mazz. forest plots were selected for investigation.

Investigation of vegetation plots

In each plot, three 20 × 20 m quadrats were selected for tree investigation (Fig. 2). First, measure the fundamental characteristics of the vegetation communities. The diameter at breast height (DBH, using a steel girth ruler (GNJZ-680)) and height of each tree were measured (using dendrometer (CGQ-1)), and the number of trees was calculated (Table S3) (Fig. 2). In each 20 × 20 m quadrat, three 5 × 5 m quadrats were selected for shrub and herb investigation (Fig. 3). Their height was measured with a 0.01 m-precision ruler and their quantity was recorded (Table S3) (Fig. 2). The coverage of non-tree vegetation was also recorded. Coverage refers to the ratio of the horizontal projection area of vegetation to the occupied area26. In grassland plots without trees, the investigated way was consistent with that used for understory vegetation in tree-containing plots. The numbers of all shrubs and herbaceous plants in the quadrat were recorded, and the species diversity is calculated using the Simpson index27.

Fig. 2
figure 2

Field investigation design and sampling methods.

Full size image

Next, measure the fundamental habitat characteristics. The altitude and slope direction of each quadrat were recorded by GPS (WS-009) and slope measuring instrument (MNom) (Fig. 2). The rock bareness degree and soil thickness were measured (Table S2). The rock bareness degree is the ratio of the exposed rock area to the quadrat area (Fig. 2). The soil thickness is the vertical distance from the soil surface to the bedrock. 9 points were selected in a 5 × 5 m quadrat to measure the soil thickness and calculate the average soil thickness (Fig. 2) (Table S2).

Plant sample collection and determination

The natural vegetation in study area is not permitted to be destructively sampled, so we referred to the “A handbook of protocols for standardized and easy measurement of plant functional traits worldwide” manual for the collection of plant samples28. For each non-tree plant, we selected 5 individuals to collect leaf samples. For each individual, 5 healthy leaves were collected (Fig. 2). These were carefully preserved and brought back indoors. Each leaf was scanned using the Epson V19 and the leaf area was measured using AutoCAD2012 (Autodesk)26. Additionally, the leaves were placed in an oven and dried at 80 ℃ for 48 h, and the leaf dry weight was measured. Finally, the specific leaf area was calculated, which is equal to the leaf area divided by the leaf dry weight18.

Soil sample collection and determination

In each 5 × 5 m quadrat, 5 points were randomly selected to collect soil samples from the 0–20 cm soil layer (Fig. 2). One soil sample was collected from each point using a ring knife to measure bulk density and saturation water content26. Submerging the perforated side of the cutting ring in shallow water, allowing the soil to spontaneously absorb water for 12 h, and then measuring the weight. The soil - filled cutting ring was dried in an oven at 105 °C for 48 h29. Then, the dry weight of the soil was recorded and the soil bulk density was calculated by dividing the soil dry weight by the volume of the cutting ring (100 cm³)26. The saturated water content is equal to the water-absorption weight divided by the dry weight of the soil26. An additional 1.0 kg soil sample was collected from each point. After air drying, screen it using a 0.15 mm sieve. The potassium dichroite oxidation method was used to determine soil organic carbon30. After catalytic digestion of the soil, the Discrete Auto Analyzer (smartchem450) was used to determine soil total nitrogen31. Use a pH meter (PHS-3 C) to measure soil pH.

Data analysis

Data organization and differential analysis

Data normality has been verified by Kolmogorov Smirnov test. The significance level for all analyses was α = 0.05. The T - test was used to analyze the differences in habitat factors, community structure factors, and specific leaf area of different vegetation communities. The two steps were completed in SPSS 26 (IBM).

Allometric relationship analysis

The allometric relationship theory indicates that the growth of various parts of organisms follows certain quantitative laws and is regulated by factors such as physiology and morphology33. Analyzing the allometric relationship between leaf area and leaf dry weight can reveal the trade-off strategies in leaf growth, help understand how plants allocate resources to adapt to the environment, and the intrinsic driving mechanisms of changes in specific leaf area32. The allometric relationship between leaf area and leaf dry weight was analyzed using the allometric growth Eq. 34. If the slope (allometric index) between two traits differs significantly from 1.0, it indicates that they have an allometric relationship. Use the smatr package in R 4.3.3 for allometric relationship analysis34.

log y = b log x + log a

where y are leaf area of each species, and x are leaf dry weight of each species, b is the allometric exponent whose sign reflects the direction of change and magnitude reflects the rate of change, and a is the y-intercept.

