Abstract: The construction of an effective drought index is the key to identifying and evaluate the actual drought conditions.Taking Xinjiang as the study area, based on the data of precipitation, air temperature and soil water content, we improved the existing standardized composite meteorological drought index (CI), used a day-by-day precipitation weighted attenuation method for rainfall processing, and developed a new comprehensive meteorological drought index (CIn index) with clear physical mechanisms and ease of application based on a water balance model.Furthermore, we evaluated the applicability of this index in Xinjiang, assessed its capability to characterize the entire drought evolution process, and analyzed the driving factors behind its variations.The results show that: ① The newly constructed CIn index solves the problem of the CI index's underestimation of meteorological drought in Xinjiang, and the CIn index evaluates drought events more accurately through drought duration and intensity, which is more suitable for meteorological drought conditions in Xinjiang.② The frequency of meteorological droughts across different time scales and drought levels based on the CIn index is characterized as spring>summer>winter>autumn, with "dry to wet" variations being dominant.Additionally, the drought conditions exhibited significant spatial heterogeneity.③ The CIn index is positively correlated with rainfall, the potential evapotranspiration and relative humidity, and negatively correlated with wind speed and average temperature.The CIn index not only overcomes the problem of underestimation of drought level that occurs in the comprehensive meteorological drought index, but also evaluates the drought trend more accurately, which can provide a theoretical basis for the prevention and control of meteorological drought.