Abstract:Against the context of climate change, with rising temperature and uneven distribution of precipitation, the production of highland barley in Ganzi is increasingly affected by drought. Based on the daily meteorological data of 12 meteorological stations in the main barley planting areas of Ganzi from 1981 to 2022, the daily Meteorological drought Comprehensive Index (MCI) was calculated. The Mann-Kendall mutation test, Morlet wavelet analysis, and Hurst exponent based on ReScaled range (R/S) analysis were employed to examine the spatial-temporal distribution and change trends of drought intensity across various growth stages in Ganzi. This analysis aims to further comprehend the impact of drought intensity on various growth stages of highland barley. The results showed that: (1) Drought intensity exhibited a decreasing trend during the tiller-heading stage, while it increased during other growth stages. Drought intensity during the head-maturity stage was significantly higher than that during the sow-tillering stage and the entire growth stage. (2) Drought intensity in each growth stage experienced abrupt mutations in both the 1990s and the 2020s, while exhibited a significant decrease during the sow-tillering stage in 1985 and the tiller-heading stage in 2002. (3) The regions with the strongest drought intensity across different growth stages were located in Derong, the southwestern part of Ganzi, followed by surrounding counties such as Batang, Xiangcheng, and Daocheng, showing a distribution pattern that increases from the northeast to the southwest. (4) Over the next decade, drought intensity was projected to show a continuous drying trend during the sow-tillering stage, head-maturity stage and the entire growth stage, while the tiller-heading stage was expected to experience a continuous wetting trend. The temporal variation characteristics of drought intensity during different growth stages of highland barley in Ganzi show significant differences, while the spatial distribution characteristics exhibit similarities.