Leaf chemistry plays a central role in structuring phyllosphere microbiomes. Plant populations often evolve genetic differences in leaf chemistry across region due to both abiotic and biotic selection pressures, including insect herbivory. Plants in invasive populations may reassociate with native specialist insects, providing an ideal system to examine how herbivory-mediated changes in plant chemistry affect phyllosphere microbiome. Here, we conducted a common garden experiment using Ambrosia artemisiifolia populations differing in leaf chemistry and reassociation history with a specialist beetle—Ophraella communa. We found that plant populations with a longer reassociation history exhibited stronger herbivore resistance and supported phyllosphere communities with higher alpha diversity and more complex composition. These changes were associated with shifts in concentrations of plant metabolites and the expression levels of corresponding biosynthetic genes. The abundance of the fungal pathogens, Golovinomyces, decreased with increasing herbivore resistance, while Pestaliopsis showed the opposite trend. Although reassociation history was linked to population latitude, climatic and soil conditions at the sites of origin also contributed to between-population variation in leaf chemistry and phyllosphere fungal community composition. Our study suggests that genetic differences in leaf chemistry among plant populations can strongly affect herbivore resistance and phyllosphere fungal communities. The observed alignment of reassociation history, chemical traits and phyllosphere fungal communities suggests that herbivore-mediated selection may be a key driver of microbial community evolution in invasive plants.

Land tillage disturbances and nutrient enrichment profoundly alter ecosystem processes and functions. Previous studies have explored the effects of tillage disturbance and nutrient enrichment on plant communities and soil properties. However, integrated studies of the effects of tillage disturbance and nutrient enrichment on multiple below-ground ecological processes and functions are needed. Here, we conducted a field experiment in the Hulunber grassland, establishing four treatments (control, tillage disturbance (D), nutrient enrichment (NPKμ) and tillage disturbance plus nutrient enrichment (NPKμD)) to examine their influences on plant communities, soil microbial communities, and carbon mineralization. Compared with the D treatment, the NPKμD treatment increased plant community biomass through a significant 13-fold rise in annual and biennial plant biomass (P < 0.01). Both the D treatment and NPKμD treatment significantly decreased the Shannon index of plant communities (P < 0.05). Microbial network complexity increased under NPKμ treatment whereas the D treatment reduced it. Both D treatment and NPKμ treatments significantly reduced soil carbon mineralization, and NPKμ exacerbated the negative effects of tillage disturbance (P < 0.05). Partial Least Squares Path Modeling showed that plant diversity, biomass and soil properties influenced soil carbon mineralization directly and indirectly via soil bacterial and fungal communities. Our findings suggest that nutrient enrichment promotes the recovery of plant community productivity after disturbance, while the recovery of plant diversity and soil microbial community structure may require a longer period. Therefore, achieving comprehensive ecological integrity characterized by stable plant community structure and healthy soil microbial communities requires long-term dynamic monitoring and targeted management strategies.

Warm temperate forests have the large potential to sequester atmospheric carbon dioxide (CO₂), while the interannual variability (IAV) of net forest ecosystem carbon exchange (NEE) in the global carbon cycle is still not fully understood. In this study, we conducted eddy-covariance measurement to investigate the IAV of carbon fluxes and concurrent influencing factors in a warm temperate natural oak forest from 2017 to 2022. Our results showed the natural oak forest was a strong CO₂ sink with an increase of 27.79 g C m⁻² a⁻¹ in annual carbon sequestration, resulting from a larger increase in annual gross primary production (GPP) than that of annual ecosystem respiration (Re). Precipitation in spring (PPT_spring) negatively influenced annual GPP, soil water content in spring (SWC_spring) negatively influenced annual Re, while the water conditions had lesser effect on annual NEE attributing to the synchronous changes of annual GPP and annual Re. Increase of temperature in autumn (Ta_autumn) delayed the end date of the growing season, leading to the increase in annual carbon sequestration. In addition, carbon fluxes did not significantly decrease under dramatic reduction of summer precipitation, indicating that warm temperate natural oak forest had a high resistance to seasonal drought. Our study helped us to better understand the mechanisms underlying forest carbon fluxes in response to drought in the context of future climate change.

