1. Introduction

The Qinghai-Tibet plateau (QTP) is the largest and highest plateau in the world (Zhang et al., 2002), the source of many of Asia’s major rivers (e.g., the Yangtze, Yellow, Mekong, Salween and Brahmaputra rivers) (Sun et al., 2012; Sati, 2020), and an important center for biodiversity that houses an array of high-elevation ecosystems (e.g., Shimono et al., 2010; Sun et al., 2017; Liu et al., 2021; Wang et al., 2021a). Despite its importance, QTP has faced multiple conservation challenges over the past decades, especially from climate changes (e.g., rising temperature, changing precipitation patterns, and an increase in extreme weather) and over-exploitation of resources by human activities (e.g., commercial harvesting and over pasture) (e.g., Law and Salick, 2005; Chen et al., 2011a, 2017b, 2021a; Nolan et al., 2018; Yin et al., 2020; Niu et al., 2021; Zhang et al., 2021). Recent studies have documented these impacts, however, a comprehensive review has yet to synthesize our understanding of how these factors interact to threaten plant diversity in QTP.

QTP plant diversity is highly threatened by temperature changes. QTP has warmed at twice the global average over the past 50 years, resulting in significant environmental changes (Liu and Chen, 2000). This warming is expected to continue at an accelerated pace in the coming decades. At warmer temperatures, species adapted to cool environments at high elevation may be forced to migrate to higher elevations in search of suitable habitats (Shivanna, 2022). This shift to higher elevations has been observed with near-treeline plant species in the Himalayas and QTP, where climatic suitability is predicted to expand into higher elevations by 2050 and 2100, particularly above 3500 m (Lamsal et al., 2017). These shifts in habitat suitability may shrink available habitat for native high-elevation species, in particular endemic and specialized plants (Wershow and DeChaine, 2018). In some cases, these shifts in habitat suitability may result in invasion of some low and middle elevation species into higher elevation ecosystems (Pathak et al., 2019). Invasive species often outcompete native species for resources, and in extreme situations increase the risk of extinction for native plants with narrow ecological niches or keystone species (Wershow and DeChaine, 2018; Kumar Rai and Singh, 2020; Chen et al., 2024a). The loss or decline of keystone species (e.g., cushion plants) in hostile subnival habitats of the QTP can have ripple effects, altering the structure and function of the entire ecosystem (Chen et al., 2024a).

Warmer climates in the QTP have also been shown to alter plant phenology, i.e., leaf-out (reviewed by Piao et al., 2019b) and flowering (Hart et al., 2014). For instance, warmer climates could lengthen the growing season and flowering duration, increasing plant abundance and reproductive success (Chen et al., 2020a). However, changes in leafing or flowering phenology may increase the risk of newly emerged tissues to freezing and thus negatively affect the overall fitness of species (Vitasse et al., 2014; Yang et al., 2020; Jin et al., 2025). Furthermore, alterations in phenology can disrupt interactions between plants and their pollinators (Gérard et al., 2020), affecting both species negatively (Scaven and Raffety, 2013). A warmer climate has also been shown to increase pressure from pests and pathogens (Singh et al., 2023).

Rising temperatures have also been shown to alter the total water budget in the QTP by promoting glacier retreat and permafrost melts, as well as altering precipitation patterns (reviewed by Chen et al., 2019a; Wang et al., 2023b; Bao et al., 2024). For example, in the Hengduan Mountains (HDM), located in the southeastern part of QTP, precipitation has been found to be more variable, with an increase in extreme rainfall events and prolonged dry periods (Ning et al., 2012; Xu et al., 2018). Reduced water availability decreases plant populations that are water-dependent (Beniston and Stoffel, 2013). This impacts the composition and productivity of plant communities, particularly in alpine meadows and grasslands (Yang et al., 2013). Warming on the QTP has also led to permafrost thaws, which cause soil instability and erosion (Yang et al., 2010b; Xie et al., 2017; reviewed by Chen et al., 2019a). Permafrost thaws alter soil structure by decreasing soil nutrient levels. This, in turn, makes it difficult for plants to root and survive (Wang et al., 2000), leading to changes in vegetation cover and soil hydrology (Wang et al., 2012). Permafrost thaws also increase the thickness of the active soil layer, resulting in more liquid soil moisture during freeze–thaw cycles. These freeze–thaw cycles significantly influence soil properties and greenhouse gas emissions (Chen et al., 2020d). For instance, freeze–thaw cycles have been shown to increase nitrous oxide (N₂O) emissions and affect soil microbial activity, which in turn affects vegetation composition and productivity. Together these changes promote shifts in plant species composition, with drought-resistant species becoming more dominant (Martínez-Vilalta and Lloret, 2016).

Anthropogenic activities, such as harvesting of wild plants, have also led to evolutionary changes in natural populations on the QTP. For example, studies have shown that intensive collection and/or overgrazing have resulted in dwarfing of Saussurea laniceps and Rheum nobile, as well as camouflaging of Fritillaria delavayi (Guo et al., 2022; Law and Salick, 2005; Song et al., 2020; Niu et al., 2021). This type of unintentional human selection on plant traits is a powerful force that drives evolutionary changes in natural populations and may incur unforeseen impacts that negatively affect fitness and regeneration of plants.

