You are viewing the site in preview mode

Skip to main content
Download PDF

Phylogenetic relatedness and plant traits influenced flowering phenology change patterns in natural habitats in China (2003–2021)

BMC Plant Biology volume 25, Article number: 654 (2025) Cite this article

Abstract

Background

Phenology research has provided important insights on the influence of climate change on ecosystems. Investigation of spatial and interspecific difference can help us to better understand the phenology change pattern. In this study, observational data for 190 species collected from 2003 to 2021 at eight ecological stations in China were assessed via linear regression to detect trends in first flowering date (FFD), air temperature, and precipitation. We then examined the relationship between FFD change patterns, air temperature and precipitation through redundancy analysis, calculated the relative importance of phylogenetic relatedness, climate change, site conditions and plant traits in explaining variations in FFD change intensity using boosted regression tree method.

Results

We found that (1) FFDs of nearly 40% of the observed species changed significantly (p < 0.05), with species showing advanced and delayed FFDs accounting for half. (2) Air temperature increased at most stations, particularly in spring and summer, while precipitation decreased in humid and subhumid temperate zones and increased during most seasons in arid temperate and subtropical zones. (3) Spatial differences were observed in FFD trends. At stations in Northeast, North, and Southwest China, which are regions with increased temperature, the percentage of species with advanced FFD was higher than that of species with delayed FFD, with the mean trend ranging from − 2.4 to − 6.5 d decade− 1. Conversely, at stations distributed in Northwest and South China, which are regions with increased precipitation, the percentage of species with advanced FFD was lower than that of species with delayed FFD, with the mean trend ranging from 1.3 to 7.1 d decade− 1. (4) Air temperature and precipitation had a stronger influence on FFD change in the temperate zone than in the subtropical zone. Climate factors with the greatest influence on FFD change patterns varied with the observation site. Interspecific variations in FFD change intensity were mostly explained by phylogenetic relatedness, although plant traits, site conditions, and climate change also had a certain effect.

Conclusions

Our research found that the first flowering phenology of large percent of the observed plants changed significantly from 2003 to 2021, showing spatial and interspecific differences across observation sites. Our research also demonstrated the importance of plant phylogeny on interspecific differences in phenological changes, plant traits such as growth form, plant height, and flowering time influence flowering phenology to a certain extent. These findings will help us to better understand phenological responses to climate change on a national scale, and help us better predict the response of various plants to climate change in the future.

Peer Review reports

Introduction

The flowering period is an important stage in the plant life cycle. Changes in flowering phenology affect plant reproduction and evolution, interspecific competition, symbiosis in plant communities, and overall ecosystem stability [1]. Numerous studies have been conducted on the changes in flowering phenology in the context of climate change. These studies indicated that the first flowering dates (FFDs) of numerous plants have significantly advanced at several sites in America and Europe [2,3,4,5,6,7,8]. In contrast, researchers have also found that FFDs have not advanced dramatically, with several species even showing delayed FFDs at the same sites [9,10,11,12]. In China, the FFDs of numerous plants have advanced in many regions over the past 30–60 years [13–14]). However, the results have varied, with the FFD of some species not being significantly advanced while that of others showing a delay [15–16]. Therefore, the trends in FFD change not only differ between regions but also between species within a region [17]. Plant traits such as growth form, woodiness, plant height, longevity, and specific leaf area reflect the response and adaptation of plants to environmental changes [18]. They are closely correlated with the phenological response of the plant to climate change [19]. Plants with different functional traits exhibit varied phenological responses to environmental changes [20–21].

In China, Ge et al. (2015) analyzed the differences in phenological change between growth forms using observational data on 104 plant species collected from 145 sites over the period 1960–2011 [14]. However, their study did not include plant traits such as plant height, flowering time, and woodiness when examining the influence of plant traits on phenological change. In addition to plant traits, previous studies have shown that phenology is phylogenetically conserved, with closely related species tending to have similar phenological rhythms [22,23,24]. Hence, plant species cannot be considered statistically independent but rather are related via phylogeny; therefore, phylogenetic relatedness should be included when examining patterns of phenological change.

Phenological change research relies on long-term observational data. Because of traffic and manpower limitations, many phenological observation sites are located in parks, botanical gardens, or other areas in or near cities. In urban areas, the “heat island effect” can contribute more than 20% to climate warming and has a direct effect on plant phenology [25,26,27,28]. This may interfere with our understanding of the changes in plant phenology in the context of global warming. In addition, phenological observation objects often include cultivated plants, such as Syzygium aromaticum, Salix matsudana, Ulmus pumila, Populus simonii, and Fraxinus chinensis while fewer plants from natural habitats have been selected for observation [29–30]. This is not conducive to understanding phenological changes in natural ecosystems. Therefore, more attention should be paid to the phenological trends of plants in their natural habitats.

In this study, based on the FFD observation data of plants with different growth forms collected from 2003 to 2021 by eight ecological stations affiliated with the Chinese Ecosystem Research Network, we calculated the FFD change intensity (number of days/year) of each species using a linear regression model and then assessed the general trend of FFD change at the country and station scales. Furthermore, we investigated the contributions of phylogenetic relatedness, plant traits, changes in climate, and site conditions to the variations in observed FFD changes using boosted regression tree analysis (BRT) according to the method presented by König et al. (2018) [19]. This study aimed to examine how phylogeny, environmental background, climate change, and plant traits influence spatial and interspecific differences in FFD trends. Specifically, we focused on the following questions: Has flowering phenology changed in plants in natural habitats over the past 20 years (2003–2021)? Have clear spatial and interspecific differences in flowering phenology changes occurred? To what extent do phylogenetic relatedness, plant functional traits, site conditions, and climate change influence spatial and interspecific variations in flowering phenology?

This study aimed to provide scientific support for understanding regional and interspecific differences in phenological changes in large regions. The outcomes of this study will improve our understanding of the mechanisms underlying phenological changes and the influence of climate change on natural ecosystems.

