The　scientific　and　accurate　evaluation　of　water　resources　carrying　capacity　has　good　social,　environmental　and　resource　benefits.　Reasonable　selection　of　evaluation　parameters　is　the　key　step　to　realize　efficient　and　sustainable　development　of　water　resources.　Taking　Zanhuang　County　in　the　North　China　Plain　as　the　research　area,　this　study　selected　fuzzy　comprehensive　evaluation　models　with　different　weights　in　the　established　evaluation　index　framework　to　explore　the　sources　of　uncertainty　affecting　the　evaluation　results　of　water　resources　carrying　capacity.　By　using　the　sensitivity　analysis　method　of　index　weight,　the　index　with　the　biggest　influence　factor　on　the　evaluation　result　is　selected　to　reduce　the　uncertainty　problems　such　as　index　redundancy　and　small　correlation　degree.　The　results　show　that　the　correlation　and　reliable　of　comprehensive　evaluation　value　obtained　by　different　weight　methods　is　different.　The　evaluation　result　obtained　by　using　the　analytic　hierarchy　process　is　more　relevant　than　the　entropy　weight　method,　and　it　is　more　consistent　with　the　actual　load-bearing　situation.　The　study　of　sensitivity　index　shows　that　water　area　index　is　the　biggest　factor　affecting　the　change　of　evaluation　results,　and　water　resources　subsystem　and　socio-economic　subsystem　play　a　dominant　role　in　the　whole　evaluation　framework.　The　results　show　that　strengthening　the　data　quality　control　of　index　assignment　and　weight　method　is　helpful　to　reduce　the　error　of　water　resources　carrying　capacity　evaluation.　It　can　also　provide　scientific　basis　for　the　improvement　of　fuzzy　evaluation　model.