Here, we provide a comprehensive review integrating decades of researches on how plant diversity on the QTP has responded to climate change and anthropogenic pressure. Our objectives are threefold: (1) to synthesize current knowledge on how warming, hydrological shifts, and human activities (e.g., harvesting, grazing) directly and indirectly affect plant traits, distribution, and ecosystem function (Fig. 1); (2) to identify critical gaps in our understanding, such as the synergistic effects of multiple stressors on endemic species; and (3) to propose a framework for future research and conservation prioritization. Our review thus not only advances fundamental knowledge but also provides a road map to inform policy and practice for mitigating biodiversity loss in the QTP and other high-elevation regions facing similar challenges.

2. Effects of anthropogenic climate change on the growth and reproduction of plants and their interactions with pollinators in QTP 2.1. Plant growth

Plant species face various stresses and may respond differently to climate change depending on the location of the population within the range of the species (Liancourt et al., 2013). In cold and humid regions (e.g., arctic tundra, European Alps, and high Himalayas), warmer temperatures have been shown to drive plant growth (Myers-Smith and Hik, 2018; Filippa et al., 2019; Anderson et al., 2020). However, forest dieback and growth declines remain a global concern, including in the QTP. Many studies have documented how tree growth and recruitment has been altered by climate change on the QTP, with particular attention paid to alpine treelines, which are believed to be important monitors of plant, vegetation, and ecosystem responses to climate change (reviewed by Miehe et al., 2025).

Climate warming impacts treelines by altering plant growth dynamics (Körner, 1998, 2012; Xie et al., 2025). Dendrological studies have indicated that even minute warming has increased radial growth in trees near treelines in North America (e.g., LaMarche et al., 1984; Payette et al., 1985; Cooper, 1986), the South American Andes (e.g., Villalba et al., 1997; Salzer et al., 2009), European Alps (e.g., Nicolussi et al., 1995; Paulsen and Körner, 2001) and the Southeastern QTP (Liang et al., 2009). However, studies on the QTP have shown that drought conditions associated with elevated temperatures have decreased tree recruitment near alpine treelines and that tree (and shrub) growth are less dependent on temperature (Esper et al., 1995; Wang et al., 2018a; Guo et al., 2018; Lu et al., 2022). These findings are consistent with research in the subtropical Andes (Morales et al., 2004). Despite these findings on the effects of climate change on radial growth of trees, we know little about how warmer temperatures affect vertical (height) tree growth. Recent projections suggest that warmer temperatures may lengthen the height-growth duration of Smith fir (Abies georgei var. smithii) on the southeastern part of QTP by 4–22 days by the end of the century, which may enhance primary growth, forest productivity and CO₂ sequestration in the region (Zhang et al., 2022a).

2.2. Plant reproduction

Human-induced climate change has also threatened plant reproductive success by influencing plant reproductive effort (e.g., flower number, nectar volume) and performance (e.g., Song et al., 2020; Zi et al., 2022; Cun et al., 2024), as well as decreasing pollination services (e.g., Eckert et al., 2010; Montero-Castaño and Vilà, 2012; Stanley and Stout, 2013). In QTP, the effects of warming on plant reproductive effort have been well documented. For example, warming in alpine meadows of the QTP has been shown to reduce reproductive allocation in many herbaceous species (Cao, 2018; An, 2021). The plants may have shifted resources to vegetative growth to compete for space and light, as necessitated by increases in species richness and abundance. Similarly, a long-term field experiment on Saussurea nigrescens in alpine meadows of the eastern part of QTP found that artificial warming reduced nectar volume per floret, floret number per capitulum, and capitulum number per plant (Mu et al., 2015). Notably, studies have shown that plant species with different life-history traits respond differently to warming experiments. For example, in an alpine meadow of the QTP, experimentally induced seasonal winter warming significantly reduced seed production in multi-inflorescence species, while single-inflorescence species remained unaffected (Liu et al., 2012). In addition, elevated temperature has been shown to increase the floral nectar sugar concentration in mustard flowers (Singh, 2013). However, our current understanding of the effects of warming on plant reproduction in the QTP are mostly derived from studies on herbaceous species in alpine meadow ecosystems. A greater understanding of the overall effect of warming on plant reproduction requires further research on additional ecosystems and/or woody species.

Pollinators, especially insect pollinators, are crucial partners of plants, yet they are reportedly in widespread decline due to anthropogenic land use changes, agricultural intensification, species invasion, and climate change (Thomann et al., 2013). Research on alpine grasslands located in wetlands in the eastern part of QTP found that overgrazing significantly reduces the taxonomic richness and abundance of the epigenic arthropod community, with the dominant taxa of Diptera and Hymenoptera particularly affected (Wang et al., 2022b). Similarly, field observations in alpine meadows and wetlands of the HDM have suggested that the population size of pollinators for Rheum alexandrae have decreased in sites overgrazed by yaks (Song et al., unpublished data). Overgrazing in the HDM has also been found to substantially reduce seed output of R. alexandrae and R. nobile by reducing pollination services (Song et al., 2020; Song et al., unpublished data).

Human activity has also been shown to affect the interaction between plants and pollinators. For example, the introduction of alien plant species or the expansion of native plant species due to human activities has been shown to cause pollinator deficiency (Richardson, 2011). Competition between native and alien plants for pollinators has been suggested to reduce pollinator visitation rate and thus decrease seed production in native plants (Qiu and Tan, 2018). Furthermore, a recent study found the spillover effect of non-native honey bees (Apis mellifera) enhances visitation fidelity of pollinator species and reconfigures pollination interaction in the alpine meadow of QTP due to an increase in competition between honey bees and native pollinator flies (Wang et al., 2024a). In the QTP, a reduction in pollinator visitation or transition of dominant pollinators would create pollination deficiency for most plant species (Thomann et al., 2013), although some plants may adapt to pollinator declines (Zhang and Li, 2008; Xiong et al., 2013; Xu et al., 2025) by evolving mechanisms that assure reproductive success (e.g., delayed autonomous selfing or prolonged floral longevity). Accordingly, one possible response to climate change on the QTP is that plant species that are less sensitive to pollinator decline may become more dominant in plant communities, changing community composition and ecosystem functions (Liu et al., 2012; Xu et al., 2023).