Carbon (C) quality of non-leaf litter is closely related to decomposition rate and plays a vital role in terrestrial ecosystem C sequestration. However, to date, the global patterns and influencing factors of non-leaf litter C quality remain unclear. Here, using meta-analysis method, we quantified the characteristics and driving factors of the initial C quality of non-leaf litter (bark, branch, flower, fruit, root, stem, and wood) with 996 observations collected from 279 independent publications, including the concentrations of total C, lignin, cellulose, and hemicellulose. Results showed that (1) only total C and cellulose concentrations significantly varied among different types of non-leaf litter; (2) C quality is higher (i.e., lower concentration) in bark, branch, root, stem and wood litter from angiosperms than gymnosperms, from herbaceous than woody plants, from broadleaved than coniferous trees, and from arbuscular mycorrhizal (AM) than ectomycorrhizal (ECM) plants (except for hemicellulose concentration); and (3) the impacts of different geographic features on C quality of non-leaf litter differed among different litter types, while soil properties generally exhibited strong impacts. Overall, our results clearly show the global patterns of C quality and associated influencing factors for different types of non-leaf litter, which would be helpful for a better understanding of role of non-leaf litter in terrestrial ecosystem C cycling and for the improvement of C cycling models.

Xanthium italicum is a globally distributed invasive weed that causes significant ecological and agricultural damage in the invaded areas. Although multiple mechanisms have been reported to contribute to its invasive success, the extent to which intraspecific differentiation and phenotypic plasticity facilitate this process in invaded habitats remains insufficiently understood. In this study, we conducted a common garden experiment with three nitrogen treatments: no nitrogen addition (NN), low nitrogen (LN: 2 g urea per pot), and high nitrogen (HN: 4 g urea per pot). Ten populations of invasive X. italicum (ten individuals per population, 100 individuals total) and native Xanthium sibiricum (excluded from the NN treatment due to seed limitations) were grown under each nitrogen treatments. Under the NN treatment, we detected significant phenotypic differences among different invasive X. italicum populations across six growth traits (root length, shoot length, crown breadth, base diameter, relative chlorophyll content, and biomass). Furthermore, when subjected to the LN and HN treatments, invasive X. italicum exhibited significantly higher phenotypic plasticity compared with that of native X. sibiricum in biomass and base diameter. Our findings suggest that phenotypic plasticity and intraspecific differentiation may play important roles in facilitating the invasive success of X. italicum in China, potentially increasing the risk of further biological invasion.

Groundwater depth is a key environmental factor influencing the composition and structure of plant communities in coastal ecosystems. However, effects of the groundwater depth on the characteristics of shrub-grass communities in muddy coastal zones remain poorly understood. In this study, we conducted a field experiment to evaluate effects of the different groundwater depth (0.54, 0.83, 1.18, 1.62, and 2.04 m), on soil salinity, soil moisture, community diversity, distribution pattern and growth of the dominant Tamarix chinensis in the muddy coastal zone of Bohai Bay. Our results demonstrated that (1) the soil moisture and salinity gradually decreased with increasing groundwater depth (P < 0.001); Compared to the 0.54 m groundwater depth, soil moisture at depths of 0.83, 1.18, 1.62, and 2.04 m decreased by 16.02%, 24.83%, 54.40%, and 61.24%, and soil salinity decreased by 43.17%, 50.82%, 63.93%, and 73.41%, compared to 0.54 m, respectively. (2) The Simpson, Shannon-Wiener, Pielou and Margalef indices of the T. chinensis communities peaked at the 1.62 m groundwater table depth; (3) The dominant shrub T. chinensis population exhibited an aggregated distribution and optimal growth of T. chinensis shrubs occurring within the groundwater table depth range of 1.18 to 1.62 m; (4) The groundwater depth affected the diversity of the plant community primarily by influencing soil salinity rather than soil moisture; the dominant shrub T. chinensis promoted diversity of plant community, but this facilitation effect was inhibited by soil salinity. Our results suggest that the optimal groundwater depth for maintaining biodiversity falls within the range of 1.18 to 1.62 m. Shallow groundwater diminishes biodiversity both directly through soil salinization and indirectly by impairing T. chinensis’ facilitation of biodiversity. Therefore, regulating optimal groundwater table depth and protecting T. chinensis are critical for biodiversity conservation and ecosystem recovery in muddy coastal areas.