Calculation of specific leaf area at community level

To explore the influence of habitat factors and community structure factors on specific leaf area, it is necessary to extend the specific leaf area from the species level to the community level. The classical Community weighted means algorithm was used to obtain the specific leaf area values at the community level in each plot35.

$$T_\text{cwm}= \sum_{i=1}^{n} P_{i} T_{i}$$

TCWM is the community trait value, Pi is the relative abundance, Ti is the trait value of the i​th species in the community, and n​ is the number of species in the community.

Path analysis

The relationships among habitat factors, community structure factors, and specific leaf area are complex. Habitat factors may directly affect specific leaf area, or indirectly influence specific leaf area by affecting the community structure. To gain a deep understanding of the relationships among these factors, path analysis was used to construct a complex relationship network among various factors. In the analysis, the specific leaf area was designated as the dependent variable, and the habitat factors and community structure factors were designated as the independent variables. During the analysis process, if the effect value is greater than 1.0, it indicates a significant collinearity between two factors. These collinear factors were screened based on the ecological significance and the model’s goodness of fit, aiming to clearly reveal the ecological mechanism with a simple model. Path analysis was completed in SmartPLS 4 36.

Results

Leaf trait characteristics of dominant non-tree species in different environments

The specific leaf area of the same non-tree species declined as the altitude rises (Table 1). In all plots, the growth rate of leaf area was significantly faster than that of leaf dry weight could not be found (Fig. 3). The allometric relationships between leaf traits of the dominant non-tree species in the same plot were consistent (Fig. 3). Above an altitude of 3018 m, all leaves of non-tree plants invested in dry weight rather than in area (Fig. 3). When the average soil total nitrogen content exceeded 3.53 mg·g− 1, the leaves invest resources in both dry weight and area. (Fig. 3).

Table 1 Specific leaf area of dominant non-tree species.
Full size table
Fig. 3
figure 3

Allometric relationship between leaf area and leaf dry weight of different species. Data with the same color indicated the same sample plot (P < 0.05).

Full size image

Vegetation structure, rock, and soil nutrient characteristics in different plots

The rock bareness degree in grasslands was significantly higher than that in forests (Fig. 4a). The trend of non-tree vegetation coverage in different vegetation types was opposite to that of rock bareness degree, but was similar to soil total nitrogen (Fig. 4a, c, d). Although rock exposure limited plant extension, an increase in soil nitrogen enabled plants to fully utilize the limited space. In grassland, the trend of Simpson index was opposite to that of soil total nitrogen (Fig. 4b).

Fig. 4
figure 4

Vegetation structure of non-tree plants characteristics, rock bareness degree, and soil total nitrogen. Different lowercase letters indicate a significant difference in each indicator between the two different plots (P<0.05).  

Full size image

Path analysis results of different vegetation types

Altitude and soil total nitrogen were key factors influencing species diversity and vegetation coverage (Fig. 5). An increase in altitude may increase the environmental stress for vegetation extension and leaf development. The impact of altitude on species diversity, vegetation coverage, or specific leaf area was similar in different vegetation types (Fig. 5). An increase in altitude led to an increase in species diversity and a decrease in vegetation coverage (Fig. 5).

Fig. 5
figure 5

Path analysis among habitat factors, non-tree vegetation community structure, and specific leaf area. “*” indicate P < 0.05. “**” indicate P < 0.01. ALT (Altitude), SD (Species diversity), VC (Vegetation coverage), STN (Soil total nitrogen), SLA (Specific leaf area). In the path diagram, the standardized beta coefficient represented the total effect. The table below the diagram presented the direct and indirect effects among various factors within different vegetation types.

Full size image

An increase in soil total nitrogen may remove the limitations on vegetation extension, but it did not necessarily have a positive effect on species diversity. The impact of soil total nitrogen on vegetation coverage was positive (Fig. 5). In grasslands, an increase in soil total nitrogen led to a decrease in species diversity (Fig. 5). However, in forests, an increase in soil total nitrogen led to an increase in the species diversity of non-tree plants (Fig. 5).

The increase in vegetation coverage intensified competition among plants, resulting in a negative relationship with specific leaf area (Fig. 5). In grasslands, the impact of species diversity on specific leaf area was negative, whereas in forests, it was positive (Fig. 5).

