Trends in Flowering Phenology of Herbaceous Plants and Its Response to Precipitation and Snow Cover on the Qinghai—Tibetan Plateau from 1983 to 2017

Based on limited controlled experiments, both advanced and delayed shifts in flowering phenology induced by precipitation and snow cover have been reported on the Qinghai–Tibetan Plateau (QTP). To clarify the impact of precipitation and snow cover on flowering phenology, we conducted a comprehensive statistical analysis of the temporal change in flowering phenology and its responses to precipitation and snow cover changes using regression models built on the largest collection of ground phenological observation data on the QTP. We found that first flowering date (FFD) for the early-flowering time series significantly advanced at the rate of −0.371 ± 0.149 days/year (p p = 0.108). Cumulative pre-season precipitation regressed with FFD positively for early-flowering time series, with the explained variation ranging from 11.7 to 49.4% over different pre-season periods. The negative impact of precipitation on flowering phenology is unexpected, because an increase in precipitation should not hamper plant growth in the semi-arid and arid environments on the QTP. However, precipitation was found to be inversely correlated with temperature. Thus, it is likely that temperature, and not precipitation, regulated flowering phenology over the study period. No relationship was found between FFD and snow-cover melt date or duration. This result indicated that snow cover may not affect flowering phenology significantly, which may be because plant flowering time was much later than the snow-cover melt date on the QTP. These findings contrast the results of controlled experiments on the QTP, which showed that precipitation regulated flowering phenology, and with other studies that showed that snow-cover melting time determined flowering dates of early-flowering species in high latitude and Arctic zones in Europe and North America, where the low-temperature environment is similar to the QTP. These findings can improve flowering phenology models, assist in the prediction of phenological responses of herbaceous plants to climate change, and forecast changes in the structure and function of the grassland ecosystem on the QTP