Human activities and global changes have led to alterations in global and regional precipitation regimes. Despite extensive studies on the effects of changes in precipitation regimes on plant community composition across different types of grassland worldwide, few studies have specifically focused on the effects of precipitation changes on high-altitude alpine steppe at community and plant species levels in the Tibetan Plateau.

We investigated the effects of growing-season precipitation changes (reduced precipitation by 50%, ambient precipitation, enhanced precipitation by 50%) for 6 years on plant community composition in an alpine steppe of the Tibetan Plateau by linking above- to belowground traits of dominant species.

We found that reduced precipitation shifted community composition from dominance by bunchgrass (primarily Stipa purpurea) to dominance by rhizomatous grass (primarily Leymus secalinus). Roots and leaf traits of L. secalinus and S. purpurea differed in their responses to reduced precipitation. Reduced precipitation enhanced root vertical length and carbon (C) allocation to deep soil layers, and decreased the leaf width in L. secalinus, but it did not change the traits in S. purpurea. Moreover, reduced precipitation significantly enhanced rhizome biomass, length, diameter and adventitious root at the rhizome nodes in L. secalinus. These changes in traits may render rhizomatous grass greater competitive during drought stress. Therefore, our findings highlight important roles of above- and belowground traits of dominant species in plant community composition of alpine steppe under precipitation change.

Adaptive convergence in floral phenotype among plants sharing a pollinator guild has been acknowledged in the concept of pollination syndrome. However, many plants display traits associated with a given syndrome, but are visited by multiple pollinators. This situation may indicate the beginning of a pollinator shift or may result in a stable situation with adaptations to different pollinators. In Salvia stachydifolia, a previous study suggested that flower shape is optimized to maximize the contribution to pollination of bees and hummingbirds. Here, we studied three additional aspects of its floral biology: sexual phases, nectar dynamics and breeding system, and examined their connection with pollinators’ behaviour to explore the presence of adaptations to bee and/or hummingbird pollination.

Using a greenhouse population, we applied five pollination treatments to characterize breeding system. To determine sexual phases, we recorded flower opening, anther dehiscence, corolla fall and stigma receptivity. Additionally, we characterized nectar volume and concentration dynamics along the day. Finally, to determine pollinator assemblage and visitation patterns, we performed field observations and recorded pollinators’ behaviour.

Salvia stachydifolia was partially protandrous and self-compatible, but open-pollinated plants attained the highest reproductive success, suggesting that reproduction is mainly dependent on pollinator activity. Bombus opifex bumblebees were the most frequent visitors, but Sappho sparganura hummingbirds dominated visits early in the morning and at dusk. Nectar was typical of bumblebee pollination. We suggest that the bee–hummingbird mixed visitation constitutes an unstable evolutionary situation, making S. stachydifolia an ideal system to understand the ecological circumstances in which pollination shifts occur.

Leaf nutrient resorption is sensitive to changes in soil nutrients. However, the effects of N deposition on nutrient resorption efficiency (NuRE) in plant macro-nutrients remain unclear. Poplar (Populus deltoids) is one of the most extensively cultivated hardwood species worldwide. We explored general patterns and dominant drivers of NuRE and stoichiometry of poplar plantations in response to N addition.

We conducted a 4-year N-addition experiment to explore NuRE and stoichiometric responses to N addition in two poplar (P. deltoids) plantations (8- and 12-year-old stands) in a coastal region of eastern China. We measured soil and foliar (green and senesced leaves) concentrations of nitrogen (N), phosphorus (P), potassium (K), calcium (Ca) and magnesium (Mg) for a series of N addition treatments including N₀ (0 kg N ha−1 yr−1), N₁ (50 kg N ha−1 yr−1), N₂ (100 kg N ha−1 yr−1), N₃ (150 kg N ha−1 yr−1) and N₄ (300 kg N ha−1 yr−1).

Consistent for (both) 8- and 12-year-old stands, N addition did not affect the NuRE and stoichiometry (with the exception of CaRE and CaRE:MgRE ratio). N resorption efficiency–P resorption efficiency (NRE–PRE) scaling slopes were consistently less than 1.0 under N addition. These results suggest that NRE generally decouples from PRE within each N treatment. Moreover, these results point to robust control of green leaf nutritional status on nutrient resorption processes as indicated by the positive relationships between NuRE and green leaf nutrient concentrations. Our findings provided a direct evidence that growth in 12-year-old poplar plantations was N-limited in the coastal region of eastern China.

Our objective was to quantify the contributions of the seed bank and the established vegetation to the species composition, functional composition and diversity, and discuss the implications of these differences in regeneration and persistence of floodplain plant communities.