Soil nutrient heterogeneity is common in nature, but few studies have tested the effects of soil nutrient heterogeneity on plant productivity in natural communities. Such effects are of particular interest in habitats where heterogeneity may be increasing due to global warming, as in high-elevation grassland on the Qinghai-Tibetan Plateau. In a three-year study, we added N, P and K to 2 m × 2 m plots in grassland to establish five homogeneous and six heterogeneous treatments that varied in patch size, patch contrast (difference in nutrient levels between patches), and number of patch types (with different nutrient levels). We measured aboveground biomass of grasses, other graminoids, legumes and forbs. Biomass of grasses was higher (over 300 g m^–2) and that of legumes was lower (about 25 g m^–2) at higher nutrient availability in homogeneous treatments. Within heterogeneous plots, grasses similarly had about 25% higher biomass and legumes had about 60% lower biomass in patches with higher nutrients, mainly at the larger patch size and sometimes more so when patch contrast was greater. Accounting for 47%–61% of the total aboveground biomass, grasses showed a roughly similar pattern to that of the whole community. An increase in soil nutrient heterogeneity is unlikely to affect plant aboveground biomass in this grassland, although it can increase biomass of grasses and decrease that of legumes. Soil nutrient heterogeneity might partly mitigate these effects if large patches without elevated nutrients persist.

Atmospheric nitrogen (N) deposition includes inorganic N (IN) and organic N (ON). IN enrichment trends to reduce species richness greater than ON, likely lowering ecosystem stability, as species richness and ecosystem stability are usually positively related. However, previous field experiments evaluating N deposition effects on ecosystem stability used either IN or ON additions, likely biasing results. We assessed the effects of IN:ON ratios (0:10, 3:7, 5:5, 7:3, and 10:0) at 10 g N m^–2 year^–1 on the temporal stability of plant community productivity in a temperate meadow grassland using 6-year (2017–2022) data from a long-term N addition experiment established in 2014. Species richness, species asynchrony, population stability, and dominant species stability were investigated to explore mechanisms underlying community stability changes. We found that IN:ON ratio showed no significant effect on community stability, although all N addition significantly reduced community stability (averaged 26.7% reduction). However, IN decreased species richness more than ON (54.1% reduction in 10:0 vs. 31.8% reduction in 0:10). IN:ON ratio showed no significant effect on species asynchrony, population stability or dominant species stability. Species asynchrony and dominant species stability were both positively related to community stability, while population stability showed no significant association. It implies that species asynchrony and dominant species stability maintained community stability across IN:ON ratios. Overall, our findings suggest that, despite IN reducing species richness greater than ON, it may be reasonable to assess N deposition effects on ecosystem stability using either IN or ON addition.

Soil dissolved organic matter (DOM) is vital in terrestrial ecosystem carbon (C) cycling; however, the regulatory effects of forest types and elevations on soil DOM dynamics in mountain ecosystems remain incompletely understood. Here, we investigated DOM content, spectroscopic characteristics, molecular traits and their potential drivers along an elevational gradient (2600-3500 m) in the Hengduan Mountains. Our results showed that soil dissolved organic C (DOC) content was higher in broad-leaf forest soil (at 2900 m and 3500 m) than in coniferous forest soil (at 2600 m and 3200 m) irrespective of elevation, with a greater amount in wet season than in dry season. Humification index (HIX) trends aligned with the DOC content, while the aromaticity index (AI) showed an inverse relationship. These patterns were linked to the quality of litter carbon sources. Molecular-level analysis of DOM suggested that lignins/CRAM-like structure compounds and tannins predominated in soil DOM, indicating that the molecular composition of soil DOM was typical of plant-derived sources in our study region. Additionally, the relative abundance of lignin compounds decreased gradually with increasing elevation during the dry season. We detected that soil properties (especially, NH₄⁺-N content) predominantly mediated DOM dynamics in dry season, whereas litter traits (i.e., leaf-DOC content) were the key factors across elevations in wet season. Overall, our results revealed litter traits and soil properties predominantly regulated soil DOM mechanism along elevational gradient, indicating that soil DOM dynamics associated with tree species in alpine mountain ecosystems may differentially influence soil C sequestration under future climate change scenarios.