Data and methods

Data sources

Phenology observation data

To monitor changes in China’s ecosystem, the Chinese Academy of Sciences established the Chinese Ecosystem Research Network (CERN) in 1998. All CERN ecological stations are located within a typical ecosystem in a specified region. The CERN stations conduct long-term observations of the ecosystem in accordance with unified monitoring protocols and observation indicators. Phenology is an important component of CERN monitoring, and includes multiple phenological periods. Phenological data were collected from eight CERN ecological stations across various ecological zones in China (Fig. 1). These stations include: Changbai Mountain Forest Ecosystem Research Station (CBF) in a humid middle-temperate zone; Beijing Forest Ecological Station (BJF) in a subhumid warm-temperate zone; Shapotou Desert Research and Experiment Station (SPD) in an arid middle-temperate zone; Maoxian Mountain Ecosystem Research Station (MXF) in the northern region of a humid middle-subtropical zone; Huitong Research Station of Forest Ecosystem (HTF) and Ailaoshan Station for Subtropical Forest Ecosystem Studies (ALF) in southern regions of humid middle-subtropical and humid southern-subtropical zones respectively; Dinghu Mountain Research Station of Forest Ecosystem (DHF) and Heshan Mountain Integrated Experimental Station of Hilly Land (HSF) in eastern regions of a humid southern-subtropical zone. At each ecological station, typical species of different growth form (tree, shrub and herbs) were selected to perform the observation, particularly the dominant species in plant community. However, the number of selected species is not uniform for the varied species richness across ecological stations. Table 1 summarizes the altitude, annual mean temperature, annual precipitation, accumulated temperature > 10 ℃, and number of observed species for each station.

For trees and shrubs, three to five individual plants of each target species were selected for permanent observation, and the FFD was defined as the day on which a branch with multiple buds had at least one fully opened flower. For herbs and grasses, three to five permanent quadrats (1 m×1 m) were selected to perform phenological observations for each target species, and the FFD was defined as the day on which 10% of the observed plants had one or more flowers fully opened for the first time. Phenological observations were performed manually daily from the beginning to end of the growing season. During the flowering period, observations were made in both the morning and afternoon to ensure that flowering events were not omitted. For more detailed description of phenological observation methods see Wu et al. (2019) “Protocols for standard biological observation and measurement in terrestrial ecosystem [31].”

FFD data from 2003 to 2021 for 190 plant species were used in this study. Among these species, 66 were trees, 59 were shrubs, 54 were herbs, and 11 were grasses (with 10 annual plants). The FFD data are presented in Supplementary Table S1. As there are missing values in the ground phenological observation data, therefore, for a specified species, if the values of first flowering date missed in more than 20% of the observation years, the phenological sequence of this species was excluded from the dataset. We also made linear regression analysis between first flowering date and the year (details were presented in Sect. 2.2.1), and interpolated the missed values according to the fitted linear regression function. The interpolated dataset was used to make the Redundancy analysis in the following data analysis procedure.

Temperature and precipitation data

A weather station was installed at each ecological station within 1 km of the phenological observation site to collect air temperature and precipitation data. Air temperature data were recorded using an air temperature sensor (Vaisala series) at the automatic weather station, and precipitation data were collected through manual observations of the precipitation collection system. The air temperature and precipitation data are presented in the Supplementary Table S2.

Site conditions

Site condition-related variables were included in the analysis to test whether site conditions had an obvious influence on spatial differences in the FFD change. Variables of site conditions include altitude, annual mean temperature, annual precipitation, and accumulated temperature (> 10 ℃). The altitudes of the observation sites were measured using a handheld GPS. The annual mean temperature, annual precipitation, and accumulated temperature were the mean values calculated based on the meteorological observation data obtained over the last 20 years. The site condition data are presented in the Supplementary Table S3.

Plant traits

As proposed by König et al. (2016), plant traits known to be biologically relevant and potentially relevant for phenological reactions should be selected to make the analysis. Plant traits such as growth form, woodiness, annual/perennial, plant height, leaf distribution along the stem, leaf fresh area per leaf dry mass, leaf dry mass per leaf saturated mass, flowering time and pollination type were included in the research of König et al. (2016). In our research, limited by the data availability, plant traits of growth form (trees, shrubs, herbs, and grasses), woodiness(woody/non-woody), annual/perennial, plant height, and flowering time were included in the analysis to understand the interspecific differences in phenological responses to climatic changes. Growth form, woodiness, and annual/perennial data for each observed species were acquired from the Flora of China (online version, https://www.iplant.cn/frps). The plant height of observed species was measured at the flowering stage. For herb, grass and short shrubs, steel tape was used to measure the plant height. For trees and tall shrubs, measuring rod or laser ranger was used to measure the plant height. The flowering time was obtained from long-term phenological observation data. The plant traits data are presented in the Supplementary Table S4.

Fig. 1
figure 1

Locations of the eight ecological stations of the Chinese Ecosystem Research Network (CBF, Changbai Mountain Forest Ecosystem Research Station; BJF, Beijing Forest Ecological Station; SPD, Shapotou Desert Research and Experiment Station; MXF, Maoxian Mountain Ecosystem Research Station; HTF, Huitong Research Station of Forest Ecosystem; ALF, Ailaoshan Station for Subtropical Forest Ecosystem Studies; DHF, Dinghu Mountain Research Station of Forest Ecosystem; HSF, Heshan Mountain Integrated Experimental Station of Hilly Land)

Full size image
Table 1 Environmental factors and number of observed species at each ecological station
Full size table

Data analysis

Analysis of FFD trends

The original FFD data were recorded in the year-month-day format. For the convenience of data statistics, we converted the original date-type data into time sequence data of the FFD, expressed as the days of year (DOY) from January 1 to the day when the first flowering event happened. We analyzed the trend of FFD for each single species using a linear regression, with “year” as the independent variable and DOY as the dependent variable [2, 32]. A negative slope (slope < 0) of the linear regression indicated an advance of FFD, whereas a positive slope (slope > 0) indicated a delay of FFD. The linear regression analysis was carried out using the “lm” function in the statistical program R version R 3.1.1 [33].

Negative time sequences (slope < 0) with a p-value lower than 0.05 were defined as “advanced,” positive time sequences (slope > 0) with a p-value lower than 0.05 were defined as the “delayed,” and both “advanced” and “delayed” were defined as “changed.” The p-values, together with the slope values of the linear regression are listed in Table S5. Based on Table S5, we calculated the proportion of species with positive, significantly positive, negative, and significantly negative trends as well as the mean trend of all species (Table 2). Then, we plotted the frequency distribution of phenological trends of the first flowering date (Fig. 2).

At each ecological station, we calculated the percentage of species with changed, advanced, and delayed FFD (Fig. 3) and the mean trend of FFD for all target species at each ecological station (Fig. 4). Slopes of linear regression equation of DOY with “year” were multiplied by 10 as the trend of FFD in each ten years. The slope was defined as the FFD change intensity.

