The　DisTrad　(Disaggregation　Procedure　for　Radiometric　Surface　Temperature)　model　shows　limited　applicability　for　sub-pixel　mapping　of　thermal　remote-sensing　images　in　densely　vegetated　areas　due　to　the　phenomenon　of　normalized　difference　vegetation　index　(NDVI)　saturation.　In　this　article,　we　compared　the　effect　of　NDVI　and　enhanced　vegetation　index　(EVI)　in　the　DisTrad　model　for　thermal　sub-pixel　mapping　in　densely　vegetated　areas　due　to　their　different　sensitivity　in　densely　vegetated　areas.　Taking　Ganzhou　in　Southern　China　as　an　example,　we　produced　250-m　thermal　remote-sensing　images　from　a　1000-m　image　using　250-m　NDVI　and　EVI　data.　After　comparing　with　the　synchronous　90-m　thermal　image　from　advanced　spaceborne　thermal　emission　and　reflection　radiometer,　we　found　that　the　EVI　can　achieve　a　better　result　than　NDVI　in　densely　vegetated　areas.