Adaptation analysis of two Asteraceae invasive plants in Lhasa, Tibet

Seed collection

The B. pilosa seeds used in this study were collected from five representative regions across different geographic locations and climatic zones in China, with the aim of assessing the germination capacity and colonization potential of this species in the high-altitude environment of the Qinghai–Tibet Plateau.

In March 2023, we first conducted an outdoor planting experiment at the experimental field of Tibet University (hereafter referred to as the “experimental field”), using seeds previously collected from Kunming, Yunnan Province (designated as KM). This preliminary trial aimed to determine whether B. pilosa could complete its full life cycle—germination, growth, flowering, and seed maturation—under the natural climatic conditions of Lhasa. The plants successfully matured and produced seeds by mid-October of the same year. These locally matured seeds were subsequently collected and labeled as LS. This result demonstrated that B. pilosa possesses the potential to establish and reproduce in the Lhasa region.

Based on this finding, we further collected seeds from four additional regions to compare the germination performance and early growth responses among different geographic sources. Detailed information on the collection sites of both plants is shown in Fig. 1. Between July and October 2023, naturally matured B. pilosa seeds were collected from the following sites: Kunming (KM, the same population as the one used for LS production; elevation ∼1,891 m, subtropical plateau monsoon climate), Mianyang (MY, ∼470 m, subtropical humid climate), Qiandongnan Miao and Dong Autonomous Prefecture (QD, ∼800 m, subtropical monsoon humid climate), and Nyingchi (LZ, ∼3,000 m, warm-humid climate of southeastern Tibet). Among these, MY and QD represent low-elevation regions characterized by warm and humid conditions, whereas KM and LZ correspond to mid- and high-elevation zones located in southwestern China and on the southeastern margin of the Qinghai–Tibet Plateau, respectively, both exhibiting typical monsoonal or warm-humid highland climates. In particular, Nyingchi experiences abundant annual precipitation and has a significantly milder and wetter climate than Lhasa. In contrast, Lhasa is located in the central part of the Qinghai–Tibet Plateau at an elevation of approximately 3,650 m. It has a typical cold semi-arid climate, with cold and dry winters and springs, a short summer rainfall season, intense solar radiation, and high ultraviolet exposure, resulting in harsh environmental conditions overall.

All seeds of T. minuta used in this study were collected from Nyingchi (designated as YJ). Nyingchi was selected as the seed source for both species due to its geographic proximity to Lhasa and because both B. pilosa and T. minuta have exhibited signs of natural establishment in the area. This facilitates evaluation of their potential adaptation and upward expansion into higher elevations. All collected seeds were stored under dry room-temperature conditions in the Environmental Laboratory of Tibet University until mid-November 2023 and were subsequently used in an overwintering field experiment conducted in the experimental plots.

Overwintering field experiment of two invasive plant species

To evaluate the overwintering survival and spring emergence performance of B. pilosa and T. minuta under natural climatic conditions in Lhasa, an outdoor common garden experiment was established in the experimental field of Tibet University in mid-November 2023. The experimental area is flat, with uniform soil type, ample sunlight, and no shading from buildings or tall vegetation, providing suitable conditions for a field-based ecological study.

Plot preparation and environmental control

Prior to sowing, the field was plowed and loosened using a small rotary tiller to reduce soil compaction and minimize differences in physical and chemical soil properties among treatment zones. Three parallel sowing strips were established to represent three burial depth treatments: 0 cm (surface sowing), one cm, and three cm.

Visible plant residues on the soil surface were manually removed before sowing, and no herbicides were applied in order to maintain the natural soil conditions. For the 0 cm treatment, shallow furrows were created to provide minimal physical barriers to prevent seed displacement by strong winter winds, without affecting soil moisture or water retention. For the one cm and three cm treatments, shallow trenches were dug to the respective depths, and seeds were sown and then evenly covered with soil to ensure consistency and reproducibility across treatments.

