https://doi.org/10.5061/dryad.z34tmpgpr

Description of the data and file structure

Changes in reproductive phenology induced by warming are happening across the globe with significant implications for plant sexual reproduction. However, the response of plant reproductive output (number of flowers and fruits) and success (successful fruits/total flowers) in response to climate change have not been well-characterized. Here, we conducted a warming and altered precipitation experiment in an alpine meadow on the eastern Tibetan Plateau to investigate the effects of climate change on the reproductive phenology and success of six common species belonging to two flowering functional groups. We found that warming advanced the start of flowering and the start of fruiting in mid-summer flowering plants. Warming reduced the reproductive output of early-spring flowering plants but did not change their reproductive success. The effects of warming and altered precipitation on the reproductive output and success of mid-summer flowering plants were year-dependent, and the fruiting phenology regulated the response of the mid-summer flowering plant’s reproductive success to climate change. The findings highlight the critical role of fruiting phenology in the reproductive success of alpine plants and imply that alpine plants may reduce their fitness by producing fewer flowers and fruits under climate warming, especially for later flowering plants.

Code/software

Linear mixed-effects (LME) models were used to assess the effects of year, warming, and precipitation changes on soil temperature and moisture, in which year, warming, precipitation changes, and their interactions were treated as fixed factors, while the block was treated as a random factor. To investigate the effects of warming and precipitation changes on the phenology and reproductive characteristics of different flowering functional groups, we also employed the LME models to assess the effects of year, warming, precipitation changes, flowering functional groups, and their interactions on key phenological events (including reproductive phenology and phenological duration), as well as on reproductive output and success. Here, year, warming, precipitation, flowering functional groups, and their interactions were treated as fixed factors, while the block was treated as a random factor. We chose a linear mixed effects model considering the year according to the Akaike Information Criterion (AIC) (Appendix S1: Tables S10). Furthermore, given the significant annual climatic fluctuations in the region [29, 38, 49] and the fact that we did not monitor soil temperature and moisture during the non-growing season, we did not consider the effects of soil temperature, soil moisture, and their annual change rates on alpine plant phenology and reproduction.

One-way ANOVA followed by Tukey’s HSD tests was used for multiple comparisons to detect differences in reproductive phenology across different years among the six treatment levels of the two flowering functional groups. We applied a similar statistical strategy using LME models when analyzing the reproductive output and success of the two flowering functional groups, in which warming, precipitation, and their interaction were treated as fixed factors, and the block was treated as a random factor. Differences were defined as significant when p < 0.05. Linear regression models were constructed to analyze the relationships between phenological events/reproductive output and success, and the changes in soil temperature and moisture, focusing on species and flowering functional groups across all plots, to evaluate the response of reproductive phenology to variations in soil temperature and moisture.

 To investigate whether the responses of reproductive phenology of alpine plants to climate change regulate their reproductive output and success, we conducted the following three analyses. First, Pearson’s correlation analysis was used to analyze the correlation between reproductive phenology and reproductive output across different flowering functional groups. Second, residuals from the regressions of reproductive phenology events or reproductive output on the soil temperature were used to test for independent effects of changes in phenology on the reproductive output. Third, piecewise structural equation modeling (SEM) was utilized to analyze the main factors influencing alpine plant reproductive success under warming and precipitation changes, as well as their underlying mechanisms. We initially formulated an a priori model encompassing all hypothesized pathways (Appendix S1: Figure S9). Furthermore, given the disparities in temperature sensitivity and drought tolerance exhibited by different flowering functional groups [29, 30, 38, 47], the impending impacts of warming and precipitation changes on their reproductive phenology and reproduction are likely to diverge significantly. Thus, the data used in the SEM were collected from each plot across two years and we ran the model separately for each flowering functional group (early-spring flowering plants and mid-summer flowering plants). We assessed model fit by d-separation tests and  test. Finally, we used Fisher’s statistics and P value to evaluate the fit of the model [61].

All the statistical analyses were conducted in R version 4.3.0 (R Development Core Team, 2022), with the ‘lme’ function from the ‘nlme’ package for all linear mixed-effects models. All Tukey’s HSD tests for multiple comparisons of differences use the ‘TukeyHSD’ function in the ‘multcomp’ package. The ‘psem’ function in ‘piecewiseSEM’ package for SEM [62].