Discussion
Although previous studies have largely examined the impacts of climate warming on plant phenology in alpine meadows on the QTP by using either manipulative experiments (Dorji et al. 2013; Wang et al. 2014; Shiet al. 2014) or remote sensing data (Zhang et al. 2018; Wang et al. 2020; Shen et al. 2015; Yu et al. 2010; Dong et al. 2013), only few studies (Chen et al. 2015) have documented the impacts of natural climate variations on plant phenology in alpine meadows in the long term. Therefore, using data from a period of 21 years, we investigated the impacts of climatic variation on phenological dynamics (green-up, flowering, fruiting, and withering periods) and growth patterns (heights) of different alpine plants from various functional groups.
We observed a significant delay in the green-up period forKobresia humilis , Stipa purpurea , and Artemisia scoparia , which is consistent with the findings of Yu et al.(2010), Chen et al. (2015), and Zhu et al. (2019), but in contrast to the observations made by Wang et al. (2020) using remote sensing modelling. Although our study sites were close to those of Wang et al. (2020) (about 80 km away), and both sites had a similar vegetation type (alpine meadow), we believe that the different results can be attributed to two major reasons. First, we observed warming and wetting trends in our research site (Zhou et al.2019), whilst Wang et al. (2020) observed warming and drying trends. This leads us to infer that different climate change patterns, associated with geometric and geographical factors, may result in different responses of plant phenology in neighboring sites, which suggests that long-time observations across different sites can provide a scientific basis for the study of alpine plants on the QTP under a changing climate. Second, we believe that different phenological observation methods have led to different findings. In our study, we used field observation, whereas Wang et al. (2020) applied remote sensing data. Additionally, the advance green-up period ofAstragalus laxmannii is consistent with the findings of Chen et al. (2015). Hence, we speculate that the green-up period of legumes on the QTP may be advanced under a warmer and wetter climate, which means that the response of spring phenology to climate change depends, among other factors, on the functional plant group.
For the withering period, Kobresia humilis , Stipa purpurea , and Artemisia scoparia showed delays, whereasKobresia humilis showed an advanced withering period, this supports the findings of previous studies, where the responses of autumn phenology were more complex (Chen et al. 2020) than those of spring phenology. Previous studies have also shown that reproductive phenology, in contrast to other phenological periods, is relatively stable in its response to climate change on the QTP (Jiang et al.2016). The trends of green-up and fruiting periods were consistent, which means that the functional groups and community scales show a consistently delayed fruiting period, which agrees with Zhang et al. (2011) and Yang et al. (2014).
The growth patterns of Kobresia humilis and Astragalus laxmannii were more sensitive to climate change than those ofStipa purpurea and Artemisia scoparia , which is inconsistent with Wang et al. (2020). This could be attributed to the measurement of growth patterns; i.e., we used changes in height, whereas Wang et al. (2020) used changes in biomass over time. The trends of plant biomass found in our study are not consistent with those observed by Liu et al. (2018) finding, i.e., the biomass of the sedgeKobresia humilis increased significantly, whereas there was no significant increase in the biomass of Stipa purpurea andArtemisia scoparia . We believe that these different responses may be due to the “root effect” (2018); Stipa purpurea with deeper roots can endure more severe droughts than Kobresia humilis with shallower roots. In our study area, Kobresia humilis may therefore benefit more significantly from wetting. However, the trends of the two forbs Artemisia scoparia and Astragalus laxmannii showed different responses, which may be related to the “functional group (legume and non-legume forb) effect”.
The responses of plant phenology and growth to climate change generally show a lag effect (Fu et al. 2019; Dong et al. 2013; Ahaset al. 2000; Li et al. 2016). For example, plants must undergo endodormancy before entering the green-up periods (Chuineet al. 2016); the spore cells begin to grow during the period of ecological dormancy, and the plant gradually recovers from dormancy and starts a new growth cycle before reaching the threshold of a certain temperature, water, or photoperiod (Fu et al. 2019). In our study, the optimum thresholds for all phenological phases and height growth patterns of most species were more than 30 days, which agrees with the results of previous studies. For example, 30–60 days of climate factors are related to the green-up period on the QTP (Donget al. 2013), and climate factors with a 6-week delay can best fit the flowering phenology of alpine plants on the QTP (Li et al. 2016).
There is evidence that the air temperature under a warming climate is the most important environmental factor affecting spring phenology, especially on the QTP (Zhang et al. 2018; Wang et al.2020; Yu et al. 2010; Wang et al. 2014; Shi et al.2014). However, in our study, it was soil temperature rather than air temperature that determined the spring phenology or the start of the rapid growth phase of our four species, confirming our hypothesis that soil temperature is the primary driver of plant growth. However, our results show that increased soil temperature may delay spring phenology or the start of the rapid growth phase because of an insufficient chilling period under a warmer climate (Yu et al. 2010; Ernakovich et al. 2014).