Experimental design and treatment setup

A completely randomized block design (CRBD) was adopted, with three sowing depth treatments (0 cm, one cm, and three cm). Each treatment included 10 replicates, with each replicate consisting of one seed row containing 50 seeds. Thus, a total of 500 seeds were sown per seed source per depth treatment (50 seeds ×10 replicates), ensuring sufficient statistical power and reproducibility.

Five seed sources of B. pilosa (KM, MY, QD, LZ, and LS) and one seed source of T. minuta (YJ, from Nyingchi) were included in the experiment. Each seed source was sown in a separate block to avoid mixing, and all plots shared the same soil and climatic conditions to minimize environmental variation. Planting rows were labeled with source identifiers, burial depth, and replicate numbers to ensure accurate tracking during subsequent monitoring.

Sowing and monitoring

After sowing, no artificial intervention was applied; all seeds overwintered under natural environmental conditions. In the following spring (March to April 2024), systematic surveys were conducted to record seedling emergence rates and plant height across all treatments and replicates. These measurements were used to assess the survival potential and early establishment ability of the two invasive species in the harsh high-altitude environment of Lhasa.

Data collection and analysis

In the experimental field, the first observation of B. pilosa cotyledon emergence occurred on May 15, 2024. However, due to the presence of residual seeds from Cosmos bipinnatus—previously cultivated in the same field—numerous volunteer seedlings emerged in early spring. These seedlings closely resembled those of B. pilosa and T. minuta during early developmental stages, making it difficult to distinguish species during initial emergence. To ensure the accuracy of species identification, formal germination data collection was delayed until approximately three weeks after sowing. At that time, seedlings had developed their first true leaves, allowing us to reliably distinguish species based on leaf morphology. From this point onward, systematic data collection was conducted. Weekly surveys were performed throughout the growing season to record the number of newly emerged seedlings within each replicate plot. To minimize potential interference and ensure data reliability, all non-experimental seedlings and weedy vegetation were manually removed during the survey period, particularly those occurring within the planting rows. If no new emergence was observed for three consecutive weeks, that date was defined as the germination endpoint for the respective treatment.

To prevent unintended seed dispersal from reproductive individuals, the experiment was terminated on August 15, 2024, when some plants had begun to develop inflorescences and seeds were approaching maturity. On that date, all experimental plants were removed from the field. Prior to removal, plant height measurements were conducted to assess early growth performance. Figure 2 shows the thriving growth of B. pilosa and T. minuta in the experimental field. For each seed source, 30 vigorous individuals were randomly selected from each burial depth treatment, totaling 90 plants per seed source.

All data processing and statistical analyses were conducted in Python v3.11.4 (The original germination count and plant height data can be found in Table S1). Germination rate was calculated as: (number of emerged seedlings / total number of sown seeds) ×100%. Plant height data were visualized using boxplots to illustrate distribution patterns and outliers. One-way ANOVA was used to test the effects of burial depth and seed source on both germination rate and plant height. When assumptions of normality and homoscedasticity were met, Tukey’s HSD test was applied for post hoc multiple comparisons. Statistical significance was set at P < 0.05.

Microclimate data collection and processing

To explore the relationship between climatic factors and the germination and early growth of B. pilosa and T. minuta under natural conditions in Lhasa, environmental data were obtained from an SP200 automatic weather station installed adjacent to the experimental field. The station has been in continuous operation since mid-November 2021, providing stable long-term meteorological observations.

For this study, we selected weather records from January 1, 2022, to August 15, 2024, encompassing multiple full annual cycles prior to and during the experiment. This time span offers valuable context for characterizing the climatic background and variability during the experimental period. The weather station recorded data automatically at 15-minute intervals.

To improve data processing efficiency and focus on key climatic variables, we used the Python programming environment (v3.11.4) to clean and process the raw dataset. Daily mean air temperature and daily cumulative precipitation were extracted and used to construct a continuous time series depicting temperature and precipitation trends in the Lhasa area during the study period (see ).

In subsequent analyses, temperature and precipitation data were treated as potential environmental variables and correlated with phenotypic traits such as germination rate and seedling height under different seed source and burial depth treatments. This allowed us to assess the influence of major climatic drivers on the colonization potential of invasive species in high-altitude ecosystems on the Qinghai–Tibet Plateau.

