Both of the component measurement and the analysis based on Air Quality Model (AQM) are essential for the consideration of PM2.5 reduction measures. In order to collect more information regarding PM2.5 components, long-term weekly measurement for inorganic ions in PM2.5 had been conducted from August 2009 to August 2011 at Meisei University, located in Western Tokyo. The performance of AQM was also clarified based on the comparison between CMAQ simulation and the observation. Observed SO42- and NO3- showed higher concentration in summer and winter season, respectively. Since the ratio of total amount of inorganic ions in PM2.5 mass concentration reached up to 75% in each February, reduction measures for the inorganic ions seemed to be more effective to decrease PM2.5 mass concentration especially in winter season. As AQM performance, a correlation coefficient of around 0.8 was obtained by the comparison with observed SO42-. However, AQM did not show a reasonable performance for NO3- and NH4＋ due to both of the sampling artifact originated in the volatilization of NH4NO3 and the model uncertainties especially for the overestimation of NO2 and O3. As a result, it was suggested for the improvement of Nitrate prediction that further observation studies for the volatilization and the precursors such as HNO3 and NH3 were important as well as the evaluation of overestimated NO2 and O3.