Evapotranspiration　(ET)　is　an　important　reference　for　scientific　management　of　farmland　irrigation,　accurate　estimation　of　crop　yields,　and　optimal　allocation　of　water　resources.　The　purpose　of　this　paper　is　to　estimate　ET　by　taking　the　farmland　of　Kai-Kong　River　Basin　in　Xinjiang　province　as　an　example.　The　method　was　based　on　the　SEBAL　model　and　the　Landsat　data;　it　estimated　the　net　radiation　flux,　soil　heat　flux　using　the　land-surface　parameters　first,　and　the　instantaneous　evapotranspiration　was　then　calculated　using　the　energy　residual　method　and　the　latent　heat　flux.　The　sinusoidal　function　method　was　used　to　extend　the　time　scale　of　the　instantaneous　evapotranspiration　to　the　daily　evapotranspiration　and　the　daily　evapotranspiration　were　verified　using　the　FAO56　Penman-Monteith　formula.　The　compared　results　shown　that　the　error　of　daily　evapotranspiration　between　SEBAL　and　FAO56　Penman-Monteith　is　within　1mm,　and　except　for　the　estimation　accuracy　of　final　growth　stage　is　about　80%,　the　estimation　accuracy　of　other　growth　stages　are　more　than　80%.　The　SEBAL　model,　coupled　with　the　Landsat　remote　sensing　data,　can　offer　an　efficient　and　adequate　alternative　to　estimate　ET　at　catchment　scale.　©　Published　under　licence　by　IOP　Publishing　Ltd.