Studies have observed that warmer temperatures on the QTP have advanced or delayed flowering and fruiting (Mo et al., 2018; Hu et al., 2020). However, most studies have not assessed the impact of changes in the timing of these phenological events on plant reproduction itself. Moreover, the few studies that have done so mainly focused on plant reproductive effort (e.g., flower number and seed output; Liu et al., 2012; Hu et al., 2020). For example, research found that warming can extend the duration of flowering by advancing the date of the first flower and delaying the date of the last flower (Chen et al., 2020a). This change in flowering phenology increases the relative cover and abundance of plants, likely through increases in seed production and overall reproductive success. Changes in plant flowering phenology, however, can also result in phenological mismatch between plants and their pollinators or herbivores (Li et al., 2011; Kudo and Ida, 2013). Although studies have shown that such phenological shifts due to global climate change are particularly pronounced in high-elevation regions (Kudo and Ida, 2013), few studies have focused on the potential phenological mismatch between plants and pollinators, or the subsequent impacts on the fitness of both groups. These mismatches may be particularly serious when either the plant or pollinator species involved is a specialist within their respective pollination system (Memmott et al., 2007; Song et al., 2014) or when the time period during which resources are available (e.g., flowering period for specific pollinators) is relatively short (McKinney et al., 2012; Song et al., 2020, 2022).

3. Plant phenology as a biological indicator of anthropogenic climatic changes in QTP 3.1. Spring greening date

Vegetation growth at high elevations and latitudes is highly sensitive to climatic warming (Lucht et al., 2002). Landscape-scale remote-sensing studies have shown that the average onset of vegetation green-up over the QTP advanced by ca. 4.1 days in responses to 1 ℃ in spring temperature and the green-up date on the QTP advanced by 0.33–0.88 d/10 years during 1980s and 1990s (Yu et al., 2010; Piao et al., 2011). In contrast, from 1999 to 2006, a delay in green-up date was detected (Yu et al., 2010; Piao et al., 2011). Such a reverse in vegetation green-up was correlated with an increase in temperature (0.09 ℃ year⁻¹) during the period 1982–1999, followed by a decrease from 1999 to 2006 (Piao et al., 2011).

Research has shown that the effect of climate change on green-up date in the QTP may be amplified by elevation and nighttime warming. Specifically, previous work has shown that across the QTP the green-up date increased by 0.8 day per 100 m increase in elevation (Piao et al., 2011). Furthermore, using both in situ and satellite observations, studies found a 1 ℃ increase in preseason daily minimum temperature (T_mini) advanced green-up date by 4 days, while an increase in preseason daily maximum temperature (T_max) did not advance green-up date (Shen et al., 2016). Thus, unlike most areas of the Northern hemisphere, where daytime temperature exerts strong controls on spring phenology and vegetation (Piao et al., 2015), on the QTP, nighttime temperature seems to exert greater control over green-up date (Shen et al., 2016). This finding is in contrast to the apparently similar ecosystem of the Arctic, where vegetation is mainly controlled by T_max (Hinzman et al., 2005; Parmesan, 2007; Shen et al., 2016). However, if T_mini continues to increase faster than the daily maximum temperature (T_max), as it has over the past few decades (Liu et al., 2006), nighttime warming will continue to strongly impact ecosystems in the QTP.

Previous studies on alpine and arctic regions have found that warming changes the reproductive phenology of plants, e.g., budburst, flowering and fruiting (e.g., Walker et al., 1999; Dunne et al., 2003; Aerts et al., 2006; Cleland et al., 2006). However, the effect of warming on plant reproductive phenology may be more complex on the QTP, which has heterogenous habitats and high levels of plant diversity. For example, a field experiment on the QTP found that the effect of artificial warming and snow addition on reproductive phenology varied with rooting depth and life-history traits (early vs. late flowering) among species (Dorji et al., 2013). In this experiment, warming delayed reproductive phenology in shallow-rooted, early flowering plants (e.g., Kobresia pygmaea), while snow addition was found to mitigate the negative effect of warming on K. pygmaea. The researchers reasoned that the effects of warming were caused by water stress and that snow addition can compensate for low levels of upper-soil moisture. These findings suggest that changes in soil environment (e.g., the timing of soil thaw and soil water availability) should be investigated as an important environmental trigger for either leafing or flowering phenology of plants in some arid, semi-arid or monsoon dominated ecosystems in the QTP (Wang et al., 2008, 2014; Dorji et al., 2013; Shen et al., 2016; Piao et al., 2019a).

3.2. Flowering date

Flowering phenology commonly occurs after leaf-out, and is consequently not easily identified by satellite images. Moreover, field work and/or experimental studies on flowering time have only begun over the last few decades. Consequently, specimen-based data recordings have increasingly been used to better understand the long-term shifts and variability in flowering phenology at either species or community level (e.g., Primack et al., 2004; Gaira et al., 2011; Hufft et al., 2018). For example, analysis of an herbarium collection of 41 species from the QTP indicate that warmer climates have caused flowering to occur earlier (Li et al., 2013). Similarly, analyses of over 10,000 herbarium specimens of Himalayan Rhododendrons collected since the 1880s indicated that mean flowering date advanced with annual warming (2.27 d earlier per 1 ℃ warming) but was also delayed in concert with fall warming (2.54 d later per 1 ℃ warming) (Hart et al., 2014). The advance in flowering date is likely due to the effect of warming on overwintering bud formation, whereas delays in flowering were likely related to the impact of higher temperatures on the chilling requirement of some species (Hart et al., 2014). An analysis of historic herbarium records (1848–2003) of Aconitum heterophyllum collected in the Indian Himalaya region adjacent to the QTP also found that over the past 100 years plants have been flowering earlier (17–25 days) (Gaira et al., 2011).

