Impervious　surface　coverage　is　a　quantifiable　environmental　indicator　correlating　closely　with　the　urban　thermal　environment.　This　study　measured　the　impervious　surface　development　in　the　subtropical　coastal　area　of　Xiamen,　southeastern　China,　and　investigated　its　impact　on　the　regional　thermal　environment.　The　Normalized　Difference　Impervious　Surface　Index　was　used　to　extract　impervious　surface　features　of　the　Xiamen　area　from　the　Landsat　Thematic　Mapper　images　of　1989,　1996,　and　2009,　respectively.　Multivariate　regression　analysis　was　implemented　to　examine　how　land　surface　temperature　(LST)　was　related　to　impervious　surface　and　to　some　other　urban　biophysical　descriptors　such　as　vegetation　and　water.　Results　show　that　the　impervious　surface　of　the　study　area　increased　by　approximately　eight　times　during　the　20　study　years.　Detailed　relationship　analysis　can　find　that　(1)　impervious　surface　is　positively　exponentially　correlated　with　LST,　while　vegetation　and　water　are　inversely　related　to　LST;　(2)　the　impervious　surface　contribution　to　regional　LST　change　can　be　up　to　six　times　greater　than　the　sum　of　vegetation　and　water　contributions　and　(3)　each　decrement　of　10%　impervious　surface　cover　with　additional　10%　green　or　water　space　could　lower　LST　by　up　to　2.9　or　2.5　°C,　respectively.　These　findings　would　facilitate　the　regional　urban　planning　strategies　to　minimize　the　local　temperature　rise　in　future　urban　growth.　©　2012　Royal　Meteorological　Society.