The relationships among FFD change intensity, plant traits, and site conditions were examined in this study. We used ANOVA to test the differences in FFD change intensity between growth forms and woodiness/non-woodiness. The results are presented in Fig. 5. Linear regression was performed to test the relationships between FFD change intensity and plant height and flowering time and between FFD intensity and site condition variables (altitude, annual mean temperature, annual precipitation, > 10 ℃ accumulated temperature). The results are presented in Table S6.

Analysis on the trends of temperature and precipitation change

We analyzed the trends of monthly mean air temperature and total precipitation in spring (March − May), summer (June − August), autumn (September − November) and winter (Prior December − February) using a linear regression with “year” as the independent variable. The R2 and p-values of the linear regressions are presented in Fig. S1 and Fig. S2. Slopes were used to represent the trends of seasonal temperature and precipitation changes and were included in the following BRT analysis. Beside the linear regression analysis, we used generalized additive model (GAM) to fit the trends of air temperature and precipitation at each ecological station. Twenty-eight GAM models were fitted respectively for variables of air temperature and precipitation. Parameters of fitted GAM models were presented in Table S7. Among these parameters, edf (effective degree of freedom) reflect the degree of nonlinearity. If the edf is higher than 1, this indicates the nonlinearity of the trend of air temperature and precipitation. The linear regression analysis was carried out using the “lm” function, and GAM was performed using the “mgcv” package, both analysis were executed in the statistical program R version R 3.1.1 [33].

Redundancy analysis between phenological data matrix and environmental data matrix

Redundancy analysis (RDA) was used to detect the relationships between phenological change trends, air temperature, and precipitation. RDA is a Principal Component Analysis (PCA) of the fitted value matrix of the multivariate multiple linear regression analysis between the response (phenology data matrix, DOY×species) and explanatory variables (air temperature and precipitation data matrix). The aim of RDA is to find a linear combination of a series of explanatory variables that can best explain the variation in the response variables [34]. Through RDA, we obtained the portion of the total variance of the response variable matrix that can be explained by the explanatory variables (portion constrained) (Table 3). We also obtained the phenology-environmental ordination plot and biplot scores for constraining variables (Table S8), which can be used to represent the role of environmental factors in explaining the variance of the response variables, that is, which variable has a greater influence on phenological changes. More details of the RDA can be found in Numerical Ecology with R (Second Edition) published by Bocard et al. (2018) [34].

Phylogenetic analysis

Because phylogenetic relatedness may exist among the studied species, we applied the following procedure to include phylogenetic information in the subsequent BRT analysis. We first used the R package “plantlist” to calibrate the Latin names of the genera and families of the studied species. Then, we created a phylogenetic tree of the studied species and linked the trait data to the species phylogeny using the function “phylo.maker” from the package V. PHYLOMAKER2 [35]. We adopted the same approach and procedure reported by Sporbert et al. (2022) [24] to account for the phylogeny based on the phylogenetic eigenvector regression by Diniz-Filho et al. (1998) [36]. This procedure was proposed by Pistón et al. (2019) [37] and Bianchini and Morrissey (2020) [38]. We first used the function “vcv.phylo” to compute a pairwise distance matrix (Table S9 is the matrix for species with advanced FFD sequence, and Table S10 is the matrix for species with delayed FFD sequence) from our phylogenetic tree, and then we performed a principal coordinates analysis (PCoA) to extract eigenvectors from this distance matrix using the function “pcoa” from the APE package v.5.8-1 [39]. The phylogenetic relationships among species were well represented by phylogenetic eigenvectors, which can be used to control for phylogenetic autocorrelation when a sufficient number of eigenvectors are included in the analysis [36, 38]. According to the research of Sporbert et al. (2021), the eigenvectors selected into the BRT analysis could explain more than 90% of the variation in the phylogenetic distance matrix. In our BRT models, the covariates for advanced FFD sequences included the first 28 eigenvectors of a total of 99 eigenvectors (Table S11), while those for delayed FFD sequences included the first 31 eigenvectors of a total of 95 eigenvectors (Table S12), both of which explained over 90% of the phylogenetic structure in the distance matrix (see Table S13, Table S14). Besides the phylogenetic relatedness among species, phylogenetic relatedness at genus level was also included in the BRT analysis to explain the inter-specific variation of flowering phenology. The phylogenetic relatedness among genus, represented by phylogenetic vectors, was also achieved using the procedure proposed by Pistón et al. (2019) and Bianchini and Morrissey (2020). The phylogenetic distance matrix at genus level for species with advanced FFD sequence was presented in Table S15, and the phylogenetic distance matrix at genus level for species with delayed FFD sequence was presented in Table S16. The number of selected phylogenetic vectors was also determined according to the proposition of Sporbert et al. (2021). For advanced FFD sequences, the first 63 eigenvectors of a total of 76 eigenvectors (Table S17) were included in the BRT model fitting, and the first 33 eigenvectors of a total of 82 eigenvectors (Table S18) were included in the BRT model fitting. Both of which explained over 90% of the phylogenetic structure in the distance matrix (see Table S19, Table S20).

Analysis of the relative importance of phylogenetic relatedness, plant traits, site conditions, and climate change on variations of FFD change intensity

We used BRT to fit the relationships between FFD change intensity, site conditions, climate change, plant traits, and phylogenetic eigenvectors. The BRT models were fitted using the “gbm” [40] and “dismo” [41] of R library. All calculations were performed using the statistical program R 3.1.1 [33]. We built two BRT models: one for FFD sequences with negative slopes and the other for FFD sequences with positive slopes. As proposed by König et al. (2016) and Sporbert et al. (2021), we set bag fraction as 0.5, and tree complexity as 2 in the procedure of BRT model fitting. Considering the small number of training data points, we selected a smaller value of learning rate (0.001) to fit the BRT according to the recommendation of König et al. (2016). In the procedure of model fitting, following the recommendation of Elith et al. (2008), we conducted a cross-validation procedure (cv.folds = 5) to optimize the number of trees with the chosen learning rate using the gbm.step function. Considering the high number of variables included in the fitting of BRT models, we used a step-wise procedure for model simplification (gbm.simplify) that reduces the number of predictor variables without substantial gains in the cross-validation error (cv.folds = 5). For the simplified (most parsimonious) models, the optimal number of trees was estimated again using the gbm.step function. Parameters such as the mean total deviance, mean residual deviance, training data correlation were used to assess the performance of the fitted BRT models (Table S21). The relative importance of site conditions, plant traits, changes in climate, and phylogenetic relatedness for the variation in FFD change intensity was deduced from boosted regression tree models (Fig. 6). For more detailed description of parameters and settings on BRT model see “A working guide to boosted regression trees” by Elith et al. (2008) [42].