We sampled all ground cover vegetation up to 1.5 m height and seed bank in 25 plots (10 m × 1 m) distributed across five sites in dry and rainy seasons in a periodically flooded savanna in the Pantanal wetland, Brazil. We evaluated the soil seed bank by seedling emergence method.

The seed bank species had traits that conferred regeneration to the communities, while persistence traits characterized the vegetation. The seed bank had higher functional richness and lower functional evenness than the vegetation. The existence of different plant traits between seed bank and vegetation allowed the coexistence of species with functionally contrasting persistence and regeneration traits, which may help maintain functional diversity. It may allow the community to be more resilient when dealing with different environmental filters such as drought, fire and flood.

The scaling relationship between nitrogen (N) and phosphorus (P) concentrations ([N] and [P], respectively) in leaves manifests plants’ relative investment between the two nutrients. However, the variation in this relationship among taxa as well as its causes was seldom described.

The analysis was based on a dataset including 2483 leaf samples from 46 genera of angiosperm woody plants from 1733 sites across China. We calculated the leaf N–P scaling exponent (β_L) with an allometric equation ([N] = α[P]β), for each genus, respectively. We then performed phylogenetic path analyses to test how the climate and soil niche conditions of these genera contributed to the inter-genus variation in β_L.

The genera living with lower soil P availability presented a more favoured P uptake relative to N, as shown by the higher β_L, suggesting a resistant trend to P limitation. Additionally, genus-wise β_L was positively correlated with soil N–P scaling exponents (β_S), implying that the variation in leaf nutrients is constrained by the variability in their sources from soil. Finally, climatic factors including temperature and moisture did not affect β_L directly, but could have an indirect influence by mediating soil nutrients. Phylogeny did not affect the inter-genus variation in β_L along environmental gradients. These results reveal that the trade-off between N and P uptake is remarkably shaped by genus niches, especially soil nutrient conditions, suggesting that the β_L could be considered as a functional trait reflecting characteristics of nutrient utilization of plant taxa in response to niche differentiation.

Understanding the patterns and drivers of carbon isotope discrimination (13Δ) in C₃ and C₄ functional groups is critical for predicting C₃/C₄ vegetation ratio from the isotopic composition of soil organic matter. In this study, we aimed to evaluate how intraspecific variation will modify functional group-level 13Δ values and the associated prediction of C₃/C₄ vegetation ratio.

We investigated 13Δ of 726 individual plants (96 species; C₃ and C₄ functional groups) and topsoil organic matter in 26 grassland communities along an aridity gradient in northern China. The fraction of C₄ contribution was calculated with mixing models that considered: (i) both intra- and interspecific effects on the 13Δ values of C₃ and C₄ functional groups; (ii) only interspecific effects; or (iii) none of these effects.

We found divergent responses of plant 13Δ at the intraspecific level to the changes of aridity across the gradient. The 13Δ of both C₃ and C₄ functional groups was negatively correlated with an aridity index, with higher sensitivity for C₃ than for C₄ functional groups. Intraspecific 13Δ variation played a key role in driving the total 13Δ variations of C₃ plants. Overlooking such intraspecific effect in mixing models led to a greatly increased fraction of C₄ contribution to soil organic carbon. A correction for the effects of intraspecific variation is therefore essential for correctly inferring C₃/C₄ vegetation ratio in the past. Our findings provide basic information for the reconstruction of past vegetation change from bulk materials in arid and semiarid biomes.

Within-species genetic and phenotypic variation have well-known effects on evolutionary processes, but less is known about how within-species variation may influence community-level processes. Ecologically meaningful intraspecific variation might be particularly important in the context of anthropogenic impacts on natural systems, such as agriculture and species invasion, because human actions can cause strong selection pressures.

In a greenhouse study, we explored intraspecific (30 accessions) and ecotypic variation (representing agricultural and nonagricultural habitats) in biomass and rhizome production in response to inter- and intraspecific competition and soil fertility of Johnsongrass (Sorghum halepense), a widespread invasive species and agricultural weed.

Contrary to our expectations and previous results, we did not find variation in biomass production among Johnsongrass ecotypes at this early life stage. However, we did find that Johnsongrass biomass varied substantially depending on competitor identity, soil fertility treatments and among accessions. Rhizomes were 11% larger in the agricultural ecotype and up to 3-fold larger in fertilized treatment; while rhizome biomass increased by ~50% when fertilized, but did not differ among ecotypes. Interestingly, in competition, Johnsongrass produced 32% less biomass and 20% less rhizome mass with a conspecific than when competing interspecifically with corn. Our results indicate species-specific competitive responses and changes in rhizome allocation in response to neighbor identity; suggesting the possibility of adaptation by Johnsongrass to shift allocation under competition.

