Purpose　From　perspective　of　risk　management　and　control,　the　ecological　risk　contributed　from　various　sources　can　give　direct　risk　information　in　selecting　priority　pollutants　to　control　in　river　ecosystem　restoration.　However,　studies　combined　receptor　model　with　risk　assessment　to　obtain　ecological　risk　in　every　source　category　that　is　still　limited.　It　is　essential　to　quantitatively　and　spatially　analyse　the　contributions　of　various　sources　to　both　the　content　and　ecological　risk　in　river　system.　Materials　and　methods　Ten　trace　metals　in　47　sediment　samples　were　collected　in　Le＇an　River,　which　is　the　main　tributary　of　the　largest　freshwater　lake　in　China,　Poyang　Lake　and　analysed　by　laboratory　analysis.　The　potential　ecological　risk　index　(PERI)　model　was　used　to　evaluate　the　ecological　risk　posed　by　sediment　trace　metals.　The　positive　matrix　factorization　(PMF)　was　utilized　to　identify　the　source　of　sediment　trace　metal　and　apportion　their　contribution.　The　ecosystem　risks　of　the　sources　were　quantitatively　analysed　by　combining　PERI　and　PMF,　termed　PMF-PERI.　In　addition,　we　conducted　a　spatial　dissimilarity　analysis　of　the　source　contribution　distributions.　Results　and　discussion　The　ecological　risk　assessment　results　showed　that　the　metals　Cd,　Cu,　and　Hg　may　cause　the　main　ecological　hazards.　Source　apportionment　of　the　trace　metal　concentrations　indicated　that　the　dominant　source　of　trace　metals　in　the　sediment　was　human　activities　(55.83%),　such　as　Pb-Zn　mining　and　agricultural　activities　(30.33%),　gold　smelting　(13.62%)　and　copper　mining　(11.89%).　Source　apportionment　of　the　ecological　risks　posed　by　the　trace　metals　indicated　that　natural　sources　were　the　main　ecosystem　risk　sources　(64.75%),　followed　by　Pb-Zn　mining　and　agricultural　activities　(14.65%),　copper　mining　(10.67%)　and　gold　smelting　(9.93%).　The　results　of　spatial　dissimilarity　analysis　indicated　that　the　significant　spatial　variation　in　the　contribution　compositions　was　related　to　the　locations　of　the　sources　and　the　river.　Conclusions　The　results　of　this　study　indicated　that　the　emission　reduction　of　agricultural　activities　should　be　given　priority　attention　for　protecting　the　watershed　environment　of　the　Le＇an　River.　In　addition,　this　study　extends　the　use　of　risk　assessment　in　receptor　model　source　apportionment.