Effects of burial depth on seed germination of two invasive Asteraceae species

Figure 3 illustrates the number and germination rates of B. pilosa and T. minuta under natural climatic conditions in Lhasa, across three burial depth treatments (0 cm, one cm, and three cm) and five seed sources. The results indicate that both seed provenance and burial depth significantly influenced the germination performance of the two species.

For B. pilosa, seeds from the original Kunming source (KM) exhibited the highest overall germination, with 301 emerged seedlings and a germination rate of 26.4%. The rates for other sources were as follows: Qiandongnan (QD) at 23.2%, Nyingchi (LZ) at 22.4%, Mianyang (MY) at 17.4%, and Lhasa (LS) at only 11.5%. Notably, the LS seeds were collected from KM plants that had completed one full growing season in Lhasa. The substantially lower germination rate of the LS source may reflect negative effects of the high-altitude environment on seed viability or maternal plant fitness.

Regarding burial depth, all B. pilosa seed sources except LZ showed the highest germination rates under surface sowing (0 cm). For instance, the KM source showed significant differences across the three depths, with germination rates of 26.4% (0 cm), 18.2% (one cm), and 15.6% (three cm) (P < 0.05). As B. pilosa seeds lack dormancy, their germination is likely influenced by external physical factors such as temperature, moisture, and light. Surface sowing can provide enhanced light exposure and better gas exchange, thus facilitating seed germination. These results suggest that B. pilosa has high germination potential in the surface soil layer, which may contribute to its successful colonization in novel habitats.

In contrast, T. minuta exhibited much lower germination potential overall. The average germination rates across the three burial depths were 4.0% (0 cm), 5.8% (one cm), and 2.2% (three cm), also showing a decreasing trend with increasing depth. Despite the low emergence, T. minuta is a typical r-selected species with high fecundity and diverse dispersal pathways, including wind, water, and animal-mediated transport (Wang et al., 2023). These traits suggest that even with low natural germination rates, the species may still establish and expand rapidly under favorable conditions.

Effects of burial depth on plant height in two invasive species

Plant height is a key phenotypic indicator of early growth performance and is commonly used in invasion biology to assess a species’ colonization potential and competitive advantage (Holman et al., 2016). In this study, standardized measurements of plant height were conducted on the final day of the experiment for both B. pilosa and T. minuta under three burial depth treatments (0 cm, one cm, and three cm) (Fig. 4).

For B. pilosa, plant height varied significantly among seed sources. Seeds originating from KM, LS, and LZ—all from high-elevation regions—exhibited greater plant height under surface sowing (0 cm), indicating stronger early growth vigor. In contrast, individuals from the low-elevation sources MY and QD were generally shorter, with minimal height variation across the three burial depths. This pattern suggests that seeds from higher elevations may be better adapted to the cold and arid conditions of the Lhasa Plateau, thus exhibiting more robust growth under similar environmental stresses.

Compared to B. pilosa, T. minuta exhibited overall taller growth across all burial depths, with consistently higher median values. In particular, some individuals exceeded 70 cm in height under the three cm burial treatment. This marked growth advantage likely reflects inherent interspecific differences in growth strategy and morphological development, rather than directly indicating superior adaptation to the high-altitude environment.

Relationship between germination and climatic factors in two invasive species

The success of plant invasion is often closely linked to local climate conditions and key environmental factors (Ni et al., 2021). To evaluate how temperature and precipitation influence the germination of B. pilosa and T. minuta, we analyzed daily climate data—specifically mean temperature and cumulative precipitation—collected by an automatic SP200 weather station located adjacent to the experimental plots.

Figure 5 presents the daily variation in air temperature and precipitation in Lhasa over the past three years (2022-2024). The data indicate that the primary growing season in Lhasa spans from May to October, during which both temperature and precipitation increase substantially, providing critical conditions for seed germination and seedling establishment (Fig. 5A). In 2024, a sustained rainfall event occurred in early May, with a total of 18 mm of precipitation recorded between May 4 and May 10 (see Table S1), effectively increasing soil moisture levels. This was followed by a rapid rise in daily average temperature, which stabilized above 15 °C by mid-May, creating favorable thermal conditions for seed germination.