Research has suggested that early flowering can help plants effectively absorb available soil nutrients during the growing season (Li et al., 2013). However, advances in flowering date may also expose these vulnerable organs to the risk of freezing damage in the early of growing season (Inouye, 2008; Wipf et al., 2009; Kirchhof et al., 2025) (Fig. 2). Moreover, changes in flowering date may affect reproductive success by causing a mismatch between pollinator availability and the timing of flowering (Hegland et al., 2009; McKinney et al., 2012). Research is needed to better understand the implications of changes in flowering phenology and consequent reproduction success on plant demography, community interaction and trophic dynamics of alpine vegetation of the QTP.

4. Changes in plant distribution in QTP due to anthropogenic climate change

Warmer temperatures may shift the distribution of plants adapted to the cold (e.g., Lenoir et al., 2008; Wiens, 2016; Liang et al., 2018). Mountain plants frequently move upward or poleward to locate conditions similar to their original habitats (e.g., Holt, 1990; Hickling et al., 2006; Lenoir et al., 2008; Steinbauer et al., 2018; He et al., 2020). However, such shifts can expose native plants to new biotic and abiotic challenges, potentially leading to unpredictable ecological consequences (Harley, 2011; Kharouba and Vellend, 2015; Cowles et al., 2016). For instance, newly arriving species may threaten community stability and local endemism (Harley, 2011; Kharouba and Vellend, 2015), decrease species abundance (Calinger, 2015), or reduce adaptive variation and genetic diversity by altering interspecific interactions (Callaway, 2007; Pauls et al., 2013). Additionally, mating systems within communities could be disrupted and hence affect reproductive success (Etterson and Mazer, 2016). Continuous warming may result in a lack of suitable habitats at higher elevations, increasing the likelihood of local extinctions (Thomas et al., 2004; Malcolm et al., 2006; Urban, 2015; Freeman et al., 2018).

4.1. Changes in species distribution on the QTP

Plant species in the QTP and adjacent Himalayas have been found to predominantly migrate upward and/or poleward in response to a warming climate (e.g., Liang et al., 2018; Li et al., 2019; Hu et al., 2019, 2021; He et al., 2020; Liao et al., 2021; Wang et al., 2021b; Chen et al., 2022; Zhao et al., 2021), consistent with observation from other mountainous regions (e.g., Kelly and Goulden, 2008; Lenoir et al., 2008; Gottfried et al., 2012; Elsen and Tingley, 2015; Steinbauer et al., 2018). For example, recent research found that all sampled species (approximately 150) in the Hengduan Mountains (HDM) exhibited upward migration from the last glacial maximum to the present, a finding that their models projected would continue at least until 2050, whereas the mechanisms and rates of these shifts vary between species (Liang et al., 2018).

Climate change poses a serious threat to endemic and endangered species. In the QTP, endemic and endangered species, which are restricted to narrow ecological niches, are particularly disposed to changes in distribution (Wen et al., 2014; Sun et al., 2017). For instance, models have projected that a warmer and drier climate on the QTP will cause Pomatosace filicula, an endemic species, to migrate to higher elevations or cooler, more humid areas. Such migrations may exacerbate habitat degradation and reduce the probability that species survive (Chen et al., 2022). Models have also projected that the distribution of plants in the QTP and surrounding areas (i.e., Incarvillea, Aconitum spicatum, Dactylorhiza hatagirea, Nardostachys jatamansi, Paris polyphylla and Valeriana jatamansi) (Rana et al., 2020, 2021), as well as in the HDM (i.e., Rheum nobile and R. alexandrae) will initially expand northward into suitable habitats, before retreating southward due to climate change (Rana et al., 2022).

4.2. Changes in community composition on the QTP

Climate warming can significantly affect plant communities by altering species composition and interspecific interactions, further impacting vegetation structure and nutrient and energy cycles (e.g., Jackson et al., 2000; Jobbagy and Jackson, 2000; Lenihan et al., 2003; Gottfried et al., 2012). For example, Salick et al. (2019) observed increases in plant species richness, abundance, and diversity on the HDM summits over a relatively short time period (seven years), primarily driven by the upward migration of native plants from lower elevations. Although mountain-top communities are particularly sensitive to these types of changes, their responses to climate change remain largely underexplored (Chen et al., 2011b; Steinbauer et al., 2018; Meng et al., 2019; Salick et al., 2019).

Plant communities on summits are predominantly composed of cushion-shaped species. These pioneer plants are exceptionally adapted to cold, nutrient-poor environments (Körner, 2003; Boucher et al., 2016) and are highly sensitive to climatic changes (Cranston et al., 2015) as well as changes to interspecific interactions (Chen et al., 2024a). As foundation species, cushion plants play a vital role in reorganizing community structures (Badano et al., 2006; Badano and Marquet, 2008), enhancing plant diversity (Cavieres and Badano, 2009; Yang et al., 2010c), preserving phylogenetic diversity (Butterfield et al., 2013), and regulating alpine soil microbiomes (Roy et al., 2013; Chang et al., 2018; Wang et al., 2020). Additionally, they contribute to the maintenance of species interaction networks, thereby preserving alpine biodiversity and ecosystem functions (Losapio et al., 2019; Losapio and Schöb, 2020; Niu et al., 2025a, 2025b). Therefore, ensuring the long-term persistence of cushion plants is vital for maintaining overall biodiversity in high mountain communities.

