Agricultural　machinery　is　an　important　non-road　mobile　source,　which　can　exhaust　multi-pollutants,　making　primary　and　secondary　contributions　to　the　air　pollution.　China　is　a　significant　agricultural　country　of　the　world;　however,　the　agricultural　machinery　emissions　research　is　at　an　early　stage,　and　an　emission　inventory　with　a　high　temporal-spatial　resolution　is　still　needed.　In　this　study,　a　comprehensive　emission　inventory　with　a　high　temporal-spatial　resolution　for　agricultural　machinery　in　China　was　first　developed.　The　results　showed　that　the　total　emissions　in　2014　were　262.69　Gg,　249.25　Gg,　121139　Gg,　2192.05　Gg,　1448.16　Gg　and　25.14　Gg　for　PM10,　PM25,　THC,　NOx,　CO　and　SO2,　respectively.　Tractors　and　farm　transport　vehicles　were　the　top　two　greatest　contributors,　accounting　for　approximately　39.9%-53.6%　and　17.4%-24.6%,　respectively,　of　the　total　emissions　of　the　five　pollutants　(except　THC).　The　farm　transport　vehicles　contributed　the　most　(81.8%)　to　the　THC　emissions.　The　county-level　emissions　were　further　allocated　into　1　km　x　1　km　grids　according　to　source-specific　allocation　surrogates.　The　spatial　characteristic　analysis　indicated　that　high　emissions　were　distributed　in　northeast,　north　and　central　south　China.　To　obtain　a　high　temporal　resolution　emission　inventory,　a　comprehensive　investigation　on　the　agricultural　practice　timing　in　different　provinces　was　conducted.　Then,　the　annual　emissions　in　the　different　provinces　were　distributed　to　a　spatial　resolution　of　ten-day　periods　(i.e.　the　early,　mid-　and　late　ten-day　periods　in　each　month).　It　was　found　that　higher　emissions　in　China　occurred　in　late　April,　mid-June　and　early　October.　In　addition,　the　emission　uncertainty　was　also　analyzed　based　on　the　Monte　Carlo　simulation.　The　estimated　high　temporal-spatial　resolution　emission　inventory　could　provide　important　basic　information　for　environmental/climate　implications　research,　emission　control　policy　making,　and　air　quality　modeling.　(C)　2018　Elsevier　Ltd.　All　rights　reserved.