Compressive　sensing　(CS)　based　data　gathering　is　a　promising　approach　to　reduce　data　sampling　and　transmission　in　wireless　sensor　networks　and　thus　prolong　WSN＇s　lifetime.　The　physical　phenomena　are　generally　nonstationary　and　thus　the　sparsity　of　sensing　data　varies　in　temporal　and　spatial　domain.　In　order　to　guarantee　the　reconstruction　accuracy　with　lower　energy　cost　due　to　the　variation　of　sensing　data,　this　paper　proposes　an　adaptive　compressive　data　gathering　scheme　containing　adaptive　measurement　and　reconstruction.　The　adaptive　measurement　is　that　the　number　of　measurements　is　tuned　adaptively　according　to　the　prediction　of　the　change　trend　of　the　sensing　data.　The　adaptive　reconstruction　is　based　on　the　Stagewise　Orthogonal　Matching　Pursuit　(StOMP)　algorithm　and　using　the　Proportional-Integrative-Derivative　(PID)　method　to　adaptively　guarantee　the　reconstruction　accuracy.　At　last,　an　adaptive　compressive　data　gathering　system　is　built　on　Crossbow　Micaz　WSN　platform.　The　experimental　results　show　that　the　proposed　scheme　can　ensure　reconstruction　accuracy　with　low　energy　cost.　©　2017　IEEE.