The QTP is widely recognized as one of the diversification centers for cushion plants, with nearly 130 cushion species identified to date (Zhang et al., 2022b). Given their ecological importance (Yang et al., 2010c; Chen et al., 2015a, 2015b, 2019b, 2021b; Liu et al., 2016; Jiang et al., 2018), changes in the dynamics of cushion species can have profound impacts on associated biota. Consequently, cushion-dominated communities at high elevations of QTP provide an excellent model for studying community dynamics and the impacts of climate warming (Huang and Wang, 1991; Zhao et al., 2011; Chen et al., 2020b, 2024a). Concerns regarding the negative impact of cushion plant degeneration on alpine ecosystems have persisted for decades (Huang and Wang, 1991), and recent research suggests that vegetation dominated by cushions is likely to decline with ongoing climate warming (Zhao et al., 2011). However, due to their slow growth rates and long lifespans, tracking the population dynamics of cushion plants poses challenges (Benedict, 1989; Morris and Doak, 1998; Chen et al., 2017a). A recent study conducted in HDM highlights that Arenaria polytrichoides, a representative cushion plant in this region, is migrating upward in response to climate warming (Chen et al., 2024a). While A. polytrichoides populations at higher elevations are expanding, those at lower elevations are declining, potentially leading to significant alterations in alpine plant communities (Chen et al., 2024a).

Changes in community dynamics are evident in ecological ecotones, particularly in treeline ecotones of high-elevation/latitude regions. Current studies on treeline responses to climate change in the QTP can be divided into three categories: (1) dendroecological studies focusing on growth responses (Liang et al., 2011); (2) population studies, which investigate spatial and elevational expansion (Baker and Moseley, 2007); and (3) succession studies, which examine forest composition. Regional dendroecological data have shown that Abies treelines in the HDM region have not exhibited significant upward movement over the past 400 years (Liang et al., 2011). A recent study using satellite-derived tree cover data found that although some treelines in the QTP have remained stable, most have advanced over the past century (Wang et al., 2022c). In addition, climate warming has been shown to accelerate successional processes at Himalayan treelines, leading to late-successional species (Abies spectabilis) quickly outcompeting pioneer species (Betula utilis), likely affecting forest composition at climatically controlled range limits (Sigdel et al., 2024). Research has also identified a climate-induced transformation of spruce-fir forests to pine and broadleaved forests on the southeastern Qinghai-Tibet Plateau (Zhang et al., 2025, Zhang et al., 2025). Such a change in foundation species is likely to have profound consequences for ecosystem functions and services, a critical issue that warrants further investigation.

Variations in the responses of treeline species to climate change can be explained by multiple abiotic and biotic factors, including species-specific climate tolerance (Harsch et al., 2009), time lags in response (Bertrand et al., 2011), and local soil and nutrient constraints (e.g., Loomis et al., 2006; Shi et al., 2006; McNown and Sullivan, 2013; Müller et al., 2016). Biotic interactions play a crucial role in shaping treeline dynamics. For example, Rhododendron shrubs in the HDM have been shown to facilitate tree establishment beyond the climatic treeline by improving micro-environmental conditions for Larix potaninii var. macrocarpa and Picea likiangensis (Chen et al., 2020c). In addition, dense shrub or herb coverage in regions adjacent to the QTP was found to hinder treeline upslope movement by intensifying interspecific competition, suggesting that species interactions can modulate treeline responses to climate warming by slowing migration rates or preventing movement altogether (Liang et al., 2016). Treeline dynamics are also significantly impacted by human activities, such as intensified pastoralism and livestock grazing. For instance, in the eastern part of QTP, human activity and grazing has been shown to deter upward shifts in treelines induced by climate warming (Wang et al., 2019).

Compared to treeline ecotones, few studies have examined the impacts of climate change on other communities or transition zones. In the sub-nival ecosystem of the QTP, soil seed assemblages were found to differ between patches, with bare ground habitats having the lowest seed diversity and density but harboring the largest empty niches (Chen et al., 2021c). These bare ground habitats are likely to experience more pronounced community recruitment changes than grasslands and rock beds when thermal conditions improve due to global warming (Chen et al., 2021c).

4.3. Changes in vegetation on the QTP

The extensive elevational range and complex topography of the QTP support a diverse array of vegetation types, spanning from tropical to polar zones (The China Vegetation Atlas Editorial Committee and Chinese Academy of Sciences, 2001). Numerous studies have predicted potential vegetation changes in China (e.g., Ni et al., 2000; Weng and Zhou, 2006; Yu et al., 2006; Wang et al., 2018; Liao et al., 2020, 2021) and identified alpine vegetation in the QTP as highly sensitive to climate change (e.g., Miehe, 1996; Yu et al., 2006; Zhao et al., 2011; Gao et al., 2013; Gao et al., 2016; Huang et al., 2016; Chen et al., 2020b; Zhu et al., 2020; Chen et al., 2021b; Zhou et al., 2022). Some modeling studies on the QTP predict a contraction of alpine meadows, steppes, and sparse/cushion vegetation and deserts, and an expansion of shrubs, broad-leaved, mixed, and coniferous forests to higher elevations and latitudes due to climate warming (He et al., 2020; Zhao et al., 2021). Other models predict changes in the distribution of mixed and deciduous broad-leaf forests in the QTP, shifting their distribution toward northern China, while grasslands, shrublands, and wooded grasslands may extend into southeastern China (Yu et al., 2006).