The linear electron transport to assimilation ratio (ETR/A) has been used to evaluate the stress status of plants and the ETR measurement accuracy. However, how ETR/A responds to short- and long-term CO₂ variations remains uncertain. Here, we assessed the short-term, instantaneous CO₂ response (seconds to minutes) of ETR/A in sunflower (Helianthus annuus) and cowpea (Vigna unguiculata) in a controlled experiment. Additionally, we compiled published data to investigate the effect of elevated growth CO₂ (long-term) on ETR/A across plant functional groups. Our results showed that ETR, photosystem II photochemical efficiency, and photochemical quenching increase while ETR/A decreases with increasing CO₂ for both sunflower and cowpea. The CO₂ sensitivity of ETR and ETR/A is greater at low [CO₂] (Rubisco-limiting condition) than at moderate to high [CO₂] (electron transport-limiting stage). The meta-analysis showed that plant functional groups and long-term CO₂ significantly affect ETR/A, which challenges the universal applicability of current empirical ETR/A thresholds across species and environments. Long-term CO₂ elevation increased A without altering ETR, resulting in a 21% reduction in ETR/A compared to ambient CO₂. Our results demonstrate an imbalance between electron transport and carboxylation under elevated CO₂, potentially constraining photosynthetic performance under future climate scenarios.

Wetland reclamation disrupts original biogeomorphic processes, making passive restoration after agricultural abandonment a key near‒natural solution. Soil organic carbon (TSOC), total nitrogen (TSN), and total phosphorus (TSP) storages serve as critical indicators of ecological restoration outcomes, closely linked to plant community succession and functional strategies, however, their drivers and influencing pathways remain unclear. This study examined soil functions (TSOC, TSN, and TSP), plant communities, and functional traits in passively restored freshwater wetlands following agricultural abandonment on China’s Sanjiang Plain. Results revealed that TSOC and TSN peaked at 14‒ and 17‒year post‒restoration, respectively, then stabilized, while TSP initially decreased before increasing. With extended restoration duration, plant communities showed increased height, coverage and biomass but decreased density and diversity, while functional traits transitioned from acquisitive to conservative strategies. Variance partitioning analysis revealed that soil function dynamics were primarily governed by plant community and functional trait interactions. Random forest models identified key drivers, while structural equation modeling delineated both direct effects of restoration duration and indirect pathways mediated by plant attributes. Specifically, synergistic declines in specific leaf area (SLA) and plant density enhanced TSOC accumulation. Coordinated reductions in SLA and stem phosphorus content (SPC) increased aboveground biomass (AGB), thereby elevating TSN but depleting TSP. A trade‒off between leaf phosphorus content (LPC) and root‒to‒shoot ratio (RSR) further modulated TSN dynamics. These findings demonstrate that passive wetland restoration facilitates soil function stabilization, with plant functional traits and community characteristics playing synergistic effects. This mechanistic understanding provides a scientific framework for optimizing restoration strategies.

Rising global temperatures are significantly affecting species distributions worldwide. Properly assessing the threat of invasive species in the context of global warming is crucial. In this study, we quantitatively assessed the potential threat of emerald ash borer (EAB) against global ash tree species (Fraxinus) under multiple future climate scenarios based on the premise of niche conservatism. Through a multidimensional comparison of overlapping distribution areas and niches forecasted by species distribution modeling, we observed that rising temperatures lead to significant shifts in the habitat ranges of both EABs and Fraxinus species, often resulting in increased overlap of both their suitable habitats and niches. These results indicate that global warming, across most climate scenarios, exacerbates the threat of biological invasions by EABs in all main distribution regions. This study highlights the critical importance of considering both invasive species and their potential hosts in predictive modeling. Additionally, our results establish a well theoretical foundation for future research and management strategies aimed at protecting vulnerable ecosystems form the expanding of invasive species.