Previous studies have shown that the daytime temperature in summer exceeded the physiological threshold of alpine plants (Sherry al. 2007; Aldridge al. 2011) and delayed reproductive phenology. However, here, it was soil temperature rather than air temperature that delayed the flowering of Kobresia humilis . This may be explained by the fact that the flowering period of this species was earlier (in late spring) and it reached a lower height (less than 10 cm). Besides, we found that the maximum air temperature delayed the fruiting of Kobresia humilis , confirming previous hypotheses (Sherry et al. 2007; Aldridge et al. 2011).
The withering periods of Kobresia humilis , Stipa purpurea , and Artemisia scoparia were advanced under a warming climate. However, these results do not support the hypothesis that autumn warming will prolong the growing season, which has been stated by several authors (Zhang et al. 2018; Liu et al. 2016). This may be related to the adaptability of these species: an earlier ending of the growth period and an advanced dormancy can help to avoid unfavorable conditions and transfer more resources to the roots for later years (Xieet al. 2015).
Regarding the period of maximum growth, air temperature rather than other factors was the most impacting important for Astragalus laxmannii . This is consistent with the findings of a previous studies (Gonsamo et al. 2018), emphasizing that the average air temperature is always lower than the optimum air temperature for plant growth, especially in alpine regions. Wang et al. (2020) speculated that the moisture in the soil limited the end of the rapid growth period, but this could not be confirmed in our study. The different responses may be due to the different climatic conditions in the two sites (Shen et al. 2015) and the different calculation methods for growth patterns (Wang et al. 2020).
Although some studies have investigated the effects of manipulative warming on reproductive phenology on the QTP (Dorji et al. 2013; Wang et al. 2014; Zhu et al. 2016), the trade-off relationships between reproductive phenology and vegetative phenology and its impact on the ecosystem in the context of climate change have been largely neglected. We confirm the hypothesis that there are significant negative correlations between reproductive and vegetative phenology (Millerrushing et al. 2008; Wang et al. 2014; Arft et al. 1999) in the four functional groups. Besides, there was a similar trade-off in the growth pattern, and significant negative correlations were also found between the length of the rapid growth phase and the intrinsic growth rate in all four functional groups, confirming the hypothesis that there is a trade-off between the length of the rapid growth phase and the intrinsic growth rate (Brienenet al. 2020) in alpine plants. However, only one of the trade-offs played a key role in regulating aboveground biomass, and the impacts of the phenological or growth processes on alpine plant biomass are species-specific: soil temperature and average air temperature affect Stipa purpurea (significantly negative) andArtemisia scoparia (significantly positive) biomass by impacting vegetative reproduction; soil temperature and maximum air temperature affect Kobresia humilis biomass by affecting reproductive and vegetative phenology, which is inconsistent with previous findings (Maet al. 2017; Xu et al. 2018). Also, soil temperature and average/ maximum air temperature affect Astragalus laxmanniibiomass by affecting the length of the rapid growth phase and the intrinsic growth rate, which does not support the hypothesis of Suonan et al. (2019).
In summary, our findings highlight the trade-offs between phenological dynamics and plant growth patterns of key species in the alpine grassland of the QTP against the background of a changing climate. We explored the role of air and soil temperature in driving observation models, which is of great significance for understanding the feedbacks of alpine plants to climate change. The functional groups cannot be ignored in examining the responses of alpine plants to climate change. Although the predicted warming of the climate may have significant positive effects on the biomass of sedges and legumes, in non-legume forbs, the growth patterns of phenological dynamics may be affected. Our findings have profound implications for the adaptation of livestock feeding systems and the sustainable delivery of key ecosystem services on the Qinghai-Tibetan Plateau.
Acknowledgements The authors would also thank the anonymous reviewers for their helpful comments, the endeavor of editors and reviewers was also appreciated. This research was financially supported by the grants from the grants from: Second Tibetan Plateau Scientific Expedition and Research Program (2019QZKK0307); National Key R&D Program of China (2016YFC0501906) and Qinghai Provincial Key R&D program in Qinghai Province (2018-NK-A2).
AUTHORSHIP: S.L., S.D., Y.F., and S.L. conceived the ideas, designed the study, and led the writing of the manuscript. S.L., B.Z., H.S., Y.X., X.G., J.X., S.W., and F.L. collected the data. S.L. and S.D. analyzed the data. All authors contributed to the drafts and revision and gave the final approval for publication.
Competing interests: The authors declare no competing interests.