Generally, rising temperatures and increased precipitation have been found to boost vegetation growth in the QTP (Huang et al., 2016). However, responses to thermal and hydrological factors may vary among different vegetation types (Yu et al., 2006; Zhao et al., 2011; Huang et al., 2016). For example, rising precipitation during a period of higher temperatures did not increase the growth of dry steppe in the QTP (from 1986 to 2000) (Huang et al., 2016). During a similarly warm period (2000–2011), declines in precipitation were found to inhibit the growth of meadow vegetation and wetter conditions were found to promote the growth of wet meadow.

Human activities also exert significant impacts on vegetation dynamics (e.g., Miehe et al., 2014, 2023; Huang et al., 2016; Chen et al., 2017b; Yin et al., 2020; Yuan et al., 2021). Grasslands, in particular, are experiencing severe degradation due to global warming and intensified human economic activities (Zhao and Zhou, 2005; Liu et al., 2008; Miehe et al., 2019; Yuan et al., 2021). Such changes threaten the balance between human activities and ecosystem functions, e.g., the upward migration of forests may reduce alpine meadows, thereby diminishing the QTP’s capacity for animal husbandry.

5. Native plant invasion in QTP as a result of climate change and human activity

Mountains have become ‘hotspots’ for biological invasions (Iseli et al., 2023). Unsurprisingly, biological invasions induced by climate change have been declared a crisis in the QTP and adjacent areas (Gao et al., 2019; Wang et al., 2023a). Climate change on the QTP increased multifaceted threats from human activities, including grazing, agricultural reclamation, and infrastructure development (Zhang et al., 2021). These activities have significantly promoted the invasion of native plants (Xiong et al., 2025), known as “native invasions” (Valéry et al., 2008), more than just promoting the spread of alien invasive plants. Thus, the research paradigm of biological invasions, especially in the QTP, requires some re-thinking: “Is the threat of alien invasions necessarily more urgent than that of native invasions?”. Current research focuses mostly on alien biological invasions (Mooney and Cleland, 2001; Charles and Dukes, 2008), however, in the QTP native species have been shown to rapidly become dominant in response to changing land-use practices or increased disturbances, causing harm similar to that of alien invasive species (Pivello et al., 2018). In addition, the current approach to biological invasions often overemphasizes scenarios where species cross geopolitical boundaries, neglecting numerous actual invasion harms caused by native invaders. Despite over a decade of advocacy and synthesis (Carey et al., 2012; Pivello et al., 2018), this conceptual framework remains in its nascent stage. Specifically, there is a lack of sufficient understanding of the mechanisms of biological invasion, potential impacts, and response strategies concerning native plant invasions in QTP (Carey et al., 2012). Furthermore, studying native plant invasions is an important pathway to promoting sustainable development: since the natural enemies of native invasive plants are locally present, biocontrol methods developed on this basis can avoid the risk of secondary invasions and demonstrate strong adaptability. In contrast, biological control of alien invasions has provided numerous cautionary examples.

One key factor for the successful invasion of native plants is their ability to produce specific compounds that inhibit the growth of other plants, a phenomenon termed ‘allelopathy’ (Wang et al., 2022a; Revillini et al., 2023). For instance, research has shown that extracts from the native invasive plant Ligularia cymbulifera can inhibit germination in common alpine forage grasses (Wang et al., 2022a). Similar research has found that extracts from the native invasive plant Euphorbia jolkinii inhibit germination and growth of forage grasses and that plant diversity was lower in invaded communities than in non-invaded communities (Wang et al., 2022d). These findings resemble those from alien plant invasion research that gave rise to the “Novel Weapons Hypothesis,” which posits that invasive species produce novel allelopathic compounds in the environment, giving them a competitive advantage over native plants (Callaway and Ridenour, 2004). In contrast to well-documented allelopathy in alien plant invasion and its effects on soil dynamics (Ehlers et al., 2020), the role and impacts of allelopathy in native plant invasion and below-ground ecological processes have received less attention.

Native plant invasions create unique soil chemical environment by releasing specific allelopathic compounds (Wang et al., 2022a). These compounds may exert selective pressures on microbial communities, allowing organisms well-adapted to these chemical conditions to thrive while eliminating others that cannot adapt (Revillini et al., 2023). This selective pressure leads to a less diverse microbial composition, resulting in reduced beta diversity and community homogenization (Gao et al., 2022). Additionally, plant invasions reshape microbial interactions, such as reducing the complexity and stability of microbial networks (Gao et al., 2022). As the primary source of this selective pressure, higher concentrations of these allelopathic compounds increase the deterministic processes in microbial community assembly (Wen et al., 2022). In addition to altering microbial communities, allelopathy also affects soil functions. For example, allelopathic compounds such as volatile monoterpenes can inhibit nitrogen mineralization rates and nitrification, as well as influence carbon cycling (Smolander et al., 2011). Thus, the accelerated invasion of native plants in QTP is rapidly reshaping microbial communities and ecosystem functioning, yet we lack substantial empirical evidence to reveal how this process unfolds (Fig. 3).

