Taking　the　＇9.3＇　military　parade　in　2015　and　two　red　alerts　of　heavy　air　pollution　in　December　of　the　same　year　as　examples,　the　characteristics　of　meteorological　factors　and　pollutant　concentration　variation　were　compared.　Based　on　the　estimation　of　pollutant　emission　reduction　under　different　periods,　the　WRF-CAMx　model　was　used　to　evaluate　the　effect　of　PM2.5　pollution　improvement.　The　results　showed　that　the　daily　average　PM2.5　concentration　(19.0　μg•m-3)　during　the　parade　(from　August　20　to　September　4)　decreased　by　60.0%　and　48.0%,　respectively,　in　comparison　to　that　before　(August　15-19)　and　after　(September　5-15)　the　parade.　The　daily　average　PM2.5　concentration　(232.3　μg•m-3)　during　the　first　red　alert　period　was　higher　than　that　of　the　second　red　alert　(216.6　μg•m-3).　The　air　quality　before　the　second　red　alert　was　better　than　that　before　the　first　red　alert.　The　proportion　of　emission　reduction　during　the　parade　was　generally　larger　than　that　during　the　red　alert　periods,　which　provided　a　controllable　and　favorable　condition　for　the　realization　of　the　＇Parade　Blue＇.　The　PM2.5　concentration　in　Beijing　decreased　by　32.4%,　17.1%,　and　22.0%　under　the　control　measures　during　the　periods　of　the　＇9.3＇　parade,　the　first　red　alert,　and　second　red　alert,　respectively.　The　higher　proportion　of　PM2.5　concentration　reduction　could　be　attributed　to　the　more　intensive　regional　emission　reduction　and　the　favorable　meteorological　conditions.　The　intensity　of　the　pollution　reduction,　the　timing　of　the　implementation　of　emergency　control　measures,　and　meteorological　conditions　were　the　most　important　factors　that　may　have　influenced　the　improvement　of　pollution.　©　2020,　Science　Press.　All　right　reserved.