Results

General trend of the first flowering phenology

Based on the p-values of the linear regression between the FFD time sequences and year, 39 FFD time sequences showed a significant advanced trend over the past 19 years (p < 0.05) while 37 FFD time sequences showed a significant delayed trend (p < 0.05), with the mean trend of FFD for all time sequences at -0.22 ± 13.6 d decade− 1 (Table 2). The FFD trend was less than 15 d decade− 1 for 80% of these phenological sequences (Fig. 2).

Table 2 Summary of phenological trends
Full size table

Trmean is the mean trend of all the time sequences; Negall and Posall: proportion of negative and positive trends, respectively; Negsig and Possig: proportion of significant negative and positive trends (p < 0.05), respectively.

Fig. 2
figure 2

Frequency distribution of phenological trends of first flowering date (FFD). Data include 197 data points on 190 species from eight sites across China. Negative values indicate advancing trends, whereas positive values indicate delaying trends

Full size image

Spatial differences in the trends of the first flowering phenology

The percentage of species with a changed FFD showed large variations between the ecological stations (Fig. 3). The percentages of species with advanced and delayed FFD also showed large regional differences (Fig. 3).

The percentage of species with an advanced FFD was higher than that of species with a delayed FFD in ALF, BJF, MXF, and CBF (Fig. 3), and the mean trends in FFD were − 6.4, − 6.5, − 3.3, and − 2.4 d decade− 1, respectively (Fig. 4). In DHF + HSF, HTF, and SPD, the percentage of species with a delayed FFD was higher than that of species with an advanced FFD (Fig. 3), and the mean trends in FFD were 7.1, 4.2, and 1.3 d decade− 1, respectively (Fig. 4).

Fig. 3
figure 3

Percentage of species with changed, advanced, and delayed first flowering date in different ecological stations (CBF, Changbai Mountain Forest Ecosystem Research Station; BJF, Beijing Forest Ecological Station; SPD, Shapotou Desert Research and Experiment Station; MXF, Maoxian Mountain Ecosystem Research Station; HTF, Huitong Research Station of Forest Ecosystem; ALF, Ailaoshan Station for Subtropical Forest Ecosystem Studies; DHF, Dinghu Mountain Research Station of Forest Ecosystem; HSF, Heshan Mountain Integrated Experimental Station of Hilly Land; N indicates the number of phenological sequences.)

Full size image
Fig. 4
figure 4

Mean trends (mean ± SD) in first flowering date across ecological stations (CBF, Changbai Mountain Forest Ecosystem Research Station; BJF, Beijing Forest Ecological Station; SPD, Shapotou Desert Research and Experiment Station; MXF, Maoxian Mountain Ecosystem Research Station; HTF, Huitong Research Station of Forest Ecosystem; ALF, Ailaoshan Station for Subtropical Forest Ecosystem Studies; DHF, Dinghu Mountain Research Station of Forest Ecosystem; HSF, Heshan Mountain Integrated Experimental Station of Hilly Land)

Full size image

Differences in the trends of first flowering phenology between plant trait groups

Among species with advanced FFD, the FFD advance intensity was higher in herbaceous plants than in trees and shrubs (Fig. 5), whereas for FFD delay intensity, no significant differences were observed between growth forms (Fig. 5).

Strong positive linear relationships were observed between FFD change intensity (both advance and delay) and plant height for herbs and grass (p < 0.1, Table S6), whereas for trees and shrubs, no clear linear relationships were observed between FFD change intensity and plant height (Table S6).

The FFD delay intensity increased significantly with later flowering time (p < 0.05) in shrubs, whereas the FFD advance intensity decreased with later flowering time in herbs and grasses (p < 0.1). No clear relationships were observed between FFD change intensity and flowering time in trees (Table S6).

Fig. 5
figure 5

Differences in FFD change intensity between growth forms. A1 indicates the difference between growth forms in FFD advance intensity, A2 indicates the difference between woody and non-woody plants in FFD advance intensity; B1 indicates the difference between growth forms in FFD delay intensity, B2 indicates the difference between woody and non-woody plants in FFD delay intensity. FFD indicates first flowering date. Different letters (a, b) indicate significant difference between plant trait groups. N indicates the number of phenological sequences in plant traits group

Full size image

Relationship between flowering phenology change, air temperature and precipitation

In the temperate zone, the mean air temperatures at CBF, BJF and SPD increased in most seasons over the period to 2003–2021 (Fig. S1), particularly in spring and summer. In the subtropical zone, the mean air temperatures during most seasons at MXF, ALF, HSF, and DHF also increased over the past 19 years. In contrast, the mean air temperature at HTF showed a slight decreasing trend in most seasons, except spring, over the past 19 years (Fig. S1). From 2003 to 2021, the total precipitation decreased in most seasons in CBF and BJF. However, the total precipitation during each season increased at SPD (Fig. S2). At MXF and HTF, the total precipitation increased in most seasons. At ALF, the total precipitation decreased in most seasons, except winter. At HSF and DHF, the total precipitation decreased in spring and summer but increased in autumn and winter (Fig. S2).

The RDA analysis (Table 3) revealed that air temperature and precipitation have great explanatory power for phenological change patterns. In addition, this power varied greatly at different ecological stations and was greater in temperate zones than subtropical zones. From the biplot scores for the constraining variables (Table S8), the variables with the greatest influence on phenological change patterns varied with the observation site. In the temperate zone, monthly mean air temperatures in spring and winter were the most important variables in CBF and BJF, whereas total precipitation in spring and summer were the most important in SPD. In the humid middle subtropical zone, the monthly mean air temperature in spring and total precipitation in summer were the most influential variables in MXF, total precipitation in summer and autumn were the most important variables in HTF, and monthly mean air temperatures in spring and autumn were the most important variables in ALF and DHF + HSF.

Table 3 Proportion of explained variance in the redundancy analysis between phenological change pattern and temperature and precipitation variables at each observation site
Full size table

Relative importance of site conditions, climate change, plant traits and phylogenetic relatedness on FFD change intensity

The pie charts in Fig. 6 present the overall contributions of the variables grouped by “plant traits,” “site conditions,” “changes in climate,” and “phylogeny” deduced from the BRT models. “Phylogeny” was the factor with the greatest importance on variations in FFD change intensity in the BRT models for both advanced and delayed FFD sequences. Growthform and plant height are the most important plant traits on variations in FFD change intensity in the BRT models.

