Response of grassland growing season length to extreme climatic events on the Qinghai-Tibetan Plateau.

Extreme Climatic Events (ECEs) are increasing in intensity, frequency, and duration as the earth warms, which greatly affects the vegetation phenology. However, the response of vegetation phenology to different types of ECEs (e.g., extreme hot, extreme cold, extreme drought, and extreme wet) has not been extensively studied. To fill this knowledge gap, we investigated the relationship between the length of growing season (LOS) of grassland and ECEs on the Qinghai-Tibetan Plateau (QTP). First, we analyzed the spatial distribution and interannual trends of phenology based on the MODIS Normalized Difference Vegetation Index (NDVI). Second, we used Coincidence Rate (CR) analysis to quantify the relationship between LOS anomalies and ECEs. Finally, we analyzed the sensitivity of LOS to the intensity of ECEs. The results indicated that the spatial distribution of LOS was closely related to local hydrothermal conditions, with longer LOS in places with more precipitation or higher temperatures during the growing season, and LOS extended by 0.28 days/year from 2000 to 2022. Moreover, we found that the CR of negative LOS anomalies to ECEs notably exhibited variations along climatic gradients, with higher CR to extreme hot generally occurring in warmer areas. Meanwhile, the CR of extreme wet increased while the CR of extreme drought decreased with increasing precipitation. We also found that the sensitivity of LOS to ECEs changed more markedly, along the climatic gradients, in alpine ecoregions compared to temperate ecoregions. Overall, the sensitivities of LOS ranked in descending order of absolute sensitivity to extreme drought, extreme wet, extreme hot, and extreme cold. This study furthers our understanding of the grassland response to ECEs under different hydrothermal conditions, which can provide valuable reference for the management and conservation of grassland ecosystems in QTP under future climate change scenarios.