The　measurement　of　aerosols　(PM1.0　and　PM2.5)　was　conducted　during　2016　and　2017　in　Beijing,　Tangshan　and　Shijiazhuang,　investigating　the　spatial　and　temporal　variations　of　aerosols　and　major　chemical　components.　The　WRF-Chem　model　was　applied　to　simulate　the　impacts　of　aerosol　direct　and　semi　direct　feedbacks　on　meteorological　factors　and　identify　the　source　of　PM2.5.　The　results　showed　that　the　average　annual　concentrations　were　63.3-88.7　mu　g/m(3)　for　PM1.0　and　81.3-112　mu　g/m(3)　for　PM2.5　at　the　three　study　cities,　and　the　average　seasonal　concentration　ratios　of　PM1.0/PM2.5　ranged　from　64.3%　to　86.0%.　PM1.0　and　PM2.5　showed　a　good　correlation　that　the　squared　correlation　coefficients　were　all　higher　than　0.9,　indicating　both　mainly　came　from　the　same　emission　sources.　Water-soluble　inorganic　ions　and　carbonaceous　components　were　major　chemical　species　in　PM1.0　and　PM2.5,　accounting　for　48.9%-54.1%　and　25.6%-27.8%　in　PM1.0,　48.1%-52.3%　and　22.7%-24.7%　in　PM2.5.　Those　chemical　species　showed　spatially　similar　characteristics　but　pronounced　seasonal　differences,　with　higher　concentrations　in　autumn　and　winter,　lower　in　spring　and　summer.　Aerosol　feedbacks　had　different　effects　on　various　meteorological　factors.　Three　study　cities　monthly-mean　incoming　solar　radiation　decreased　by　40.6　W/m(2),　82.2　W/m(2),　38.4　W/m(2),　and　49.9　W/m(2);　planetary　boundary　layer　height　reduced　by　54.0　m,　109　m,　32.2　m　and　85.2　m;　temperature　at　2　m　decreased　by　0.5　degrees　C,　0.8　degrees　C,　0.5　degrees　C　and　1.3　degrees　C;　relative　humidity　increased　by　1.5%,　2.6%,　1.3%　and　4.7%　in　April,　July,　October　and　January,　respectively,　while　wind　speed　changes　were　relatively　smaller　than　above　factors.　Additionally,　the　major　sources　of　PM2.5　in　January　were　identified　as　transportation　in　Beijing,　while　industrial　and　domestic　sources　in　Tangshan　and　Shijiazhuang.　The　obtained　results　will　provide　more　in-depth　and　comprehensive　understanding　of　aerosol　pollution　and　management　strategies.　(C)　2018　Elsevier　Ltd.　All　rights　reserved.