Fig. 6
figure 6

Relative importance (%) of plant traits, site conditions, climate change, and phylogeny on FFD advance intensity, deduced from boosted regression trees. Models were fitted separately for (a) advanced phenological sequences (the chart on the left) [n = 102] and (b) delayed phenological sequences (the chart on the right) [n = 95]. Pie charts represent overall importance of the four variables categories. Temp_accu, accumulated temperature; Temp_annual, annual mean air temperature; Precip_annual, mean annual precipitation; Precipchange_spring, change trend of total precipitation in spring; Precipchange_summer, change trend of total precipitation in summer; Precipchange_autumn, change trend of total precipitation in autumn; Precipchange_winter, change trend of total precipitation in winter; Tempchange_spring, change trend of monthly mean air temperature in spring; Tempchange_summer, change trend of monthly mean air temperature in summer; Tempchange_autumn, change trend of monthly mean air temperature in autumn; Tempchange_winter, monthly mean air temperature in winter

Full size image

Discussion

In this study, the proportion of significant phenological sequences in our study was similar to that of Ge et al. (2015) [14], who found that 43% of spring and summer phenological events changed significantly between 1960 and 2011 based on 1263 phenological observations of 112 species collected across 145 locations in China. However, the observation period in our study was much shorter than that of Ge et al. (2015) [14]. This indicated that obvious phenological changes occurred from 2003 to 2021. Furthermore, the proportion of significantly delayed phenological sequences (18.8%) in our study was much higher than that reported by Ge et al. (2015) (1.2%) [14]. This might be because the numbers of phenological records in temperate zones collected by Ge et al. [14] were much larger than those in the subtropical zone and more species had advanced spring and summer phenological events in temperate zones [43]. This discrepancy may have led to a statistical bias in the proportion of significantly delayed and advanced phenological events. Therefore, to fully understand trends in phenological events at large scale, more attention should be paid to the collection of phenological observation records in subtropical zones.

In RDA analysis, the variance explained by air temperature and precipitation in the temperate zone was higher than that in the subtropical zone. This indicates that air temperature and precipitation have stronger influences on phenology in temperate zones, particularly in regions with increasing temperatures [44]. In humid and subhumid middle temperate zones, the monthly mean air temperature in spring and winter had the most important influence on the phenological change pattern of CBF and BJF. This was consistent with the conclusion proposed by Mo et al. (2011) [45], who stated that temperature has the greatest influence on phenology in temperate and cold regions. SPD is located in an arid middle temperate zone and experiences extreme drought; thus, water is the most important factor controlling plant growth [46]. This explains why precipitation in spring and summer had the most significant influence on the phenological change patterns in SPD. This conclusion is consistent with the research of Crimmins et al. (2010) [47], who considered precipitation to be the most important variable explaining flowering phenology changes in dry regions. The relationships among phenological change patterns and air temperature and precipitation were more complex in the subtropical zone than in the temperate zone. MXF and HTF are located in the humid middle subtropical zone, and due to the influence of monsoon, drought often occurs in late spring and early summer in Southwest China before the monsoon arrives. Thus, precipitation is the most important factor affecting flowering phenology. In tropical and near-tropical regions, temperature and photoperiod vary little throughout the year, plants have high heat requirements, and changes in plant phenology are mainly caused by temperature change [45]. This might explain why the monthly mean air temperature in spring was the factor with the greatest influence on flowering phenological changes in DHF and HSF. Although ALF is located in the southern subtropical zone, temperature is the most important variable controlling phenological changes because of the high altitude.

Flowering phenology is phylogenetically conserved, closely related species tending to flower or leaf out at similar times [22,23,24]. In our research, phylogenetic relatedness had the most important influence on FFD change intensity compared with plant traits, site conditions, and climate change trends. This indicated the change trend of flowering phenology is phylogenetically conserved, closely phylogenetic related species show similar phenological response to climate change. One possible mechanism for the conservation of flowering phenology change might be that related species generally have similar morphological traits, such as flower size, shape, aroma and nectar content, which determine the success of pollination [48], so closely phylogenetic related species tend to have similar phenological response to climate change and blossom at the same time to attract pollinators [49]. Another possible mechanism might be that closely phylogenetic related species have the almost same genetic basis, possess similar evolutionary direction, and easier to generate approximate phenological response to climate change [50]. This finding highlights the importance of phylogenetic relatedness, which not only influences the timing of phenological events but also the intensity of phenological changes. Phylogenetic relatedness should be considered as an important factor in the analysis of phenological patterns. However, our conclusion differs from that of König et al. (2018) [19], who concluded that phylogeny was not important for changes in phenology. This difference might be related to the distinct geographical scale and constructed phylogenetic tree between our study and that of König et al. (2018) [19]. In the research of König et al. (2018), the phenological observation data was collected at a larger geographical scale (Europe and North America) compared with that of our research, geographical isolation and divergent evolution may lead to more distant phylogenetic relatedness among species. In the research of König et al. (2018), the phylogenetic information was extracted from the Daphne phylogenetic tree [51], which is based on a backbone family phylogeny of the Angiosperm Phylogeny Group III. Whereas in our research, the phylogenetic tree was constructed based on the combination of GBOTB for seed plants [52] and the clade in Zanne et al.’s (2014) [53] phylogeny for pteridophytes. This may lead to the difference of the relative importance of phylogenetic information in explaining the interspecific difference of flowering phenology change trend between our research and that of König et al. (2018).

Plant height is closely related to the competitive ability of an ecosystem [54]. To enhance their competitive ability, small grasses and herbs may show stronger phenological responses to environmental changes than larger plants [19]. Previous studies have demonstrated that plant height is positively correlated with the onset of flowering [22, 24, 55–56]. In this study, we found that plant height and FFD change intensity exhibited different relationships between growth forms. Plant height and FFD change intensity were positively correlated for herbaceous plants but did not show a clear relationship for trees and shrubs. In general, shorter species may respond faster to environmental changes [57], and they are more likely to flower earlier than taller ones in herbaceous plants [55]. Plant height or size should be included in the analysis of interspecific differences in phenological response to climate change because this trait greatly influences species interactions, survival, and reproduction [58].

Growth form has an important influence on interspecific variation in phenological changes [14, 19]. In our research, FFD advance intensity varied significantly with the growth form, with herbaceous plants having a significantly higher FFD advance intensity than trees and shrubs. This finding is consistent with that of Ge et al. (2015) [14] but inconsistent with that of König et al. (2018) [19]. This discrepancy might be related with the bias in the number of observed species with different growth form in the phenological observation dataset, and the diverse trends of climate change at local scale. In addition, we found that the FFD delay intensity was not significantly different between the growth forms. The exact reason for this deserves further investigation.