Some microbes have adapted to allelopathy and can degrade specific allelopathic compounds, thereby alleviating plant growth stress (Li et al., 2015). Research also shows that invaded communities have higher resource availability (e.g., total carbon content) compared to non-invaded communities, which can lead to higher bacterial richness in invaded communities (Gao et al., 2022). This may be related to the role of bacteria in degrading allelopathic substances and forming carbon sources. Because specific compounds select adaptive microbes, these microbes often carry specific functional genes involved in the degradation process (Liu et al., 2023). Therefore, selecting specific microorganisms or constructing synthetic microbial communities to degrade allelochemicals represents a promising and sustainable approach to addressing native plant invasions in QTP (Liu et al., 2023; Wang et al., 2024b).

Extensive consequences of native species invasion may involve multiple degradation processes of native vegetation, such as the upward shift of alpine treelines and the decline of dominant communities (Fig. 4a). Although studies have found that the elevational distribution of regional treelines can be limited by shrubs such as Rhododendron (Chen et al., 2020c), some treeline species (e.g., Larix) have been found to exhibit strong allelopathic effects that inhibit the growth of plants (Kong, 2007; Yang et al., 2010a). Similar situations have also been observed for cushion plant communities that have suffered invasions from native plant species (Chen et al., 2024a). Such invasions can lead to significant changes in plant and microbial composition and abundance (Chen et al., 2024a; Zhang et al., 2025b). During this process, invasive, non-cushion plants have been shown to exert allelopathic effects on cushion plants, inhibiting the regeneration of their seedlings (Chen et al., 2024b) (Fig. 4b). Together these findings indicate native plant invasions are mediated by a complex array of abiotic and biotic interactions. Understanding the consequences of these processes requires further investigation into these phytochemical–microbe relationships. However, it is important to note that not all the aforementioned vegetation degradation processes can be attributed to native species invasion. Clearly defining native plant invasion requires broad and solid investigation, which is currently the most lacking aspect.

6. How has commercial harvesting affected plant diversity and evolution on the QTP?

Recent estimates suggest that approximately 24,000 plant species are used for medicinal purposes worldwide (Pironon et al., 2024), but only a little over 3200 species are cultivated commercially. The commercial harvesting of alpine medicinal plants is common in QTP, resulting in various impacts on plant diversity (Olsen, 2005; Klein et al., 2008). Notable species include Aconitum heterophyllum, Allium stracheyi, Angelica glauca, Bergenia stracheyi, Dactylorhiza hatagirea, Fritillaria delavayi, Nardostachys jatamansi, Neopicrorhiza scrophulariiflora, Rheum australe, R. tanguticum, Rhodiola crenulata, Saussurea laniceps, and others (e.g., Ghimire et al., 2005; Ma and Zhuang, 2010; Wang and Li, 2017; Sharma et al., 2022).

Some medicinal plants are more heavily harvested than others due to a variety of factors. Medicinal plants have high economic value. Accordingly, species with high market demand and high prices are particularly susceptible to overharvesting. For example, the harvesting of Fritillaria cirrhosa bulbs is particularly intense due to its high market value, with prices reaching up to USD 715 (CNY 5000) per kilogram (https://www.zyctd.com, accessed in September 2024). When the harvesting seasons of Fritillaria and Nardostachys overlap, harvesters often prioritize the former (Chang, 2022). Harvesting of medicinal plants is also influenced by resource availability. When the population of a medicinal plant drops below a certain threshold, collectors transfer to alternative or substitute species. For instance, the snow lotus Saussurea laniceps was once heavily harvested due to its unique appearance and medicinal properties. However, as its population declined, collectors turned to the more abundant, though less conspicuous, Saussurea medusa. Harvesting is also influenced by target accessibility. Improved transportation routes, such as roads, enhance access to previously remote regions, making it easier for collectors to reach and harvest plants. For example, the expansion of road networks has opened up new populations of Bergenia ciliata to commercial harvesting (Pyakurel et al., 2018).

Commercial harvesting can cause rapid and irreversible population declines. Previous research has estimated that if more than 10% of Nardostachys jatamansi individuals are harvested, the ramet density will be unable to recover to pre-harvest levels for several years (Ghimire et al., 2005). Similar population declines have also been reported in other alpine medicinal herbs, such as Picrorhiza kurrooa, Aconitum heterophyllum, and Fritillaria roylei in the Western Himalaya (Uniyal, 2017). Eleven plant species subject to large-scale collection in the eastern part of QTP face similar issues, with nine of these experiencing resource depletion. In some areas, the fatality rate of harvesting reaches as high as 78% (Ma and Zhuang, 2010). Notably, about 10% of medicinal species from the QTP are listed in the ‘China Species Red List.’

Commercial harvesting can also reduce the biodiversity of the community indirectly. For example, the harvesting of Ophiocordyceps sinensis has been shown to decrease species diversity, coverage, and aboveground biomass of surrounding grassland plants (Xu et al., 2013). One explanation for this finding is that harvesting has changed interactions between plants, animals, and microbiomes. This has also been found in communities where harvesting removes key species from a network, e.g., a pollination network. In pollination networks, plant species within a community share pollinators, leading to either dependence or competition among them. Generalists plants, species that provide significant rewards to various pollinators, play a crucial role in maintaining the stability of the pollination network (González et al., 2010). Over-harvesting these plant species may remove such a key species from a network, reducing the stability of the community.

Continuous and targeted harvesting of wild plants has been shown to exert strong selective pressure on specific traits and induce phenotypic changes (Darimont et al., 2009). Furthermore, traits favored by harvesters, e.g., size, color, flowering season, and seed maturation time, often influence survival and reproduction. For example, plants harvested before flowering and set seed have lower capacities to reproduce successfully and maintain population sizes. These changes may lead to evolutionary changes within these populations, e.g., selecting for smaller plants that reproduce quickly.

