As one of the most destructive and costly natural disasters, drought has far-reaching negative effects on agriculture, water resources, the environment, and human life. Scientific understanding of propagation from meteorological to hydrological drought is of great significance for accurate forecasting of hydrological drought and preventing and mitigating drought disasters. The objective of this study is to analyze the spatio-temporal variational characteristics of propagation from meteorological drought to hydrological drought and the associated driving mechanisms in the eastern Qilian Mountains using the standard precipitation index (SPI), standardized runoff index (SRI), and drought propagation intensity index (DPI). The results show that there has been meteorological humidification and hydrological aridification in the upper reaches of the Shiyang River Basin over the last 56 years; especially in the 2000s, the intensity of hydrological drought was the strongest and the intensity of meteorological drought was the weakest, indicating the propagation intensity of meteorological drought to hydrological drought was extremely strong during this period. The changes of meteorological and hydrological dry–wet are different, both on seasonal and monthly scales. The meteorological dry–wet is shown to have had a significant effect both on the current and month-ahead hydrological dry–wet, where the one-month lag effect was most obvious. The relationship between meteorological and hydrological droughts also vary in space: Hydrological aridification in the Huangyang River, and the rivers east of it, was greater than that in the western tributaries. The drought propagation intensities from west to east showed a decreasing trend, excluding the Huangyang River. Climate and land-use changes are the main factors affecting the propagation from meteorological drought to hydrological drought. When the natural vegetation area accounted for between 76.3–78%, the cultivated land area between 0.55–3.6% and the construction area between 0.08–0.22% were a peer-to-peer propagation process from meteorological drought to hydrological drought in the upper reaches of the Shiyang River.

Vegetation is one of the most important indicators of climate change, as it can show regional change in the environment. Vegetation health is affected by various factors, including drought, which has cumulative and time-lag effects on vegetation response. However, the cumulative and time-lag effects of drought on different terrestrial vegetation in China are still unclear. To address this issue, this study examined the cumulative and time-lag effects of drought on vegetation from 2001 to 2020 using the Standardized Precipitation Evapotranspiration Index (SPEI) in the Global SPEI database and the Normalized Difference Vegetation Index (NDVI) in MOD13A3. Based on Sen-Median trend analysis and the Mann–Kendall test, the change trend and significance of the NDVI from 2001 to 2020 were explored. The Pearson correlation coefficient was used to analyze the correlation between the SPEI and NDVI at each cumulative scale and time-lag scale and to further analyze the cumulative and time-lag effects of drought on vegetation. The results show the following: (1) The NDVI value increased at a rate of 0.019/10 years, and the increased area of the NDVI accounted for 80.53% of mainland China, with a spatial trend of low values in the west and high values in the east. (2) The average SPEI cumulative time scale most relevant to the NDVI was 7.3 months, and the cumulative effect demonstrated a high correlation at the scale of 9–12 months and revealed different distributions in different areas. The cumulative effect was widely distributed at the 9-month scale, followed by the 12-month scale. The correlation coefficients of cumulative effects between the SPEI and NDVI for cropland, woodland and grassland peaked at 9 months. (3) The average SPEI time-lag scale for the NDVI was 6.9 months, and the time-lag effect had the highest correlation coefficient at the 7-month scale. The strongest time-lag effect for cropland and grassland was seen at 7 months, while the strongest time-lag effect for woodland was seen at 6 months. Woodland had a lower time-lag effect than grassland at different scales. The research results are significant for their use in aiding the scientific response to drought disasters and making decisions for climate change precautions.