The intensity of FFD advances in early flowering plants is usually higher than that in late-flowering plants [59, 60]. König et al. (2018) also found that early flowering grasses showed more intense advances than late-flowering ones [19]. In this study, we also found a clear negative relationship between FFD advance intensity and flowering time for herbaceous plants, which is consistent with the conclusions of Dunne et al. (2003) [59] and König et al. (2018) [19]. For shrubs, a significant positive relationship was detected between FFD delay intensity and flowering time. We speculate that the increased temperature and precipitation in late spring and early summer may produce unsuitable environments for blossoming, thus strengthening the intensity of the FFD delay.

Regarding longevity (annual/perennial), Fitter and Fitter (2002) [61] found that annual plants were more likely to flower earlier than congeneric perennials as temperature increased. In our study, the relative importance of longevity was very low in the BRT analysis, which may have been due to the limited number of observed annual plants.

In this study, site conditions and changes in climate contributed to the variation in FFD change intensity; however, the relative importance of site conditions and climate change was much lower than that reported by König et al. (2018) [19]. An interesting finding in our research is that site conditions have a more important influence on FFD delay intensity than on FFD advance intensity. The linear regression analysis also revealed stronger correlations between the FFD delay intensity and site conditions than between the FFD advance intensity and site conditions. Furthermore, the intensity of FFD delay decreased with the increase of altitude for trees and shrubs, whereas for herbs and grass, the intensity of FFD delay increased with the increase of altitude. We speculated that the harsh environmental conditions at high altitudes lead to a short growth season, which does not allow trees and shrubs to delay the flowering event; otherwise, they would not be able to complete their life cycle. In contrast, herbs grow fast, are more plastic in terms of the timing of their internal physiological and chemical dynamics, and may respond faster to environmental changes [57]. Therefore, herbs may be more likely to show a stronger FFD delay intensity than trees and shrubs.

In general, phylogenetic relatedness and differences in biological and ecological characteristics among species lead to variation in flowering phenology in response to climate change. Furthermore, environmental factors such as climate, soil, and topography have distinct effects on plant growth in different ecological zones. All these factors lead to interspecific and regional differences in phenological responses to climate change [62].

At last, we should be aware that in this study, very limited phenological observation data from the Northwest and Southwest China (particularly the Tibetan Plateau) was included in the analysis, and the number of species observed at different ecological stations varied considerably, which may bring uncertainty to this research.

Conclusion

Our research found that the first flowering phenology of nearly 40% of the observed plants changed significantly from 2003 to 2021, FFD showed spatial and interspecific differences, and flowering phenology trends varied across observation sites. Our research also demonstrates the importance of plant phylogeny on interspecific differences in phenological changes. In addition to phylogenetic relatedness, plant traits such as growth form, plant height, and flowering time influence flowering phenology to a certain extent. These findings will help us to better understand phenological responses to climate change on a national scale. The conclusion about the relative importance of phylogenetic relatedness and plant traits on interspecific differences in phenological change will help us better predict the response of various plants to climate change in the future. Phylogenetic relatedness, plant traits, trends in climate change, and site conditions should all be considered as important factors when analyzing the mechanisms and predicting the patterns of plant phenology change. Our research also found that the phenological change patterns in the subtropical zone were more complex than those in the temperate zone. Thus, to fully understand phenological patterns in China, additional phenological observations should be performed in subtropical zones.

Data availability

Data is provided as supplementary information files.

References

  1. Matthews ER, Mazer SJ. Historical changes in flowering phenology are governed by temperature × precipitation interactions in a widespread perennial herb in Western North America. New Phytol. 2016;210(1):157–67.

    Article  PubMed  Google Scholar 

  2. Bradley NL, Leopold AC, Ross J, Wellington H. Phenological changes reflect climate change in Wisconsin. Volume 96. PNAS; 1999. pp. 9701–4.

  3. Abu-Asab MS, Peterson PM, Shetler SG, Orli SS. Earlier plant flowering in spring as a response to global warming in the Washington, DC, area. Biodivers Conserv. 2001;10:597–612.

    Article  Google Scholar 

  4. Bradley AV, Gerard FF, Barbier N, Weedon GP, Anderson LO, Huntingford C, Aragão LEOC, Zelazowski P, Arai E. Relationships between phenology, radiation and precipitation in the Amazon region. Global Change Biol. 2011;17:2245–60.

    Article  Google Scholar 

  5. Gonsamo A, Chen JM, Wu C. Citizen science: linking the recent rapid advances of plant flowering in Canada with climate variability. Sci Rep. 2013;3:2239.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Szabó B, Enik V, Czúcz B. Flowering phenological changes in relation to climate change in Hungary. Int J Biometeorol. 2016;60:1347–56.

    Article  PubMed  Google Scholar 

  7. Menzel A, Ye Y, Matiu M, Sparks T, Scheifinger H, Gehrig R, Estrella N. Climate change fingerprints in recent European plant phenology. Global Change Biol. 2020;26(4):2599–612.

    Article  Google Scholar 

  8. Büntgen U, Piermattei A, Krusic PJ, Esper J, Sparks T, Crivellaro A. Plants in the UK flower a month earlier under recent warming. P Roy Soc B-Biol Sci. 2022;289:20212456.

    Google Scholar 

  9. Ahas R, Aasa A, Menzel VG, Fedotova H, Scheifinger H. Changes in European spring phenology. Int J Climatol. 2002;22(14):1727–38.

    Article  Google Scholar 

  10. Rosenzweig C, Casassa G, Karoly DJ, Imeson A, Liu C, Menzel A, Rawlins S, Root TL, Seguin B, Tryjanowski P. Assessment of observed changes and responses in natural and managed systems. In: Parry ML, Canziani OF, Palutikof JP, van der Linden PJ, Hanson CE, editors. Climate change 2007: impacts, adaptation and vulnerability. Contribution of working group II to the fourth assessment report of the intergovernmental panel on climate change. Cambridge: Cambridge University Press; 2007. pp. 79–131.

    Google Scholar 

  11. Neil KL, Landrum L, Wu JG. Effects of urbanization on flowering phenology in the metropolitan phoenix region of USA: findings from herbarium records. J Arid Environ. 2010;74(4):440–4.