Correspondingly, the emergence of seedlings was observed of both species around May 15, suggesting that this period constituted the primary germination window (Fig. 5B). Notably, both species demonstrated considerable tolerance to environmental stress. Despite the harsh winter and spring conditions in Lhasa—characterized by temperatures consistently below 0 °C, minimal precipitation, and intense ultraviolet radiation—experimental observations and climate data confirmed that seeds of both species remained viable throughout the winter and were able to germinate rapidly once favorable conditions returned in the spring. These findings suggest that under the ongoing trend of warming and increased precipitation on the Qinghai–Tibet Plateau, the colonization potential of invasive species in this region may continue to rise, thereby increasing their risk of expansion and establishment.

Effects of seed provenance and maternal environment on germination

This study revealed significant differences in seed germination performance among B. pilosa provenances. Seeds originally sourced from Kunming (KM) exhibited the highest germination count, reaching a total of 301 seedlings across the three burial depth treatments, markedly higher than that of the LS provenance (115 seedlings), which originated from KM seeds that had undergone one full growing season and natural maturation under Lhasa conditions. The substantial decline in germination rate of the LS provenance suggests a profound influence of the maternal environment on seed quality. This phenomenon is attributable to maternal effects, where the ecological conditions experienced by the mother plant—such as temperature, radiation, and soil nutrient availability—can significantly affect the developmental quality and vigor of the offspring seeds, subsequently influencing their germination behavior and early growth (Nguyen et al., 2021).

In the high-altitude environment of Lhasa, environmental stressors such as low temperatures, high ultraviolet radiation, and reduced oxygen availability likely impair the resource accumulation capacity of mother plants, resulting in reduced seed vigor. In contrast, the mild and humid climate of Kunming is more favorable for maternal development and resource allocation, thus producing seeds with higher viability and germination potential.

Moreover, seeds collected from other external provenances, such as QD, LZ, and MY, also exhibited higher germination rates compared to the LS provenance. This finding suggests that seed populations in the initial stages of invasion may possess greater ecological adaptability. Previous studies have reported that increasing altitude imposes significant constraints on seed germination and seedling survival due to cold temperatures, drought, and intense radiation (Zait, Konsens & Schwartz, 2020). Our results further support the view that high-altitude environments exert strong environmental filtering effects on the establishment of invasive plant populations.

Effects of burial depth on seed germination and plant growth

Burial depth had a significant effect on both germination rate and plant height in B. pilosa and T. minuta. For B. pilosa, seeds from all sources consistently showed the highest germination rates under the 0 cm surface sowing treatment, with a gradual decline observed at one cm and three cm depths. For instance, the germination rates of seeds from the KM provenance were 26.4%, 18.2%, and 15.6% at 0 cm, one cm, and three cm depths, respectively. This trend is consistent with patterns reported in other Asteraceae species. For example, Artemisia ordosica exhibited a germination rate of 87.6% at 0.5 cm depth, which dropped sharply to 9.7% at one cm and was nearly zero at three cm (Zheng et al., 2005). Similarly, Xanthium spinosum showed significantly higher emergence under shallow burial, with maximal rates at 1–3 cm and near-zero germination below five cm (Tao et al., 2022). These findings suggest that shallow burial enhances soil warming and diurnal temperature fluctuation near the surface, which promotes germination by accelerating thermal cues, rather than differences in oxygen availability. The surface-sown advantage observed in B. pilosa thus reflects a broadly shared trait among Asteraceae species adapted to open or disturbed habitats.

In terms of early seedling growth, B. pilosa seeds from mid- to high-elevation provenances (KM, LS, and LZ) exhibited greater plant height under surface sowing compared to those from low-elevation sources (MY and QD), particularly at 0 cm depth. This pattern may reflect a form of pre-adaptation, whereby seeds originating from colder, drier environments possess enhanced vigor under the harsh climatic conditions typical of the Lhasa Plateau.