Invasive plants may alter soil fungal communities in a way that improves their growth. Nitrogen (N) content of soil affects the symbiosis between plants and arbuscular mycorrhizal fungi (AMF), further determining plant growth. Yet, it is unclear whether altered AMF communities change the dependence of invasive and native species on N-form, and whether N forms alter the invasive plant–AMF interaction (PSI_M).

Two synthetic plant communities, including four Solidago canadensis individuals and four native plant species, were inoculated with AMF spores from S. canadensis-invaded soils and adjacent non-invaded soils, and were provided with nitrate, ammonia or glutamate. After their growth, the performance of the two plant communities in treatments of AMF origin and N forms, and the pathways of the N forms affecting S. canadensis growth and PSI_M were evaluated.

Solidago canadensis had no obvious N-form dependence in any of the AMF inoculations. Native plant species showed weak N-form dependence, but invasive AMF could remove their N-form dependence. In the absence of N, AMF did not affect growth of S. canadensis and the native plants. In contrast, with N addition, invasive AMF significantly increased belowground and total biomass of the invasive plants but not those of the native plants. Positive PSI_M of S. canadensis was also evidently greater than that of native plant species and was realized through directly or indirectly regulating phenotypic traits including plant height, leaf number and number of rhizomes. Our findings emphasize the importance of plant–AMF interactions and a unique N-acquisition strategy during plant invasions.

Dispersal through space or time via dormancy is one of the primary processes whereby organisms can influence the environment they experience. In plants, strong evolutionary correlations are expected between the two kinds of dispersal because both are performed by the seeds and play comparable adaptive roles. In this paper, we investigated these evolutionary correlations using amphicarpic plants, which simultaneously produce aerial seeds with high spatial dispersal propensity and subterranean seeds that do not disperse.

We investigated the variation in dormancy and germination in aerial and subterranean seeds of two amphicarpic legumes (Vicia amphicarpa L. and Lathyrus amphicarpos L.) and two closely related homocarpic taxa (Vicia sativa L. and Lathyrus cicera L.) by estimating germination percentages following different combinations of dormancy breaking treatments (i.e. dry after-ripening, cold stratification and physical scarification).

Our results showed complex interactions between spatial and temporal dispersal. Right after dispersal, aerial seeds were more dormant than their subterranean counterparts, but this trend reversed with after-ripening, as seeds developed physical dormancy. Seeds of homocarpic plants germinated at higher percentages than those of their amphicarpic congeners and lost dormancy homogeneously with after-ripening. Conversely, amphicarpic seeds exhibited varied dormancy strategies modulated by both physiological and physical dormancy. These are expected to increase variation in emergence timing, providing multiple levels of diversifying bet-hedging. This strategy might be adaptive under highly unpredictable conditions by enabling plants to rely on historically favorable sites in good seasons without preventing spatial and temporal migration.

It is important to explore the underlying mechanisms that cause triphasic species–area relationship (triphasic SAR) across different scales in order to understand the spatial patterns of biodiversity.

Instead of theory establishment or field data derivation, I adopted a data simulation method that used the power function of SAR to fit log-normal distribution of species abundance.

The results showed that one-step sampling caused biphasic SAR and n-step sampling could cause 2n-phasic SAR. Practical two-step sampling produced triphasic SAR due to the Preston and Pan effects in large areas. Furthermore, before exploring biological or ecological mechanisms for the nature phenomenon, we should identify or exclude potential mathematical, statistical or sampling reasons.

Functional group composition of a plant community is mainly driven by environmental factors and is one of the main determinants of grassland biodiversity and productivity. Therefore, it is important to understand the role of plant functional groups (PFGs) in mediating the impact of environmental conditions on ecosystem functions and biodiversity.

We measured plant biomass and species richness (SR) of grasslands in 65 sites on the Mongolian Plateau and classified 157 perennial herbaceous plants into two main PFGs (namely grasses and forbs). Using the random forest model and ordinary least squares regression, we identified that environmental factors (i.e. aridity index, soil total nitrogen [STN] and pH) were significantly related to the SR and aboveground biomass (AGB) of PFGs. We then used structural equation modeling to explore the relationship between the identified environmental factors and community SR and biomass, and the role of PFGs in driving this relationship.

We found that aridity index had unimodal relationships with both AGB and SR of the PFGs and the whole community. All SR and biomass metrics were significantly related to STN and pH. The relationship between aridity index and community biomass was mediated by an increase in the AGB of grasses. The influence of STN and pH on community SR was mainly due to their regulation in the SR of forbs. Our results indicate that community composition and the identity of the PFGs play a key role in linking environmental factors to ecosystem functioning.