Grazing is widely used in mountain land, which changes soil structure through feeding, trampling, excreta return, redistributing solar radiation, surface runoff, and then affects soil moisture (SM) and soil elements. However, research of interaction between topography and exclusion duration on relationship between soil and vegetation characteristics is scarce. The study was carried out to explore effect of topography and livestock exclusion duration on soil properties, correlation between soil and vegetation characteristics. The results showed that: (i) SM peaked at 3 years of exclusion. Water use efficiency, soil organic carbon, soil phosphorus, soil available nitrogen and soil available phosphorus were found to be directly proportional to duration of livestock exclusion and inversely correlated with slope. Soil nitrogen and N/P were directly proportional to duration of livestock exclusion and slope. C/N was inversely correlated with duration of livestock exclusion and slope. C/P was directly proportional to duration of livestock exclusion, and the change with slope was not obvious. Soil properties in sunny slope were greater than in shady slope. (ii) Aspect and slope positively affected the relationship between soil properties and aboveground biomass significantly. The effects of livestock exclusion on relationship between aboveground biomass, plant species richness and soil properties were insignificant. (iii) Livestock exclusion of sunny slope is more beneficial to soil nutrient accumulation than shady slope. Livestock exclusion played an opposite role to topography in regulating the relationship between soil and vegetation characteristics. Therefore, grazing management on complex topography is conducive to regulating soil nutrients and further coordinating vegetation growth.

Understanding differences in disease position (i.e. the average height of infected leaves) among fungal pathogens is crucial for predicting and managing plant diseases. However, we know little about how disease position varies across disease and host plant types, and whether the local climate (i.e. temperature and precipitation) affects disease position. Here, we investigated disease position in herbaceous plants across key grassland ecosystems in China, including the Qinghai-Tibetan Plateau, Inner Mongolia Plateau, and North China Plain. We tested how fungal pathogen characteristics (e.g. disease types and pathogen lifestyles), host plant characteristics (e.g. biomass, natural height and plant growth type), and climatic conditions (e.g. mean annual temperature [MAT] and precipitation [MAP]) affected disease position. Disease position tended to be higher for biotrophic versus necrotrophic pathogens, and this pattern was strongest in forbs and legumes. Disease position was also environment-dependent; higher temperatures and precipitation significantly increased disease position, but these effects varied among disease types. For both biotrophic and necrotrophic pathogens, larger host plants had lower mean disease positions. In this study, we provide evidence for how disease types and climatic conditions impact disease position; our findings emphasize the importance of disease position for understanding patterns of infection and managing disease outbreaks in a changing world.

Litsea coreana Levl. var. sinensis has developed four heteromorphic leaf types to adapt to the local environment. This study investigated functional traits of four heteromorphic leaf types associated with leaf morphology, anatomy, photosynthesis, and activity compounds, to elucidate their niche differentiation within a single tree. Lanceolate (La) type had the largest leaf length (LL = 8.4 cm), leaf shape index (LSI = 2.8), leaf perimeter (LP = 18.5 cm), but low palisade tissue thickness (TPTT), light saturation point (L_SP), and light compensation point (L_CP), allowing them suited for varied positions and mild, humid climates. Ovoid (Ov) and orbiculate (Or) types had highest leaf thickness (LT = 0.2 mm) and high TPTT (66.9 and 63.9 μm), high chlorophyll (0.695 and 0.696 mg g⁻¹), high net photosynthetic rate (P_Nmax = 8.1 and 6.6 μmol m⁻² s⁻¹), high total flavonoid content (TFC = 45.2 and 47.7 mg g⁻¹), indicating their adaptation to top and edge canopy positions with high temperatures and light conditions. Oblanceolate (Ob) type had largest SLA (117.2 cm² g⁻¹) and high dark respiration rate (R_d), but low TPTT and chlorophyll, which is important for shade adaptation in the lower canopy. Overall, LSI increased with the increases of LL, the higher value of TPTT, the greater value of P/S (Palisade/Spongy) was observed, and as the increases of L_SP, P_Nmax gradually increased. Three principal components were extracted with a cumulative contribution of 75.2%, of which the Ov type had the highest value. Our findings provided evidence that the presence of heteromorphic leaf types facilitates the utilization resources of different microhabitat by L. coreana Levl. var. sinensis.