Although T. minuta exhibited generally lower germination rates across treatments, it demonstrated a remarkable growth advantage during the early development stage. In particular, individual plants under the three cm burial treatment reached heights exceeding 70 cm. This rapid growth and structural establishment suggest that once germination occurs under favorable conditions, T. minuta could quickly attain a competitive advantage, highlighting its ecological impact potential despite initial limitations during the germination phase.

Colonization potential of two invasive species in Lhasa and discrepancies with model predictions

Although the Qinghai–Tibet Plateau is widely regarded as an “ecological barrier” to biological invasions due to its high elevation (Chu et al., 2024), intense ultraviolet radiation, and arid climate, our findings demonstrate that both B. pilosa and T. minuta are capable of completing germination and early establishment under natural conditions in Lhasa. Meteorological data revealed a sharp temperature increase and concentrated rainfall beginning in mid-May, which provided a critical “climatic window” for seed germination in both species, corroborating the observed synchronized seedling emergence in the field.

However, current ecological niche models (e.g., MaxEnt), which are widely used to predict species’ potential distributions, did not identify the Lhasa region as a highly suitable area for either species (Qi et al., 2022; Yang et al., 2023). This discrepancy may be attributed to limitations in input variables, spatial resolution, and underlying assumptions of the models. Previous studies have shown that invasive plants often overcome original environmental constraints through niche shifts or intraspecific variation (Petitpierre et al., 2012). Additionally, microhabitat effects should not be overlooked. Localized moist environments along farmland edges, near buildings, or beside roads in Lhasa may provide more favorable conditions than the macroclimatic averages recorded by meteorological stations. These findings underscore the necessity for cautious interpretation of species distribution model outputs, particularly in dynamic and topographically complex high-altitude ecosystems.

Invasion risk under plateau warming trends and management implications

Against the backdrop of global climate change, the Qinghai–Tibet Plateau is experiencing a marked trend of warming and increased precipitation, characterized by rising annual mean temperatures, intensified summer rainfall, and an extended growing season (He et al., 2023; Yu et al., 2024). Our findings demonstrate that seeds of both B. pilosa and T. minuta remain viable throughout the winter and spring—despite prolonged sub-zero temperatures and minimal precipitation—and are able to germinate rapidly when favorable conditions return the following year. This highlights their strong environmental tolerance and overwintering capacity. In particular, T. minuta, as a typical r-selected species, possesses high reproductive potential and diverse dispersal strategies (Wang et al., 2023), enabling it to exert significant ecological influence once germination occurs. B. pilosa also has the ability to spread over long distances via animals or human-mediated vectors. There is already evidence that ecologically similar species have successfully colonized high-altitude regions; for instance, Ageratina adenophora has spread widely across the Himalayan Mountains (Negi et al., 2023a; Negi et al., 2023b). As environmental barriers in the plateau continue to weaken, the risk of plant invasion in Lhasa and across the broader Qinghai–Tibet Plateau is likely to increase.

In summary, under the overarching trend of global warming, the ecological barriers that once limited biological invasions at high altitudes on the Qinghai–Tibet Plateau are gradually diminishing. The potential distribution range of invasive species is likely to shift toward higher elevations and latitudes. As a densely populated and logistically connected city in the plateau region, Lhasa faces heightened risks of exotic plant introductions. Once established, such species may pose long-term threats to native ecosystems. Therefore, we recommend strengthening invasive species monitoring and risk assessment in Lhasa and other key areas across the plateau, with particular attention to high-risk zones that lie outside the range predicted by existing ecological niche models. Future modeling efforts should integrate warming scenarios, species-specific physiological and ecological traits, and microhabitat heterogeneity to improve the accuracy of invasion forecasts and enhance policy responsiveness.

Permission for sample collection

B. pilosa and T. minuta is not listed on the IUCN Red List of Threatened Species, and the collection area is not within a protected zone. Consequently, no specific permissions or licenses were required for sample collection from authorities.