Discussion

Variation of specific leaf area of non-tree plants driven by community structure

In the face of environmental stress, plants need to reallocate energy to resist stress rather than for growth and reproduction37. Therefore, areas with low vegetation cover exhibited high species diversity, and this phenomenon has also been observed in the grasslands of the Qinghai-Tibet Plateau38. However, in grasslands, non-tree plants have almost filled the space of non-exposed rocks. Due to the lack of resources, environmental stress increases, and an increase in both vegetation cover and species diversity made the environment unable to support non-tree plants with a high specific leaf area. These non-tree plants need to enhance their resistance and lifespan with a low specific leaf area.

In forests, high vegetation cover limited the specific leaf area increase of non-tree plants. However, differently from grasslands, forests have a stronger ability to conserve water and soil, and a low rock exposure rate provides development space for non-tree plants39. An increase in the niche dimension under low coverage not only increased species diversity but also enabled the environment to support a high specific leaf area. This is a phenomenon that has not been clarified in other alpine areas yet, because the unique climate and soil conditions in this alpine area allow the survival of tree forests40 (Song 2021).

Leaf adaptation strategy of non-tree plants implies vegetation degradation risk

The environmental stress due to the increase in altitude made it difficult for relatively large non-tree plants to survive41. However, it provided more living space for a variety of small plants, resulting in high species diversity in areas with low vegetation cover. From the information on the investigated species and the relationship between vegetation cover and species diversity, this increase in diversity may not be a positive phenomenon. Because the increase in diversity is accompanied by the succession of the shrub-grass structure towards a single herbaceous structure. Especially in areas above 3000 m, shrubs were no longer seen in grasslands and Quercus rehderiana Hand. -Mazz. forests. The species of shrubs in Pinus yunnanensis Franch. forests were gradually becoming homogeneous, and the shrubs of the Ericaceae family have become dominant. However, the vegetation community with homogeneous species shows poor resistance to environmental stress as the altitude increases42. Therefore, each species no longer allocated resources to the extension of leaf area. Instead, resources were invested in increasing biomass for coping with environmental stress. This was an important reason for the negative correlation between altitude and specific leaf area, indicating that non-tree plants were under great survival pressure. This phenomenon has been observed in other alpine areas43. However, our study further explained that this change was due to the allometric relationship of leaf traits (Fig. 6).

Fig. 6
figure 6

The key factors drive alteration in non-tree plant leaf adaptation strategy.

Full size image

The increase in soil total nitrogen promoted the resource allocation that took into account both leaf area and leaf dry weight, but when the nitrogen content exceeded 3.53 mg·g− 1, the number of dominant species decreased. This indicated that although the increase in nitrogen alleviated nitrogen limitation, it reduced the niche dimension11. Dominant species obtained more nutrients, reducing the pressure of underground competition. Instead, they concentrated resources on extending leaf area to compete for above-ground resources, and enhance their own growth and reproduction, leading to an increase in vegetation cover. Eventually, the types of dominant species become homogeneous. Despite the specific leaf areas of dominant species and some auxiliary species were similar, differences in the approaches of leaf resource allocation may lead to different adaptation strategies. The reduction in the types of dominant species is detrimental to the stability of plant communities. It may indicate vegetation degradation44. In grasslands, due to the high degree of rock bareness degree, this phenomenon has occurred earlier compared to forests. Although the increase in nitrogen alleviated the environmental stress on some species, it also exacerbated the environmental stress on auxiliary species. Other research has also been found that nitrogen has an important impact on vegetation cover in grasslands of other alpine areas, but its impact on diversity is not significant45. Perhaps compared with this study, there were more dominant species, and the ultimate dominant species among various species has not yet been determined. Therefore, in alpine regions, simply increasing the soil nitrogen content may instead pose greater challenges to vegetation restoration efforts (Fig. 6).

Conclusions

In alpine rocky desertification areas, merely observing the changes in the specific leaf area and species diversity of non-tree plants will overlook the degradation risk of non-tree vegetation communities, especially when specific leaf area is difficult to clarify the plant adaption strategy under environmental stress. Because there are significant differences in leaf resource allocation strategies among species with similar specific leaf areas. Moreover, improving the development of non-tree plants through nitrogen addition led to the leaf resources allocation of dominant species to both leaf dry weight and leaf area, making their above-ground competitiveness stronger. Eventually, the species of the vegetation community were succession towards few dominant species, increasing the risk of vegetation degradation. Caution should be exercised when implementing nitrogen addition for vegetation restoration in alpine areas.