    Article  Google Scholar 

  12. Templ B, Templ M, Filzmoser P, Lehoczky A, Baksiene E, Fleck S, Gregow H, Hodzic S, Kalvane G, Kubin E, Palm V, Romanovskaja D, Vucetic V, Zust A, Czucz B. NS-Pheno team. Phenological patterns of flowering across biogeographical regions of Europe. Int J Biometeorol. 2017;61:1347–58.

    Article  PubMed  Google Scholar 

  13. Dai JH, Wang HJ, Ge QS. Multiple phenological responses to climate change among 42 plant species in Xi’an, China. Int J Biometeorol. 2013;57:749–58.

    Article  PubMed  Google Scholar 

  14. Ge QS, Wang HJ, Rutishauser T, Dai JH. Phenological response to climate change in China: a meta-analysis. Global Change Biol. 2015;21(1):265–74.

    Article  Google Scholar 

  15. Bai J, Ge QS, Dai JH. The response of first flowering dates to abrupt climate change in Beijing. Adv Atmos Sci. 2011;28(3):564–72.

    Article  Google Scholar 

  16. Wang L, Zhang Q. Analysis of phytogeographic characteristics of typical alpine grassland steppe in Qinghai-Tibetan plateau recently 20 years. Plateau Meteorol. 2018;37(6):1528–34.

    Google Scholar 

  17. Parmesan C, Yohe G. A globally coherent fingerprint of climate change impacts across natural systems. Nature. 2003;421:37–42.

    Article  CAS  PubMed  Google Scholar 

  18. Lauterbach D, Römermann C, Jeltsch F, Ristow M. Factors driving plant rarity in dry grasslands on different Spatial scales: A functional trait approach. Biodivers Conserv. 2013;22:2337–52.

    Article  Google Scholar 

  19. König P, Tautenhahn S, Cornelissen JHC, Kattge J, Bönisch G, Römermann C. Advances in flowering phenology across the Northern hemisphere are explained by functional traits. Global Ecol Biogeogr. 2018;27:310–21.

    Article  Google Scholar 

  20. Cleland EE, Chiariello NR, Loarie SR, Mooney HA, Field CB. Diverse responses of phenology to global changes in a grassland ecosystem. Volume 103. PNAS; 2006. pp. 13740–4.

  21. Morin X, Lechowicz MJ, Augspurger C, O’keefe J, Viner D, Chuine I. Leaf phenology in 22 North American tree species during the 21st century. Global Change Biol. 2009;15:961–75.

    Article  Google Scholar 

  22. Bolmgren K, Cowan PD. Time- size tradeoffs: a phylogenetic comparative study of flowering time, plant height and seed mass in a North temperate flora. Oikos. 2008;117:424–9.

    Article  Google Scholar 

  23. Davies TJ, Wolkovich EM, Kraft NJ, Salamin N, Allen JM, Ault TR, Betancourt JL, Bolmgren K, Cleland EE, Cook BI, Crimmins TM, Mazer SJ, McCabe GJ, Pau1 S, Regetz1 J, Schwartz1 MD, Travers SE. Phylogenetic conservatism in plant phenology. J. Ecol.2013; 101:1520–1530.

  24. Sporbert M, Jakubka D, Bucher SF, Hensen I, Freiberg M, Heubach K, König A, Nordt B, Plos C, Blinova I, Bonn A, Knickmann B, Koubek T, Linstädter A, Masková T, Primack RB, Rosche C, Shah MA, Stevens AD, Tielbörger K, Träger S, Wirth C, Römermann C. Functional traits influence patterns in vegetative and reproductive plant phenology-a multi-botanical garden study. New Phytol. 2022;235:2199–210.

    Article  PubMed  Google Scholar 

  25. Zhang XY, Friedl MA, Schaaf CB, Strahler AH. Climate controls on vegetation phenological patterns in Northern mid-and high latitude inferred from MODIS data. Global Change Biol. 2004;10:1133–45.

    Article  Google Scholar 

  26. Luo ZK, Sun OJ, Ge QS, Xu WT, Zheng JY. Phenological responses of plants to climate change in an urban environment. Ecol Res. 2007;22(3):507–14.

    Article  Google Scholar 

  27. Primack RB, Higuchi H, Miller-Rushing AJ. The impact of climate change on Cherry trees and other species in Japan. Biol Conser. 2009;142(9):1943–9.

    Article  Google Scholar 

  28. Wang H, Zhong S, Tao Z, Dai J, Ge Q. Changes in flowering phenology of Woody plants from 1963 to 2014 in North China. Int J Biometeorol. 2019;63(5):579–90.

    Article  PubMed  Google Scholar 

  29. Wang HJ, Dai JH, Ge QS. The Spatiotemporal characteristics of spring phenophase changes of Fraxinus chinensis in China from 1952 to 2007. Sci China-Earth Sci. 2012;55:991–1000.

    Article  Google Scholar 

  30. Gao XY, Dai JH, Zhang MQ. Responses of variations of plant ornamental period to climate change in the West suburbs of Beijing from 1965–2014. Geographical Res. 2018;37(12):2420–32.

    Google Scholar 

  31. Wu DX, Zhang L, Song CY, Zhang SM. Protocols for standard biological observation and measurement in terrestrial ecosystem. Beijing: China Environmental Publishing Group; 2019.

    Google Scholar 

  32. Zheng JY, Ge QS, Hao ZX, Wang WC. Spring phenophases in recent decades over Eastern China and its possible link to climate changes. Clim Change. 2006;77:449–62.

    Article  Google Scholar 

  33. R Core Team. R: A Language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2014.

    Google Scholar 

  34. Borcard D, Gillet F, Legendre P. Numerical ecology with R, 2nd edition. Use R! series, Springer International Publishing AG. 2018.

  35. Jin Y, Qian H. V.PhyloMaker2: an updated and enlarged R package that can generate very large phylogenies for vascular plants. Plant Divers. 2022;44(4):335–9.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Diniz-Filho JAF, de Sant’Ana CER, Bini LM. An eigenvector method for estimating phylogenetic inertia. Volume 52. Evolution; 1998. pp. 1247–62.

  37. Pistón N, de Bello F, Dias AT, Götzenberger L, Rosado BH, de Mattos EA, Salguero-Gómez R, Carmona CP. (2019). Multidimensional ecological analyses demonstrate how interactions between functional traits shape fitness and life history strategies. J. Ecol.2019;107:2317–2328.

  38. Bianchini K, Morrissey CA. Species traits predict the Aryl hydrocarbon receptor 1 (AHR1) subtypes responsible for Dioxin sensitivity in birds. Sci Rep. 2020;10:1–11.

