Abstract: Meteorological drought indice is one of the indices to measure the degree of agricultural drought.This study selected four drought indices:Precipitation Anomaly Percentage（PA）,Standardized Precipitation Index（SPI）,Standardized Precipitation Evapotranspiration Index（SPEI）,and New Standardized Precipitation Evapotranspiration Index（n SPEI） constructed through SPI and SPEI.Utilizing meteorological data from 1961 to 2020 in the northern Xinjiang agricultural oasis region,comparisons were made in correlation,regression fitting,and the distribution of drought frequency at different levels to optimally select a drought indice that provides a more realistic description of drought conditions.Furthermore,the meteorological drought variation characteristics in the research area were analyzed.The results indicate:（1） The drought standard of PA underestimated drought severity,SPI was insensitive to drought exacerbation caused by temperature rise,and the Thornthwaite method used in SPEI calculation overestimated the impact of temperature on drought.At the same time,nSPEI overcame the shortcomings of SPI and SPEI,showing the best applicability to the research area.（2）Over nearly 60 years,meteorological drought in the research area exhibited fluctuating characteristics over time.Since the 21st century,spring and summer have shown wet-dry variations,while autumn exhibited a weak dry-wet variation. In terms of drought intensity,spring and summer have strengthened,while autumn has slightly weakened.On the drought scale,drought has been predominant,with localized drought increasing in spring and decreasing in summer and autumn.Regarding drought frequency,extreme droughts are more common in spring,and severe droughts are more common in summer.