The snow lotus Saussurea laniceps has long been heavily harvested for medicinal purposes. Research on snow lotuses has suggested that harvesting has selected for smaller plant size over the past century. In contrast, the less conspicuous congener S. medusa has not shown this trend (Law and Salick, 2005), however, as S. laniceps has become rare, S. medusa and other congeners are now more heavily harvested. Research has also suggested that commercial harvesting has influenced the evolution of color in Fritillaria delavayi, a medicinal herb from the QTP (Niu et al., 2021). Specifically, researchers found that in heavily harvested populations F. delavayi are well camouflaged, however, in less harvested populations the plants are green. Harvesting may have similar effects on underground traits. For example, human harvesting is thought to have decreased root size in the household plant Nardostachys jatamansi (Chang, 2022). All these studies suggest that human actives are changing biodiversity in this region and driving plant evolution in an unexpected and dramatic way.

7. Perspectives

Although recent research has characterized how plant diversity in the QTP responds to climate change and human activity, we know little about the mechanisms that underlie these responses, acclimation and adaptation. The QTP offers a unique opportunity to understand these gaps in our knowledge. Specifically, we recommend that the following areas be further investigated.

1. Plant growth and reproduction: Little is known about how alpine forbs in the QTP have evolved in response to climate change in the Anthropocene. We believe that studies on ecological adaptations of high mountain plants and their responses to climate change would benefit from interdisciplinary collaborations between anatomist, botanist, and ecologists. Studies on how plant reproduction has been affected by climate change may require a different approach. Most knowledge on plant reproduction in QTP is species-specific or localized. Thus, studies are required that examine how climate change and human activity affect plant reproduction in diverse ecosystems over large spatiotemporal scales. Importantly, climate change may drive evolutionary adaptations that optimize plant fitness, e.g., floral longevity. Therefore, to identify taxa that may be susceptible to environmental perturbations, it is crucial to determine whether the plastic responses of plants’ reproductive traits to environmental changes are phylogenetically constrained.

2. Plant phenology: Long-term programs are required that carry out ground observations of both leaf-out and flowering phenology of either woody or herbaceous taxa in the QTP. We recommend that these programs focus on the cold and/or arid distribution limits of plants sensitive to changes in temperature and/or moisture conditions. Studies that aim to gain insight into the mechanisms underlying the overall phenotypic response of plants to climate changes should utilize logistic support from field research stations in the QTP to carry out field experiments.

3. Plant distribution range: One of the ecosystems most sensitive to climate change on the QTP is found on mountaintops. However, our understanding of how plant diversity and distribution of mountaintop ecosystems respond to climate change remains very limited and localized. Studies are needed that detect how species, communities, and vegetation currently distributed on the high mountaintop respond to climate change, with particular attention on the degradation of some foundation species, such as cushion plants and the associated changes in the functions of such a ‘sky island’ ecosystem.

4. Native plant invasion: Future studies should investigate the impact of native plant invasions on belowground processes (e.g., soil microbial diversity, nutrient cycling, and soil enzyme activities). These studies should focus on identifying and characterizing allelopathic compounds produced by invasive native plants and their models of action. Understanding these chemical interactions can aid in developing management practices to mitigate their impacts. We also recommend that studies examine multi-trophic interactions to gain insights into how native plant invasions affect herbivores, pollinators, decomposers, as well as the ecological consequences of these invasions on ecosystem functions.

5. Commercial harvesting: Understanding the ongoing threat of harvesting medicinal plants on the QTP requires more field studies. Our hope is that field studies may provide insights into effective conservation strategies and warning systems that protect over-harvested plant species. Researchers should also strive to develop harvesting strategies that promote sustainable development while recognizing the cultural and economic significance of traditional medicinal plants in local communities.

Acknowledgements

Hang Sun acknowledges the Second Tibetan Plateau Scientific Expedition and Research program (2024QZKK0200), the Key Projects of the Joint Fund of the National Natural Science Foundation of China (U23A20149) and Yunnan Key R&D Program (202403AC00028) for supporting the field excursion, samples collections and ecological experiment in QTP. Rest co-authors acknowledge the Yunnan Innovation Team Project (202305AS350004 to Yang Yang), the Young Academic and Technical Leader Raising Foundation of Yunnan Province (202205AC160053 to Jianguo Chen), the CAS “Light of West China” Program (xbzg-zdsys-202319 to Bo Song), Yunnan Revitalization Talent Support Program “Young Talent” Project (to Yazhou Zhang), National Youth Talent Support Program (to Yang Niu) and Post-doctoral (oversea) Fund of Ministry of Education of China (to Zihan Jiang). We also thank Hongyan Jin (Kunming Institute of Botany) for her assistance during the preparation of this manuscript.

CRediT authorship contribution statement

Yang Yang: Funding acquisition, Writing-original draft, Writing-review & editing. Jianguo Chen: Funding acquisition, Writing-original draft, Writing-review. Bo Song: Funding acquisition, Writing-original draft, Writing-review. Yazhou Zhang: Funding acquisition, Writing-original draft, Writing-review. Yang Niu: Funding acquisition, Writing-original draft, Writing-review. Zihan Jiang: Funding acquisition, Writing-original draft, Writing-review. Hang Sun: Conceptualization, Supervision, Funding acquisition, Writing-original draft, Writing-review & editing.

Declaration of competing interest

Hang Sun, Zihan Jiang, Bo Song and Yang Niu are the Editors and Yazhou Zhang is the junior Editor of Plant Diversity and were not involved in the editorial review or the decision to publish this article. The other authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.