    Article  Google Scholar 

  39. Paradis E, Schliep K. APE 5.0: an environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics. 2019;35:526–8.

    Article  CAS  PubMed  Google Scholar 

  40. Ridgeway G, Edwards D, Kriegler B, Schroedl S, Southworth H, Greenwell B, Boehmke B, Cunningham J. gbm: Generalized boosted regression models (R package version 2.1). 2013.

  41. Hijmans RJ, Phillips S, Leathwick J, Elith J, Dismo. Species Distribution Modeling. R package version 0.9-1.2013.

  42. Elith J, Leathwick JR, Hastie T. A working guide to boosted regression trees. J Anim Ecol. 2008;77(4):802–13.

    Article  CAS  PubMed  Google Scholar 

  43. Xu J, Fan HD, Ni J. Meta-analysis of plant phenological change in China during 1950 and 2015. J Subtropical Resour Environ. 2019;14(2):1–11.

    Google Scholar 

  44. Miller-Rushing AJ, Primack RB. Global warming and flowering times in Thoreau’s concord: A community perspective. Ecology. 2008;89:332–41.

    Article  PubMed  Google Scholar 

  45. Mo F, Zhao H, Wang JY, Qiang SC, Zhou H, Wang SM, Xiong YC. The key issues on plant phenology under global change. Acta Ecol Sin. 2011;31(9):2593–601.

    Google Scholar 

  46. Tao Z, Wang H, Liu Y, Xu Y, Dai J. Phenological response of different vegetation types to temperature and precipitation variations in Northern China during 1982–2012. Int J Remote Sens. 2017;38:3236–52.

    Article  Google Scholar 

  47. Crimmins TM, Crimmins MA, David BC. Complex responses to climate drivers in onset of spring flowering across a semi-arid elevation gradient. J Ecol. 2010;98(5):1042–51.

    Article  Google Scholar 

  48. Westoby M, Leishman MR, Lord JM. On misinterpreting the phylogenetic correction. J Ecol. 1995;83:531–4.

    Article  Google Scholar 

  49. Boulter SL, Kitching RL, Howlett BG. (2006) Family, visitors and the weather: patterns of flowering in tropical rain forests of northern Australia. J.Ecol.2006; 94:369–382.

  50. Johnson SD. Climatic and phylogenetic determinants of flowering seasonality in the cape flora. J Ecol. 1993;81:567–72.

    Article  Google Scholar 

  51. Durka W, Michalski SG, Daphne. A dated phylogeny of a large European flora for phylogenetically informed ecological analyses. Ecology. 2012;93:2297–2297.

    Article  Google Scholar 

  52. Smith SA, Brown JW. Constructing a broadly inclusive seed plant phylogeny. Am J Bot. 2018;105:302–14.

    Article  PubMed  Google Scholar 

  53. Zanne AE, Tank DC, Cornwell WK, Eastman JM, Smith SA, FitzJohn RG, McGlinn DJ, Moles BCO, Reich AT, Royer PB, Soltis DL, Stevens DE, Westoby PF, Wright M, Aarssen IJ, Bertin L, Calaminus RI, Govaerts A, Hemmings R, Leishman F, Oleksyn MR, Soltis J, Swenson PS, Warman NG, Beaulieu L. Three keys to the radiation of angiosperms into freezing environments. Nature. 2014;506:89–92.

    Article  CAS  PubMed  Google Scholar 

  54. Gaudet CL, Keddy PA. A comparative approach to predicting competitive ability from plant traits. Nature. 1988;334:242–3.

    Article  Google Scholar 

  55. Sun S, Frelich LE. Flowering phenology and height growth pattern are associated with maximum plant height, relative growth rate and stem tissue mass density in herbaceous grassland species. J Ecol. 2011;99:991–1000.

    Article  Google Scholar 

  56. Segrestin J, Navas ML, Garnier E. Reproductive phenology as a dimension of the phenotypic space in 139 plant species from the mediterranean. New Phytol. 2020;225:740–53.

    Article  PubMed  Google Scholar 

  57. Atkin OK, Loveys BR, Atkinson LJ, Pons TL. Phenotypic plasticity and growth temperature: Understanding interspecific variability. J Exp Bot. 2006;57:267–81.

    Article  CAS  PubMed  Google Scholar 

  58. Huang L, Koubek T, Weiser M, Herben T. Environmental drivers and phylogenetic constraints of growth phenologies across a large set of herbaceous species. J Ecol. 2018;106:1621–33.

    Article  Google Scholar 

  59. Dunne JA, Harte J, Taylor KJ. Subalpine meadow flowering phenology responses to climate change: integrating experimental and gradient methods. Ecol Monogr. 2003;73:69–86.

    Article  Google Scholar 

  60. Menzel A, Sparks TH, Estrella N, Koch E, Aasa A, Ahas R, Briede A. European phenological response to climate change matches the warming pattern. Global Change Biol. 2006;12:1969–76.

    Article  Google Scholar 

  61. Fitter AH, Fitter RS. Rapid changes in flowering time in British plants. Science. 2002;296:1689–91.

    Article  CAS  PubMed  Google Scholar 

  62. Wang LX, Chen HL, Li Q, Yu WD. Research advances in plant phenology and climate. Acta Ecol Sin. 2010;30(2):447–54.

    Google Scholar 

Download references

Acknowledgements

We are grateful to the people who made great contribution to the collection of phenological observation data at ecological stations.

Funding

This work completed with the financial support from the Ministry of Sciences and Technology the People’s Republic of China (Grant/Award Number: 2022YFF1300103), Chinese Academy of Sciences (Grant/Award Number: KFJ-SW-YW043-4, KFJ-SW-YW037-01, XDA26010101-4).

Author information

Authors and Affiliations

  1. Key laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, Beijing, China

    Chuangye Song, Lin Zhang, Yuan Jia & Dongxiu Wu

Authors
  1. Chuangye Song
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Lin Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Yuan Jia
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Dongxiu Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Chuangye Song made the formal analysis, methodology, writing original draft, review and editing.Lin Zhang reviewed and edited the manuscript.Yuan Jia made the figures of this manuscript.Dongxiu Wu presented the conceptualization, reviewed the manuscript and edited the text.

Corresponding authors

Correspondence to Chuangye Song or Dongxiu Wu.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Song, C., Zhang, L., Jia, Y. et al. Phylogenetic relatedness and plant traits influenced flowering phenology change patterns in natural habitats in China (2003–2021). BMC Plant Biol 25, 654 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12870-025-06572-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12870-025-06572-0

Keywords

BMC Plant Biology

ISSN: 1471-2229

Contact us