Plant-associated symbiotic mycobiomes play critical roles in crop adaptation to harsh environments. However, knowledge of wheat mycobiomes in alpine ecosystems remains limited. This study investigated the diversity, composition and functional adaptations of symbiotic fungi associated with the wheats cultivated in alpine areas on the Pamir Plateau and adjacent lowland regions. Our results revealed distinct fungal community structures between the alpine and lowland habitats and across different plant compartments (stems, roots and rhizosphere), with higher fungal diversity observed in roots and rhizospheres compared to stems. Taxonomically, Eurotiomycetes predominated in alpine samples, while Tremellomycetes were more abundant in lowland areas. Fungal co-occurrence network analysis exhibited a higher proportion of positive associations among fungal taxa in alpine environments, supporting the stress gradient hypothesis that environmental stresses enhance mutualistic relationships. Functional analyses demonstrated that saprotrophic fungi dominated both regions; however, alpine fungi were more inclined toward endophytic and saprotrophic strategies, whereas pathogenic and parasitic fungi were prevalent in lowlands. These distinctions suggested that the harsh environmental conditions in alpine regions may drive plant-associated fungi toward biotrophic strategies as adaptive responses. Our findings highlight how environmental factors shape symbiotic fungal community composition and function, offering insights into utilizing these microbial communities for sustainable agriculture in challenging alpine conditions.

The home-field advantage (HFA) hypothesis predicts that litter decomposes faster in its home environment than elsewhere. Given the critical role of HFA effect in mediating terrestrial carbon and nutrient cycling, it remains unclear how nitrogen (N) deposition affects this effect. Here, we conducted a 12-month field experiment involving reciprocal litter translocation, with four N addition levels (0, 30, 50 and 100 kg N ha^–1 year^–1), in Quercus acutissima and Pinus massoniana stands in a subtropical urban forest and a mini meta-analysis of the effect of N addition on HFA in global forests. Results showed that a significant positive HFA of 3.98 was observed, but N addition (particularly with high rates) tended to weaken it, by decreasing it to –6.12 at 100 kg N ha^–1 year^–1. Conversely, the combined HFA index increased with bacterial community dissimilarity between stands. Such relationships were due to that N addition changed soil bacterial diversity and composition in P. massoniana stand and led to the convergence of bacterial communities between stands, mainly by lowering soil pH. Similarly, our experiment and meta-analysis showed that the species-level HFA index and its response to N addition also decreased with N addition rate (P < 0.05), suggesting the generalization of how N addition affects HFA. Our findings suggest that soil acidification driven by N inputs indirectly reduces HFA through altering bacterial communities, which helps to better understand and predict the dynamics of HFA effect and the cycling of carbon and nitrogen in forest ecosystems under N deposition.

Artificial light at night (ALAN) is an emerging component of global change and may increase the risk of plant invasion. However, the effects of different intensities of ALAN on the growth of invasive and native plants remain unclear. We conducted a controlled experiment in which five pairs of invasive and native plants from different families were grown separately under three light regimes: ambient light, low ALAN, and high ALAN. Our study showed that the total biomass of both invasive and native plants increased significantly under low-intensity ALAN. However, under high-intensity ALAN, the total biomass of invasive plants increased significantly, whereas the biomass of native plants significantly decreased. These findings indicate that invasive plants can better utilize light energy and have more effective photosynthetic responses under ALAN, while the photosynthesis of native plants is inhibited. The leaf dry matter content and leaf nitrogen content of invasive plants were significantly higher than those of native plants under ALAN, which significantly improved the photosynthetic nitrogen use efficiency of invasive plants. This indicates that invasive plants have stronger phenotypic plasticity and nitrogen-distribution strategy under ALAN. In summary, the enhanced physiological response of invasive plants under different intensities of ALAN may contribute to their continued spread and dominance in the ecosystem.

Plant community composition influences soil carbon (C) storage and stability in coastal wetlands, but such effects remain unclear in the non-growing season. In this study, the soil C content, density and stability were examined across five coastal plant communities—Spartina alterniflora, Suaeda salsa, Phragmites australis, mixed S. alterniflora–S. salsa communities and bare flat in the non-growing season in Yancheng, Jiangsu Province, China. The S. alterniflora community exhibited elevated soil organic and inorganic C contents, owing to its high biomass, strong C retention capacity. The P. australis community showed higher dissolved organic C and microbial biomass C contents, possibly driven by increased soil moisture and inorganic nitrogen (N) that promote microbial decomposition of plant residue. The S. salsa community had the lowest soil organic C density due to its low aboveground biomass, soil moisture and inorganic N and jointly microbial effects. The highest soil inorganic C density was observed in bare flat, which was associated with its high soil moisture. The lowest resistance index of C in P. australis community was associated with low electric conductivity, high C and N availability and bacterial effects. Soil C fractions, densities and resistance index of C decreased with soil depth, likely reflecting reduced water and N availability that constrain root and microbial activities. The results suggest that the S. alterniflora community enhances soil C accumulation, while P. australis community accumulate more labile C fractions, evoking low C stability due to interaction between soil physicochemical and microbial properties.

Large flowers are typically more conspicuous to pollinators and are associated with more nutritional rewards than small flowers. Flower size can also determine flower temperature, which can be a reward for flower-visiting insects. Nectarless Royal irises offer overnight shelter and morning warmth to male Eucera bee pollinators. A dark spot on their lower tepals (black patch) may act as a visual cue and contribute to flower heating. Here, we examined the association of floral display (i.e. flower size and black patch size) with flower temperature and female fitness (i.e. seed set) in Royal irises populations across an aridity gradient. First, we tested pollinator preference using artificial flowers of varying sizes. Next, we assessed associations between flower and black patch size, flower warming rate and female fitness. Finally, we manipulated flowers in the field to determine if the black patch influences heating. Pollinators preferred larger artificial flowers for overnight shelter. However, selection for larger flowers was found only in the population with the smallest flowers. No association was found between flower/black patch size and floral heating in natural populations, and the black patch did not affect flower heating. Flowers reached the temperature threshold for bees to start flying (18 °C) 10–35 min faster than ambient air. We conclude that the large flower size in the Royal irises serves as a visual signal, advertising for night shelter and flowers—independently of the size—heat up faster in the morning than ambient air. Flowers thus, potentially offer a ‘head start’ to the flower-dwelling bees, allowing them to warm up more quickly than in outside conditions.

Plant traits are influenced by evolutionary and environmental factors co-operating across varying spatial and temporal scales. While significant progress has been made in understanding aboveground-belowground trait relationships in terrestrial plants, little is known about how plant above- and belowground traits perform in marsh wetlands at large scales, particularly for traits related to clonal architecture and resource acquisition strategies. We measured above- and belowground traits of 15 occurring, common clonal plant species in nine marsh wetlands in northern China, and obtained data of soil physicochemical properties and climates. We found a crucial role of soil moisture in shaping traits of wetland clonal plants. Across the nine wetlands, all traits except those of leaves showed higher values in the high- than in the low-moisture areas in the low-precipitation areas, but this trend was reversed in the high-precipitation areas. In particular, clonal plants showed longer rhizome internodes and higher rhizome internode biomass in the higher-moisture areas, thereby displaying a guerrilla architecture. Moreover, most wetland clonal plants also exhibited larger specific leaf area, showing an acquisitive strategy of resource uptake. These findings deepen our understanding of the ecological strategies of wetland clonal species, and provide insights for the conservation and restoration of marsh wetland vegetation.